CIRDAP

The Centre on Integrated Rural Development for Asia and the Pacific (CIRDAP) is a
regional,intergovernmentalandautonomousinstitution.Itwasestablishedin1979atthe
initiative of the countries of the AsiaPacific region and the Food and Agriculture
Organisation(FAO)oftheUnitedNationswithsupportfromotherUNorganisationsand
donorcountries/agencies.ThemembercountriesofCIRDAPareAfghanistan,Bangladesh
(host state), India, Indonesia, Iran, Lao PDR, Malaysia, Myanmar, Nepal, Pakistan,
Philippines,SriLanka,ThailandandVietnam.

The main objectives of the Centre are to: (i) assist national action, (ii) promote regional
cooperation,and(iii)actasaservicinginstitutionforitsmembercountriesforpromotion
of integrated rural development through research, action research, training and
information dissemination. Amelioration of rural poverty in the AsiaPacific region has
beentheprimeconcernofCIRDAP.

The programme priorities of CIRDAP are set under four Areas of Concern: 1) Agrarian
development; 2) Institutional/infrastructural development; 3) Resource development
including human resources; and 4) Employment. Within these areas of concern, the
thematic areas are: Poverty alleviation through participatory approaches with emphasis
on social sector development (e.g. health, education and nutrition); Employment
generation through microcredit support, infrastructure development and local resource
mobilisation; GONGO collaboration; Gender issues; Governance issues; and
Environmentalconcernsforsustainableruraldevelopment.

OperatingthroughdesignatedContactMinistriesandLinkInstitutionsinmember
countries,CIRDAPpromotestechnicalcooperationamongnationsoftheregion.Itplaysa
supplementaryandreinforcingroleinsupportingandfurtheringtheeffectivenessof
integratedruraldevelopmentprogrammesintheAsiaPacificregion.

EDITORIALBOARD

Chairperson

DurgaP.Paudyal

Editor

K.A.S.Dayananda

Members

T.Abdullah,DevelopmentConsultant
Q.K.Ahmad,BangladeshUnnayanParishad
MomtazUddinAhmed,DhakaUniversity
MohammedFarashuddin,EastWestUniversity
S.K.Singh,CIRDAP
HosseinShahbaz,CIRDAP

EditorialAssistant

S.M.Saifuddin

ISSN10185291

AsiaPacificJournalof
RuralDevelopment

VOLUMEXIXJULY2009 NUMBER1

CENTREONINTEGRATEDRURALDEVELOPMENTFOR
ASIAANDTHEPACIFIC

ASIAPACIFICJOURNALOFRURALDEVELOPMENT
VolumeXIX,July2009,Number1
ISSN10185291
CIRDAP2009

Theviewsexpressedinthispublicationarethoseofthe
author(s)anddonotnecessarilyreflectthoseofCIRDAP

Publishedtwiceayear

Priceincludingpostage
AnnualSubscription(TwoIssues):
Tk.300forBangladesh
US$20forAsiaPacificCountries
US$35forOtherCountries

Advertisementsareinvitedfromprofessionalbodies/institutions

Thefollowingtariffchargeswillapply

a) Innerbackcover:US$200
b) Fullpagetext:US$100
c) Halfpagetext:US$50

PublishedbyDurgaP.Paudyal,DirectorGeneral,onbehalfof
CentreonIntegratedRuralDevelopmentforAsiaandthePacific(CIRDAP)
ChameliHouse,17TopkhanaRoad,GPOBox2883
Dhaka1000,Bangladesh

Printedby:NizamPrinters&Packages,Dhaka

ASIAPACIFICJOURNALOFRURALDEVELOPMENT
VolumeXIXJuly2009Number1
CONTENTS
Articles

ICTandEmpowermentofRuralandDeprivedWomeninAsia
D.A.Patil,A.M.Dhere,C.B.Pawar

Page

SpatialAssessmentofPotentialforGroundwater
23
DevelopmentinLowerChaoPhrayaRiverBasin
SudsaisinKaewrueng,MukandS.Babel,AshimDasGuptaandShahriarM.Wahid
MigrationofRuralPeopletoUrbanAreas:
AStudyofThreeUpazilasinMymensingh
GulsanAraParvin,MamunMuntasirRahman,FarhanaYasmin

43

AccessibilityofWomentoProductiveResourcesin
FarmHouseholdsofKadunaState,Nigeria
S.A.RahmanandF.A.Ajayi

59

EmpiricalAnalysisonRuralHouseholdsBorrowingBehaviour:
TheCaseofCentralJava
TakashinoNina

67

ValueChainsandStandardsinShrimpExportfromBangladeshand
ThailandtoJapan:AComparativeStudyonSafetyCompliances
MohammadTajUddin

89

SimulationofGrowthandYieldofMaizeunderWaterStressImposed 109
duringCriticalGrowthPeriodsinNakhonRatchasimaProvince,Thailand
KiattiyosThongsagaandS.L.Ranamukhaarachchi
DecentralisationonAccesstoRuralWaterSupplyServices:135
AStudyofSixVillagesinWestBengal,India
IndranilDe
CarpSeedTransportinBangladesh
MahmudHasan

155

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

ICTandEmpowermentofRuraland
DeprivedWomeninAsia
D.A.Patil*,A.M.Dhere**,C.B.Pawar***

Abstract
The Role of Information and Communications Technologies (ICTs) as a tool for
development has attracted the sustained attention of the government and NGOs. The
millennium declaration adopted in 2000 underscored the urgency of ensuring that the
benefits of new technologies, especially ICTs, are made available to all. ICTs can be a
powerful catalyst for political and social empowerment of women and the promotion of
gender equality (United Nations 2005:3). This paper examines some key ICTbased
initiativesinAsiaingeneralandIndiainparticularwhicharespecificallydesignedfor
the empowerment of rural and deprived women. It shows how ICTbased initiatives
empower rural and deprived women; it tries to clarify a theoretical dilemma between
meaning and measurement of empowerment and provides a holistic definition of
empowerment. Finally it focuses on the twofold need to address some of the social and
technicalchallengesaswellaspolicysuggestionstouseICTsproactivelyandeffectively
topromotegenderequalityandfortheempowermentofruralanddeprivedwomen.The
study used observations, focus group discussions and review of documents to collect
secondaryandprimarydata.

1.0 Introduction
Women constitute half of the worlds population, perform nearly twothirds of
itsworkhours,receiveonetenthoftheworldsincomeandownlessthanone
hundredthoftheworldsproperty.Asper2001Census,23percentofwomenare
intheworkforce,94percentofallworkingwomenareintheinformalsector.
Womencarryadisproportionatelygreaterburdenofworkthanmenandsince
womenareresponsibleforagreaterburdenofworkthanmenandsincewomen
are responsible for a greater share of nonSNA (System of National Accounts)
workincaseofeconomy,theyenterlabourmarketalreadyoverburdenedwith
work(UnitedNations2005).
CorrespondingAuthor.AssistantProfessor,CentreforEquity,SocialJusticeandHuman
Development(CESJHD)Yashada,Pune07,Maharashtra,India,Email:
dhanrajpatil17@rediffmail.com
** Lecturer,InidraCollegeofCommerceandScience,Pune33,Maharashatra,India.Email:
prof.amardhere@gmail.com
*

***

Lecturer,SangolaCollegeSangola,Solapur,Maharashatra,India.Email:
shekhar102030pawar@rediffmail.com

In case of India, as according to 2001 Census, out of the total population of

1,027,015,247thefemalepopulationwas495,732,169accountingfor48.2persons
of the total population. But, their participation in economically productive
activities are often underestimated. For last two decades, womens work
participation rate increased from 22.73 per cent in 1991 to 28.6 per cent during
2001(Census2001).
Inthepatriarchalsociety,womenhavebeensubjectedtoexploitation,atrocities,
maledominance and genderbias. The Food and Agricultural Organisation
(FAO) has also emphasised strengthening and motivating them to claim their
rights, particularly the women at the grassroots. Gender gaps in access to
resources, services and information are a major obstacle to womens
empowerment. In this context ICTs are emerging as a key facilitator for
empoweringwomen,especiallyruralanddeprivedwomen.Thepresentstudyis
anattemptinthiscontexttostudytheroleofICTsfortheempowermentofrural
anddeprivedwomen.

2.0 BackgroundofStudy
In the recent past, ICTs have been added to the women and gender equality
debate. ICTs are being presented as a tool having potential to benefit womens
empowerment and a number of ICT projects that specifically target women
have been established in several developed and underdeveloped countries.
BeforegoingtostudytheroleofICTsinwomenempowerment,isnecessaryto
understandwhatisICT.

2.1WhatisICT?
Information and Communication Technologies (ICTs) are a diverse set of
technological tools and resources to create, disseminate, store, bring value
additionandmanageinformation.TheICTsectorconsistsofsegmentsasdiverse
as telecommunications, television and radio broadcasting, computer hardware,
software and services and electronic media, for example, the internet and
electronicmail(UnitedNations2005:1).

2.2RoleofICTsingenderempowerment
Empowermentofwomeninthecontextofknowledgesocietiesentailsbuilding
uptheabilitiesandskillsofwomentogaininsightintotheissuesaffectingthem
andalsobuildinguptheircapacitytovoicetheirconcerns.InthiscontextICTs
areemerging asa powerful toolforgender empowerment in many developing

countries. There has been a rapid growth in the ICT sector since the late 1980s
andtheuseofICThasdramaticallyexpandedsincethe1990s.Accordingtothe
WorldBank,teledensityinIndiahadreached3.8percentofthepopulationby
2001(Jain2006).
2.2.1ExamplesfromAsia
InAsiasomeoftheICTinitiativeshavebeenplayingvitalroleforempowering
ruralanddeprivedwomen.Forexample,theLumbiniCMCprojectinNepalwas
established in April 2004. The community multimedia centres CMC combining
newtechnologieslikecomputers,internetandspecialisedsoftwareapplications
withtraditionalmedia;itworksmainlyforeducation,healthandadultliteracy.
LikewisetheInformationandCommunicationTechnologyAgencyofSriLanka
(ICTA)startedaprojecttousetechnologytodocumentviolenceagainstwomen
andeducatingwomenontheirlegalrightswiththehelpofwebportal.However
theeHomemakersprojectinMalaysiawaschosenasarunnerupingenderand
ICT awards; it is a marketing website; the project also developed a Distributed
Work Management Applications (DWMA) using mobile telephone for the
empowerment of deprived women. Besides these examples some of the other
keygenderICTinitiativeshavealsobeenpresentedinthenextsection.
2.2.2 ExamplesfromIndia
India has initiated concrete policies to promote gender parity in information
technologyandanumberofprojectshavebeenlaunched.Forexample,aproject
setupbytheIndianInstituteofTechnology(IIT)Chennaiwhereatotal80per
centofthenewkioskssetuparerunbywomen,manyofwhomhavehadvery
littleornoacquaintancewithtechnologyearlier(seetheChart).Womensgroups
NGOs and other sections of the civil society are also playing a vital role in
empoweringwomenintheregion(VijayalaxmiandBhavani2006).
Indiashop,anecommercewebsite(2005)inTamilnadu,hasbeendesignedtosell
products made by rural womens cooperatives and NGOs. The Dhan
Foundation (2004) and Swayam Krishi Sangam (2004) are using ICTs such as
handhelddevicesandsmartcards,toimprovemicrofinanceprojectstoempower
poorwomen.TheSelfemployedWomensAssociation(SEWA2004)hasseveral
ICT projects for women including communitylearning centres, a school of
Science and Technology for selfemployed women and the Theliphone project,
whichprovidesmobilephonestowomenintheinformalsector.i

Chart:SomeSelectedGenderBasedRuralICTInitiativesinIndia

Sl.
No.
1.

2.

3.

4.

5.

6.

7.

Target
Group

Implementing Sponsoring
Agency
Agency

Project
(Initiator)

Area

Community
RadioDeccan
Development
Society

Deccan
Develop
ment
Society

Medak
District
Andhra
Pradesh

Dalit
Women

DDS

Kudumbha
shree

Govern
mentof
Kerala

Kerala(991
Panchayats
and58
municipa
lities)

Familyasa
wholeand
womenin
particular

Govern
mentof
Kerala

Mahila
Samakhya,
Karanataka
andITfor
change,
Benglore

Three
talukas
(Humsur,
Nanjansud
and
Periyapatn)
inMysore
district

SelfHelp
Group
women

Mahila
ITfor
Smakhya
change
Karnataka Karnataka
(MSK)

Change
initiatives
and
UNESCO

Buduria
Municipa
lityNorth
24Paragans
district
West
Bengal

Rural
Women

ProjectEnter
PriceToeHold

ASCENT,
CLRI

Belgaum
District,
Karnataka

Rural
artisans
specially
women

CLRI

PuttingICTsin
thehandsof
thewomen

Datamation
Foundation
(Nonprofit
Trust)

RuralAreas
ofKanpur,
Lacknow
etc.

Rural
Margina
lised
women

Datamation TheWorld
Foundation Bank
Charitable Microsoft
Trust

SEWA

Self
Employed
Womens
Association

Ahmeda
bad,
Gujarat,
Western
Partof
India

Informal
Sector,
Women

SEWA
and
Partner
NGOs

Ministryof
Agriculture
(GoI)

TamilNadu

Name

Mahiti
Manthana

Nabanna

Yearof
Starting

UNESCO
1998

SIRDI
2002

2005

UNESCO
Chagne
initiatives,
Kolkatta
2003

NLDP
UNDP

1999
20002

2001

SEWA

1977

8.

TamilNadu
Womenin
Agriculture
(TANWA)

Rural
Women

Directorate DANIDA
of
Govern
Agriculture mentof
India

1994
2003

Sl.
No.

Name

Target
Group

Area

Implementing Sponsoring
Agency
Agency

Yearof
Starting

Networkof
Entrepre
neurship
and
Economic
Develop
ment
(NEED)

Sitapurand
Barabanki
blocksof
Uttar
Pradesh

Rural
population
(Women
and
Children)
forhealth

PICs

10. PravaraVillage
ITProject
(PRAGATI)

KVKC
(Krishik
Vikas
Kendra)

Ahemed
nagar
District,
Maharashtra

Rural
Population
mainly
women(for
health)

Conver
gent
communi
cations

Convergent
communi
cations,
Pravara
Group

1999

11. ICTEnabledLife
Skilland
Sexuality
Educationfor
AdolescentGirls

Centrefor
Womens
Develop
mentand
Research

Kancheepur
amDistrict,
Chennai

Dalit
Adolescent
girls(for
educatio)

Centrefor
Womens
Develop
mentand
Research

Centrefor
Womens
Develop
mentand
Research

2004

12. Computeron
Wheels:
Education
Through
MobileVans

Vidya
Ruralareas
Pratishthans inallIndia
Instituteof
Information
Technology
(VIIT)

Rural
Population

Vidya
State
Pratish
Govern
thans
ment
Instituteof
Information
Technology

13. Economic
Empowerment
ofMinority
Muslimsin
India

Datamation
Foundation

Suppressed
minority
womenof
thecountry

Damation
foun
dation

UNESCO

14. IndiraSoochna
Shakti(ISS)

UNDPand
Planning
Commi
ssionand

Students
specifically
girlsand
women

CHIPS

Govern
mentof
Chattis
garhand
UNDP

9.

PACS(Poorest
AreasCivil
Society)
Programme
Empoweringthe
ruralpoor

Project
(Initiator)

Govt.of
India

Seelampur
Zaffarabad

246Village
Counsils
(Panchayat),
headquarter
villagesof
Chattisgarh

NEED

2002

2000

2001

2001

Source:http://ictrd

SelfhelpgroupsofruralwomeninAndhraPradeshhavebeensosuccessfulin
marketing their products at home and abroad that the major MNCs (multi
nationalcorporations)wanttousetheirsellingskills.
Thus,withthesefactsinmindpresentpaperlooksattheavenuescreatedbyICT
enabledservicesandopportunityforwomensempowerment.

3.0 ReviewofLiterature
There are very few studies which have been conducted to analyse the role of
ICTsforwomenempowerment;someoftherelevantstudiesareasfollows:
CEEWA(CouncilfortheEconomicEmpowermentofWomenofAfrica)(2005)in
his study in Uganda found that as an enormous source of information, ICTs
constituteapowerfullearningtoolthatprovidesaccesstomarketinginformation
that can help womens business succeed. ICTs like mobile telephony, can also
offerdirectandinexpensivemeansofcommunicationforwomensorganisations
andenablethemtoshareknowledgeonaquickandcollectivebasis.However,
accesstoICTsisrestrictedbecauseofoppressivegenderrelations,socialcultural
barriers, distance to the ICT facilities, poor infrastructure and costs for access
ICTs.
Jain (2006) observed that, ICT has played an important role in changing the
conceptofworkandworkplace.Newareasofemploymentsuchasnetworking,
i.e.workingfromadistance,arebecomingfeasiblewithnewtechnology.
Dalal (2006) argues that,while there is recognition of the potential of ICTas a
tool for the promotion of gender equality and the empowerment of women, a
genderdividehasalsobeenidentifiedreflectedinthelowernumbersofwomen
accessing and using ICT compared with men. Unless this gender divide is
specifically addressed, there is a risk that ICT may exacerbate existing
inequalitiesbetweenwomenandmenandcreatenewformsofinequality.
The study by the Vijalakshmi and Bhavani (2006) examines the nature and
sourceofgendervariationsininternetuseandhowculturalfactorscontributeto
such variations. The study provides insight into the gender variations in using
theinternetspaceandformingonlinerelationshipsandonlinediscourse.Onthe
whole,thestudysupportsthefactthattheinternetprovidesanopportunityfor
women to construct their identities to challenge certain traditional norms and
reconstructtheirlives.
The findings of Moser (1993) indicate that the assumption where access to ICT
will automatically empower women economically or otherwise is questionable.
Onthecontrary,womenssocialpositioninsocietyhasnotchangedmuch.

4.0ConceptualFramework
While studying the concept of empowerment it is necessary to understand this
conceptinaholisticmanner,whichwillprovideaconcreteideatotheresearcher
toselectspecificindicatorsfortheconceptofempowerment.

4.1Understandingempowerment:Dilemmaofmeaningand
measurement
Thissectionprovidessomekeyapproachesofempowermentandalsoitgivesa
holisticdefinitionofempowerment.

4.2Empowermentandpower
Many past studies, empowerment is looked at in relation to how people
understandpower.Itisalsopresentedasaprocessthroughwhichwomengain
powerovermen.Thisspecificunderstandingofempowermentispossibleoneof
thereasonsmenresittheconcept.Inothercases,empowermentispresentedasa
processthatenableswomengainaccesstodecisionmakingandmakingchoices
abouttheirownlives(HuyerandSikoska2003).

4.3Feministapproachofempowerment
Feminist understanding of empowerment includes the idea of the personal as
politicalthereforecallingforaconsiderationofempowermentasexperiencedby
women within the public and private spheres (Rowlands 1997). Empowerment
couldalsooccuratapoliticalsocialoreconomiclevel.Economicempowermenthas
to do with access to resources, and is key to women because their subordinate
position is entrenchment in their level of access to and control over resources.
Social empowerment is about challenging social and cultural structures. While
politicalempowermentinvolvestherightsandabilitiesofpeopletoparticipateas
equalsindecisionmakingprocesses(Clement1994).

4.4Empowermentbydevelopmentalternatives
Another view to empowerment has been put forward by the development
alternatives with Women for a New Era (DAWN).3 DAWN argues that for
women to be empowered their strategic and practical gender needs have to be
addressed(Moser1993).
DAWNusesthesestrategiestoaddressissuesofinequalitybetweenwomenand
menbecauseitrecognisesthatwomensempowermentisnotagiventhingand
that women have to work collectively to overcome the structural inequalities
(Longwe1994).

4.5EmpowermentbyLongwesframework
ThereisalsotheLongweWomensEmpowermentFramework.Thisframework
examines the extent to which projects lead to womens empowerment.

According to Longwe, only by addressing the five levels at which gender gaps
exist `empowerment can happen. The five levels include: welfare, access,
conscientisation (awareness), mobilisation and control. Welfare addresses the
lowest level at which a development intervention may hope to close a gender
gap;whilecontrolislevelatwhichthereisabalanceofpowerbetweenwomen
and men and neither has dominance. Access considers equality of access to
resourcesandconscientisationisalevelofawarenessraising.Forwomentotake
appropriateactiontoclosegendergapsorinequalitiestheremustberecognition
that their problems stem from inherent structural and institutional
discrimination.Participation(mobilisation)isconcernedwiththeextenttowhich
women have been able to take part in decisionmaking process alongside men.
But what does empowerment mean? Although defining empowerment is
subject to debate, a common understanding is necessary for the sake of this
discussion. This paper defines empowerment as a multidimensional, gradual
process by which women become aware of their subordinate position and the
powerstructuresatplayintheirenvironment;thengoaheadtodosomethingto
overcome these constraints. This process enables women to become selfreliant
andgaincontrolovertheirownlives;andactonissuestheyconsiderimportant
but without infringing upon the rights of others; this process may also lead to
collectivechange(Rowlands1997andFreire1972).Thus,itentailsdeveloping
the capacities of women to overcome social and institutional barriers and
strengthening their participation in the economic, political and social processes
soastoproduceanoverallimprovementintheirqualityoflife.

5.0 ResearchQuestions
For centuries, women in this country have been socially and economically
handicapped. They have been deprived of equal participation in the socio
economic activities of the nation especially the condition is worst among the
ruralanddeprivedsectionwomen.
Women need ICT for the same reasons as men to access information of
importancetotheirproductive,reproductiveandcommunityrolesandtoobtain
additional resources. Based on the above review of literature and conceptual
framework,thisstudyistryingtoanswerthefollowingresearchquestions:

CanICTsplayanimportantroleinwomenempowerment?
Ifyes,thenisitempoweringruralanddeprivedsectionswomen?
Ifyes,whatarechallengesandobstacles?
Howtoovercomewiththeseissues?

In this context this paper attempts to examine various ICTbased experiments

for the empowerment of rural and deprived women in different parts of India
andAsia.

6.0 Methodology
6.1Objectives

TounderstandtheroleofICTsforwomenempowerment.

To study how ICTs are helpful for the empowerment of rural and
deprivedsectionswomen.

To suggest policy implications for the effective implementation of

furtherICTbasedprogrammes.

6.2 Database
ICTsareplayinganimportantrolefortheempowermentofruralanddeprived
women. We examine the contribution of these technologies in the process of
empowerment at the Asian level for the periods198182 to 200405. The paper
uses just released 20042005 report of United Nations on Women and Beyond,
UNDP (2001) Human Development Report and the proceedings of United
NationsEconomicandSocialCommissionforAsiaandthePacific(UNESCHP):
withSpecialReferencetoGenderandDevelopment(2003).
The study is also based partially on the primary data of the Warana Wired
Village Project (WWVP) district Kolhapur, Maharashtra state of India, which is
oneoftheoldestICTbasedprojectinAsia.However,discussionswithruralICT
mangers of WWVP, Kolhapur, members of Women Association of SEWA,
Gujarat and experts of NIRD, Hyderabad immensely helped us to understand
sociotechnicalhurdlesandpossiblesolutionsfortheeffectiveimplementationof
ICTprojects.Theprimarydataforthisstudyhasbeencollectedthroughpersonal
interviews, focus group discussions and observations during the period July to
September2008.

7.0 ResultsandDiscussion
7.1ICTsandempowermentofruralanddeprivedsectionsofwomen:
Someprominentinitiatives
A number of ICT initiatives can be cited all acrossAsia that demonstrates how
women can acquire ICTrelated skills and use the technology and at the same
time have control over their use. In this section, efforts have been made to

highlight key initiatives that clearly depict how women can be empowered
economically, socially and politically. Keeping in view the plight of rural
women, who are more unaware of new technologies than their urban
counterparts, government and nongovernment organisations (NGOs) are
providing special packages for them. Let us take an overview of some of the
prominent ICTbased initiatives pertaining to the empowerment of rural and
deprivedsectionswomen.
7.1.1ICTandwomenseconomicempowerment
In Asia, a number of credible models exist that could be replicated to address
womenseconomicissuesthroughtheuseofICTs.
7.1.1.1SelfEmployedWomensAssociation(SEWA)
IthasbeenorganisingwomenintheinformalsectorinIndiasince1972.4Itwas
oneofthefirstorganisationsgloballytorealisethepotentialofusingICTsforthe
productive growth of the informal sector. SEWA is establishing Technology
Information Centres in 11 districts of Gujarat, India to provide computer
awarenesstrainingandbasiccomputerskillsfortheirbarefootmanagers,built
the capacity of women organisers and leaders and strengthen their members
microenterprises.Itnowrunsprogrammesthatdevelopwomensabilitiesinthe
use of computers, radio, television, video, the telephone, fan machines, mobile
phone and satellite communication. Electronic networking is expected to
strengthen the connections between the various cooperatives working in
different sectors and areas, and currently enables the provision of content
tailored to the needs and environment of particular groups of villages. In
addition,membersofSEWAareabletoaccessgovernmentschemesandtapinto
newmarkets.Inthesecondphase,thecentreswillalsosupporttheeducationof
girls(www.sewa.org).5
7.1.1.2GrameenPhones,Bangladesh
It is the best known of the ICTenabled businesses with a high percentage of
womenowners/operatorsinBangladesh.6

10

Grameen Banks Women Phone

In 1996, the Bank setup Grameen Phones, Bangladeshs first cell phone network.
Grameen Phones is particularly not worthy because of the economic empowerment
that it has brought to poor, largely uneducated women.7 From among them has
more than two million predominantly women borrowers; the Bank management
selects Village Phone Operators to whom the phone is provided as an in-kind loan.
The operators resell wireless phone services (incoming and outgoing) to fellow
villagers.
The women operate their phone businesses while doing household chores or
operating another business. They have created a phone culture among women by
enabling their access to communication tools from which they might otherwise be
excluded. They have also shown that poor, largely uneducated women can master
the skills and run a small business. Women phone operators have achieved
economic and social empowerment within their households and communities.
Source: www.telecommos.com

7.1.1.3Demonstratorapplicationgrantscheme,Malaysia
Among the 74 projects funded, a number were aimed to benefit women, and
includedwomenscapacitybuildingintheuseofICTs.Forexample,theproject
calledTCentreforTelenetworkingandTelecommutingwasdesignedtoguide
200participantsmainlywomenandyouths,tolearnandacquiretelenetworking
skillsandtoenablethemtoadaptteleworkingasanewmodeofwork.Another
projectthatwasfundedundertheDAGSschemeistheEHomemakersProject,
the concept of working from home by providing resources and a platform for
homemakers and home workers to teletrade and teleexchange. A Womens
Electronic Networking Training (WENT) Award winner 2003, this project
provides basic ICT skills training to disadvantaged women to enable them to
participate effectively in this knowledgebased economy. The project prepares
themtoworkathomethroughothersofttrainingsandempowermentexercises
(http://ehomemakersnet).
7.1.1.4Smile(SavitriMarketingInstitutionforLadiesEmpowerment)
ItisavoluntaryorganisationinPune.Thisprojecthasincreasedliteracylevelof
underprivilegedwomenthroughtheusageofICT.Internethasalsohelpedthem
market their various products like soft toys, candles, bags, utility items, etc.

