DSpace at VNU: Becoming and being urban in Hanoi: Rural-urban migration and relations in Viet Nam

This article was downloaded by: [Georgetown University]
On: 10 August 2013, At: 08:10
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

The Journal of Peasant Studies
Publication details, including instructions for authors and
subscription information:
/>
Becoming and being urban in Hanoi:
Rural-urban migration and relations in
Viet Nam
Nguyen Tuan Anh , Jonathan Rigg , Luong Thi Thu Huong & Dinh
Thi Dieu
Published online: 01 Mar 2012.

To cite this article: Nguyen Tuan Anh , Jonathan Rigg , Luong Thi Thu Huong & Dinh Thi Dieu (2012)
Becoming and being urban in Hanoi: Rural-urban migration and relations in Viet Nam, The Journal
of Peasant Studies, 39:5, 1103-1131, DOI: 10.1080/03066150.2011.652618
To link to this article: />
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or

howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at />

The Journal of Peasant Studies
Vol. 39, No. 5, December 2012, 1103–1131

Becoming and being urban in Hanoi: Rural-urban migration and
relations in Viet Nam

Downloaded by [Georgetown University] at 08:10 10 August 2013

Nguyen Tuan Anh, Jonathan Rigg, Luong Thi Thu Huong and Dinh Thi Dieu

The discourse of the rural-urban migrant is that of a sojourner in the city, a man
or a woman who will almost inevitably return to his or her rural roots and
re-engage with farming and village living. In this paper we ask whether ruralurban migrants can ‘become’ urban and shed their identification as temporary
denizens of the city. We develop a conceptual framework that provides five entry
points to explore this process of becoming urban, and then apply the framework
drawing on the experiences of migrants to Viet Nam’s capital, Hanoi. We argue
that even when migrants do return to their homelands they do so with altered
priorities and on different terms. The experience of migration was not infrequently transformative and life-changing. While migrants may not ‘become’
urban in the fullest sense, their homeland had become a space of familial origin
and emotional identification, not a place where people necessarily sought to
reside, work, raise their children and build their lives.
Keywords: Viet Nam; migration; urbanisation; livelihoods; rural-urban relations

The discourse of the rural-urban migrant
In Asia, the main contributing factor to urbanisation remains internal rural-urban
migration (UN Habitat 2009, 25). One view of rural-urban migrants in Asia is of a
dislocated, disenfranchised and marginalised proletariat, displaced from rural areas to
urban contexts. In more radical interpretations, this is explicitly linked to the disruptive
effects of capitalist transformation. Davis refers to the ‘brutal tectonics of neoliberal
globalization’ and the ‘forcible incorporation into the world market of the great
subsistence peasantries of Asia and Africa’ with an end result of ‘rural ‘‘semiproletarianization’’, [and] the creation of a huge global class of immiserated semipeasants and farm labourers lacking existential security of subsistence’ (Davis 2006, 174,
and see Glassman 2004). Other scholars are more positive about the motivations for,
and the outcomes of migration seeing the process as livelihood enhancing and socially
empowering. As Bird and Deshingkar categorically state in their review of circular
migration in India, ‘there is overwhelming evidence that internal migration can lead to
positive change in both sending and receiving areas’ (2009, 3 [emphasis added]), a view
that Ploeg and Ye Jingzhong (2010) also endorse in their study of China.1

1

‘This circular pattern [of migration in China] has a specific set of characteristics that
positively feed into the process of industrialisation and simultaneously help to avoid major
social and political unrest and the high levels of deprivation typically associated with ‘misery
belts’ elsewhere’ (Ploeg and Ye Jingzhong 2010, 514).
ISSN 0306-6150 print/ISSN 1743-9361 online
Ó 2012 Taylor & Francis
/>


Downloaded by [Georgetown University] at 08:10 10 August 2013

1104

Nguyen Tuan Anh et al.

Unlike the nineteenth century urbanisation process in Western Europe and
North America and currently in much of Africa and Latin America (Ploeg and Ye
Jingzhong 2010), most rural-urban migrants in Asia do not relinquish their attachments – material, emotional and symbolic – to ‘home’ in the countryside. Every
lunar New Year, China experiences what has been called the world’s greatest human
migration (Chang 2008) as scores of millions of urban migrants return to their rural
homes.2 The same occurs in Viet Nam over Tet, in Thailand during the Songkran
holidays, in the Philippines at Christmas, and in Indonesia during Idul Fitri. From
this has emerged a discourse of the Asian migrant that stresses their status as
sojourners in the city. In this schema, rural-urban migrants, who are mainly young
and single and increasingly female, leave home as part of a family strategy to support
the livelihood of the natal household. They engage in work which is usually marginal
and low paid, often informal, and rarely upwardly mobile. Such migrants never
throw off their identity – both self-expressed and externally reified – as ruralites;
farmers on the make. Their stay in the city remains a sojourn, where return is
inevitable, and as such they never fully become members of urban society. Migrant
work is one component of an inter-locking livelihood, where income generation in
the city must be understood in the context of livelihoods ‘back home’. Children,
futures, parents, identities and aspirations remain rooted in the countryside, even
while these migrants live and work in the city.3
Asian rural-urban migrants, therefore, constitute a large and significant portion
of the urban population but they are not urbanites in the classic sense. Such ruralurban migrants occupy what has been conceptualised as an interstitial – or ‘inbetween’ (see Resurreccion 2005, Agergaard et al. 2011) – social and economic space.
They are denizens of the urban, but not its citizens; they live in the city, but belong to
the countryside They work in the industrial and service sectors, but at core remain
peasants (farmers); they are out of place, therefore, in multiple ways. For governments, the rural-urban migrant represents a problem of management in the context
of rapidly expanding urban populations and creaking city services; for scholars, they
are a hard-to-pin-down population who are difficult to categorise and a challenge to
research given their mobility; and for NGOs, migrants are the flotsam of capitalism
who need to be protected and supported.

And yet it is also clear that not all migrants live on the edge of subsistence,
scraping a living at the margins of the Asian miracle. Some build productive and
rewarding lives in the city. They may arrive as refugees or speculative sojourners
from the countryside, and yet make a relative success of their stay, so much so that
some, over time, make the transition from rural migrants, to city sojourners, to
urban citizens.4 It is this process of ‘becoming urban’ that the following paper seeks
to illuminate. How and why do some migrants manage to make this transition, and
what prevents – or inhibits – others from doing so? And how can we measure and
2
China has some 130 million rural-urban migrant workers: ‘Together they represent the largest
migration in human history; three times the number of people who emigrated to America from
Europe over a century’ (Chang 2008, 12). Including rural-rural migrants, the total number on
the move at the end of 2008 numbered 225 million (Ploeg and Ye Jingzhong 2010, 515).
3
For various studies and reports that reflect this discourse, see UNDP (2009, 71–2), GSO
(2006, 2), Anh Dang et al. (1997, 313), Martin (2009, 10), Cai Fang et al. (2009, 8–10), We Ha
et al. (2009, 4), and Tirtosudarm (2009, 29).
4
For a parallel example considering the case of Vietnamese sojourners in Laos and Thailand,
see Hardy 2005.


