Livelihood outcomes in the residential cluster and dyke programme in an giang providence vietnam case study in tan an and vinh hoa commune tan chau district an giang province

Livelihood Outcomes in the Residential Cluster and Dyke Programme in an Giang Providence,VietnamCase study in Tan An and Vinh Hoa Commune, Tan Chau District, An Giang Province Pham Xuan

Livelihood Outcomes in the Residential Cluster and Dyke Programme in an Giang Providence,Vietnam

Case study in Tan An and Vinh Hoa Commune, Tan Chau District, An Giang Province

Pham Xuan Phu, An Giang University (AGU), Vietnam

Master Thesis No 41

Master Thesis in Rural Development with Specialization

in Livelihood and Natural Resource Management

Livelihood Outcomes in the Residential

Cluster and Dyke Programme in An Giang Province, Vietnam

C a s e s t u d y i n T a n A n a n d V i n h H o a C o m m u n e s , T a n C h a u D i s t r i c t ,

A n G i a n g P r o v i n c e

Pham Xuan Phu, An Giang University (AGU), Vietnam

Master Thesis in Rural Development with Specialization in Livelihoods and Natural Resource Management

Master Thesis No 41 | Hue City, Vietnam | November 2007 | ISSN: 1403-7998

Department of Urban and Rural Development | Swedish University of Agricultural Sciences

Tan Chau district is the area most heavily flooded and most affected by river bank erosion by annual floods in An Giang province The floods have caused considerable and serious damage

to humans and property in the district As in other flooded provinces in the Mekong Delta,

An Giang Government implements a program for stabilizing livelihoods and promoting sustainable improvements of the lives of communities “living with floods” by building residential clusters and dykes in order to solve the damage in the longer term This is a new program, and there have been studies by Ausaid and Care on residential clusters in An Giang, Dong Thap, Long An in Provinces in the Mekong Delta, Vietnam This study looks more specifically at changes in household livelihoods and chooses a district which has not yet been studied The aim of the study is to understand how people’s livelihoods change when they move and the advantages and disadvantages of moving The study explores livelihoods of people who moved and did not move to residential clusters and dykes; to understand the impact of the resettlement program on livelihoods after resettling as well as to explore how the livelihood strategies and conditions for those living in the residential cluster and dyke have changed, before and after the move

The study was carried out in two communes of Tan Chau district including Vinh Hoa and Tan An commune, by using PRA tools and interviewing households A total of 72 households were interviewed to get information about the characteristics of the household such as manpower, household activities, income, capital, employment, their coping with disasters of floods and changes of living before and after the resettlement

The results show that people who live in the residential cluster and dyke area were mainly poor and subject to preferential policy (beneficiary of social welfare) (64% fleeing from flood

of which, 53% fleeing from flood and 11% households subjects to preferential policy, and 36% fleeing from river bank erosion) Most of them were of low education level (17% illiterate and 61% at primary education level) Living conditions were not sufficient as only 55% households used clean water and 14% households had usable semi septic toilet Their job opportunity and total working days were lower than before with 89% household found it difficult to find jobs, and this made their income unstable Therefore, while the new cluster creates a safe place for them it also reduces the opportunities of their livelihood

Key words: Flood, residential cluster and dyke, resettlement, livelihood, risk, vulnerability, natural hazard

Dedication

To my wife Ngo Thi Bao Tran, my daughter and my mother Nguyen Thi Hue, my father Pham Xuan Tai, my brother Pham Xuan Tuan, Pham Xuan Ngoc, Pham Xuan Quy, my sister Pham Thi Kim Dung and my mother in law Tang Nguyet Thu, my father in Law Ngo Kien Lam

The study was financed by SIDA within the support to the RDViet project implemented by SLU, the Department of Urban and Rural Development, Swedish University of Agricultural Sciences in collaboration with the Department of Rural Development, Hue University of Agriculture and Forestry

I would like to express the deepest gratitude to Dr Britta Ogle, International coordinator

of the SIDA project

My heart-felt thanks are also to my thesis supervisor Prof Adam Pain, Department of Urban and Rural Development, SLU, Vice supervisor Dr Ian Christoplos, Dr Malin Beckman Department of Urban and Rural Development, SLU and Mr Vo-Tong Anh as a co-advisor, Dean of Agriculture and Natural Resources, Faculty of An Giang University, for providing me valuable comments, conductive suggestions and encouragement during the master course

I also wish to express my sincere gratitude to professors and lecturers of this course who gave me their useful knowledge and technical assistance

Associate Prof Le Duc Ngoan, Ms Nguyen Anh Phuong (Hue Agriculture Forestry University); Prof Vo Tong Xuan, Mr Doan Huu Luc, Mr Vo Tong Anh (An Giang University) who gave me precious opportunity and encouragement during this course

My department of Soil and Natural Resources, Agriculture and Natural Resources Faculty, An Giang University who encouraged and assisted me during the course, especially Miss Ho Thi Thu Ha, Ms Ngo Thuy Bao Tran, Mr Le Van Lenh, Mr Pham Duy Tien,

Mr Nguyen Thanh Son, Ms Pham Huynh Thanh Van, Mr Nguyen Van Kien, Mr Pham Anh Dung, Mr Bui Van Sang, Mr Ho Van Thao and three students Nguyen Phuoc Nguyen, Doan Thanh Huong, Pham Van Phuoc for their great help in the survey and accomplishing the research

The officers from Flood and Storm control Committees of An Giang province and Tan Chau district for providing useful information

The commune officers at Tan An, Vinh Hoa in Tan Chau district who cooperated and assisted me in carrying out this research

I am very grateful to thank all my classmates of M.Sc course on Rural Development who shared their knowledge and experience with me during my study and my stay in Hue City, a cultural heritage city of the world

My deep thanks are also to my family for their love encouragement and support to me particularly to wife Ngo Thuy Bao Tran who encouraged, helped and inspired me to finishing the course

