master thesis in rural development Master Thesis No 40

Master Thesis in Rural Development with Specialization

in Livelihood and Natural Resource Management
issn 1403-7998

Factors Influencing Access to Credit of

Households in Rural Areas of Vietnam
Case study of Tan Linh Commune, Ba Vi
District, Ha Tay Province

Nguyen Thi Thu Phuong, Hue city, Vietnam

Department of Urban and Rural Development

Swedish University of Agricultural Sciences
Factors Influencing Access to Credit of
Households in Rural Areas of Vietnam
Case study of Tan Linh Commune, Ba Vi District, Ha Tay
Province

Nguyen Thi Thu Phuong, Hue city, Vietnam-2006

Master Thesis in Rural Development with Specialization in Livelihoods and Natural

Resource Management

Master Thesis No 40 | Hue City, Vietnam | September 2006 | ISSN: 1403 7998

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

ABSTRACT
This thesis aims to explore the factors influencing access to credit because a sufficient
access to credit can improve agricultural production condition of households in rural areas.
The terms and conditions under which credit is transacted may vary substantially across
different transactions, depending on the characteristics of the borrower and lender and the
relationship between them. Consequently, in order to improve access to credit of rural
households, factors of borrowing level of households need to be studied.
For this purpose, a fieldwork was conducted in Tan Linh commune, Ba Vi district Ha
Tay province with the objectives to examine (a) the credit demand and factors of
borrowing level of rural households, (b) the behaviour of a formal lender in rationing
credit, (c) the influence of credit on rural household production. Data collection methods
used were a household survey, group discussions, key informant interviews and data and
information from business statistics as well as other secondary sources provided by the
district and commune branches of banks.
The results show that there exists a large demand for credit by the local farmers,
especially those portions of the amount needed but not applied for, due to risks and
uncertainties particularly for fear of rejection by formal lenders. This mind set among loan
and non-loan borrowers were found in all household groups but was most prevalent
among poor households. Farmers were rational in deciding which sources to ask for a
particular kind of loan. The amount of credit applied for by households to VBARD,
dependency ratio, number of sick persons, owned farming area, the ownership of Land
Use Certificates (LUCs) are factors that influence credit rationing by VBARD. Credit
status influences on agricultural production, particular for livestock production.

2
ABSTRACT IN VIETNAMESE
Nghiên cứu này nhằm khảo sát các nhân tố ảnh hưởng đến việc tiếp cận tín dụng của hộ
bởi một sự tiếp cận thích đáng có thể góp phần cải thiện điều kiện sản xuất của hộ nông
nghiệp ở các vùng nông thôn của Việt Nam.
Những điều khoản và quy định của một hợp đồng tín dụng biến đổi đáng kể giữa các
giao dịch tín dụng khác nhau, phụ thuộc vào đặc điểm của từng người vay, từng người
cho vay và mối quan hệ giữa họ. Kết quả là để tăng cường tiếp cận tín dụng của các hộ
gia đình ở nông thôn, các nhân tố ảnh hưởng đến mức độ vay mượn của hộ bắt nguồn từ
cả hai bên: người vay và người cho vay cần phải được nghiên cứu.
Để tìm hiểu vấn đề này, một khảo sát thực địa đã được tiến hành ở xã Tản Lĩnh,
huyện Ba Vì, tỉnh Hà Tây với các mục tiêu cụ thể sau: (a) nghiên cứu nhu cầu tín dụng
và các nhân tố ảnh hưởng đến mức độ vay mượn của hộ, (b) nghiên cứu phương thức
cho vay của một tổ chức cho vay chính thức trong việc hạn lượng tín dụng, (c) nghiên
cứu ảnh hưởng của vốn vay đối với sản xuất của hộ. Các phương pháp được sử dụng để
thu thập số liệu bao gồm điều tra hộ, thảo luận nhóm và phỏng vấn cá nhân. Bên cạnh
đó, số liệu thứ cấp về hoạt động kinh doanh của ngân hàng cấp xã, cấp huyện, các báo
cáo kinh tế xã hội hằng năm của xã cũng đã được thu thập và phân tích trong luận văn.
Kết quả nghiên cứu chỉ ra rằng hiện nay người nông dân xã Tản Lĩnh có nhu cầu lớn
đối với tín dụng cho cả mục đích sản xuất và tiêu dùng trong đó đáng phải quan tâm đến
lượng tín dụng thực tế người dân cần nhưng đã không nộp đơn để xin vay do lo ngại rủi
ro trong sản xuất và sợ bị ngân hàng từ chối. Vấn đề này đã xuất hiện ở cả hộ vay và hộ
không vay trong cả ba nhóm hộ khá, trung bình và nghèo tại địa phương nhưng nó đặc
biệt phổ biển đối với những hộ nghèo. Người nông dân suy xét có lý trong việc quyết
định nguồn tín dụng nào họ xin vay cho mỗi một mục đích sử dụng vốn vay cụ thể.
Lượng tín dụng hộ nộp đơn xin vay, tỉ lệ người phụ thuộc, số người ốm, diện tích đất
canh tác, và vấn đề sở hữu giấy chứng nhận quyền sử dụng đất là những nhân tố góp
phần ảnh hưởng đến quyết định hạn lượng tín dụng của ngân hàng Nông nghiệp và Phát
triển nông thôn. Mức độ thoả mãn nhu cầu tín dụng (lượng tín dụng thực tế được vay so
với nhu cầu) có ảnh hưởng quan trọng đối với sản xuất nông nghiệp, cụ thể là cho hoạt
động chăn nuôi.

3
ACKNOWLEDGEMENT
First of all I want to thank the people of Tan Linh Commune, the site of this study whose
patience and willingness to share their life stories and time made this study possible. More
importantly, the four months time living with them were significant days in my life that
led me to understand more the real value of life.
I would like to express my deep gratitude to my supervisor Assoc. Prof. Ranjula Bali
Swain, for her professional advice and constant encouragement especially for the time that
I was sick. Also, my profound gratitude goes to Dr. Pham Bao Duong – my supervisor
whose guidance, moral support and understanding put much knowledge, experience and
ardent efforts in my thesis. Likewise, my gratitude goes to Dr. Minh Ha Fagerstrom for
her constant encouragement both in my academic and personal life.
I am especially grateful to RDViet Project and the people behind it - Dr. Britta Ogle,
Assoc. Prof. Le Duc Ngoan, and Ms. Huynh Anh Phuong - for their help and financial
support which enabled me to come to Tan Linh Commune to do the study and enrich my
experience.
To my classmates who shared their friendship and life during our stay in Hue City.
Special appreciation goes to Mr. Tu and Ms. Suong for their help and friendship.
Lastly, my endless gratitude and appreciation to my parents, my aunt, and my younger
brother – for tirelessly supporting and believing in me…THANK YOU!

Nguyen Thi Thu Phuong

Vietnam, 2006

4
LIST OF CONTENTS
ABSTRACT
ABSTRACT IN VIETNAMESE
ACKNOWLEDGEMENT
1 INTRODUCTION...........................................................................................................................................................8
2 LITERATURE REVIEW ..........................................................................................................................................10
3 STATE OF RURAL FINANCE IN VIETNAM.........................................................................................14
3.1 Reforms in Vietnam ...........................................................................................................................................................14
3.1.1 Agricultural reform..................................................................................................................................................14
3.1.2 Financial reform ........................................................................................................................................................15
3.2 Overview the development of rural financial market in Vietnam ...................................................16
3.2.1 Vietnam Bank for Agriculture & Rural Development (VBARD)..........................................16
3.2.2 Vietnam Bank for the Social Policy (VBSP).......................................................................................17
3.2.3 People's Credit Funds (PCFs).......................................................................................................................17
3.2.4 Rural Shareholding Bank (RSHB) .............................................................................................................17
3.2.5 Informal financial sector .....................................................................................................................................18
4 METHODOLOGY.......................................................................................................................................................19
4.1 Data collection .......................................................................................................................................................................19
4.1.1 Collection and review of secondary data..............................................................................................19
4.1.2 Mapping...........................................................................................................................................................................19
4.1.3 Group discussions...................................................................................................................................................20
4.1.4 Key informant interview......................................................................................................................................22
4.1.5 Household survey....................................................................................................................................................22
4.1.6 Observation...................................................................................................................................................................22
4.2 Data analysis ...........................................................................................................................................................................23
5 STUDY AREA ............................................................................................................................................................. 24
5.1 Physical characteristics ..................................................................................................................................................25
5.2 Economic situation .............................................................................................................................................................26
6 RESULTS AND DISCUSSION ...................................................................................................................... 30
6.1 Participation of rural households in the credit market in Tan Linh Commune...................30
6.1.1 Credit demand in Tan Linh Commune....................................................................................................30
6.1.2 Diversity in purposes of credit demand in Tan Linh commune...........................................32
6.1.3 Financial institutions and current access to credit of local people..................................34
6.1.4 Interaction between formal and informal sector.............................................................................36
6.1.5 Factors of borrowing of rural households............................................................................................38
6.2 Formal lender’s behaviour in responding to credit demand of local people.........................43
6.2.1 Credit rationing situation....................................................................................................................................43
6.2.2 Factors of credit rationing by VBARD.....................................................................................................44
6.3 Credit and livestock production in Tan Linh commune.........................................................................49
6.3.1 Credit-constrained and unconstrained households.....................................................................50
6.3.2 Credit constrained households and livestock production .......................................................52
7 CONCLUDING REMARKS .............................................................................................................................. 55
8 REFERENCES............................................................................................................................................................ 58

5
LIST OF TABLES AND FIGURES

TABLES

Table 1: Criteria of the community for classifying households.........................................21

Table 2: Ha Tay in comparison with other provinces in Red River Delta.......................25
Table 3: Means of some important variables of surveyed households...............................28
Table 4: Credit demand of local farmers (million VND).................................................31
Table 5: Borrowing purposes by wealth categories..........................................................32
Table 6: Credit sources and average amount of loans......................................................35
Table 7: Numbers of loans by credit sources...................................................................36
Table 8: percent of loans by formal and informal sectors.................................................38
a,b
Table 9: Indicators of borrowers by the level of borrowing from formal lender (Duncan
sig. 0.05).....................................................................................................39
Table 10: Indicators of borrowers by the level of borrowing from informal lenders
a, b
(Duncan sig.0.1).......................................................................................41
a, b
Table 11: Means of indicators by credit status of households to VBARD (Duncan sig.
0.05............................................................................................................45
a, b
Table 12: Means of some indicators by level of rationing by VBARD (Duncan sig.
0.05)...........................................................................................................48
Table 13: Means of some indicators by constrained and unconstrained households..........51
Table 14: Some indictors of constrained households by value of livestock production
a, b
(Duncan sig. 0.05)....................................................................................53

FIGURES

Figure 1: Location of Ha Tay province in Red River Delta............................................24

Figure 2: Sources of borrowing.......................................................................................37

6
LIST OF ABBREVIATIONS

PCF People's Credit Fund

SBV State Bank of Vietnam
SOE State-owned Enterprise
VBARD Vietnam Bank of Agriculture and Rural Development
VBP Vietnam Bank for the Poor
VBSP Vietnam Bank for the Social Policies
VLSS Vietnam Living Standard Survey
VND Vietnamese dong (16,000 VND = 1 US$ in 2005)
VWU Vietnam Women's Union
LUCs Land Use Certificates
JLGs Joint Liability Groups
RCFs Central Credit Fund
RCFs Regional Credit Funds
PCFs Provincial Credit Funds
RSHB Rural shareholding bank.

7
1 INTRODUCTION

There exists a large demand for credit in rural areas of developing countries where poverty
remains dominant and accumulation capacity is quite low. In these areas, rural households
need credit to abate constraints of their budget to finance expenditures at critical times but
the fact is that it is difficult for them to borrow money as they desire. This makes the
widely held belief that credit can help households in rural areas of developing countries
improve their livelihood condition seem like a weak assumption. This is especially so as
farmers’ access to credit sources can even be non-existent. It reflects an apparent conflict
between development professionals’ perception and the reality in which they work.
In a Press Release in 2004, the International Fund for Agriculture Development
(IFAD) stated that over one billion people lack access to basic financial service, depriving
them of the means to improve their income, secure their existence, and cope with
emergencies. It has been estimated that only five percent of farmers in Africa and about
fifteen percent in Asia and Latin America have had access to formal credit; and on an
average across developing countries five percent of the borrowers have received eighty
percent of the credit (Vega 1983, Adams & Vogel 1986, Braveman & Guasch 1986, cited
in Bali Swain, 2001). Lack of agricultural credit where it is needed most reflects
imperfections of rural credit markets in developing countries. Consequently, the
availability of funds to carry out timely purchases of cash inputs into agricultural
production, as well as to buy capital equipment like ploughs and water pumps has long
been regarded as one of the critical constraints inhibiting rising productivity in small-farm
agricultural (Ellis, 2000). The severity of this constraint is typically thought to reside in the
poor functioning of rural financial markets (Hoff et al, 1993; Besley, 1995, cited in Ellis,
2000). In particular, private markets in loanable funds, or credit markets, operate unevenly
if at all, in rural settings. In rural Africa, there remains a generalized problem of low rural
credit availability (Bigsten & Kayizzi- Mugerwa, 1995, cited in Ellis 2000). In rural Asia,
private money lending exists, but tends to be associated with personalized transactions in
interlocked markets that can place the borrower in a permanent state of obligation to the
lender (Bhaduri, 1986; Basu, 1994, cited in Ellis 2000). Government and NGOs have, of
course, for decades tried to overcome these market failures, but their success at doing so
has tended to be intermittent and uneven (Ellis, 2000. page 74).
There are many reasons for the existence of credit market failures, amongst which the
high costs of setting up banking operations in rural areas, the difficulty and cost of securing
adequate information on potential borrowers, the risk of default on loans and the absence
of collateral to put up against loans are the most frequently identified (Ellis, 2000). A lot of
studies (Adhikari, 2005; Barret et al., 2000; Bardhan & Dayton-Johnson, 2000; Hackett,
1992) conducted in rural areas of developing countries show that there exists low
population densities and significant heterogeneity of rural areas in terms of equity of
resource distribution, resource endowments, production, and consumption. These can
make the provision of formal financial services costly. Consequently, lenders often
confront the problem of information asymmetry. Rationing of credit demands becomes
necessary for lenders (Stiglitz & Weisse, 1981).
For a developing country like Vietnam, rural household’s access to credit to make an
optimal decision in their production activities is a very pressing issue. Vietnam is an
agricultural country where 80 percent of the population live in rural areas with their main
source of income derived from agriculture. The economy including its rural sector after
20 years of adoption of economic reforms is recorded with remarkable success. It is worth

8
mentioning its success in quelling hyperinflation, maintaining macroeconomic stability and
achieving a high economic growth rate of 7.5 percent per annum on average, while
halving the poverty rate. Reforms in agriculture and rural areas have helped to rapidly
increase agricultural output. Per capita food production was boosted from 303 kg in 1990
to 444 kg in 2000, and revenues from agricultural exports were more than quadrupled
from US$ one billion in 1990 to US$ 4.3 billion in 2000, (Document No.2685/VPCP-
QHQT, 2002). In the light of many emerging opportunities for expanding production,
farm households, being newly defined as autonomous economic units – have had an
extraordinary demand for funds to finance for their production (Duong & Izumida 2002).
However, despite the efforts of the Vietnamese Government in developing financial
systems, a large number of rural households are being excluded from access to institutional
credit services. The VLSS (Vietnam Living Standard Survey) provides an evidence of
reliance of rural household on the informal markets: 73 percent of the total loans of the
households were obtained from informal sources of which moneylenders accounted for 33
per cent while friends and relatives accounted for 40 percent of the total loans, (Dat,
2001). Thus the formal rural credit market in Vietnam has limitations in reaching out to
the greatest part of rural population to meet strong credit demand for improving their
livelihood conditions. It is a reason for why the informal credit market plays a significant
role in rural communities. There remains a lack of studies of this situation in Vietnam.
In the context of this problem, the research was set up by the idea that factors
influencing access to credit need to be explored because a sufficient access to credit can
improve the agricultural production condition of households in rural areas. More
concretely, the objectives of this research are to examine (a) the credit demand and factors
of borrowing level of rural households, (b) the behaviour of a formal lender in rationing
credit, (c) the influence of credit on rural household production.
The thesis is organized as follows. Chapter 2 reviews the previous literature on rural
credit markets. Chapter 3 presents an overview of the state of rural finance in Vietnam.
Chapter 4 describes methods used to collect data. Chapter 5 investigates the context of the
study site. Chapter 6 shows evidence from the field and then discuss it. Chapter 7 draws
the main concluding remarks.