The Journal of Peasant Studies

1105

Downloaded by [Georgetown University] at 08:10 10 August 2013

assess this progressive process of urban insinuation? As it has become common to
observe, the world recently reached ‘an invisible but momentous milestone’: for the

first time in human history, the greater portion of the globe’s population live in cities
and towns, and the world is now more urban than it is rural (UNFPA 2007, and see
Satterthwaite 2006). However, as we briefly outline below, the contours of the
transformation are not as clearly delineated as this striking and resonant claim
suggests. In terms of numbers, living conditions, livelihood pathways and futures,
for example, there are gaps in our knowledge of the migrants who make up such a
significant part of the burgeoning but often highly unstable urban population.
Becoming urban: establishing a heterodox conceptual framework
One of the key reasons why the contours of the urban transition in Asia are indistinct
is because it is often unclear quite when and how rural-urban migrants become
urban, and when they should be counted as urbanites, rather than as ruralites. It is
not unusual for Asian households to have representation in both urban and rural
spaces and associated economies, and they embrace both urban and rural sensibilities. The aggregate data, as noted above, may identify an epochal shift in the
balance of the rural and urban populations, yet very significant numbers do not fit
into such neat, categorical boxes. A question addressed towards the end of this
section as well in the conclusion of the paper is whether this large population of
‘neither-rural-nor-urban’ should be counted as occupying a third, ‘in-between’
category.
In this paper, we adopt a heterodox position in terms of how we view the process
of becoming urban and, by association what we understand by ‘urban-ness’ and,
therefore, ‘rural-ness’. We view the process as consisting of multiple, often overlapping, tendencies which we group here into five categories or entry points. Each
provides a different perspective to view – or approach – the integration of rural
migrants into the urban fabric, and each draws on scholarship in different fields, as
Table 1 sets out (Table 1). At a very general level, the legalistic approach links with
geographical literatures; livelihoods, to economic and development studies literatures; identity and behaviours to anthropology; and networks and associations to
sociology. As will become clear, none offers a neat solution or answer to the
challenge and to the question of how and when a migrant becomes urban. This is
to be expected because, as the paper will illustrate, ‘becoming’ urban is multiple,
contingent and – and importantly – reversible.
Perhaps the simplest, and certainly the neatest, way to answer the question,

‘when does a rural migrant become urban?’ is to take a legalistic or bureaucratic
approach. A rural migrant becomes an urbanite when s/he is officially registered, or
counted, as such. This takes the transition from rural to urban as simple and
complete, achieved through the tick of a bureaucratic box. Second, it is possible to
take a livelihoods approach in which a rural migrant (and household) becomes urban
when he or she builds a livelihood which is predominantly urban in provenance
(location), character and weighting. So rural migrants might be counted as urbanites
when their livelihoods are predominantly based in an urban or peri-urban location,
and involve livelihood activities that can be regarded as emblematic or constitutive
of urban living. A third way to approach the question is in terms of identity, both
self-expressed and/or externally reified. Does a migrant in an urban context consider
him or herself rural or urban, sometimes expressed in terms of ‘home’, and how is


1106

Nguyen Tuan Anh et al.

Table 1.

Becoming urban: a conceptual framework.

Views of ‘becoming’
urban
Legalistic

Downloaded by [Georgetown University] at 08:10 10 August 2013

Livelihoods

Identity and
identification
Social and
cultural behaviours
Social networks
and associations

Illustrative
Counted as urban
bureaucratically; household
or individual registration
records them as urban in
terms of residence; census
and population records treat
them as urban
Urban livelihoods in terms of
provenance (location in an
urban context) or character
(associated with the urban)
Migrant self-identifies as an
urban, city dweller and it
treated as such by other city
dweller
Norms of behaviour, attitude,
dress, consumption patterns
and social networks
emblematic of urban life
Social networks and
association, and more
generally social capital,

is urban in location and
distribution

Supporting literatures
Wu and Treiman 2004;
Wong and Wai-Po 1998;
Wong and Rigg 2010;
Deshingkar 2006

de Haan 1999; Rakodi and
Lloyd-Jones 2002
Silvey and Lawson 1999;
Andersson 2001
Thompson 2004, 2007

Andersson 2001; Curran and
Rivero-Fuentes 2003

that migrant treated by the rest of the urban population: as someone ‘out of place’,
or as a fellow urbanite? Fourth, it is possible to focus on the behaviours, views,
consumption patterns, preferences and everyday practices of migrants to determine
whether they remain rural in their living patterns and practices, or have become
urban. Finally, we can map the social networks and associations – the configurations
of social capital – that migrants deploy and ‘use’ to give meaning to their lives.
Where is the centre of gravity of such networks, and who or what constitutes these
associative relations?
It is the first two of these approaches (the legalistic/bureaucratic and the livelihoods
approaches) to the question of becoming urban which have the most purchase in
policy terms for the simple reason that they are the most easily measured and tracked:
Where are you registered as living? What do you do? And yet, as we expand in a

moment, it is these two approaches which are arguably the most problematic because
they do not adequately deal with the mixed, partial and contingent processes by which
migrants and their wider families become, over time, insinuated into the urban fabric.
It is through unravelling urban/city and rural/village identities, behaviours,
consumption practices and the geographies of their social networks that we come to
better understand migration as a social process. In the remainder of this section, we
focus particularly on the shortcomings of the first two approaches in Table 1, while
touching more briefly on the limitations of the others.
The legalistic or bureaucratic approach suffers from the fact that there are many
tens of millions of people living in urban areas of Asia who are still officially ‘rural’.
If there is one lesson to be drawn from the experiences of countries across Asia, it is


Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1107

that administrations have failed to keep up with the increasingly mobile nature of
living and livelihoods. From Mumbai to Ho Chi Minh City and Bangkok, official
data capture only a small portion of what is going on as bureaucrats struggle with
categories that are not fit for purpose and with populations who are far more agile
than they are. National censuses often do not pick up short-term movements and
population registers are either lacking or provide only rudimentary coverage: ‘For
the most part, migration data remain patchy, non-comparable and difficult to access’
(UNDP 2009, 28, and see Newland 2009, 10). In an assessment of the situation for
Asia, Kundu writes that ‘studies on internal migration are seriously constrained by
the fact that no international organisation systematically collects or tabulates even
the basic demographic information. . .in a cross sectionally and temporally comparable manner’, something he consider a ‘tragedy’ given the sheer number of

migrants and their importance in policy terms (Kundu 2009, 14). As we discuss
below, the issue of getting to grips, in simple numerical terms, with the scale of ruralurban migration in Asia is even more pressing in countries with household
registration systems, notably China and Viet Nam (World Bank 2003, 22–23, Wu
and Treiman 2004). In summary, when people are mobile, their residence ill-defined
and temporary, censuses inadequate, social structures fluid, and livelihoods are
ranged across space, there is a real empirical challenge of knowing the scale and
nature of the ‘problem’ (as it is usually framed), and taking a legalistic approach has
considerable shortcomings as a result.5
Taking a livelihoods approach to becoming urban also suffers for a range of
equally trenchant reasons: because there are many people in urban areas pursuing
agricultural pursuits (in the form of various forms of urban agriculture); there are an
increasing number of factories (‘urban’ livelihoods) in rural areas;6 and it is not
unusual for migrants to embrace livelihood complexes that are both rural and urban
in their provenance and character. The desire to put people into occupational boxes
(‘worker’, ‘farmer’), not least for ease of categorisation, overlooks the mixed and
fluid nature of livelihoods in much of the Global South.
When it comes to considering the process of becoming urban in terms of
identities, migrants show an ability simultaneously to embrace multiple identities.
Treatment by others is also often ambiguous. The same is true of urban ‘behaviours’,
where research has clearly revealed that ‘urban’ behaviours, norms and attitudes are
not tied to urban spaces, but have colonised the countryside as some rural villages
have become socially urban (see Thompson 2004, 2007). Finally, mapping networks
of association and their configuration does not provide clarity in terms of the relative
importance of such associations in functional terms on the one hand, and their
cultural and social significance on the other.
In briefly setting out the shortcomings attached to all these approaches to
understanding the process of becoming urban, the intention is not to conclude that the
task is an impossible one. Rather, it emphasises that the development process,
embodying as it does a number of inter-locking social, economic, political and
geographical processes, is never ‘complete’, and is rarely neat. These five lenses offer

5

‘Despite our ability to establish these broad contours of movement, what we know is dwarfed
by what we don’t know. Unfortunately, migration data remain weak. It is much easier for
policy makers to count the international movements of shoes and cell-phones than of nurses
and construction workers’ (UNDP 2009, 28).
6
This is one of the particular characteristics of the Asian urbanisation/industrialisation
experience (see Rigg 2001).


1108

Nguyen Tuan Anh et al.

complementary insights into the multiple and progressive ways in which rural migrants
insinuate their way into the urban fabric as they make the transition from ruralites to
urbanites, from denizens to citizens of the city, from sojourners to residents. The process
is probably never fully complete for the migrant – perhaps only for their children – but
problematising the discourse of the rural-urban migrant in this way permits us to open
up an avenue for analysis that pays attention to the process of becoming. In particular, it
helps to get away from the dichotomous (rural-urban) or trichotomous (ruralinbetween-urban) approach to understanding migrants and migration.