LIST OF CONTENTS

ABSTRACT 2

ACKNOWLEDGEMENT 4

LIST OF ABBREVIATIONS 7

LIST OF TABLES, FIGURES AND BOXES 8

1 INTRODUCTION 9

1.1 Background 9

1.1.1 The situation of flooding in the Mekong Delta 9

1.1.2 Implementation of government policy on building residential clusters and dykes 11

1.1.3 Advantages and disadvantages in residential clusters and dykes 12

1.1.4 Background of socio-economic situation in study site 12

1.2 Problem statement 14

1.3 Research Objectives 15

1.4 Research Questions 15

1.5 Scope and Limitations 15

2 REVIEW OF RELATED LITERATURE 16

2.1 Theories and basic concepts 16

2.1.1 Vulnerability 16

2.1.2 Livelihood 16

2.1.3 Livelihood assets 17

2.2 Development of migration in the world and settlement in residential clusters and dykes in the Mekong Delta 18

2.2.1 Development of migration in the world 18

2.2.2 History and development of settlement in residential clusters and dykes in Mekong Delta 19

2.3 Impact of floods and responses to floods 20

2.3.1 Support for recovery from flood disasters 20

2.3.2 Impact of floods 20

2.3.3 How the household’s traditionally managed flooding, and present responses 21

3 RESEARCH METHODOLOGY 22

3.1 Rationale of study site 22

3.1.1 Selection criteria of sites 22

3.1.2 Sampling Criteria 22

3.2 The description of research sites 24

3.2.1 History of moving into residential clusters and dykes 26

3.3 Vinh Hoa Commune 27

3.3.1 Physical characteristic and economic status 27

3.3.1.1 The residential cluster at township Vinh Hoa centre 28

3.4 Tan An commune 28

3.4.1.1 Physical characteristic and economic status 28

3.4.1.2 The residential dyke at Tan An Commune 28

3.5 The difference between residential clusters and dykes 29

3.6 Conceptual framework analysis 29

3.7 Data collection 31

3.8 Data collection methods 31

3.8.1 Secondary data 31

3.8.2 Primary data 31

3.9 Data analysis methods 33

4 RESULTS AND DISCUSSION 34

4.1 The building program of clusters and dykes in Tan Chau district 34

4.2 Resettlement process in the residential clusters and dykes in the district 35

4.2.1 Approval process of resettlement in the residential clusters and dykes 35

4.2.2 Approved households for resettlement in the residential clusters and dykes 36

4.3 Who moved or did not move to residential clusters and dykes? 37

4.3.1 Household’s decision making on moving into the residential clusters 38

4.3.2 Household decision making on not moving into the residential clusters 39

4.4 Households livelihood inside the residential clusters and dykes 43

4.4.1 General information of interviewed households in the residential cluster and dyke 43

4.4.2 Household living conditions in the residential cluster and dyke 44

4.5 Factors influencing livelihoods after resettlement 46

4.5.1 Peoples’ livelihood activities before and after resettlement 46

4.5.2 Job opportunities and total working days of main household labourers 49

4.6 SWOT analysis 53

4.7 Risk to people before and after moving to residential clusters and dykes 54

4.7.1 Risk to human life and people’s assets 54

4.7.2 People’s concerns about some implicit dangers in residential clusters and dykes 54

4.8 People’s opinion to stabilize in residential cluster and dyke 55

5 CONCLUSIONS 57

6 REFERENCES 58

7 APPENDICES 61

LIST OF ABBREVIATIONS

AusAID: Australian Agency for International Development BBS: Bangladesh Bureau of Statistics

CARE: CARE International in Vietnam

ECHO: European Community Humanitarian Office

FAO: Food and Agriculture Organization

GDP: Gross Domestic Product

IUCN: The World Conservation Union

MDPA: Mekong Delta Poverty Analysis

MDPPA: Mekong Delta Participatory Poverty Assessment MOC: Ministry of Construction

MOF: Ministry of Finance

MOLISA: Ministry of Labour, Invalids, and Social Affairs NGOs: Non Government Organizations

PC: People’s Committee

PRA: Participatory Rural Appraisal

SWOT: Strength Weakness Opportunity Threat

UN: United Nations

USD: American Dollar (USD 1 = VND15, 830 as June, 2005) VND: Vietnam Dong

LIST OF TABLES, FIGURES AND BOXES

TABLES

Table 3: Local people’s participation in planning and implementation of the residential

Table 6: Household assets of house and land before moving to residential clusters and dykes 38

Table 10: RRA results with groups of households outside the clusters and dykes in Tan An

Table 18: Households opportunity to find jobs before and after moving to residential

Table 19: The income differences of surveyed households after moving into the cluster & dyke 51

Table 22: Local people’s opinions of the choice to stay in the residential cluster/dykes or to

FIGURES

Figure 4: The project of construction of secure houses in the condition of regular flooding

Figure 6: Location of selected residential cluster and dyke in Tan Chau district, An Giang

BOXES

Box 1: Why do people move to the resident clusters and dykes, instead of choosing other

Box 3: Why do people (Vinh Hoa commune) not move to the resident clusters and dykes? 41

1 INTRODUCTION

1.1.1 The situation of flooding in the Mekong Delta

Seven of the 12 Mekong Delta Provinces are in flood-prone areas: Dong Thap, An Giang, Long An, Kien Gang, Tien Gang, Can Tho and Vinh Long In the seven provinces above, the flooded area covers around 2 million ha (accounting for about 50 percent of the Mekong Delta area) with around 11 million people (about 1.9 million households) for these seven provinces

Figure 1: The Mekong River Delta

The number of flooded communes in these 7 provinces is 610, of which during the flood season 77 are deeply submerged (over 3 meters in depth), 112 are submerged to an average level (from 2 to 3 meters deep), 329 are shallowly submerged (under 2 meters) and 82

communes are in eroded estuaries or coastal areas (MOC, 2002) Similar to other provinces in

the Mekong Delta, floods in An Giang cause prolonged deep inundation and river bank erosion In this study area, people living in the residential clusters and dykes were poor, they came from the deeply flooded areas with at least 1.2 meters water level or river bank erosion

of the Tien and Hau rivers and on the temporary houses This is still a big problem so they are a target for resettlement Extremely devastating floods were found in the years 1994, 1995,

1996, 2000, 2001, and 2002 Particularly, the flooding in 2000 caused the highest damage and was the largest flood compared to severe floods of 1961, 1966, 1978 and 1996 Since the

recognition that “human beings cannot change what ‘mother nature’ has given” was appreciated, the

strategies of dealing with flood were rethought and the concept of “living with flood” was

introduced by the Government (Dang Quang Tinh & Pham Thanh Hang, 2003) This concept is based on traditional practices of local people and has been adopted as a strategy for the Mekong Delta According to Fforde et al., 2003, rural people are fully for dealing with floods as the most frequent disaster, which lasts from July to December every year Located along rivers, they used natural materials for building houses, such as flimsy wood, or made a living by both catching fish and shellfish and collecting wild vegetables

According to Dao Cong Tien (2002), floods in the Mekong Delta last around six months, between July and December every year, one month later than flooding in the upstream of the Mekong Delta Compared with upstream and flash flooding in the sloping areas of Central and Northern Mountain areas, flooding in the Mekong Delta occurs more gently but lasts longer The flooding season starts from July to the end of August, becomes highest from September to October and then declines from November to December Though the flood levels from 2001 to 2005 were almost as high as 2000, the impact of the floods was not devastating as in 2000

There are positive impacts of flooding Not only do the floods bring sediments to the rice fields but they also make the soil more fertile (diluting sulphate), improve soil quality (land reclamation), enable the development of aquaculture as well as balance the ecology and promote ecotourism areas that are characteristic for the Mekong Delta, such as Dong Thap Muoi (Plain of Reeds), and Tram Chim (Bird Yard), National Park This 4 million ha region with nearly 3000 ha of cultivated areas supplies more than 50% of staple food and 60% of fish production for the entire nation, accounting for 27% percent of the total GDP of Viet Nam

In addition, the poor people who live in this area are able to earn their living during the floods by catching fish, shellfish and crab, growing water vegetables such as Lotus, Neptunia, Water Caltrop or collecting Sesbania flower to increase food for their family

Table 1: Major losses of property from floods and bank erosion in An Giang

Property from floods and

bank erosion (million

in 2000 caused the deaths of 501 people - most of them were children, and brought about serious loss estimated over VND 4,200 billion; the flood in 2001 caused the deaths of 407 people, and a total loss estimated over VND 1,594 billion Particularly, children were vulnerable during the flood disasters The last records showed that most deaths during the floods were of children (Appendix 1)

1.1.2 Implementation of government policy on building residential clusters and dykes

Dang Quang Tinh and Pham Thanh Hang (2003) found in “Living with Floods in the Mekong River Delta of Viet Nam” that poor households were highly vulnerable to daily difficulties and occasional shocks that may strike an individual, a family or a community Because their income level was very low and unstable, they had low saving capacity and were unlikely to be able to resist unexpected shocks (such as natural disaster, loss of harvest, of job

of labour resources, of health, and so on) In addition to crop damage and household poverty, human loss was another important consideration (Fforde et al., 2003) Day care centres were set up to take care of infants and small children during flood season, enabling their parents to work for earning a living

Being an annually flooded province, An Giang has to consider serious damage to human life and property, about 33% of overall damage (major losses of property from floods and bank erosion is relatively high) compared to other provinces in the Mekong Delta (Vietnamese Yearbook, 2001) Tan Chau is a river bank erosion area, which is affected deeply by floods in

An Giang province In order to reduce the damage, the State proposed research policies and developed several programs about living peacefully with floods, such as the program of developing traffic and building up residential clusters of the Mekong delta (Decision No 99/TTg, 09/02/1996), or the program of giving poor residents loans to raise their house foundation or to make houses on poles to avoid the flood (Decision No 256/TTg, 24/4/1996) However, these programs did not bring about the expected effect of “to live and work in peace and contentment” in residential clusters and dykes due to the lack of previous practical experience of the Government, the Ministries, and sectors at all levels (Fforde et al., 2003) There were more than 60 guiding documents of clusters development having been issused by related authorities of ministries and sectors in respect to the most appropriate