9
2 LITERATURE REVIEW

Agricultural production has specific characteristics. Agricultural production takes time.

The lag between the start of production and the realization of output ranges from a few
months to several years. In this environment, credit transaction serves to finance
production and to permit farmers to consume before harvest. Moreover, the agricultural
process depends on a host of external factors, many of which are not under the control of
the farmer. When production is risky and insurance markets are incomplete, credit
transaction serves a valuable role by permitting people to smooth consumption in the face
of a randomly fluctuation stream of income (Bardhan, 1999). The seasonality of
agricultural expenditures and revenues implies that a household not only has to satisfy an
annual cash income constraint, with total expenditure less or equal to total revenue but
also to balance its budget during the lean season when there are high expenditures for
consumption and input purchase and few revenues (Sadoulet & de Janvry, chapter 6,
1995).
The terms and conditions under which credit is transacted may vary substantially across
different transactions, depending on the characteristics of the borrower and lender and the
relationship between them (Bardhan, 1999). A credit contract is defined simultaneously by
the amount, purpose, cost of borrowing, nature of collateral and the repayment schedule.
These are determined by negotiation between borrowers and lenders in the rural financial
market. So, factors influencing access to credit of rural households will be those that affect
both decision making of households and behaviour of lenders in responding to the
demand to borrow.
Thus, in order to increase access to credit, one has to identify the factors that influence
the borrower’s choice. Many policymakers assume that borrowers base their decisions on
one particular loan component, that is, the interest rate. However, focusing on interest
rates alone is not sufficient to explain borrowers’ choices for or against a particular financial
intermediary. Ranking lenders according to interest rates and transaction costs only,
implies that borrowers most likely will make use of those lenders with lowest rates and
costs. But borrowers’ decisions to accept or reject are influenced by other factors as well
(Zander, no date). Some borrowers might be afraid of prohibition or untrustworthiness of
the lender because they formerly have had overdue debt. Some others fear refusal due to
lack of collateral. Loan duration also influences borrowing decisions - whether it is
compatible with the life cycle or not. Some borrowers who are illiterate or have low
education find it difficult to cope with the complex procedure of credit applications typical
of many formal lenders. Mismatch of the credit offered and the actual loan needed is
another factor influencing decision making of household in participation in rural credit
market
On the other hand, we have to understand how lenders, especially formal lenders
respond to the borrowing demands of households. Stiglitz and Weiss (1981) applied an
imperfect information paradigm to explain behaviours of lenders. The imperfect
information paradigm concerns the behaviours of rational actors in an environment where
information is imperfect and costly. The asymmetric information is understood as
transactions in which one party has more information than the other about the
transactions. In the rural credit market, the assumption is made that banks cannot
differentiate high risk and low risk applicants because the banks don’t know information
regarding to investment choice, honesty, capacity and willingness to repay the loans of
potential clients which the loan applicants know well about themselves. Base on this

10
paradigm, Stiglitz and Weiss explored whether imperfect information leads to credit
rationing and they also analyzed the strategic behaviour of borrowers and lenders in a
credit market with asymmetric information. In particular, the bank chooses the optimal
interest rate under the consideration of its expected profit function, and lends money to
applicants at this optimal interest rate level so as to maximize its profit. At the optimal
interest rate which the bank chooses, the demand for funds exceeds the supply of funds.
Unsatisfied borrowers would offer to pay a higher interest rate to the bank to receive the
loans; the interest rate they offer can be bid up until demand equals supply. Stiglitz and
Weiss argued that the banks would not lend to an individual who offer to pay more than
optimal interest rate because banks could think that such a loan is likely to be a worse risk,
decreasing the expected return. So credit is rationed. As a result, some applicants obtain
the loan they apply for while others are turned down. Stiglitz and Weiss (1981) showed
that the interest rate is not the only term of the contract which is important. The amount
of the loan and the amount of collateral the bank demands of loan applicants will also
affect both the behaviour of borrowers and the distribution of borrowers. Increasing
collateral requirement could increase the riskiness of the bank’s loan portfolio, either by
discouraging safer investors, or by inducing borrowers to invest in riskier project and
therefore could decrease the bank’s profit (Stiglitz & Weiss, 1981)
It is also important to make clear the terms of access to credit. Access to credit implies
that the borrowing action of someone was accepted by the lender. It means that access to
credit depends simultaneously on the borrowing action (borrower’s choice) and the
lender’s decision. It is worth noting that households with access to credit may or not may
be credit constrained. They are credit constrained if they borrowed less than what they
desired and not credit constrained if their demands were met. The same is true for
households with no access to credit at all. The households with no access to credit may be
credit constrained if they had demand for credit but they could not obtain it, or not credit
constrained if they have no demand for credit.
To finance the purchase of inputs, the farm households must either dip into savings or
obtain credit (Diagne et al., 2000). Therefore, with limited access to credit, the budget
balance becomes a constraint where expenditures have to remain less than or equal to the
sum of revenues during a period, accumulated savings, and credit availability. Hence credit
constraint limits the optimum production or consumption choices, (Sadoulet & de Janvry,
1995). When a household is in a credit constrained situation, it means that they cannot
obtain the amount of credit they desire at the prevailing interest rate, nor can they obtain
credit by offering to pay a higher interest rate. Facing such a situation they have to
consider carefully how to invest, what inputs they should buy and of course they only buy
inputs for production after their basis needs had been satisfied. In other words, with
limited access to credit farmer’s production decisions will depend on their consumption
decisions. Where farmers are poorer these motivations are stronger – liquidity constraints
are more likely to be binding so that production and consumption cannot be financed by
savings and the costs (in terms of utility, health, and even survival) or fluctuation in an
already low level of consumption are very high (Bardhan, 1999).
In fact, many recent empirical findings show that access to credit has a significant
impact upon agricultural production as well as household economic welfare. By using data
collected in Tunisia for econometric estimates, Foltz (1998) showed that better access to
credit market will improve the profitability of a great number of farmers and the presence
of credit market constraints does impinge on farm profitability. The various evidence
supporting this assertion can be found from the papers by Carter (1989), Duong and &
Izumida (2002), Feder et al (1989), Feder et al (1990), Fisman (2000), Godquin &
Sharma(2004), Kochar (1997), Quach et al (2005), Zeller et al (2005).

11
 By using an econometric framework for analyzing credit rationing, Zeller (1994)
examined determinants of credit rationing for informal lenders and formal credit groups in
Madagascar. Empirical results showed that formal groups obtain and use information about
the creditworthiness of the credit applicant in a similar way to informal lenders. Land as
criteria for loan rationing neither plays a role for informal lenders nor for members of the
groups. Informal lenders and group members can obtain information about the wealth,
indebtedness, and income potential of the loan applicant. The use of the leverage ratio
(which is defined as the ratio of outstanding debt over income) as a significant determinant
of loan rationing is less regressive than the use of land as collateral that has been identified
as the overriding determinant for access to formal credit contracted directly between the
bank and the individual borrower (Zeller, 1994).
Diagne (1999) used the concept of credit limit to identify determinants of access to and
participation of households in informal and formal credit market in Malawi. His results
show that (1) there exists severe credit constraint in Malawi, (2) the composition of
household assets is much more important as a determinant of household access to formal
credit than the total value of household assets or landholding size. However, landholding
size remains a significant determinant of access to informal credit. (3), the unobserved
program-specific attributes captured by the program dummy variables are the most
significant factors that influence household decisions to participate in a credit program.
These unobserved program-specific attributes include the types of loans provided and
restriction on their use, as well as other educational and social services provided by the
programs.
With the idea that credit rationing might not be as high as it is generally assumed in
literature and farm households may have a low demand for credit, Bali Swain (2002)
investigated the extent of effective credit rationing by the formal sector in the rural
markets of Puri. Her study used three different econometric models. The first one assumes
that formal credit is the cheapest source of credit for all households and that all
households have a positive demand for formal credit at the existing interest rate. The
second model was constructed with the assumption of the probability to borrow from the
formal sector is jointly determined by the demand for credit and the decision of the bank
on access. The third one relaxes both these assumptions and the household chooses
between borrowing from the formal or the informal sector. The results of employing three
econometric models for analysis of 1052 surveyed-households coming from 66 villages
showed that the three models estimated different results but in general there existed a high
demand for credit and a high degree of effective credit rationing by the formal sector in
Puri.
By using a micro-econometric analysis of household surveys, Duong and Izumida
(2002) examined the rural household participation in the Vietnamese rural credit market;
the formal lender’s lending behaviour in response to the credit demand of households and
impact of credit on household production. The empirical results showed that due to
segmentation of rural credit market in Vietnam, households are rational in deciding the
sources from which they should ask for particular kinds of loans. It was found that, total
farming area and total value of livestock are decisively the determinants of borrowing by
households from the formal financial institutions. Reputation, the dependency ratio of
households and the amount of credit applied for by the household are identified as the
determinants of credit rationing by the bank.
Through the use of cross-sectional data from the two households surveys undertaken in
1992/1993 and 1997/1998 to derive empirical evidence, Quach et al (2005) showed that
in Vietnam household credit contributes positively and significantly to the economic

12
welfare of households in terms of per capita expenditure, per capita food expenditure and
per capita non-food expenditure. The positive effect of credit on household economic
welfare is regardless of whether they are poor or better-off households. They also found
that credit has a greater positive effect on the economic welfare of poorer households and
that the age of the household head, the household size, land ownership, and savings and
the availability of credit at village level are key factors that affect household borrowing.

13
3 STATE OF RURAL FINANCE IN VIETNAM

3.1 Reforms in Vietnam

Vietnam's transition process is well-known in the economic world as 'Doi Moi'. Doi Moi
literally means 'change' and 'newness' and is the Vietnamese term for reform and
renovation. It is aimed at restructuring Vietnam's legal, regulatory, administrative,
investment and foreign trade apparatus and policies to transform its centrally-planned
economic system into a market economy with 'socialist characteristics' (Bryant 1998).

3.1.1 Agricultural reform

Before the year 1986, all land, natural resources and productive activities in Vietnam were
controlled by the State. The State allocated equipment and raw materials for production
and organized agriculture under a collective system. In 1980, all the country had 126
thousand of cooperatives that attracted 65.6 percent of the total number of households
into collective production. However, the collective model did not stimulate enthusiasm of
farmers because the cooperatives were not able to bring about sufficient income to
guarantee for basic needs of members of cooperatives. As a result, growth rate of
agricultural production was low, on average at 1.9 percent per year. In 1980, Vietnam had
to import 1.6 million tons of food, (MARD, 2004)
From the social and economic crises of period of 1981-1985, the 6th Party Congress of
the Vietnam Communist Party adopted a reform policy in order to move towards a
market economy. The reform process led to liberalization of a centrally planned economy
that was no longer appropriate for the development of the national economy. There were
several fundamental reforms in the agricultural sector. Since 1988 households have been
considered as autonomous, independent economic units. It means that households
themselves are fully responsible for their own production as well as consumption decisions.
It is a contrast to the period before Doi Moi, when major units of production and business
were cooperatives and state – run farms.
Property right is an important element in the reform. Collectives were dismantled in
1988 and land was distributed among households. Under the Land Law in 1993 (amended
in 1998), farm households had the right to use the land distributed to them and that right
could be renewed. Farmers have a long-term land use certificate which they could use to
transfer, to exchange, to lease, to inherit or to mortgage. Thanks to the reforms in the
agricultural sector, agricultural production sharply increased in output. The growth rate of
agricultural production was stable at 3-4 percent. While up to the 1980's Vietnam used to
be a food importing country, in 1999 it was the second largest rice-exporting nation in the
world. In 2000 food output equivalent to paddy was 35.64 million tons which was an
increase of 6.42 million tons over 1996. The growth rate of food output was higher than
population growth rate (1.8%). Therefore, national food security was assured even in bad
weather conditions. Further, rural life was improved substantially with average income per
capita in 2001 increasing to twice that of 1993 (Cuc, 2003).
Although Vietnam achieved a drastic decrease in poverty between 1993 and 1998,
poverty remains a largely rural phenomenon, with 90 percent of the poor living in rural
areas and with 45 percent of the rural population living below the poverty line (Putzeys,
2002). It is also worth noting that individuals living just above the poverty line are easy to
force below the poverty line again because they lack the capacity of accumulation and

14
access to credit to cope with risks and shocks. Currently, Vietnam has 12 million
households who are involved in agricultural production in which 5.4 million households
are poor (Dao, 2001). Overall capital accumulation of these households is very low. In
2000, the average of accumulation per household was 3.2 million VND per year.
There still exists however a contradiction between the use of mechanized agricultural
production and high redundancy of laborers in rural areas. For example in the Red River
Delta, it takes 300 working days of manual laborers to finish one ha but if using machine,
this one ha only needs 50 days. It means that there is a redundancy of 250 working days
which is equivalent to 250 laborers without jobs (Cuc, 2003). To solve this contradiction,
industry and services in rural area should be developed to regulate agricultural laborers to
non-agricultural sector. However, this process has required a large capital to create non-
farm works.
The other current shortcoming of Vietnam agricultural production is limited
competitiveness of agricultural products in the world market. This weakness can be found
in all products of cultivation and husbandry with low quality, high cost, few categories,
and limitation of food safety (Cuc, 2003). This reflects backward processing technologies.
Investing into modern processing manufacturing can solve two problems. Firstly, it might
improve quality of agricultural products which will lead to increase of competitiveness of
agricultural products in the market, and secondly it can create more non-farm jobs for
redundant agricultural laborers. However, this process requires capital accumulation. In the
context of low accumulation capacity in rural area, access to credit would be a vital mean
for solving the problems.

3.1.2 Financial reform

The financial system before Doi Moi was a “mono-bank” system. It means that the State
Bank of Vietnam (SBV) was entrusted with both the functions of a central bank and a
commercial bank. As the role of central bank, SBV regulated monetary policy, managed
the national reserves and issued banknotes. As commercial bank, it mobilized savings from
public and allocated credit to enterprises. At that time, rural finance was composed of the
State Agricultural Bank, credit cooperatives and the informal sector. The State Agricultural
Bank, which was actually a part of the SBV, supplied long-and short-term funds to stated
owned enterprises and cooperatives. The funds came from government budget and the
savings of the public. The implementation of the Renovation reforms brought changes to
the banking system of Vietnam. In 1988, the banking system shifted from a mono-
banking system to a two-tier system, whereby responsibilities of a central bank were
separated from those of commercial banks. Under the two-tier system, the State Bank of
Vietnam retained the responsibilities of the central bank, specifically responsibilities for
monetary policies. Commercial banking responsibilities were transferred to the two newly
created state-owned commercial banks.
Historically, The Government of Vietnam has pursued a policy for channeling
subsidized credit through the formal state-owed sector, principally Vietnam Bank for
Agricultural and Rural Development (VBARD) and Vietnam bank for the poor (VBP). In
1998, the policies of the government aimed to: increase capital available through formal
financial institutions for rural lending to extend outreach to more rural households;
increase capital for medium-term and long-term loans; eliminate preferential access to
credit by state-owed enterprises; apply interest rate policies that mobilized domestic
savings; reduce directed lending and preferential loans, increasing the market allocation of
capital; and simplify lending procedures.