Downloaded by [Georgetown University] at 08:10 10 August 2013

Urbanisation, migration and registration in Viet Nam
From 1975 through to the early 1990s, the rate of urbanisation in Viet Nam was low,
and rural-urban migration stunted (Table 2). According to official data, there was no
increase in levels of urbanisation between reunification in 1976 and 1990, with the
figure remaining close to 20 per cent throughout the period. As in China, this was

because a strictly enforced household registration system (ho khau)and its coupling
with access to social goods from education to health and food security strongly
discouraged people from leaving their place of registration (see below).7 With the
introduction of doi moi (‘renovation’, economic reform) in Viet Nam in 1986, the
elemental link between an individual’s ho khaui and their access to the state subsidy
system has been progressively eroded, rural-urban migration has accelerated, and
rates of urbanisation have increased. The Viet Nam Household Living Standards
Survey (VHLSS) shows that between 1993 and 1998 seasonal migration increased
six-fold (de Brauw and Harigaya 2007, 434), a large proportion of this human tide
flowing to Ho Chi Minh City and Hanoi (Figure 1).
Table 2.

Urban population of Viet Nam (1976-2020).

Year

% urban

1976
1977
1978
1979
1980
1981
1982
1984
1985
1986
1987
1988

1989
1990
1991
1999
2005
2009
2020

20.6
20.1
19.7
19.5
19.1
18.6
19.1
19.1
19.2
19.3
19.7
19.9
20.1
20.8
21.6
24
26.4
29.6
35–45

Sources: Anh Dang et al. (1997, 317), UNICEF (2010) Coxhead et al. (2010).

7

For a comparative discussion of China and Viet Nam, see McGee (2009).


The Journal of Peasant Studies

1109

Downloaded by [Georgetown University] at 08:10 10 August 2013

Over the period since reunification in 1975, it is possible to identify three broad
periods in Viet Nam’s urbanisation history: 1975–1986; 1986–1993; and 1993present. From 1975 to 1986, urban population growth was largely through natural
increase. To a significant extent, mobility was controlled by the ho khau system and
when it did occur, was usually officially sanctioned and mostly rural-rural rather
than rural-urban. As a result levels of urbanisation remained largely unchanged. In
1986, with the introduction of doi moi the opportunities for movement began to
grow, although it was not until the early 1990s, and in particular the promulgation of
the 1993 Land Law, that the institutional barriers to migration had eased sufficiently
to see a marked increase in rural-urban migration. This ushered in the third period of
urban growth, from 1993 to the present day during which there has occurred a
marked increase in urbanisation and the relative size of the urban population, driven
by rural-urban migration.
The household registration system in Viet Nam
Until the late 1980s, Viet Nam’s household registration system largely ‘fixed’ people in
space. In the north of the country this was instituted over the course of the 1950s, and
was then extended to the south with reunification and the creation of the Socialist
Republic of Viet Nam in 1975. The household registration system served several
purposes. It was a powerful tool of surveillance and political control; it assisted in the
management of resources under a socialist system where residence in an area secured

access to services; and for the central state, the household registration system was an
important means by which the state could ‘plan’. As Hardy (2001, 192) writes,
Even when one died, the ho khau was still of importance. Unregistered residents were
not entitled to commune land for burial. Before doi moi [reform], the link between
identification and access to rights and services was all embracing. . . . To live without a
ho khau was to live without the rights granted to Vietnamese citizens under the law. And
the ho khau. . .was intimately tied to place of residence. Rights were granted in the place
of registered residence, and in that place alone.

To be without a ho khau was not to be entertained lightly. It alone guaranteed
access to the services of the state; quite literally, without a ho khaui one could not eat
(see Hardy 2001, 194). In time, the ho khau system was codified into four registration
categories (GSO 2006, and see Leaf 1999, 305):8
– KT1: a person registered in the district where he/she resides
– KT2: a person not registered in the district where he/she resides, but registered in
another district of the same province/city9
– KT3: a person from another province/city who has temporary registration in their
place of destination for a period of one year, after which the KT3 registration has
to be re-issued. (Since July 2007 the requirement to re-register has been lifted.)
8
In reviewing documents of the city authorities, statistical agencies and the Hanoi Communist
Party it is evident that while definitions of KT1 and KT2 are fairly clearly established and
agreed, those for KT3 and KT4 vary considerably. It is not particularly surprising, therefore,
that interpretations of the latter two categories also differ.
9
There are two forms of KT2 registration: KT2 ‘arrived’ (or KT2 de^n) and KT2 ‘left’ (or KT2
di),
 the latter held by the authorities in the migrant’s place of departure and the former in their
place of arrival.

Downloaded by [Georgetown University] at 08:10 10 August 2013

1110

Figure 1.

Nguyen Tuan Anh et al.

Northern Viet Nam.

– KT4: a person from another province/city who has temporary registration in their
place of destination for a period of six months, after which the KT4 registration
has to be re-issued. (Since July 2007 the requirement to re-register has been lifted.)
There have been important modifications to the ho khau registration system
over time, in particular its loosening, reflecting the reality of everyday lives and
living in Viet Nam during the reform era.10 Since July 2007, a migrant with KT3 or
KT4 registration can, after one year of living stably in their new location, request

10

The relevant laws, decrees and circulars are: Decree No. 51-CP issued by the Government on
10 May 1997 on Registration and household management; the Law on Residence issued on 29
November 2006; Decree 107/2007/N D-CP issued on 25 June 2007 on guiding the
implementation of the Law on Residence; Decree 56/2010/N D-CP issued on 24 May 2010
on Amendments and Supplements to a number of articles of Decree No 107/2007/ND-CP;
Circular No.52/2010/TT-BCB on 30 November 2010 issued by Ministry of Public Security on
Detailed provisions for the implementation of some articles of the Law on Residence; Decree
No. 107/2007/ND-CP dated 25/06/2007; and Decree No. 56/2010/ND-CP dated 24/5/2010.

Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1111

re-registration to KT1 with the important proviso that the owner of the house where
they have been living supports the application.
Temporary migrants in Hanoi should be classified as KT2, KT3 or KT4, with the
majority in the latter two classes. However, partly because the ho khau system
remains in place at the same time as its logic has been undermined by the economic
reforms of doi moi, there is very good reason to believe that there is a widening gap
between official data on residence and de facto residence, with a substantial and
growing ‘floating’ population (see Dapice et al. 2010, GSO 2006). As the UNDP
observed in 2010, ‘there is a significant gap in data on internal migration [in Viet
Nam], which carries widespread implications for understanding and measuring the
parallel processes of migration and development, as well as for exploring how
migration can be used to enhance Viet Nam’s socio-economic development’ (UNDP
2010, 5). An indication of the size of the under-reporting of migration flows is
evident from the data on population on the one hand and number of workers
employed in enterprises in Ho Chi Minh City on the other: the former grew by 7.5%
between 2002 and 2005; the latter by 39% (Dapice et al. 2010, 3). There is every
reason to suppose that a similar discrepancy applies to Hanoi, and Viet Nam’s other
larger urban centres such as Danang. Across the country, it has been suggested that
this floating population numbers between 12 and 16 million which, if broadly
correct, represents between 13% and 18% of Viet Nam’s population (UNDP
2010, 5).
While the formerly tight link between registration status and well-being (through
access to state services) has been cut, migrants with one of the three temporary

residency classes (KT2, KT3 and KT4) nonetheless are limited in their access to
medical, educational and other social services as well civil rights including the right
to vote. Those with KT4 residency can only legally stay in a location for six months
(extendable) and, until 2010, could not own land titles.11 At the same time, it is
widely reported that re-registering (changing residency status) is time-consuming and
burdensome, although unequally so across the country (UNDP 2010, 7). In light of
this, it is not surprising that many migrants do not register their arrival in a new
place (or their boarding house landlords do not do it for them) and, if they do, then
they do not re-register after six months. A new Law on Residence was introduced in
2007 which, on paper, has loosened some of these restrictions but there is evidence
that it has been unevenly adopted, leading to a degree of confusion among migrants
as to their rights (UNDP 2010, 8).12
Methods and research context
The discussion that follows is based on fieldwork undertaken in Hanoi, Viet Nam’s
capital, between September and December 2010, drawing on detailed interviews with
30 migrants across the city.13 As outlined above in general terms, because of
uncertainties about the data on internal mobility not only in Viet Nam but across

11

This has since been relaxed.
Our migrant interviewees were similarly confused: they were unclear of their rights and
responsibilities under the revised registration legislation.
13
While Hanoi is Viet Nam’s capital city it is second in terms of population and economic
weight to Ho Chi Minh City in the south.
12


Downloaded by [Georgetown University] at 08:10 10 August 2013

1112

Nguyen Tuan Anh et al.