11

ThroughInternet,thereisgreaterawarenessandexposuresandmarketreachfor
theproducts.
7.1.1.5DatamationFoundation
It started a project in 2003 in Seelampur area of Delhi for Muslim women. The
project localised appropriate communication and information networks by
settingupanICTcentreataChennai.Thishelpedlinkresourcepoorwomento
the information and tolls for knowledge management. It also helped establish
buyersellerlinkagestowardseradicationofabsolutepoverty.Ithasestablished
its standing in the community and became a big attraction for the women of
Seelampur. People drop in to consult on matters other than computer training.
TheICTcentrehascreatedselfconfidenceinwomen,creatiedawarenessabout
theirinterestandhelpedthemtakecollectivedecisions.
7.1.1.6Observationsfromtheaboveexamples

Women should be trained with business management/development skills: Women

areculturallynotexposedtoestablishingandrunningabusiness.Infact,in
mostcountries,tradeisgenerallyamaledomainaswellandbusinessescan
succeed or fail by whom you know and do business with. Hence, women
will not only lack business management skills but also business
development/marketingskillsandabusinessnetwork.Womenshouldeither
besupported(metoredbutnotmadereliant)withthenecessaryexpertiseby
a expert or be trained specifically on these skills. For almost all of the
examplesabove,womenwhoareenabledtorunsmallbusinessesarealmost
always provided with a captured market of peers and fellow
villagers/communitymembers,asmallbutstillnecessarytostartoutsidethe
capturedorfamilymarketwhichwillprovidesustainablepath.

Grameens development approach and system of accountability: Grameens

approach, on the other hand, operates on the notion that a woman knows
best on what she has to do given her situation, capacity and needs. What
Grameen spearheaded was a development approach that placed resources
into the hands of poor women and empowered them with control and
decision making over the use of these. In setting up their system of
accountability among their borrowers, Grameen successfully challenged
certain cultural norms and entrenched genderpower dynamics. Thus, in
rural areas, women are generally not perceived to have any meaningful
income generation capacities, and hence they are relegated mainly to

12

householddutiesandcheaplabour.Withoutthepowertoworkandearna
good income, their voices are silenced. Hence economic independency is
veryimportantfacetfortheempowermentofwomen.Inthiscontext,onthe
basisoftheaboveinitiatives,itisclearthatICTsareplayingvitalroleforthe
empowermentofruralanddeprivedsectionswomen8.
7.1.2 ICTandwomenssocialempowerment
Womans social disempowerment is often strongly linked to her isolation from
informationthatsheneedsandthisincludesherability,opportunityandspace
(bothvirtualandnonvirtual)tocommunicateinherownlanguagewithothers
for this information (www.wikibooks). However social empowerment includes
two basic areas like health and education during the last ten years several ICT
based initiatives have been implemented to empower rural and deprived
women.Wewillstudysomeofthekeyinitiativesinthisregard.

WWVP has been helpful to reduce caste and

class-based communication barrier
This ground reality is expressed by agricultural dalit women from Warana region of
South Maharashtra, India. During field work it was observed that before Warana
Wired Village Project (WWVP) the interactions between inter and intra caste - class
group women was less. According to the dalit women Initially I was interested to
know this magical instrument of computer and internet so frequently I used to visit the
WWVP centre regarding agricultural work and like me many other agricultural
womens from different strata also used to visit the centre, apart from agricultural
topics we were discussing on various topics with fellow agricultural women.
The study showed that WWVP has been helpful to minimise the socio-cultural
communication barriers at the village level. This process will create healthy
atmosphere for the social empowerment of rural and deprived women.
Source: Patil D. A. 2007

7.1.2.1Project,Nabanna
InIndia,newinitiativeshavebeentakentoputawebbasedinformationsystem
tostrategicuseforthesocialempowermentofpoordeprivedwomenofBaduria,
aruralregioninNorth24ParanganasDistrictintheIndianStateofWestBengal.

13

Project Nabanna: Benefits for women of an ICT centre in India

As women became involved in the Baduria ICT centre in West Bengal, India they
reported that they gained more respect in their local communities as a result of the
ICT skills acquired at the centre learning to use a computer and accessing and
distributing information to local people. This resulted in greater respect at both the
family and community levels. Younger women felt they were able to approach the
job market with greater confidence. There was also an emergence of solidarity;
since women learned to use computers together at the ICT centre, they often
discussed their problems, creating a sense of unity among them and bringing forth
leadership qualities.
Source: United Nations 2005

After knowing these cases now some initiatives have been observed, especially
runningforhealthandeducationalempowermentofruralanddeprivedwomen.
7.1.2.2SisuSamarakshak,Andhrapradesh
The programme aims at harnessing ICT to accelerate achievement of goals in
womens health childrens growth and development in the early years through
enhanced community monitoring mechanisms and bridge the digital divide.
ThisprojectissponsoredbyGovernmentofAndhraPradeshandUNICEF.Itis
an ICT initiative to impart knowledge to illiterate communities on health,
hygiene and sanitation using audio and culturally appropriate images. The
projectcomponentincludesinformationpages(Usingaudio,picture,videoand
touch tone screen), especially for poor illiterate and semi literate women
(www.ict4d.org).
7.1.2.3Yuva.com:ICTEducationTrainingProgramme
The main objective of this programme is to train rural educated youth (mainly
women, lower income group people and deprived classes) who have passed
SSLC with annual income less than Rs.24,000/ to become an entrepreneurs in
Belguam, Bidar, Chitradurga and Mandya Districts of Karnataka
(www.ict4d.org).
7.1.2.4 Observationsfromtheaboveinitiatives

Technical andsocial limitations: As far as rural ICT initiatives are concern, it

was found that due to womens heavy workloads and multiple roles that

14

limit their available time to use the ICT centres; male attitudes towards
womens use of technology and to women who visit a mixed sex public
facility; the lower educational levels of women compared to those of men
and therefore their lack of literacy skills; the lack of relevant context for
womenintheirlocallanguages;andtheirlackofdisposableincomeforfee
payingcentres,areallgenderbasedfactorsthatcontainthewayofwomen
empowerment through ICTs. Though the numbers of ICT initiatives are
limited, this is a beginning of the silent revolution, especially in traditional
agrariancountrieslikeIndia,SriLanka,Bangladeshetc.
7.1.3ICTandpoliticalempowermentofwomen
Peoplearoundtheworldareusingnewtechnologiesinunprecedentedwaysfor
networking,movementbuilding,politicalparticipationandadvocacy.ICTshave
been applied as agents of change in enabling women to participate directly in
politicsandciviclife.
7.1.3.1SEWAsinitiative
In India, SEWA organises electronic discussions through the panchayati raj
(villagegovernanceinstitutions).Inthesediscussions,villagewomenoftenpose
questionsthatareansweredpromptlybyapanelofexperts.Throughtranslation
modules,responsesgotothewomenintheirvernacularlanguage(Nath2001).
7.1.3.2Womentowomenvideoproject,Pakistan
In the years 1993 to 1994, in the Northwest Frontier Province of Pakistan, the
Deutsche Gesellschaft for Technische Zusammerabeit (GTZ) with the Public
Health Engineering Department of Pakistan introduced a community based
WaterandSanitationProjectthatusedtechnologytoeffectivelyaddressgender
issuesandincludewomeninthedesignandplanningprocesses.Thewomento
womenvideoprojectprovidedwomenwiththeopportunitytospeakacrossthe
physicalboundariesofthepudah(customofseclusion),discusswhatwaterand
sanitation infrastructure could be afforded and participate in choosing the best
watersupplyoptionfortheirhouseholds(www.wikibooks.com).
7.1.3.3 ESeva,AndharaPradesh
A project that possibly captures the potential of ICTs in enabling womens
politicalandcivicparticipationistheeseva(eservices)project,whichbeganin
thedistrictofWestGodavariinAndhraPradesh,India.TheprojectusesICTsto
provide access to various CitizentoCitizen (C2C) and CitizentoGovernment

15

(C2G) services to the people living in rural areas. Under this project, web
enabled rural kiosks termed eseva centres were established at the mandal (a
subdistrictunitofadministration)level.Theuniquethingaboutthesecentresis
that they are run and managed by rural women as information leaders to help
bridgethegenderdivide.
7.1.3.4 Observationsfromtheaboveexamples

Focusonadministrativegoalratherthanmaingoal: It is important to note that

while egovernance initiatives are promising, many focus only on
administrativeefficiencyandgoalandfailtoenablethecitizensfullpolitical
participation in consultative and decisionmaking bodies or mechanisms.
This means that at best such efforts render women as consumers of
informationandnotequalstakeholdersindevelopment.

Questions of relevance: The SEWA model and the Women to Women Video
project show that giving women voice alone is not sufficient if there is a
commitment to addressing gender quality. The voice expressed must be
heardandrespondedtowithsubstantiveaffirmativeaction.Otherwise,even
egovernmentserviceswilllosetheirrelevancetowomeninthelongrun.

8.0 ConclusionandPolicyImplications
Itisimperativeforasustainabledevelopmentindevelopingandoverpopulated
countries like India and other Asian countries, that women have access to
education and appropriate needbased technologies. ICT has been promoted as
an vital tool in ensuring that marginalised groups, particularly women are
includedinthedevelopmentoftheglobalinformationsociety.WhileICTalone
cannot end gender inequality, it cap help catalyse social change and
empowerment(www.ictrd.net.in).
General Assembly resolution 58/146 stressed the need to ensure that rural
womenhaveaccesstoandfullyparticipateintheareaofICT.Effectiveaccessto
and use of ICT can improve rural womens leadership and participation in
community and economic development activities (United Nations 2005:14). On
the other hand rural women are not at the lowest level of the digital gender
divide.AccordingtofindingsoftheITU,therearemanychallengestoimplement
ICTprogrammesespeciallyfortheempowermentofruralanddeprivedwomen.

16

8.1ChallengesofICTuseforwomensempowerment
8.1.1Technicalproblems
Patil (2007) observed that, in India the utilisation of ICT for women
empowerment is in the takeoff stage and obviously there are many field and
technicalproblems.Followingaresomeofthekeyconstraints:

Insufficientregionalspecificinformation

Inadequateinternetconnectivityandelectricity

Inadequatesupportfromfacilitator

LackofInfrastructurefacility

Lackofconstructionofcontentinlocallanguage

Limiteddurationofprojects(especiallysponsoredprojects)

Problemofmonitoringandevaluation

Lackofoperatorsknowledge

LackoftheinvolvementofotherICTchannels(radio,televisionetc.)

Lackofmarketingmanagement/developmentskills.

8.1.2 Socialproblems
WhilestudyingtheproblemsfortheeffectiveimplementationofICTprojectsfor
women empowerment, it is also necessary to reduce social problems because
socialproblemsaredeeplyrootedinsociety.Someofthekeysocialproblemsare
asfollows:
i. Orthodoxandtraditionalattitude
ii. Lackofenablingenvironmentandawareness
iii. Patriarchaldominatedsociety
iv. Lackofeducationandtraining(particularlyruralanddeprivedwomen).
v. Povertyandhugedomesticandagriculturalwork.
vi. Lackofhealthyparticipationfromallstrataofthesociety.
vii. Lackofneedbasedapproach.
viii. Menhegemony
8.2StrategiesforeffectiveuseofICTsforwomenempowerment
Several studies (Adhiguru 2004; Patil 2007; Dhemeja and Mendury 2002; Jain
2007; Dalal 2007, United Nations 2005), and the views expressed by ICT

17

managers,ruralwomenassociationsandexpertssuggestfollowingstrategiesfor
the effective utilisation of ICT for the empowerment of rural and deprived
women.
8.2.1 EnablingenvironmentforICT
There is a need to create an information friendly environment, which supports
andencouragesstrategiestopromotewomensequalaccesstoandopportunity
to benefit from ICT projects as well as creating a regulation and policy
environmentwhichsupportswomensuseofICTs.
8.2.2 Contentdevelopment
Developing content which speaks to womens concerns and reflects their local
knowledge and which is of value for their daily lives, business enterprises and
familyresponsibilities.
8.2.3 Needbaseddevelopmentofsoftware
To improve capacity of women managers/operators of information services to
identifyandsupplytheinformationneedsofusersbasedonthehumancapacity
(literacy,numeracy)ofbeneficiariesandtoutilisetheinformation.
8.2.4 Bridgingthedivide
Strategic development of ICT to address socioeconomic dividers: age, gender,
community,landholdingandgeographyareessential.
8.2.5 Lowcosttraininginstitutes
An emphasised by the advancement of women in 2002, the main barriers to
womensuseofICTcontinuetobelackofpropertrainingespeciallyamongthe
rural and deprived women. In this context Schools, Panchayat offices and
Women SelfHelp Groups (SHGs) should impart ICT training. Students can be
effectivenocosttrainersNSS/NCCstudentscanplayakeyrole.9
8.2.6 Marketingdevelopmentandmanagementskills
Nowadays many ICT projects are not working effectively; marketing
developmentandmanagementskills,thesetwofactors,areveryimportantand
most of the rural and deprived women are lacking in these skills. Hence these
skillsarenecessarytobeincludedinICTtrainingandmodules.

18

8.2.7 Developmentofpublicaccesscentres
One of the strategies adopted to increase access of remote, rural and
marginalisedgroupstoICTisthedevelopmentofpublicaccesscentres,suchas
publicphones,telecentres,libraries,informationcentresorcybercafs.
8.2.8 ICTcentrescanbepartofexistinginstitutions
For the popularity and socially convenient use of ICT centres, the ICT centres
can be part of existing institutions such as health centres, school, community
centres,panchayatandICDScentres.
8.2.9 VenueandtimeofICTcentres
ToensurethatwomentakefulladvantageoftheseICTcentres,itisimportantto
make the venue comfortable and safe. In many cases, the location of and
arrangements around public access centres are decided without keeping the
constraints on women in mind such as inappropriate opening times (including
evening),securityissuesandlackoftransport.
8.2.10 WomentobeinvolvedindesigningICTpolicies
EquitableaccesstoICTtechnologyandtheautonomytoreceiveandproducethe
information relevant to their concerns and perspectives are critical issues for
women. They therefore need to be involved in decision making of new ICT
policies. It is important to engender ICT policy to ensure that women,
particularly rural and poor women, benefit from ICT. Gender must be
considered from the start of project design. Only then can ICT policies and
projects properly address the gender digital divide and further contribute to
womensoverallempowerment.
8.2.11

Participationofdifferentagenciesneeded

For theeffective implementation of ICT projects, no single agency can effectively

deliverthiscriticalinput.Needforaproactiveparticipationbytheprivatesector,
NGOsandothercivilsocietyorganisationsarebeingincreasinglyfelt.

8.2.12Smallandmediumruralenterprises
For the economic empowerment of rural and deprived women, the
establishmentofSmallandMediumEnterprises(SMEs),SmallScaleIndustries
(SSIs),etc.canbeencouraged.
8.2.13Projectplanning
GenderconcernshavetobeaddressedininitialstagesofICTprojectstoensure
that the needs and priorities of both women and men are appropriately

19

considered and that gender equality goals are embedded in project design. A
studyundertakenbytheWorldBankin2002of80ICTprojectsconcludedthat
gender issues were rarely articulated in product design and implementation,
often because donors do not request this information. Hence involving all
stakeholders in the target community, including both women and men, in the
earliestphasesofprojectdesignisimperative.
8.2.14FundsforICTfordevelopment
Gender perspectives should be incorporated into the allocation and
implementationoffundstosupportICTfordevelopment.Issuesofimportance
include increasing the allotment of funds for womenspecific projects and the
consideration of gender perspectives in all aspects of the management of such
funds.Attentionshouldbepaidtoensuringthatsufficientfundsaredirectedto
capacitybuildingwhichisaccessibletowomenandmen.
8.2.15GenderandICTawards10
To create awareness and encourage ICT programmes and managers, it is
necessarytogivegenderICTawards.Inthisdirectiongenderequalityadvocates
have also initiated awards to highlight good practices. Gender equality
advocateshavealsoinitiatedawardstohighlightgoodpractices.Thegenderand
ICT awards programmed was inaugurated during the WSIS in Geneva in
December2003.Thisawardsprogrammewasconceivedtorecognisegenderand
ICT initiatives globally and provide further impetus for mainstreaming gender
perspectivesinthefieldofICT.
8.2.16 ResearchongenderequalityandICT
Research on gender equality and ICT gives vital contribution for the policy
making. Continued research on gender equality and ICT issues at the national
regionalandgloballevelsanddocumentationofgoodpracticesalsocontributes
to deepening the understanding of practitioners and policymakers about the
interplay of these technologies with gender equality and the empowerment of
women. One recent example is a report by UNDP and UNIFEM entitled,
Bridging the Gender Digital Divide, examines opportunities and challenges
faced by women, especially rural and deprived women. This shows the
importanceofresearchtounderstandthegrassrootslevelhurdlesandprovides
proper direction to design effective ICT policies for the empowerment of rural
anddeprivedwomen.

20

Notes:
DedicatedleadersinMaharashtraandGujarathavecontributedtothedevelopmentofseveral
successful cooperatives; late Tatyasaheb Kore is one of them who developed Warana region
through different cooperatives. Warana Wired VillageProject(WWVP) was implementedby
PrimeMinistersSpecialITTaskForcein10thAugust,1998,whichisknownasoneoftheoldest
ICTprojectinAsia.ThefirstauthorofthisarticledidhisPh.D.onWWVPin2007.
1

2
The term informal sector is used mainly with reference to developing countries. The ILO
defines it as an unorganised sector in which economic activities take place outside the
framework of public and private sector establishments. In India this sector cuts across well
defined crafts or industry conglomerates like cottage and household industries, Khadi and
village industries, handlooms, handicrafts, coir, sericulture etc. set up all over the country in
rural,semiurbanandurbanenvironments.

DAWN is concept that arose out of southern Womens awareness of the need for an
alternative model of development. DAWN calls for a model of social advancement which is
equitable, participatory, holistic and sustainable and responds to peoples needs. Although
DAWN emerged out of an NGO forum in 1985, its philosophy reflects the dissatisfaction of
manydevelopingcountriesatthetimewithunfavourabletermsoftrade,protectionismandthe
conditionalityoftheWordBanksStructuralAdjustmentProgrammes(SAPs).

4
SEWA is one of the oldest women organisations which were working in various parts of
Gujarat,mainlyforruralanddeprivedwomen.

As a union, SEWAs current outreach stands at about 530,000 women members. For more
informationonSEWAs,seehttp://www.sewa.org.

Grameen effectively raises consciousness by laying down certain rules of conduct that
borrowers are required to follow. These range from boiling water and setting up their own
safe household water supply to nonpayment of dowry for marriages. Grameen therefore
interfered and challenged cultural norms and given power dynamics within the community
andthefamily.
6

Twentyfive per cent of telephone operators are men, a much larger percentage than their
representationintheborrowingpopulationasawhole(5%).AstheGrameenBankchoosesthe
villageoperators,themalemanagementoftheBankmadeagenderbaseddecisioninselection
ofadisproportionatenumberofmaleoperators.

GrameenTelecomsVillagePhoneProgramme:AMultimediaCaseStudy.TeleCommons
DevelopmentGroup.March2000.See:http//:www.telecommons.com/villagephone/

9
WENTwasanAsiaPacificregionalannualtrainingforwomenintheuseofICTs,andwasa
project that was implemented for five years continuously. At the end of those five years, an
awardwasfundedbytheUnitedNationsEconomicandSocialCommissionforAsiaandthe
Pacific(UNESCAP)asaformofacknowledgementtothemostsuccessfulWENTgraduatewho
appliedherleaningsfromWENTforaselectedcommunityorwithinherownorganisation.

The Gender and ICT Awards is a collaborative project between the APC WNSP and the
GlobalKnowledgePartnership(GKP).TheinauguralawardsceremonytookplaceinGenevain
conjunctionwithWSISinDecember2003,wherechangeinitiativesacceptedtheirawardasthe
winner in the category of Advocacy and Networking. For more information on Nabanna,
IndiaandotherGenderandICTAwardwiningprojects,visithttp://www.genderawards.net
10

21

References
Adhiguru. 2004. Institutional Innovations for Using Information and Communication
Technology. Policy Brief 18. National Centre for Agriculture Economics and Policy
Research,NewDelhi.
Ceewa.2005.CaseStudyofRuralICTInitiativeinUganda,StudyReport,Uganda.
Census.2001.DirectorateCensus,NewDelhi.
Clement, A. 1994. Computing at Work: Empowering Action by low level users. In
CommunicationoftheACM.37,1:5363.
Dalal,P2006.UseofICTforWomenEmpowermentinIndia,Anonymous.
Dhameja and Medury. 2002. Information and Communication Technology in the
GlobalisationEra:TheSocioEconomicConcerns.ResearchReport,IGNU,NewDelhi.
Freire,P.1972.PedagogyoftheOppressed.London:PenguinBooks.
Huyers, S. and Sikuska, T. 2003. Overcoming the Gender Digital Divide: Understanding
ICTs and their Potential for the Empowerment of Women. United Nations
INSTRAWVirtualSeminarSeriesonGenderandICTS.No.1.
Jain, S. 2006. ICTs and Womens Empowerment: Some Case Studies from India,
Anonymous.
Longwe, S. H. 1994. From Welfare to Empowerment. In Office of Women in Reviews,
UNICEF.
Moser, C. 1993. Gender Planning and Development: Theory, Practice and Training.
London:Routledge.
Narayan, D. (ed.). 2005. Measuring Empowerment: CrossDisciplinary Perspectives. The
WorldBank.WashingtonDC.
Nath,V.2001.EmpowermentandGovernancethroughInformationandCommunication
Technologies:WomensPerspective,KnownetInitiative,Anonymous.
Patil, D. A. 2006. Information and Communication Technology (ICT) and Rural
Development:LessonsfromRuralIndia.IASSI,Quarterly,Vol.25,No.2.
Patil, D. A. 2006. New Communications Perspectives for Cooperatives and Rural
Development.JournalofCommunicationStudies,Vol.23.
Patil, D. A. 2007. A Study of Communication Networks and Development in Kolhapur
District,Ph.Dthesis(Sociology),SubmittedtoShivajiUniversity,Kolhapur.
Rowlands, J. 1997. Questioning Empowerment: Working With Women in Honduras,
Oxford: Oxfam GB. UNDP. 2001. Human Development Report, Making New
TechnologiesWorkforHumanDevelopment.London:OxfordUniversityPress.
United Nations. 2005. Women 2000 and Beyond, Division for the Advancement of
WomensDepartmentofEconomicandSocialAffairs,UnitedNations.
Vijayalaxmi, P. and Durgabhavani, V. 2006. Internet Use, Indian Culture and Gender
Variations,MediaAsia,Vol.33,No.34.
http://Ict4d.net.in
http://ehomemakers.net
http://sewa.org
http://telecommons.com
http://wikibooks.com

22

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

SpatialAssessmentofPotentialforGroundwater
DevelopmentinLowerChaoPhrayaRiverBasin
SudsaisinKaewrueng*,MukandS.Babel**,
AshimDasGupta**andShahriarM.Wahid**

Abstract
IntheLowerChaoPhrayaRiverBasinofThailand,developmentofthegroundwater
resourceshasplayedavitalroleinmeetingthebasicwaterrequirementfordomestic
purposes as well as for economic development through its use in the industrial and
agriculturalsectors.Amidstreportsoffallinggroundwaterlevels,landsubsidenceand
saltwaterintrusion,thepotentialforgroundwaterdevelopmenthasbeeninvestigated
with full consideration to spatial distribution of current uses, exploitability and
suitability,anddemandforwater.Owingtothenatureofgroundwaterdevelopment
and nonavailability of sufficient pertinent data, several analytical tools and
techniqueswereusedtoestimatethecurrentuses.Whileseveraldistrictshavealready
reached the critical threshold of exploitation, considerable potential for development
existsinonefourthofthe127districtsinthebasin.Astructuredapproach,basedon
zoningofpotentialareas,isrecommendedforfuturegroundwaterdevelopment.

1.0Introduction
Groundwaterisaninvisibleresource.Sinceitiswidelydistributedinarangeof
waterproducing geological structures (aquifers) and much less dependent on
recent precipitation than surface sources, it can provide a uniquely reliable
source of goodquality and lowcost water for domestic, industrial and
agriculturalpurposes.ManyAsiancountriesheavilydependongroundwaterfor
domesticwatersupplyandtakeadvantageoftheresourcetofacilitateeconomic
activity.TherapidindustrialisationandurbanisationtakingplaceinmanyAsian
countries,however,haveintensifiedthestressplacedonthispreciousresource.
In response to the growing demand for water, groundwater has been depleted
throughexcessiveabstraction,andresultantlandsubsidencehasbeenobserved
insomeareas.Aquifercontaminationwithvarioustypesofpollutantsincluding
saltwater intrusion in coastal aquifers also poses serious challenge. The Lower
ChaoPhrayaRiverbasininThailandisacaseinpoint.