15
 In 8/2000 SBV issued new regulations under which deposit rates were set by reference
to commercial bank’s short term lending rate for reliable clients and lending rates were
capped within maximum spread above the deposit rate. In 6/2002 SBV further revised
interest rate policy removing the cap on the spread for all institution except for VBP.
In October 2002, the Government issued a Decision on the Establishment of the Social
Policies Bank of Vietnam (VBSP). This Bank opened on January 2003, taking over the
subsidized loan portfolio of VBP. By June 2003, the preferential lending schemes of all
other bodies were transferred to the Social Policies Bank. It was envisaged that subsidized
lending would continue through this channel for the foreseeable future. This was one of
basic reforms of financial and banking system in Vietnam

3.2 Overview the development of rural financial market in

Vietnam

3.2.1 Vietnam Bank for Agriculture & Rural Development

(VBARD)
VBARD is foremost in the formal rural financial service sector. It provides over 75
percent of the credit extended by formal financial institutions to rural households. The
VBARD branch network generally extends to the district level, but goes to the commune
level in highly populated areas. In the past few years, VBARD has introduced a number of
innovations to extend its outreach. In 1995, VBARD started a program of poverty-
oriented lending, drawing from the government's Preferential Credit Fund. At the end of
1995, the fund was transferred to VBP. However, VBARD still lends to poor households.
In 2000, roughly four million households accounting for 30 to 40 percent of rural
households accessed VBARD loans of which 20 percent are poor households. To date,
VBARD has been the largest provider of financial services to rural areas and the second
largest provider to poor households. By the end of year 2003, VBARD provided credit to
0.8 million of poor households accounting for 15 percent of total of poor households
taking formal loans in the whole country (Hung, 2005).
Under the current regulations of SBV, VBARD can provide up to 10 million VND
without collateral. For loans larger than 10 million VND, households have to offer
collateral. Land use certificates (LUCs) were used as the sole collateral of households in
rural area which the banks will hold until the repayment is completely performed.
However, currently all over the country only 60 percent of total rural households has
LUCs (Hung, 2005).
Group lending techniques have been employed by VBARD to increase its outreach in
rural areas and providing to joint liability groups (JLGs) is prevalent. The JLGs serve as
social collateral which replaces physical collateral to increase the rate of loan repayment
(Izumida & Duong, 2001). This lending technique has been applied by support of local
organizations such as Farmer Union and Woman Union
Currently, VBARD has been maintaining lending interest rate at 1 percent for short-
term loans and 1.15 percent for medium and long term loans.

16
3.2.2 Vietnam Bank for the Social Policy (VBSP)
VBSP was established in 2003. It operates as a non-profit organization. Its major
responsibility is to provide loans to poor households and promote the poverty alleviation
program. Since its establishment, VBSP has been the largest provider of financial services
to rural poor households in Vietnam. By the end of the year 2003, 2.8 million of poor
household accessed VBSP with total loans value approximating 10,349 billion VND
(Hung, 2005). The average size of loan is 3 million VND with 36 months of maximum
duration. In addition, VBSP also provides micro-credit to support job creation and poor
students. The capacity of VBSP in mobilizing deposits is limited due to lending at a low
interest rate of 0.5 percent per month. The operational capital of the bank mainly depends
on State’s budget.
To get the loans from VBSP, applicants have to be members of the JLG which is
organized by local social organizations as Woman Union, Farmer Union and so on. These
organizations have played a very important role in selecting good borrowers, in guiding
borrowing procedures, and in monitoring loan repayments (Izumida &Duong, 2001).

3.2.3 People's Credit Funds (PCFs)

People's Credit Funds (PCFs) were established as commune-based savings and credit
cooperatives. The PCF model is operating under the Cooperative Law. In comparison
with VBARD and VBSP, PCFs has closer relation with farmers. The PCF network is
comprised of one Central Credit Fund (CCF), 21 Regional Credit Fund (RCFs), and 961
PCFs operating in 53 out of 61 provinces throughout the country. The PCF network
covers over 1000 communes, accounting for 11 percent of the total communes in the
country (Hung, 2005). According to recent reports, the PCF network currently provides 5
percent of the outstanding credit in the formal rural financial sector and covers 6 percent
of the rural households, of which 5 percent are the poor households. Reports indicate that
the loan repayment rate throughout the system has remained high (above 98%). The
savings mobilization capacity of the PCF system has substantially improved since its
inception. The modus operandi of the PCFs is to mobilize people's capital and in turn lend
it to other members. Lending and deposit procedures are simple and appropriate to the
level of formal education among the rural population. Moreover, the diversification of
credit and savings products has brought about a wide range of options to households that
have different levels of income.

3.2.4 Rural Shareholding Bank (RSHB)

RSHBs are the result of the reorganization or merger of rural credit cooperatives. The
main advantage of such banks is that their lending procedures are simple, with staff relying
on their knowledge of and close relationships with borrowers who are often family or
friends. It is also the staff who help the applicants complete the required documents. The
reliability and low cost of this process are brought about by the dual roles of most staff, as
technical personnel and shareholders. These banks are very cautious about expanding their
lending to the poor, and therefore very doubtful in terms of effort to outreach to this
sector (Hung, 2005)

17
3.2.5 Informal financial sector
Despite the great efforts to improve credit provision to rural areas, the formal financial
sector cannot fully meet rural households’ demands for loans, leaving a large market for the
informal credit sector to operate. Private moneylenders seem to dominate the provision of
capital to rural areas. Loans from informal sources (e.g., moneylenders, relatives and other
private individuals) represent at least 50 percent of total loans. It is believed, however, that
many households borrowing from the formal sector also borrow from the informal sector
so that, in reality, the informal sector could still represent as much as 70 percent of total
loans. Moneylenders charge interest rates (ranging from three percent to 10 percent a
month) that are much higher than those charged by formal financial institutions (Hung,
2005). Relatives, friends and neighbours often are the first sources when household seek
for credit. These loans are provided without collateral or written contract. Interest rates are
recorded low and in many cases are free. The purposes of these loans are quite diverse for
both consumption and production.

18
4 METHODOLOGY

Field research was undertaken in 11 villages of Tan Linh commune, Ba Vi distric, Ha Tay
province. The understanding about nature of things was constructed by looking at the
overall picture of 11 villages and then tracing the details concerning questions of the what,
the why and the how. The field survey was initiated by field visits and discussions with
key local authorities. Next, group discussions, key informant interview and a household
survey were used to collect data covering a variety of topics related to land, water
resource, production patterns, human resource (labour, age, education and health), credit
and agricultural output. Information about credit is the main component including credit
demand, sources and use of loans, disadvantages in getting the loans from the formal sector
and level of borrowing compared to demand.

4.1 Data collection

4.1.1 Collection and review of secondary data

The secondary data was collected from province-level authorities and commune-level
authorities including annual reports about the socio-economic situation and production
performance, development plans, reports about programming and land use of the
commune, reports regarding the number of borrowers, the purposes of loans, the kinds of
loans, outstanding debts and repayment ratios of VBARD and VPSP in 2005, reports of
financial intermediaries such as Farmer Union, Woman Union, etc. who support the
banks in lending to farmer groups.

4.1.2 Mapping
Mapping was undertaken with the participation of four local staff including a local land
survey officer, a water officer, a local extension worker and a head of farmer union.
During this process the researcher played the role of facilitator and observer. A map of the
research site was drawn by them indicating boundaries between villages, natural resources,
farming patterns, irrigation system and location of inhabitants. The map also reflected the
dimension and scope of objects to be investigated. Topographical data coming from the
map pointed at basic information on soils, use of land for particular purposes, location of
rivers and forest, agro-ecological zones, water availability and infrastructure system (road,
banks, markets, tourist areas, schools, and health facilities). The purpose of using this tool
was to collect general information which gave support for understanding the overall
picture of the site. Moreover, the map was used as a visual aid to facilitate farmers in group
discussion.

19
4.1.3 Group discussions
a. Farmer group
Two groups of 10 farmers each were selected randomly by the heads of villages to discuss
together within each group. One included farmers who are representative of four villages
locating at the edge of Tan Vien Mountain. The other included farmers who are
representative of seven villages bordering the town of Ba Vi district. They were
interviewed collectively because they had a common experience, coming from similar
background. So they are seen as representing a wider population of the commune.
This method was used to collect data relating to (1) production patterns of local
households (2) credit demand and purposes of borrowing (3) their disadvantages in access
to each credit source (4) the reasons for these disadvantages, (5) how they made borrowing
decisions in the context of credit constraints (6) their expectations regarding terms of
collateral, loan duration, lending method, and (7) how credit constraints influence their
production. In addition, seasonal calendar was used in discussion process to explore more
information. Rural livelihoods are inevitably dominated by the seasons. Each period in the
year has its specific characteristics, necessarily impacting on local people’s life. In this study,
this tool was used to list all main production activities and their cycles over 12 months.
These yearly cycles are important in determining labor availability, agricultural materials
demand, times of crop and animal diseases, times of leisure after harvest, food shortages
and variation in cash flow. Besides, the map which had been drawn by four local staff as
presented above was used as visual aid in discussion process to explore more information
regarding location of resources (natural, physical, financial, social resources), their
availability and linkages, distances and access to resources which served their production
development.
In this study, group discussion seemed appropriate for collecting data because (1) it was
good for eliciting a lot of information quickly, identifying and exploring beliefs, attitudes
and behaviours of local farmers (2) it provided in-depth information about how people
think about their credit status, reasons and factors of being credit constrained households,
(3) it suggested a guidance in setting up a framework for oriented research and
constructing an appropriate questionnaire for the household survey.

b. Village heads group

Eleven village heads who had good knowledge of local people were invited to discuss
in a group. The purpose of this discussion was to identify criteria for classifying households
in the Tan Linh commune into wealth categories. The discussion was organized as
follows: First, the author asked all village heads to define what ‘wealth’ is in the
community and with respect to this definition, how many different types of households
existed within in the community? This question was debated among village heads and they
came to conclusion that in fact, in their community, there were three different kinds of
households, namely better-off households, medium households and poor households.

20
 Table 1: Criteria of the community for classifying households

ID Indicators Better off Medium Poor

1. Income 20-30 million VND/year 10-15million 4-7 million VND/year
VND/year
2. Durables (TV, Sufficient durable with high insufficient or sufficient serious lack and very
motorbike, radio, bikes,
quality durable but lower low quality
etc)
quality
3. Production instrument Having sufficient necessary Enough necessary Lack of essential
and farming machine
instruments for their own instruments but they do instruments for
production. They even have not work well production or their
leased them to earn more quality is bad
money
4. Number of cattle 5–6 2-3 0-1
5. Number of poultry 200-300 per year 50-70 per year 30 per year
6. Number of pigs 40-60 per year 10-15 per year 2-4 per year
7. House type Houses have two floors or solid house with flat Tile –roofed house has
solid ones with flat roof roof been downgraded
8. Accumulation (savings having insurance or lending to having very low having to borrow from
and insurance)
relatives and friends accumulation others
9. Extra job/secondary doing business, tourism, small doing business, Cultivate for the better
job
providing decorative plants building worker off, bricklayer, porter

10. Access information access to various sources of
information to be quick-
witted to changes; take
advantages of opportunities
and options
11. Relationship within the Having a wide circle of
community
acquaintances. Very often
participate in merrymaking
and meetings
12. Education graduated from secondary
school or high school
13 Children/ members of reaching adulthood
family
Source: Village heads group discussion in Tan Linh commune, Nguyen Thu Phuong,
2005
good access to Lack of information
information relating to even they do not care
disease, breeding, prices, much what are changing
market, etc to adapt to because in case of
fluctuation knowing about it, they
are still not able to
change their situation.
being expansive and having a feeling of
confident to have inferiority, keeping
relation with others away from contact with
community
graduated from graduated from primary
secondary school or school
high school
being at the age of having the old, sick
schooling people, young children

Next, the author and village heads determined what a better-off households is, focusing
particularly on its characteristics. Then the characteristics of poor households were
discussed; and last came the characteristics of medium households. After that, all distinctive
characteristics of each group which were defined via discussion among village heads were
written on poster paper size as a final result of the discussion. This result on the one hand
was used in the discussion of two farmer groups which were presented above to probe
farmers’ perception to specific problems of each category. On the other hand, it was used
to classify 100 households which were selected randomly as a sample for household survey
into three categories: better-off households, medium households, and poor households. An
assistant with a good command of banking and finance assisted the author in facilitating the
group discussions.

21
4.1.4 Key informant interview
In this study, key informants including leaders of the commune, the vice-director of
VBARD at the commune level branch, the director and staff of VBSP at district level, the
head of Farmer Union, and the head of Woman Union were those considered to be
influential, prominent, or well-informed people in the community and in organizations or
institutions; they were selected for interviews on the basis of their expertise in areas
relevant to the research. This method was used to collect information regarding (1)
policies and regulations on lending and mobilizing savings of the banks; (2) the assessment
process for the approval of a loan; (3) methods that the banks use to clean unworthy credit
client; (4) advantages and disadvantages in lending to local people. A checklist with open-
ended questions was prepared to focus on the issues to be explored.
Besides, six farmers belonging to credit constrained households (identified based on the
result of household survey) were selected as key informants for probing understanding
about credit constraint status. These farmers were asked about the process of borrowing,
the ways they use loans, their decision in the context of credit rationing by the bank, and
the influence of credit constraint on their production. Some stories told by them were
presented in the text of the study. Interviews with these farmers were conducted after the
household survey.

4.1.5 Household survey

The purpose of household survey was to collect specific information of each household for
gaining an in-depth understanding of the human resource, assets, social position, financial
resource, credit status, production patterns, and income sources of households. In this
study, a household survey was used as a major method to collect data on a range of core
topics including (1) business production activities of household (2) asset, means of
production, land and land use right (3) human resource (4) number of loans applied for
and actually received, including information on amounts involved, interest, collateral
requirement, period and source of the credit (5) credit status for all sources (6) capacity of
savings and use of loans (7) cash cost for production and value of production. One
hundred households were selected randomly in a list of 2.218 households. The list was
provided by people’s committee. Most of the selected households were interviewed by the
author. The rest was done by two assistants who have economics backgrounds.
After finishing the household survey, the author classified the sample of the 100
households into wealth categories. A list of 100 household heads was given to 11 village
heads. Based on the criteria for classifying households constructed in the group discussion,
village heads grouped 100 households into three categories: 19 better-off, 54 medium and
27 poor households.

4.1.6 Observation
Personal observation was used to bridge data gaps and verify information obtained from
different used methods. This method was combined with all methods above in the process
of collecting data. The employment of observation method was able to gain first-hand
experience of local farmer behaviours meanings, relationships, and contexts and to
systematically describe these.

22
4.2 Data analysis

The data from the household survey was coded and organized into sub-subjects including
general information, asset, financial sources, livestock production, cultivation, non-farm
income.
The data will be presented by cross tabulation, graphics, maps and pictures as evidence
for making arguments. Data from the survey, group discussion and key informant
interviews constitute the core data for analysis.

23
5 STUDY AREA

Tan Linh commune, Ba Vi district, Ha Tay province was selected as the site for this study
on the basis of the fact that this site has the same agricultural production patterns as other
provinces in Red River Delta; secondly it is well-connected with other provinces in
region especially Ha Noi city to take advantages of technology, information and market
which

Figure 1: Location of Ha Tay province in Red River Delta

motivate farmers to expand production and sell products; thirdly Tan Linh remains one
of seven poor mountainous communes of Ba Vi district which implies that local
households have a low capacity for accumulation. They could be households that are
having strong demand for credit for development of their economy.