Asia, many scholars have chosen to work with primary data, using insights gained
from micro-level studies to reflect on the wider picture (Kundu 2009, 14). This,
clearly, has its own deficiencies, not least the challenge of scaling up and questions
over the ‘representativeness’ of insights gained from such research. Nonetheless,
large scale surveys – such as, in the Viet Nam case, the VHLSS – do not shed much
light on the real extent of migration to major urban areas or, and more importantly,
the nature of migrant lives in the city.
Following discussions with local scholars and pilot interviews among migrants in
different parts of the city, we arrived at a sampling framework with four key criteria.
To begin with, all the interviewees were to be from rural contexts, with family
livelihood backgrounds in farming or farming-related activities. Second, rather than
focusing our attention on one district or ward, we took the decision to distribute our
interviews across the city, covering the districts – though unequally – of Cau Giay,
Hai Ba Trung, Dong Da, Ha Dong, Hoan Kiem and Thanh Xuan (Figure 2). We
took this decision because each district has developed over different periods,
attracting different migrants at different times, and offering different employment
opportunities. We decided to take their ‘location’ in the city as their place of
residence, rather than their place of work or employment, although more often than
not the two effectively coincided (Figure 2).
Third, we wanted a rough balance between male and female interviewees
because of the different ways that men and women connect, through the life cycle,
with their natal communities.14 And finally, we took the decision to select migrants
on the basis of their year of first arrival, planning to select around ten interviewees
from each of three periods (Table 3). The first period (1986–1992) spans the start of
doi moi (reform) in 1986 (and the relaxation, around 1988, on restrictions on travel

and control of spatial mobility through the ho khau) and ends just before the
promulgation of the new Land Law in 1993, which gave farmers the ability to
transfer, exchange, lease and mortgage their land, thus permitting greater
livelihood flexibility and opening up greater scope for migration (Le Bach Duong
and Khuat Thu Hong 2008). It was following the introduction of the 1993 Land
Law that rates of migration and urbanisation accelerated – although data tracking
these trends, as outlined above, are problematic. The second period runs from the
introduction of the Land Law in 1993 through to 1999, the early reform years
during which Viet Nam was one of Asia’s poorest countries and the poverty rate
even at the end of this period was 37 percent for the country as a whole, and 46
percent in rural areas (Figure 3). The final period spans the years from 2000 during
which the economy expanded at such a rate that the country is now close to lower
middle-income status (VNN 2010), with the poverty rate falling to around 15
percent by 2006, one quarter of the figure in 1993. The decision to select our
interviewees in this manner was partly because we wanted people of varying ages
who had migrated to Hanoi during different development ‘eras’. We hypothesised
that length of time in the city would have a bearing on the depth of their connection
to urban living and life, and we also suspected that different development forces
would be acting on their initial decision to migrate, given changing rural conditions

14

In several of the cases we were, in effect, interviewing a migrant couple so although our first
point of contact might have been a man or a woman, the resulting interview included
discussions with both husband and wife, and sometimes their children too.


Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

Figure 2.

1113

Hanoi and respondents’ places of occupation and residence.

and urban opportunities. That said, and as the discussion which follows shows,
these assumptions often break down and generational factors can be more
important than length of time in the city in shaping migrants’ engagement with the
urban context.
The actual selection of informants was based on a snowball sampling strategy,
initially drawing on the links of the Vietnamese authors of this paper (including their
students living in areas of the city with concentrations of migrants), and then using
the contact networks of the interviewees to identify possible further informants,
while at the same time meeting our sampling strategy. As Figure 2 shows, in most
cases the places of work and residence of the migrants were co-located; in those
instances when this was not the case, we conducted the interview at the time and
place most convenient for the respondent. The fact that all our migrants had mobile
phones made arranging interviews much easier than would have been the case even
five years ago. For the interviews themselves we had a checklist of topics to cover


1114

Nguyen Tuan Anh et al.

Table 3.

Interview sampling frame.

Date of first arrival in Hanoi
1986–1992

1993–1999

2000–2007

Total

5
6
11

5
5
10

3
6
9

13
17
30

Male
Female
Total

1986–1992

Downloaded by [Georgetown University] at 08:10 10 August 2013

District of residence
Cau Giay
Thanh Xuan
Dong Da
Hai Ba Trung
Hoan Kiem
Ha Dong
Thanh Tri
Tu Liem
Total

1993–1999

2000–2007

Male

Female

Male

Female

Male

Female

Total

2
–
1
2
–
–
–
–
5

1
–
1
1
1
1
1
–
6

2
2
–
–
–
–
–
–

4

1
3
–
1
–
–
–
1
6

–
2
1
–
–
–
–
–
3

4
1
1
–
–
–
–
–

6

10
8
4
4
1
1
1
1
30

although, as is often the case, discussions not infrequently veered onto new and,
sometimes, unexpected ground.15 We opened each interview explaining what the
research was about and then gathered information of a contextual nature: age, place
of birth, date of arrival in Hanoi, conditions in source settlements, education, family
and marital status, registration, and so forth. We spent considerable time detailing
the migration and livelihood histories of each respondent, and then turned to
exploring their links with ‘home’ – remittances, and returns to and the nature of their
engagement with their place of origin. This was followed by an often equally long
discussion of respondents’ engagement with the city, before turning to questions
concerning their plans for the future. We concluded by thanking each respondent,
asking if we could return to follow up questions if need be, either by phone or in
person.
The field research took as its starting point, ‘the migrant’. We then scaled up
from the migrant to their wider family/household and, from here, sometimes to
their village of origin or urban social milieu. In taking this approach we deviated
from much livelihoods research which takes ‘the household’ as the starting point
and then disaggregates, where necessary, to the individual. We took this approach
because we were interested in how migrants sometimes re-shape their engagement

with other social units (‘communities’, ‘families’) over time, due to their migration
experiences.
Following each interview, a summary of the issues and the discussion was
compiled. The interviews themselves were taped, and then transcribed, translated
and cross-checked against our summary statements. We then undertook a second
round of interviews, both face-to-face and by telephone, to follow up on particular
15

For example, how opportunities for women’s employment was leading to a modest
individualisation of work and reward, and a re-working of the functioning of the patriline and
women’s role and place in the patriline.


Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1115

Figure 3. Viet Nam poverty rates, 1993–2006.
Source: JDR (2007, 4).

issues, to confirm areas where we had further questions or doubts, and in some
instances to interview other members of migrant families. Finally, we also undertook
two excursions to the countryside, to Thanh Hoa and Nam Dinh, to visit the areas,
settlements and natal households from which selected migrants originated (see
Figure 1).
Becoming urban
We now turn to applying the conceptual framework outlined above through the
experiences of our interviewees. We are interested not only in understanding how

migrants may progressively become urban, but also those forces, barriers and
predilections that might prevent them from doing so.
Applying a legalistic/bureaucratic approach to becoming urban
Among our interviewees, the balance of registrations showed a small majority of
migrants falling into the KT3 and KT4 categories (Table 4). There were also two
migrants with no registration whatsoever. What is striking is that these migrants,
with temporary residency registration status in Hanoi, had on average been living
and working in the city for over 13 years.16 Based on our interviews, residency status
provides little explanatory traction when it comes to delineating migrants’
integration into urban space.
While registration does not have the legal and material gravitas that it once did, it
nonetheless bestows certain benefits and, as we will see, also has certain problematic
elements. The main benefit for our interviewees, at least among those who had
brought families with them or were intending to do so, was that it provided greater
16

There are some quite specific categories of migrant who fall by the wayside: university
students, for example, whose university registration expires, who stay in town but can’t
transfer this to another city-based unit and, at the same time, are no longer registered in their
home village. Marriage migration is another instance where migrants may become de facto
unregistered. (Personal communication: Andrew Hardy.)