Lecturer,KasetsartUniversity,Bangkok,Thailand.Email:Sudsaisin.Kaewrueng@ait.ac.th
AsianInstituteofTechnology,Pathumthani,Thailand.Email:msbabel@ait.ac.th;
ashim.dasgupta@gmail.com;swahid@gmail.com;
*

**

23

The basin, located in the lower central plain, holds the largest storage of
groundwaterinThailand(RamnarongandWongsawadi1998).Waterisheldin
confined and unconfined aquifers. The largest of these aquifers Bangkok
AquiferSystemcontainseightseparateconfinedaquifersintheUpperTertiary
to Quaternary strata covering Bangkok and its six adjoining provinces:
Ayutthaya, Pathum Thani, Nonthaburi, Nakhon Pathom, Samut Sakhon and
SamutPrakarn.Theeaseofexploitationowingtotheartesianconditionofthese
aquifers and rapid spread of energised pumping technologies mean that
groundwaterdevelopmenthasbeenveryrapidintheLowerChaoPhrayaRiver
Basin.Currently,itisoneofthemostvitalandvaluableresourceforagriculture,
municipalandruralwatersupply,andindustryinthebasin.
UNEP (2003) reported that outside the Bangkok aquifer system, agrowells are
extensivelyinstalledinfloodplainsandinburiedchannelsofthemajorriversfor
agricultural use during the dry period. In irrigation projects in the basin area,
with the rotational supply in command areas, conjunctive use of surface and
groundwater is common as farmers have realised the advantages to increased
productivityoftimelyirrigationandsecurityofapplication.Mostagrowellsare
privately owned and often information on their numbers and amount of
groundwaterextractionisnotavailableorinaccurate.ChulalongkornUniversity
(2002) reported, based on field survey conducted in 1998, that an estimated
17,100agrowellshavehelpedthefarmerstocopewitherraticrainfallandlong
periodsofdryspellandincreaseagriculturalproductivity.Growingthreecrops
inayearorevensevencropsintwoyearshasbecomepossibleduetotheuseof
theseagrowells.
GroundwaterisalsoextensivelyusedbyindustriesinBangkokandsurrounding
areas because of selfcontrol over water supply and lessthandesired level of
access to public piped water, unreliable or insufficient or low quality of piped
watersupply,andlowerpriceofgroundwatercomparedtothatofpipedwater
supply(DGR2004).Theuseofgroundwaterfordomesticsupplyiswidespread
in the basin. This is evident from the records of a large number of agencies
involved in domestic water supply. In municipal areas of Bangkok and the
adjoining provinces of Nonthaburi and Samut Prakarn, the Metropolitan
Waterworks Authority (MWA) is responsible for providing piped water. The
ProvincialWaterworksAuthority(PWA)provideswaterinthemunicipalareas
of remaining 11 provinces in the basin area (Martin et al. 2004). Outside MWA
andPWAserviceareas,pipedwatersystemsaremanagedbyfourgovernment
agencies: Public Works Department (PWD), Department of Health (DoH),

24

Department of Groundwater Resource (DGR) and the Office of Accelerated

Rural Development (ARD). The wells operated by these agencies, known as
Village Piped Water System (VPWS), are now operated by local administration
under technical supervision of DGR, an agency under the Ministry of Natural
Resources and Environment (MONRE) responsible for groundwater
managementinThailand.
The management of extensive groundwater exploitation presents, arguably, a
complex set of issues related to the spatial and temporal dimensions of its
occurrence and use. Already, overpumping has resulted in the falling of
groundwaterlevelupto60minBangkokMetropolitanareaanditsvicinity.This
has triggered land subsidence and salt water intrusion. Though the Thai
government has implemented remedial measures to counteract the crisis and
overall situation has slowly improved in strictly controlled areas, problems of
groundwater overuse and quality deterioration still persist and will pose
environmentalthreatsinthefuture(UNEP2001).
Objectives of the Study: Considering the significance of sound groundwater
management on social and economic fabric of the Lower Chao Phraya River
Basin,thisstudyisundertakentospatiallyassessgroundwaterusepotentialin
thebasin.
The specific objectives are: (1) to comprehensively estimate the current
groundwater use by various water use sectors, (2) to evaluate the quantity of
exploitable resource of suitable quality to meet societal demand for
groundwater, and (3) to identify areas with potential for groundwater
developmentinthebasin.

2.0MethodologyofInvestigation
2.1Studyarea
Thestudyareacovers12provincesintheLowerChaoPhrayaRiverbasin:Chai
Nat, Sing Buri, Ang Thong, Lop Buri, Suphan Buri, Ayutthaya, Pathum Thani,
Nonthaburi, Nakhon Pathom, Samut Sakhon, Samut Prakarn and Bangkok
(Figure1).ThebasinrepresentsanareaofQuaternarydepositsofsilt1530min
depthovertoppingthesoftmarineclayslaiddownwhentheareawasavastbay
oftheSouthChinaSeaabout6,000to8,000yearsago(Sinsakul1997reportedin
Kohnhorstetal.2002).

25

Figure1:TheStudyArea

Theaquifersystemislargelyrechargedbyrainfallandseepagefromrivers.In
the lower central plain where Bangkok is situated, about half of the area is
coveredbythickmarineclay,anditisestimatedthatonly56percentofrainfall
reachestheaquifer.Smallsubbasinsthathavebeenidentifiedduringpetroleum

26

explorationinthecentralplaincontainalluvialfansandfluviatilesedimentsthat
arealsogoodaquifersystems(Martinetal.2004).
At present, about 1.21 million ha land in the study area is under irrigation
through the Great Chao Phraya Project (GCPP). Based on monthly rainfall
recordsin182stationsduring19522002,itisobservedthattheaverageannual
rainfallis1,087mmandvariesbetween680and1,520mmfromyeartoyear.The
totalrainfallisnormallyhigherintheBangkokarea,andlessintheareaslocated
farther inland. About 87 per cent of the total annual rainfall occurs in the six
months of the rainy season (MayOctober). Often, September gets the highest
rainfallinayear,whileJanuarygetsthelowest.
The study is divided into two parts: Firstly estimation of current quantity and
extent of groundwater exploitation, and secondly, assessment of groundwater
developmentpotential.

2.2Estimationofgroundwaterexploitation
Estimation of groundwater exploitation in the study areais complicated by the
differentpatternsofgroundwateruseandcategoriesofinstalledwellsaswellas
variation in corresponding available data type. Table 1 lists different types of
registered (with Department of Groundwater Resources, DGR) and non
registeredwellsusedtoabstractgroundwaterin2002alongwiththenumberof
eachkindofwells.
The private and public wells are classed based on ownership as well as well
depth.Shallowwellsaredefinedasthosewellsthatpenetrateupto30mdepth
anddeepwellspenetratebeyond30m.AccordingtotheGroundwaterAct1977,
deepwellsarerequiredtoberegisteredwithpermittedpumpageforaspecified
purpose (domestic, agricultural and industrial wells) while shallow wells need
notberegistered.Thenonregisteredwellsconstitutethemajority(89.3%)ofthe
totalgroundwaterwells.Detailsofdifferenttypesofwellsusedinthreedifferent
usecategoriesaresummarisedhere:

Domesticuse:Normally, shallow wellsare used fordomestic water supply

in rural areas. They are both private and publicowned. Though private
deep wells or tubewells are reported in the National Rural Development
(NRD2C)DatabasedevelopedbytheCommunityDevelopmentDepartment
(CDD), Ministry of Interior (MOI), Thailand, often they are not registered
withtheDGR.Thepublicdeepwellsdevelopedbygovernmentagencies

27

Table1:NumberofGroundwaterWellsinStudyArea

Categoryofwell/
utilisation

Province
ChaiNat
SingBuri
AngThong
LopBuri

28

SuphanBuri
Ayutthaya
PathumThani
Nothaburi
NakhonPathom
SamutSakhon
SamutPrakarn
Bangkok
Total
Percentage

RegisteredwithDGR1/
Agricul Indus
Domestic
tural
trial
Private Private
Private
deep
deep
deepwell
well
well
6
0
35
326
109
122
46
27
81
14
18
136

NonregisteredwithDGR2/
Sub
total

41
557
154
168

Domestic
Private Public
shallow shallo
well
wwell
1,970
615
610
167
813
129
4,057
695

Private
deep
well
7673
4394
5744
2362

Public
deepwell
(VPWS)
1,649
738
630
1,825

1
8
11
16

PP
WS
4/
0
1
0
0

PW
A3/

Agricul Sub
Total
tural
total number

Agro
ofwells
wells
5/
,260
16,168 16,209
745
8,663
9,220
,782
10,109 10,263
,854
11,809 11,977

Percent
age

18.10
10.30
11.46
13.38

98
392
804
252
144
404
444
643

51
30
28
4
25
13
3
51

206
492
789
230
468
1,261
1,142
684

355
914
1,621
486
637
1,678
1,589
1,378

6,057
2,489
890
174
1,073
146
38
0

1,235
251
22
2
168
27
15
0

6553
371
146
12
1849
178
172
0

1,875
927
341
33
1,681
366
553
72

17
50
57
0
30
44
0
0

0
15
37
2
15
4
2
10

,821
24
2
5
941
96
9
0

19,558
4,327
1,565
238
5,757
861
829
82

19,913
5,241
3,186
724
6,394
2,539
2,418
1,460

22.24
5.85
3.56
0.81
7.14
2.84
2.70
1.63

3,573
3.99

359
0.40

5,646
6.31

9,578
10.70

18,317
20.46

3,326
3.71

29,454
32.89

10,690
11.94

234
0.26

86
0.10

17,859
19.94

79,966
89.30

89,544
100.00

100.00

Remark:PWAisProvincialWaterworksAuthoritywhichprovidespipedwaterfordomesticandindustrialpurposesinmunicipalareas
PPWSisprivatepipedwatersystem
VPWSisvillagepipedwatersystem
Source: 1/dataobtainedfromDGR
2/dataobtainedfromNRD2Cdatabase
3/dataobtainedfromPWA
4/dataobtainedfromDepartmentofWaterResources(DWR)
5/dataobtainedfromRoyalIrrigationDepartment(RID)andintegratedwithNRD2Cdatabase

(PWD, DOHand ARDare not registered with the DGR. These public deep
wells are used to supply water beyond the tapped water serviced area of
PWA through the Village Piped Water System (VPWS). Additionally,
Private Piped Water System (PPWS) are sometimes established by private
groundwater plant operators outside the tapped water serviced areas of
PWA with permission from the Department of Water Resources (DWR) of
theMONRE.

Agricultural use: Most of the agrowells are shallow type and serve
extensivelyassourceofsupplementalandsmallscaleirrigation.

Industrialuse:Therearetwotypesofgroundwateruseforindustrialwater
supply: one is groundwater extraction by private deep wells owned by
individualfactoriesandtheotherisextractionbyMWAandPWAtosupply
waterthroughtheirpipedsupplysystems.

With much difference in data availability of various categories of groundwater

wellsanduses,differenttypesofestimationtoolsandtechniquesareadoptedto
estimateexploitationfordifferenttypesofwells/uses.Theseareexplainedhere.
Groundwaterusebyregisteredwells:TheDGRprovidesregistrationofprivate
deep wells in the study area. During the registration process, the well owners
specify purpose (domestic, agricultural and industrial) and extent of proposed
groundwateruse.Basedonthesespecifications,theDGRregistersthewellsfor
permitted pumpage rates. Available data on the number of wells and
correspondingpermittedpumpageratecanbeusedtocalculatethegroundwater
use by registered wells. However, it is observed that the actual pumpage often
differs from the permitted pumpage. Therefore, Groundwater Pumpage
Coefficient (GPC) ratio of actual pumpage to the permitted pumpage was
estimated from sample wells and used to correct overall estimation of
groundwaterusebyregisteredwells.
To estimate GPCs, actual and permitted pumpage data from 697 domestic, 73
agriculturaland4,434industrialwellsinsevenprovinceswereobtainedfromthe
DGR and analysed. The overall sample size was taken within significant
confidencelevelat99percent(standarderrorof0.01).TheannualaverageGPCs
werefoundtobe0.859,0.669and0.655fordomestic,agriculturalandindustrial
wellsrespectivelywithanoverallaverageof0.691(Table2).Thisfactindicates
thattheactualpumpageinthedomesticwaterusedsectorishighestbutremains
lessthanthepermittedpumpage.

29

Table2:EstimatedGroundwaterPumpageCoefficients(GPC)

Province

Domesticwells
No.ofwells

Ayutthaya

120

GPC

Agriculturalwells
No.ofwells

0.75

GPC

Industrialwells
No.ofwells

GPC

10

0.732

455

0.678

PathumThani

117

0.943

0.722

619

0.534

Nonthaburi

40

0.515

0.11

225

0.72

NakhonPathom

139

1.005

25

1.189

738

0.627

0.972

1,065

0.82

SamutSakhon
SamutPrakarn

20

1.111

850

0.553

Bangkok

260

0.717

28

0.591

482

0.651

Total

697

0.859

73

0.669

4,434

0.655

TheestimatedGPCscomparewellwithearlierstudybytheJapanInternational
Cooperation Agency, JICA (1995) which has estimated overall GPC of 0.67 for
Bangkokanditsvicinityareas.
Groundwaterusebynonregisteredwells:Nonregisteredwellsincludeprivate
nonregistered shallow, deep and agrowells as well as wells managed by
organisations other than DGR. Data on these wells are limited. Several
techniqueswereusedtoestimategroundwaterusebythesewellsdependingon
type of available data. Details of these techniques and their application are
providedhere:

Domestic use through private/public shallow or deep nonregistered wells:

Theseareusedtosupplydomesticwaterinruralareas.Groundwateruseby
these wells was estimated by multiplying the pumping rate
(0.71m3/day/well) proposed by Chulalongkorn University (2002) and
numberofsuchwells.

DomesticusethroughVPWSandPPWS:Mostruralinhabitantsaresupplied
by the public deep wells through VPWS. Others use PPWS supplied by
privatedeeporshallowwellsintheircommunities.Groundwaterexploited
byVPWSorPPWSiscomputedbyanalysingdataonnumberandaverage
size of households served by VPWS or number of users of PPWS, average
household size and average rate of per capita consumption (120 litre/day).
Data on households served by VPWS and average household size were
collected from NRD2C database while data on number of users of PPWS
wereprovidedbytheDepartmentofWaterResources(DWR).

30

Domestic use through MWA and PWA supply: MWA and PWA use
groundwater to supply water through their piped water distribution
systems. Both these agencies maintain data on annual pumpage and the
number of users of their services. These were obtained from respective
officestoestimategroundwaterabstractionbythem.

Industrial use: The industrial use of groundwater was estimated by

analysing the amount of groundwater supplied by MWA and PWA. Both
these agencies use surface and groundwater to serve their clienteles:
households, factories and business houses. The MWA and PWA keep
recordsongroundwaterpumpageandsurfacewatersupplyaswellsasthe
totalamountofpipedwatersuppliedtotheirclienteles.Theratiosofground
and surface water supplied by MWA and PWA service providing stations
were used to differentiate between ground and surface water used by
clienteles. Sum of groundwater use by factories and business houses were
takenasindustrialwaterusefromgroundwatersources.

Agricultural use: To determine agricultural use of groundwater, field

surveys were carried out to estimate pumping hours and rates from agro
wells in rainfed and irrigated paddy farms which dominate agriculture in
the study area. Pumping rates were estimated for 30 agrowells during the
wet and dry seasons of 20042005. Additionally, 19 and 14 farmers in rain
fed and irrigated areas respectively were interviewed in August 2004 to
validateestimationofpumpinghoursandrate.Thisinformationalongwith
data on number of agrowells was used to estimate groundwater use
throughagrowells.

2.3Assessmentofgroundwaterdevelopmentpotential
Groundwaterusevariesconsiderablyoverspaceinthestudyarea.Inthelower
part of the study area extensive use of groundwater is reported and has been
attributedforvariousenvironmentalproblems,suchasloweringofgroundwater
levels, land subsidence and salinity intrusion into the aquifer system. These
problems led to the formulation of various mitigation measures by the
Government of Thailand, such as promulgation of the Groundwater Act,
definingcriticalzones,licensingandmeteringofgroundwaterwells,introducing
groundwater use and preservation charges. Nevertheless, considering limiting
options of alternate sources of water, it is argued that demand of groundwater
for domestic use will persist in the future. This is particularly relevant for
attaining a number of Millennium Development Goals (MDG) of the United
Nations related to water supply. Hence, groundwater development potential is

31

assessed considering aquifer yield, its exploitability, use and quality of supply,
andtwovariablesrelatedtofuturewaterdemandofthedomesticsector.These
areexplainedhere.
Use to Yield (UTY) Ratio: The UTY ratio identifies the present state of
groundwater use against naturally limiting level of sustainable yield. For
calculation of UTY ratio, district was taken as the lowest management decision
unit. It is assumed that groundwater use in a district is homogeneous. Hence,
estimated groundwater use data were reaggregated at district level within
GeographicalInformationSystem(GIS).
Sustainableyieldorsafeyieldisthemaximumamountofgroundwaterthatcan
becontinuouslywithdrawnwithoutcriticallyloweringthegroundwaterlevelor
exceeding natural recharge. Yield was estimated based on amount of natural
recharge through hydrogeological units by rainfall. Based on analysis in the
lower parts of the study area, DGR (2002) estimated that 10.44 per cent of the
annual rainfall contributes to natural recharge. Generally, 10 per cent of the
annual rainfall contributes to natural recharge into unconsolidated rocks of
extensive and highly productive aquifers. For consolidated sediments of
extensive and highly productive, extensive but less productive, and local and
unimportantaquifers,thenaturalrechargeamountsto5,3and2percentofthe
annual rainfall respectively (Ramnarong and Wongsawadi 1998). The
hydrogeological map was originally created in digital format in GIS; it was
overlainandintersectedwiththeadministrativedistrictboundariesofthestudy
area. The average annual rainfall amount was derived from historical data to
prepare the isohytal map of annual average precipitation. These were re
sampled to pixel scale using GIS technique. Information on hydrogeology and
rainfallforeachpixelwasusedtoestimatethepotentialaquiferyieldofthepixel
polygon. Sustainable aquifer yield of a district was then computed based on
proportional aerial contribution of individual polygons yields to the district
basedonthefollowingequation:
YT=a1y1+a2y2+...+anyn

where,YT=totaldistrictsustainableaquiferyield(m3/y);an=areawithin
the district containing a particular hydrogeological unit (km2); and yn =
sustainable aquifer yield of a selected pixel (m3/y/km2) based on the natural
rechargefromrainfall.

32

For interpretation of UTY ratios, scale proposed by Wehmann et al. (2003) was
adopted (Table 3). If the UTY ration falls below 0.25 in a district, groundwater
usepotentialinthedistrictisconsideredtobeverygoodwhileifitsvalueexceed
0.90 the district has the poorest potential for development. The UTY ranges of
0.250.50 and 0.510.90 are considered to indicate good and moderate potential
fordevelopment.AUTYmapwaspreparedbyoverlayinggroundwateruseand
yieldmaps.However,theuncertaintyinaquifersystemshasnotbeenevaluated
andthephysicalpropertiesoffieldwellstudiesaretherebynotused.
Table3: Guidelines for Assessing Potential for Groundwater Development
BasedonUsetoYieldRatio

Usetoyieldratio

Potentialfordevelopment

<0.25

Verygood

0.250.50

Good

0.510.90

Moderate

>0.90

Poor

Source:Wehmannetal.2003

Demand for development: Two variables are argued to dictate the demand for
groundwater development in the future: inhabitants access to domestic water
and availability of sufficient water to meet domestic water requirement.
Consideringthefactthatgroundwateristhemajorsourceofdomesticwater,itis
arguedthatdistrictswhereinhabitantshavelessaccesstodomesticwaterare
better candidates for groundwater development. It is proposed that districts
wherelessthan50percentinhabitantshaveaccesstodomesticwaterisgraded
asverygoodpotentialareasforgroundwaterdevelopmentwhiledistrictswith80
percentinhabitantshavingaccesstodomesticwatershouldbegradedaspoorfor
groundwaterdevelopment(Table4).
Table4: Guidelines for Assessing Potential for Groundwater Development
Based on Demand for Development Arising from Lack of Access to
DomesticWater

Percentageofpopulationwithaccessto
Potentialfordevelopment
domesticwater
<50%

Verygood

5065%

Good

6580%

Moderate

>80%

Poor

33

To determine the percentage of inhabitants having access to domestic water,

information on public wells or Village Piped Water System (VPWS) were
assessed.CorrespondingdatafortheurbanareaswerecollectedfromMWAand
PWA while for the rural areas data were collected from NRD2C database of
Thailand for the year 2003. Data of 6,655 villages in 127 districts covering the
studyareawereextractedfromNRD2Cdatabase.
Table5: Guidelines for Assessing Potential for Groundwater Development
BasedonDemandforDevelopmentArisingfromLackofSufficient
DomesticWater

Potentialfordevelopment

Percentageofvillage/districtwithmorethan
10%householdsgettinginsufficientwater
>10%

Verygood

7.510%

Good

57.5%

Moderate

<5%

Poor

Furthermore, it is argued that even if the inhabitants have access to domestic

water,theymaynothavesufficientquantitytomeetbasicdomesticneeds.This
is particularly relevant for rural areas. Thus, the villages within the study area
were differentiated in accordance with household access to sufficient water (45
litres/capita/day). The threshold for water sufficiency is taken as that opted
during NRD2C data collation. If more than 10 per cent households (HH) in a
villagelacksufficientwater,itisconsideredwaterinsufficientvillagewhileifless
than10percentHHinavillagelacksufficientwateritistermedwatersufficient
village(CDD2003).Sincedistrictistakenasthelowestspatialmanagementunit
inthisstudy,villagedatawerescaleduptodistrictlevelbasedonthepercentage
ofwaterinsufficientvillageinadistrict(Table5).
Table 6: Guidelines for Assessing Potential for Groundwater Development
BasedonExploitability

Wellyield(m3/h)

Extractiondepth(m)
<50

50150

>10

Verygood

Verygood

Good

510

Verygood

Good

Moderate

25

Good

Moderate

Poor

<2

Moderate

Poor

Poor

34

>150

Exploitability of groundwater: Data of mean well yields and depths were

assessed to determine the degree of groundwater exploitability. Data of 5,615
wells were collected from the DGR and analysed. Adopted criteria to
quantitativelyassessexploitationpotentialareshowninTable6.
Suitability of groundwater: To determine the suitability of groundwater for
domestic use, data on Specific Capacity and Total Dissolved Solids (TDS) were
assessed.
A higher Specific Capacity and lower TDS indicate better suitability for
groundwater extraction (Table 7). Data of 5,543 representative wells were used
duringtheanalysis.
Table 7: Guidelines for Assessing Potential for Groundwater Development
BasedonSuitabilityofGroundwater

Specificcapacity
Totaldissolvedsolids(TDS)(mg/L)
3
(m /h/m)
<500
5001,500
>1,500
>3.5
Verygood
Verygood
Good
2.03.5
Verygood
Good
Moderate
0.52.0
Good
Moderate
Poor
<0.5
Moderate
Poor
Poor

Overall groundwater development potential: For quantitative assessment of

potentialforgroundwaterdevelopment,anumericalvalueisassignedbasedon
thelevelofpotentialasverygood(4),good(3),moderate(2)andpoor(1)forall
fiveindicatorsandthentheywereaggregatedtoprovideanoverallpicturesof
districts. All five indicators were given equal weights during aggregation to
avoid biasness of decision. It is proposed to identify four classes of overall
potentialforgroundwaterdevelopment:verygood,good,moderateandpoor.If
adistrictsecuredoverallscorebetween16and20,itisconsideredtohavevery
goodpotentialforgroundwaterdevelopment.Likewise,scoresof1216,812and
less than 8 were taken as indications of good, moderate and poor potential of
groundwaterdevelopmentrespectively.Resultingoverallgroundwaterpotential
wasmappedtovisuallyaidmanagementdecision.

3.0ResultsandDiscussion
3.1Groundwaterexploitation
The total groundwater exploitation was estimated to be 3.625 million m3/d or
1,323millionm3in2002.Ofthis,1.432millionm3/dwasabstractedthroughdeep

35

wells registered with the DGR indicating that about 60 per cent of the
groundwater withdrawal was by the wells which were not registered with the
DGR(Table8).
Agriculture represents the largest user of groundwater. It consumed 1.767
million m3/d (48.75%) through combined use of private deepwells and agro
wells.Agrowellsrepresentthepreferredmodeofabstractioninagriculture.The
average pumping rate of agrowells was estimated to be 385 litres/minute for
wells in the rainfed areawhileit is800 litres/minute in theirrigatedareas. On
averageagrowellsintherainfedareawereoperatedfor1,110and1,450hours
duringthewetanddryseasonsrespectively.Ontheotherhandagrowellsinthe
irrigated area were operated for 285 and 385 hours during the wet and dry
seasonrespectively.Fieldsurveydatarevealedthatabout65percentofinstalled
agrowells in the irrigated area remains operational during the wet season.
Industrial and domestic sectors consumed 0.969 and 0.888 million m3/d of
groundwaterannuallyrepresenting26.74percentand24.51percentofthetotal
groundwateruserespectively(Table8).
It is observed that about 89 per cent of the total used groundwater was
abstracted by private operators while the reminder was abstracted by public
serviceproviders.Amongserviceproviders,MWAandPWAdistributed74,778
m3/d(2.06%)and55,255m3/d(1.52%)ofgroundwatertodomesticandindustrial
sectors respectively during 2002. Nonregistered private deep/shallow and
public shallow wells together abstracted an estimated 38,004 m3/d (1.05%) of
groundwaterfordomesticpurpose.VPWSandPPWSsupplied262,099and4,256
m3/dofgroundwaterrespectivelyfordomesticuserepresenting7.35percentof
thetotalabstraction.
Among provinces within the study area, groundwater use varied from 2.13 to
13.55 per cent. Nonthaburi Province consumed lowest groundwater while
SuphanBuriprovincethehighest.Provincesintheupperpartofthestudyarea
arealsohighgroundwaterusers(Figure2andTable1).
On the other hand, in the lower part of the study area higher amount of
groundwater isabstracted for domestic and industrial purposes.Pathum Thani
andSamutSakhonarethelargestgroundwaterusersfordomesticandindustrial
purposes,respectively.

36

Table8:EstimatedGroundwaterUse(m3/d)in2002

Province

RegisteredwithDGR
Domestic

Agri
cultural

Indus
trial

NonregisteredwithDGR
Subtotal

Agricul
tural

Domestic
Private
Public Private
shallow shallowwell deepwell
well

Public
deepwell
(VPWS)

PPW
S

PWA1/

MW
A1/

Sub
total

Indus
trial

Agro
wells

Total
Percent
volume
age
ofuse

Sub
total

PWA&
MWA

430

2,869

3,299

1,399

437

6,063

15,795

23,694

406,181

429,875

433,174

11.95

SingBuri
AngThong
LopBuri
SuphanBuri
Ayutthaya
PathumThani
Nothaburi
Nakhon
Pathom

1,362
133
508
1,641
38,862
204,181
31,302

728
264
2127
590
1,498
891
47

6,277
2,180
17420
3,736
124,102
159,947
29,397

8,367
2,577
20,055
5,967
164,462
365,019
60,746

433
577
2,880
4,300
1,767
632
124

119
92
493
877
178
16
1

3,740
4,384
2,139
4,818
147
28
0

11,822
15,682
27,688
44,521
38,492
19,805
14,736

8,287
4326
2,514
11,150
5,351
1,159

486
0
0
0
402
400
400

03/

24,887
25,061
35,714
65,666
46,337
22,040
15,261

226,718
245,213
322,098
411,754
20,665
5,922
1,224

3,603
1,354
1,046
7,745
2,738
1,241
0

255,208
271,628
358,858
485,165
69,740
29,203
16,485

263,575
274,205
378,913
491,132
234,202
394,222
77,231

7.27
7.57
10.45
13.55
6.46
10.88
2.13

24,612

1,068

66,696

92,376

762

119

1,274

41,705

11,341

836

56,037

97,880

8,396

162,313

254,689

7.03

SamutSakhon
SamutPrakarn
Bangkok
(m3/d)
(106m3/y)
Percentage

58,945
53,740
93,439
509,155
185.8
14.05

169
23
1,250
8,655
3.2
0.24

246,456
147,586
107,533
914,199
333.7
25.22

305,570
201,349
202,222
1,432,009
522.7
39.50

104
27
0
13,005
4.7
0.36

19
11
0
2,362
0.9
0.07

39
5
0
22,637
8.3
0.62

10,597
19,456
1,800
262,099
95.7
7.23

23,047

67,175
24.5
1.85

835

402
03/
495
7,6033/
4,256 7,603
1.6
2.8
0.12
0.21

34,641
13,631
19,901
7,246
9,898
0
379,137 1,758,532
138.4
641.9
10.46
48.51

25,502
0
3,630
55,255
20.2
1.52

73,774
379,344 10.47
27,147
228,496
6.3
13,528
215,750
5.94
2,192,924 3,624,933 100.00
800.4
1,323.1

60.50
100.00

ChaiNat

37

Note:MWAisMetropolitanWaterworksAuthority

PWAisProvincialWaterworksAuthority
PPWSisprivatepipedwatersystem
VPWSisvillagepipedwatersystemorpublicdeepwells.

