Sie sind auf Seite 1von 15

World Development Vol. 71, pp.

7993, 2015
2013 Elsevier Ltd. All rights reserved.
0305-750X/$ - see front matter
www.elsevier.com/locate/worlddev

http://dx.doi.org/10.1016/j.worlddev.2013.11.002

RuralUrban Migration, Household Vulnerability,


and Welfare in Vietnam
LOC DUC NGUYEN, KATHARINA RAABE and ULRIKE GROTE *
Leibniz University Hannover, Germany
Summary. This paper investigates the interaction of migration, vulnerability to poverty, and welfare of rural households in three
provinces in Central Vietnam. It addresses three questions. (1) To what extent do shocks motivate rural household members to move
to urban areas?, (2) Are migrants in the new urban settings better o in terms of working conditions and quality of life?, and (3) What
is the eect of migration on rural households welfare and vulnerability to poverty? The analysis uses panel data of 2200 households from
rural Vietnam covering the period 20072010, and a tracking survey of 299 migrants from 2010. The empirical evidence from a probit
model shows that migration, especially migration for employment, is a livelihood support strategy for households exposed to agricultural
and economic shocks. Migration for education is more likely observed among households with higher human capital and being nancially better o. Nevertheless, the probability of migration decreases with the employment opportunity in the village. Migrants perceive
themselves to be better o at the place of destination, but income losses from shocks of their rural households may reduce their employment quality. The results from dierence-in-dierence specications with propensity score matching techniques suggest that migration
has positive income growth eects, and that these eects are more pronounced in provinces with fewer job opportunities. These eects
help not only migrant households moving out of poverty, but it also improves the poverty situation in rural areas.
2013 Elsevier Ltd. All rights reserved.
Key words migration, vulnerability to poverty, dierence-in-dierence, propensity score matching, Vietnam

1. INTRODUCTION

become a strategy for households in rural areas of Vietnam


to reduce income uctuation. However, a substantial share
of individuals and households who migrated in search for better income opportunities could not improve their living conditions. These problems arise from the lack of knowledge and
experience when living in modern cities. Additionally, the
inadequate implementation of labor laws (Le, Tran, &
Nguyen, 2011), or the limited access to aordable health care
services, among others (United Nations Population Fund
(UNFPA), 2010) made the migrants become vulnerable in
their destinations. Moreover, the 2008 global economic crisis
aggravated the vulnerability of migrants. Some migrants
stopped sending remittances or returned to their households
at the place of origin (Oxfam & VASS, 2010). These challenges
are likely to aect the motivation of migration and the welfare
of rural households.
The main objective of this paper is to assess the success of
migration as a livelihood support and risk coping strategy of
rural households who sent some of their household members
to urban areas during the period 20082010. However, not
only original households in the rural areas are analyzed, but
also migrants from the urban areas. Therefore, an employment quality index (EQI) is developed that deploys information on a variety of indicator variables of working and living
conditions to quantify the success of individual migrants in
the city. The eectiveness of migration as livelihood support
strategy of rural households is determined by comparing the
changing welfare outcomes of migrant and non-migrant

Rural households in developing countries face several types


of unpredictable events threatening their livelihoods. Among
others, these include economic shocks like price uctuations,
and natural disasters like droughts or oods. Due to the absence of basic social safety nets and sucient and comprehensive insurance schemes, rural people, especially the poor, often
have to cope with the eects of shocks and associated risks on
their own. Specically, they diversify their livelihoods, save for
precautionary purposes, or join mutual support networks
(Dercon, 2002).
Migration is one livelihood strategy that households in ecologically vulnerable communities pursue to diversify their income sources and to overcome the adverse welfare eects of
social, economic, and institutional constraints in their places
of origin (Ezra, 2001; Tongruksawattana, Schmidt, & Waibel,
2010). Migration increases household income and smoothes
income uctuations mainly through remittances (Stark &
Bloom, 1985).
Like in other developing countries, rural households in Vietnam, whose livelihoods deeply depend on agriculture, face
substantial income variability because of climate change and
price uctuations in the context of rapid liberalization and reform processes. Moreover, the gap between government support and the loss through damage have increased over time
(see Appendix Figure 1). Therefore, rural residents smooth
their consumption through savings, mutual support, or private
transfers including remittances (Newman & Wainwright, 2011;
Phung & Waibel, 2009).
Over the past decade, Vietnam has experienced an exponential increase in the movement of people both within and across
its borders. By meeting the demand for labor created by industrial development and foreign direct investments following the
Doi Moi reforms, migration plays an important role in Vietnams economy, and contributes to poverty reduction (Cu,
2005, chap. 5; Dang, Tacoli, & Hoang, 2003). Migration has

* The research was nanced by the German Research Foundation (DFG)


under the umbrella of the research project Impact of shocks on vulnerability to poverty: Consequences for development of emerging Southeast
Asian economies (DFG FOR 756). The authors would like to thank
Mulubrhan Amare, Lena Hohfeld, Dean Jollie, and Hermann Waibel as
well as two anonymous reviewers for their valuable comments.
79

80

WORLD DEVELOPMENT

households by means of the propensity score matching


procedure.
Past studies on migration in Vietnam are based on Household Living Standard Surveys which are not appropriate to
study migration as only ocially registered migrants are included in the sample. This current paper builds on (1) a panel
data set of about 2200 rural households from three provinces
in Vietnam and (2) a tracking migrant survey of 299 migrants.
Given this unique data set, the results provide a new perspective on migration in Vietnam.
The paper is structured as follows. Section 2 reviews the
main theories of migration and related empirical evidence.
Section 3 describes the data set and methodology applied to
assess the success of migration in rural and urban Vietnam.
Section 4 presents and discusses the empirical results of the panel data analysis and the migrant survey analysis. Finally, Section 5 concludes.
2. LITERATURE REVIEW
The economic literature on migration assumes that individuals or households rationally consider various locations and
choose the place that maximizes the expected gains from
migration. The expected gains from migration depend on a
large number of factors such as personal characteristics and
experiences, social networks, wealth, or reduced vulnerability
to poverty. Dierent theories and models have advanced to account for their importance.
In his laws of migration, Ravenstein (1885, 1889) linked
migration patterns to conditions of labor force surpluses and
decits, with people moving from surplus labor areas to decit
labor areas in order to improve their living conditions. He also
developed the idea of the pull and push factors in order to
explain the forces driving migration. Pull factors are social,
economic, political, or environmental incentives at the place
of destination, such as job opportunities, better education,
and living conditions. Push factors are incentives at the place
of origin that force people to out-migrate. Specic factors include insucient job and employment opportunities, insecurity regarding political, social, or economic conditions, or
the loss of wealth (Lee, 1966). Other classical migration models exist from Sjaastad (1962) and have been further developed
by Harris and Todaro (1970) and Mincer (1978).
Stark and Bloom (1985) developed a fundamentally dierent
theory of migration called the New Economics of Labor
Migration (NELM). According to this approach, migration
decisions are joint family decisions, although this does not imply that household members move jointly. In fact, households
decide on the migration of few household members so as to
maximize and smooth household income and ensure sustainable livelihoods through the spatial/local diversication of
household resources such as labor. Migration is thus a strategy
for managing and minimizing risk to household income and
survival.
One of the important contributions of NELM is the link between migration decision and risks. The costs of migration
(including those associated with risks and opportunities) are
shared among household members, thus creating a co-insurance system between migrant and non-migrant household
members. The co-insurance system involves (1) family support
to the migrant in the case of need (risk) in the destination area
and (2) migrant support to the family via remittances to facilitate risk coping at the place of origin. In addition to risk and
wage dierentials, models also link migration to social capital,
the existence of functioning social networks among migrants,