1116

Nguyen Tuan Anh et al.

Table 4.

Residency classification of interviewees.

Residency
classification

Number of
interviewees

% of
sample

Average length of
time in Hanoi (years)

6
3
7
9
2
3
30

20
10
23
30
7
10
100

20

20
14
13
9
6
14

KT1
KT2
KT3
KT4
No registration
Unknown
Total

Downloaded by [Georgetown University] at 08:10 10 August 2013

Source: migrant interviews, 2010.

security in terms of access to education and health. While it was possible for
migrants with KT3 and KT4 registration to have their children educated within the
state system in Hanoi, a number of migrants remarked that they were prioritised
last with the risk that their children would be denied access to the school of their
choice. Mrs Hong (interview #022, 30.9.2010), who arrived in Hanoi from Nam
Dinh around 100 km south of Hanoi in 1988 and was running a tea stall in Ha
Dong district at the time of the interview while her husband was a xe om
(motorcycle taxi) driver, told us that she was willing to relinquish her rights to land
in her home village in order to buy land in Hanoi and obtain KT1 residency in the
city. Mrs Hong and her husband traded security in terms of land in their natal
village for the future of their children, gaining for them a superior education in

Hanoi and, they hoped, better employment prospects and long-term prospects.17 In
this regard they would seem to have been successful: their 26 year-old son had just
completed an IT degree at the National University of Ho Chi Minh City while
their daughter, who was a year younger, had obtained a high status job in a
fashion company. Schools in Hanoi are superior to those in the countryside, and
education in such a school significantly raises the chances of passing the national
university entrance examination, thereby gaining access to one of the country’s
more prestigious universities. This, in turn, raises the chances of accessing a good
job on graduation which, in the context of Viet Nam’s limited social security
system, helps to secure parents’ livelihoods in old age.
A second negative characteristic of a temporary registration (KT3 or KT4) is that
the holder could not, until very recently, buy land or a house.18 The attractions of
owning land in Hanoi are significant, not least because of its investment value (see
Table 5) and the attractions of having a foothold in the capital. This was something
that Mrs Hong also emphasised when we re-interviewed her with her daughter,
Tham (interview #022b, 10.11.2010). If only she could buy a house in Hanoi, she told
us, ‘everything then will be OK’ because it would secure the futures of her children.
Having a house in a rural area, she went on to explain, does not help: it is only of use
17

Thuoc, a migrant from Ha Tinh, had done much the same. He had relinquished his right to
land in his homeland to secure residency status in Hanoi. Thuoc and his wife did this ‘in order
to make it convenient for our [only] child to attend school’ (interview #025, 7.10.2010).
18
Migrants were formerly caught in something of a Catch 22: without a house or land in the
city migrants could not register as permanent residents; and without permanent residency,
they were barred from buying land and having a ‘Red Book’.


Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1117

when one is old and thinking of retiring. Mr and Mrs Hang (interview #009,
21.9.2010), an electrician and cleaner respectively, borrowed 75 million VND19 from
an assortment of friends and relatives to buy a 50m2 plot of land in Hanoi in 1998
which they were struggling to pay off – but which they nonetheless viewed as an
excellent investment.20 The challenge even for middle-income migrants is that the
cost of land in Hanoi has increased much more quickly than have wages (Table 5).
Quite a number of our respondents who had arrived in Hanoi ten or more years
previously lamented their foolishness in not buying land when they had the chance.
Mrs Hop (interview #016, 24.9.2010), who first arrived in Hanoi in 1998 but still
squats on land from which she and her family were at imminent risk of being evicted
at the time of the interview, told us: ‘I have lived her for ten years, and can’t buy a
single square metre’.
In contrast to a number of reports on Viet Nam’s registration system (e.g. UNDP
2010), our interviewees did not notably remark that re-registration was particularly
troublesome or expensive. In the main it was that few saw any great advantage
coming from re-registering. They either did not have children in the city or managed
to get them into city schools even without permanent residence, and for most the
prospect of buying land was a distant dream in any case. There was, however, also a
rather surprising risk articulated by a number of the interviewees; namely, the risk
that a migrant might lose access to his or her land in their ‘home’ village should they
become, officially, too urban.21 It seems that rural districts and provinces sometimes
treated this issue rather differently. Some of our respondents saw no likelihood that
their land would be re-allocated; others most certainly did, and kept their temporary
status in Hanoi so that their land in the countryside would not be taken away from
them. We return to this issue in the context of livelihoods in the next section.

Table 5.

Hanoi’s land prices, 1989–2010.

Date

Cost (per metre2)

Location

1989
1998
1999
2002
2007
2010
2010
2010
2010

1 million VND
1.5 million VND
1.4 million VND
10 million VND
10 million VND
60 million VND
100 million VND
120 million VND
200 million VND

Hoan Kiem
Cau Giay
Thanh Xuan
Cau Giay
Tu Liem (peri-urban Hanoi)
Hoan Kiem (but in a poor area with ‘drug addicts’)
Cau Giay (away from main road)
Cau Giay
Cau Giay (close to main road)

Source: migrant interviews, 2010.

19

VND¼Viet Nam dong. At the time of the research, US$1¼20,000 VND; £1¼30,000 VND.
Because Mr and Mrs Hang (interview #009, 21.9.2010) do not have permanent residency in
Hanoi they had to buy the land in the name of Mr Hang’s brother to circumvent the legal
requirement to have residency.
21
In a similar vein, the French archives reveal migrants in the colonial era begging to be
allowed to pay their poll tax in the home village, as it was that payment that ensured their
share of village land and their continued connection to the village. Personal communication:
Andrew Hardy.
20


Downloaded by [Georgetown University] at 08:10 10 August 2013

1118

Nguyen Tuan Anh et al.

It has been suggested (e.g. UNDP 2010) that there are large number of ‘KT0’
migrants in the larger cities of Viet Nam – migrants without any registration
whatsoever, living beneath the administrative radar. We did not find this, although it
should be noted that our approach was not designed specifically to pick up on the
issue and our sample was a small one. We were generally impressed by the level of
knowledge that ward leaders had of their areas of jurisdiction and those residing
there. Our sample did, however, support the contention that in Viet Nam’s larger
cities there is a large ‘floating’ population of migrants who have lived in the city for
many years but remain classified as temporary residents (Coxhead et al. 2010). There
is the possibility that intentional (under)reporting errors are used to maintain a basis
of power among local officials, and that these are multiplied many times and then
transmitted upwards to be reflected in aggregated statistics that bear little
resemblance to the number of temporary registered migrants (personal communication, Michael Leaf).
In terms of this paper, and overall, what the discussion so far suggests is that
taking a legalistic approach to becoming urban, at least in Viet Nam, is problematic
because of the way that the household registration system creates a context where
registration status does not map onto the reality of a migrant’s engagement with the
city. Even though our 30 respondents had lived in Hanoi for an average of 14 years,
the large majority still held temporary registration status. In other countries,
registration may not be quite so formulaic as it is in Viet Nam, but the wider point
that governments are poorly equipped to deal with the growing mobility of people
has wider resonance. In a review of the situation in India, Bird and Deshingkar write:
‘Negative government attitudes combined with ignorance created by inadequate data
sets has led to the widespread neglect of migration as an important force in economic
development’ (Bird and Deshingkar 2009, 6). There is also the important point that
while our interviewees in general saw no particular advantage from re-registering –
so that bureaucratic hurdles were ignored or avoided rather than traversed – the

absence of correct registration can become a problem should migrants have, for
example, to deal with the police or the courts.
Urban livelihoods and rural security
[From the] vast array of facts and figures . . . [there are] certain principles or laws which
appear to me to guide all migratory movements . . . none of these currents of migration
can compare in volume with that which arises from the desire inherent in most men to
‘better’ themselves in material respects. (Ravenstein 1889, 286)

While the majority of our interviewees may still have held temporary residency
status in the city (or held no registration), and therefore could be counted rural in these
narrow, bureaucratic terms, for the large majority their livelihoods were firmly urban
in provenance and character. While the literature on migration in Asia may stress the
‘inter-locking’ nature of rural and urban livelihoods, among our respondents there
were only seven who combined rural and urban activities in a studied and therefore
strategic manner.22 For the remainder, their livelihoods were focused on the city, and
were focused full time on the urban context. Even the seven cases where households
combined farm and non-farm work, the balance of return was very much city-centred
and the farm tended to ‘tick-over’ in a subsidiary fashion.
22

Namely: interviews #006, #007, #008, #016, #018, #020 and #028.


Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1119

While our sample is not large enough to draw any statistical inferences about the

nature of those who combined farm/rural and non-farm/urban livelihoods, they
represented a balance between men and women, they spanned a range of ages and
dates of arrival in Hanoi, and were engaged in a variety of urban pursuits (Table 6).
The single notable feature was that none had KT1 or KT2 registration. This, though,
can be seen simply as a reflection of the respondents’ continued livelihood
engagement with their rural villages of origin.
One theme to recur time-and-again was the fact that it was difficult, even
impossible, to sustain a reasonable standard of living on the basis of farm work
alone. In line with Ravenstein’s ‘law’, our respondents left their rural roots in the
main ‘to ‘better’ themselves in material respects’. Even households with quite large
holdings of land expressed the view that they had little choice but to migrate and find
work in the city. Mr Nhieˆn (interview #020, 28.9.2010) owned seven sao (one
sao¼360m2) of land in Nam Dinh but was selling sandals from a bicycle along with
other members of his extended family; only in this way, he said, could he pay his
daughter’s university tuition fees of 700,000 VND per month. From his land he
could earn 14 million VND (two harvests of rice); from their work in Hanoi, he and
his wife, in a good month, could make 7 million VND. It was common for migrants
to turn over their land to relatives to farm – not infrequently without any rent being
paid – and closing up their house, absenting themselves from the village except for
periodic returns for major festivals and celebrations (see below).
What is striking, however – and this resonates with other Asian countries – is
that among the migrants we interviewed, notwithstanding the general view that
agriculture was physically hard, socially unattractive and remuneratively marginal,
the great majority still retained ownership of their land in the rural villages from
which they originated, even if they did not farm it. Only a very small number had
sold their land. De Brauw and Harigaya argue that migrants in Viet Nam keep hold
of their land not least because land rights are uncertain (2007, 431). We propose
that there is also a set of considerations that more broadly link to livelihood
security.
While there is no doubt that rural livelihoods have become increasingly marginal

in Viet Nam (as elsewhere in Asia), they were often still ‘sustainable’ in the sense that
land provided the possibility of basic subsistence security. Urban livelihoods were
usually more remunerative, and provided the opportunity for accumulation, but they
were also risky. Thus, in part, our respondents were keeping hold of their land to

Table 6.

#006
#007
#008
#016
#018
#020
#028

Migrants and their households with inter-locking livelihoods.
Age

Gender

Year of first
arrival in Hanoi

Occupation

Registration

57
29
26

61
57
47
34

M
F
M
F
M
M
F

1990
2003
2004
1998
1990
1998
1997

Porter
Seamstress
Factory worker
Boarding house owner
Xe om driver
Sandal seller
Junk trader

KT4

KT3
KT4 (?)
KT4
KT4
No registration
KT4

Source: migrant interviews, 2010.


1120

Nguyen Tuan Anh et al.

Downloaded by [Georgetown University] at 08:10 10 August 2013

ensure livelihood security (even when they did not farm the land), but had largely
abandoned farming so that they could achieve the possibility of upward livelihood
mobility.23 Thus most of our rural migrants, in terms of their current, prevailing
livelihoods were pursuing urban activities in urban space, with only a minority
combining this with agriculture in their villages of origin. But most were also
keeping a livelihood toehold in the countryside. Whether this toehold would be
reactivated was often far from certain. For many of the migrants in their 50s,
intended return was equated with retirement, but not with a re-agrarianisation of
living and livelihoods.
In terms of livelihoods, therefore, we found most of our respondents leading
urban lives and pursuing urban livelihoods; but it was also true that these
same migrants often remained rural in terms of their identities, to which we now
turn.
City versus village identities, urban versus rural behaviours

From our interviews, it emerged that one of the most persistent elements of the rural
in the urban was reflected in migrant identities and, although to a lesser extent, in
how urbanites regarded and treated rural migrants.
Mr Thinh (interview #026, 7.10.2010) was from Thanh Hoa, around 150 km
south of Hanoi, and arrived in Hanoi in 1996. His son and daughter were at school
and university in Hanoi respectively, he had a livelihood as a butcher which
embedded him thoroughly in the city, he had bought land and a house in Hanoi, and
had even gone so far as to sell the family house in his home village (but not his
agricultural land). On a number of grounds, therefore, Mr Thinh had insinuated his
way deeply into the fabric of the city. Even so, he told us:
I love my homeland [queˆ hu’o’ng].24 However, I do not love agricultural production.
Agricultural production is a hard job. . . . I was even a good farmer. But I do not like
farming. The homeland is where I was born. The homeland brought me up. The
homeland is in my heart. . . . The young and capable people should choose cities, the old
should live in the countryside. When you are of working age, you should live in cities.
When you retire you should live in the countryside. (Interview #026, 7.10.2010)

In this interview, and during a follow-up interview with Mr Thinh and his
daughter (interview #026b, 10.11.2010), it was clear that notwithstanding their long
engagement with Hanoi this family still identified strongly with their rural and
village roots. This was not, however, as farmers but rather as people from Thanh
Hoa. It was their rural and village roots, rather than their peasant origins which they
valued and which they were keen to preserve. Mr Xoai (interview #004, 6.9.2010)
was a tofu maker who, like Mr Thinh, had also lived in Hanoi for a long period,
23

As well as livelihood security there is also the issue of security following death. While the
government has been trying for some years to encourage cremation (burial uses valuable
farmland), among older generations to be buried in one’s homeland is highly important and
this usually requires land. (The government has also tried to promote cemeteries rather than

individual burial plots on farmland, albeit not terribly successfully.) Personal communication:
Andrew Hardy.
24
Queˆ hu’o’ng is variously translated as ‘homeland’, ‘native place’ and ‘native land’. ‘Home’ is
preferred here rather than ‘native’ because of its social intimations of belonging. The French
translation pays natal, however, is closer to the Vietnamese meaning of the term.


The Journal of Peasant Studies

1121

since 1987, and showed a similar emotional attachment to his rural roots. Mr Xoai
told us that he would ‘definitely’ return to Bac Ninh – after another five more years
in Hanoi, he said.25 He put this in poetic terms, reciting the second verse of a popular
Vietnamese song:

Downloaded by [Georgetown University] at 08:10 10 August 2013

Queˆ hu’o’ng la` chum kh e’ˆ ngot`
˜ˆ nga`y
Cho con tre`o ha´i moi
Queˆ hu’o’ng la` du’o`’ng
di

 hoc`
` bu’o`’m va`ng bay
Con v e’ˆ ro’p
Your homeland is a bunch of sweet starfruit
For you to gather each day.

Your homeland is the path from school
Which you follow home, shaded by yellow butterflies.
(interview #004, 6.9.2010; Lyrics by Do˜ˆ Trung Quaˆn; music by Gia´p V
an Thach)
`

Although Mr Xoai did not go on to recite the rest of the song, the end suggests
that if you do not remember your home village, you cannot become a ‘good person’
(thanh nguoi, meaning a moral person). It is, therefore, quite a stark statement on the
Confucian value of attachment to home.26
The pull of the rural, or more particularly the village/homeland, is strong for
migrants, even after a decade or more in Hanoi. This was never, however, expressed
in terms of a desire to return to farming and agriculture; often, quite the opposite.
The frequent references in Vietnamese poetry and literature, particularly in the
north, to que huong – homeland – which we also saw in the way in which our
respondents talked about their villages of origin, should not be interpreted as
necessarily indicating an instrumental attachment to place. Jacka writes of similarly
conflicting emotions that attachment to home on the one hand and the attractions of
the city on the other embody for young women migrants in Beijing:
While the dynamism of the city exerts a powerful pull, my interlocutors indicated that
the desire not to return to the countryside—because it is too poor, villagers’ thinking is
too ‘backward’ and ‘feudal,’ farming is too draining, and they would not be able to
readjust to rural life after living in the city—is often an even stronger motivation for
remaining in Beijing. (Jacka 2005, 64)

Thuoc, 38 years old at the time we interviewed him, had arrived in Hanoi from
Ha Tinh in 1998 (interview #025, 7.10.2010). He did not complete secondary school
because of his family’s straightened circumstances, and came to the city in search of
work. For the first three or four years after he arrived he found a series of low paid,
menial and marginal jobs – as a cart puller, bricklayer’s assistant and dish-washer.