24
 Table 2: Ha Tay in comparison with other provinces in Red River Delta
Provinces
Items Ha Tay Hung Hai Bac Vinh Hai Nam Thai
Yen Duong Ninh Phuc Phong Dinh Binh
Total area 2,192 923 1,648 808 1,371 1,519 1,641 1,545
Population (million)
2.49 1.11 1.70 0.98 1.15 1.75 1.95 1.84
Average per capita GDP (USD) 273 329 380 362 360 536 248 251
Contribution of agricultural sector in
GDP (%) 35 35 30 29 26 16 37 48
Contribution of industrial and
constructional sector in GDP (%) 36 33 41 43 45 39 25 18
Contribution of services and
comercial sector in GDP (%) 30 31 28 28 29 45 38 33

Population work in agricultural

sector (million people) 0.96 0.47 0.68 0.37 0.45 0.45 0.73 0.79

Average area of agricultural land

per capita (m2) 1,282 1,349 1,544 1,397 1,468 1,621 1,458 1,312
Total value of agricultural
production (billion VND) 5,034 3,227 4,295 1,876 2,440 2,112 3,942 4,635

Total number of pigs (1000 head) 1,225 519 787 473 496 588 716 906

Total number of cattle (1000 head) 133 36 68 60 155 23 39 51

Source: General Statistics Organization, 2003

5.1 Physical characteristics

Ha Tay is an agricultural province of northern Vietnam, located in the Red River Delta
and shares a border with six provinces, namely Hanoi, Hai Hung, Ha Nam, Hoa Binh and
Phu Tho. This transitional zone, which links the Northwest and the Northern Midland
with provinces in the Red River Delta, is made up of the two towns of Son Tay and Ha
Dong, together with twelve districts. Ha Tay ranks first in terms of area and population in
which the number of people working in agricultural sector is biggest; however, average
area of agricultural land per capita is the lowest. Agricultural production of Ha Tay
obtained the highest value compared to those of other provinces, accounting for 35
percent of GDP of the province while industrial and service sectors contributes to GDP a
lower percent. The average per capita GDP of this province is much lower than that of
some other provinces in the Region, (Table 2). Population density is highest in Ha Dong
2 2
Town with 6,161 persons/km and lowest in Ba Vi district with 587 persons/km
Ba Vi District is a mountainous area located far from the province centre with a land
2
area of 428km where ethnic groups like the Kinh, the Muong and the Dao co-reside.
Because of its terrain, Ba Vi is shaped into specific areas: seven mountainous communes,
12 hilly communes and 12 plain communes. The district population averages at 253,229
persons, 95 percent of which live in rural areas with a total farming land area of more than
11,000 ha.

25
 Tan Linh commune is one of seven mountainous communes of Ba Vi district. It is 14
kilometres from the district centre and 60 km from Hanoi. Generally, Tan Linh is a half
mountainous and half plain area. The commune’s topography is characterized by a
considerable elevation in the south and low-lying areas in the north. The commune is
made up of three distinct regions: the mountains areas, the hilly areas and the plains. Its
terrain extends from South to North and from East to West. 45 percent of the commune
land area is covered by hills and mountains. The rest is plain.
Tan Linh is situated in a region of typical monsoon climate of the Northern Delta. The
annual average temperature measures 23.4 percent. February is the time when the
0 0
temperature reaches its lowest level (17 C) and even 7 C at particular points in time,
resulting in a foggy and misty weather. The spell of hottest weather can be found in June
0
and July when temperature amounts to about 28.5 C. Average rainfall is 2377mm/year,
highest in July with 520.5mm and lowest in December with 13.8mm. Moisture and
humidity averages 86 percent. The annual floods make their appearance from June to
October while the period between January and April witnesses local drought. In winter,
Tan Linh comes under the influence of north-easterly monsoon while in summer it is
affected by south-westerly winds which play havoc with crops and livestock.
Generally, the soil is characterized by the property of being soft and light which suits
well the need for orchard trees and long-harvesting crops. The alluvial soil found now in
valleys or riverbank is usually aluminous. This kind of soil is spent for cultivating rice and
cash-crop. Besides, the areas between valleys and hills are filled with exhausted grey soil
(source: local authority).
The water needed for agriculture completely depends on the natural sources distributed
through the difference in height between the mountainous, hilly and plain areas. This
system of water supply includes four lakes and four dykes. Together with the two streams
Cai and Thon Chu they form a natural system of irrigation and drainage for the
commune. The local canals have been in use for a long time and were mainly made of
soil. Only 1130m of Hoc Cua has been made in concrete while several patches of canals
have shown signs of complete degradation.
According to a local report of planning, the local traffic system in general is still
underdeveloped with the 87A trail and the entry to Ao Vua being the only roads that have
been surfaced. The remaining is dirt and mud. Some roads, especially those in residential
areas, are narrow to such an extent that they pose a lot of difficulties to the daily
transportation of goods and the travelling of residents, especially during the rain season.

5.2 Economic situation

In Tan Linh commune, agriculture plays a vital role in the local economy with 87 % of
the households depending on it for their livelihood. In 2001, the production value of
agriculture makes up 70 percent of the total local GDP (source: local authority - key informant
interview).
Cultivation is still the basis of the local economy; 90 percent of local households in the
commune are involved in cultivation. Almost local people grow rice however this product
is just for household consumption. In 2001, total yield of rice was 3052.7 tons and the
average amount of rice per capita stood at around 317 kg. The reason for this poor
performance was said to have come from a low area of cultivation per capita of 545 sq
meters, added by the fact that a considerable area of land for rice planting is too flooded in
the rain season to be cultivable. The irrigation system is basic which means the irrigation

26
and drainage work being completely dependent on the natural conditions. Besides
growing rice, local farmers cultivate other varied cereal crops as well as short-term
industrial plants, such as maize, sweet potatoes, cassava, peanut, soybean, and vegetables.
These farm products and also by-products are favoured sources of nutritious feeds.
However, most of them are kept for raising livestock for household use not for selling in
the market. It can be inferred that at household level, cultivation is not considered a direct
income source, and the products are only enough for home consumption and raising
livestock. There is integration between crops and livestock production at household level.
In 2005, livestock accounted for 80 percent of the agricultural output. Raising cattle,
pig and poultry is common but livestock production is small scale. There are a few large
commercial livestock production units. On average, each agricultural household keeps
1.16 head of cattle mainly for reproduction and draught which can be sold when farmers
need cash. Most of the local cattle such as cows, buffaloes and goats feed themselves on
local pasture. When the rain season comes, the grassland is flooded with water, leaving the
cattle to be fed by rice straw and agricultural by-products. The largest sum of money is
spent on the purchase of breeding animals and veterinary care, and the locals do not spend
too much money on cattle feeds. Compared with other livestock, the cattle are considered
a more valuable household asset and they can easily be ‘converted’ to cash in case of
emergency.
Pig-raising is extremely important in the household economy of commune. The
average number of pigs per agricultural household is five. However large-scale pig
production is not common. Products coming from cultivation such as sweet potatoes,
cassava, maize, and rice bran are the main sources of feed for raising pig. In addition, local
farmers also buy compound feed and use small household resources that are available
everyday, like household waste, leaves, and vegetables for raising pig. Profitability is not
high but pig raising provides a means of saving.
Results of the group discussion with farmers showed that raising poultry and fish are
important sources of supplementary income for them. Within the past two years however,
although the locality has not been affected by the avian flu epidemic, the amount of raised
poultry is kept at a modest level, mostly for family use, due to a dramatic fall in the market
price for poultry.

27
Table 3: Means of some important variables of surveyed households

Indicators Classified households

Unit Sample poor HHs medium better- off
n=100 n=27 HHs HHs n=19
n= 54
Age of household head years 46.35 41.93 48.50 46.53
school
Education of household head 8.42 7.89 8.59 8.68
years
Family size people 4.71 4.19 4.96 4.72
Number of labor people 2.91 2.33 3.15 3.05
Number of dependent people 1.89 1.85 1.98 1.68
Number of sick person people 0.12 0.15 0.13 0.05
*
Social status 0.13 0.07 0.13 0.21
Total farming area 360m2 12.86 8.37 13.48 17.48
Million
Total demanded credit 16.98 12.84 17.43 21.60
VND
Million
Total formal credit 4.50 2.22 4.69 7.22
VND
Million
Total informal credit 4.99 3.47 4.19 9.44
VND
Million
Total credit 9.49 5.65 8.88 16.66
VND
Million
Total value of livestock production 36.46 16.09 42.46 48.31
VND
Million
Total value of livestock sold in the market 15.08 7.00 17.63 19.35
VND
Million
Total cash cost for livestock raising 8.25 5.47 8.61 11.18
VND
Million
Total value of cultivation 7.49 4.01 8.43 9.77
VND
Million
Total value of cultivation sold in the market 1.96 0.54 2.55 2.24
VND
Million
Total cash cost for cultivation 2.22 1.87 2.29 2.49
VND
Million
Non-farm income 9.04 3.93 10.39 12.45
VND
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006
*Social status reflects position of household head in the society. It is defined as 1, if
household head hold position, as 0 if not.

Non-agricultural supplementary incomes are not yet widespread. Some better-off

households invest in farming machineries such as rice milling, tractors and threshing-
machines for lease. Of the whole commune, twelve households work in the small scale
industry, mainly making furniture, moulding and welding. The two places of interest Ao
Vua and Ho Tien provide a basis for the development of local services and trade. In 2001
the whole commune had 131 households working in such services as transportation,
construction material supply, motorbike repair, merchandise retail and cattle slaughtering.
A high percentage of local residents, especially males, manage to find a hired work in

28
proximal regions or in urban areas, either as porters or builders. This, however, is not
steady and permanent employment so the income it brings is usually unstable, (Annual
report of people’s committee, 2001)
Table 3 presents an overview of key variables in the household survey. Generally, the
result shows that there are differences in age, education, labor, dependents, social status
and credit access between household groups in which poor households are those having
the poorest-resources of human capital (age, labour, and education), natural capital (land),
financial capital (credit and savings), social capital (reputation and social status) which could
result in the lowest value of their agricultural production and lowest non-farm income.
The biggest value of production belongs to households which hold the most advantages in
labour, capital, land, and social relation. They are the better-off in the community. It is
noteworthy that no households of the sample have savings in the formal sector. Only 10
households have the small savings by lending to their friends or relatives without interest.
According to the 2005 annual report by VBARD, of the seven mountainous communes
(including Tan Linh) with 15 thousand households, only 6 percent have deposits in
VBARD. From Table 3, it is evident that raising livestock provides major income source
in all household groups while the value of cultivated products sold in the market is really
modest.
The result of household survey is completely consistent with the result of group
discussion and what was found from reports of the commune.

29
6 RESULTS AND DISCUSSION

6.1 Participation of rural households in the credit market in

Tan Linh Commune

6.1.1 Credit demand in Tan Linh Commune

In the group discussion, farmers discussed together and reported that in order to increase
income sources and improve living standards, they have to expand their production. In
order to expand their production, they need capital for investment. Initially, they often
think about available funds including savings and agricultural products that can be
converted in to cash easily. Very often, these funds are not enough to finance for their
production on a larger scale and they have to rely on capital coming from outside the
household. Therefore, for local people, credit demand happens to be an indispensable
condition for expanding production. Moreover, all of them said that lean harvest, disease
and risks could happen any time, so they also have to borrow to cover consumption
expenditures at critical times.
Credit demand was calculated by the amount of loans applied by household from all
credit sources plus demanded amount of credit that has not been asked yet by households.
In Table 4, we can see that average amount of total credit demanded per household is 17.4
million VND in which the average amount of credit they applied to credit sources is 11.1
million VND (making up more than 63% of the total). The remaining 37 %, though still
demanded, has not been applied to lenders because of many reasons that will be presented
in the following parts. It is worth noting that there was a sharp difference in the credit
demand between three kinds of household. The group of better-off households has the
largest amount of credit demand. On average, each better-off household wanted 21.7
million VND for both consumption and production. The ratio of the amount of credit
demand that had not been applied by better off households to the total credit demand that
they desired was smallest. It implies that when they need credit they did not hesitate to ask
the lenders for borrowing money. This contrasts with poor households’ situation. The
average amount of credit demand of a poor household was the smallest at 12.8 million
VND. It is interesting to note that the ratio of the amount of credit demand that had been
not applied for by poor household to total credit demand that they actually need from all
credit sources was the biggest, more than 50%. This indicates that poor households did not
dare to ask lenders for credit possibly because they were afraid of refusal or risks in their
production although actually they had demand for credit. Between better-off households
and poor households were medium households whose credit demand was higher than the
poor household but lower than the better-off households.

30
Table 4: Credit demand of local farmers (million VND)

Credit demand Total sample Better-off Medium Poor

(N=100) (n=19) (n=54) (n=27)
mean sd mean Sd mean sd Mean sd

Average amount of credit was asked

11.15 11.26 18.08 14.64 11.10 11.00 6.36 5.41
by household from all credit sources

Average demanded amount of credit

6.00 6.24 3.63 4.99 6.59 7.41 6.48 3.76
has not been asked by household

Total demanded credit per household 17.15 13.67 21.71 16.56 17.69 14.60 12.84 7.24

Source: Household survey in Tan Linh Commune. Nguyen Thu Phuong, 2006

The question is why there are differences in credit demand among the groups. To
answer this question, data collected from the discussion of the village head group about
characteristics of the three kinds of households were analyzed for interpretation. The
group of 11 village heads discussed and came to conclusion that better-off households
developed livestock on a much larger scale than the others. These households are
interested in investing in intensive production, doing business, leasing agricultural
machinery and tourism services. Hence, they are often the households that had demand for
largest amount of credit to finance business and production activities. Village heads
reported that better-off households are the only ones that have accumulation in savings
and insurance; their fixed capital in the form of production instruments and farming
machinery is biggest. Besides the economic benefits, there are other advantages in terms of
health, education, social relations and information access which could make them more
credit-worthy in lenders’ eyes. These potential advantages explained why they do not
hesitate to raise their credit demand to lenders. The income and livestock raising scale of
medium household is much less than those of better-off but there are not substantial
differences regarding social relations, information access, education and health care
between better-off households and medium households. The only significant finding is
that most of the children of medium households are at the age of schooling. The
implication is that school fees may become a burden to their budget and it could hinder
them from having as much credit demand as better-off households for intensive livestock
production. On the contrary, results of group discussion showed a different picture of poor
households. They are households that have the lowest income level, lowest value of fixed
capital and lowest capacity in livestock production. They often have small children, elderly
and seriously sick persons to take care of. Because of poor economic conditions, they
could hold an inferiority complex, keeping away from contact with community so their
social relations are narrow. In addition, their capacity in production is weakest so they
demand the smallest amount of credit in comparison with the other households. They
even do not dare to voice their demand for credit to lenders because they were afraid of
being turned down.
Of the 100 households surveyed, 93 households said that they had demand for credit.
In 93 households that had demand for credit, 29 households got an amount of credit that
equals their demand. The remaining 64 households had a gap between demand and credit

31
secured. These figures, along with the analysis of total credit demand presented above
shows the strong demand for credit by local people.

6.1.2 Diversity in purposes of credit demand in Tan Linh

commune.
It is necessary to study what purposes the local people demand credit for. According to the
result of farmer group discussion, they need credit for many different purposes in which
credit for raising livestock is the most common in the community. Besides, some
households demand credit for purchasing farming machines to lease and serve their own
production. Some others need credit to provide tourist services such as restaurant,
boarding houses, and souvenir shops. It is also worth noting that credit plays an important
role in consumption purposes. Members of groups discussed and came to the conclusion
that the local people also need credit in the emergencies to pay for foods, school fees,
medical fees, and former debts. Some of them want to borrow money to buy durables or
to build a house. These showed that the purposes of loans in the community are really
diverse relating to both consumption and production.

Table 5: Borrowing purposes by wealth categories

Purposes Sample % of loans

Number of loans % of loans Better off Medium Poor
I. Farm/business
1. Raising livestock 84 47 20 52 27
2. Breeding facilities 8 4 13 75 13
3. Purchasing farming machinery 3 2 33 67 0
4. Doing business 8 4 13 88 0
II. Consumption
5. Medical fees 15 8 7 60 33
6. School fee 18 10 22 61 17
7. Durable equipment 3 2 0 67 33
8. House building 9 5 33 44 22
9. Food 9 5 11 33 56
10. Weddings and funerals 14 8 0 64 36
11. Paying debts 7 4 29 14 57
12. Fee for applying for children’s job 2 1 100 0 0
III. Total number of loans 180
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006

In table 5, the result of household survey showed that there are 84 loans for raising
livestock coming from both formal and informal sectors, accounting for 82 percent of total
loans for production purposes. The number of loans for raising livestock of better-off
household group was the smallest, accounting for 20.2 percent of total loans for livestock
production but this figure among the medium households was biggest, at 52.4 percent. It
can be explained by the fact that some better-off households, which did not have
limitations of self-finance, could expand their livestock raising scale by their savings and

32
products sold in the market. So their number of loans for raising livestock was smallest. In
the meantime, medium households could be the households that wanted to increase their
scale of livestock farming but they lack accumulation for investment. So their loans for
livestock raising is biggest.
27.4 percent of total loans for raising livestock were borrowed by poor households.
The figure showed that poor households needed more loans for livestock farming than the
better –off who have good self-finance but less than medium households who have
potential for increasing the scale of livestock raising. Loans for the construction of
breeding facilities accounted for 4.4 percent of total loans in which 75 percent were
borrowed by medium while those of poor households and better-off households are equal,
at 12.5 percent. This could be explained by the fact that better-off household had built
solid breeding livestock facilities while medium households started building, and poor
households have not enough capacity to invest into breeding livestock facilities as a large
scale. Investment into leasing farming machines and doing business is rather new. It seems
that only some better-off and medium households have grasped these opportunities. In
Table 5, we see poor households did not have any transactions for such purposes. It is
interesting to recognize that there are no transactions for cultivation purposes. The same
result also was found in data collected from group discussion. This unusual fact will be step
by step interpreted in following sections.
Loans for consumption purposes (Table 5) show that school fees, medical fees,
weddings and funerals are the most common reasons. Purposes for house building and
food were found in all groups. Most of these loans were borrowed by poor households
and medium households. Poor households hold the majority of transactions for purposes of
buying food and paying urgent debts. This suggested that poor and medium households
had to borrow to finance their basic needs while better-off households could cover such
expenditure without having to borrow.