non-migrants and return migrants, and migration institutions


(Massey, 1990).
Migration studies also depart from the NELM approach to
identify the factors behind migration and the well-being of migrants. For instance, Agesa and Kim (2001) used data from
Kenya to identify the determinants of migration decisions.
Their results show that migration is relatively more likely
among workers facing a positive urban to rural earnings dierence, suggesting that skilled workers self-select to migrate to
urban areas. Giesbert (2007) reports evidence from Western
Kenya according to which the propensity to migrate depends
on education and migrant networks, but not on household
wealth. Ezra (2001) nds that individuals belonging to economically poor households in ecologically vulnerable communities have a higher propensity to out-migrate than those from
less vulnerable regions in Ethiopia.
Several empirical studies investigated the impact of migration on rural households welfare but with ambiguous results.
Evidence from Thailand suggests that migration reduces income inequality mainly through changes in the distribution
of productive assets (Garip, 2010). Another study from Thailand reveals that poor rural households tend to produce poor
migrants, which could be one of the reasons for the continuous
existence of a wide ruralurban gap in welfare (Amare,
Hohfeld, & Waibel, 2012). Similarly, Azam and Gubert
(2006) report ndings from Mali and Senegal according to
which remittances cause rural households to reduce their work
eort, which reduces the eectiveness of migration as a poverty reduction instrument. Fuente (2010) uses household panel
data from Mexico for the period 19982000 to assess how
likely households with a high level of vulnerability to poverty
receive remittances. However, contrary to the expectation,
households with a higher level of vulnerability to poverty have
a lower probability to receive remittances.
The working and living conditions of migrants in the destination places are also analyzed in several studies. Shah (2000)
uses four indicators to assess the degree of success of migrant
workers in Kuwait. The indicator variables include (1) objective measures such as the migrants salary and job permission
and (2) subjective measures regarding job satisfaction. Among
others, migrant workers are asked to indicate whether the job
is the same or better than expected prior to migration. The results show that human and social capital are the main factors
contributing to the success of migrants. Akay, Bargain, and
Zimmermann (2011) use general health questionnaires to identify the factors that aect the subjective well-being of rural
workers, migrants, and local urban workers in China. Their
study nds that the well-being of migrants positively depends
on the length of the migration period, the quality of working
conditions, and the existence of community ties. Amare et al.
(2012) calculate employment quality indices for migrants in
Thailand. They conrm that human capital is a major factor.
Along with government support, human capital can improve
living and working conditions of migrants in the city.
In Vietnam, most empirical studies on migration are based
on data from three surveys of the General Statistical Oce
in Vietnam (GSO): (i) the population census which is conducted every 10 years, (ii) the Vietnam Living Standard Survey (VLSS) from 199394 and 199798 which was replaced
by the Vietnam Household Living Standard Survey (VHLSS)
published biannually from 2002 onwards, and (iii) the migration survey from 10 provinces conducted in 2004 in cooperation with the United Nations Population Fund (UNFPA)
(GSO & UNFPA, 2005).
Based on the population census of 1999 and 2009, GSO
(2011) analyzed the patterns and trends of internal migration,

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

dening a migrant as a person who has changed the living


place in the last 5 years. The results show that ruralurban
migration has become an important trend with an estimated
rate of 8.9% in 2009, directed especially to central cities or
provinces with high levels of industrialization. Migrants
have become younger and have better living conditions, but
migration also widens the socioeconomic gaps between rural
and urban areas. GSO and UNFPA (2005) analyzed the
socioeconomic conditions of migrants in the cities based on
the 2004 migration survey (see UNFPA, 2010). The results reveal that the majority of migrants move because of economic
reasons. This is also supported by Dang et al. (2003) and
Niimi, Pham, and Reilly (2009), both arguing that rural outmigrants shifted to urban areas to benet from increased economic opportunities. Although the dominant share of migrants feels better o and sends remittances to their families,
migrants face many problems in the destination area. One of
the biggest problems concerns Vietnams complex household
registration system. The system controls and monitors the
changes of peoples residence in Vietnam by classifying them
into dierent residential categories, each being associated with
certain rights and obligations. The registration system may reduce the benets from migration by constraining the access of
internal migrants to basic public services such as education or
health services in the absence of registration (Niimi et al.,
2009). Furthermore, even though incomes increase after
migration, the average incomes of migrants are still lower than
the incomes of the local residents in the destination areas. In
addition, many migrants are employed on a temporary basis
without formal labor contracts or social protection.
Using the VLSS 199394 and 199798, De Brauw and
Harigaya (2007) nd that migrant households expenditure
levels exceed those of non-migrant households by approximately 5%. Nguyen, Tran, Nguyen, and Oostendorp (2008)
use the VHLSS 20022004 panel data set to assess the eect
of migration on household expenditures and inequality in
rural areas. The study shows that migration is a highly selective process, strongly inuenced by household and village
characteristics. Furthermore, migration positively aects
household expenditures, while at the same time increasing
the degree of income inequality in rural areas. In a followup study with the VHLSS 20042006 panel data, Nguyen,
Den Berg, and Lensink (2009) conrm the positive household
expenditure eects of migration, but also report a slight decline in poverty and inequality. Pincus and Sender (2008)
voice concern about studying migration based on the data
sets of the VLSS and VHLSS, since only the ocially registered households being for at least 6 months in the survey
location are covered. Migrants without a permanent residence status are ignored in the sample. Based on their own
small survey of workers in VHLSS enumeration areas, they
conrm that the number of unregistered workers is considerable so that the ocial data sets underestimate the actual
number of migrants. Further shortcomings of the ocial surveys are that it is not possible to link the migrants with their
original households so that the impact of remittances on the
welfare of rural households cannot be assessed. Moreover,
temporary migrants are not captured so that the results are
likely to underestimate any internal migration trends.
The present study is motivated by the existing ambiguous results based on unreliable data sets. It seeks to provide new
empirical evidence regarding the extent to which migration is
a successful livelihood support strategy for rural households
in Vietnam. It uses a unique data set linking rural households
with their migrating members in urban cities.

81

The remainder of this paper addresses three research questions. First, to what extent do shocks motivate rural household members to move to urban areas? Second, are migrants
in the new urban settings better o in terms of working conditions and quality of life? Third, what is the eect of migration
on rural households welfare and vulnerability to poverty?
3. DATA AND METHODOLOGY
(a) Data
The study uses a panel data set that contains information on
a random sample of 2200 households 1 from the three provinces Dak Lak, Thua Thien Hue, and Ha Tinh in Vietnam.
Household data were collected in 2007, 2008, and 2010. The
questionnaires for the household survey covered a broad set
of questions regarding the socio-demographic and economic
conditions of the sampled households. Among others, specic
interest was with the migration experience of the household
and the household members, with the composition of the income source portfolio, with borrowing and lending patterns,
and especially, the exposure to demographic, social, economic,
and agricultural shocks. In addition, village heads were interviewed in 2007 and 2010 in order to collect general information about their village such as village population,
employment structure, infrastructure characteristics, and resource use patterns.
For analyzing the motivations of migration and evaluating
the impacts of migration on household welfare, a migrant
and a migrant household need to be dened. In this paper, a
migrant is a household member having lived outside of the village for at least 1 month in the year 2008 and/or 2010. A migrant household is a household that has at least one migrant in
that period. Since a household member may have migrated out
already in 2007, the household could have benetted from
remittances having an eect on the per capita income variable. To avoid such endogeneity problems, households with
migrants in 2007 are dropped from the sample. 2 The remaining sub-data set consists of about 1711 households. Of these,
443 are migrant households with 890 migrants, including both
ruralrural and ruralurban migrants. Since this paper focuses
on ruralurban migration, the 158 ruralrural migrant households are also dropped from the data set. The remaining 285
ruralurban migrant households are used for the analysis.
Almost 60% of those migrant households have at least one
member who migrated in search of a job, and 33% migrated
for educational reasons. The remaining 8% indicated other
reasons like followed the family or went to help another
household.
In addition to the household survey, a migration survey of
299 migrants is the basis for this study. The migration survey
is a tracking survey in which the respondents are migrant
household members of the surveyed rural households. The migrant status of the household members was determined in a
two-step procedure. First, information from the rural household survey in 2008 was used to construct a list of migrant
household members. Second, the migrant status of the household members was conrmed during the rural household survey in 2010. The respective information was then directly
transmitted to the migrant survey team, which implemented
the migrant survey at the same time in Ho Chi Minh City
and its two surrounding and highly industrialized provinces
Dong Nai and Binh Duong which have the highest rates of
net migration (UNFPA, 2010).