1/isthewholeabstractionidentifiedtobefordomesticpurposes
2/isthewholeabstractionidentifiedtobeforbusinessandindustrialuses
3/standsforMWAsupplyinBangkok,NonthaburiandSamutprakarnprovinces

37

Figure2:SpatialDistributionofGroundwaterWellInstallation

3.2Groundwaterdevelopmentpotential
Groundwaterdevelopmentpotentialwasassessedconsideringaquiferyield,its
exploitability, use and quality of supply, and domestic water demand. Yield is
takenasequaltonaturalrecharge.Occurrenceofalluvialaquifer(Qfd)through
majority of the study area generally indicates condition for high recharge rate
(Figure1).
Anaturalrechargeof4.825millionm3/d,orabout7.9percentofaverageannual
rainfallof1,087mm,wasestimated.Ingeneral,groundwaterexploitabilityand
suitabilityaregoodinmostofthedistricts.Ofthe127districtsinthestudyarea,
118 districts have exploitability exceeding 5m3/h within 150m from the ground
surface. Groundwatersuitability, in terms of TDS and Specific Capacity, shows
that 122 districts have Specific Capacity exceeding 2m3/h/m while still
maintainingTDSbelow1,500mg/L,thethresholdforfreshwater.

38

Figure3:GroundwaterDevelopmentPotentialBasedon(a)UsetoYield(UTY)
Ratio, (b) Demand for Water due to insufficient domestic water
supply,(c)GroundwaterExploitability,and(d)OverallPotentialfor
FutureDevelopment

(a)

(b)

(c)

(d)
39

Table9:Number of Districts in Provinces Differentiated Based on Potential

forGroundwaterDevelopment

Province No.of Potentialfor

dis
development
tricts

Chai
Nat

Sing
Buri

Ang
Thong

Lop
Buri

11

Suphan
Buri

10

Ayutth
aya

16

Pathum
Thani

Non
thaburi

Nakhon
Pathom

Samuth
Sakhon

Samuth
Prakan

Bang
kok

40

Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor
Verygood
Good
Moderate
Poor

Useto
yield
ratio

4
4
3
1
3
4
2
1
3
5
5
4
2

3
1
3
1
2
3

3
1
1
2

1
2

4
23
5
4
8

Numberofdistrictsbasedon
Exploi
Lackofaccess
Lackof
tability
todomestic
sufficient
watersource
amountof
domestic
water

4
4
1

2
4
7
2

6
6

1
1
1

5
6

1
5
5

5
5
1
1
5
5

1
4
8
2
2
2
1
3

6
1

3
11

2
5

16
9

1
4

3
3
2

4
4

2
6

6
4

1
4
1
1

6
5

1
1
1

1
2

2
1

5
1

4
2

1
4
1

35

3
40

40

40

Suita
bility

6
2

10
1

10

16

6
1

5
1

2
1

3
32
3
2

Integrated
Potential

4
4

1
5

4
3

3
7
1

4
6

1
14
1

1
3
3

4
2

6
1

5
1

22
17
1

However, the estimated overall Use to Yield (UTY) ratio (around 0.75) is a
pointer for opting strategic management regime in the future. Comparatively
higherUTYvalues(UTY>0.9)inthelower(eastofBangkokandwestofSamut
Sakhon, Nakhon Pathom) and middle of the upper parts (Pathum Thani, Ang
Thong, Suphan Buri,Sing Buriand Chai Nat) of the studyarea indicate higher
exploitationbyindustrialwellsand/oragrowells(Figure3(a)).
Inhabitants in seven districts, out of 127 in the study area, have poor and
moderateaccesstodomesticwaterindicatingthatthesedistrictsholdverygood
orgoodpotentialforgroundwaterdevelopment(Table9).
In32districtsinhabitantshaveinsufficientsupplyofdomesticwaterindicating
thattheywillhavedemandforgroundwaterdevelopmentinthefuture.Overall,
considering all the indicators, 87 districts have good or very good potential for
groundwaterdevelopment(Figure3(d)forlocation).However,aclosescrutiny
reveals thatin 16 districts of these87districts UTY ratiois very highimposing
restrictiononfurthergroundwaterextraction.

4.0 Conclusions
Groundwater is abstracted in the Lower Chao Phraya Basin of Thailand by a
number of agencies in the public sector as well as private well owners for
agricultural, industrial and domestic uses. Data and information necessary to
quantify the amount of abstraction are inadequate and poses considerable
challenge for sustainable development and management of the resource. This
study applied a number of techniques involving data on withdrawal and/or
consumption trends to estimate the groundwater use at district level in the
LowerChaoPhrayaBasin.Analysisandresultsindicatedthatanestimated1,323
million m3 groundwater was abstracted in 2002 and about 50 per cent of the
water was consumed by the agricultural sector. Furthermore, majority of the
populationintheruralareasaredependentongroundwaterfordomesticuse.
The extensive use of groundwater brings forth the obvious question of
sustainability of groundwater use in the future. Hence, this study attempted to
assess the potential for groundwater development considering availability and
pressing demand for the resource. An indicatorbased evaluation scheme was
appliedtospatiallydifferentiatethepotentialfordevelopment.Resultspointout
that several districts of the Lower Chao Phraya Basin are in a critical stage of
groundwater use and further development is not warranted. However, 25 per
centofdistrictshaveverygoodpotentialforfurtherdevelopment.

41

It is suggested that, though sustainable yield were obtained from empirical

relationship between rainfall and recharge, a differential management strategy
that recognises the varying potential for development among districts can be
adopted. To further consolidate the findings, safe yield of the aquifer is
recommendedtobeestimatedbasedonthehydrogeologyofthebasin.

References

CDD (Community Development Department). 2003. Guideline for National Rural

Development Database (NRD2C) questionnaire. Community Development
Department,MinistryofInterior,Bangkok(inThai).Pp.71.
Chulalongkorn University. 2002. Groundwater management in upper part of the lower Chao
PhrayaRiverbasin.FinalMainReport(inThai).Pp.431.
DGR(DepartmentofGroundwaterResources).2002.Effectsofgroundwaterrechargeonland
subsidenceandgroundwaterquality:Mathematicalmodelstudy,DraftFinalMainReport
(in Thai). Department of Groundwater Resources, Ministry of Natural Resources
andEnvironment,Thailand.
DGR (Department of Groundwater Resources). 2004. Effect of groundwater over pumping
mitigation: Mathematical model study, Final Report. Department of Groundwater
Resources,MinistryofNaturalResourcesandEnvironment,Thailand.
JICA (Japan International Cooperation Agency). 1995. The study on management of
groundwaterandlandsubsidenceintheBangkokMetropolitanareaanditsvicinity,final
mainreport.Pp.376.
Kohnhorst,A.,L.Allan,P.Pokethitiyoke,andS.Anyapo.2002.GroundwaterArsenicin
central Thailand. In: Sustainable Environmental Sanitation and Water Services.
Preprintsof28thWEDCConferenceKolkata(Calcutta)India.Pp.123125.
Martin,R.,R.Savage,andR.Pyvis.2004.Astudytodevelopastrategytoexpandruralwater
infrastructurewithintherestructuredwaterandwastewatersectorinThailand,Ministry
of Finance, Draft Final Report. PublicPrivate Infrastructure Advisory Facility
(PPIAF).Pp.105.
Ramnarong,V.andS.Wongsawadi1998.PotentialofgroundwaterinThailand,Journalof
HydrologistAssemblyVol.2,Pp240276(inThai).
Sinsakul,S. 1997. Country Report: Late Quaternary Geology of the Lower Central Plain,
Thailand. Inter. Symp. on Quaternary Environmental Change in the Asia and
WesternPacificRegion,Oct.1417,1997,U.ofTokyo,Tokyo,Japan.
UNEP (United Nations Environment Programme). 2001. Bangkok state of the
environment,EnvironmentalQualityManagementandControlDivision,Officeof
the Permanent Secretary for the Bangkok Metropolitan Administration. BMA.
Pp.104.
UNEP (United Nations Environment Programme). 2003. Groundwater and its
susceptibilitytodegradation:aglobalassessmentoftheproblemsandoptionsfor
management,Div.ofEarlyWarningandAssessment,UNEP,Kenya.Pp.138.
Wehmann,H.A.,S.V.Sinclair,andT.P.Bryant.2003.Ananalysisofgroundwateruseto
aquiferpotentialyieldinIllinois,GroundwaterSection,IllinoisStateWaterSurvey.
Pp.30.

42

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

MigrationofRuralPeopletoUrbanAreas:
AStudyofThreeUpazilasinMymensingh
GulsanAraParvin,MamunMuntasirRahman*,FarhanaYasmin*

Abstract
This study is an approach of that facts finding, which would create a basis for further
policyformulationtoconvertmigrationasacatalystofdevelopment.Throughempirical
studiesitintendstoseekansweroftwoqueriesthatarewhomigratefromruraltourban
areasandwhatarethedifferencesbetweenmigrantandnonmigranthouseholdsliving
intheplaceoforigini.e.ruralareas.Ithasbeenfoundthattheoverwhelmingmajorities
(60%) are migrating within the age1635 years to seek job, and they are primarily
migrating to Dhaka city. It has been also noticed that there is wide difference between
migrantandnonmigrantfamiliesindifferentsocioeconomicindicators.Familieshaving
migrant members are in a better position in housing condition, income and land
ownership.

1.0Introduction
Accordingtoscholars,definitionandclassificationofmigrationisnoteasyand
straight forward. In the traditional definition, migration means movement of
peoplefromoneplacetoanother,temporarilyorpermanentlyinsearchofbetter
life,livelihoodortoavoidthreatoflifeandlivelihood(Siddiqui2005).Whatever
the definition is migration is one of the crucial phenomena in both developed
and developing countries though the nature and facts are different in different
places. In Asia among the developing countries Bangladesh bears significant
attention in the issue of migration, which is recognised as an important
livelihood strategy for the people of Bangladesh (Nazem 1998 and Siddiqui
2005).BeingoneofthemembersofLeastDevelopedCountries(LDCs),lowlevel
of economic growth, lack of employment generation, rapid population growth,
highpopulationdensity,wideruralurbandisparities,frequentnaturaldisaster,
political unrest and corruption have made the people of Bangladesh bound to
searchforanewplaceforlivelihood.
Both in the matter of internal and international migration, Bangladesh attain,
greatconcernofpolicymakersandresearchers.Internalmigrationintheformof
rural to urban migration (4.5%) is primarily responsible for the rapid urban

UrbanandRegionalPlanningDepartment,BUET,Dhaka.Email:gulsanaraparvin@urp.buet.ac.bd;
mamunmuntasir@urp.buet.ac.bd;farhanayasmin@urp.buet.ac.bd

43

growth,whichisestimatedat3.2percentwithinyear1991to2001(BBS2003and
Nazem1998).StudyconductedbySkeldon(2005)hasclaimedthatinBangladesh
40percentofruralhouseholdssendadultmemberstoseekworkintownandin
some areas this figure is more than 80 per cent. On one side ruralurban
migration is the principal cause of rapid urban growth and on the other side,
international migration is playing vital role in national economy by reducing
unemployment and adding significant amount of foreign exchanges (Sidiqui
2005andDailyIttefaq,27June2007).
Therefore, due to being key feature of national socioeconomic structure,
migration issues should be diagnosed properly and adequately. At this stage a
numberofscholarsdominatedbyRitaAfserandTasneemSiddiquihavemade
greatcontributions.ButstillitisnoticedthatlikemanyotherLDCsBangladesh
has poor data base and researches related to various dimensions of migration
(Afsar 2000). Therefore, though migration issue is well addressed by some
legendaryscholars,itisnotoldyetandmigrationstillcompriseswideinterestof
researchers.

2.0ObjectiveofStudy
Though migration is not a new phenomenon in developing country in 21st
century, it has emerged as the top of policy agenda in many countries and
Bangladeshisoneofsuchcountries(Skeldon2005).Itisadvisedthatinthelight
ofthechangessweepingthroughruralBangladesh,thelinkbetweenmigration
anddevelopmentisneededtobereexamined(Afsar2005).Beforetheinitiative
of exploring this link it is necessary to be familiar with the forms and facts of
ruralmigration.Inthiscontext,thisstudyaimstoinvestigatetheissueslikewho
migrate, where they migrate, for what purpose and what are the differences
betweenmigrantandnonmigrantfamiliesintheplaceoforigin.Thisstudyisan
approach of these facts finding, which would create a basis for further policy
formulationtoconvertmigrationasacatalystofdevelopment.

3.0MethodsofInvestigation
Beingconcerntothelimitationoftimeandresourcesthispaperdelimitsitsscope
of investigation in the three upazilas (administrativeareabelowthedistrictlevel) of
Mymensingh, which is one of the largest districts of Bangladesh, is not well
addressedinmigrationrelatedstudies.Therefore,thisstudyhasconcentratedto
Mymensingh District. This district is located at the north end of the country
having Meghalaya of India at its top north. Out of 12 upazilas of Mymensingh

44

threeupazilas named Muktagacha, Haluaghat and Gafargao have been selected

asstudyareasfromthreesidesofthedistrict.Theseupazilashavebeenselected
on the basis of the information provided by the officials of upazila parishad
about the number of families having migrant members. Among the three
upazilas,onehaslargenumberoffamilieshavingmigrantmembers,anotherhas
mediumconcentrationsofmigrantfamiliesandthirdonehaslowconcentration
ofmigrantfamilies.Afterfinalisingtheselectionofupazilas180households(60
from each upazila) have been selected randomly for questionnaire survey. No
statisticalmethodhasbeenfollowedtodeterminesamplesize
Throughempiricalstudiesusingfieldinvestigationandquestionnairesurveyit
intendstoseekansweroftwobroadquestions.Thesequeriesare:
i)

Whomigratefromruralareas;and

ii) Whatarethedifferencesbetweenmigrantandnonmigrantfamilies?
Toexaminethefirstquerythispaperexaminesvariousformsofmigrants.Here,
forms indicate issues such as the ratio of migrant and nonmigrant families,
profile of migrants, trends and purpose of migration, and the destination of
migrants. On the other hand, to answer the second question, it attempts to
investigate the fact like how different socioeconomic issues vary between
migrantfamily(familieshavingatleastonemigrantmember)andnonmigrant
family.Atthisstageitwouldbetriedtodepictcomparativepictureofmigrant
andnonmigranthouseholdsinthecaseofvarioussocioeconomicissuessuchas
familytypeandsize,housingcondition,incomeandoccupation,landownership
andaccesstovariousservicesandfacilities.Itshouldbenotedthatheresurvey
hasbeenconductedattheplaceoforigin,wherefamilymembersofthemigrants
live.Hereunder,thefollowingsectionanalysisbasedonprimarydatahasbeen
presented.

4.0ResultsandDiscussions
4.1Whomigratefromruralareas?
InBangladeshtherearealreadysomeresearchesthathaveinvestigatedvarious
characteristics of migrants and have given the answer of the questionwho
migratefromruralareas.Buttherearelittleattemptstoseekthisanswerforall
parts of the country. Therefore, this study has focused on one of the districts,
which has not got much attention at this regard. The three upazilas of
Mymensingh District studied under this researchare differentin some casesof

45

their socioeconomic and physical condition though there are similarities in

manyaspects.InordertopresenttheoverallconditionofMymensinghDistrict,
it has been tried to select three different upazilas having different settings.
Thoughsocioeconomic,physicalandevenpoliticalcontextofalocalityisvery
important in occurrence of migration from an area here the characteristics of
migrantshavenotbeenpresentedlocalitywise.
4.1.1Extentofmigrationinstudyarea
In order to know the extent of migration in the locality, it has been tried to
identifythenumberofhouseholdshavingatleastonemigrantmemberandthe
households which have no migrant members. Here, the household that has at
least one member who has migrated outside form the locality is called as
Migrant Household and in contrast household that has no migrant (out
migration)memberinthefamilyistermedasNonMigrantHousehold.
Table1:DistributionofHouseholdAccordingtoMigration

Migration
Householdhavingmigrantmember
Householdhavingnomigrantmembers
Total

Numberofhousehold
71
109
180

Percentage
39
61
100

Source:FieldSurvey,January,2007

AttheintroductionsectionitisalreadymentionedthatinBangladesh40percent
of rural households have adult members who have migrated to towns for
livelihood. This study area is not exception from that. The table demonstrates
that39percentofthehouseholdshavemigrantmembers.Itshouldbenotedthat
there are total 108 migrant members migrated from 71 households. It denotes
thataverage1.5personsaremigratingfromeachhousehold.
Figure 1 presents interesting result by showing variation in the number of
migrant and nonmigrant households in three upazilas. Here, Gafargao upazila
has the highest concentration of migrant households. About, 60 per cent of the
households of Gaforgao have at least one migrant member. It should be
mentioned that this upazila is located by the side of Brahmaputra River and
almost every year people of this area are needed to face severe flood and river
erosion.Duetothenaturaldisasterpeopleofthisareahavehighertendencyto
migrate for seeking job. It is interesting that though they are vulnerable in
naturaldisaster,theyarenotmigratingwithfullfamilyratheroneor2members
aremigratingfortheirsecuredlivelihood.

46

Percentage

Figure 1: Migrant and Non-Migrant Households in Different Upazilas

120%
100%
80%
60%
40%
20%
0%

60%
40%
Haluaghat

37%
85%
63%
15%
Muktagacha

Gafargao

Names of Studied Upazilas

% of NonMigrants
Households
% of Migrant
Households

Note: Migrantfamilymeansthefamilyhasatleastonemigrantmemberinthefamily

NonMigrantFamilymeansthefamilywhichhasnomigrantmember

Source:FieldSurvey,January,2007

Unlike Gafargao, Muktagacja has the lowest number of migrant households,

whichisonly15percentofthetotalhouseholds.Muktagachaisawellserved
upazilaofMymensingh.Itisoneoftheoldestupazilasconsistsof1municipality
(town)and10unions(ruralareas).Thisupazilahas4collegeand40highschools
whichisexceptionallybetterthatmanyupazilasofBangladesh.Moreover,there
are a number of mills and factories (15 biscuit factories and 12 oil mills)
providing job opportunities to the local people. Theses social and economic
facilitieshavesignificantlyreducedtheoutmigrationfromthisarea.
Lastly Haluaghat is an upazila that is relatively in a remote location. It
accommodates a large number of tribal communities, Hindu and Christian
households. Number of migrant families in this area is neither so large nor so
small (40%). It has been noticed that those who migrate from this are mostly
belong to the nonMuslim community and these people migrate for better
livelihoodandeducation.
Duringexaminingtheextentofmigrationfromthestudyareaithasbeenclearly
noticedthatsocialandeconomicpotentialitiesofanareaprimarilydeterminethe
areas level of out migration. Next hereunder, the characteristics of migrant
membershavebeenpresented.

47

4.1.2Characteristicsofmigrant
Duringidentifyingthecharacteristicsofmigranttheirage,educationlevel,time
periodofmigration,purposesanddestinationofmigrationhavebeenexamined.
Followingsectionspresentsmigrantscharacteristics.
Table2:DistributionofMigrantAccordingtoTheirAgeand
EducationLevel

Percentage

Literacy

Percentage

1.85

Illiterate

12.03

16to25

26.85

Signature

0.92

26to35

34.26

Primary

5.55

36to45

18.52

Madrasha

0.92

46to55

12.04

Secondary

21.3

56to65

3.70

HigherSecondary

23.15

Above65

2.78

Graduate

28.7

100.00

PostGraduate

7.4

Total

100

Agegroup
Lessthan15

Total
Note:Totalmigrantmembersare108andallaremalemembers
Source:FieldSurvey,January,2007

4.1.2.1Ageandliteracylevelofthemigrants
The age group analysis of the outmigrated population indicates that majority
(around 92%) have migrated during their age of 16 to 35 years. Studies
conducted by Afsar in 2000 and 2001 (cited in Afsar 2005) have got the similar
findings. In fact, 1635 is the age of the highest productivity and enthusiasm
when people can take any dynamic decision and migration is one of such
decisions. This age cluster resembles the reasons of migration from the study
areas where majority of the migration occurred due to service and education,
usuallyincurredbythisageclass.
The education qualification of the migrated population shows that large
proportion (about 57%) of the migrants have a considerable level of higher
education starting from higher secondary to postgraduate. About 44 per cent
migrants have passed Secondary and Higher Secondary School Examinations
and more than one third of the migrants (36%) have graduated. If we compare
this scenario with the purpose of migration, it is evident that most of these
people have shifted out from their locality to obtain further education and to

48

searchforjobtoearnabetterlivelihood.Thisalsodepictsthelackofsufficient
higher level educational institutions (especially college and tertiary institutes)
andinsufficientserviceindustryfortheeducatedmass.
TrendsofMigration

No. of Migrant (person)

Figure 2: Number
of Migrated
Persons
in Different
Time
Fi g.2 No.
of Mi grate
d Pe rson
i n Di ffe
re nPeriod
t Ti m e
60

Pe ri od

55

50
40

28

30
20

10

13

81
19

85
19
86
19

90
19
91
19

5
06
00
00
20
20
e
-2
v
01
96
bo
A
20
19

95
19

T ime P eriod

Source:Fieldsurvey,January,2007

The trend of migration in the study area has been examined on the basis of a
time line since 1980 to 2006. The data show that the tendency to migrate was
verylowin the decade19801990 when onlyaround4 per centof total present
day migration was taken place (total 108 persons). A sudden boom in shift of
localinhabitantstodifferentlocationoutsidetheirlocalitytookplacesince1996
(28personsi.e.26%oftotal)anditreachedtothepeakin2005whenmorethan
half of the total migrant members (51%) migrated from the locality. Rapid
flourishofgarmentindustries,latherprocessingindustries,shrimpculturesince
1990to2000andlaterrealestatebusinessinDhakacitymighthavecontributed
to the dramatic rise of out migration flow. It should be mentioned that among
the migrants majority have migrated to Dhaka city and for services. Though
thereisnoclearexplanationofsharpfallintheflowofmigrationinyear2006it
is thought that the political unrest and frequent labour unrest in garment
industriesinthewholeyearof2006aretheresponsiblefactorsbehindsuchfall.
4.1.2.2Purposeanddestinationofmigrants
It is known that people migrate from one place to another for seeking a better
livelihoodortoescapefromhazard,riskorsuchkindofthreatsoflife.Earlierit
has been mentioned that three studied upazilas of Mymensingh District have

49

three different settings. So, the purposes or reasons of migration from these
different areas should be different. But it is noticed that there is not much
variationsamongthethreeupazilasinthecaseofthepurposesofmigration.
Table3:PurposesofMigrationfromRuraltoUrbanAreas

Purposes
Services
Jobinworkshopsorfactory
Shopkeeping

Numberofmigratedpersons
77
4
2

Percentage
71
4
2

Rickshawpuller
Business
Education
Migratewithhusband

1
6
16
2

1
6
15
2

Total

108

100

Source:Fieldsurvey,January,2007

Like the normal feature of migration here people are primarily migrating for
seekingbetterlivelihood.Morethan80percentofthepeoplehavemigratedfor
different economic activities. The main reason forcing people to leave their
dwellingsaredominatedbypullfactorslikeformalservicesectoremployment.
Accordingtothemigrants,theyaremostlymigratingforservices.Hereitshould
be noted that this service sector includes jobs in different garments industries,
factories, real estate companies and different business offices. Majority of these
migrants,havesecondaryandhighersecondarylevelofeducation.So,theyare
not able to have a higher position in the service sectors. But those who have
graduationlevelofeducationhavemigratedforgovernmentservices,whichare
mostlyteaching.Inthelocalityseekingbetterorhighereducationisidentifiedas
another important purpose of migration (15%). Gafargao and Haluaghat which
are relatively deprived in higher educational facilities accommodate more
numberofmigrantswhohavemigratedforeducationpurpose.Nowitisneeded
toknowwherethesemigrants,gotoseektheirlivelihoodandeducation.
Since social and economic opportunities coupled with other services and
amenities are available in the larger towns and metropolis, most of the cases
destinationsofthemigrantsarecities,especiallyDhaka.Thesurveyfindingsalso
shows that almost half of the total outmigration (around 48 per cent of total
migrant and 54 per cent of migrant within the country) from Mymensingh
regionhasbeentoDhaka,thecapitalofBangladeshandthehubofmajortrade,
commerce and industry. Moreover, it is also interesting to notice that since

50

Mymensingh Townis thenearest urban centre andoffers better living facilities

and constitutes the attraction elements such as employment, entrepreneurship
facility and better education facilities, a considerable part (around 20%) of the
total outmigration has been within the district of Mymensingh, especially
districtsadar(centre).Someotherinternalmigrationisalsoobservedthoughitis
negligible.