33
6.1.3 Financial institutions and current access to credit of
local people
The above sections have been spent discussing credit demand and diversity in borrowing
purposes. In this part, we continue to study the financial institutions that are providing
loans to satisfy local people’s credit demand, and examine current access to credit of local
households. The rural credit market in Vietnam has a dual structure in which formal and
informal finance exist side by side. The formal sector is led by the Vietnam Bank for
Agriculture and Rural Development (VBARD) that has been evolving to cover an
increasing share in the market (Izumida & Duong, 2001). Data collected from surveying
randomly 100 rural households in Tan Linh commune showed the same result to their
findings. In Table 6, we see there are 44 credit transactions of VBARD in 2005,
accounting for 76 percent of the total transactions from all formal sources, with the
average borrowing amount per household being 7.6 million VND which is much larger
than those of VBSP and public funds (Woman Union Fund, Farmer Union Fund, etc.
More than 90 percent of total credit transactions of VBARD are to finance livestock
production while purposes of borrowing of local people are really diverse as mentioned
above. The implication here is that many borrowing purposes other than livestock raising
would not be supplied by VBARD. It is also interesting to note that there are different
borrowing levels among three kinds of household groups. Better–off households which
have demand for the largest amount of credit discussed above have the biggest average
loan size at 13.5 million VND. However, this figure was much less than that they desired
(Table 4). In addition, they only occupied 18 percent of total households borrowing from
VBARD. Medium households constituted 61.4 percent of total households borrowing
from VBARD with an average loan size at a modest 6.6 million VND per households
while their demand for credit presented in Table 4 is 17.7 million VND. It means that to
obtain credit amount as they demanded, better-off and medium households have to
borrow from other financial sources rather than the formal sector.

34
Table 6: Credit sources and average amount of loans

Total surveyed households: 100

Number of poor households: 27
Poor Medium Better-off
Financial % of poor Average % of Average loan % of Average loan
Total
sources HHs to loan size medium size better-off size
number of
total of (million HHs to (million VND) HHs to (million VND)
borrowing
borrowing VND) total of total of
households
HHs borrowing borrowing
mean sd mean sd mean sd
HHs HHs
Formal sector 58
1. VBARD 44 21 5.50 2.35 61 6.59 2.48 18 13.50 12.57
2. VBSP 10 30 3.33 1.53 40 5.75 0.96 30 4.67 2.08
3. Funds 12 25 0.17 0.06 67 0.93 0.98 8 0.20 -
4. Project 3 0 0.00 0.00 67 22.50 10.60 33 15.00 -
Informal
sector 79
1. Relatives 31 39 3.82 4.92 39 5.17 4.82 23 19.57 14.18
2. Friend 47 23 2.12 2.44 62 2.46 1.86 15 2.76 2.77
3.
30 37 2.14 2.71 43 7.15 10.46 20 3.83 2.48
Moneylenders
Source: Household survey in Tan Linh commune, Nguyen Thu Phuong, 2006

Poor households occupied 20.5 percent of total credit transactions from VBARD with
smallest average amount of credit in comparison with three groups, standing at 5.5 million
VND per household. Other formal lenders in the commune provide really limited loans
regarding both the number of transactions and loan size. In 100 surveyed households,
there are only 10 households who got loans from VBSP, 12 households who borrowed
from public funds and three households who borrowed from the project that support dairy
cow raising. Another thing that should be noted is although the VBSP loans target the
poor households, from the results of the household survey, of total of households got loans
from VBSP, up to 70 percent are medium and better-off households while only 30
percent are poor households with the smallest average amount of credit of roughly more
than 3.3 million VND. Besides, of all the households that got loans from public funds,
only 25 percent are poor households with smallest average amount of credit. For loans
coming from cow-raising projects, no poor household is involved. This implies that the
support programs possibly do not start with the poor. Interestingly, of the three kinds of
households, the most active households in the rural market are not better-off households
but medium households who have the largest demand for numbers of loans as analyzed
above. Medium households are active in both formal and informal sectors. As can be seen
from Table 4, they occupied 38.7 percent of total households borrowing from relatives,
61.7 percent of those borrowing from neighbours and friends and 43.3 percent of those
borrowing from moneylenders with average size of loans are larger than those of the rest.
On the contrary, generally better-off households hold a modest ratio of participation in
informal rural market. However, in the case of having demand for credit, better-off
households often got loans of a large size from the informal sector, a well illustrated by the
importance of relatives, as a source of credit.
In summary, the local credit market was constituted by both formal and informal
sectors. There were sharp differences in access to credit among three kinds of households

35
in which medium households were the most active in the market while poor households
had the most limited access to credit in formal sector as well as informal sector.

6.1.4 Interaction between formal and informal sector

In this study, the sources of credit were grouped into two categories. Formal lenders
include the Vietnam Bank for Agricultural and Rural development (VBARD), Vietnam
Bank for Social Policies (VBSP), public funds (fund for Woman Union, fund for Farmer
Union, etc.) and projects. Informal lenders include relatives, neighbours and friends and
moneylenders.

Table 7: Numbers of loans by credit sources

Source Number of loan Percent of total

I. Formal sector 72 40
1. VBARD 44 24
2. VBSP 10 6
3. Funds 15 8
4. Project 3 2
II. Informal sector 108 60
1. Relatives 31 17
2. Friend 47 26
3. Moneylenders 30 17
III. Total (I+II) 180 100
Source: Household survey in Tan Linh commune, Nguyen Thu Phuong, 2006

In the previous sections, we have discussed and found out that rural households in Tan
Linh commune participated in both formal and informal institutions. The next question
that should be raised is how important is the informal sector as a source of credit. This
question will be examined by looking at the proportion of loans taken from the informal
sector (Table 7). Of the 180 loans of 100 households that were randomly surveyed, about
40 percent came from formal sources. Among formal sources, VBARD is the major source
that constitutes about 24 percent of loans. Next, it was VBSP which occupied 6 percent of
loans. Table 7 shows that 60 percent of total loans came from informal sources. Among
informal sources, relatives and friends constitute 43 percent of loans with interest rate
equaling zero. 17 percent of loans came from moneylenders at an average interest rate
higher than those of VBARD. This implies that local people still depend considerably on
informal sources for their credit demand.

36
Figure 2: Sources of borrowing

Column
1 2 3 4 5 6 7 8 9 10

2 9 hhs

3 (Non borrowers)

4 12 hhs

5 (Formal sector only)

Row

6 33 hhs

7 (Informal sector only)

9 46 hhs

10 (Both formal and informal sectors)

Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong 2006

Again, the borrowing simultaneously from both sectors and domination of informal
sector were reported in Figure 2. Of 100 households sampled 79 percent had transactions
with the informal sector, 58 percent of those households had credit transactions with the
formal sector and 46 percent of those surveyed households had transactions with both
sectors. This can be explained by the rationing action of formal lenders and borrower’s
choice for a particular loan.
It is pertinent to ask what the borrowing from the informal sector is used for. Are
informal loans used to finance the purposes that were not provided by formal lenders?
Table 8 appears to indicate that this is indeed the case. Loans are classified into two
categories. Farm/business loans include raising livestock, purchasing farming machinery,
and doing business.
Consumption loans include medical fee, durables, school fee, housing building,
weddings and funerals, paying debts and fee for applying for children’s job. Table 8
indicates that business loans were financed by both sectors. Regarding business loans,
raising livestock purpose has strongest demand with a dominating proportion of 67 percent
of total loans from the formal sector. Total loans for other business purposes are really
small and financed mainly by the informal sector. It is easy to see that all of loans for
consumption purposes came from the informal sector.

37
 Table 8: percent of loans by formal and informal sectors

Business loans Consumption loans

Raising livestock

farming machinery

Medical fees

funerals

for children’s job

Fee for applying
Purchasing

Doing business

School fee

Durable

House building

Food

Weddings and

Paying debts
Sources

Formal sector 62 2 3 0 5 0 0 0 0 0 0
Informal sector 30 1 5 15 14 2 9 9 14 7 2
Percent of total
Formal sector 67 67 38 0 26 0 0 0 0 0 0

Informal sector 33 33 63 100 74 100 100 100 100 100 100

Source: household survey in Tan Linh Commune, Nguyen Thu Phuong 2006

Table 8 shows that all loans for food, medical fee, house building, wedding and
funerals, etc were provided by the informal sector. Only loans for school fees were
financed by the formal lenders, but only 26.32 percent of total of those loans. An
explanation for this special case was that VBSP has lending policies that enable poor
households to send their children to schools. The finding here is that formal lenders in the
commune only provide loans for business purposes, mainly for raising livestock while
borrowing purposes of farmers are diverse. In other words, the source of a loan, from
formal or informal sector, depends on the purpose of the loan. It means that farmers are
rational in deciding which lending source to use for a particular borrowing purpose.

6.1.5 Factors of borrowing of rural households

A credit contract including many terms such as the amount, the duration, the purpose, the
collateral and the repayment schedule that are set up by agreement between borrower and
lender. This implies that factors influencing the level of access to credit of households
regarding number of loans and loan size would come from both sides of the contract:
borrowers and lenders.
We begin with factors that were derived from borrowers by the question as of what are
the factors that affect the borrowing by households. When there is a demand for credit, a
household may borrow from the formal or informal sector. In many cases, they can
borrow from both sources at the same time, depending on their demand and perceived
relative probabilities of obtaining credit from each sector. To examine factors of
borrowing by households from each sector, the sample was decomposed into two main
borrowing sources: formal and informal. Among the sample households, 58 households
borrowed from formal sector while 79 households borrowed from the informal sector.
In Table 9, the level of borrowing from formal sector is classified into five groups:
group 1 includes households that borrowed less than three million VND; group 2 includes
households who borrowed from three million to six million VND; group three: from six
million to nine million VND, group 4: from nine million to 15 million VND, and group
5: more than 15 million VND.

38
 a,b
Table 9: Indicators of borrowers by the level of borrowing from formal lender (Duncan
sig. 0.05)

Indicators Unit <3 million 3-6 million 6-9 million 9-15 million >15 million
n=8 n = 20 n = 12 n = 11 n=7
Mean SD Mean SD Mean SD Mean SD Mean SD
10.8 49.7 48.8
1. Age of HH head Years 47.25 9.24 53.17 13.66 7.49 47.14 7.67
6 5 2
school
2. Education of HH Head 8.75 1.16 8.00 1.45 7.50 3.06 8.55 0.93 9.71 1.60
years
3. Family size People 5.38 1.19 4.85 1.31 4.58 1.73 4.45 1.21 5.29 0.95
4. Number of labor People 3.25 1.04 3.40 1.19 3.25 1.36 2.64 0.92 3.86 0.69
5. Number of dependants People 2.25 1.67 1.45 1.23 1.33 0.78 1.82 1.33 1.43 0.98
6. Number of the sick People 0.13 0.35 0.20 0.52 0.00 0.00 0.00 0.00 0.00 0.00
7. Social status 0.13 0.35 0.15 0.37 0.17 0.39 0.18 0.40 0.43 0.53
a a b
8. Reputation 1.75 0.71 2.00 0.73 1.83 0.39 2.00 0.63 2.57 0.53
2 15.0 15.0
9. Total farming area 360m 12.00 5.00 10.00 12.00 8.00 8.00 14.00 7.00
0 0
10. Total value of Million 21.72 21.0 31.6 43.6 46.14 30.2
a 18.74 35.78 26.43 28.00 b
livestock VND 0 8 2 2
Million
11. Total informal credit 4.50 4.50 4.12 5.27 2.42 2.8 3.73 4.15 4.23 4.11
VND
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006

Table 9 shows a difference of borrowing level by households from the formal source.
Of 58 households who borrowed from formal lenders, only seven borrowed more than 15
million; however, almost all of their indicators are superior to the rest. They have biggest
number of labourers, highest social status and reputation. The owners of these households
fulfil the highest level of formal schooling, and they are also the youngest. Total
production value of livestock of household that can be considered as proxy for household’s
capacity for livestock production stand at the highest compared to the rest. There is not
any sick person in these households and their number of dependants is rather low, at the
second position in the five categories of borrowing level. The question is why they have
the biggest borrowing from formal lenders. These households have great number of
advantages regarding education, labour, age, and capacity for livestock production. Such
advantages could motivate them to expand their production on a larger scale. So they have
biggest demand for credit. Interestingly, they are also the ones that obtain the biggest
amount of loans that also depend on the lender’s decision. Having demand for credit is
completely different from taking loans from the bank. In many cases, farmers have demand
for credit but they cannot obtain credit from formal lenders. This could be because they
did not ask lenders for borrowing for fear of being refused, or actually they could have
been turned down by formal lenders. It implies that in order to take a loan from any
source, firstly farmers have to claim their credit demand. It is time now to revisit the issue
of households taking the biggest loan from the formal lenders and answer the question
why they dare to claim their credit demand while some others do not although they
actually need it. The three indicators of reputation, social status and total production of
livestock are analyzed to find out their differences. Reputation is the indicator reflecting
the households’ economic position as well as their prestige in a village; social status is the

39
indicator defined as 1 if the head holds any position in the community, and as 0 if not;
total production value of livestock reflect household’s capacity in raising livestock. In
Table 9, regarding the households that borrowed a biggest amount of loans, their social
status and reputation are biggest. Besides, they do not have the sick in their family to take
care of, and total production value of livestock is much larger than the rest. These all could
make them more confident in asking formal lenders for borrowings. At the same time,
such advantages make them more credit-worthy in lender’s eyes. So their probability of
taking loans from formal lenders is highest.
Now we look at the households that borrowed with smallest amount of credit. There
are eight households belonging to this group. We compare characteristics of households
borrowing at smallest amount of credit with households that borrowed at the biggest
amount of credit to find out the differences between them. Table 9 shows that there are
not substantial differences in terms of age, education level, and family size between the
households who borrowed less than three million VND and household that borrowed
more than 15 million VND. What is worth noting here is the difference in terms of
number of number of dependants, number of the sick persons, reputation, social status and
total production value of livestock of these households vis-à-vis other households,
especially those that borrowed with the highest amount of credit. They have the highest
number of dependants, the highest number of the sick person, and the lowest total value
of livestock production. These imply that they may have had to borrow money for
consumption purposes rather than production purposes. But as discussed earlier, loans are
seldom sanctioned by the banks for purposes other than production, except for some cases
that are in line with specific directives from the government. So they hardly take
consumption loans from formal lenders. On the other hand, their indicators of reputation
and social status are lowest which are in the sense that they are poor and do not hold
position in the community. Such disadvantages could make them less confident in
claiming their credit demand to formal lenders as well as being less credit-worthy in the
appreciation of lenders. So their borrowing level from formal sector is smallest.
Next, we find out the difference between households that borrowed less than three
million VND with households that borrowed from three to six million. What is most
noteworthy is that these households have fewer dependents and sick people than those that
borrowed less than three million VND. Meanwhile, their indicators including the number
of labourers, social status, reputation and total production value of livestock, exceed those
of households with smallest amount of borrowing. Consequently, these result in bigger
loan amounts borrowed.
By comparing indicators between households that borrowed from three million to six
million VND with those borrowing from six million to nine million, again we can see that
the number of sick people could be an important factor that influences borrowing by
households from the formal sector. Table 9 shows that although households belonging to
group 3 have lower education level, fewer labourers, lower reputation, lower social status
than those coming from group 2, they obtained more credit from formal sector. It is
worth seeing that there is not any sick person in their family. It implies that number of
sick person could be a factor hindering borrowing from the formal sector by households.
In summary, looking at overall cross-tabulation of Table 9, we can see that, under the
increase of borrowing level from less than three million VND to more than 15 million
VND, indicators of number of sick person and number of dependents tend to decrease,
while indicators social status, reputation and total production value of livestock have a
tendency to increase. It implies that the number of dependents, sick person, reputation,
social status and total value of livestock could be the factors influencing borrowing from

40
 formal lenders by rural household in Tan Linh commune. Duncan test shows that only
reputation and total value of livestock have statistically significant differences at 5 percent
among group 1, group 3 and group 5. The other indicators including the number of
dependents and sick people and social status are statistically insignificant.
A similar method of analysis was also employed to examine factors influencing
borrowing from informal sector by households.