2001, the Government issued Decision No 1548/2001/QD-TTg about investing to raise the house basis to overcome floods for the Mekong delta residential clusters and dykes in 2002 (MOC, 2002) Construction investment to develop the residential clusters and dykes is one of the main goals of socio-economic development of the Mekong Delta in 2001-2005 This is a quite new policy of the Government The target of the policy is to improve life quality of residents in this area, in order to guarantee for local people that it is not necessary to emigrate and that people have a comfortable and stable life and can live and work in peace and contentment (MOC, 2002)

To restrict the damages from floods, the construction and development of the residential clusters and dykes is a relatively new practice in the Mekong Delta of Vietnam (Decision No 105/2002/QD-TTg issued on 02/8/2002 by the Prime Minister on the Government policy

to allow households in flooded areas to buy on credit the houses and housing foundations in residential clusters/dykes of the Mekong Delta) Several government agencies and non-governmental organizations have been involved in providing assistance to communities living

in these residential clusters and dykes with the aims of stabilizing livelihoods and promoting sustainable improvements of the lives of communities ‘living with floods’ After the historically severe flood in the Mekong River in the year 2000, Vietnamese Government introduced a new concept of ‘living with floods’ that has become the major strategy for disaster mitigation This strategy was based on a full realization of the non-preventable nature

of the floods in the Mekong River Delta as well as their both positive and negative impacts

on economic development and on people’s lives It is interesting to note that ‘living with floods’ had been practiced by the people in the Mekong River Delta of Viet Nam for a long

time before it became a strategy of the Government For instance, as stated in the provincial program 31 dealing with production development, employment and raised income of people living in flooded areas, that introduction of appropriate aquatic production systems, off-farm opportunities as well as small-scale services for the people was put in practice This program had been initially effective as people basically knew how to use natural resources in a sustainable way through suitable farming production systems during the flood season Extensive freshwater shrimp farming or cultivating neptunia, for example, were among significant sources contributing to their raised income

1.1.3 Advantages and disadvantages in residential clusters and dykes

A number of outcomes expected to be achieved include better employment, concrete housing, and safety Thus, the intended impact of residential clusters and dykes program on people‘s livelihood in flood areas was focused on the changes of livelihood assets and the use

of livelihood assets to cope with vulnerability The construction and development of the residential clusters and dykes in the Mekong delta can assist the poor with the means to protect their livelihoods against shocks and to build up and diversify their livelihood activities (Dao Cong Tien, 2002) In 2002, one of CARE International observers’ studies (about policy

of building in residential clusters in Mekong Delta, Residential clusters and dykes research in

An Giang, Dong Thap, Long An provinces in Mekong Delta, Viet Nam, Fforde et., al 2003) indicated that research on the residential clusters and dykes programs was needed and that no study had been conducted to look into the effectiveness of these programs from a multi-sectoral and long-term perspective In 2003, research funded by ECHO and CARE

dykes believed that they would miss out on work if they lived in the new clusters because people hiring labourers would prefer to use labour who lived nearby Also they complained that in the cluster areas they lacked basic facilities such as clean water, latrines, electricity and waste management to provide a hygienic environment

Although the residential clusters help people to live safely and they could provide good livelihood conditions because they are aimed at making people's lives more stable by providing a safe home, they also have posed some limitations such as lack of job opportunities, increase social evil, polluted environment, among others (Care, 2003) The long-term needs must also be considered when designing and implementing these programs (Chitose Noguchi, in CARE, 2003)

1.1.4 Background of socio-economic situation in study site

Located in the Mekong Delta, An Giang is an upstream Province where the two Tien and Hau rivers, deriving from the great Mekong River, start and which is regularly under the influence of disaster and annual flooding seasons Many floods have caused heavy damage to human lives and property in the Mekong Delta in general and in An Giang in particular Especially in the year 2000 flooding season, the floods were estimated as the most severe within the last 70 years in the Mekong Delta, and those consecutive floods in 2001 and 2002 have caused great damage to production, infrastructure, property of people and government, and human lives (especially children) The heavy losses have not only influenced local socio-economical situation, creating difficulties to thousand of households, but also the environment and health of people in the flood affected areas

Realizing the objective necessity of living with floods in an upstream Province and overcoming the challenges by nature, since 1996, An Giang province has actively implemented Decision No 99 by the Prime Minister on improvement of water management and communication systems and the construction of residential clusters/dykes in the Mekong Delta It also implemented Decision No 256 by the Prime Minister on facilitating for thousands of poor farmer households to borrow for raising foundations According to the management board for residential clusters/dykes in An Giang province, the results between

1996 and 2000 show that 54 residential dykes have been built and nearly 8,700 household have been provided with stable shelters In the late 2001, the Prime Minister issued Decision No.173 on socio-economic development in the Mekong Delta in the period 2001- 2005 and Decision No 1548 on the investment in foundation construction for making residential clusters/dykes in deeply flooded areas in the Mekong Delta in 2002 Together with the other seven Provinces in the Mekong Delta, An Giang has launched the program of building residential clusters/dykes, according to which in 2002 the province has invested in building

30 clusters and 52 dykes to provide housing foundation for 18,154 households to live stably and securely during flooding seasons in a total area of 390 ha, at a total of VND 337 billion

In 2003, An Giang continued to provide funds for the construction of 35 clusters and 48 dykes to provide housing foundation for 13,155 households in a total area of nearly 244 ha and at a total cost of VND 225 billion Currently, there are 75 clusters and 122 dykes, the total number of housing foundation is 7061, and the total area is 802 ha in An Giang province, but the number of resettled households is only 4051

The study area is Tan Chau district, An Giang province Districts and communes surveyed were selected based on consensus between the researcher and local authorities In

An Giang province, one district from the most flooded area and with bank erosion was selected In Tan Chau district, two communes from the most flooded area and residential clusters were selected

Tan Chau district in An Giang province is an agricultural and upstream area of the Mekong Delta (Figure 1) It is heavily affected each year by floods and river bank erosion The district has abundant and diversified natural resources (land, water, aquatic products) and

a convenient geographical position with a good land and river transport network

Tan Chau district is in a position of strength and has specific characteristics compared with other districts in the province because it has the Vinh Xuong international port that facilitates both domestic and international trade With these potential advantages, Tan Chau is now developing rapidly and is one of the districts that plays important roles in the economy compared to other districts in the province and in defence security in An Giang Province The economy of the district is based on agricultural production, with an economic general growth rate of 10.9% per year, of which aquaculture accounts for 4.6% per year, and industry-construction accounts for 4.6% per year Traditional professions of the district include carpet weaving and carpentry The rate of poor households in the district (including prioritised target groups) compared to other districts is relatively high (1,094 households), approximate in 3.2% of the total poor households and households subject to preferential policy in the province Poor households in this situation have been issued with poverty certification and currently do not have a house or have only unstable housing Neither do the households who are subject to preferential policy like the Vietnamese Mother – Heroes

Figure 2: Administration Map of An Giang Province

Residential clusters and dykes have been invested and developed rather fast in Mekong Delta

provinces including An Giang, Dong Thap, Vinh Long, Kien Giang, Long An, Tien Giang,

Can Tho with a total cost funded by Government in 2001 – 2005 worth VND 3,200 billion

(Appendix 2) An Giang province received the most investment in the Mekong Delta with

residential clusters and dykes built in 11 districts and towns at a the total cost of VND 190

billion in 2002 (MOF, 2002) Although construction of residential clusters and dykes finished

in 2002 in An Giang Province, from 2002 to 2003 nobody moved into residential clusters and

dykes Until 2004, the number of households resettling in residential clusters and dykes was

still low After the construction period has passed, the residential clusters and dykes have

revealed many disadvantages such as resettling of households was still slow, toilets did not

function, clean water was lacking and the environment was polluted (Fforde et al., 2003:70)

There have been some previous studies about residential cluster and dyke in the Mekong