Table4:DestinationsofMigrants

Destination
(InsideBangladesh)
Dhaka

Numberofmigratedpeople
52(54%ofmigrationinside
thecountry)

Mymensingh

21

Muktagacha

Destination
(Outside
Bangladesh)
Dubai

Numberof
migrated
people
4

SaudiArab

Italy

CharAlgi

France

Chittagong

USA

Valuka

Narayanganj

Ghorashal

Gazipur

Shalotia

Khulna

Sylhet

Kishorganj

Norail

Tangail

Total

97

Total

11

Source:Fieldsurvey,January,2007

Table 4 reveals that only 10 per cent of total migrant members (108) have
migrated to outside the country and the place is dominated byMiddle East.In
fact,BangladeshexportscontactlabourmostlytoMiddleEastandSouthEastern
countries. Therefore, the people having relatively low level of education
(secondaryandhighersecondarylevel)trytoseekjobasalabourinDubaiSaudi
Arabia and such other countries. In contrast, people having a better level of

51

educationandhaveanyreferencepersontoEuropeancountriesandUSAtryto
migratethere.
From the above sections various characteristics of migrant have been known.
Nowinthefollowingsectionithasbeentriedtopresentacomparativepictureof
migrant and nonmigrant families on the basis of different socioeconomic
conditions.

4.2Whatarethedifferencesbetweenmigrantand
nonmigrantfamilies?
Inordertoassessthedifferencesbetweenmigrantandnonmigranthouseholds,
a comparative analysis has been done. To compare between the households
some variables like family type and size, their housing condition, income and
occupation, land ownership and access to various services and facilities have
beenconsidered.
4.2.1Familytypeandsizeofmigrantandnonmigranthouseholds
Inthispartfamilytypeandsizeofmigrantandnonmigranthouseholdshave
beencomparedandithasbeentriedtoexplorethedifferences.
Table5:FamilyTypeandSizeofMigrantandNonmigrantHouseholds

Familytype

Nonmigranthousehold

Migranthousehold

No.of
household

Percentageof
total

No.ofhousehold

Percentageof
total

Singlefamily
Jointfamily
Total

88
21
109

81
19
100

42
29
71

59
41
100

Familysize

No.of
household

Nonmigranthousehold

24
59
1014
1519
20+
Total

47
51
5
3
3
109

Migranthousehold

Percentageof
total

No.of
household

Percentageoftotal

43
48
4
3
3
100.00

12
42
10
6
1
71

17
59
14
8
1
100.00

Source:Fieldsurvey,January,2007

52

The above table depicts that the tendency for migration is greater for the joint
families. Therefore, majority of the migrant members (59%) belongs to joint
family. This might be attributable to greater demand for different facilities and
amenitiesaswellaseconomicinsolvencythatprovokethejointfamily(usually
comprising greater family members) members to leave their dwellings to meet
uplargerfamilydemand.Itisinterestingtofindfromtheabovetablethatfamily
sizeisaninfluentialcatalystbehindmigration.Probablytheincreasingeconomic
demandofthelargefamilieshavecompelledtheiradultpopulation(dominantly
male population) to shift elsewhere for better life as well as to support their
families in the locality concerned. The contrast situation is observed for small
families. It is very surprising that most of the migrant households (73%) have
large family size that is composed of 514 members. In contrast, overwhelming
concentrationofnonmigranthouseholds(81%)areinthesinglefamilytypeand
their family size varies within 29 members. The percentage of migrant
householdshavingmorethan10members(23%)ismorethandoublethanthat
of nonmigrant households (10%). In addition with the higher family demand,
another responsible factor of having higher migrant members in jointlarge
familiesishavingalternativemalepersontotakecarerestofthefamilymembers
intheplaceoforigin.
4.2.2 Housingconditionofmigrantandnonmigranthouseholds
Housing condition of the migrant and nonmigrant households has been
compared on the basis of major structural condition of their living place.
HereundertheTable6presentsthecomparativepicture.
Table6:HousingConditionofMigrantandNonMigrantHouseholds

Housing

Nonmigranthousehold

Migranthousehold

No.of

Percentageof

household

total

Kutcha

82

75

26

36.62

Semipucca

24

22

34

47.89

condition

No.ofhousehold

Percentageof
total

Pucca

11

15.49

Total

109

100.00

71

100.00

Source:Fieldsurvey,January,2007

Itisinterestingtoobservethatmorethantwiceofthedwellingsofnonmigrant
families (75.23%) are temporary in nature (kutcha structures) than that for

53

migrant families. In contrast, ownership of pucca (concrete and brick built)

dwellingsisthreetimeshigheramongthemigranthouseholdsthanthatofnon
migranthouseholds.Allthesereflectthechangeinlivingstandardandaffluence
of the migrant families. Here, it should be added that the income level of the
migranthouseholdsisalsohigherthanthatofnonmigranthouseholdsanditis
presentedinthefollowingdiscussion.
4.2.3Incomeandoccupationofmigrantandnonmigranthouseholds
Here,underthisanalysisthemonthlyincomeandthemainoccupationofboth
migrantandnonmigranthouseholdshavebeenidentifiedandcompared.
Table7:IncomeandOccupationofMigrantandNonmigrantHouseholds

Income(inTk.)

10012000
20013000
30015000
500110000
>10000
Total
Occupation
types
Agriculture
(including
fisheriesand
livestock)
Service
Business
Daylabour
Rickshaw
puller
Artisan
Total

Nonmigranthousehold
No.ofhousehold
Percentageof
total
13
12
30
28
39
36
22
20
5
4
109
100.00
Nonmigranthousehold
No.ofhousehold
Percentageof
total
54
50

Migranthousehold
No.ofhousehold
Percentage
oftotal
9
13
11
15
17
24
23
32
11
15
71
100.00
Migranthousehold
No.ofhousehold
Percentageof
total
46
33

19
16
15
3

17
15
14
3

22
8
3

2
109

2
100

2
71

31
7
4
4
3
100

Source:Fieldsurvey,January,2007

Theincomelevelofthefamiliesinthestudyareas,arehigherforthehousehold
having migrated people. As observed, the share of lowincome families (Taka
1001 Taka 3000 per month) is more among the nonmigrant families (40%)
comparedtomigratedones(28%).Ontheotherhand,incaseofahigherincome

54

level (Taka 5001Taka 10000 per month), the migrated families (32%) clearly
dominateovernonmigratedones(20%).Inaddition,thefamilieswithmigrated
membershaveasmuchas3timesshare(15%)tononmigratedfamiliesamong
the highest income families (> Taka10000).All these demonstrate the affluence
of the families with migrated members and show the financial inducement for
migration.
Inthecaseofoccupationpatternofmigrantandnonmigranthouseholds,itwas
assumed that migrant would have less affiliation with agriculture but in the
study area the fact is not like that. Almost half of the both migrant and non
migrant households are dependent on agriculture. But in the service sector
concentrationofmigranthouseholdsishigher(almostdoublethanthatofnon
migrant). In the previous section it has been also mentioned that people of the
areaaremostlymigratingforthepurposeofservices.Anotherimportantfactin
thecaseofoccupationpatteristhatthedependencyondaylabouristhreetimes
higheramongthenonmigranthouseholdscomparingwithmigranthouseholds.
It denotes that migrant households are relatively less engaged in low profile
economicactivitiesandtheirincomelevelalsosupportthisfact.
4.2.4Landownershipandaccesstoservicesandfacilities
Half of the both migrant and nonmigrant households are dependent on
agriculture and the average land ownership of the migrant and nonmigrant
households is also almost same in amount, which are 2.98 and 3.15 acres
respectively. It should be mentioned that both migrant and nonmigrant
households of Gafargao upazila are significantly in a better position in land
ownership in compare with other two upazilas. Land ownership is average 5.7
acreperhouseholdinGafargao,whereas1.8acreand1.2acreinHaluaghatand
Muktagacharespectively.
Table8:AverageAmountofLandOwnershipofMigrantand

NonMigrantHouseholds

Studyarea

Assets(inacre)ofmigrant
households

Assets(inacre)nonmigrant
households

Haluaghat

1.543

2.195

Muktagacha

2.294

0.109

CharAlgi

5.12

6.43

Average

2.986

3.151

Source:Fieldsurvey,January,2007

55

4.2.5Accesstoservicesandfacilities
Here, services and facilities include only water supply and sanitation facilities,
electricity connection at home and access to credit facility and institutional
support(mostlytrainingfacilitiesandagriculturalextensionservices).
Table9:AccesstoCreditFacilityandInstitutionalSupports

Issues

Nonmigrantfamily(totalno.109)

Yes(%)

No(%)

Total(%)

Yes(%)

Migrantfamily(totalno.71)
No(%)

Total(%)

Accesstocreditfacility

56

44

100

44

56

100

Accesstoinstitutional
supports

96

100

87

13

100

Note:HereInstitutionalsupportsmeanssupportfortrainingandskilldevelopment,motivation,
agriculturalextensionservices,etc.
Source:Fieldsurvey,January,2007

Before conducting field investigation it was assumed that the migrant

households would have better access to various services and facilities due to
theirbetterincome.Butintheissueofwatersupplyandsanitationconditionand
intheelectricityconnectionnonoticeabledifferencehasbeenidentifiedbetween
migrantandnonmigranthouseholds.Ithasbeenfoundthataverage70percent
of the households of the study area have no electricity connection but the
situation is better in Muktagacha which is relatively well served upazila. In the
issue of water supply and sanitation,majority(above 80%) have access to pure
drinkingwaterandhygienictoiletfacility.Motivationalactivitiesandcampaign
programme of both the Government and NonGovernment Organisations
(NGOs) have contributed to such positive situation in water supply and
sanitation.
Since after emergence of microcredit programmes for the poor, a number of
NGOs are providing credit to the poor and middle income group. So there is
credit supply in the rural areas though it is not always sufficient and problem
free. Due to having relatively higher income and cash earning more than half
(56%) of migrant households are not taking credit from the available sources.
Interestingly, it is noticed that both in the case of migrant and nonmigrant
households, overwhelming majority have agreed that they have access to
institutional supports. But the support is delimited within the agriculture
extensionservicesmostly.

56

Intheabovesectionsithasbeentriedtoseektheansweroftwoquestionsthat
arewhothemigrantsareandwhatarethedifferencebetweenmigrantandnon
migranthouseholdsintheplaceoforigin,whichisruralBangladesh.Basedon
alltheanalysisandoutcomefollowingconclusioncanbedrawn.

5.0Conclusion
By conducting field level intensive study, this research not only examines few
principalcharacteristicsofmigrantsbutalsocomparesmigrantandnonmigrant
familiesofthreeupazilasofMymensinghDistrict.Studyrevealsthatsince1990s
outmigration from the region has been increased rapidly and similar with the
nationalaverage39percentofthefamiliesoftheareahaveaverage1.5person
migrantmembers.Thesepeoplehavemigratedprimarilyfortheirlivelihoodand
in some cases for better education. Interestingly, Dhaka is the destination of
morethanhalfofthesemigrants(54%).Comparativeanalysisbetweenmigrant
and nonmigrant families shows that dependency on the tertiary occupation
(services) is higher among the migrant families, comparing with nonmigrant
families and in the higher income level migrant households have more
concentrationthannonmigranthouseholds.Noremarkabledifferencebetween
migrant and nonmigrant households has been noticed in the aspects of the
accesstoelectricity,watersupplyandsanitation.
Asoneofthemostimportantobservationsofthisstudy,itisnoticedthatinthe
caseofincome,housingconditionandlandownershipmigrantfamiliesareina
better position. With the use of remittance the migrant families are improving
theirlivingstandard.ItissaidthatmigrationisshapingBangladeshisocietyand
remittances have been causing silent economic revolution in Bangladesh
(Siddiqui2003andAziz2001,citedinIOM2005).Thereforemigrationshouldbe
encouraged. But internal migration which is primarily rural areas to Dhaka
oriented is needed to be controlled, since it is causing unplanned urbanisation,
congestion and environmental degradation. International migration should be
enhanced by proper education and skill generation of the young generation.
Government should have adequate attention and take pragmatic action to
convertmigration(bothruraltourbanandinternational)asacatalystofnational
economicdevelopment.
This study does not give any attention to the problems, potentials and
consequencesofmigrations.Butitistruethatmigrationinduceshugeproblems
and potentials in the arena of social, economic and physical matters of living.
Therefore, it is necessary to keep continuing the diagnosis of the migration to

57

minimiseandcopewiththeinducedproblemsandontheotherhandtoaccrue
thepotentialsforenhancingdevelopmentpace.Forsuchkindofdiagnosisthis
research would be a foundation. This is expected that this research would
facilitate further research related to migration and migrant. Furthermore, it
would assist policymakers to accrue benefit from migration and to offset its
adverseimpactsintheeconomyandsociety.

References
Afsar, Rita. 2000. RuralUrbanMigrationinBangladeshCauses,ConsequencesandChallenges.
Dhaka:TheUniversityPressLimited.
BBS. 2003. Population Census 2001: National Report Provisional, Bangladesh Bureau of
Statistics.GovernmentofPeoplesRepublicofBangladesh,Dhaka.
IOM.2005.DynamicsofRemittanceUtilisationinBangladesh.IOMResearchSeriesNo.18.
InternationalOrganisationforMigration,Switzerland.
Nazem, Islam, Nurul. 1998. Changing Faces of Urban Areas in Bangladesh. In Bayes
Abdul and Muhammad Anu (eds.), Bangladesh at 25: An Analytical Discourse on
Development.Dhaka:UniversityPressLimited.
Skeldon, Ronals. 2005. Migration and Migration Policy in Asia: A Synthesis of Selected
Cases. In Tasneem Siddiqui (ed.), Migration and DevelopmentProPoorPolicy Choices.
Dhaka:UniversityPressLimited,
Siddiqui, Tasneem. 2005. Introduction. In Tasneem Siddiqui (ed.), Migration and
DevelopmentProPoorPolicyChoices.Dhaka:UniversityPressLimited.
Siddiqui,Tasneem.2005.InternationalMigrationasaLivelihoodStrategyofthePoor.In
Tasneem Siddiqui (ed.), Migration and DevelopmentProPoor Policy Choices. Dhaka:
UniversityPressLimited.

58

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

AccessibilityofWomentoProductiveResourcesin
FarmHouseholdsofKadunaState,Nigeria
S.A.Rahman*andF.A.Ajayi*

Abstract
Women in Nigeria form an active labour force contributing about 80 per cent of
agricultural labour, but rarely own the means of production. They are the key human
resources in farm production. This study examines the accessibility of women to farm
resources in farm households in two geographical regions of Kaduna State in Northern
Nigeria.Amultistagerandomsamplingtechniquewasusedintheselectionofasample
of 240 farm households. Ordinary least squares (OLS) regression model was used in
analysingthedatathatwerecollectedthroughinterviewschedule.Itwasobservedthat
women who were less involved in farm decision making had low accessibility to
productive resources compared to those who were more involved in the farm decisions.
Thevariablesincludedintheregressionmodelexplained51and58percentofvariation
in the rate of women accessibility to farm resources in northern and southern parts of
KadunaStaterespectively.

1.0Introduction
In Nigeria women are not homogenous because of the enormous variety of
socioeconomicand cultural situationsin which they may find themselves. The
specificsoftraditionalgenderbasedsubordinationwould,therefore,varyacross
the country. In spite of the heterogeneity, however, Nigeria womens
subordination, has typically limited their access to and control over such
productiveresourcesasland,labour,creditsandotherfarminputs(Rahmanand
Haruna1999).
The doctrine that women are not expected to work on farms, especially in
Muslim communities of northern Nigeria, has met with considerable empirical
challenges. According to Saito (1992), by 1990, women in northern Nigeria
accounted for 22 per cent of the farm labour, either on their own farms or as
hiredlabour.Intheotherpartsofthecountrymenmakemajorfarmdecisions,
althoughwomendecideonwhatcropstogrowontheirownfields(Okorji1991).
Women make decisions on daily management of farms and households (Saito
1995).Gabriel(1991)reportsa16hoursworkingdayforAfricanfarmingwomen
atcertaintimesoftheyear,whileWhatmore(1991)pointedoutthatnomaterthe
KeffiFacultyof Agriculture(ShabuLafia Campus),NasarawaState University,NasarawaState,
Nigeria.CorrespondingEmail:shehurahmandr@yahoo.com

59

extentofwomensagriculturalactivities,thereislittlevariationintheextentto
whichdomesticlabourissharedbyothermembersofthehouseholds.
The overall feature of Nigerian womens status is essentially that of
marginalisation, which is best explained within the context of farmhousehold
productive relation (Rahman et al. 2004). Women in Nigeria form an active
labour force, but they rarely own the means of production. They, however,
contribute about 80 per cent of the labour force in agriculture (Ingawa 1999;
Mgbada2000;Rahmanetal.2004).
Commercialisation of land has overlooked womens cultivation rights.
According to Quisumbing (1994), there has been a great disparity between
womenandmeninthesizeoflandholdings,duetopopulationpressure.There
arealsoconstraintsonwomenlabourtime,astheycannotcallonthelabourof
other household members in the way the men can (Malena 1994). Women
interestandinvolvementinfarmdecisionmakingandproductionhavealotof
implicationsfortheiraccesstoandcontroloverresources(Morna1990;Rahman
andAlamu2003).
The general aim of this paper is to examine the nature and extent of women
accessibilitytoproductiveresourcesforfarmproduction.Specifically,thepaper:

determinestherateofwomeninvolvementinfarmdecisionmaking.

assessestheaccessibilityofwomentoproductiveresources.

identifiesfactorsdeterminingtheaccessibilityofwomentoproductive
resources.

2.0Methodology
A survey was conducted on farm households in two geographical regions of
Kaduna State (northern and southern parts). The two regions were chosen to
allowforcomparisonbetweenwomeninMuslimcommunities(innorthernpart
ofKadunaState)andwomeninChristiancommunities(inthesouthernpartof
the state), especially in the aspect of womens involvement in agriculture. The
study used mainly primary data. The relevant primary data were obtained in
February and March, 2003. The data were collected using interview schedule
administered to women and household heads by 2 researchers and 4 trained
enumerators.
Multistagerandomsamplingtechniqueswereusedintheselectionofasample
of 240 farm households. Three Local Government Areas (LGAs) were first

60

randomly selected from each of the two geographical regions. Secondly, four
villages were randomly selected per L.G.A. Thirdly; there was a random
selectionofsamplehouseholdsfromtheselectedvillages.Ineachofthetwenty
four(24)selectedvillages,tenfarmhouseholdswererandomlyselectedgivinga
totalof240sampledfarmhouseholds.
The survey data include detailed module on household size, farm size, farm
inputs, farm decisions, womens socioeconomic characteristics and their
accessibilitytofarmresources.Analysisofthedatawasdoneusingdescriptive
statisticssuchasmeanandpercentage;andbyusingmultipleregressionmodels.
The ordinary least squares (OLS) regression model was used in analysing the
factorsthatdeterminethelevelofaccessibilityofwomentoproductiveresources
(land,labour,fertiliser,creditsandotherfarminputs).Thismodelisspecifiedin
formofdoublelogarithmicfunctionasfollows:
LogRA=log0+1logFD+2logVC+3logCP+4NC+5logES+6logAG+
7logIL+logU
RA=Rateofaccessibilityofwomentoproductiveresources(%)
FD=RateofInvolvementinFarmDecisions(%)
VC=ValueofContributiontofarmproductionbythewomen(N)
CP=Cooperativeparticipation(years)
NC=PresentNumberofChildrenbythewoman.
ES=Educationalstatus(years)
AG=Ageofthewoman(years)
IL=LevelofIncomeofthewoman(N)
0=Constantterm
07=Regression
U=Errorterm

3.0ResultsandDiscussion
3.1Womensinvolvementinfarmdecisionmaking
As women are the key human resources in farm production, their productivity
depends partly on the rate of their involvement in farm decisionmaking (Saito
1995). This study revealed that women in southern part of Kaduna enjoy more
decisionmaking power than women in northern part of Kaduna. Women in

61

northernpartofKadunaStatewereinvolvedinfarmdecisionmakingattherate
of 25.07 per cent while their counterpart in the southern part of the state rated
56.09percent(Table1).Consideringthetworegionstogether,therateobserved
was40.58percent.InnorthernpartofKadunaState,womenhadbelow50per
cent rate of involvement in all the identified aspects of farm decisions. In
southernpart,womenhadabove50percentinvolvementrateindecisionswith
regards to harvesting, transporting, selling, consuming, processing and storing
of produce, but less involved below (50%) for selection of enterprises, size of
enterprises, inputs procurement and allocation. It is significant to note that
decisionsandresponsibilitiestakenbywomenarerelatedtothefarmtasksthey
perform.
Table1:WomenInvolvementinFarmDecisionsinNorthernand

SouthernPartsofKadunaState
Decisionswithregardsto:
Selectionofenterprises
Sizeofenterprises
Inputsprocurement
Inputsallocation
Harvestingofproduce
Transportingofproduce
Sellingofproduce

Rateofinvolvement(%)
NorthernKaduna SouthernKaduna
14.40
49.17
12.50
46.04
11.46
44.17
13.54
47.92
20.00
50.42
18.13
52.08
28.54
64.38

Average
31.79
29.27
27.82
30.73
35.21
35.11
46.46

Consumingofproduce
Processingofproduce
Storingofproduce

48.13
48.75
35.21

62.08
74.17
70.42

55.11
61.46
52.82

Average

25.07

56.09

40.58

Note:The rate of involvement of a woman in farm decisions was computed based on Five levels
(verymuchinvolved,muchinvolved,moderatelyinvolved,lessinvolvedandnotinvolved).
The levels were scored 4,3,2,1 and 0, respectively; and considered for the past five years to
giveamaximumscoreof20.
Source:FieldSurveyData,2003

The enormous variety of socioeconomic and cultural situations in the State

might have influence the rate of womens involvement in the farm decision
making.ThenorthernpartoftheStateisaHausaMuslimdominatedzonewhile
the southern part is nonHausa Christiandominated. The socioeconomic life
styles of women in the HausaMuslim communities are strictly guided by
Islamicprincipleswhichfavourlessworksforwomenonfarms.Thisisnotthe

62

case with other nonHausa Muslim women who are freely engaged in both
agricultural and nonagricultural activities. In the nonHausa communities
womenownfarmsandplayamoreactiveroleinfarmactivities.

3.2Womenaccessibilitytoproductiveresources
Thefactthatwomengenerallydonotcontrolalotofmoneyinthehouseholds,
thus,theiraccesstoproductiveresourcesisusuallylimited.Giventhesuperior
financial status of men, they therefore control the means of production better.
This study observed very low accessibility rate of women to productive
resources. In the northern part of Kaduna, the rate was 3.42 per cent; in the
southernpartitwas19.73percent;whiletheoverallwas11.58percentasshown
inTable2.
Table2:WomenAccessibilitytoFarmProductiveResourcesinNorthernand
SouthernPartsofKadunaState
Productiveresources
Farmland
Labour
Fertiliser
Credit
Otherfarminputs
Average

NorthernKaduna
3.17
5.56
2.74
2.38
3.25
3.42

Accessibilityrate(%)
SouthernKaduna
32.31
21.15
15.38
10.58
19.23
19.73

Average
17.74
13.36
9.06
6.48
11.24
11.58

Note:Otherfarminputsincludeseeds(localorimproved)andchemicals

(herbicidesor/andinsecticides)
Source:Fieldsurveydata,2003

Women in southern part of Kaduna had more access to farm land (32.31%),
labour (21.15%) and other farm inputs (19.23%). The womens poor financial
status and poor access to credit facilities do not allow them to acquire farm
inputs.Thislimitsthescaleoftheirfarmproduction.Womenmaybewillingto
adoptnewfarmingpracticesorbuyequipmentandotheragriculturalinputs,but
they have no power to decide without the husbands approval. Moreover the
husbands control the family finances in the majority of cases, which make
situationevenmoredifficultforthewomen.

3.3Determinantsofwomensaccessibilitytofarmresources
Improvementsintherateofwomenaccessibilitytoproductiveresourcesrequire
examining the socioeconomic constraints that hinder the accessibility to the

63

resources.Thevariablesincludedintheregressionmodelexplained51,58and
63 per cent of variation in the rate of women accessibility to farm resources in
northern and southern part of Kaduna, and both regions, respectively. This
could be attributed to the traditional genderbased subordination which varies
acrossthestate.
Values of contribution to farm production by women in terms of labour and
fund relate positively and significantly to the rate of accessibility to farm
resourcesinboththenorthernandsouthernpartofKadunaState.
Table3: Factors Determining Women Accessibility to Farm Resources in
NorthernandSouthernKadunaState
Estimatedcoefficient
SouthernKaduna

Factors
NorthernKaduna
Constant
FD
VC
CP
NC
ES
AG
IL
R2

Both regions

0.512
(1.053)
0.286
(1.008)
0.116*
(2.983)
0.098
(1.122)
0.064*
(2.818)
0.080
(1.103)
0.034
(1.313)
0.161
(1.401)

0.693
(1.241)
0.432*
(4.126)
0.134*
(3.817)
0.141
(1.003)
0.043
(1.131)
0.111
(1.254)
0.017
(1.248)
0.432*
(4.243)

0.417
(1.382)
0.379
(1.121)
0.234*
(3.004)
0.105*
(2.576)
0.024
(1.500)
0.098
(1.052)
0.014
(1.066)
0.274*
(2.998)

0.51

0.58

0.63.

Note: *Significantat5percentlevel

Figuresinparenthesesaretvalues

Source: Computed from field survey data, 2003

In the southern part of Kaduna, income level was identified as another

significant factor that positively influenced the rate of involvement in the farm
decision making(Table 3). Considering the two regions of theState together,it
was observed from the regression that contribution to farm production, co
operativeparticipationandincomewereidentifiedtobesignificantlyrelatedto

64

therateofwomenaccessibilitytoproductiveresourcesinKadunaState(Table3).
Participationofthewomenincooperativescouldbeasourceofimprovementin
theirsocioeconomicstatus.

4.0 Conclusion and Recommendations

Despite the significant contribution of women to agricultural production, they
have not been given the opportunity to realise their full potential in terms of
theirlabourproductivity.Thisisasaresultoftheirlimitedaccesstoandcontrol
oversuchproductiveresourcesasland,labour,creditsandfertilisers.Thisstudy
hasrevealedthatinKadunaState,womenwerelessinvolvedinfarmdecision
making, and they had low accessibility to farm resources. Income and co
operative participation of women were identified as the significant factors that
positivelyinfluencedaccessibilityofwomentofarmresources.
Thisstudy,therefore,recommendsthefollowings:

Women oriented policies and programmes should be introduced to

improve womens access to productive resources such as land, credit
and appropriate technologies so as to enhance food production,
distribution and consumption. Such policies should be the types that
will encourage women to struggle collectively, especially through
formationofcooperatives.

Developing and reinforcing research and information gathering

exercises on village communitylevel sociocultural financial conditions
that affect the rural women. This will enable improvement in the
traditionalgenderbasedsubordinationinvariouscommunitiesforgood
welfareofwomen.