82

WORLD DEVELOPMENT

Because the majority of migrants work in the informal sector and frequently change their contact details, the migrant
survey team succeeded in interviewing 299 out of 600 potential
migrant respondents. Questions addressed the migration process, the shocks, risks, and socio-economic situation of migrants in the destination area, and the type and nature of
links between the migrants and their rural households. Among
the 299 interviewed migrants, there are 233 wage-employed, 15
self-employed, and 51 other migrants with irregular jobs. To
measure the employment quality index to quantify the success
of these migrants in the city, only wage-employed migrants are
explored.
(b) Methodology
This paper estimates three basic models, one for each research question. The rst two models involve the estimation
of probit and linear regression specications. The third model
is estimated by means of the propensity score matching and
dierence-in-dierence procedure in order to control for unobservables between migrant and non-migrant households.
(i) Model 1: The migration decision model
In a rst step, a non-linear probability model is estimated
that links the household migration status in 2010 and 2008
to household and village characteristics in 2007, respectively.
The model is dened as in Eqn. (1).
PrDij;2010;2008 1 FXij;2007 ; Zj;2007 ; FEProv :

The dependent variable captures the probability that household i in village j is a migrant household in 2008 and/or
2010. Here, the binary dummy Dij,2010,2008 equals one if household i in village j had at least one migrant household member
in 2008 and/or 2010 and zero otherwise.
The probability of being a migrant household in 2008 and/
or 2010 is a function of observable household characteristics X
in 2007 (i.e., Xij,2007). The vector specically contains information on (1) socio-demographic household factors (HHCij,2007)
such as the gender and education of the household head and
household members and the share of dependent household
members, (2) household wealth (WIij,2007) such as per capita
income, household indebtedness, income sources, or land
holdings, and (3) the households exposure to demographic,
social, economic, or agricultural shocks (SEij,2007). Vector X
of household i in village j is accordingly dened as
Xij;2007 HHCij;2007 ; WIij;2007 ; SEij;2007 :

In addition to household characteristics, the probability of


being a migrant household is also linked to a set of observable
village characteristics Z. Being equal to
Zj;2007 RCj;2007 ; Distj;2007 ;

the vector of village characteristics includes information on


road conditions (RCj,2007) and the distance to the district
headquarters (Distj,2007) in 2007. Finally, province xed eects
FEProv are included to capture the migration eects of province-specic unobservables.
Since households decide that their members migrate out for
several reasons, the motivation for each purpose may be
dierent. Therefore, we estimate three alternative equations
presenting household migration decisions (i) for all types of
reasons; (ii) for employment, and (iii) for education.
(ii) Model 2: Measuring employment quality of migrants
One of the main objectives of this paper is to identify the
factors aecting the livelihood of migrants in the urban

destination area. Closely related to Shah (2000) and Amare


et al. (2012), we assess the livelihood situation of migrant m
by using k indicator variables Ik of the migrants employment
and living conditions, which we then combine into a composite
indexthe employment quality index (EQI). The EQI combines information from a set of subjective indicator variables,
which indicate whether migrants perceive their (1) income to
be stable, (2) working conditions to have improved since their
last job, and (3) living conditions to have improved since they
have left the rural area. The EQI also includes objective indicator variables, specifying whether migrants have (1) accumulated savings, (2) above average income levels, and (3) a
written employment contract.
Each indicator variable Ik equals one if the underlying condition holds in 2010 and zero otherwise. The EQI of migrant m
is then dened as an unweighted average of the subjective and
objective dummy variables, dened as
EQIm;2010

6
1X
I mk;2010 :
6 k1

The EQI assumes values between zero and one, with employment quality being better for larger values.
Given the specication of the EQI, we dene a linear model
that links the EQI to a set of (1) individual characteristics of
migrant m (MCm,2010) and (2) characteristics of the household i that migrant m belongs to (HHCim,2010). The characteristics of migrants include gender, age, education level, the
length of migration period, the type of job, job characteristics, and the way of getting the job. Household characteristics
cover the loss from shocks that a household might have faced
in 2010 like income loss and asset loss due to shocks (SEim,2010).
Variables related to ethnicity and whether a household
belongs to a political or social organization are also added
to the model. The EQI model is then given as in Eqn. (5).
EQIm;2010 FMCm;2010 ; HHCim;2010 ; SEim;2010 :

(iii) Model 3: Evaluating the impact of migration on rural household welfare


The third model seeks to quantify the eect of migration on
rural household welfare. First, the average treatment eect on
the treated is identied (Heckman & Navarro-Lozano, 2004),
where treatment refers to the households migration status and
the treatment eect arises by comparing outcome of households with migrants against that of comparable households
without migrants. It specically follows model (6).
ATT EY1i  Y0i jD 1 EY1i jD 1  EY0i jD 1:
6
ATT abbreviates the average treatment eect on the treated,
which measures the impact of migration on the outcome of migrant households. D is a binary dummy variable that equals
one if the household has at least one migrant and zero, otherwise. Y1i and Y0i denote the outcome of household i with and
without migrants, respectively.
It is impossible to compute the outcome of the migrant
household in case no one migrated (Y0i|D = 1), as this
variable is unobserved. In order to solve this problem,
this paper employs the method of Propensity Score
Matching (PSM). Under the assumption of conditional
independence, this method pairs the set of observable
characteristics of migrant households to some group of
comparable non-migrant households by creating propensity
scores (Rosenbaum & Rubin, 1983). The outcome of migrant

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

households that they would have if they had not migrated is


estimated from the outcome of the comparison groups of
non-migrant households. Specically, the outcome of the
migrant household in case no one migrated (Y0i|D = 1) is
approximated by the observable value of the outcome of the
comparable non-migrant household E(Y0i|D = 0).
In this paper, we use the migration decision model (1) to
estimate the propensity score, and the Nearest-Neighborhood
and Kernel matching methods to estimate the outcome that
migrant households would have had in case none of their
household members had migrated.
However, the standard PSM method controls for selection
on observable variables, but cannot account for unobserved
variables and their simultaneous eect on the probability of
migration and the outcome variable (Rosenbaum & Rubin,
1983). We therefore use the method of dierence-in-dierences
with PSM to eliminate the eect of unobserved (time-invariant) variables on the outcome variable (Smith & Todd,
2005). This approach also helps to address the endogeneity
problem that usually precludes the identication of the outcome eects of migration.
Given this, the impact of migration on household outcome
growth is rewritten as
ATT Y12010  Y12007 jX2007 ;D 1  Y02010  Y02007 jX2007 ;D 0;
7
where ATT denotes the average treatment eect on the treated. The subscripts 2007 and 2010 denote the baseline and
the year of migration, respectively. To assess the variability
of matching estimators, we present bootstrap standard errors
based on 1000 replications.
(iv) Outcomes
In this paper, the outcomes are household income growth,
vulnerability to poverty, and poverty and inequality indices
for indicating the change in household welfare.
+Income growth: It is the observable change in income of
migrant households from 2007 to 2010.
+Vulnerability to poverty (VTP): It is measured as the
probability that a household (or an individual), whether currently poor or not, would nd itself poor in the future. It measures the household consumption with respect to the
consumption poverty line (Chaudhuri, 2003; Dercon, 2002;
Klasen, Lechtenfeld, & Povel, 2013). The impact of migration
on the poverty indices of migrant households is calculated as
follows:
DVTP PVTP12010 jD 1  PVTP02010 jD 1;

VTP12010

where
is the estimated vulnerability to poverty of migrant households in 2010, and VTP02010 the counterfactual vulnerability to poverty which was estimated previously by PSM
procedure.
+Poverty and inequality indices: These include three Foster
GreerThorbecke (FGT) poverty indices and the Theils L and
Theils T inequality indices (Haughton & Khandker, 2009).
The impact of migration on the poverty indices of the migration household are calculated as follows:
DP PE12010 jD 1  PE02010 jD 1;
E12010

where
is the observed per capita expenditure of the migrant household, and E02010 is the counterfactual per capita
expenditure which was estimated previously.
We also measure the impact of migration on poverty and
inequality for the total rural population capturing also synergy