Delta such as AusAID, Care, Fforde et al The difference between their research and this

study is that they have a broader policy focus, while the focus of this study is a case study of

two communes with deeper household level analysis The Fforde´s study shows a range of

active interventions by the Vietnamese Government and international donors to support

measures to help cope with the extreme flood in Angiang, Dongthap and Longan provinces

AusAID analyzes the poverty situation in the Mekong delta including some provinces that

are influenced by annual flood Residential clusters and dykes is a relatively new program,

with different outcomes regarding the clusters and dykes respectively The residential clusters

and dykes have advantages and disadvantages for people The purpose of the research is to

understand the opportunities and constraints for people to develop stable livelihood in the clusters and dykes

The general objective of this study is to focus on the issue of understanding the impact of the residential clusters and dykes on the livelihood of the residents The specific objectives are to: 1) Analyze critically the process of implementation in resettlement

2) Understand the vulnerability context in relation to flooding before moving into the residential cluster and dyke

3) Understand household’s livelihood strategies within the vulnerability context before and after moving into the residential cluster and dyke

4) Compare household livelihoods before and after moving into the residential cluster and dyke

The following questions were investigated in relation to the specific objectives:

1) To analyze critically the process of implementation in the resettlement

• Who moved to the residential cluster and dyke? And who did not? Why?

• What was the process of implementation of resettlement?

2) To understand the vulnerability context in relation to flooding before moving into the residential cluster?

• What is the nature of risk in relation to flooding?

• Who are affected by these risks?

3) To understand household livelihood strategies within the vulnerability context before moving into the residential cluster and dyke

• What are households’ experiences in facing the risks of flooding in comparison to the multitude of risks they face?

4) To compare livelihood household before and after moving into the residential cluster and dyke

• How does the livelihood of households change when living in the residential cluster and dyke?

The research also has the following limitations:

1) This research did not have the objective of studying residential cluster and dyke in terms of architectural and construction techniques and was restricted to analyzing the livelihood issues existing in the residential cluster and dyke

2) The accuracy of the study was dependant upon primary data collected through interviewing small-scale households and key informants

3) The research mainly compared household livelihoods before and after living in residential clusters and dykes Some relatively limited discussions were held with households living outside residential clusters and dykes More research is needed to compare households living inside and outside at the same time

4) The impact of residential cluster and dyke construction is not evaluated regarding the natural environment If the natural environment is affected it will make the household’s living and livelihood unstable

2 REVIEW OF RELATED LITERATURE

This chapter reviews theories and the main concepts and related studies revolving around poverty and vulnerability, including floods, shocks, seasonality of employment opportunities, that are important for the livelihoods of people Comparing and contrasting these related studies, this chapter will discuss knowledge and ideas relevant to this study

2.1.1 Vulnerability

According to Brooks (2003), vulnerability means a set of conditions and processes that

determine the susceptibility of humans or human systems to be adversely affected by a shock

or hazard Also Brooks (2003) cited in Few et al (2004:18), describes vulnerability as a

condition of susceptibility shaped by exposure, sensitivity and resilience Folke et al (2002)

argue that the focus of the understanding of vulnerability is the concept of resilience The resilience of poor people represents their ability to withstand the impact of the trends and shocks (e.g (i) the former is defined as gradual environment degradation, oppressive political systems or deteriorating terms of trade, (ii) while the latter is sudden changes such as natural hazards, market prices) Resilience varies greatly from household to household even in one locality because it is determined by two characteristics of peoples’ livelihood such as the assets they possess and the services provided by infrastructure and institutions Vulnerability as defined by DFID (1999) stems from the negative external environment in which people exist such as shocks (e.g floods, droughts, storms), trends (e.g population, economic, resources), and seasonal shifts (e.g employment opportunities, prices, and production) This research uses vulnerability as a concept because vulnerability helps to understand the extent to which shocks/trends force people to change their livelihood Shocks force those who are exposed to transfer ‘prematurely’ into new forms of livelihood for survival Which new forms of livelihoods can households/individuals choose? It is clear that their choices depend on the resource base utilized by each household Their process of responding with choices will progress in a new vulnerability context where they also have to manage uncertainty and constraints

The word “livelihood” is used in many fields, but the term as used in the DFID ‘Sustainable Livelihood Guidance Sheet’ is understood as follows: “A livelihood comprises the capacities, assets and activities required for a mean of living A livelihood is sustainable when it can cope with and recover from stresses and shocks and maintain or enhance its capacities and assets both now and in the future, while not undermining the natural resource base” (Chamber & Conway, 1992)

Another definition was found in the guideline in FAO’s Corporate Document Repository, which suggests that “livelihood” does not just mean the activities that people carry out to earn a living It means all the different elements that contribute to, or affect, their ability to ensure a living for themselves and their household This includes: (i) the assets that the household owns or is able to gain access to- human, natural, social, financial and physical; (ii)

the activities that allow the household to use those assets to satisfy basic needs; (iii) the different factors that the household itself may not be able to control directly, like the seasons, natural disasters or economic trends, that affect its vulnerability; (iv) policies, institutions and processes that may help them, or make it more difficult for them, to achieve an adequate livelihood (FAO, 2002) It is clear that four elements in concert direct the household process

of making decisions, especially in choices of income generating activities in which the complexity, diversity, and the flexibility is reflected

In summary, the Sustainable Livelihoods Approach put the livelihoods of poor people at the centre of analysis and action It focuses on the main factors that affect poor people's livelihoods and the typical relationships between these factors It is a means of stimulating thought and analysis, and it needs to be adapted and elaborated depending on the situation This study is to focus on the influence of the process of clusters and dykes that affects how poor people create and protect a livelihood for themselves and their households

(i) Natural capital comprises the land, water, and biological resources that are utilized by people to generate means of survival (Ellis, 2000) It is clearly important to those who derive all or part of their livelihoods from natural resource-based activities such as farming, fishing, gathering in forests, etc (DFID, 1999) In term of natural processes (e.g Fires, flood, earthquake, seasonality, storms), there is a close relationship between natural capital and the vulnerability context in which many of the shocks devastate the livelihoods of the poor (DFID, 1999) Actually, in the case of Tan Chau district, there are two different kinds of shocks; both the floods and the shocks involved in the changed livelihood in the clusters and dykes What are the roles of which in households’ livelihoods?

(ii) Physical capital comprises what is created by economic production processes Buildings, irrigation canals, roads, tools, machines, and so on are physical assets (Ellis, 2000) Hence, a given state of infrastructure as well as physical property will bring households advantages or disadvantages

(iii) Human capital: At a household level, human capital is a product of the amount and quality of labour available with its education, skills, and health (DFID, 1999; Ellis, 2000) The poor have their own labour as key livelihood assets (FAO, 2002) Human capital is increased

by investment in education and training, as well as, by the skills acquired through pursuing one or more occupations (Ellis, 2000) Emphasizing education and skills, it is clear that gaining improvement in human capital is not easy and quick, especially to the peasants who are confronted by shock and risks In fact, both subjective and objective factors play a role in household decision making in terms of investment in education and learning a skill Accordingly, this choice by itself is a process of building/improving the household’s capacity

In the study site, what is the state of household’s human capital? In addition, what is its impact on household choices in terms of livelihood activities after resettlement? What are the households’ choices in investment in education and learning a new skill and which are the factors that influence these choices?