65

References
Gabriel,T.1991.TheHumanFactorinRuralDevelopment.BelhavenPress:London.
Ingawa, S. A. 1991. Welcome Address at the National Workshop for Women in
Agriculture, held in FACU Headquarters, Sheda, Abuja, Nigeria. 31st August2nd
September.
Malena, C. 1994. Gender issues in integrated pest management in African agriculture. NRI
SocioEconomicSeries5.NaturalResourcesInstitute.Chatham,U.K:
Mgbada, J. U. 2000. Production of staple crops by rural women in Enugu and Ebonyi
States:lessonsforenhancingpovertyalleviationprogrammes.InOlowu,T.A.(ed.).
AgriculturalextensionandpovertyalleviationinNigeria Proceeding of the 6th Annual
NationalExtensionSocietyofNigeria.Pp.1012.
Morna,C.I.1990.Womenfarmersemergefromtheshadows.Africanfarmers,No.3April.
Okorji,E.C.1991.Acomparativestudyoftheroleofwomenintraditionalandmodern
organizations in Nigeria. In Ijere. M.O. (ed.) Women in Nigerian Economy. Acena
publishers.Enugu.
Quisumbing, A. 1994. Gender Differences in Agricultural productivity: a survey of
empirical evidence. Discussion paper series No. 36, Education and social Policy
Department,WorldBank,Washington,D.C.,U.S.A.
Rahman,S.A.andI.M.Haruna.1999.Determinantsofwomenseconomiccontribution
to the farm sector in Nasarawa state, Nigeria. A paper presented at the National
Workshop of Society for International Development, held at Institute of
Administration,AhmaduBelloUniversity,Zaria,Nigeria.3rdto4thNovember.
Rahman, S. A. and J. F. Alamu. 2003. Estimating the level of women interest in
agriculture:TheapplicationofLogitRegressionmodel.NigerianJournalofScientific
research,4(1):4549.Vol.4,No.1,Pp.4549.
Rahman, S. A., J. Gabriel, and N. D. Marcus. 2004. Gender differentials in labour
contributionandproductivityinfarmproduction:EmpiricalevidencefromKaduna
stateofNigeria.PaperpresentedattheNationalConferenceonTheFamily,heldat
NewTheatrecomplex,BenueStateUniversity,Makurdi,Nigeria.1st5thMarch.
Saito, K. A. 1992. Raising the productivity of women farmers in subsaharan Africa:
Overview Report vol. 1. Population and Human Resources Development
Department,WorldBank,WashingtonDC.,USA.
Saito, K. A. 1995. Raising the productivity of women farmers in subsaharan Africa;
Agricultural and Environmental challenges. In Srivastava, J.P. and Alderman,
H. (Eds.) Proceedings of the 13th Agricultural sector symposium. The World Bank.
Pp.147152.
Whatmore, S. 1991. Women in agriculture. Journal of Rural Studies, Vol. 7, No. 1 & 2
(specialissue).

66

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

EmpiricalAnalysisonRuralHouseholdsBorrowing
Behaviour:TheCaseofCentralJava
TakashinoNina*

Abstract
The aim of this paper is to empirically examine determinants of households borrowing
demand and how borrowers choose the lending source by using the data set on rural
villagers financial transactions in Central Java. The main findings are summarised as
follows:(1)villagersintheresearchareahaveaccesstotheformalfinancialinstitutions;
(2) the observed loan amount is determined by investment opportunities and
consumption needs, such as housing expenditure, and the level of asset holdings; (3)
borrowers chose the lender depending on both material interest rate and psychological
cost or trust relationship; (4) the poor tend to borrow money from traditional selfhelp
saving group (arisan in Javanese) more frequently than from formal institutions or
friendsandrelatives.

1.0Introduction
Access to credit is essential to improve the living standard of rural villagers in
lowincomecountri.1However,formalfinancialinstitutions(banks,cooperatives
andsoon)cannotalwaysachievehighrepaymentratesuccessfully,andhence,
the policymakers need to explore a better design for such institutions (Zeller
and Meyer 2002). Knowledge about rural households borrowing, such as how
much financial needs they face and from what sources they borrow money,
would provide much helpful insight to designing an effective rural financial
institution.
Theaimofthispaperistoempiricallyexaminethedeterminantsofhouseholds
borrowingdemandandhowborrowerschoosethelendingsourcebyusingthe
data set on rural villagers financial transactions in Central Java. Same as the
villagers in other lowincome countries, rural households in Indonesia can
borrow money from two kinds of lender, formal and informal. Financial

Center for Experimental Research in Social Sciences, Hokkaido University, Japan. E-mail:
ntakashino@lynx.let.hokudai.ac.jp
1 Credit is a means to enable investment by solving a liquidity problem. The liquidity problem
arises from the fact that outlays triggered by the investment precede (expected) future returns
(Petrick2005).

institutions under governments control, such as banks and cooperatives, are

called formal lenders. Their transactions involve a written contract, filling
documents and proposal of collateral. On the other hand, traditional selfhelp
savinggroup(arisaninJavanese),moneylenders,andvillagerswhoprovideloan
tofriendsandrelativesarecalledinformallenders.Theirtransactionsareoutof
the governmental regulation and usually based on an oral contract without
takingcollateral.
Existingliteraturesproposetwodifferentviewsontheroleofeachsectorinlow
incomecountries.Whilesomestudiestreatinformalsectorsassecondarysources
for the borrower who has been rejected his/her loan offer from formal sector
(Bell, Srinivasan and Udry 1997; Hoff and Stiglitz 1990), others see them as
possible favourable sector because transaction cost of informal sector can be
cheaperthanformal(Kochar1997).ThisstudywillshowthatJavanesevillagers
chose a lender depending on the situation, and hence, the latter view is
applicableforthecaseoftheresearchsite.
SomestudiesdescribeborrowingbehaviourobservedinruralareasofIndonesia
(Hamada2006;Mizuno1999:199226)2,butempiricaldataonborrowingdemand
andsectorchoiceofruralhouseholdislimited.Anotherimportantcontribution
of this paper is to fill this gap and to show under what conditions villagers
wouldchooseborrowingfrominformalsectors.
The mainfindings are summarisedasfollows:(1) villagersin the researcharea
haveaccesstotheformalfinancialinstitutions;(2)theobservedloanamountis
determined by investment opportunities and consumption needs, such as
housing expenditure, and the level of asset holdings; (3) borrowers chose the
lender depending on both material interest rate and psychological cost or
relationship of trust; (4) the poor tends to borrow money from arisan more
frequentlythanfromformalinstitutionsorfriendsandrelatives.
The organisation of this paper is as follows: After describing an overview of
sample hamlets and socioeconomic conditions of sample households in the
second section, the data on borrowing behaviour and asset management are
summarisedin the thirdsection. Analytical frameworks on the empirical study
2

Hamada (2006) points out rural villagers are familiar with informal lenders and more closely
relatedwiththemthanwithformalinstitutions.Mizuno(1999:pp.199226)comparestheroleof
formalandinformalfinanceontheprovidingloanstosmalltextileenterprisesinruralWestJava.
Mizunofoundthatthetransactioncostofformalfinanceishighbecauseitrequiresborrowersto
submitcollateralandhenceinformalfinanceabsorbsborrowingdemands.

68

andestimationresultsarediscussedinthefourthsection.Finally,summaryand
conclusionsaregiveninthefifthsection.

2.0OverviewofSampleHamlets
2.1Geographicalconditions
To investigate households borrowing behaviour, this paper employs a dataset
compiledfromasamplehouseholdsurveyconductedfrom2001to2005bythe
researchersofJSPSDGHECoreUniversityProgramme.About120householdsin
fourhamletsintheruralareaofYogyakartaprovinceweresurveyedassamples3
(Table1).HamletAandBarelocatedonahillsideawayfromYogjakartacity.
Ontheotherhand,hamletCandDareclosetoYogyakartacity.(Distancetothe
cityis40km,32,4kmfromhamletA,Band10kmfrombothCandD.)

Table1:BasicDataofSampleHamlets

Hamlets
Areacondition
Distancefromcitykm

622
115

B
hilly
32.4
traditional
/rainfed
641
130

5.41
129.5

Compound
Others
Averagesizeoffarmingland(ha)

Irrigationsystem
Numberofpopulation
Numberofhousehold
Averagenumberofhousehold
member
Totalareaofland(ha)
Paddyfield
Dryfield

A
hilly
40
rainfed

C
lowland
10

D
lowland
10

traditional technical
604
163

430
97

4.93
191.5
22.1

3.71
44.5
23.0

4.43
41.3
31.0

127.0
2.5

144.2
25.2

17.1
4.4

6.8
3.5

1.10

1.28

0.14

0.32

Source:SubejoandIwamoto,Noriaki,2003

Geographical differences between hilly area and low land are affected
householdseconomymainlyintwoaspects.Oneiswatersupplyandanotheris
opportunitiestogetofffarmjob.DuringthedryseasonpeopleinhamletAand
B suffer from a water shortage and a decrease in income because almost all of
themarefarmers.WaterconditionofhamletBisslightlybetterthanA.HamletC
3

Details of the survey are given in Iwamoto, Hartono and Fukui. (2003) and Subejo and
Iwamoto(2003).

69

is semiirrigated while D is fully irrigated. Some farmers in Hamlet D raise

catfishandityieldshighprofitwithhighrisk.SincehamletCandDarelocated
neartothecity,villagershavemoreopportunitiestobeengagedinofffarmjob.

2.2Householdsriskandinvestmentneeds
Table 2 shows the characteristics of family members of the sample households.
Although sample households family structure is different, age of household
head is about 50, and each household consists of four to five members on
average.SeeingfromhigherlevelofschoolinginhamletCandDthanthoseinA
andB,financialneedsforinvestmentoneducationmightbemoreinthehamlets
inthelowlandarea(CandD).

Table2:CharacteristicsofSampleHouseholds

A
B
C
Householdcharacteristics

Ageofhouseholdhead
48.2
49.9
59.6
Familymember
4.6
4.3
4.6
Numberoflabourers(man/hh)
3.1
3.0
3.2

Ratioofgraduates1)
Elementaryschool(%)
64.2
58.3
52.9
Juniorhighschool(%)
32.5
34.3
19.6
Highschool(%)
3.4
5.0
22.5
University(%)
0.0
2.5
5.0
Householdincome(Rp/hh)
5,447,712
4,873,281
8,847,360
Agriculture(Rp/hh)
3,400,448
3,123,402
1,897,347
Offfarmjob(Rp/hh)
1,513,794
1,347,316
6,394,178
9,216
500
115,594
Landrent(Rp/hh)
Remittance(Rp/hh)
524,255
402,063
440,241
RatioofAgric.income(%)
62.4
64.1
21.4
Numberofsamplehouseholds
85
80
87
(households/3years)
Note:1)Graduateratioperfamilylabourers
2)ExchangerateatAug.2003was1USdollar=8,500rupiahs
Source:JSPSDGHECoreUniversityProgramme20012003

54.6
4.0
2.6

46.6
11.8
32.4
9.3
15,247,219
7,828,038
6,351,973
136,649
930,558
51.3
77

Level of annual household average income from 2001 to 2003 is around 8.5
million rupiah (around 820 dollars) and the sources of household income are
varied.Therefore,financialneedsforconsumptionsmoothingorinvestmentare
alsodifferentdependingoneachhouseholdscharacteristics.Villagersearntheir
incomefromagriculture,offfarmjob,landrent,andremittance.

70

Villagers face different level of agricultural production risks. As mentioned

above, farms in the hilly area (hamlet A and B) are less irrigated, and villagers
rely on rainfed condition for agricultural production. They produce rice only
once a year in rainy season and crops like cassava and corns in other seasons.
However, the yield potential on average is lower than lowland area (hamlet C
and D) and villagers often suffer from unexpected bad yields because of water
shortage.Therefore,farmersintheareafacemuchincomefluctuationandrisk.
On the other hand, farmers in lowland area can well utilise irrigate water and
achievecomparativelyhighyieldsofpaddy.Eventhoughtheycanmitigaterisk
by using irrigation, many sharecropping contracts are still observed there.
Reasonmightbethefarmerswanttoshareriskscausedbycropdiseasesandrats
attack (the damage is specifically observed in lowland area). Another possible
reason is that most of the sharecropping contracts are based on personal
relationshipamongfriendsandrelatives(Fukui,Hartono,andIwamoto2002).
Manyhouseholdsraiselivestocksuchascows,goats,andchickensinallsample
hamlets.ParticularlyinhamletD,somehouseholdsraisecatfishorotherfishes
on farms. Although fish farming requires high initial cost for constructing
fishpond and operational cost for buying young fish and its feed, the return of
fish farming is very high. Since fish meal, the feed of fish farming, is also
favouritefoodofrats,cropdamagecausedbyratsisespeciallysevereinhamlet
D(Iwamoto,HartonoandFukui2003).
Earnings from offfarm employment provide villagers stable income and
mitigateincomefluctuation.Somevillagersareemployedingovernmentsector,
private companies and factories as permanent worker. They earn much and
stable income throughout the year with less income fluctuations. Such
opportunities are found more in the hamlet C and D (lowland area) where
villagerscancommutetothecentreofthecity.Thiscircumstancemayhavean
impactondemandforinvestmentonschoolingbecausehighschoolenrolmentis
usually required to get permanent job. Hence, villagers desire for getting
permanent job might increase their financial needs. Also, if villagers have
opportunities to get job as parttime workers, they can deal with any income
dropbyearningincomeexpost.
Another comparatively stable income source is land rent. But only small
numbersoflandownersareincludedinsamplehouseholds.Whilealmostallthe
farmers in hilly area cultivate their own land, a typical farmer in lowland area
(hamletCandD)employstenantcontract.Therefore,somevillagersinhamletC
andDearnincomefromlandrent(Table2and3).

71

Remittance takes two forms: One is permanent annual (or monthly) transfer at
constantamountandanotheriscasuallysendingtohelpsharpdropinincomeor
unexpectedconsumptionneeds.Table2showsthatsomehouseholdswouldget
remittanceanditmighthavedecreasedborrowingdemand.

Table3:LandOwnershipandTenancy

5,056

2,539

410

1,433

1,945

1,423

5,056

2,539

2,355

2,857

Owned(squaremeter)

16,696

3,162

174

268

Leasedin(squaremeter)

1,023

375

Total(squaremeter)

17,719

3,310

174

268

2,590

1,387

523

562

31

30

30

28

Paddy
Owned(squaremeter)
Leasedin(squaremeter)
Total(squaremeter)
Othercrops

Homegarden(squaremeter)
Sample(household)

Source:JSPSDGHECoreUniversityProgramme2003

3.0HouseholdsBorrowingBehaviour
3.1Overviewoncreditmarket
Table4describesthehistoryofsamplehouseholdsborrowingfor3yearsfrom
September2000toAugust2003.Householdscanborrowmoneyfromtwokinds
of lenders, namely formal and informal ones. Financial institutions under
governmentscontrol,suchasbanksandcooperatives,arecalledformallenders.
Their transactions involve written contract, filling documents and requiring
proposal of collateral. Onthe other hand, traditional selfhelp savinggroup (or
arisan in Java), moneylenders, and villagers who provide loan to friends and
relativesarecalledinformallenders.Theirtransactionsareoutofgovernmental
regulationsandusuallybasedonoralcontractwithouttakingcollateral.
Sorting all borrowings by the lenders type, Table 5 clarifies different
characteristicsintheircontracts.Villagersborrowedmoneyfromvariousformal
lenders. Accessible banks are BRIUD (Bank Rakyat Indonesia Unit Desa), Bank
Kredit Desa (BKD), BUKOPIN, BMT, and other BPRs (Bank Perkreditan Rakyat).

72

SomevillagersmayalsohaveaccesstofarmerscooperativesorKUD(Koperasi
Unit Desa) or employees cooperatives. Monthly interest rate of formal
borrowing is cheap (around 2%), but they require collateral in order to select
repayable borrowers and overcome information scarcity. Hence, almost all the
villagers can borrow money from the formal lender only if they own enough
assets.

Table4:NumberofCreditTransactions(Unit:No.ofhouseholds)

2001
a.Formal
b.Informal
c.Both
d.No
Total(a+bc+d)
2002
a.Formal
b.Informal
c.Both
d.No
Total(a+bc+d)
2003
a.Formal
b.Informal
c.Both
d.No
Total(a+bc+d)

5
2
0
26
33

5
3
0
25
33

5
5
1
24
35

1
9
0
20
30

0
6
0
24
30

1
16
1
14
32

2
8
1
24
35

3
7
1
24
35

4
18
2
13
37

Total

8
6
3
22
34

8
11
2
16
37

8
8
2
19
37

16
25
4
92
137

16
27
3
89
135

18
47
6
70
141

Source:JSPSDGHECoreUniversityProgramme20012003

Assets used as collateral include certificate of land, motorbike, bicycle, sewing

machine, certificate of income, personal recommendation, car, TV, and so on.
Addingtothesetraditionaltypesofloans,manyformalinstitutionsprovideloan
programme for the poor and small enterprises. Takesra Kukesra, Tabungan
KeluargaSejahtera(Takesra)danKreditUsahaKeluargaSejahtera(Kukesra)aresome
of the programmes nationally conducted for the poverty alleviation between
1994and2001.
When villagers borrow money from friends and relatives, they do not need to
prepareanydocumentsorcollateralbecausetheirpersonaltrustsenablethemto
have a contract orally. Besides, lenders usually allow the borrowers to

73

rescheduletheirloanswhentheycannotpay.Becausetheamountofloancannot
get over the lenders personal asset holdings, it is typically smaller than banks.
Comparingloansfromfriendsandthosefromrelatives,villagerstendtoborrow
moremoneyfromrelativesthanfriendsbasedonstrongersocialtieortrust.The
lenders do not require any interest payments to friends and relatives, and this
kind of loan transaction is common in developing countries in Southeast Asia
(Fafchamps 1999). Most of the lenders live in the same hamlet with the
counterpart borrower, but particularly for the case of lending between non
relatives,someoflendersliveoutsidethehamlets.
Table5:CreditTransactionsSortedbyLenderType

No.

Averageamount

Interestrate

Term

Collateral

(Rp.)

(%/month)

(month)

(%)

Formal

48

3,000,000

2.33

14.2

90

23

3,526,087

2.07

18.0

91.3

BPD

6,000,000

2.13

36.0

100.0

BUKP

1,600,000

1.89

14.5

50.0

BMT

4,171,429

2.95

10.9

85.7

BUKOPIN
BPR

475,000

2.79

5.3

100.0

2,250,000

2.25

5.0

100.0

Bank
BRI

4,333,333

2.33

16.0

100.0

Cooperative
KUD

Other

13
3

1,319,231
633,333

1.85
2.50

12.5
10.0

23.1
33.3

Other
Programme

10

1,525,000

1.65

13.2

20.0

121,429

2.17

12.3

Takesra

112,500

2.30

9.5

0.0

Other

133,333

1.99

16.0

0.0

Informal
Relatives

22

1,730,000

0.00

4.5

Friends

33

569,697

0.50

6.1

99
222

109,414
1,034,649

3.92
2.50

3.0
23.4

Arisan
Total

Note:*Informalloansareoftenrescheduled
Source:JSPSDGHECoreUniversityProgramme20012003

Selfhelp saving group, called arisan in Java, is also another source of loans.
Group members (from 5 to 30 people) have regular meeting, and they enjoy
lotterywithsomefees.Atthesametime,theycollectsavingsfromthemembers

74

andpooltheminordertoprovideloantoanymemberfacingfinancialneeds.In
thesamemanner,withtheloansbetweenfriendsandrelatives,thetransactions
are based on mutual trust and oral promising requiring no collateral or fixed
date of repayment. But it requires interest payment (around 4% monthly) at
every meeting. Amount of loan on an average is smaller than those of other
typesmentionedabovebecausetheamountofcollectedsavingisverysmall.
Although some of the villagers have a little information regarding informal
moneylendersaroundsamplehamlets,nosamplehouseholdhaseverborrowed
moneyfromthem.Only31householdsoutof118haveanyinformationonthe
contract form of money lenders (such as interest rate or loan term) when
interviewed in 2004 survey. According to information, although levels of
interest rate were different for each money lender, they were very expensive
ranging from 30 to 300 per cent. The loan term was short and some of them
visited the borrower daily or weekly. Some villagers pointed out that traders
tendedtolendmoneyasthesecondjob.

3.2Motivationsofborrowing
Villagers borrow money for consumptions or investment (Table 6). Borrowing
for consumptions smoothing are classified into the ex ante strategy for big
consumption needs and expost dealing with drop in income or unexpected
expenditure. Sample households tend to borrow money for the expost
consumption smoothing, such asdailyconsumption, medical expenditures and
soonratherthanfortheexanteborrowingforbuyingdurablegoods.
Whenvillagersborrowmoneyforconsumption,theyseemtochooseborrowings
from informal lenders, because informal lenders care usually borrowers
immediate needs whereas formal lenders evaluate productivity of borrowers
project. Therefore, many villagers ask for loans to the formal lenders when the
borrowers want to use resources for investment purposes. On the other hand,
when loan is used for medical expenditures or buying durable goods and its
amountisbig,loansfromfriendsandrelativesareactivelychosen.Ifloanisused
forunprofitablepurposesandtheneedissmall,theytendtoborrowfromarisan.
Whenloanpurposehascharacteristicsbothasinvestmentandconsumption,like
loan for education and housing, formal and informal loans are used. Table 6
showsthatsomevillagersborrowmoneyfrominformalsourceeventhoughthey
useitforinvestmentanditsuggesttwopossiblesituations,oneislowereffective
cost of informal loans and the second one is constrained access to cheap
formalloans.

75

Table6:CreditTransactionsSortedbyPurpose

No.oftransactions
Informal

Friend Arisan

Formal
Coop.

Average

amount
Total

Bank Programme

ofloan

Relative

(Rp.)

Consumption

Daily

29

46

154,935

Medical

12

2,470,000

Social

10

59,500

Durable

1,208,889

Debt

2,950,000

Help

95,000

Education

19

42

568,571

Housing

15

2,385,000

Agriculture

17

30

236,333

Fishfarm

10

1,780,000

Trading

16

945,313

Other

11

23

2,186,739

Investment

Other

4,248,333

Total

22

33

99

14

47

222

1,034,649

Source:JSPSDGHECoreUniversityProgramme20012003

3.3Savingandasset
Withdrawing asset holdings is alternative device to deal with financial needs.
Households having less own asset rely on borrowing facing a financial needs.
Thevillagerscanaccumulatetheirwealthinvariousforms,suchasland,durable
goodslikeTV,motorbikesorsewingmachines,financialasset(goldordeposit),
livestock, trees, etc. There is a slight difference between the two study areas
(table7).Livestockandtreesplayimportantroleasassetinhillyarea,whereas,
manyvillagersownmotorbikesandoftenusethemascollateralforbankloan.
Somevillagersmaypreferborrowingtowithdrawing,whileothersmaynot.The
amountofloanassuchaspurposeofneedsmightdeterminethechoice.Figure1
illustrates the households answers to the question how do you deal with
variousamountsoffinancialneeds.Villagerschoosethemostfavourablewayto
deal with each range of the financial needs from the alternatives including

76

borrowing from relatives, nonrelative friends, arisan, formal lender, and using
savingorasset4.

Table7:AssetHoldings(Unit:Rupiah)

Agriculturaltools
Livestock
Cow
Goat
Chicken
Other
Trees
Durablegoods
TV
Bicycle
Motorbike
Car
Financialasset
Gold
Deposit
Other
Total
Samplesize

A
267,815

4,142,458
600,597
87,024
0
3,620,000

403,241
3,190
655,172
0

376,690
476,207

B
334,377

4,984,434
380,645
90,485
2,708
3,460,000

553,333
12,667
1,914,167
0

344,367
425,633

C
134,100

1,605,000
176,667
137,167
20,733
674,167

428,167
270,370
4,769,167
0

654,583
317,600

D
205,393

96,429
189,286
127,429
21,196
836,897

528,213
285,594
3,941,829
1,551,724

286,552
127,241

17,241

80,000

50,000

523,703

10,649,636
31

12,582,816
30

9,237,720
30

8,721,486
29

Source:JSPSDGHECoreUniversityProgramme2003

The figure shows when facing small amount of needs (less than 0.5 million
rupiah),manyvillagerswanttoandareabletosatisfytheneedsusingtheirown
saving where 40 households out of 120 choose to use saving. Borrowing from
friends and relatives or arisan is also chosen for small needs reflecting its easy
borrowing. On the other hand, they do not choose selling asset or borrowing
fromformalloanbecauseofitsexpensivetransactioncost.
For the relatively large amount of need (0.5 to 1 million rupiahs) many
households are unable to meet by using saving, hence choose borrowing from
relativesorformallenderandalsosellingfixedassetsasgoat,trees,motorbike,
andTV.

The definition of using saving here is to sell liquidity asset including deposit, gold, small
livestock like chickens whereas using asset means to sell fixed asset, such as trees, motorbike,
largelivestocklikecow.

77

Figure1:SectorChoiceStrategies

Fam ily

Friend

Form al

Arisan

Saving

Asset

60

N o . o f A n sw e rs

50

40

30

20

10

0
100,000

1,000,000

10,000,000

100,000,000

A m ount of N eeds

If the villagers need to prepare money over 1 million rupiahs, formal loan and
sellingassetsplayimportantroletomeetuptheneeds.Furthermore,thefigure
suggests choosing more asset selling than formal loan because of higher
transaction cost (including risks to loose collateral if default). Villagers tend to
avoidsellinglandorhouse,andonlytwohouseholdschooseit.Thefactsimply
theirdesiretoretaintheirlandandthattheydouselandascollateralonlyifthey
aresuretorepaytheloan.

78


Table8:CopingStrategiesforBigFinancialNeeds

Copingstrategy(%)

Amount

Financialneeds
No.

Asset

Borrowing

Wedding

6,525,000

100

25

Disease

6,338,846

13

85

62

Funeral

1,670,000

100

100

Education

3,133,889

18

89

28

Housing

16,555,556

89

33

Note:TotalpercentagesofAssetandBorrowingareover100becausesomehouseholdsuse
bothofthem
Source:JSPSDGHECoreUniversityProgramme20012003

The discussion above is based on the answer to virtual needs, Table 8 depicts
how the villagers coped with actual needs for big economic shocks happening
from 2001 to 2003. For example, they face big financial needs for ritual events,
diseases,andinvestmentonschoolingorhousing.Theycanraisefundthrough
deposit,livestocksaleforaforeseeableeventlikeweddingorinvestment,andit
coversmostpartoftheexpenditureatlowrateofborrowinginsuchoccasions.
However, because many of the villagers cannot cover unexpected event like
diseasesorfuneralsbytheirownassets,theyneedtoborrowmoney.