83

eects of migration on the poverty and inequality in the villages:


DI IZ12010  IZ02010 ;

10

where IZ12010
and IZ02010 is

is calculated directly for the entire data sample,


measured through the estimated counterfactual
expenditure per capita of migrant households and observed
per capita expenditure of non-migrant households.
4. RESULTS AND DISCUSSION
This section presents the econometric results of the study.
Section (a) presents the evidence on the factors driving migration. Section (b) discusses the employment quality index and
identies the factors that inuence employment quality. Section (c) presents the evidence on the eect of migration on rural household welfare.
(a) Explaining the households migration decision
This section discusses the factors inuencing the households
migration decision by the probit model (1). Table 1 shows the
results. Summary statistics of the models variables are presented in Appendix Table 6.
In general, the coecient estimates show the expected signs,
and the signicance properties are robust to alternative model
specications. 3 The evidence suggests that migration is the
preferred livelihood support strategy of households having
experienced agricultural and economic shocks. This is especially true for migration for employment, implying that an
agricultural or economic shock forces a rural household to
send out its member(s) to nd a job in urban areas to compensate for the loss in households income. Agricultural shocks include oodings, droughts, crop pests or livestock diseases,
whereas economic shocks relate to job loss, collapse of business, strong increase of input prices, or strong decrease of output prices. Demographic and social shocks like e.g., illness or
death of a household member, theft or conict with neighbors
in the village are in contrast not likely to incite a migratory response within a household. Interestingly, these shock variables
are not statistically signicant in the case of migration for education (Appendix Table 6).
Considering the socio-demographic household characteristics, the propensity of migration in general, and migration
for employment in specic, signicantly increases with the
age of the household head. Thus, the older the household
head, the more likely his or her member(s) migrate out to nd
employment. This variable becomes statistically insignicant
in case of migration for education. Moreover, the propensity
of migration signicantly decreases with the relative number
of household dependents. This nding suggests that the propensity to migrate is higher in households that are characterized by a larger share of productive laborers.
Consistent with the predictions of migration theory, migration depends on the level of human and social capital. It appears that the probability of migration increases with the
share of household members with completed secondary education. It is more pronounced in case of migration for education
when both, the share of household members with completed
secondary education and the share with professional training
increase. These variables become statistically insignicant in
case of migration for employment. This means that migration
for employment does not necessarily require a higher education level. Looking at the descriptives of the sample, we nd
that indeed 44% of all migrants lack secondary school

84

WORLD DEVELOPMENT
Table 1. Determinants of household migration (probit regression)
Variables

Household characteristics in 2007


HH experienced demographic shocks (1-yes, 0-no)

All migration decision Job migration Education migration decision


decision
(1)
(2)
(3)
0.108
(0.084)

0.143
(0.100)

0.155
(0.143)

HH experienced social shocks (1-yes, 0-no)

0.096
(0.180)

0.178
(0.317)

0.252
(0.342)

HH experienced agriculture shocks (1-yes, 0-no)

0.146*
(0.087)

0.220**
(0.108)

0.249
(0.158)

HH experienced economic shocks (1-yes, 0-no)

0.273*
(0.159)

0.392**
(0.190)

0.322
(0.216)

Female headed HH (1-yes, 0-no)

0.146
(0.122)

0.105
(0.143)

0.101
(0.205)

Age of HH head (years)

0.016***
(0.003)

0.019***
(0.004)

0.003
(0.007)

Share of HH members w/completed secondary school

0.178***
(0.044)

0.054
(0.054)

0.390***
(0.064)

Share of HH members w/completed high school or professional education

0.041
(0.039)

0.010
(0.046)

0.191***
(0.062)

HH members belong to political or social organization (1-yes, 0-no)

0.187*
(0.109)

0.164
(0.139)

0.281
(0.199)

1.553***
(0.216)

1.966***
(0.263)

2.190***
(0.430)

Log of monthly HH per capita income (PPP $ in 2005)

0.042
(0.046)

0.002
(0.048)

0.120*
(0.072)

HH engaged in o-farm activities (1-yes, 0-no)

0.073
(0.087)

0.050
(0.098)

0.262*
(0.147)

0.091***
(0.034)

0.133***
(0.041)

0.031
(0.060)

0.032
(0.098)

0.243**
(0.124)

0.047
(0.186)

0.143
(0.112)

0.151
(0.128)

0.112
(0.181)

0.209***
(0.053)

0.219***
(0.058)

0.102
(0.079)

Ha Tinh province (1-yes, 0-no)

0.535***
(0.142)

0.517***
(0.183)

0.271
(0.200)

Thua Thien Hue province (1-yes, 0-no)

0.253
(0.155)

0.407**
(0.181)

0.062
(0.228)

1.836***
(0.310)

2.225***
(0.329)

2.520***
(0.538)

1432
231,09
0.00
0.16
582.86

1326
177.32
0.00
0.17
387.04

1245
144.28
0.00
0.27
184.47

Dependency ratio

Log of land per capita (hectare)


HH is indebted (1-yes, 0-no)

Village characteristics in 2007


Village road condition (1-good condition, 0-bad condition)
Log distance from village to district headquarters (km)

Constant
Number of observations
Wald chi2(18)
Prob > chi2
Pseudo R2
Log pseudolikelihood

Standard errors are clustered at the village level.


Source: Own calculations based on the DFG Rural Village Survey 2007 and the DFG Rural Household Survey 2007, 2008 and 2010.
*
Denotes statistical signicance at 10%.
**
Denotes statistical signicance at 5%.
***
Denotes statistical signicance at 1%.

Dependents are household members below the age of 10 years and above the age of 65 years. The dependency ratio is the number of dependents relative
to the total number of household members.

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

qualications. 4 Households with membership in political or


social organizations also display a larger propensity to migrate
in general. However, the results turn statistically insignicant
in case of migration for employment and education as well,
indicating that social capital is not really important for the
migration decision.
As regards the measures of household wealth, households
with larger per capita income levels in 2007 are more likely
to be engaged in migration for education. They may consider
migration as an investment strategy. For migration for
employment, the income variable becomes statistically insignicant. When looking at the level of income of the households in the data set, it can be seen that households with
members migrating for education have a relatively higher level
of per capita income as compared to households with members migrating for nding a job. Furthermore, a quarter of
all rural households with migrants for employment are poor
in 2007. These observations support the idea that if households support migration out of desperation, then this only applies to households with migrants who look for employment.
Surprisingly, the variable Household engaged in o-farm
activities is only statistically signicant in case of migration
for education. Therefore, the more the households are engaged
in o-farm activities, the less likely they allow their members
to migrate out for education. In general, households with
more land per capita are less likely to be migrant households
because they need more laborers for agricultural production.
This is especially true for the case of migration for employment but not for the case of migration for education.
Considering indebtedness, migration for employment is
more likely observed among households with open nancial
debt positions. Especially rural households thus send a migrant out expecting that remittances will facilitate the repayment of their outstanding debts. In this case, migration for
employment is more likely a desperation strategy than an
investment strategy. 5
Overall, the descriptives show that three quarters of all migrant households have only one migrant, while 18% have two
migrants, and 7% have three or more migrants. About half of
all migrants send remittances, no matter how many migrants a
rural household has. But interestingly, of those migrants who
send remittances, a quarter is better o, while three quarters
are not better o. Again, this evidence conrms that migration
is likely to be a desperation strategy.
Gender does not have any signicant eect on the migration
decision. This is true even when separating into migration for
employment and migration for schooling. In general, we nd
that about 60% of all migrants are male and 40% female. Half
of the migrants for education are male and around 60% of the
migrants searching for jobs are male. The average age of a migrant is 24 years with female migrants being on average
2 years younger than male migrants. With respect to education, the male migrants have on average one additional year
of schooling, but overall, 44% of all migrants lack secondary
school qualications. This is true for male and female migrants alike. 15% of the female migrants are from poor households, whereas this share amounts to 11% for the male
migrants.
Finally, the households decision to migrate for education is
not inuenced by any village characteristics like remoteness
and road infrastructure development. However, households
near to the district headquarters have a higher propensity to
send members to migrate for employment. Moreover, migration does depend on unobserved provincial eects. Specically, the evidence suggests that households from the Ha
Tinh and Thua Thien Hue provinces are more likely to be