(iv) Financial capital refers to stocks of money to which the household has access This is chiefly likely to be savings, and access to credit in the form of loans (Ellis, 2000) Financial capital for household livelihoods sometimes is not only in form of money Each household converts it into forms of productive physical assets such as breeding facilities, mechanic, traffic vehicles, etc that depends on choices of investment What do they use it for and why will be examined in this study

(v) Social capital: Social capital was “defined by Moser (1998) as ‘reciprocity within communities and between households based on trust deriving from social ties’” (Ellis, 2000: 36) In the guidance sheets on sustainable livelihoods from DFID (1999), the importance of social capital seems to be considered as “resource of last resort” – a buffer that can help households to cope with a shock and “a safety net to ensure survival during periods of intensive insecurity.” Also, in social networks households develop knowledge and share that knowledge

In brief, a household bases its livelihood on the five types of assets above, that they own or can access, to build livelihood activities involving income generating income and improving capacity for their livelihood

residential clusters and dykes in the Mekong Delta

2.2.1 Development of migration in the world

There are different definitions and explanations for what compels people to migrate At one extreme there is involuntary migration This denotes extreme economic and often social harship, and is undertaken mostly by landless or land-poor, unskilled and illiterate poor labourers Here people do not have any choice of the place or type of work that they undertake Migration for survival is well documented in AP (Ramana Murthy, 1991; Reddy 1990; Rao,1994) Nearly all of the studies have identified the main drivers of migration as the worsening situation of dry land agriculture created by drought, crop failure and poor terms of trade More recently, the idea of migration as a coping strategy (Davies, 1996) is gaining acceptance This is migration that is integral to people’s coping, survival and livelihood strategies (Rao,2001; Conroy et al, 2001; Mosse et al, 2002) and not just a response to emergencies

Seasonal migration that is undertaken to improve the economic position of the household,

or accumulation migration, is also being noted by recent research in India For example Rao (2001) refers to this kind of migration in his study of Annathapur and Rayadurga districts in Andhra Pradesh Type 1 is migration for coping and survival Type 2 is defined as migration for additional work/income It takes place when the work in the village is over, normally after harvesting all crops Type 3 is migration for better remuneration or a better work environment or opportunity to use skills or acquire new skills They observe that there is a continuous transition between the different types Coping migration can become accumulative over time, as information improves, skills are acquired and relationships with employers stabilise Middlemen may be eliminated and migration becomes an altogether remunerative option for the poor

An associated question that has often been addressed in the literature is whether there is evidence that migration can reduce poverty This is not easy to measure or understand because poverty levels are not constant and the effects of migration can not be isolated

On the other hand, Kothari’s (2002) review of migration studies finds that migration can both reduce and perpetuate poverty

BBS (Bangladesh Bureau of Statistics), 1998-1999, argues that migration contributes to the livelihood of the poor It is assumed that the extreme poor people are more likely to migrate

to other parts of the country Studies have shown that displaced people initially try to relocate themselves within the village, then in neighbouring villages and gradually move to urban areas when no other livelihood option is available in their known rural surroundings Initially, they try to earn a living from labour and other off-farm activities, but once they are in an urban setting they look for jobs in different parts of the informal sector (Mamun, 2003)

2.2.2 History and development of settlement in residential clusters and dykes in Mekong Delta

According to Diep Dinh Hoa (1998), the resettlement in the Mekong delta has aimed at the stabilisation of people ‘s livelihood in the short term and hoped that this stabilisation would give a knock–on impulse to subsequent development But the organisation of migration has not yet been taken into account during the process of resettlement It may have advantages at first, but gives rise to many problems during the process of long term settlement John Soussan, et.al 1999, argue that in order to succeed in resettlement, it needs the participation

of people in land use planning and needs to have a good organisational structure in local government If so it can lead to sustainable livelihood improvement with both increased incomes and a major growth in the assets owned For example, top-down planning for building up residential clusters in Vinh Hoa village of An Giang province failed as a result of

no local people’s participation at its early stage This lead, not only to a waste of financial resource but to failed resettlement

According to Huynh Lua (1987), although the settlement history of the Mekong Delta began centuries ago, the major influx of people to the area occurred after the land was opened for agriculture development in 1975 when the war ended Dating back to the first days of settlement, the popular living areas of the Mekong Delta dwellers were high strips of land along the river, where there was soft land and fresh water for living and cultivation (Fforde et al., 2003) When doing research on socio-economic and environment development in flooded areas of the Mekong Delta, Dao Cong Tien (2004) emphasizes that residents dwelling

in Mekong Delta belong to three types of groups: (i) those on dykes, along rivers, canals and ditches were the most common types In the past couple of years, this group includes those who live in residential clusters and dykes along the rivers or canals, which were constructed

by the local government based upon people’s hope In recent years, since the land capital for building up such clusters was limited, the so-called top-down policy of local government for resettlement, launched without local people’s participation has become a major concern for

those in clusters, grouped on higher land, were also relatively common, (iii) the third was in dispersed places, in jungles or in the middle of fields, with the houses built right on the family field or on an unclaimed plot of land Dao Cong Tien (2004) also showed that Clusters and Dykes were two main forms of dwelling in the Mekong Delta in general historically and for residents in the flooded areas in particular such as (i) the cluster was the centre in order to develop with natural, socio-economic events, along with demand and accessible capacity of popular living in flood areas (ii) the residential areas on the dykes were more dispersed as they are narrow and extended There was little land for cultivation, low quality of technical infrastructure, waste collection and sanitary quality was low in comparison with the clusters and the dwellings were separated from public services such as health services, schools and

markets as compared with the clusters However, people were interested in living in residential dykes rather than in residential clusters because they wanted to live close to rice-fields, canals, and rivers for parking small boats and could cultivate home gardens and raise animal husbandry to increase their income (Dao Cong Tien, 2004)

2.3.1 Support for recovery from flood disasters

The case of Bangladesh shows that people’s livelihood after the flood changed when they migrated and resettled, especially in the case of poor people This was focused on the changes

of livelihood assets and the use of livelihood assets to cope with vulnerability (Beck, 2005) Credit is not yet used as a major tool for livelihood development in the residential clusters and dykes but international experience suggests that the role of credit is very important for

credit can provide the poor with the means of support against shocks and to build up and diversify their livelihood activities

Omer Hayati, Abed Elbagi & Nurain Khartoum (2005) argue that if credit is to fulfil its social objectives of bringing financial services to the poor, it is important to know the extent

to which its wider impacts contribute to poverty reduction The impact of credit on the livelihoods of poor people through social networks that contribute to poverty reduction needs

to be examined Social networks play an important part in helping clients escape from poverty Access to social networks provides clients with a defence against having to sell physical and human assets and thereby protects household assets

Christoplos citied in the Tsunami Evaluation Coalition (2006) showed that the primary concerns of disaster affected people include assests, occupation structure, production scheme, shelter, etc People need help to rebuild their lives Dang Quan Tinh & Hang (2003) argue that the poor experience the greatest difficulties in recovering due to five main factors: (i) the poor have houses of poorer quality, which therefore suffered greater damage than the houses belonging to the better-off (ii) The poor used more resources to repair and strengthen their houses in proportion to their total resources (iii)The poor had a higher degree of production failure in animal husbandry due to diseases caused by lower standards in sanitation and less production knowledge (iv) The poor have a less diversified household economy and are more dependent on paddy production (v) The poor often have lower labour capacity and poorer health conditions

floods improve water and soil quality, kill insects and diseases on plants and livestock and feed underground water (iii) floods create great aquatic resources owing to a more favourable water environment, food sources and abundant aquatic species.(iv) Floods together with rivers and the sea create a system of interlacing canals and ditches and waterlogged ecological areas that makes river traffic easy However, floods also have major disadvantages and pose multiple risks to humans, including (i) flood affected markets, households, and individuals, particularly the loss of agriculture production, reduction in employment opportunities and loss of assets, which cause major reductions in household incomes and wealth (ii) the Mekong Delta suffered 407 deaths and around VND 2,284 billion (USD 147 million) of property loss in the

1994 flood 217 deaths and property loss of VND 2,182 billion (USD 141 million) in the 1996 flood and 453 deaths and property loss of VND 4,000 billion (USD 258 million) in the 2000 flood

Fforde et al., (2003) argues that government policy for construction of residential clusters and dykes in the Mekong Delta is a kind of flood-avoiding program This study shows that people‘s living in residential clusters and dykes creates a safe place for them but reduces the opportunities of their livelihood

2.3.3 How the household’s traditionally managed flooding, and

Traditional, low technique, coping and flood management strategies to enhance safety and incomes are used in the area People know how to cope with flooding to maintain seasonal crops, how to plant ‘floating’ rice, how to exploit the aquatic sources, how to build small scale border embankments to protect the summer-autumn crop and how to evacuate to high grounds during big floods (Dang Quang Tinh & Pham Thanh Hang, 2003)