He and his wife, however, managed to save enough to set up a restaurant, which they
were running at the time of the interview, with Thuoc also working part-time as a xe
om (motorcycle taxi) driver. In 2009, he relinquished his rights to agricultural land in
his homeland in order to secure Hanoi residency, and get his only child, a son, into
school:

25
26

By which time, Bac Ninh may well have been engulfed by Hanoi!
I am grateful to Andrew Hardy for pointing out this link.


1122

Nguyen Tuan Anh et al.

Interviewer: Have you bought land or a house in Hanoi?
Thuoc: [No, but] probably in the future. At present, we cannot buy. We only have
the intention.
Interviewer: Have you got enough money to buy a house?
Thuoc: Not yet. . . We will wait to see how our business fares in the future.
Interviewer: How many times per year do you visit your homeland?
Thuoc: Three or four times per year. . ..

Downloaded by [Georgetown University] at 08:10 10 August 2013

Interviewer: Why do you visit your homeland?
Thuoc: [To] visit my parents, and my brothers and sisters. My sentiments for my
homeland are very strong, so I visit my homeland often.

Interviewer: Do you prefer to live here or live in your homeland?
Thuoc: I prefer to live here because there is a lot of work here to do, and not much
work in my homeland.
Interviewer: Do you intend to return and reside in your homeland in the future?
Thuoc: Yes, when I retire I will return to my homeland.
Interviewer: When will you return and settle in your homeland?
Thuoc: When my son has grown up, I will retire and I will return and live in my
homeland.
Interviewer: Do you think your son will settle in Hanoi or in your homeland?
Thuoc: If he does well at school, he will live in Hanoi.
There is also the question of whether this pull of the rural and the homeland was
likely to be transmitted inter-generationally. Sons and daughters who had been
educated in Hanoi, who had few friends in their parents’ home villages, who may
never had spent more than a week or two in the village over Tet (lunar New Year),
and who probably did not know how to farm, have different identities from their
parents, identities which are more urban than rural.
In terms of how migrants are treated by other urbanites, this was rarely conveyed
to us as particularly problematic. Most still had their non-Hanoian accents which
marked them out as originating from the provinces, and certain occupations – such as
junk scavenging, xe om driving, portering, and selling goods from bicycles – are classic
migrant livelihoods (see DiGregorio 1994 for an early study). The men and women
undertaking these types of work are, almost without exception, migrants. There was
also a sense among some of our interviewees that such work was demeaning. Mr Vinh
(interview #006, 14.9.2010) was a porter in the local market, selling his labour day-today and hour-to-hour, doing ‘the hardest job in the city’, as he put it. Mr Vinh had
arrived in Hanoi in 1990 from Thanh Hoa’s Quang Xuong district, one of that
province’s poorest areas, leaving because his young family were hungry and travelling
to Hanoi was the only thing he could think to do. In the two decades he had spent
labouring in Hanoi while his wife continued to manage the farm he had, remarkably,
managed to put three of his four children through university.27 When we came to ask
27

Throughout this 20 year period of city work, Mr Vinh remained a short-term ‘temporary’
migrant in Hanoi, with KT4 registration.


Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1123

whether we could take a photo of him for our records he rather reluctantly agreed,
but asked us not to publicise the photograph because he did not want his children’s
friends, colleagues or employers to see what he did to get by: it might be
embarrassing. But while such work might be regarded as demeaning, we did not
find that such differences stigmatised migrants in a manner that has been alluded to
in some other countries, such as Thailand and India (see Korinek et al. 2005, Bird
and Deshingkar 2009), possibly because such a large proportion of Hanoi’s
population are migrants, the children of migrants, or the grandchildren of
migrants.28 One does not have to dig very deeply into the past to uncover the
rural origins of most of the city’s population.
Linked to the issue of urban/city and rural/village identities, is that of behaviour.
Thompson, working in Malaysia, writes of the degree to which his kampong (village)
has become ‘socially urban’ in the way in which people consume, dress, behave and
interact: ‘In contrast to the pervasive rhetoric that reinforces a sense of rural–urban
difference through nostalgic fixation, this paper argues that. . . kampong [villages] in
Malaysia today have become as much urban as rural spaces [with] respect to
production, consumption and social interaction. . .’ (Thompson 2004, 2357). Like
Derks’ work in Cambodia, the penetration of television and the mass media into the
countryside, has ‘brought the city closer than ever to the world of Cambodia’s

villagers’ (2008, 142). At the same time, urban attitudes can be found in rural areas.
The issue, perhaps, is that while the designations ‘rural’ and ‘urban’, which are often
paired with ‘traditional’ and ‘modern’, used to have a strong(er) spatial and
geographical designation, this has been eroded with increasing mobility. People –
like the xe om driver Mr Viet (interview #018, 28.9.2010), the junk scavenger Ms
Lien (interview #017, 24.9.2010), and the porter Mr Vinh (interview #006, 14.9.2010)
– might be visibly ‘rural’, but this was more a measure of the work they did, and
therefore their class, than where they lived. The reverse was also true: urban
sensibilities and behaviours are colonising rural spaces and settlements as those who
have worked in urban areas and in factories bring their habits and tendencies back to
their rural places of origin.
It is often when people return to the countryside that the progressive
transformations in behaviour that arise from living and working in the city
become evident. Tham (interview #022b, 10.11.2010), the daughter of Mrs Hong a
migrant from Nam Dinh, had married a Nam Dinh man and from time-to-time
had to visit her husband’s village and parents-in-law. In talking about the
differences between people from Hanoi and the outer provinces, Tham said that
country folk were rather shy and did not interact smoothly with Hanoians, who
were more confident, worldly and articulate. Never having farmed, she knew
nothing about agriculture and when she visited her husband’s village felt out of
place, surrounded by rice fields – a fish out of water. The other villagers in her
husband’s village regarded her as a Hanoian, but she tried to fit in by dressing
appropriately and eating the simple food set before her. She didn’t wear dresses or
skirts, she told us, because she thought this would be interpreted as meaning she
was work shy and unwilling to take on manual jobs around the house. When it
came to clothes and fashion, Tham said that rural people tried to dress in a
modern way, but often a little too brashly – they ‘follow trends but in a country
manner’, she said. This hints, once more, at differences which are as much about
28

Annuska Derks (2008) also writes of Phnom Penh, Cambodia’s capital, being a ‘city of
migrants’.


Downloaded by [Georgetown University] at 08:10 10 August 2013

1124

Nguyen Tuan Anh et al.

class as they are about geography. It may be possible to be modern in the rural;
but to be urbane – to be sophisticated – is a slightly different thing.
Tham may have become a new urbanite, but the transition was not always
successfully navigated. Linh (interview #008, 14.9.2010), a young migrant from Yen
Thanh district, Nghe An province, had arrived in Hanoi in 2004 and was working in
a joint stock garment factory at the time of our interview. He told us that he had no
desire whatsoever to work in the paddy field, the occupation that he was born to.
(No man of his age, he told us, stays at home to farm: only the ‘incompetent’ and the
‘stupid’.) At the same time, he was intimidated by people with better qualifications or
higher education than he had achieved. On reviewing this interview, we sensed that
Linh was a migrant caught between two worlds. He professed a desire to be modern
and move on from his farming and rural roots; and yet he actively avoided mixing
with those who might enable him to achieve this. He was a frustrated young man; his
girl friend had just dumped him – an event which he blamed on his rural roots, his
rustic manner, and his limited education.
There are now subtleties, fine gradations of difference, that the designations
‘urban behaviours’ and ‘rural behaviours’ do not adequately accommodate. To be
urban encompasses as a set of social and cultural practices and a way – or grammar –
of living (Bourdieu 2000). Urban and rural in terms of behaviour are not mutually
exclusive categories; and neither are they tied to place.