4.0EmpiricalAnalysisonChoiceofBorrowing
4.1Analyticalframework
To investigate the households borrowing behaviour, the author empirically
examined the determinants of household borrowing choice assuming that each
household chooses a lending source after assessing their loan demand (which
depends on economic situation)for the time. Denoting the loandemandas D*,
andtheamountofloanthehouseholdactuallyborrowedasD,Tobitestimation
methodisappliedtoestimateloandemandasfollows:
(1) D*=1X1+u1
D=D*ifu1>1X1
=0otherwise
X1 is a vector of explanatory variables. It includes variables related to
householdsfinancialneeds,suchasageandsexofthehouseholdhead,amount
of asset holding, event dummy, and location dummy (Table 9 summarises the

79

definitions of various explanatory variables). 1 is a coefficient vector which

reflects the effects of these variables. u1 is an error term. Although the panel
dataset is available for the estimation, the dataset is treated as pooled data
becausethenumberofpositiveamountofborrowingistoosmalltomakepanel
estimation.

Table9:IndependentVariablesforLoanDemandEstimation

Definition
Householdhead
Age
Ageofthehouseholdhead
Sexofthehouseholdhead:
Sex
Male=0,Female=1
Land
Ownedlandarea(1,000m2)
Asset
Assetholdings(millionrupias)
Labourandeducation
Labour
No.oflabour(1570yearsold)
Highschool Ratioofhighschoolgraduated
University
Ratioofuniv.graduated
=1ifconstructedhouse
House
Illness
=1ifsufferedfromillness
Student
No.ofstudents
Selfreportevaluationon
Distrust
neighbourstakeadvantage
Location

HamletA
=1forhamletA
HamletB
=1forhamletB
HamletC
=1forhamletC
Year

2001
=1fortheyearof2001
2002
=1foryear2002

Ave.

S.D.

53.62

12.83

0.06

0.25

23,621.49 117,930.37
9.70
8.42

3.05
1.27
0.10
0.18
0.03
0.10
0.03
0.16
0.07
0.30
0.48
0.73

Min.

19.0

Max.
80.0

0.0

1.0

0.0
0.1

0.0
0.0
0.0
0.0
0.0
0.0

1,577,562.0
42.4

7.0
0.8
0.6
1.0
3.0
4.0

2.31

1.49

1.0

5.0

0.26
0.25
0.27

0.28
0.37

0.44
0.43
0.44

0.45
0.48

0.0
0.0
0.0

0.0
0.0

1.0
1.0
1.0

1.0
1.0

Source:JSPSDGHECoreUniversityProgramme20012003

Explanatoryvariablesareselectedonthebasisofdiscussioninprevioussections.
The sign of coefficient on asset holding is not predictable because it may have
both positive and negative impacts on borrowing demand. It may stimulate
borrowingdemandtoinvest,andatthesametime,enablevillagerstodealwith
the consumption needs decreasing demand of borrowing. The impact of event
dummies are clearly predicted as positive because the villagers face much
financial needs for ritual events such as wedding or funeral ceremony. The
coefficients of location dummies would reflect differences in productivities or

80

investment demand among four hamlets. Taking hamlet D as a reference,

coefficientonthedummyforhamletA,BandC(withlowerproductivitythan
hamletD)areexpectedtobenegative.
It is assumed that observed borrowing amount reflects the equilibrium of loan
demand and supply. If the borrowers face any constraint and cannot borrow
money, the estimated coefficient might be unreliable. In such a case, both
demand and supply functions needed to be estimated at the same time 5 .
However, no sample household have been rejected loan offer from a bank and
only 4 out of 115 households suppose that a bank will reject their loan offer
(Table 10)6. Therefore, it is plausible to estimate borrowing amount assuming
thatobservedborrowingreflectstheequilibriumofloandemandandsupplyin
thestudylocations.

Table10:PossibilityofBorrowingMoneyfromBank

Beenrejectedloanoffer
Wouldberejectedif
offered

ReasonwhyNo

Yes

No

Neither

115

110

Nocollateral

Lowprofitability

Source:JSPSDGHECoreUniversityProgramme2004

Usingtheresultsofthefirstestimationonborrowingdemand,itisasexamined
how the borrowers choose the lending source. Suppose the borrowers face
unorderedthreechoicesofborrowing:(1)borrowingfromfriendsandrelatives,
(2)arisan,and(3)formallender.Andtheprobabilitiesofborrowingfromeachof
theabovelendersaredenotedasP1,P2andP3respectively.LetX2bethevector
of the characteristics variables which determines the borrowers choice. Then
multinomial logit estimation method is employed to estimate Pj (j = 1, 2, 3) as
follows(Maddala1983:35):

ForthecaseofIndia,itisassumedthatnotallborrowerscanborrowasmuch moneyas
theywant.Therefore,existingstudyestimatesbothdemandandsupplyfunctionsatthesametime
(Bell,Srinivasan,andUdry1997,andKochar1997).
6
Many households in other developing economies have been rejected loan offer from a
bank. For example, Duong and Izumida (2002) reports high ratio of rejection for the case of rural
householdsinVietnam.
5

81

(2)P1=exp(21X2)/[1+exp(21X2)+exp(22X2)]
P2=exp(22X2)/[1+exp(21X2)+exp(22X2)]
P3=1/[1+exp(21X2)+exp(22X2)]
Householdswithoutanyborrowingexperiencewerediscardedfromthesample.
DetailsoftheexplanatoryvariablesareshowninTable9.EstimatedvalueofD*
orborrowingdemandofthehouseholdisalsousedasanexplanatoryvariable.
Because the coefficients represent the favourability of borrowing from each
categorycomparedtoborrowingfromformalloan,itwouldchangedepending
onthecostofborrowing.

Table11:DegreeofDistrusttoNeighbours

SelfreportagreementtoNeighbourstakeadvantageofyou.

No.of
answers
Share(%)

(5)Strongly
agree

(4 Agree

30

99

8.4

27.7

0.8

(1)Strongly
disagree

Total

54

171

357

15.1

47.9

100.0

(3)Neither (2)Disagree

Source:JSPSDGHECoreUniversityProgramme20012003

The cost of borrowing from friends and relatives would involve not only
material interest payment but also psychological factor. As shown in the
previoussection,villagersusuallydonotrequireinterestpaymentfortheloans
to friends and relativesin the researchfields. If thechoice of villagers depends
only on material cost, all borrowers will borrow money from friends and
relatives. However, the collected data show that some of the villagers prefer
borrowing from friends and relatives while others do not. It implies that an
alternativedeterminantofborrowingfromfriendsandrelativeisapsychological
factor,suchastrustrelationshipbetweenthem.Togettheindexofthedegreeof
distrust, villagers rated their agreeableness on a fivepoint scale (1 = strongly
disagreeto,5=Stronglyagree)tothestatementNeighbourstakeadvantageof
me(Table11).
Expectedsignofthecoefficientforthisvariableontheprobabilityofborrowing
from friends and relatives would be negative because those who distrust
neighbourswouldnotpreferborrowingfromthem.Addingtothiseffect,social
norm whichencourages villagers to help widows would be other determinants
(expectedtohavepositiveimpactforthehouseholdwithfemalehouseholdhead).

82

For the case of arisan loan, the household without much asset might borrow
moneyeasierthantherichvillagersbecausethisloanisoriginallysetuptohelp
thepoor.Fromtheinterview,villagerstoldthatwhentherearemanyapplicants
forloan,theywouldgiveloantoamemberwhohasurgentneeds.Thisimpliesa
negativeimpactoftheassetholdingsonthepossibilityofborrowingfromarisan.
Loan size would be another determinant. When villagers borrow money from
theformallendingsourcesuchasbank,theyhavetopaytheinitialcost.Itmeans
thattheaveragecostofborrowingbecomeshigheriftheyborrowsmallamount
of loan. Therefore, given the average cost of another choice to be constant,
borrowing from formal lender would be more favourable when the borrowing
amountisincreased.

4.2Estimationonborrowingamount
The result of estimation for the borrowing amount (1) shows the structure of
financial demand and supply in the research area (Table 12). The positive
coefficient on age squared means that the elder villagers face more monetary
needs. The result shows no significant difference between households with
femaleheadandothersonborrowingdemand.
Anothersignificantfactorshownintheresultisthelevelofassetholdings.From
the view point of the demand, the positive coefficient suggests that the more
assetsthevillagershave,themoretheirchancestoinvest.Fromthepointofloan
supply,itmightreflectmoreavailabilityofgettingloans.Asanalternativefactor
affecting investment opportunities, the number of student is positively
associated with the amount of borrowing, reflecting the villagers high
expectationonthereturntoschooling.Ontheotherhand,landholdinghasno
significant impact on their borrowing amount. This reflects low return to
additional agricultural investment such as chemical input. Although demand
factors on consumption such as housing and illness have positive effect on
borrowing, these are not statistically significant. Possible reason might be that
loanisnotthesolechoiceforthevillages:Villagerscanutilisetheirownassetto
dealwiththiskindofbigconsumptionneedsasshowninprevioussection.
TheimpactofthelocationdummiesisestimatedbasedonthatofthehamletD.
ThecoefficientsofthedummiesforhamletA,B,andCarenegativeshowinglow
demand of borrowing in theses hamlets compared with hamlet D where the
agriculturalprofitabilityishighwiththewellorganisedmodernirrigation.

83

Table12:BorrowingFunction

Coefficient

pvalue

Characteristicsofhouseholdhead
Age
Agesquared
Female

487.4

0.226

4.4

***

0.006

1283.0

0.290

Land

27.7

0.228

Asset

169.2

0.094

3.5

0.253

Assetsquared
Humancapital

Labour

143.2

0.541

Highschool

1105.1

0.516

University

6110.8

**

0.024

806.1

**

0.047

Student
House

863.3

0.589

Illness

1263.8

0.153

Locationdummy

HamletA

2304.7

**

0.064

HamletB

2845.6

***

0.002

HamletC

2230.7

***

0.007

Yeardummy

2001

1344.3

0.627

2002

2055.5

***

0.002

Constant

11870.7

***

0.006

4172.5

***

0.000

Samplesize

325

No.ofpositiveobservations

135

1403.4

Likelihood

Note:*,**,***denotessignificanceof10%,5%,1%respectively
Source:Authorsestimation

4.3Choiceoflendingsource
The result of estimation for equation 2 shows the determinants of borrowers
choiceonlendingsource(Table13).Inthetable,firstrowshowstheimpactsof
determinants to choose friends and relatives as a lender rather than formal
institutions. Positive coefficient on age and negative coefficient on age squared
imply that the age has a diminishing effect on borrowing from friends and

84

relatives.Theturningpoint(ormaximumpointoftheimpact)isaround55years
old. It might reflect the two facts that 1) for young villagers, more experience
enables them to form good relationship for the personal borrowing, and 2) the
elder villagers over 55 years old tend to play a role as lender rather than
borrower. As mentioned in previous section, the positive impact of female
household suggests the effect of underlying social norm to help widows. This
effectisalsoshownforthecaseofarisanborrowingwithweaksignificance(in
secondcolumn).

Table13:ChoiceBehaviourofLendingSource

Friendsandrelatives
Coefficient

Pvalue

Householdhead

Arisan
Coefficient
Pvalue

0.331

0.079

0.097

0.604

Agesquared

0.003

0.064

0.001

0.722

Female

2.636

0.094

1.823

0.211

Distrust
Borrowing
(estimates)
Land

0.632

***

0.004

0.200

0.315

0.378

0.130

0.524

**

0.050

0.024

0.307

0.162

0.067

Asset

0.006

0.890

0.080

0.105

Employee

0.343

0.219

0.049

0.826

HamletA

0.896

0.441

0.927

0.551

HamletB

2.445

**

0.033

3.170

**

0.011

HamletC

0.336

0.710

0.967

0.209

Yeardummy

Age

Locationdummy

2001

0.918

0.207

1.082

0.145

2002

0.681

0.383

0.485

0.529

Constant

6.234

0.195

2.127

0.667

Samplesize
No.ofpositiveobs.
LRstatistics
PseudoRsquare
Likelihood

130
40
86.6
0.55
98.5

74

***

Note:*,**,***denotessignificanceof10%,5%,1%respectively
Thecoefficientsareestimatedbasedontheborrowingfromformallender
Source:Authorsestimation

85

The results show that the probability of borrowing from friends and relatives
increasedwhentheborrowerstrustneighboursmore(thecoefficientondistrust
was positive). In other words, those who distrust neighbours prefer the loan
from formal financial institutions. Similar tendency is shown for the case of
borrowingfromarisaninthesecondcolumn.Theseresultsimpliedinterestrate
isnottheonlyfactorthataffectsvillagerschoicebutsocialrelationshipalsohas
animpactontheirbehaviour.
Estimated results suggested that the villagers chose formal lender when they
need big amount of loan. Negative impacts of borrowing amount on informal
borrowing (both columns) mean positive impact of that on formal borrowing.
Asmentionedabove,becausetheyhavetopaytheinitialcostfortheformalloan
application, the cost of borrowing from formal institution will become less for
thebiggeramountofloan.
Fortheimpactoflandholding,avillagerwithlargerlandtendstochooseformal
lender more than any other sources. Negative coefficient of asset holdings on
arisanborrowingmeanspoorhouseholdsmorelikelytochooseborrowingfrom
arisanratherthanfrombank.Positiveassociationbetweenformalborrowingand
employeestatusmighthappenbecauseemployeeswithstablesalarycanapply
loanwitheasierprocessofevaluation(butthesignificanceoftheimpactisweak).

5.0SummaryandConclusions
Themajorfindingsaresummarisedasfollows:1)villagersinthestudyareahave
access to the formal financial institutions; 2) the observed loan amount is
determined by investment opportunities, consumption needs (housing
expenditure, and the level of asset holdings); 3) borrowers chose the lender
depending both on material interest rate and psychological cost or trust
relationship;4)thepoortendtoborrowmoneyfromarisanmorefrequentlythan
fromformalinstitutionsorfriendsandrelatives.
Empiricalanalysisshowsthatthevillagerscanborrowmoneyfromanysourceif
they want. However, the villagers do not actively borrow for the purpose of
productiveinvestment.Possibleexplanationisthatmostofthevillagersdonot
have investment opportunities from which they can expect enough returns.
Particularly,itisrevealedthatsuchtendencyisobservedforthehouseholdwith
fewer asset holdings. Therefore, training to manage small business or
introductionofcommercialcropswouldraisetheirproductivityandactivatethe
villagersborrowing.

86

Reference

Bell, C. T., N. Srinivasan, and C. Udry. 1997. Rationing, Spillover, and Interlinking in
Credit Markets: The Case of Rural Punjab. Oxford Economic Papers, New Series, 49:
557585.
Duong, B. P. and Y. Izumida. 2002. Rural Development Finance in Vietnam: A
MicroeconometricAnalysisofHouseholdSurveys.WorldDevelopment,30:319335.
Fukui, S., I. S. Hartono, and N. Iwamoto. 2002. Risk and Rice Farming Intensification in
RuralJava.TheJapaneseJournalofRuralEconomics,4:3243.
Hoff,K.andJ.E.Stiglitz.1990.ImperfectInformationandRuralCreditMarketsPuzzles
andPolicyPerspectives,WorldBankEconomicReview,4:23550.
Iwamoto, N., I. S. Hartono, and S. Fukui. 2003. Farm Household Economy and Its
Sustainability A Case Study in Yogyakarta Province, Central Java. Department of
AgriculturalandResourceEconomics.WorkingPaperSeriesNo.03H002
Jaffee, D., J. Stiglitz. 1990. Credit rationing. In Friedman, B. M., Hahn, F. H. (eds.),
HandbookofMonetaryEconomics2,Elsevier,Amsterdam:838888.
Kochar, A. 1997. An empirical investigation of rationing constraints in rural credit
marketsinIndia.JournalofDevelopmentEconomics,57:339371.
J.Leith,C.Porter,SMERUInstitute,P.Warr.2003.IndonesiaRiceTariff,PovertyandSocial
ImpactAnalysis.March,
http://www.odi.org.uk/prspsynthesis/Indonesia_Final_PSIA.doc
Maddala. 1983. Limited Dependent and Qualitative Variables in Econometrics. Cambridge:
CambridgeUniversityPress.
Petrick, M. 2005. Empirical measurement of credit rationing in agriculture: a
methodologicalsurvey.AgriculturalEconomics,33:191203.
Zeller, M., R. L. Meyer (eds.). 2002. The Triangle of Microfinance. Financial Sustainability,
Outreach,andImpact.Baltimore:JohnsHopkinsUniversityPress.
Subejo and Noriaki Iwamoto. 2003. Labor Institutions in Rural Java: A Case Study in
Yogyakarta Province. Working Paper Series No. 03H01. Department of Agriculture
andResourceEconomics,theUniversityofTokyo.

87

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

ValueChainsandStandardsinShrimpExportfrom
BangladeshandThailandtoJapan:AComparative
StudyonSafetyCompliances
MohammadTajUddin*

Abstract
This study analyses the food safety compliance performances of different stakeholders in
value chains of Bangladesh and Thailand from mother shrimp collection to consumers
plate. Data and information collected through field surveys, interviews, and interactions
withdifferentstakeholdersinBangladesh,ThailandandJapanwereanalysedandusedas
the basis of the paper. The results reveal that the competent authority monitors the
hygiene and sanitation condition of the buyerdriven value chain activities, whereas
processorscumexportersimplementHACCPproceduresabout85percentto90percent
in all stages of production, distribution, processing and export of shrimp to assure the
quality standards. Shrimp suppliers certificate and farmers and region code in
Bangladesh while movement document and fry movement document in Thailand are
usedastraceabilitytools.Boththecountrieshavealreadystartedworkingforcertification,
ecolabelling and sustainable seafood trade. However, it is recommended to ensure chain
traceabilitytogetherwithimprovedmanagementtoincreaseproductscompetitiveness.

1.0Introduction
Shrimp as a fisheries commodity represents a great importance in the
international trade (Xinhua 2008). Its a popular food item consumed through
out the globe, although once it was considered as a luxury foodstuff by many
people. Developed nations such as USA, Japan and European countries are the
major importers of shrimp products, whereas, developing nations, especially
SouthEastAsiancountries,actasthemainshrimpsuppliersoftheworld.
Shrimp is considered significant to both Bangladesh and Thai economy. In
Bangladesh, the production of shrimp by aquaculture method is almost an
exportorientedactivityproducinganaverageoutputof53,000tonsandearning
US$530 million, which issome8.0 percent of the countrys totalexports in the
200708fiscalyear(July2007toJune2008).Shrimpisthesecondlargestforeign
exchange earner after garments. The sector employs nearly 1.3 million rural
Associate Professor, Department of Agricultural Economics, Bangladesh Agricultural University,
Mymensingh,Bangladesh.Email:tajbau@yahoo.com

89

people in forward and backward linkage activities like hatching, nursing,

producing,harvesting,processing,exportingandotherlogisticsupportservices.
Ithasthehighestpotentialforfurtherdevelopment,enoughtocultureshrimps
to earn US$1.5 billion by 2010 (BFFEA 2007). Bangladesh is already among the
top 10 exporters of shrimp in the world and accounts for some 3 per cent of
global production. The EU (45%), USA (35%) and Japan (4%) are the worlds
majorimportersofshrimpfromBangladesh.
InThailand,thecontributionofshrimpfarmingsectorhasbeenveryimpressive.
Thailand has emerged as one of the largest seafood suppliers to global market
andatpresent,thesectoraccountsfor30percentoftheworldtradeinshrimps.
Thailands shrimp production was about 600,000 tons and export value was
US$2.28billionin2007(www.fisheries.go.th).Themajormarketsforshrimpare
theUSA,Japan,EU,CanadaandSouthKorea.
Nowadays, import countries have developed standard criteria and a full
certificationsystemfortheshrimpproductionline,fromfarmtotable.Inviewof
the liberalisation of global trade and increasing demand by consumers in
developed countries, the food industries have been confronted with the
challenge of competitiveness in qualityoriented international market where
commodities, production areas, and brands compete with each other.
Competitiveness in food production in the near future will be more dependent
on the reliability of the safety and quality of the food and acceptability of
productionprocedures(Rouf2004).
On the international level, buyers and consumers are increasingly demanding
that shrimp is produced in compliance with recognised codes of conduct
regarding food safety, human rights, fair labour practices and environmental
protection. Contemporary regulation of food safety incorporates principles of
qualitymanagementandsystemicperformanceobjective(Ponte2007).However,
the export of shrimp to developed country markets is becoming increasingly
difficult because of the emerging sets of food safety and agricultural health
standards, along with buyers changing requirements (Ito 2005). Evidently, the
importance of the EU market for this particular export sector of Bangladesh is
indeedveryhigh.Anydisruptioninthismarketwasboundtohavesevereand
important implications for this exportoriented sector of the country, and
negativemultiplierimpactforthenationaleconomy(CPD2006).
However, there is an abundance of literature on the food safety standards and
regulations in shrimp industry in the world, but there have been no attempts

90

particularly to compare quality standards and emerging challenges in shrimp

exports to Japan practiced through the value chains in both Bangladesh and
Thailand.Therefore,thisstudydealswithJapansshrimpimportsanditstrends;
and measures of different stakeholders in shrimp industry in Bangladesh and
Thailand in order to comply with the safety standards of Japanese importers
andbuyers.

2.0MaterialsandMethods
Data and information for this study were collected from both primary and
secondary sources. In Bangladesh, primary data were collected in 2007 from
GeneralManagers,ExecutiveDirectorsofprocessingplantsandshrimpfarmers
inKhulnaDistrictfollowingrandomsampling.Semistructuredinterviewswere
also conducted with personnel of other stakeholder organisations such as
Bangladesh Shrimp Farmers Association, Bangladesh Frozen Foods Exporters
Association, Department of Fisheries and Export Promotion Bureau and a
researcher. In Thailand, primary data were gathered by personal interview in
2008 using semistructured interview schedule with parties actively involved
with the value chain of shrimp industry. Information has also been gathered
through individual discussion with the Thai coastal and freshwater fisheries
officers, labour department officials, Thai Frozen Foods Association and
university professors. In Japan, a university professor, one research personnel,
onerepresentativefromafishimportcompanyandanumberofshrimpbuyers
wereinterviewedin2008.Besides,secondarydataandinformationfromvarious
organisationsaswellasfrompublishedandunpublishedsourcesofgovernment
agenciesandtradeorganisationsinBangladesh,ThailandandJapaninthetype
of documents, reports, handouts, notifications, etc. having relevance with this
studywerealsoconsulted.Inadditiontothese,followuptelephonicdiscussions
wereconductedwithotherknowledgeablepersonnel.Thedataandinformation
from all these field surveys, interviews, communications and discussions were
summarisedwhichwereanalysedandusedtowritethispaper.

3.0ResultsandDiscussion
3.1Seafoodsafetystandardsandmeasuresimplementedby
differentstakeholders
Asfoodsafetyisreceivingincreasedattentionintheinternationalmarkets,itis
essential to maintain appropriate quality of the exported food items through
proper quality control and monitoring measures of respective publicprivate

91

stakeholders.IFPRI(2003)mentionedfoodsafetyisaffectedbythedecisionsof
producers, processors, distributors, food service operators, and consumers, as
wellasbygovernmentregulations.
Intheseafoodindustry,asinotherfoodindustries,thefoodsafetymanagement
systemknownasHACCP(HazardAnalysisandCriticalControlPoint)hascome
tobewidelyadoptedintheUS,EUcountries,Canadaand,toalimiteddegree,
Japan (Ito 2007). During the 1990s, developed countries made a strong shift
towardrequiringtheHACCPapproachtoassuringfoodsafety.UnderHACCP,
companies are responsible for analysing how hazards such as foodborne
pathogensmayentertheproduct,establishingeffectivecontrolpointsforthose
hazards,andmonitoringandupdatingthesystemtoassurehighlevelsoffood
safety (IFPRI 2003). It is designed to prevent hazard through inspection at
multiple points in production processes, representing a departure from point
inspectionofendproducts(Cato1998).
In response to the increasingly emerging food safety requirements, the
governmentandtheprivatesectorarejointlyimplementingaseriesmeasuresto
protect shrimp industry in Bangladesh. As a member of Codex Alimentarius
Commission(which states thatcountriesshould promote responsible fish trade
in such manner and environment that is hygienically acceptable, safe and also
meetsqualityrequirements)togetherwiththeagreementslikeSPS(Sanitaryand
Phytosanitary,toapplymeasuresnecessarytoprotecthuman,animalandplant
life and health), S&D (Special and Differential provisions, to call developed
countries to take into account the special needs of developing countries in the
preparation and application of SPS measures) and TBT (Technical Barriers to
Trade,topreventtheuseofunjustifiedtechnicalbarrierstotradeanditincludes
measures to protect consumer against deception and economic fraud),
Bangladesh has undertaken stringent measures to improve quality assurance
practices as well as management practices by strong compliances of HACCP
guidelines to produce safe fish and fish products. Declaring HACCP as a new
mandatory procedure for exportoriented shrimp processing plants in March
1998,thegovernmenthasstrengthenedtheFishInspectionandQualityControl
wing of the Department of Fisheries (DoF) to provide statutory support to the
processingplants,tomonitorprocessingandqualityassuranceactivitiesandto
fulfiltheHACCPrequirements(DoF200506).
Bangladesh shrimp industry practices standard in compliance of HACCP to
producesafeandqualityproducts.ItisevidentthatBangladeshshrimpindustry

92

implements HACCP procedure about 85 per cent in all stages of production,

distribution,processingandexportofshrimpinordertosupplysafeandquality
products to the importers and consumers. However, the shrimp industry was
foundtofaceanumberofproblemsincludingshortageofqualityseedandfeed,
illegalgrabbingofgovernmentlandbynonfarmers,impositionofturnovertax,
natural hazards, disorganised production practices, harmful effects on coastal
environmentandsoilfertility,etc.
Thai Ministry of Agriculture and Cooperatives through the Department of
Fisheries(DoF)implementedavoluntaryHACCPfishinspectionprogrammein
1991 for the first time, which was finally become mandatory in fishery
establishmentsin1996(www.fao.org/decrep/008).NowallprocessorsunderDoF
approval implement about 90 per cent HACCP procedure which increases
consumers confidence in quality and safety of Thai seafood. The
implementation of HACCP in Thailand shrimp industry has helped processing
factoriestoupgradetheiroperations,developbetterplans,andhandlefishmore
systematically.