There were several reasons why flood damage was reduced from 2001 to 2005 First, the levels of flood were reduced because there were great changes in the awareness level of both local authorities and residents from big losses to small losses from flood prevention to exploitation of floods Second, the Government and local people had the experience gained from the two large floods in 2000 and 2001, such as seeding Summer-Autumn crops earlier

flood kindergartens was created to protect children during the flooding period Besides, the Government of Vietnam decided to build houses in higher parts of the flood areas and encourage people to resettle in these residential clusters and dykes areas The idea of residential clusters and dyke was to make people's lives more stable by providing a safe home (Noguchi, 2003) The residential clusters and dykes program has played a role in reducing vulnerability to floods, promoting education and improving health But living conditions in the residential clusters and dykes have not shown signs of achieving sustainable livelihood To

my understanding, an unstable livelihood means that they lack jobs, or their income is unstable, which increases vulnerability because of less buffers to respond to shocks

3 RESEARCH METHODOLOGY

In this chapter, the research methodology consisting of site and respondent selection, methods

of data collection and data analysis is described

An Giang province is one of the areas most heavily flooded and affected by river bank erosion

in the Mekong Delta It is also one of the main agricultural production areas in Mekong Delta It covers 353.551 hectares, of which 281.862 hectares are devoted to agriculture and 264.284 hectares to paddy cultivation (statistical yearbook, 2005) It includes eleven districts and one city with over 2.2 million people Besides providing agricultural products for the domestic market, it mainly contributes to exported agricultural products such as rice, aquaculture, and fruits The Mekong Delta percent contribution as part of the total in Vietnam, accounts for 10 percents of GDP, 50 percent of rice production, and 52 percent of aquatic production It occupies 90 percent of exported rice, and 60 percent of exported aquatic products Although An Giang is the granary of rice of Mekong Delta, the poverty rate

is still over 13.15 percent based on the official poverty line of MOLISA in 2006

3.1.1 Selection criteria of sites

The study area is Tan Chau district, An Giang province Districts and communes surveyed were selected based on consensus between the researcher and local authorities Criteria for

districts that are effected by severe damage from flood and river bank erosion, (3) If there are many clusters and dykes in communes, the cluster and dykes where people have lived for over one year should be chosen, (4) the number of residents in the selected clusters and dykes was relatively high compared with other clusters and dykes in the District, in order to have enough sample size for the survey

In An Giang province, the program of building residential clusters and dykes has two models: The first is residential cluster and the second is residential dyke Currently, there are

75 clusters and 122 dykes, the total number of housing foundation is 7061, and the total area

is 802 ha in An Giang province The number of resettled households is however only 4051

In this study, a residential cluster in Vinh Hoa and a residential dyke in Tan An were chosen

to represent the residential clusters and dykes in Tan Chau district (Appendix 12) because they fulfilled the above criteria These communes are severely affected by flood and bank erosion and the clusters and dykes were constructed among the earliest (2002) of the clusters and dykes in Tan Chau district;

In Tan Chau district, the construction of the residential clusters and dykes finished in midyear

2002 After that the local authorities have been moving people to these areas At the end of

2002, only a small number of people had moved to these residential clusters and dykes and thus there was not enough sample size for household survey (See Table 2) Therefore, the researcher decided to choose the households moving in 2003 for understanding their

livelihood before and after living in the residential clusters and dykes The reasons to select

the interviewed households are: (1) these households are considered to be the first households

living in the residential clusters and dykes and they have some experience of living in the new

place; (2) the time is quite enough to understand people’s livelihood changes This study did

not select a sample from the household groups moving in 2004 and 2005 because these

households did not have enough living time in the residential clusters and dykes for the

Finishing construction

No of housing foundation

Years of moving

Number of resettlement

Town

Source: People’s Committee of Tan Chau district, 2006

The target groups who moved into the residential clusters and dykes were: (1) households

subject to preferential policy; (2) poor households fleeing from floods, and (3) poor households fleeing from river bank erosion Formerly, the places they lived were flooded to at

least 1.2 m, and many of them lived along the Hau River’s “third or fourth of canals” or

other remote areas and did not have a stable house

The residential clusters were built at about 2-3 km distance from their old houses, and 4-5

km distance for the dyke The distance between the old place of residence and the new place,

meant that when the farmers resettled in the residential clusters and dykes they reduced their

access to off-farm activities Also, there was a sudden change of lifestyle from a rural

orientation to a semi-urban that they often found very difficult to adjust to

Tan Chau is an upstream district affected by annual floods It is one of the deeply submerged

districts in An Giang province It has many clusters and dykes, constructed for flood

protection (Figure 3 and 4)

Figure 3: Flooding map of An Giang Province, 2000

Source: Department of irrigation in An Giang province, 2000

Figure 4: The project of construction of secure houses in the condition of regular flooding in An

Giang in the period of 2001 - 2005

Source: Department of construction in An Giang province, 2000

Tan Chau district was selected because it is one of the districts in the most flooded area

and because it is not a politically sensitive area in terms of national security and defence So far

there has been no specific research on people’s livelihoods in the residential clusters and dykes

in the area The aim in this research is to study the impact of the residential clusters and dykes

in the study area for people’s livelihoods The total natural land area is 16110 ha, with a of

population 161685 people; Tan Chau District consists of 9 communes and one town Lying

the country Cambodia; to the east to An Phu and Chau Doc districts; west to Dong Thap

Province ; and to the south to Phu Tan district (Figure 5)

Tan Chau (2002)

Chau Doc

(2003)

Cho Moi (2003)

Long Xuyen (2005)

Tinh Bien

(2004)

An Phu (2003)

Figure 5: Location of study site

3.2.1 History of moving into residential clusters and dykes

Tan Chau lies between the Tien and Hau Rivers, upstream in An Giang province It is affected by the annual floods of the Mekong River When the floods come, the wet land and river erosion caused by the floods heavily affect the livelihoods of local people According to the Provincial Board of Flood Prevention and Control, the biggest concern is the river-bank erosion According to the statistical yearbook, from 1982 to 1999, there has been more than

new places, causing a property loss of over 90 billion VND and great disadvantages for local people In the big flood in 2000, there were five occasions of erosion only in one year, close

to the Administrative Center of Tan Chau District, collapsing 115 houses and forcing 35 other households to re-allocate urgently This flood caused the damage of over 11 billion 500 million VND From 2001 to 2006, the erosion area has been 390 hectares, the number of households that have had to move away have been 1000 Especially in one year, from 6/2005

to 6/2006, the river bank erosion was 7920 meters long including one state work/project in Tan Chau District The floods are annual but the impact varies between big and small floods Big floods have a large impact on river bank erosion causing deaths and property damage, and affects livelihoods in this place, but this is not the case with the small floods In such an unpredictable situation, the budget for building a high dyke against floods and river bank erosion has been limited Besides, An Giang Province does not have much unused land to allocate to the households as new economic zones, there are only the residential clusters and dykes In this study, the reason to select one cluster and one dyke is to compare the livelihood between the two models (See table 2)

Study area

Residential cluster

at Vinh Hoa Center

Residential dyke

at canal Tan An

Figure 6: Location of selected residential cluster and dyke in Tan Chau district, An Giang province

3.3.1 Physical characteristic and economic status

Vinh Hoa commune is located about 15 km north-west of Tan Chau district centre and is accessible by roads The population in 2004 was 11,882 people, and the population density was 578 people/ km2 which is 4.5 times higher than the overall population density of Tan Chau district At present, there are 326 poor households in the commune (14.6%) There are four clusters and dykes in Vinh Hoa commune In 2005 production activities in Vinh Hoa commune consisted of around 68 percent of household income from crops, 10-11 percent from livestock production, 6-7 percent from rural services, 16 percent from aquaculture Agriculture plays a vital role in the local economy with 87 % of the households depending on

it for their livelihood 30 percent of the households have their own land for farming and 70 percent of the households make a living from farm labour on other people’s land