Social networks and obligations
The final way in which we consider the process of becoming urban, is in terms of the
social networks, associations and obligations that structure migrants’ everyday lives.
The question we sought to illuminate in our interviews was: do the spatial and social
configurations of these networks indicate the continuing importance of rural living,
or do they constitute a shift towards urban living?
Linh’s case (interview #008, 14.9.2010), mentioned above, shows how social
networks may become re-sited but not re-worked. His former school friends had all
left the village so that when he did return home it was ‘boring’ – there was simply no
one there.29 At the same time, however, he had not built up a social network that
might be described as urban in configuration; he mixed with other migrants, many
from his own province, and rented a room close to his place of work in Phuong Mai
Ward, which he shared with five other young migrant men. In the year prior to the
interview he had not returned home at all (even during the Tet holiday), but this
extended period of living in Hanoi was not translated into a social network that
could be viewed as indicative of a deep engagement with the urban.
Linh can be contrasted with the experience of Ms Hang (interview #009,
21.9.2010), the 34 year-old cleaner who had arrived in Hanoi at the end of 2000 and
managed, with her husband, to buy land in the city (see above): their social networks
in the city had substantially thickened over time, revealing a more profound
embedding in the social fabric of the city. While she returned to her rural roots for
major events and festivities like Tet, Independence Day, and her sister’s wedding, her
29

The observation that many villages in the Red River Delta consist largely of the older
generations and children of school age was mentioned time-and-again. Younger cohorts of
working age have left to find work elsewhere, more often than not in urban or peri-urban
settings.

Downloaded by [Georgetown University] at 08:10 10 August 2013

The Journal of Peasant Studies

1125

stock of social capital in Hanoi had developed, while that in her home village had
gradually diminished. She had joined her local area committee, helped to organise
children’s festivities in her ward connected with the autumn full moon festival, and
had good and well developed relations with her neighbours. To be sure, she also told
us that despite her best efforts Hanoians seemed hard to get to know, but she was
actively and successfully embedding herself in urban society.
There was often a strong sense of obligation to a migrant’s rural origins. We saw
this operating in both a symbolic and a material manner. Regarding the former, the
role of patrilineal traditions among the Kinh (Viet) of the Red River Delta area
(where most of our migrants originated), was significant. The importance of ‘home’
and ‘homeland’ (see above), emotionally important for many, was made yet more
important in the Vietnamese context because a patriline is recorded, confirmed and
celebrated in genealogies inscribed in ancestral altars (Bryant 2002, and see
Friedman et al. 2003, Rydstrom et al. 2008). The eldest son has ritual obligations,
which connect dead members of a patriline (a line of descent from father to son) with
those who are yet to become. Migration, clearly, can disrupt these familial connections and obligations particularly when migrants are male and especially if they
are eldest sons. It creates a social and cultural context where ‘return’, periodic
though it may be, and continuing contact and the performance of certain rites
become important. While there is evidence that patrilineal traditions have become
frayed (or re-worked), because of the combined effects of a long period of war,
attempts to create a new Vietnamese society based on socialist and communist
principles, and the ‘modernisation’ effects of the reform (doi moi) process, they are
nonetheless significant in shaping the connections that migrants maintain with
home.30

Mr Viet (interview #018, 28.9.2010), the 57 year-old xe om driver introduced
earlier, had been living in Hanoi for 20 years but as an eldest son felt a strong sense
of obligation in terms of the maintenance of the family patriline. He returned to the
village for all major and many minor events, celebrations and festivities, and
maintained close contacts with his village, partly because of his obligations as an
eldest son – and this despite the fact that as he had four daughters and no son, so it
will be his brother’s son who will take on responsibility for the patriline on Mr Viet’s
death. But not everyone is quite so assiduous as Mr Viet; Ms Duyeˆn’s (interview
#024, 7.10.2010) father is an eldest son from Thanh Chuong, Nghe An (300 km
south of Hanoi), and shuttles between his children in different parts of the country
(from Ho Chi Minh City in the south to Hanoi in the north), only occasionally
returning to his birthplace to worship at the ancestral altar. For him, the obligations
of being an eldest son only extended to periodic visits, and not to maintaining a more
substantial social and ritual engagement with the village.
Of more general relevance were the material obligations that migrants felt
towards their rural roots, and the roles that these had in shaping social associations.
For husbands and wives who were separated by migration, this was characteristic;
indeed, it was the very justification for migration. Daughters, and sometimes sons,
30

There is reason to suggest that with doi moi, patrilineal principles may have become more
important, rather than less. While Viet Nam may be a transitional and modernising society,
certain elements that might be viewed as emblematic of tradition have been revitalised, such as
life cycle ceremonies. Doi moi has loosened the old communist strictures on acceptable practice
while also redirecting responsibilities for social protection back to the family.


Downloaded by [Georgetown University] at 08:10 10 August 2013

1126

Nguyen Tuan Anh et al.

also felt an obligation towards their natal families. This was reflected in the
remittance of money by children to ageing parents to support their existences in the
countryside, by wives and husbands to their partners to finance the reproduction of
the family, and by parents to sons and daughters to subsidise their education, and
sometimes by migrants to the natal community more generally in the form of
donations to maintain, for example, the ancestral house. Perhaps the most striking of
such instances was that of 18 year-old Ha.
Ha (interview #010, 21.9.2010) was working as a waitress in a restaurant when
we interviewed her, having arrived in Hanoi from Hau Loc District, Thang Hoa
province (149 km south of Hanoi) six years previously. Her situation was difficult, and had been almost throughout her time in the city. She had no registration
papers, was working in a restaurant and sleeping on the restaurant tables at night
(after a succession of marginal and poorly paid jobs), and had been forced to
leave home to find work in 2004 at the age of 12. Her life in Hanoi was largely
shaped by the demands that her mother, a single parent, placed on her. Her elder
sister and husband had run up considerable debts during an unsuccessful foray to
Malaysia to find work, and they were both working as contract labourers in
China at the time of our interview. There was continual pressure on Ha to remit
money home, and her mother and sister kept regular tabs by mobile phone. Even
though she rarely returned home (just once in two years, she told us), Ha’s
obligations and contacts with home were strong and continuing. Ha, in tears, told
us that her mother didn’t care for or support her; she was just a source of income
for the wider family and her own future and prospects were being sacrificed in the
process. She wanted to take a vocational course to acquire some skills that might
get her out of the dead-end jobs she had been forced to take on; but she couldn’t
afford to because all her spare income was channelled back to her mother and
sister.
A common theme among those poorer migrants with material obligations in

rural areas was for their social networks to extend back to their villages of origin,
rather than to weave their way through Hanoi society. Mr Thanh (interview #027,
7.10.2010), from Nam Dinh (90 km south of Hanoi), was a building site porter and
had been working in Hanoi for six years. He was sharing a grubby basement room
with two other migrants from the same village and could neither afford the money to
bring his family to Hanoi, nor the time to build any meaningful social life in the city
– he rarely went out, and in the evenings stayed in his room and watched TV. He
earned some 3 million VND per month (US$150) and was able to save about 1
million VND (US$50) which he either took home himself or sometimes gave to a bus
driver friend heading to his village. All the surplus income he earned was channelled
back to the village to support his wife and son, and invest in land.31 While his social
networks were focused on his family in the village, the village itself had become
socially hollow due to wide-scale and sustained migrant absences. Mr Thanh told us
that among his age group, the majority of men had left the village, mostly to go
south, to work as fishers, builders and in various manual occupations. The only time
when returning home would mean seeing his old school friends and peers was during
the annual Tet holiday when most came back ‘home’; at other times the village was

31

He had purchased a house plot (but not yet built a house), and also spent 20 million VND
buying a fish pond in the village.