Table 10: Indicators of borrowers by the level of borrowing from informal lenders
a, b
(Duncan sig.0.1)

Indicators Unit <1 million 1 -3 million 3-10 million 10 million

n=4 n=29 n=29 n=17
MEAN S.D. MEAN S.D. MEAN S.D. MEAN S.D.
Age of household head Years 46.00 6.32 46.31 10.97 46.31 11.98 44.29 7.42
Education of household head School years 9.25 1.26 7.97 1.72 8.31 2.07 8.82 1.81
Family size People 4.75 0.96 4.03 1.30 5.10 1.11 4.88 1.63
Number of labor People 3.00 1.15 2.55 1.06 3.14 0.95 2.82 1.42
Number of dependent People 1.75 1.71 1.62 1.27 2.00 1.07 2.06 1.34
Number of sick person People 0.00 0.00 0.14 0.44 0.14 0.36 0.25 0.58
Social status 0.00 0.00 0.10 0.31 0.21 0.41 0.18 0.39
a a b
Reputation 2.25 0.50 1.62 0.73 1.93 0.59 2.24 0.75
Total formal credit Million VND 4.50 3.70 3.31 3.90 5.60 8.47 4.01 6.47
b a b b
Total value of livestock Million VND 44.01 19.25 19.66 15.57 39.36 25.32 48.08 42.04
2
Total farming area Sao (360m ) 15.67 2.95 8.87 5.79 14.04 7.53 14.96 8.97
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006

Level of borrowing from informal lenders is classified into four groups. Group 1:
households that borrowed less than one million; group 2: households that borrowed from
one to three million; group 3: households that borrowed from three to ten million; and
lastly group 4: households that borrowed more than 10 million. Of the total 100
households surveyed randomly, 79 borrowed from informal lenders. Of 79 households
who borrowed from informal lenders at the same time 46 households borrowed formal
lenders.
As analyzed earlier, informal sector provides loans for both purposes of consumption
and production as loans for smoothing consumption cannot be obtained from formal
sector. This means that informal sector covers all loans for consumption purposes. In table
10, firstly we look at group 1 which borrowed a smallest amount of credit from the
informal sector. The outstanding characteristics of these households were that the
dependency ratio is lowest, the reputation indicator which was used to measure their
economic status and prestige is highest, and there is no sick person in their family. These
indicators reflect that their economic status was better than the others, with fewer
dependents and less expenses spent taking care of sick person. In addition, it can also be
seen that, their production value of livestock is rather high, standing second of all groups.
These imply that they could satisfy their basis needs and expand their production without

41
having to borrow at critical times. So they do not have to rely much on the informal
sector.
Let’s now observe the characteristics of group 4 which include households who
borrowed the biggest amount of credit from informal lenders. Although these households
have advantages of high social status and reputation to take the loans from formal sector,
the actual amount of loans coming from formal lenders is far from meeting their credit
demand. This can result from the fact that their dependency ratio and number of ailing
family members are the highest. It is completely consistent with data collected by
interviewing bank staff. Both of the bank clerks who were interviewed, one from
VBARD and the other come from VBSP, said they hardly lend to households who have
seriously sick members because in such cases risk of loss is high. In that context, informal
sector seem to be a logical alternative for such household to get loans for production
activities and consumption. However, the domination in the informal sector of these
households could be due to cheaper credit or even credit with zero-interest which is
available from alternative informal sources including relatives, friends and moneylenders.
The results of the survey show that of 108 loans coming from informal sector, 78 are loans
with zero interest, accounting for 72 percent of total informal loans.
Next, we temporarily omit group 1, and make comparison among the remaining three
groups. Under the increase of borrowing level from three million to more than ten
million, indicators of dependency ratio, number of sick person tend to increase. This
implies that the households with many dependents and sick person could have had to
borrow with a bigger amount of credit from informal lenders to smooth their
consumption. Besides, making comparison in terms of the livestock value indicators of the
three groups, we found that the bigger the amount of credit from informal lenders, the
greater the production value of livestock. It is consistent with the finding mentioned in
Table 8, up to 33.4 percent of total loans from informal sector is to be spent on
business/farm purposes mainly for raising livestock. So the dependency ratio and decision
for expanding livestock production could be the factors influencing borrowing level of
households from informal sector.
In summary, the dependency ratio, the total value of livestock and the number of sick
person are likely to be the main factors that influence the amount of loans that households
borrow from the informal sector. Duncan test shows that the difference of the total value
of livestock between group 2 and the other groups is statistically significant at 10 percent.
The differences of number of sick persons and dependency ratio among groups are
statistically insignificant.

42
6.2 Formal lender’s behaviour in responding to credit
demand of local people

6.2.1 Credit rationing situation

With respect to the formal institutions, there are two major players in Tan Linh commune
today: the VBARD and the VBSP. VBARD is the sole state-owned commercial bank in
Tan Linh commune with the main business being to provide financial services to the rural
sector. The lending interest rates of the VBARD to the rural sector are basically
determined by the same level as those of other commercial banks. On the contrary VBSB
lends subsidized credit to poor households who are defined and determined under strict
criteria regulated by the government. However, loans coming from VBSP are really
limited regarding loan size and the number of loans, and they are distributed under
government planning. Besides VBSP, other formal sources including credit projects of
NGO and funds such as those raised by the woman union and farmer union do not appear
to have a lot of weight in Tan Linh commune. Hence, other than the VBARD, these
institutions will not be referred in the credit rationing analysis.
Of 100 households surveyed randomly, 73 households said they had demand for credit
from VBARD. However, of these 73 households, only 48 had actually applied to
VBARD for borrowing while the rest of 25 households had not because they were afraid
of being refused. Of 48 households who applied for credit from VBARD, only 22
households (46%) obtained the credit amount they applied for; 22 households said the
actual credit amount they received was less than that they applied for; 4 households had
their application turned down by VBARD. So obviously credit rationing by VBARD
exists in Tan Linh commune. It is interesting to have interviews with some farmers that
were rationed by VBARD. A farmer told his story when he wanted to buy a productive
cow costing about 12 million. He had available two million VND that he got from the
sale of 50 chickens. He applied to VBARD for a loan of 10 million but after visiting his
farm, the bank clerk only agreed to lend him five million, which was not enough to buy a
12-million VND cow so he had to borrow two more million from his uncle to buy a
nine-million cow. Another story came from a farmer who wanted to raise pigs on large
scale. He said his friends had succeeded in raising pigs at large scale so he wanted to follow
his example. He saved enough money to invest into breeding facilities, but he needed
credit from VBARD for breeding pigs and feed. He applied to VBARD for the credit
amount of 40 million that was accompanied by a detailed report of production and
repayment plans. But his application was rejected after the bank had assessed his economic
condition. The concerns from the two above conversations as well as other credit rationed
households were probed deeply in the interview with bank staff. The vice director of
branch at commune level of VBARD said that her branch takes responsibility for lending
to seven mountainous communes of Ba Vi district including Tan Linh commune. Due to
limited number of bank staff, one bank officer have to undertake all work such as
receiving applications, assessing economic condition, lending required loans, collecting
debts and monitoring loans of borrowers in the whole commune. Tan Linh is an extensive
area with 2,218 households, so with a limited number of bankers, it is really difficult to
determine whether a client is credit-worthy or not. It takes too much time and expense to
collect accurate information about the client. Hence, in the case of a household that shows
signs of high risk in repaying loan, VBARD staff often refused to lend or lend them with

43
an insufficient amount of credit, less than what client required. The question then is what
factors contribute to households being credit-rationed by VBARD?

6.2.2 Factors of credit rationing by VBARD

With respect to participation in borrowing from VBARD, the 100 households in the
survey are classified into four groups. Group 1 includes households that took the loans
with an amount equalling their applied sum. Group 2 includes households that took the
loans whose sizes are smaller than what they applied for. Group 3 includes households that
did not applied to VBARD although they had demand for credit from VBARD. Group 4
includes households that had no demand for credit from VBARD.
To get the loan from VBARD households having demand for credit first have to apply.
Based on the households’ application, the bank will make an assessment of potential
client’s situation to decide whether or not to provide a loan of the demanded amount by
the households. It means that a credit transaction carried out depends first on the
borrowers’ demand for credit and their application to the bank; second on the bank’s
assessment and decision about the application. Two questions arise here: what are the
factors that influence the decision of the bank in rationing credit? Is it true that these very
factors hinder actions of households in applying to VBARD for credit?

44
 a, b
Table 11: Means of indicators by credit status of households to VBARD (Duncan sig.
0.05)

Indicators Applied to VBARD for loan Non-applied to VBARD for loan

Non-rationed Rationed Credit demanded HHs not
households households households demanded credit
n=22 n=26 n=25 n=27

Mean Sd Mean Sd Mean sd Mean Sd

Age of household head 51.09 12.91 48.73 8.00 41.24 11.41 44.93 6.13
Education of household head 8.27 2.45 8.31 1.46 8.32 1.80 8.74 1.43
Family size 4.82 1.47 4.88 1.37 4.40 1.19 4.73 1.12
b b a
Number of labor 3.41 1.22 3.00 1.23 2.48 0.87 2.81 0.92
Number of dependent 1.45 1.22 1.88 1.21 1.92 1.38 2.22 1.40
Number of sick person 0.00a 0.00 0.27b 0.53 0.22a 0.52 0.00 0.00
Social status* 0.27b
0.46 0.12 0.33 0.04 a
0.20 0.11 0.32
Reputation** 1.82b 0.66 2.15b 0.54 1.44a 0.58 2.22b 0.64
b b a b
Owned farming area 10.02 8.54 8.34 5.15 3.81 3.95 8.13 7.24
LUCs*** 0.55b 0.51 0.58b 0.50 0.20a 0.41 0.70b 0.47
b a b
Total value of livestock 34.39 22.96 45.28 38.59 19.02 23.57 45.80 25.47
Total assets 47.66b 29.71 58.54b 41.10 27.67a 28.66 60.97b 27.91
Total savings 0.82 3.84 0.35 1.41 0.00 0.00 2.32 6.33
Total informal credit 2.99 3.22 7.46 12.44 4.04 4.18 5.09 8.62
Total formal credit 8.08 8.59 7.46 4.95 1.42 6.00 1.61 3.59
Total credit 11.07 10.24 14.92 11.23 5.46 6.61 6.70 9.10
Total credit demand applied to VBARD 7.80b 8.60 15.08c 8.70 0.00a 0.00 0.00 0.00
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006
*Social status reflects position of household head in the society. It is defined as 1, if
household head hold position, as 0 if not.
**Reputation reflects the household’s economic position as well as prestige. It is defined as
1 if household is poor, as 2 if household is medium and as 3 if household is better-off. The
classification was conducted by village heads under the criteria discussed by them in the
group discussion.
***LUC reflects the ownership of land use certificate of a household. It is defined as 1 if
household has LUC, as 0 if not.

To answer the two questions above, we began by making the comparison between
non-rationed households and rationed households. Looking at Table 11 we can see that
credit rationed households demanded credit from VBARD twice as much as non-rationed
households. In addition, the total value of livestock of rationed households is much bigger
than that of non-rationed households. On the other hand, as mentioned earlier, raising
livestock which occupied 91 percent of the total loans from VBARD is the key source of
income for local farmers. This implies that rationed households demanded credit from
VBARD to expand their livestock production to a larger scale. The indicator of total value
of livestock reflects the ability of a household to repay the loan in the sense that livestock
can be sold easily to convert to cash in case a due date approaches. Hence, generally the
higher the value of livestock, the greater the chance of repaying the loan. However, this
indicator only is never sufficient for lender to make their decision to lend. It becomes

45
clearer when comparisons are made between group 1 and group 2. Compared with those
in group1, households in group 2, though having higher total value of livestock which is
expected to increase the probabilities of repaying the loans, are actually those who are
rationed by the bank. This is also the case when it comes to indicators of reputation and
total assets value. Although group 2 has bigger total assets value and higher reputation
which reflect their better economic situation and prestige, households belonging to this
group are credit-rationed by VBARD. This is completely consistent with what bank staff
said in the in-depth interview. The bank officer put it that the decision to provide a loan is
often based on the combination of a series of different observations to assess a client.
According to her own as well as her colleagues’ experience, the level of wealth which is
reflected by the total assets value and production scale is not enough to guarantee the
safety of a loan.
Returning Table 11 to continue the comparison of other indicators between group 1
and group 2, we see that there are no substantial differences regarding the household
head’s age and years of formal schooling between the two groups. It is worth noting that
although there is no difference between the two groups in terms of family size, rationed
households group (group 2) have smaller number of labourers and bigger number of
dependants than those of non-rationed households group, which means the dependency
ratio of rationed households is higher than that of non-rationed households. This poses
the risks of using loan for purpose other than production and lacking labour to carry out
effectively the expansion of production. Higher dependency ratio partly explained their
limitation in terms of savings and insurance compared with households in group 1.
Moreover, households belonging to group 2 have the highest number of seriously ill
persons which necessitates the use of the loan for buying medicine and therefore implies
unstable economic situation. Duncan test shows that the difference of the number of sick
person between rationed group and non-rationed group is statistically significant at 5
percent. For that very reason, in all likelihood the number of sick person and number of
dependant are important indicators which contribute to households’ possibility of being
rationed by VBARD.
Another important reason for rationing by VBARD concerns the amount of credit
households applied for. Evidence can be found in the fact that households applying for an
average credit amount of 15.6 million VND are rationed by VBARD while households
belonging to non-rationed group (group 1) only applied for on average 7.8 million VND.
The difference in the applied amounts between the two groups is statistically significant at
5 percent. Under the regulations issued by the State Bank of Vietnam (SBV), rural
households are allowed to borrow up to 10 million VND without collateral from
VBARD. For loans whose sizes are more than 10 million VND, borrowers have to
undergo collateral procedures. In Vietnam, the Land Use Certificate (LUC) is the main, if
not the only asset which rural households could offer as collateral. The problems here
include first the uncertain existence of land market because land use prices determined by
the government in LUCs are lower than those in market transactions; second, according to
bank staff’s experience it is so difficult to sell a debtor’s land in case of a default because
nobody in the community has the heart to buy the houses of a neighbour who is
undergoing a difficult economic situation. Moreover, it is even more difficult to find
buyers from urban area as well as any other places. So there are few cases when land is
liquidated in the event of a default. It may serve only as a kind of threat for rural
borrowers to comply with credit contracts (Duong & Izumida, 2002). In addition,
according to commune report, issuing LUCs to households is taking place rather slowly
with a lot of conflicts and lawsuits. Up to 31/12/2005, only 49 percent of total 1,932

46
agricultural households in the commune received LUCs. Obviously, these disadvantages
lead to the rationing of credit by the bank against loan size over 10 millions.
In summary, among households that applied to VBARD for loans, those that
demanded credit amounts of more than 10 million VND to expand their livestock
production were rationed by VBARD. These households have production scales bigger
than that of non-rationed households but they have disadvantages regarding to the number
of dependants, the number of sick persons and area of owned farming land. In addition,
because of weakness regarding their collateral, VBARD responded to their credit demand
by action of rationing credit.
It is difficult to identify “good borrowers” and this requires the bank to use a variety of
screening devices (Stiglizt &Weiss 1981). Is it possible that the use of such screening
devices results in the local households having no access to VBARD for credit? Table 11
shows it is a considerable problem that the 25 households who actually have demand for
credit from VBARD did not apply to the bank. What negative factors contribute to these
households having no access to credit source from VBARD? Indicators of reputation, total
value of livestock, and total asset value in table 11 showed that of the four groups, they are
poorest. They owned the smallest farming area, the smallest scale of livestock production,
the lowest social status and the lowest value of having LUCs, while the number of
dependents and the number of sick people are high. As earlier analyzed, the given
disadvantages are the indicators which bank staff used to assess potential risks of a client.
Hence, these lead to a serious problem affecting households’ ability to access VBARD: the
fear of having their application declined by the bank. Duncan test shows that apart from
the number of dependents, the differences of all remaining indicators including the
number of labourers, the number of sick persons, social status, owned farming area, LUCs,
total value of livestock, total assets and reputation between the unapplied credit group and
the rationed and non-rationed groups are statistically significant at 5 percent.
In addition, results from the discussion of farmer groups showed that it is the short
duration applied by VBARD in most credit contracts that really sharpens their problem.
Currently, most of the loans are for animal husbandry, mainly reproductive cows, but the
duration seems incompatible with their life cycles. The average amount of 7 to 10 million
VND that the bank lends them is only enough for buying one cow. The loan duration is
12 months while it will take at least 18 months to have and raise a mature calf, the
condition being there is no risk. Short durations make farmers worry about how they can
manage to repay the loan to VBARD within 12 months. Besides, the annual reports of
VBARD branch at the commune level also showed that VBARD has reduced sharply
medium-term loans and increased short-term loans. In 2005, the short-term loans have
increased by 15 percent while the medium-term loans have decreased by 35 percent
against those of 2003. Hence, short duration could be a reason for farmers having no
access to VBARD although they need credit to expand their livestock production.