In 2006, the production value of agriculture made up 72 percent of the total local GDP (source: local authority - key informant interview) Vinh Hoa commune has plans to expand the development of crops and aquaculture as well as to improve product quality and to lower the costs of production It is also planning to build specialized high-quality raw material areas

to serve the agro-business, and increase the export value of food including vegetables, beans and frozen seafood Apart from rice, efforts will be made to develop aquaculture (basa, tra and lobsters) while carrying out large-scale commodity-based livestock production, with special focus on beef cattle, dairy cows, pigs

3.3.1.1 The residential cluster at township Vinh Hoa centre

The residential cluster at Vinh Hoa township is located about 13 km north-west of Tan Chau district centre and is accessible by roads In the 3.3 ha of cluster area there are 143 housing foundations Construction started in July 2002 and by now the basic foundations and house construction in the cluster have been completed The number of resettled households up to now is 141, and 98% of the total foundations are occupied All of the resettled households are

of Kinh ethnicity and make their living from agricultural practices There are 30 percent households have their own land for farming and 70 percent households make a living from farm labour on other people’s land

3.4.1.1 Physical characteristic and economic status

Tan An commune is located about 10 km north-west of Tan Chau district centre and is accessible by roads and waterways using boats The population in 2004 was 20,590 people and the population density was 813 people/ km2 which is 5.2 times higher than the overall population density of Tan Chau district At present, there are 260 poor households (9.23%) There are 5 dykes in Tan An commune In 2005 the production activities in Tan An commune consisted of around 70 percent of household income from crops, 15-16 percent from livestock production, 5-6 percent from rural services, 20 percent from aquaculture Agriculture plays a vital role in the local economy with 89 % of the households depending on

it for their livelihood In 2006, the production value of agriculture made up 75 percent of the total local GDP (source: local authority - key informant interview)

3.4.1.2 The residential dyke at Tan An Commune

The residential dyke at Tan An commune is located about 9 km north-west of Tan Chau district centre and is accessible by roads and waterways using boats The area consists of 11.46

ha and the number of housing foundations is 263 Construction started here in June 2002 and

by now all the basic foundations and house constructions in the cluster have been completed The number of resettled households up to 2006 is 255 households and 97% of the total foundations are occupied

3.5 The difference between residential clusters and dykes

• A residential cluster is a concentrated residential area with a size of 2-3 ha for from 100 to

200 households, adjacent to rice fields, appropriate for habitation and livelihood activities and equipped with public welfare facilities (According to documents issued by the Government (2002)

• A residential dyke is a residential area established and based on a main canal, a canal with first grade or main roads (According to documents issued by the Government (2002)

• Residential Cluster, according to Dao Cong Tien (2004), is a concentrated residential area

constructed to avoid floods It has a basic infrastructure system including electricity, roads, medical station, school and market People living in clusters are poor; their jobs come mainly from hired non-farm activities (e.g wage labour, hired labour (sand for building), hired labour, lottery seller, hair cutter, vendors and small traders They only have a house and no land for activities such as cultivation or animal husbandry

• Residential Dykes are dispersed, narrow and extended along a canal They are close to cultivation, rice-fields, fruit gardens, fishponds, livestock-houses or parking places for small boats People living on dykes are poor Their jobs come from on-farm (hired agriculture such

as rice harvesting, pesticide spaying, land preparing, bean harvesting) and off-farm (e.g natural fish catching, natural vegetable collection) They have a small area of land for animal husbandry and crop cultivation

Dao Cong Tien (2004) found that people are more interested in living in residential dykes

as compared to living in residential clusters because they prefer to live close to rice-fields, canals, and rivers for parking small boats Besides, staying at the residential dykes; people can cultivate in their home garden and raise animal husbandry to raise their income

Although people prefer living in residential dykes rather than in residential clusters, the Government has tended to emphasise the development of residential clusters in order to manage people, to invest in infrastructure and to develop the socio-economy more easily in these areas

This study applied a conceptual framework of the vulnerability context to the process of resettlement The context of vulnerability comes from different issues such as: floods, shocks, trends, seasons and the changed policies The study looks at the vulnerability of people in Tan

An and Vinh Hoa communes in this context The sustainable livelihoods framework identifies five types of livelihood assets, which consist of human capital, physical capital, natural capital, finance capital, social capital, helping people to meet their needs In the context of this study the assets are; (1)

Natural : floods, water resources, (2) Physical: electricity, roads, schools,

(5)

Social: community, family, social organizations

Analyzing livelihood strategies involves understanding how people use and combine their resources to meet short-term and long-term needs It also involves understanding how people cope with floods The analysis focuses on the perception of households

An understanding of the livelihoods of different groups of people is also important for planners/ policy makers to identify programs and activities to improve livelihoods by increasing alternatives in order to reduce risks Strategies can include combinations of

activities such as on-farm (hired labour agriculture, animal husbandry, cultivation), off-farm employment and non-farm (small traders, wage labours, vendors) The study analyzes household contexts and status of their natural resources and available experiences that the household used to ensure their best livelihood strategies Besides, current issues on how the households face problems and expectation of improved livelihood strategies are investigated What did they do to be able to overcome constraints/vulnerability/risk?; How did they solve their own problems?; who supported them?; what is the process of resettlement implementation?

In addition, the study analyzes livelihoods before and after moving to the residential cluster/dyke The study tries to show factors that influence the household resettlement process The above is analyzed based on the achieved results from practical field work on the impact of residential cluster/dyke program (Figure 7)

Figure 7: Schematic presentation of the research

Source: Adapted from DFID, Pham Xuan Phu, 2006

3.7 Data collection

For this study the following data was collected :

− Household information such as: age, education, manpower, gender of main labour

− Household occupation, income, household expenses and employment (Occupation meaning people who have a specific job (e.g bricklayer, barber, teacher), Employment meaning

whether people are working or not (e.g they have jobs)

− Living conditions in the residential clusters/dykes such as: housing foundations, houses, clean water, semi septic toilets, electricity, living environment

− Household strategies to deal with the annual flood

The data collection methodology was a combination of qualitative and quantitative methods and included the following: The primary data was gathered in four steps In the first step, general information was gained from key informants and group discussions with the Management Board of the residential cluster and dyke and Tan An and Vinh Hoa commune People’s committees In this step, the available documents and reports of the study site were also useful references to the history of the area In the second step, two groups, people who were living inside and outside of the resettlement area, were invited to work with Participatory Rural Appraisal (PRA) In the third step, a questionnaire was used for household interviews Finally, the fourth step was in-depth interviews which used a reference checklist

have occurred (Nabasa et al., 1995) In this study, a time line tool was used to record changes

over a period of the historical development of residential cluster and dyke; and the history of the local community It was used with a group of key informants and older members of the community The results from this can be seen in Appendix 4

These are diagrammatic representations of different activities, problems and opportunities which occur during different times of the year and which have an effect on people’s lives showing the seasonality of agricultural and non agricultural workload, food availability, human diseases, gender-specific income and expenditure, water, forage, credit and holidays and so on (Chambers, 1993) In this study, the seasonal calendars were used to relate

livelihood activities such as employment, production and income The results from this can

be seen in Appendix 15

This tool shows institutions, organizations, groups and important individuals found in the village as well as the villagers’ viewpoints on these In addition, the diagram explains who participates in these groups in terms of gender and wealth The Institutional Relationship Diagram can also indicate how close the contact and cooperation between those organizations and groups are (Chambers, 1993)

In this study, a Venn diagram was used to describe the administrative structure of the villages The participants were asked to list the names of institutions, which play an influential role in the community and to rank them according to importance After identifying the importance of each institution, the respondents were asked to identify the relation and the impact of the institutions in the community The results from this can be seen in Appendix 5

This technique can be used either with items important to the community such as crops or with problems perceived by the community In either case each item is compared to others

and assigned a ranking (Nabasa et al., 1995) Particularly, it is useful in discovering and

prioritizing problems presented by groups of farmers It can also bring out the perceived seriousness of farmers’ problems In this study, this tool was used to assess problems/challenges

of the livelihood in the resettlement villages (Appendix 6)