85

engaged in migration for employment in comparison to households from the Dak Lak province. 6 Consistent with the argument of UNFPA (2010), this may reect cross-province
dierences in economic development and cross-province dissimilarities in employment prospects and income opportunities. Nevertheless, the results do not show these eects in the
case of migration for education.
(b) Assessing the well-being of migrants in the destination areas
Subjective and objective indicator variables are used to measure the working and living conditions of migrants in urban
areas. These indicator variables are separately reported for migrants who are working in industry, production, and in the
service sector; and for migrants according to the number of
migration years.
According to the subjective indicator variables, the majority
of migrants in industry and production as well as in the service
sector (1) perceive their income to be stable and (2) report
improvements in working and living conditions. Maybe
reecting the eect of exaggerated expectations, migrants with
less than one year of migration experience are least likely to
perceive the outcomes of migration positively. The share of
satised migrants is highest among those with 35 years
of migration experience. For migration periods in excess of
5 years, migrants are again less likely to positively evaluate
working and living conditions.
Considering the objective indicator variables, migrants in
the service sector are more likely to have savings than migrants
in industry and production. This sectoral eect is consistent
with the observation regarding the higher average daily wage
in the service sector (see Appendix Table 7).
In general, only roughly every second migrant has a written
employment contract. The probability of having a written contract increases with the years of migration experience and is
slightly higher for migrants in industry and production. A
written employment contract inuences working and living
conditions as it provides migrants access to social protection
programs (see also GSO & UNFPA, 2005; Oxfam & VASS,
2010; UNFPA, 2010).
Approximately half of the surveyed migrants report income
levels below the sample average. Pointing to the greater vulnerability of recent migrants, below average income levels
are reported by three-quarter of all migrants with less than
one year of migration experience. The share of migrants with
below average income decreases with the number of migration
years, being lowest for the group of migrants with 710 years
of migration experience (21%).
Finally, the indicators from Table 2 are used to construct a
composite employment quality index (EQI) for the sample of
wage-employed migrants, which account for 78% of all migrants in the destination areas. In order to account for possible
dierences in the relative importance of subjective and objective indicator variables, we present three employment quality
indices according to Eqn. (4). These reect the unweighted
average of (1) the three objective migration assessment criteria
(hereafter referred to as objective EQI), (2) the three subjective
migration assessment criteria (hereafter referred to as subjective EQI), and (3) the set of both objective and subjective
assessment criteria (hereafter referred to as aggregate EQI).
The use of subjective and objective indicator variables yields
dierent conclusions regarding the average share of migrants
that is satised with the living and working conditions (see
Appendix Figure 2). Here, the distribution of the subjective
EQI is skewed to the left pointing to a larger share of satised
migrants. It thus appears that it is the environment in general

86

WORLD DEVELOPMENT
Table 2. Migrants working and living conditions by occupation and length of migration period (% of total)
Industry/production sector

Public/private service sector

Length of migration period (years)


<1

13

35

57

710

>10

Subjective assessments
Income is stable
Working conditions have improved
Living conditions have improved

71.7
68.1
85.8

70.6
73.5
86.0

57.6
69.7
69.7

71.7
60.4
83.0

86.7
75.0
91.7

78.9
73.7
86.8

61.8
70.6
85.3

60.0
73.3
100.0

Average

75.2

76.7

65.7

71.7

84.5

79.8

72.6

77.8

Objective assessments
Migrant has accumulated savings
Migrants with above average income
Migrants have written employment contract

31.9
52.2
55.8

50.0
51.5
47.8

45.5
24.2
48.5

22.6
37.7
45.3

45.0
60.0
51.7

36.8
68.4
68.4

52.9
79.4
55.9

73.3
73.3
66.7

Average

46.6

49.8

39.4

35.2

52.2

57.9

62.7

71.1

Note: Regardless of the choice of indicator variable, the number of observations equals 233.
Source: Own calculations based on the DFG Migrant Survey 2010.
*
The subjective indicator variables reect perceptions as reported by migrants.

Average income is computed across all migrants with employment in either the industry and production or service sector.

Table 3. Explaining the degree of migrants employment quality


Variables

(1)

(2)

0.020
(0.005)

0.019
(0.005)

0.021***
(0.005)

Gender (1-female; 0-male)

0.160***
(0.039)

0.158***
(0.040)

0.141***
(0.041)

Debt (1-yes; 0-no)

0.008
(0.046)

0.002
(0.046)

0.006
(0.047)

Age (years)

0.018***
(0.005)

0.024***
(0.004)

Length of migration period (years)

0.033*
(0.018)

Job in service sector (1-yes, 0-no)

0.030
(0.041)

0.029
(0.041)

0.020
(0.042)

Pay to get a job (1-yes; 0-no)

0.040
(0.055)

0.032
(0.054)

0.062
(0.055)

How to nd a job (1-introducing from friends or relatives; 0-others)

0.062
(0.042)

0.061
(0.043)

0.061
(0.043)

0.018
(0.095)

0.046
(0.098)

0.045
(0.090)

HH members belong to political or social organization (1-yes, 0-no)

0.023
(0.155)

0.026
(0.156)

0.030
(0.161)

Income loss from shocks (2005 $ PPP)

0.000*
(0.000)

0.000**
(0.000)

0.000
(0.000)

Constant

0.282
(0.216)

0.345
(0.218)

0.016
(0.197)

228
0.000
0.211

228
0.000
0.202

228
0.000
0.171

Number of observations
Prob > F
Adjusted R2

Note: The estimates are derived from an OLS regression specication.


Standard errors are adjusted for heteroscedasticity.
Source: Own calculations based on the DFG Migrant Survey 2010 and DFG Rural Household Survey 2010.
*
Denotes the statistical signicance at the 10% level.
**
Denotes the statistical signicance at the 5% level.
***
Denotes the statistical signicance at the 1% level.

***

(3)

Year of schooling (years)

Household characteristics in 2010


Ethnic (Kinh & Hoa = 1, others = 0)

***

0.071***
(0.014)

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

and factors inuencing living conditions in particular that


inuence the extent to which migrants perceive their stay in urban areas to be a success. The aggregate index is also slightly
skewed to the left, suggesting that a relatively larger share of
migrants is very satised with the living and working conditions in the urban destination area. The objective EQI, however, is normally distributed and is thus most suitable to
identify the factors that aect the EQI for the sample of
wage-employed migrants (model 2). The descriptive variables
used in this model are presented in Appendix Table 8.
Table 3 summarizes the estimation results based on three
alternative sets of independent variables. In the rst alternative set, all variables are included. Since the variables length
of the migration period and age are correlated, one of the
two variables is removed from the second but included in the
third alternative set, and vice versa. The results show that on
the one hand, migrants being female, better-educated, older,
and with longer migration periods are more likely to report
a higher objective EQI. This means that they are more likely
to have accumulated savings, above average income, and a
written employment contract. On the other hand, indebtedness
and the way they got the job do not statistically aect their

87

objective EQI. Having to pay for a job does not seem to guarantee a higher objective EQI as indicated by the negative sign.
The gender eect reects the fact that women have more stable and predictable working relations. Indeed, around 60% of
the female migrants have a job with a written contract, as
compared to 40% of the male migrants. The gender eect
may also reect the dierent spending behavior. In fact, the
descriptive information conrms that female migrants generally have higher savings than their male counterparts. The gender eect, however, does not reect above average incomes, as
women are paid lower salaries as compared to men.
In general, the types of jobs being conducted by female and
male migrants in the city are very diverse. 54% of all women
work in industry/production, mainly in weaving, but also in
textile and electronics factories that are more likely to provide
written contracts and stable employment. The remaining 46%
women work in the service sector in jobs like as accountant in
banks (14%), tailor (7%), waiter, sales person, hair dresser, or
cleaner/housemaid. In comparison, only 37% of all men work
in industry/production, including the weaving sector (12%)
and to a very small extent electronics and textiles factories.
Otherwise, men are more likely to be employed in the service

Table 4. Dierence-in-dierence estimates of the impact of migration on household income growth


Outcome variables

Treatment

Control

Dierence in average outcome ATT

General migration
Income growth (Kernel)
Income growth (Nearest-Neighborhood)

0.56
0.55

0.36
0.28

0.20*(0.09)
0.27*(0.11)

Migration for employment


Income growth (Kernel)
Income growth (Nearest-Neighborhood)

0.56
0.55

0.37
0.43

0.19*(0.09)
0.12(0.12)

By province categoriesa (Kernel)


Ha Tinh province
Thua Thien Hue province
Dak Lak province

0.88
0.40
0.17

0.54
0.28
0.08

0.34*(0.14)
0.12(0.14)
0.09(0.15)

Standard errors (in parentheses) are bootstrapped using 1000 replications of the sample. Estimates are derived by means of the dierence-in-dierences
matching technique based on propensity score matching.
Source: Own calculations based on the DFG Rural Household Survey 2007, 2008 and 2010.
a
These are calculated under all migration.
*
Denotes the statistical signicance at the 10% level.