47
 a, b
Table 12: Means of some indicators by level of rationing by VBARD (Duncan sig.
0.05)

Indicators <5 million 5 -10 million >10 million

n=8 n=10 n= 8
mean s.d. mean s.d. mean s.d.
a
Total demanded credit amount from VBARD 9.75 0.71 12.70a 2.50 23.34b 11.94
Age of household head 48.00 6.30 48.40 8.46 49.88 9.73
Education of household head 7.63 2.07 8.40 1.07 8.88 0.99
Family size 4.63 1.77 5.00 1.41 5.00 0.93
Number of labor 2.75 1.16 3.40 1.35 2.75 1.16
Number of dependent 1.88 0.99 1.60 0.97 2.25 1.67
Number of sick person 0.13 0.35 0.20 0.63 0.50 0.53
Social status* 0.13 0.35 0.00 0.00 0.25 0.46
**
Reputation 1.88 0.64 2.40 0.52 2.13 0.35
Owned farming area 8.96 4.38 8.17 5.06 7.92 6.48
***
LUCs 0.75 0.46 0.60 0.52 0.38 0.52
Total value of livestock production 31.54 28.85 45.28 26.20 59.02 56.32
Total asset 44.27 33.62 60.49 26.68 70.38 59.91
Total savings 0.25 0.71 0.7 2.21 0 0.00
a b
Total informal credit 2.06 2.64 5.45 5.59 15.38 19.86
Total formal credit 7.63 3.11 9.09 4.84 5.25 6.18
Total credit amount from both sectors 9.69 5.54 14.54 7.44 20.63 16.73
Source Households survey in Tan Linh commune, Nguyen Thu Phuong, 2006
*Social status reflects position of household head in the society. It is defined as 1, if
household head hold position, as 0 if not.
**Reputation reflects the household’s economic position as well as prestige. It is defined as
1 if household is poor, as 2 if household is medium and as 3 if household is better-off. The
classification was conducted by village heads under the criteria discussed by them in the
group discussion.
***LUC reflects the ownership of land use certificate of a household. It is defined as 1 if
household has LUC, as 0 if not.

The questions then are whether there are differences among 26 households together
belonging to the rationed group and if this is the case, what the differences are. In Table
12, 26 credit rationed households were classified into three groups according to the level
of rationing by VBARD which was determined by the applied credit amount minus the
actual credit amount that the households borrowed from the bank. Group 1 includes
households rationed by the bank at a level of below five million VND; group 2:
households rationed at a level of five to ten million VND; group 3: households rationed at
level of more than 10 million VND. First we look at households belonging to group 3
which are the most rationed by VBARD. The first interesting thing is that they had the
biggest demand for credit from VBARD. Their credit demand with average amount of
more than 24 million VND poses a stark contrast against those of the rest of two groups,
and this difference is statistically significant at 5 percent. Secondly, although their total
credit amount from both sectors is the biggest, the amount coming from formal - mainly

48
VBARD - is the smallest. Thirdly, they have the biggest value of livestock production, the
biggest value of total assets, the highest mean value of social position in the community
and a high reputation which reflect their economic status and social relationship that are
the best of three groups. The fourth noteworthy point is that they are the households with
the smallest owned farming area, the highest dependency ratio, the highest number of sick
person, the lowest value of having LUCs and have no savings.
Comparing characteristics of the other two groups, once again we see negative relation
between credit demand and level of credit rationing by the bank. A comparison between
the two groups showed that the bigger amount of credit they applied for, the higher the
level of credit rationing by VBARD they have to suffer. There are, however, positive
relations between livestock production scale, total assets value and credit demand. The
higher the value of livestock and asset, the bigger the demand for credit were applied by
households. This could be a suggestion that rationed households really have credit demand
for expanding their livestock production which could bring about an increase of their
assets. It is also interesting to see that the larger the demand for credit applied by
households the bigger the credit amount farmers borrowed from informal sector. The
difference of informal credit amount between group 1 and group 3 is statistically
significant at 5 percent, which means that rationing credit by VBARD forced them to
borrow more from informal lenders for raising livestock. Besides, it is important to note
that households belonging to the least rationed group (group1) have the lowest number of
sick persons, the biggest area of owned farming land and the highest value of having LUCs
which prove to be favourable factors in the bank staff’s approval of a loan. However,
Duncan test shows that the differences of these indicators among the three groups are
statistically insignificant.
In summary, the series of evidence given indicated that households having big demand
for credit are those who are well on their way to expand their livestock production.
However, disadvantages regarding collateral, dependency ratio, and an ailing situation,
result in their credit application being rationed by VBARD. The level of rationing could
depend on the credit amount that they applied, the dependency ratio, the owned farming
area, and the ownership of LUCs.

6.3 Credit and livestock production in Tan Linh commune

As mentioned in preceding section, people in the commune mostly rely on agricultural

production. Although all of them are involved in cultivating rice and cash crops, the
output of these activities is only enough to serve family consumption and to supplement
for feed sources. Livestock production is considered as the major source that brings about
cash income for local people. This explains why of the total of 103 credit transactions for
farm/business purposes from both informal and formal sectors, 92 transactions (89%) are
for raising livestock. Meanwhile, there is not any application for cultivation purposes
claimed by local households. This part is intended to contribute to an improved
understanding of the linkages between access to credit and agricultural productions,
particular for livestock production in Tan Linh commune. The connection between access
to sufficient amount of credit and agricultural production will be explored by first
comparing indicators of agricultural output between credit constrained households and
credit unconstrained households and second by showing differences in production among
constrained households.

49
6.3.1 Credit-constrained and unconstrained households
First and foremost, we have to determine whether a household is constrained or not.
Barham et al., 1996 defined that a household was classified as being credit-constrained if
they applied and were rejected; or did not apply due to insufficient collateral, high
transactions cost or fears of risk, or received a loan less than requested or wanted.
In this study, the determination of whether a household is credit-constrained or not is
based on survey responses under a questionnaire. Based on the result of the group
discussions and secondary data, a questionnaire was constructed with the list of all credit
sources currently available including formal and informal sources in the community.
Households were asked if they had pursued loans from each of these sources. If yes, in case
of rejection they were asked how much they applied and what the explanation for the
rejection was. In the case of acceptance they were asked how much they applied, how
much they received. If they had not pursued loans, they were asked why not. After
households had answered all the above-given questions, they were asked if credit amount
they borrowed from all sources had satisfied their credit demand for expenditures
regarding both consumption and production purposes. If it had, they are not credit
constrained. If it had not, they are credit constrained. In other words, a farm household
will be credit-constrained if they request more amount of credit than the combination of
the formal and informal market are willing to supply or they have demand for credit but
they cannot obtain it.
In that sense, the household survey showed that 64 were credit constrained, 36 were
unconstrained. In Table 13, we do not see substantial differences between credit-
constrained households and credit-unconstrained households in terms of family size,
number of labourers, farming area, education and social position of household heads
except that credit-constrained households have a greater number of sick people. The
factors that lead to sharp differences between constrained households and unconstrained
households are their credit demand and the actual amount of credit they borrowed. Table
13 also shows that constrained households demanded an amount of credit three times as
much as that of unconstrained credit households, at 21.5 million while there was no
substantial difference in the actual credit amounts between them. Further, Table 13 shows
that credit-unconstrained households have a much bigger value of livestock production
than that of credit-constrained households. In addition, their total savings, total value of
agricultural output including cultivation and animal husbandry, and non-farm income are
also considerably greater compared to credit-constrained households. These imply that
liquid sources of unconstrained households are bigger than those of constrained
households. This is the reason why the unconstrained households do not have to rely as
much on the capital outside to finance their expenditures as credit-constrained households

50
Table 13: Means of some indicators by constrained and unconstrained households

Unconstrained Constrained
Indicators Unit n= 36 n= 64

mean s.d. mean s.d.

Age of household head Years 47.81 10.01 45.53 10.48
Education of household head School years 8.47 1.93 8.39 1.71
Family size People 4.74 1.12 4.69 1.37
Number of labor People 3.00 1.01 2.86 1.15
Number of dependents People 2.00 1.37 1.83 1.29
Number of six person People 0.03 0.17 0.18 0.46
Social status 0.14 0.35 0.13 0.33
Total farming areas 13.45 5.88 12.53 8.01
Total savings Million VND 1.89 5.88 0.14 0.91
Total credit amount Million VND 8.91 12.32 9.81 8.61
Total demanded credit amount Million VND 8.91 12.32 21.53 12.06
Total value of livestock production Million VND 46.73 32.80 30.68 27.31
Total value of livestock sold in the Million VND 16.19 17.15 14.46 19.34
market
Total cash cost of livestock production Million VND 7.71 8.61 8.55 12.18
Total value of cultivation Million VND 8.85 5.31 6.73 4.54
Total value of cultivation sold in the Million VND 2.86 4.85 1.43 2.03
market
Total cash cost for cultivating activities Million VND 2.20 0.86 2.23 1.37
Non-farm income Million VND 11.42 14.21 7.70 9.20
Total value of agricultural production Million VND 55.58 34.96 37.41 29.57
Total value of products sold in the Million VND 19.06 18.19 15.89 19.65
market
Total cash cost for agricultural Million VND 9.91 8.49 10.78 12.48
production
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006

It is interesting to realise that although unconstrained households have bigger value of

livestock production than that of constrained ones, their total cash cost for agricultural
production particular for livestock production within the year of 2005 is lower than that of
credit-constrained households. It was explained by the fact that most loans were for buying
reproductive cows. For the unconstrained households, they had bought cows in previous
years which would not increase cash cost in 2005 and now these cows are reproducing,
which would increase their value of livestock production. On the contrary, credit-
constrained households could be at the starting point of the investment. Table 13 showed
that these households demanded a large amount of 21.5 million VND to finance their
production but they were rationed to an amount of 9.8 million VND which was much
lower than what they desired. They spent their limited loans buying reproductive cows in
2005 which would increase their cash cost but have not yet increased value of livestock
production because it takes time to have mature calves. Interviews with two farmers who
belonged to 64 credit-constrained households to investigate their status of production
showed that having access to insufficient loans really hindered their production plans. In
one case, the respondent said in early of 2005 he applied to VBARD for an amount of 16

51
million VND with the intention of buying a good quality reproductive cow and raising
two swines with the expectation that he would have a calf and around 48 breeding pigs in
early 2006. Selling them in the market would expectedly bring him about 15 to 16 million
VND - enough to repay the principal to the bank. However, the bank only lent him eight
million VND, which was just enough for him to buy a cow of mediocre quality. Until the
end of 2005, this cow was just pregnant while due date was approaching. He said it is
likely that he will have to borrow from moneylender and sell all a herd of 50 chickens to
repay the loan on schedule to VBARD and then sell the calf to repay the moneylender. It
means that credit constraint status pushed the farmer into a more difficult situation and
hindered the farmer from expanding his production scale. In the second story a woman,
by end of the year 2004, had a pregnant cow. At that time she wanted to buy another
with the expectation that she would have two mother cows and two calves at the end of
2005. The woman applied to VBARD for a loan of 10 million VND. She thought that
selling two calves would bring about 12 million VND which would be enough to repay
the loan. But in fact, after visiting her house, VBARD only lent her five million VND.
She had to borrow from her friend two more million, intending to buy a little cow at
seven million VND. At the same time, however, her husband suddenly got seriously ill
and she had to spend four million on medical care. The remaining three million VND was
used to buy a little buffalo. These two cases are really typical for the actual situation of
farmers who lack access to sufficient loans. Low capacity of accumulation plus rationing by
the bank make their budget constrained. With a limited budget they have to change their
production plans to adapt to risks of production (for example the cow’s illness or untimely
pregnancy) and risks of their illness situation. In other words, their production decisions
were affected by lack of sufficient loans and consumption decisions.
In summary, the analysis of the evidence above shows that the total amount of
demanded credit, savings amount, total value of livestock, and number of sick person have
a definite effect on the credit status of households. The households classified as credit-
constrained really had great credit demand for expanding scale of livestock production.
However, in the event of credit constraints they could not obtain sufficient loans to
intensify their production.

6.3.2 Credit constrained households and livestock production

Table 14 presents indicators of credit-constrained households via the value of livestock
production, on the basis of which credit constrained households were classified into 5
groups. Group 1 includes households whose value of livestock production is less than six
million VND; group 2: from 6 to 12 million VND; group 3: from 12 to 32 million VND;
group 4: from 32 to 60 million VND and lastly, group 5: more than 60 million VND.
Looking at Table 14 from the left to the right under the increase trend of livestock
production value, we found that first, education level (number of years of schooling) of
households’ heads exists in a positive relation with their value of livestock production to
the extent that households whose heads reached the highest education have the highest
value of livestock production. On the contrary, household heads of group 1 who reached
the lowest education have the lowest value of livestock; second, households’ family size
and number of labours increased from left to right with the biggest number of labourers
and family size in group 5 and the smallest ones in group 1 while the dependency ratio and
the number of sick person decreased clearly and obtained the smallest value at group 5. It
means that the higher the number of labour and level of education, the greater the chance
of expanding production scale while the number of dependents and sick person hindered
households from increasing the scale of livestock raising. These could be explained by the

52
 fact that households with a higher level of education and abundant labour sources could
have advantage in making the right decision and have the potential of labour to implement
their decisions. While the dependency ratio and the number of sick person reflect
disadvantages for expanding production because it implies that the more the number of
dependent and sick person the more the expenditure for school fee, medical and food
which make their budget more constrained. Possibly in that context, they have to use their
money for consumption purposes instead of investment. However, the differences of these
indicators among groups are not statistically significant. Hence, it could be that the
indicators of education level, number of labour, number of sick person, number of
dependents and family size reflecting human resources of households correlated with value
of livestock production.

Table 14: Some indictors of constrained households by value of livestock production

a, b
(Duncan sig. 0.05)

Indicators 0- 6 million 6-12 million 12 - 32 million 32- 60 million >60 million

n =12 n = 10 n = 16 n = 17 n=9
mean s.d. mean s.d. mean s.d. mean s.d. Mean s.d.