SWOT analysis is a tool used for analysing an organization and its environment It is the first stage to help planners focusing on respective key issues of the area SWOT stands for strengths, weaknesses, opportunities, and threats Strengths and weaknesses are internal factors Opportunities and threats are external factors (Hillson, 2003) In this study, SWOT Matrix will help identify some strengths, weaknesses, opportunities and threats considered to be important by the local people in the village The results from this can be seen at Appendix 14

The questionnaire is based on the discussions with the commune leader regarding the characteristics of the economy-society-environment of the sites Then the questionnaire was pre-tested with 10 households before the actual interviews The interviewers included the students of the Department of Rural Development, the staff of the Agriculture and Natural Resources Faculty of An Giang University, and the researcher

b In-depth interviewing

After the PRA exercise and household interviews were finished, the researcher identified main issues that needed to be studied through more in depth-interviewing This tool used the checklist focusing on livelihoods

3.9 Data analysis methods

Data analysis methods used the qualitative and quantitative methods including: (i) the results from PRA are presented in tables, figures and diagram The information is cross-checked by triangulating, (ii) the data from the households surveys was analyzed by computor software systems The software used for inserting and analyzing data includes Excel and SPPS 13.0 (Statistical Packages for the Social Services)

Figure 8: Research Diagram

Source: Pham Xuan Phu, 2006

Research objectives Selection study areas Data acquisition

Group discussion

- Natural resource

- History of community

- Farming constraints & opportunities

- Traditional livelihood in relation

- Policy relations residential clusters

- Annual reports of local office

Interview

4 RESULTS AND DISCUSSION

To understand the livelihoods of households in the residential cluster and dyke, the differences in the influence of the resettlement program on people before and after resettling investigated The results from PRA exercises and the household survey are presented in this chapter In particular, the PRA exercises provided the data relating to households and the resettlement process A SWOT was conducted to study the perceived strengths, weaknesses, opportunities, and threats in the study site The household survey provided information about the main characteristics of households and the living condition of people such as clean water, toilet and electricity Moreover, factors having influence on resettlement such as employment, occupation, income, were identified Finally, these results were combined to explore livelihoods of people who live in the residential cluster and dyke

district

According to Mr Do Thanh Chung, the Management board of residential clusters/dykes in Tan Chau district, support was needed every year for poor people in deeply flooded areas When there were no residential clusters and dykes, the local government had support programs for the poor such as loans of 3- 5 million/households to upgrade the houses in order to be safe during flood seasons However, he claimed that they do not use the loan for this purpose Instead, they spend money buying television, cassette player or drinking Therefore, the program does not have a good impact for local people whereas their debts keep increasing Seeing that situation, the local government suggests to the government to plan support programs for these people Scientists and the government have reviewed the damage in terms of human loss and damaged property after the big flood in 2000 and discuss about it Only 2001 did the government issue decision 256/TTg about building residential clusters for poor people in deeply flooded areas Mr Do Thanh Chung says that there is no participation of local people when selecting places to build residential clusters and dykes It is very difficult to get their attendance as they live scattered in deeply flooded areas As a result

of that, it is the government that decides places for building residential clusters and dykes if it thinks those places are suitable The PRA results (see table 3) show that local people do not know anything about the places to build residential clusters and dykes; they just follow the instructions from the government which asks them to move into new places This finding is similar to that of the Care report (2003), which suggests that the interviewed resettled households said that they attended a meeting in which the residential cluster dyke construction plan was disseminated, but they did not understand it They only heard and did not know what the construction design looked like as the officer did not present illustrative pictures or documents In the meeting they were interesting in how to get a loan for buying a

these places and in case they do not have money to pay for housing foundation in six years, they will come back to the old places It can be seen here that the government and local people have not had an agreement with each other Yet many researches show that the

resettlement is successful when there is participation of local people (Diep Dinh Hoa, 1998; John Soussan, Anjan Datta and Alexandra Clemeltt, 1999; CARE International in Vietnam,

2003; Noguchi Chitose, 2003)

Table 3: Local people’s participation in planning and implementation of the residential clusters and

dykes program

Do you know the places which the

government decides for building residential

clusters and dykes?

100% of responses do not know There is no participation of local people when selecting places

to build residential clusters and dykes Have you participated in planning the use and

building in residential clusters and dykes?

100% of responses have not participated Source: PRA exercise, Pham Xuan Phu, 2006

in the district

4.2.1 Approval process of resettlement in the residential clusters and

dykes

From the PRA and group discussion results (see figure 9) we know that resettlement process

has been implemented according to the following flow chart:

Figure 9: Approval process of resettlement in the residential clusters and dykes

Source: PRA exercise, Pham Xuan Phu, 2006

According to Mr Do Thanh Chung, the Management board of residential clusters/dykes

in Tan Chau district, the members of the approval Council include: (1) representative of the

Communist Party organization, (2) representative of local authorities, (3) representative of

People’s Council and (4) representative of mass organizations including: Fatherland Front,

Women’s Union, Farmers’ Association, Youth Organization and Veteran’s Association The

aim of this process was to ensure implementation according to the criteria set forth by the

Government However, the process also complicated the administrative formalities, with the

inclusion of a six step procedure, and it took from three to six months from household

submission of application to the signing of a contract with the Bank This was also one of the

main reasons that households were tardy for resettling in residential clusters/dykes though

housing foundation and houses had already been built (CARE, 2003)

Commune committee approved these applications

District committee approved these applications

4.2.2 Approved households for resettlement in the residential clusters and dykes

An Giang People’s Committee promulgated the Decision No 1706/2002/QD-UBND about the policy of purchasing house foundation and house on credit in residential clusters and residential dykes The target of this policy included preferential groups (Target group No 2.1, 2.2, 2.3 and 2.4 in Table 4) and other target groups, of which, the preferential groups had priority in being approved for resettlement If there were vacant foundations left, households

of other target group in need of foundations would be considered by the Commune PCs

Table 4: Approved households for purchasing house foundation and house on credit

Target group No 2.1

Households subject to preferential policy, poor households fleeing from floods and bank erosion (with poverty certification)

Target group No 2.2

Households undergoing removal from the site due to land clearance who are

in need of resettlement

Target group No 2.3

Households subject to preferential hardship policy (without poverty certification), without housing foundations for overcoming floods and having only temporary accommodation (through discussion and election)

Target group No 2.4 Households of target group 2.1 but on poverty line (through discussion and election)

Other target groups

Aside from the four above mentioned target groups, if there are vacant foundations left, households in need of foundations shall be considered by the Commune PCs The eligibility would be based on hamlet or village proposals and shall be reported to the People’s Council, Fatherland Front and other mass organizations in a transparent and public manner

Source: Management board of residential clusters/dykes in Tan Chau district, 2005

2700/2005/QD-UBND about revising and supplementing the target of purchasing house foundation and house on credit in residential clusters/dykes (Table 4) Poor households of this policy had to be identified in terms of the poverty line of MOLISA According to MOLISA

2005, the current poverty line for the 2001-2005 period is divided into three levels: for people in urban area it is 150,000 VND per month, in rural area it is 100,000 VND per month and 80,000 VND for those who reside in mountainous regions and islands The above decision mentions the preferential households and the other target groups The preferential households comprise four target groups (Table 4) in which target group 2.1 has the first priority in being approved of moving into the residential clusters and dykes Meanwhile, the three remaining target groups will be approved later

However, 18 percent of households interviewed reported that their income was above the poverty line but they were still approved to be resettled (Table 5) The reason was the limitations of local officials in evaluation and identification of poor households by (AusAID, 2004) They did not understand the poverty status for every case in their locality Even some local officials did not know about the poverty line that was applied to their commune Besides, some local officials display arbitrary power by ignoring the disadvantaged groups in communities and replacing them with medium households close to the local officials The list

of poor households identified in the poverty assessment process was made based on the