Table 5. Dierence-in-dierence estimates of the impact of migration on vulnerability to poverty, poverty, and inequality in 2010
Outcome variable

Treatment

Control

Dierence in average outcome ATT

Migrant household
Vulnerability to poverty (VTP)
Head count index (P0)
Poverty gap index (P1)
Poverty severity (P2)

0.21
0.21
0.05
0.02

0.22
0.26
0.07
0.03

0.01(0.02)
0.05*(0.03)
0.02*(0.01)
0.01*(0.00)

Whole sample
Vulnerability to poverty (VTP)
Head count index (P0)
Poverty gap index (P1)
Poverty severity (P2)
Theils L
Theils T

0.29
0.32
0.09
0.04
0.18
0.19

0.30
0.33
0.1
0.04
0.17
0.17

0.01(0.00)
0.01*(0.01)
0.01*(0.00)
0.00*(0.00)
ns
ns

ns, not signicant. Standard errors (in parentheses) are bootstrapped using 1000 replications of the sample. Estimates are derived by means of the
dierence-in-dierences matching technique based on propensity score matching.
Source: Own calculations based on the DFG Rural Household Survey 2007, 2008 and 2010.
*
Denotes the statistical signicance at the 10% level.

88

WORLD DEVELOPMENT

sector as security guard, technician, electrician, plumber, or as


sales person.
As regards the set of household characteristics, the objective
EQI is lower for employed migrants belonging to households
with higher income losses due to shocks in 2010. These migrants are more likely to work in unstable working relations,
without any contracts, and in lower paid jobs not allowing
them to accumulate any savings. The ndings also reveal that
ethnicity does not have any signicant inuence on the objective EQI. This result is also not surprising as 96% of all interviewed migrants belong to the majority ethnic group Kinh or
Hoa (Vietnamese or Chinese).
(c) Eect of migration on rural household welfare
The evidence suggests that (1) migration is partly attributable to household-specic economic factors, and (2) migrants
do not fare equally well in terms of living and working conditions in urban destination areas. Against this background, this
section presents the results on the impact of migration on rural
household welfare from dierence-in-dierence estimations
with propensity score matching. The study also estimated
the impact of migrants employment quality on rural household welfare, but the result is not statistically signicant.
Summarized in Table 4, the results show a large positive and
signicant eect of migration on rural households income
growth at least during the period 20072010. Dependent on
the matching method (Kernel or Nearest-Neighborhood),
households income increased by 2027%. Similarly, migration
for employment also positively and signicantly aects income
growth of original households (Kernel matching method).
Consistent with our expectations, the income growth eect
of migration is particularly pronounced for households from
the Ha Tinh province, while no signicant eects exist for
households from the Thua Thien Hue and Dak Lak provinces.
This result points to the importance of migration as a source
of income growth in structurally weak provinces with poor
employment and job opportunities (cf. UNFPA, 2010 and
GSO, 2011). 7
Table 5 presents the impact of migration on vulnerability to
poverty, poverty, and inequality in 2010. The vulnerability to
poverty and poverty indices are measured depending on the
migration status of a migrant household and for the whole
data sample, while Theils L and Theils T inequality indexes
are only estimated for the whole data sample.
The results show that migration does not make rural households less vulnerable, but it can help a household moving out
of poverty and also reducing the depth and severity of poverty. Similarly, with respect to the whole population, migration also has a positive impact on the poverty status (in
terms of the poverty head count index and poverty gap index)
at the place of origin compared to the case no one migrated.
The ndings are thus consistent with previous studies of
Nguyen et al. (2008, 2009) showing that migration can help
rural households having higher income and lifting people
out of poverty. However, while they show a slight increase
in inequality at the place of origin, we nd some synergy
eects meaning that also non-migrant households in the
villages indirectly benet.
Finally, when comparing the poverty status of migrant
households in 2007 and 2010 in a poverty transition matrix
(see Appendix Table 9), we nd that migration shifted 57 migrant households out of poverty and pushed 17 households
back into poverty. For 38 households, the poverty status did
not change in that period. Similarly, migration also made 38
migrant householdsfrom a total of 68 vulnerable households

in 2007less vulnerable, while only 12 households became


more vulnerable in 2010.
5. CONCLUSIONS
This paper investigated the interaction of shocks, the vulnerability to poverty and welfare of rural households and rural
urban migrants in Vietnam. It provides responses to the three
questions: (1) To what extent do shocks motivate rural household members to move to urban areas?, (2) Are migrants in the
new urban settings better o in terms of working conditions
and quality of life?, and (3) What is the eect of migration
on rural households welfare and vulnerability to poverty?
The analysis is based on (1) a rural household panel data set
from three provinces in Vietnam (Ha Tinh, Thua Thien
Hue, and Dak Lak) and (2) a tracking migrant survey from
the Binh Duong and Dong Nai provinces as well as the Ho
Chi Minh City. To avoid endogeneity problems, a sub-data
set of 1553 households was created by dropping all pre-2008
migrant households and all ruralrural migrant households.
To explore the rst question on the motivation of migration,
three probit models are estimated, namely for: (i) the whole
sample of migrant households, (ii) just for migration for
employment, and for (iii) migration for education. 92% of
all migrant households indicate employment opportunity
and education to be the two main pull factors. At the same
time, the empirical evidence suggests that ruralurban migration for employment is a livelihood support strategy for households coping with agricultural and economic shocks like
droughts, oods or loss of job, or with nancial debts. Migration for education is more likely to be observed among households with a large share of members having higher education
levels and that are nancially better o and do not face any
shocks. They consider migration rather as an investment strategy. Rural households who are engaged in o-farm activities
try to involve their family members in these activities. Migration, then, is less likely to happen, especially with respect to
migration for education. Similarly, the probability of migration for employment decreases for households with large landholdings, or being engaged in agricultural production since the
rural households seem to prefer using them as their own laborers as compared to hiring laborers. In general, the probability
of migration decreases with the employment opportunity in
the village, as evidenced e.g., for the Dak Lak province with
plenty of jobs available in the coee sector, as compared to
the Ha Tinh and Thua Thien Hue provinces. This nding suggests that encouraging rural labor market development can reduce ruralurban migration. Gender does not seem to have
any signicant eect on the migration decision.
With respect to the second research question, the descriptive
and econometric results show that migration coincides with
general improvements in the living and working conditions
of wage-employed migrants in the urban destination area.
Nevertheless, explicit training and wage standards might be
useful instruments for still improving migrants situation in
the urban areas. Migrants being female, better-educated, older
and, with longer migration periods are more likely to report a
higher objective employment quality index. However, households income losses due to shocks may negatively aect a migrants situation in the city. Thus, savings schemes could be a
useful instrument for smoothing income uctuations e.g., from
shocks across all groups of migrants as well as across migrant
households in rural areas.
As regards the third question on the eect of migration on rural households welfare, the results from dierence-in-dierence

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

specications with propensity score matching techniques suggest that migrant households directly benet from migration,
especially migration for employment, through positive
income growth eects. These eects help not only migrant
households moving out of poverty, but they also improve
the poverty situation in rural areas in general. Thus, also nonmigrant households seem to indirectly benet from remittances

89

of migrant households. The econometric results do not nd


any signicant eects of migration in terms of improving
the vulnerability to poverty and inequality. The descriptive
ndings identify a quarter of the households to be vulnerable
in 2007; half of them moved out of poverty in 2010. Further
research should focus on the dynamics of vulnerability to
poverty.