Age of household head 40.92 11.29 47.40 14.85 45.38 9.96 47.18 8.27 46.78 8.69
Education of household head 7.75 1.96 7.90 2.38 8.44 1.36 8.65 1.37 9.22 1.48
Family size 4.17 1.53 4.30 1.49 4.88 1.50 4.82 1.29 5.22 0.67
Number of labor 2.33 0.89 2.50 0.71 2.81 1.42 3.24 1.25 3.33 0.87
Dependency ratio 0.38 0.27 0.35 0.27 0.42 0.21 0.32 0.21 0.33 0.19
Number of sick person 0.25 0.62 0.20 0.42 0.33 0.62 0.06 0.25 0.00 0.00
Total farming area 6.49a 2.61 7.71a 2.63 12.09 6.19 17.20b 9.79 17.92b 7.88
a a b b b 12.6
Total credit amount 4.06 3.54 6.10 5.69 11.06 9.01 11.76 7.11 15.71
1
11.7 12.4 13.8
13.48a 16.90a 20.26c b b
Total demanded credit amount 5.16 8.52 26.65 30.00
4 8 9
Ratio of credit amount and credit
0.28a 0.19 0.32a 0.18 0.49b 0.19 0.44 0.15 0.50b 0.20
demand
Total value of cultivation 2.95a 1.10 4.25a 1.77 7.73b 4.60 8.90b 5.20 8.64b 4.22
Total value of cultivation sold in
0.64 0.61 0.64 0.62 1.28 1.00 2.15 2.98 2.26 2.91
the market
Source: Household survey in Tan Linh Commune, Nguyen Thu Phuong, 2006

Next, we observe the relation between total farming area and livestock output. Table
14 shows that total farming areas of constrained households have a tendency to increase
from left to right which is in direct ratio to the value of cultivation. It means that the
bigger the farming area, the higher the value of cultivation. On the other hand, value of
livestock production has the same increase tendency as the value of cultivation. In
addition, a small value of cultivation was sold in the market in Table 14 suggests that
almost products coming from cultivation activities were reserved as food resources for
people and animal. It means that cultivation could be an important food source for
livestock. In other words, larger farming areas bring about bigger value of cultivation
which in turn, increases food sources for livestock. As a result, a larger farming area could

53
support an increase of livestock production. Duncan test shows the statistically significant
differences at 5 percent of total farming area and value of cultivation among groups.
It is really important to note that the total demanded credit amount increased
substantially from group 1 to group 5 in which households having biggest value of
livestock production demanded the largest amount of credit at 30 million VND. The same
thing is true for total credit which was actually borrowed by households; however, these
figures are much less than what households desired. The employment of Duncan test
shows that the differences of the demanded credit amount and the total actual credit
amount among groups are statistically significant at 5 percent. In addition, the use of ratios
of credit amount to credit demand gives support to understand the level of credit
constraint of the households. The ratio of total credit amount against demand tends to
increase gradually from left to right. It reached minimum value in households having
smallest value of livestock production and maximum value in those having biggest value of
livestock. The difference of this ratio among groups is statistically significant at 5 percent.
In context of credit constrained households, this means the more the credit demand of
households were satisfied, the greater the chance they stand of increasing their value of
livestock production. Hence, it is evident that level of borrowing (credit) has positive
effect on livestock production.
In summary, for credit constrained households, education level, the number of
labourers, the number of sick persons, the total farming area, the actual credit amount and
the ratio of credit amount to credit demand could be the factors influencing agricultural
production, particularly for raising livestock in the community. The above analysis shows
the existence of large credit demand of local farmers and the important role of satisfying
this demand to expand their production.

54
7 CONCLUDING REMARKS

This thesis attempts to draw a clear picture of credit performance in Tan Linh commune,
focusing on participation of local people in the credit market and behaviour of formal
lender- VBARD in response to the credit demand and relation between credit status and
agricultural production of households. The study was built on the work of reviewing
relevant literature and analyzing data coming from a household survey, group discussions
and individual interviews. As a whole, the study result shows and analyzes similarities and
contrasts between household groups which were classified by wealth level and credit
status.
Regarding credit demand the study indicates that there exists a large demand for credit
by local farmers in which better off households have demand for largest amount of credit
but fewest number of loans, medium household are active in both sectors with biggest
demand for number of loans but have much smaller loan size than that of the better off.
Poor households have the smallest demand for credit regarding to loan size but they
dominate in number of loans for consumption purposes. The differences in credit demand
are explained by completely different economic characteristics of each household group
which make up their demand and influence their decision in borrowing. Further, the
study attempted to find out the real credit demand which take into account the amount of
credit demand that was not raised by local farmers although actually they had. This was
found in all household groups but was prevalent in poor households. The results indicated
that the poor groups only applied 50% of their total credit demand. The rest they did not
apply to lenders because they feared being refused.
It is very necessary to see linkages between credit demand and borrowing purposes.
More importantly, that is the exploration of which credit demand for each particular
purpose was satisfied. The scheme shows that depending on the purpose of borrowing,
local people pursued different sources of credit to satisfy their credit demand. This is
because they were rational in deciding which sources to ask for a particular kind of loan.
Borrowing purposes of local farmers are quite diverse while the formal sector is specialized
in lending for production purposes, mainly livestock raising. Hence borrowing from the
informal sector becomes logical options for local farmers to cover expenditure purposes
other than production. Moreover, the results indicate that local farmers also borrowed for
production/business purposes from the informal sector. In addition, 46 percent of total
borrowers took the loans from both sectors formal and informal at the same time. Why
this combination exists in their loan portfolios is presented in the results above and
explained by firstly, because of credit rationing by the formal lender, local farmers are
forced to borrow more from the informal sector to satisfy their demand, and secondly,
local people have no other choice except borrowing from informal sector to cover their
consumption expenditures at critical times. The result showed that up to 72 percent of
total informal loans came from relatives and friends with zero interest rate and without
complicated procedures as those of formal lenders. In summary, when local people have
demand for credit, their access to formal or informal sectors depends on each particular
loan and borrowing from informal was to fill the gap left by the formal lenders.
When local farmers have a demand for credit, it is evident that the number the number
of dependents, the number of sick person, the total value of livestock (a proxy for
household’s capacity of livestock production), the reputation and social status of household
could be the factors that influence decision making, as well as actual borrowing level of
households from the formal sector. From the results, it was found that the household

55
group that borrowed the biggest amounts from formal lenders is the group that had the
biggest number of labour, the highest years of schooling, the biggest value of livestock,
and not any sick person in their family. Further, their indicators of reputation and social
status are highest which could make them more confident to ask formal lenders. Such
advantages could motivate them to make the decision of borrowing to expand their
production in larger scale. By contrast, the household group that borrowed with the
smallest amount of credit from formal lenders is the group that had the biggest number of
dependents, the biggest number of sick person, and the lowest value of livestock
production. In addition, their indicators of reputation and social status are lowest which
are in the sense that they are poor and do not hold position in community. Such
disadvantages could make them lack confidence to claim their credit demand to formal
lenders as well as being un-credit worthy in the eyes of the lenders. So their borrowing
level is smallest. It also was found that the decision of expanding livestock production,
number of sick person and dependency ratio are likely to be the factors influencing to
households’ decision to borrow from informal sources. This implies that the economic and
social situation have a substantial effect on farmers’ self-confidence to make the decision of
applying to formal lenders. It again confirms the reasons for why in many cases they have
demand for credit but did not apply to formal sector.
In response to the strong demand for credit of local households, one formal lender –
VBARD rations the size of loans. The result indicates that credit-rationed households are
those that have the largest demand for credit. The majority of them are the poor and
medium households who want to develop intensive livestock production but they lack
accumulation of savings. The results showed that the amount of credit applied by
households to VBARD, the dependency ratio, the number of sick person, the owned
farming area, and the ownership of LUCs could be the factors that influence credit
rationing by VBARD. It is important to note that these factors also hindered 25 of 100
selected households, mainly the poor from applying for credit to VBARD although
actually they had demand for credit. It means that the poor were forced to borrow from
informal sources in the cases of emergencies or to cancel their production plan. There is
preliminary evidence that the short duration that VBARD applied is incompatible with life
cycles of livestock. It is also the reason why the locals, especially the poor, are afraid of
applying for credit to VBARD.
As a result, the households who were rationed credit by the formal lender and the
households who did not apply for credit to formal lender constituted the majority of credit
constrained households group which accounted for 64 percent of the sample. There were
sharp differences between these households and the unconstrained households regarding
the amount of credit they applied for, the capacity of livestock production, the total value
of agricultural production, the non-farm income, the savings and the number of sick
person which influenced the credit status of households.
It is important to know that for credit constrained households, education background,
family size, number of dependents, total farming area and ratio of credit amount to credit
demand are factors influencing agricultural production especially raising livestock. The
higher the ratio of credit amount to credit demand, the bigger the value of livestock
production. This means the more the credit demands of households were satisfied, the
greater the chance they stand of increasing their value of livestock production. It suggests
that rural households cannot optimize their livestock production in the context of being
credit constrained, and this results in inefficient production.
It is obvious that VBARD has been leading the local credit market and has apparently
made considerable effort in providing credit to rural household. This organization,

56
however, has also revealed its limitations regarding staff resource to cover an extensive
area, shortage of medium- and long-term loans, mobilization of small savings and scheme
of liquidating a collateral. Other formal organizations including VBSP, funds, and the
cow-raising projects do not appear to have a lot of weight in Tan Linh commune and are
limited in reaching out to the poor.
The implication here can be that rural financial systems need to be developed to
improve the access to credit of households, which brings about development of household
economy in particular and rural economy in general. To develop the financial system in
rural areas, it is necessary to create a more competitive rural financial market, which forces
VBARD to strengthen its institutional capacity. It is also pertinent to raise the question of
whether banks can overcome effects of asymmetric information to meet the demand of
rural households. Robinson (2001) showed that “ At BRI’s unit desa system (Bank Rakyat
Indonesia) and at BancoSol, the banks can differentiate among high-and low- risk
applicants with a high degree of accuracy, and interest rates are maintained at a level that is
attractive to low-risk borrowers. Incentives in the form of products, prices, services and
options to reborrow at larger loan sizes motivate low-risk borrowers to select these banks.
New borrowers are given small loans; loan size increase as capacity and willingness to
repay promptly are demonstrated. Credit rationing – either because of risk or because of
lack of funds – is unnecessary.” Whether this can be a potential solution for rural credit
market in Vietnam to cope with imperfect information need to be examined further.

57
8 REFERENCES

Adhikari, B. (2005), Property rights and natural resource: socio-economic heterogeneity and
distributional implications of common property resource management. Environment
Department, The University of York.
Bali Swain, R. (2001). Demand, segmentation and rationing in the rural credit markets of Puri.
Department of Economics Uppsala University. Accessed in November 2005 from the website:
http://72.14.235.104/search?q=cache:HLyYEhXsNQoJ:www.diva
portal.org/diva/getDocument%3Furn_nbn_se_uu_diva-749-1__fulltext.pdf
Bali Swain, R.. (2002).Credit rationing in rural India - Choice of lender and credit rationing.
Department of Economics Uppsala University. Accessed in July 2006 from the website:
http://72.14.235.104/search?q=cache:sALdWZUTKWEJ:www.nek.uu.se/StaffPages/
Publ/P275.pdf
Bardhan, P. & Dayton-Johnson, J. (2000), ‘Heterogeneity and commons management’,
Department of Economics, University of California, Berkeley, USA.
Bardhan, P., & Udry, C. (1999) Development microeconomics. Oxford: Oxford University Press.
Barham, B. L., Boucher, S., & Carter, M.R. (1996) Credit constraints, credit unions and small-scale
producer in Guatemala. World Development, Vol. 24, No. 5, pp.793-806
Barret, C. B., Bezuneh, M., Clay, D. C., & Reardon, T. (2000) Heterogeneous constraints,
incentives, an income diversification strategies in rural Africa Dept. of Agricultural, Resource
and Managerial Economics, Cornell University.
Bryant, J. (1998) Communism, poverty, and demographic change in North Vietnam. Population
and Development Review 24(1998): 235-69.
Carter, M. R. (1989), the impact of credit on peasant productivity and differentiation in Nicaragua.
Dept. of Agricultural Economics, University of Wisconsin-Madison, USA.
Cuc, N. S. (2003) Vietnam Agriculture and rural area in the renovation period (1986-2002).
Statistics Publisher.
Dat, T. T. (2001) Analysis of informal credit sector in Vietnam General survey in rural area,
agricultural and aquaculture sector 2001, Statistics publisher
Diagne, A. (1999) Determinants of household access to and participation in formal and informal
credit markets in Malawi. Food Consumption and Nutrition Division, International Food
Policy Research Institute
Diagne, A., Zeller, M., & Sharma, M (2000) Empirical measurements of households’ access to
credit and credit constraints in developing countries: methodological issues and evidence. Food
Consumption and Nutrition Division, International Food Policy Research Institute.
Duong, P. B. & Izumida, Y. (2002) Rural development finance in Vietnam: a microeconometric
analysis of household surveys. World Development Vol. 30, No. 2, pp. 319-335.
Ellis, F. (2000). Rural livelihoods and diversity in developing countries. Oxford University Press
Feder, G., Lau, L. J. Y., &Luo, X. (1989). Agricultural credit and farm performance in China.
Journal of comparative economics. Journal of Comparative Economics 13, 508-526
Feder, G., Lau, L. J. Y., &Luo, X. (1990). The relationship between credit and productivity in
Chinese agricultural: a microeconomic model of dis-equilibrium. American Journal of
Agricultural Economics, 72, 1151-1157
Fisman, R..(2001) Trade credit and productive efficiency in Developing countries. World
Development Vol.29, No. 2
Foltz, J. (1998) Credit market constraints and profitability in Tunisian agriculture. Dept. of Ag.&
Applied Economics University of Wisconsin-Madison.
Godquin, M., & Sharma, M. (2004). If only I could borrow more! Production and consumption
credit constraints in the Philipines.Accessed in December 2005 from the website
http://www.adres.polytechnique.fr/JD/pdf/Godquin_Sharma.pdf.
Hackett, S.C. (1992). “Heterogeneity and the Provision of Governance for Common-Pool
Resources.” Journal of Theoretical Politics 4(3): 325-342.
Hung, D.V. (2001) Informal Credit Market and Micro-finance in Vietnam. Priorities and Strategies
in Rural Poverty Reduction: Experiences from Latin America and Asia Presented at the Japan
Program/INDES 2001 Conference – Japan.
Hung, D.V. (2005) development of microfinance market in Vietnam, Labour and Society
Publisher.

58
Izumida, Y., & Duong, P. B. (2001) Measuring the progress of rural finance in Viet Nam. Saving
and development. Savings and Development, Quarterly Review – No. 2
Kochar, A. (1997) An empirical investigation of rationing constraints in rural credit markets in
India. Journal of Development Economics, Vol. 53, 339-371
MARD, (2004) A summary report of the structure shift of agriculture and rural economy in
Vietnam after twenty years of Doi moi adoption
Nuryartono, N., Zeller, M. & Schwarze, S. (2005). Credit Rationing of Farm Households and
Agricultural production: Empirical Evidence in the Rural Areas of Central Sulawesi, Indonesia
Putzeys, R. (2002) Microfinance in Vietnam: three case studies. Belgian Technical Cooperation,
Hanoi, Vietnam
Quach, M. H., Mullineux, A.W., & Murinde, V. (2005) Access to credit and household poverty
reduction in rural Vietnam: a cross-sectional study. The Birmingham Business School, The
University of Birmingham Edgbaston
Robinson, M.(2001) The microfinance revolution. Sustainable finance for the poor. The World
Bank.
Sadoulet, E., & de Janvry, A. (1995) Quantitative development policy analysis. Baltimore and
London : The Johns Hopkins University Press.
Socialist Republic of Vietnam, 2005, The Comprehensive Poverty Reduction and Growth Strategy
(CPRGS)
Stiglitz, J.E., & Weiss, A. (1981). Credit rationing in markets with imperfect information. The
American Economic Review, Volume 71 (3), 393-410
Zander, R. Barriers to Credit Access in Rural Sri Lanka. Accessed in November 2005 from the
website:http://72.14.235.104/search?q=cache:iHoQsGcS0vsJ:library.wur.nl/way/catalo
gue/documents/FLR12.pdf
Zeller, M. (1994) Determinants of credit rationing: a study of informal lenders and formal credit
groups in Madagasca. World Development, 22(12), 1895-1907

59