NOTES
1. See Hardeweg and Waibel (2009) for details on the data collection
procedure.
2. Any remaining potential endogeneity problems are considered to be
fairly small due to the following reasons: (i) 50% of all households have
migrants of only up to three years, and only 20% of all households
have members who migrated for a longer time period (>7 years). Thus,
the share of households with migrants prior to 2007 that may have
benetted from remittances is relatively small. (ii) It might take a
transition period of a few years for a migrant to settle down in the new
place of destination so that remittances only arise after such a
transition period. (iii) In our data set, only half of all migrants have
actually sent remittances.
3. Alternative specications control for the correlation of dierent
types of shocks and of wealth variables. For instance, the per capita
income correlates with the type of household activities, so that
model (1) was separately estimated for each of these wealth
variables.

4. With respect to the rural-rural migrants, the equivalent share amounts


to 60%, showing that a higher percentage of migrants with below
secondary school education move to rural areas as compared to ruralurban migrants.
5. With respect to the descriptives, we nd from our household data set
that 75% of all rural-urban migrants are from rural households with
nancial debts. While three quarters of these migrate for nding a job, one
quarter migrates in order to study. But even 45% of these latter migrants
for education send remittances. Thus, some of the indebted households
may consider migration for education also as an investment strategy, even
though they may face nancial restrictions.
6. Being located in the Central Highland region of Vietnam, the Dak Lak
province provides plenty of jobs in the coee sector.
7. The number of out-migrants from the Ha Tinh province is signicantly higher than that from the Thua Thien Hue and Dak Lak provinces
(GSO, 2011).

REFERENCES
Agesa, R. U., & Kim, S. (2001). Rural to urban migration as a household
decision: Evidence from Kenya. Review of Development Economics,
5(1), 6075.
Akay, A., Bargain, O., & Zimmermann, K. F. (2011). Relative concerns of
rural-to-urban migrants in China. Journal of Economic Behavior and
Organization, 81, 421441.
Amare, M., Hohfeld, L., & Waibel, H. (2012). Ruralurban migration and
employment quality: A case study from Thailand. Asian Development
Review, 29(1), 5880.
Azam, J. P., & Gubert, F. (2006). Migrants remittances and the
household in Africa: A review of evidence. Journal of African
Economies, 15(2), 426462.
Chaudhuri, S. (2003). Assessing vulnerability to poverty: Concepts,
empirical methods and illustrative examples. Department of Economics,
Columbia University, <http://info.worldbank.org/etools/docs/library/
97185/Keny_0304/Ke_0304/vulnerability-assessment.pdf> Accessed
May 2011.
Cu, C. L. (2005). Rural to urban migration in Vietnam. In H. T. Ha, & S.
Sakata (Eds.). Impact of socio-economic changes on the livelihoods of
people living in poverty in Vietnam (ASDP 71, pp. 115143). Institute of
Developing Economies, Japan External Trade Organization.
Dang, N. A., Tacoli, C., & Hoang, X. T. (2003). Migration in Vietnam: A
review of information on current trends and patterns, and their policy
implications. Paper presented at the Regional Conference on Migration,
Development and Pro-Poor Policy Choices in Asia, Dhaka, Bangladesh.
De Brauw, A., & Harigaya, T. (2007). Seasonal migration and improving
living standards in Vietnam. American Journal of Agricultural Economics, 89(2), 430447.
Dercon, S. (2002). Income risk, coping strategies and safety nets, Discussion
paper, 22. Helsinki: World Institute for Development Economics
Research, United Nations University (UNU-WIDER).
Ezra, M. (2001). Ecological degradation, rural poverty, and migration in
Ethiopia: A contextual analysis, Working Paper, 149. New York:
Population Council, Policy Research Division.
Fuente, A. D. L. (2010). Remittances and vulnerability to poverty in rural
Mexico. Journal of World Development, 38(6), 828839.

Garip, F. (2010). The impact of migration and remittance on wealth


accumulation and distribution in rural Thailand, Working paper,
0010. Weatherhead Center for International Aairs, Harvard
University.
Giesbert, L. (2007). Seeking opportunities: Migration as an income
diversication strategy of household in Kakamega district in Kenya,
Working paper, 58. German Institute of Global and Area Studies,
<http://www.giga-hamburg.de/dl/download.php?d=/content/publikationen/pdf/wp58_giesbert.pdf> Accessed August 2011.
GSO (2011). Migration and urbanisation in Viet Nam: Patterns, trends and
dierentials. Findings from the 2009 population and housing census.
General Statistics Oce, <http://vietnam.unfpa.org/webdav/site/vietnam/shared/Census%20publications/7_Monograph-Migration-Urbanization.pdf> Accessed April 2012.
GSO & UNFPA. (2005). The 2004 Vietnam migration survey: Internal
migration and related life course events. United Nations Population
Fund and General Statistics Oce, <http://countryoce.unfpa.org/
vietnam/drive/InternalMigration_RelatedLifecourseEvents_GSO1206_
e.pdf> Accessed October 2011.
Hardeweg, B., & Waibel, H. (2009). Collecting data to measure vulnerability to poverty: An overview. Paper presented at the Workshop of
Research Unit 756 of the Deutsche Forschungsgemeinschaft (DFG),
Hue City, Vietnam.
Harris, J., & Todaro, M. (1970). Migration, unemployment and development: A two-sector analysis. American Economic Review, 60(1),
126142.
Haughton, J., & Khandker, S. R. (2009). Handbook on poverty and
inequality. World Bank, <http://issuu.com/world.bank.publications/
docs/9780821376133> Assessed April 2011.
Heckman, J., & Navarro-Lozano, S. (2004). Using matching, instrumental
variables and control functions to estimate economic choice models.
Review of Economics and Statistics, 86(1), 3057.
Klasen, S., Lechtenfeld, T., & Povel, F. (2013). Using household surveys
to capture vulnerability: Issues and challenges. In S. Klasen, & H.
Waibel (Eds.), Vulnerability to poverty (pp. 80114). New York:
Palgrave Macmillan.

90

WORLD DEVELOPMENT

Le, B. D., Tran, G. L., & Nguyen, T. T. P. (2011). Social protection for
ruralurban migrants in Vietnam: Current situation, challenges and
opportunities, Research report, 08. Sussex, UK: Centre for Social
Protection, Institute of Development Studies.
Lee, E. S. (1966). A theory of migration. Demography, 47, 4956.
Massey, D. S. (1990). Social structure, household strategies, and cumulative causation of migration. Population Index, 56(1), 326.
Mincer, J. (1978). Family migration decisions. The Journal of Political
Economy, 86(5), 749773.
Ministry of Finance. (2010). The agriculture insurance pilot program in
Vietnam.
Newman, C., & Wainwright, F. (2011). Income shocks and household riskcoping strategies: Evidence from Rural Vietnam, Discussion paper no.
358. Institute for International Integration Studies, <http://
www.tcd.ie/iiis/documents/discussion/pdfs/iiisdp358.pdf>
Assessed
April 2012.
Nguyen, V. C., Den Berg, M. V., & Lensink, R. (2009). The impact of
working migration and non-working migration on household welfare,
poverty and inequality: New evidence from Vietnam. Asia-Pacic
Development Journal, 16(1), 5992.
Nguyen, N. P., Tran, N. T. M. T., Nguyen, T. N., & Oostendorp, R.
(2008). Determinants and impacts of migration in Vietnam, Working
paper 01. Development and Policies Research Center (DEPOCEN),
<http://depocenwp.org/upload/pubs/TranNgoMinhTam/Determinants
%20and%20Impacts%20of%20Migration%20in%20Vietnam_DEPOCENWP.pdf> Assessed October 2011.
Niimi, Y., Pham, T. H., & Reilly, B. (2009). Determinants of remittances:
Recent evidence using data on internal migrants in Vietnam. Asian
Economic Journal, 23(1), 1939.
Oxfam & Vietnam Academic of Social Science (Oxfam and VASS) (2010).
The Impacts of the global economic crisis on migration patterns in
Vietnam: Findings from rapid assessments in ve provinces and cities,
Research report. Oxfam GB Publisher, <http://policy-practice.oxfam.
org.uk/publications/download?Id=364515anddl=http://oxfamilibrary.
openrepository.com/oxfam/bitstream/10546/112523/1/rr-impacts-globaleconomic-crisis-migration-patterns-vietnam-230210-en.pdf> Assessed
November 2011.
Phung, D. T., & Waibel, H. (2009). Diversication, risk management and
risk coping strategies: Evidence from rural households in three
Provinces in Vietnam: Paper no. 25. In: Proceedings of the German
development economics conference, Frankfurt, Germany.
Pincus, J., & Sender, J. (2008). Quantifying poverty in Vietnam: Who
counts?. Journal of Vietnamese Studies, 3(1), 114150, <http://
www.jstor.org/stable/10.1525/vs.2008.3.1.108> Assessed November,
2012.
Ravenstein, E. G. (1885). The laws of migration. Journal of the Statistical
Society of London, 48(2), 167235.
Ravenstein, E. G. (1889). The laws of migration. Journal of the Royal
Statistical Society, 52(2), 241310.
Rosenbaum, P., & Rubin, R. (1983). The central role of the propensity
score in observational studies for causal eect. Biometrika, 70(1),
4145.

Shah, N. M. (2000). Relative success of male workers in the host country,


Kuwait: Does the channel of migration matter?. International Migration Review, 34(1), 5978.
Sjaastad, L. A. (1962). The costs and returns of human migration. The
Journal of Political Economy, 70(5), 8093.
Smith, J. A., & Todd, P. E. (2005). Does matching overcome LaLondes
critique of on experimental estimators?. Journal of Econometrics, 125,
305353.
Stark, O., & Bloom, D. E. (1985). The new economics of labor migration.
The American Economic Review, 75(2), 173178.
Tongruksawattana, S., Schmidt, E., & Waibel, H. (2010). Shocks and
coping actions of rural households: Empirical evidence from Northeast
Thailand. Paper prepared for the CPRC international conference 2010:
Ten years of war against poverty hosted by the Brooks World Poverty
Institute, University of Manchester, UK.
United Nations Population Fund (UNFPA) (2010). Internal migration:
Opportunities, challenges and social-economic development in Vietnam.
The United Nations Population Fund in Viet Nam, <http://unfpa.org/
webdav/site/vietnam/shared/Migration%20Main%20Paper_ENG_
FINAL.pdf> Accessed August 2011.

APPENDIX

Figure 1. Natural disaster loss and govt budget for disaster prevention.

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

91

Table 6. Summary statistics of variables as included in the probit specication (1)


Variables

Obs*

Mean**

Std. Dev.

Min

Max

Dependent variables in 2008 or 2010


Migrant HH (1-yes, 0-no)
Employment migrant HH (1-yes, 0-no)
Education migrant HH (1-yes, 0-no)

1432
1432
1432

0.19
0.10
0.04

0.39
0.31
0.21

0
0
0

1
1
1

Independent variables in 2007


HH experienced demographic shocks (1-yes, 0-no)
HH experienced social shocks (1-yes, 0-no)
HH experienced agriculture shocks (1-yes, 0-no)
HH experienced economic (1-yes, 0-no)
Female headed HH (1-yes, 0-no)
Age of HH head (years)
Dependency ratio
HH members w/completed secondary school
HH members w/completed professional education
HH members belong to political or social organization (1-yes, 0-no)
Log of monthly HH per capita income (PPP$ in 2005)
Household engaged in o-farm activities (1-yes, 0-no)
Log of land per capita (ha)
Household is indebted (1-yes, 0-no)
Village road condition (1-Good condition, 0-Bad condition)
Log distance from village to district headquarters (km)
Ha Tinh province (1-yes, 0-no)
Thua Thien Hue province (1-yes, 0-no)
Dak Lak province (1-yes, 0-no)

1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432
1432

0.42
0.04
0.25
0.07
0.17
47.23
0.31
0.65
1.76
0.68
4.07
0.50
2.38
0.69
0.48
2.31
0.31
0.35
0.34

0.49
0.21
0.43
0.26
0.37
14.62
0.28
0.92
1.30
0.47
1.02
0.50
1.37
0.46
0.50
0.86
0.46
0.48
0.47

0
0
0
0
0
0
0
0
0
0
1.49
0
8.01
0
0
1.61
0
0
0

1
1
1
1
1
91
1
6
8
1
6.80
1
2.02
1
1
4.32
1
1
1

Source: Own calculations based on the DFG Migrant Survey 2010, DFG Rural Village Survey 2007, and the DFG Rural Household Survey 2007, 2008
and 2010.
*
121 observations had been dropped to control outliners.
**
For binary variables, the mean refers to the share of migrants for which the dummy is equal to 1.

Table 7. Daily income of migrants with wageemployment (percent of total)


Daily wage income (1000 VND)
<50
51100
101150
>150
Stdev (1000 VND)
Median (1000 VND)
Average (1000 VND)

Industry/production sector (N = 106)

Public/private service sector (N = 127)

Total sample (N = 233)

37.7
52.8
7.5
1.9
44.1
66.7
65.1

41.7
33.1
15.7
9.4
56.2
66.7
76.2

39.9
42.1
12.0
6.0
51.3
66.7
71.2

Source: Own calculations based on the DFG Migrant Survey 2010.

92

WORLD DEVELOPMENT

Subjective EQI

30

40

10

20

Percentage of migrants (%)

60
40
20

Percentage of migrants (%)

80

50

100

Objective EQI

-.5

.5
EQI

-.5

1.5

.5
EQI

1.5

10

Percent

20

30

Aggregate EQI

.2

.4

.6

.8

EQI

Source:

Own calculations based on the DFG Migrant Survey 2010.

Figure 2. Migrant distribution of employment quality indices.

Table 8. Summary statistics of variables as used in the EQI models (6)


Variables

Obs*

Mean**

Std. dev.

Min

Max

Dependent variable
Objective EQI

228

0.50

0.33

Migrant characteristics
Years of schooling
Female migrant (1-yes; 0-no)
Debt (1-yes; 0-no)
Age (years)
Job in service sector (1-yes, 0-no)
Length of migration period (Years)
Pay to get a job (1-yes; 0-no)
How to nd a job (1-Introduced from friends or relatives; 0-Others)

228
228
228
228
228
228
228
228

10.72
0.53
0.31
23.98
0.54
3.11
0.82
0.33

3.82
0.50
0.46
5.02
0.50
1.44
0.39
0.47

1
0
0
14
0
1
0
0

19
1
1
47
1
6
1
1

Household characteristics
Ethnicity (1-Kinh, Hoa; 0-Others)
HH members belong to political or social organization (1-yes, 0-no)
Income loss from shocks (2005 $ PPP)

228
228
228

0.96
0.97
207.11

0.18
0.17
543.79

0
0
0

1
1
4510.38

Source: Own calculations based on the DFG Migrant Survey 2010 and DFG Rural Household Survey 2010.
*
Fifteen cases had been dropped from the analysis.
**
For binary variables, the mean refers to the share of migrants for which the dummy is equal to 1.

RURALURBAN MIGRATION, HOUSEHOLD VULNERABILITY, AND WELFARE IN VIETNAM

93

Table 9. Changes in vulnerability to poverty and poverty status in period 20072010


2010

2007

Non-poor
Poor
Total

Non-poor

Poor

Total

156
57
213

17
38
55

173
95
268

Non-vulnerability

Vulnerability

Total

188
38
226

12
30
42

200
68
268

2010

2007

Non-vulnerability
Vulnerability
Total

Source: Own calculations based on the DFG Rural Household Survey 2007, 2008 and 2010.

Available online at www.sciencedirect.com

ScienceDirect

Das könnte Ihnen auch gefallen