3.2ValuechainsofshrimpinBangladeshandThailand
Shrimps value chain involves a combination of different actors (located along
different points of the shrimp industry) with their institutional arrangements
andactionsfrombroodshrimpcollectiontoenduse.Thevaluechain,according
to Kaplinsky and Morris (2001), describes the full range of activities which are
required to bring a product or service from initiation, through the different
phases of production (involving a combination of physical transformation and
the input of various producer services), delivery to final consumers, and final
disposalafteruse.
Value chains of shrimp for Bangladesh and Thailand are presented in the flow
charts 1 and 2 respectively. In Bangladesh, shrimp produced in the farm is
marketedtooverseasconsumersthroughanumberofchannelsthatisfromfarm
to village traders, purchasing agents, depot owners and final delivery to the
processing plants (Flow Chart 1). All the steps of valueadded activities are
strictly controlled under GMP (Good Manufacturing Practices) and SSOP
(StandardSanitationOperatingProcedure).CompetentauthoritylikeFIQC(Fish
Inspection and Quality Control wing under the Department of Fisheries,
Government of Bangladesh) and some independent international quality
assurance organisations specifically France based SGS (Societe Generale de
Surveillance),UnitedKingdombasedLloydsandDenmarkbasedBalticControl

93

perform inspection, testing, certification and verification services following the

Codexguidelines,codeofpractice,standardsanddirectivesofEuropeanUnion
Food Law, HACCP regulationsand requirements ofother importcountries. To
ensure safety and quality of the shrimp, Department of Fisheries monitors the
hygieneandsanitationconditionsofthefoodthroughqualitymanagementand
processorientedsupervisionthroughouttheshrimpvaluechain.

FlowChart1:ValueChainofShrimp,Bangladesh

by air

by refrigerated van

Export

by ship or air
Products inspected by competent authority (i.e., FIQCFish Inspection and Quality Control) and certified by third
party certification agence
certification agency

Processing Plant

Negotiated and verified as per import order instruction by the

local agency of the import country
by rickshaw/van/pick-up/boat

Servicing Centre
Purchasing Agents

Village Traders

Depot Owners

Facilities monitored
by DoF

by rickshaw/van/pick-up/boat

Shrimp Farm/ Gher

Monitored by DoF

by air/rickshaw/van

Nursery

Monitored by DoF

by rickshaw or van

Hatchery

Monitored by DoF

by rickshaw or van

Brood Checking Centre

Checked and certified by Department of Fisheries

by local boat

Brood Shrimp

Collected/caught from the sea by trawler and

monitored by marine surveillance check post

Source:AdoptedfromBFFEA,BangladeshandFieldSurvey2007

94

FlowChart2:ValueChainofShrimp,Thailand
Flow Chart 2: Value Chain of Shrimp, Thailand
Farm

Brood Stock
Hatchery & Nursery

GAP, inspected, sampling fry, feeds

GAP: Good Aquaculture Practices

Feed, drug manufacturers,
distributors, suppliers

FMD

GMP: Good Manufacturing Practices

HACCP: Hazard Analysis and Critical Control Point
MD : Movement Document, FMD: Fry Movement Document

Licensed, registered,
inspected etc.

Farming

GAP, inspected, sampling shrimp/fish, feeds

MD

Collector/Pre-processor

GMP, inspected

MD
GMP/HACCP, inspected,
sampling finished products

Processor
Product

Distribution/Export

Control of Department of Fisheries

Fish market

Foreign buyers

Domestic consumers

Table

Source: Department of Fisheries, Thailand and field survey 2008

Source:Department of Fisheries,ThailandandFieldSurvey2008

In Thailand, it is compulsory that all seafood processors and exporters must

implement CoC (Code of Conduct), GMP (Good Manufacturing Practices) and
HACCP to produce quality shrimp. Thai Department of Fisheries (DoF), as
competent authorityforfishery products exportedfrom Thailand, monitors the
establishments and their products regularly, administers several programmes
andactivitiestoensurethatseafoodsafetyismaintainedproperlythroughfarm
totable approach (Flow Chart 2). To ensure credible inspection and controls
throughout the value chain, competent authority covers all relevant aspects of
hygiene, public health and, also animal health. In order to sustain the Thai
shrimp industry and to maintain consumer confidence in product safety, it
applies standard practices to all sectors of the value chain from primary
production to the market place and to apply a single standard for products
whethertheproductsaredestinedfordomesticoroverseasmarkets.
UnderCoCguidelines,Thaishrimpindustryappliesbestpracticesateachstep
of the value chain to meet consumer expectations for quality, food safety,
environmental protection and taste. Thai shrimp industry and the government

95

agencies are working hard in hand to guarantee full compliance with import
countries strict requirements and international standards. DoF audits and
certifieshatcheries,farms,distributorsandprocessingplantsinordertoensure
that stakeholders follow the standardised practices. Guideline on GAP
emphasisesproductionofshrimpwhichisfresh,clean,freeoftherapeuticagents
and chemicals and noncontaminated with diseases and dirt (DoF 2008). A
combination of stakeholders like hatcheries, nurseries, shrimp farms, feed
companies and processing plants as well as international marketing companies
allow the Thai shrimp industry to develop a wellorganised, fully integrated
shrimpindustry(Wyban2007).

3.3Globalvaluechaingovernance
Theglobalvaluechainapproachplacesparticularemphasisonthecoordination
of different actors along the chain of activities involved in the production,
processing and distribution of products. It highlights the linkages between
enterprises, how their activities are coordinated and the role of lead firms in
determining what is to be produced, how and by whom (Humphrey 2005).
Governanceoccurswhenonefirmfollowsparameterssetandenforced(through
monitoring and sanctions) by another. At any point in a value chain, the three
key parameters to be specified are : (i) what is to be Producedproduct design
and specification, (ii) how it is to be producedprocess specifications, and (iii)
how much is to be produced and when production scheduling and logistics
(HumphreyandSchmitz2001).
Global farm (i.e., import countries/companies or supermarkets) negotiates with
the local farm (i.e., processorscumexporters/suppliers) on product price,
product quality, reliability of delivery, speed of response, product design and
product performance/specification standards. Fulfilling such demand
specificationbylocalfarmsinexportcountriesthroughthecollectiveinitiatives
by the actors of value chain make local shrimp products more competitive in
themarket.
Normally,localagentsofimportcountries(companiesorsupermarkets)inspect
the operational culture in the processing plants. They provide specification on
colour, shape, size, packaging materials etc. Japanese shrimp trading farm also
provides precise product specification and aska the exporters to ensure that
these specificationsare met based on consideration such asquality, designand
branding together with safety and standards. The Japanese importers
demand the processorscumexporters to produce and export following their

96

specifications.However,theserefertoanimportantgoverningroleofthebuyers
or import countries/companies (as they are regulating the supply network and
defining product specifications) on value chain of shrimp of export countries.
Although Gereffi (1999) and Gammage et al. (2006) argued that shrimp
aquaculture in Bangladesh exhibits a buyerdriven commodity chain, Islam
(2008) concluded that shrimp commodity chain is neither buyerdriven nor
producerdriven, but a twindriven commodity chain in which the wealthy
buyers control supply network, while a thirdparty certifier and some
environmentalgroupsdefinetheregulatoryaspectsofproduction,codification,
certification,andtosomeextentmanagementaspects.

3.4Japansshrimpimportsanditstrend
Shrimpisthelargestimportgroupbothinvolumeandinvalueinthecategory
of imported fisheries products to Japan. Followed by USA, Japan is the second
largestshrimpbuyerintheworld.About70percentoftheimportedshrimpgo
to the restaurant and catering market, and the other 30 per cent is sold by
supermarkets for home consumption (Jonker et. al. 2005). Table 1 shows that
Japanstotalshrimpimportsvolumehavebeenfluctuatingwithdecliningtrend
in recent years, due to increase in imported cost and price together with the
issues of very precise safety and quality requirements. Another reason also
pointed out by one Japanese professor that recently numbers of aged persons
areincreasingandtheyareconsumingreducedamountofshrimp.Asshownin
Table1:JapansShrimpImportsbyProductinDifferentYears
(inMetricTons)

Productform

2003

2004

2005

2006

2007

Raw/frozen

233,195

241,445

232,443

229,952

207,257

Prepared/preserved

33,361

39,692

42,181

50,013

48,156

Cooked/frozen

13,927

16,745

17,051

18,269

17,893

Dried/salted/inbrine

1,977

2,351

2,008

2,035

1,648

Cooked&smoked

453

618

422

414

324

Sushi(withrice)

92

341

263

204

144

Live

293

383

271

184

167

Fresh/chilled

19

33

19

0.4

301,078

275,589

Totalquantity

283,317

301,608

Source:DatacompiledfromInfofish2008

(http://www.infofish.org/marketreports/shrimp)

97

294,658

the Table 1, Japans overall import of shrimp during 2007 declined (by 9%) to
275,589 MT as against of 301,078 MT in 2006. In fact, imports in 2007 were the
lowestbyJapansince2003.Japansimportsintheformofraw/frozenshrimpare
decliningsince2005andatarecordlowin2007.Thereisslightlyapositivetrend
inimportsoftheprepared/preservedandcooked/frozenshrimpsin2006.
3.4.1MarketsharesofdifferentcountriesinJapan
Frozenshrimpimportsfromthemajorexportcountries(mostlylocatedinAsia)
toJapanareshowninTable2.Vietnam,Indonesia,IndiafollowedbyThailand
(ranks as the fourth largest exporter) and China are major frozen shrimp
exporters to Japan since 2003. As mentioned by Joseph (2008), the share of
Myanmar and Bangladesh to Japans shrimp imports also dropped mainly on
account of the presence of an antibiotic in shrimp imported from the South
Asian region. Only Thailand and China managed to perform better in terms of
shipments to Japan in 2007. However, the value of frozen shrimp import
decreased by 6 per cent slipping to 204,661 million yen in 2007 as against of
227,202millionyenin2006.
Imports of frozen shrimp are still dominated in Japan by Vietnam followed by
Indonesia since 2004. There are higher imports of raw frozen shrimp (shellon
and peeled) from the top two suppliers Vietnam and Indonesia in 2004 and
2003 respectively. Imports from Thailand and China (where a shift from black
tiger shrimp farming to vannamei or white shrimp is being practised) are
increased. It is noteworthy that Thailand is the second largest producer of
vannameishrimp(morethan90percentofthecountrystotalshrimpproduction)
inAsiaafterChina.TheNation(2008)cited,however,whiteshrimphasalower
exportpricethanblacktigershrimp,whichwilldirectlyaffectthecountrystotal
export value even though export value has increased in Thailand in 2007. Thai
shrimp and other fishery products would have promising prospects in the
Japanese market. Exports of Thai shrimp products to Japan are expected to
increase by the end of this year after winning recognition in both quality and
standard among consumers as they are confident of the quality and safety
standard(Mathaba2007).
Japan has long been the second largest (consuming 33 per cent of Vietnams
seafoodexport)marketforVietnam,aftertheUSA.BothThaiandJapanesefirms
haveinvestedheavilyinVietnamfisheryprojects,whereshrimpandcatfishare
thetwomajorexportspecies(www.siamcanadian.com).VietnamgainedUS$746
million in 2007 in exports to Japan, 21.1 per cent of the countrys total seafood

98

export value (www.vasep.com.vn). Vietnams shrimp exports to Japan have

recently been increased by 2030 per cent (www.fishupdate.com). The Ministry
of Fisheriesin Vietnam is taking measures to increase the safety of Vietnamese
aquaticproductsinordertomeetincreasingdemandfromforeignmarkets.One
ofthemeasuresistodevelopaquaculturecommunitiesthroughoutthecountry,
whoseproductionprocesses,fromtheselectionofbreeds,feedandmedicinesto
processingandexport,willbeputunderstrictsupervision.
Table2:JapansFrozenShrimpImportsinDifferentYears
(inMetricTons)

Countryoforigin

2003

2004

2005

2006

2007

Vietnam

Indonesia

India

Thailand

China

Russia

Myanmar

Canada

Greenland

Bangladesh

47,626
(20.4)
52,367
(22.5)
28,191
(12.1)
16,803
(7.2)
20,494
(8.8)
8,641
(3.7)
5,377
(2.3)
8,938
(3.8)
9,180
(3.9)
3,004
(1.3)

55,506
(23.0)
48,623
(20.1)
31,572
(13.1)
17,192
(7.1)
22,609
(9.4)
8,911
(3.7)
7,630
(3.2)
8,469
(3.5)
7,683
(3.2)
3,415
(1.4)

54,573
(23.5)
45,574
(19.6)
26,309
(11.3)
18,398
(7.9)
24,092
(10.4)
10,382
(4.5)
7,519
(3.2)
8,054
(3.5)
7,527
(3.2)
3,194
(1.4)

51,133
(22.2)
43,665
(19.0)
28,546
(12.4)
20,097
(8.7)
22,810
(9.9)
9,518
(4.1)
8,847
(3.8)
8,665
(3.8)
6,788
(3.0)
4,001
(1.7)

40,041
(19.3)
37,080
(17.9)
27,025
(13.0)
26,380
(12.7)
23,997
(11.6)
8,903
(4.3)
8,021
(3.9)
7,554
(3.6)
5,427
(2.6)
2,568
(1.2)

Others

3,2630
(14.0)

29,835
(12.4)

26,821
(11.5)

25,882
(11.3)

20,261
(9.8)

Totalquantity(MT)

23,3251

241,445

232,443

229,952

207,257

(100.0)

(100.0)

(100.0)

(100.0)

(100.0)

Totalvalue(Mil.Yen)

225,409

216,820

213,858

227,202

204,661

Note:Figuresinparenthesesindicatepercentages
Sourceofdataassembling:SWR2008(http://www.swr.ucsd.edu/fmd/sunee/shrimp)

99

3.5Valueaddedproductsandbuyersrequirements
In all developed countries food habits are changing. The consumers are no
longer willing to depend on excessive time for preparing food. Heat and eat
fishery products have proved well to be perfect niche for seafood at the retail
level, because these are convenient to the consumers. In catering and retail
sectors in the USA, EU and also in Japan, market demand for specific value
addedproductshavebeenrising.Valueadditionalwaysfocusesonmarketand
consumer expectations. Among seafood, shrimp exhibits the widest range and
highest degree of value addition. In consideration of health aspect, every food
itemshavebeendiversifyingindevelopedcountriesoftheworld,wherecaloric
counts,dietaryplans,nutritionalfacts,servingsizeonseafoodpacksandrecipies
incustomerslanguageareusefuladditiontovalueaddedproducts.Inclusionof
these will definitely increase the presence of producers products on
supermarketsandotherretailoutletsofpotentialmarkets(BFFEA2007).
Aside from meeting the sanitary and phytosanitary requirements of an import
country,allpartiesinthesupplychainmustmeetincreasingconsumerdemand
for quality and safety assurance of produce. They must also meet additional
requirementsofimportersandretailersintermsofcompliancewiththerelevant
standards of the ISO (International StandardSetting Organisation). The factors
important to Japanese consumers responded by different stakeholders of
supplying countries are: organic product, uniformity in size, precise grading,
and accuracy in weight, colour separation, freshness, good shape and quality.
Jonker et. al. (2005) also identified some standards of Japanese customers, such
asfreshness,quality,visualperfection,taste,stabilityofsupply,andlowprices.
Allstakeholdersinthevaluechainmustensureandmaintainthesestandardsas
mandatory requirements for Japanese customers. Such compliances eventually
will prevent rejection of shrimp consignment as well as ensure product
acceptability by importers and ultimately by retailers and consumers. The
processingplantsinBangladeshandThailandareinvolvedinprocessingvalue
addedproductsfollowingstringentsizing,weight,cleanness,colourgradingand
general quality standards. The quality of such valueadded products exported
from both the countries is also being appreciated for maintaining taste and
meetingconsumerspreferencesaswell.

3.6Traceabilityrequirementsinshrimpexport
Traceability of food at present emerges as a vital area of competition and a
crucialdemandtrendacrosstheworldmarket.UnderEUlaw,traceabilitymeans

100

theabilitytotrackanyfood,feed,foodproducinganimalorsubstancethatwill
be used for consumption, through all stages of production, processing and
distribution (EC 2007).According to the ISO, traceability refers to the ability to
trace the history, application, or location of an entity by means of recorded
information. This principle is known as one step up and one step down. For
ensuringtraceability,exportersshouldmaintaintransparentandcrediblerecords
ofthetradedproductsthathereceivesfromthefarmersorsuppliers.
Both Bangladesh and Thai shrimp industries have meantime taken a good
number of steps to implement this issue through internal traceability (i.e.,
recordinginformationonrawmaterials,intermediateandfinalproductswithin
a fish plant) and external traceability (i.e., making documentation through
marketchainfromfrycollectiontoconsumerplate)inordertocomplywiththe
requirementsoftheimportcountries.Processorwillguaranteetothebuyersthat
his products are safe for human consumption and will not pose any health
hazards to consumers. Food business operators will ensure that all stages of
production, processing and distribution of food under their control satisfy the
relevanthygienecondition.
InThailand,movementdocument(MD)isissuedbyDepartmentofFisheriesat
hatcheries, farms through to processing plant to cover all cultured or imported
species including ornamentalfish, whereas fry movement document (FMD) is
practisedfordeliveringfrytobuyerinordertotracebacktooriginwhendrug
residue or disease is detected. Bangladesh shrimp processing plants initially
ensure shrimp suppliers certificate and farmers and region code where
suppliers certify that the shrimps are not contaminated by any kind of
pathogenicbacteriaandnochemicaldrugsareusedduringrearing;theshrimps
are harvested, preserved and transported in a hygienic manner. Besides, in the
processingplantsofbothBangladeshandThailand,allinnercartonorpolybag
and master carton are labelled and printed with all necessary information and
specificationssuchas,approvalnumber,FDA(FoodandDrugAdministration)
registration, place/country of origin, processors name or ID, lot number,
date/methodofproduction,etc.fortheforeignbuyerssothattheycouldbeable
totracetheoriginofeachproduct.

3.7Certification,ecolabellingandsustainableseafoodtrade
In recent years, the landings of fish from the worlds oceans have gradually
declined as stocks have been progressively over fished, and at the same time,
demand for seafood has been steadily rising (Greenpeace 2008). While

101

consumers and supermarkets in the developed countries are becoming more

safety conscious and adopting more sustainable purchasing patterns,
certificationandecolabellingofseafoodprovidesassurancetomeettheneedsof
futuregenerations.
It may be pointed out that ecolabelling is increasingly perceived as a way to
improve the management of fisheries and conservation of biodiversity. If
fisheries management improves due to efforts undertaken to comply with
certificationcriteria,thepotentialbenefitstofisheriessectorwillbemuchhigher.
However, the potential usefulness of ecolabelling schemes may be constrained
duetothedifficultiesfacedbythelowincomecountriesintermsofcompliance,
and developing necessary organisational capacity. Though ecolabelling is
considered to be an informal international environmentpromoting tool and in
mostcasesthesearevoluntaryinnature,itsintroductioninshrimpexportsfrom
Bangladesh needs to be actively considered since many developed countries
havealreadyadoptedthisschemes.TheinabilityorunwillingnessofBangladesh
to do this may lead to weakening of its competitive strength and erosion of its
globalmarketshare(CPD2006).
Certificationof aseafoodproduct indicates ifit was producedin a sustainable,
healthy, socially responsible and environmentallyfriendly way. It should
address four main areas: food safety and quality, social impacts of fish
farming on local communities, environmental issues and economic feasibility
(George2007).
Governments in both countries strongly prohibit shrimp fry collectors not to
catchfryfromtheseatocontrolstockstatus.Strictsupervisionandmonitoring
system are needed to protect mother shrimps at the sea and there should be a
clearcut law regarding catching of mother shrimp. To maintain biodiversity,
trawlers fishing in the sea are using turtle and other nontargeted species
excluding device to reduce unwanted catchingand to preventdamage to other
species.Governmentscanmakeanimpactthroughpolicymaking;fishfarmers
can use sustainable techniques in fish harvesting so that future catches and
harvest can be assured, and processorscumexporters have a responsibility to
purchase fish from the sustainable sources. Finally, ecosystem based fishery
management approach should be encouraged for sustainable harvesting to
ensurefishavailabilityfortodayandalsoforfuturecatchesandharvests.
For internationally certified organic aquaculture in Bangladesh, SIPPO (Swiss
Import Promotion Programme) took a pilot project in 2004 at the district of

102

Satkhira,inSouthWestofBangladeshandnexttothenationalmangroveforest.
Farmers produce shrimp by using traditional methods based on the best
utilisationsofthesun,therichwatersofthenearbySundarbanandlargespace
of land under a unique programme of reforestation, without use of any
ingredients such as compound feed, chemicals, fertilisers. The shrimp grows
naturallyjustwithnaturalfoodcomingfromthisfantasticnaturalfoodreservoir.
With such initiatives, Bangladesh would be able to earn substantially by
exportingshrimptomaintaincompetitivenessintheglobalmarket(BFFEA2007).
Thai Department of Fisheries has developed specific projects to enhance the
shrimp culture sector by improving various aspects such as environmental
quality,shrimpproductsafety,andtrustmarketing.Oneoftheprojectsnamed
seawater irrigation systems for marine shrimp culture that produces
appropriate quality seawater for shrimp culture, collects effluent from shrimp
ponds,andtreatstheeffluentbeforedischargeintonaturalwater.Theobjectives
ofthesystemsaretoproducesustainableshrimpculture,maintainenvironment
andbalanceecologicalsystem(DoF2008).

3.8ComparativefeaturesofbothBangladeshand
Thailandshrimpindustries
Thailandemergesastheworldsleadingexporterofshrimpwith30percentof
the market share. As the buying countries demand that the food must be
produced,processedandhandledwithnecessaryhygienemaintenance,shrimp
standards and safety programme in Thailand is maintained by CoC and GAP
guidelines. In sharp contrast, Bangladesh produces only three per cent of the
global production. All the activities from brood (mother) shrimp collection to
finishedproductsuptoexportarestrictlycontrolledunderGMPandSSOP.
Thailand has developed three open auction markets which facilitate shrimp
farmers to get the fair price for their products whereas there is no open
auctioning of shrimp in Bangladesh. Normally brokers/agents of traders in
Bangladesh buy shrimp from farmers field paying farmgate price, where
farmershavelessbargainingpower.
There are four types of shrimp being exported from Bangladesh, i.e., sea water
blacktigershrimp(Penaeusmonodon63%oftotalexport)andfreshwaterprawn
(Macrobrachium rosenbergii19% of total export), sea water brown/grey shrimp
(Metapeneaus monoceros10% of total export), and sea water white/pink shrimp
(Penaeus indicus8% of total export). On the contrary, white shrimp (Penaeus

103

vannameimorethan90%ofitsexport)andblacktigershrimp(lessthan10%of
itstotalexport)fromaquaculturearethemajorfisherycommoditiesinThailand.
Boththecountriesfollowtraditionalextensivemethodforshrimpaquaculture.
Thailand has four up to date and wellequipped laboratories for testing and
detectionofbacteria,virus,antibioticsandheavymetals.Theycarefullyexamine
theshrimpproductspriortoshipmenttoassureoptimumqualityinaccordance
with import countries requirements and international standards. Recently,
Bangladesh has been equipped with sophisticated equipments for all
microbiological, antibiotics, heavy metal and virus tests. Earlier, Bangladesh
usedtosendsamplestoThailandandSingaporefortherequiredtests,whichare
nowpossibleinBangladesh.
In comparison to 42 largescale modern feed mills in Thailand, Bangladesh has
only six large modern feed mills and 20 small scale crude feed producing
factories.
In realising the importance of quality and safety of fish and fishery products,
Thai Department of Fisheries, as competent authority, operates several
programmes to ensure food safety. In addition to FIQC in Bangladesh, SGS,
Lloyds and Baltic Control, as international independent company, perform
comprehensive inspection and superintendence both at the company and at
shippingportsasperinstructionoftheimportcountriestoensurefoodsafety.
WhileBangladeshdidnotexplorethepossibilitiesofexportdiversificationand
still it remains happy with the export of frozen shrimp only, Thailand is
continuing exports with its diversified range of product frozen, canned, dried
andsaltedshrimps.
Traceability has become an important factor in food safety which also helps to
definetheresponsibilitiesoffoodoperators.Thailandshrimpindustryensuresit
through both movement document and fry movement document whereas
Bangladeshstartstoexecuteitbyshrimpsupplierscertificateandfarmersand
regioncode.

4.0Conclusions
Both government and private sectors of import countries set standards and
requirements for the shrimp exporters. This study analyses the compliance of
exporters of both Bangladesh and Thailand with these standards from mother
(brood) shrimp collection and aquaculture farm to the consumers table. It is

104

concluded that different stakeholders of shrimp industry in Bangladesh and

Thailand are working concertedly to comply with the recognised codes of
conduct. However, shrimp imports volume in Japan is decreasing since 2003;
demand for valueadded products is increasing. Therefore, countries with
stringent quality assurance programmes would also be able to sell more
valueadded products to Japan. Bangladesh has a possibility to expand its
shrimp trade with Japan establishing value innovation through export
diversificationofproductsandmeetingfoodsafetystandardsproperly.
Bangladeshshouldinvestonwaystoimprovethepresentstatusofshrimptrade
and make this avenue into the real business sharing and experiencing the
knowledge from Thailand. There is a bright prospect of shrimp culture in
Bangladeshrecoveringtheillegallyoccupiedgovernmentlandanddistributing
thosetotherealshrimpfarmersandutiliseingthefullcapacityoftheprocessing
plantsinordertoaugmentitsproductionandexportvolume.InfactBangladesh
can be benefited more from increased shrimp cultivation with the present
technology and could fetch more export earnings using its full potential of
processing plants and cultivable land. However, in order to minimise the
socioeconomic problems and social tension, cultivation must be expanded
following a clear land use policy based on land topography, soil quality and
other environmental factors. Among others, new technology in shrimp disease
prevention, new seed production systems, better management and
environmental friendly aquaculture practices for sustainable seafood trade,
ensuringchaintraceabilitywitheffectivefarmregistrationprocedureaswellas
strict supervision and monitoring system at every stagein the export chain are
essentialtosatisfythesafetystandardsandrequirementsofdevelopedcountry
consumers.
Recently, Thailand has started to implement a new approach, i.e., shrimp
cluster integrating together all related supporting industries such as ice
producer,coldstorage,animalfeed,packaging,transport,institution/association
etc.,tomaintainproductssupplysmoothwithbetterqualityandlowcosttothe
market. It is also recommended to start the feasibility of such approach in
Bangladesh to improve management, meet quality standards and increase
productscompetitiveness.

105

References

BFFEA. 2007. Shrimp & fish news, Newsletter of Bangladesh Frozen Foods Exporters
Association,MarchJuneedition,Dhaka,Bangladesh.
Cato, J. C. 1998. Economics of hazard analysis and critical control point (HACCP)
programmes,FAOFisheriesDepartment,FAO,Rome.
CPD.2006.FishtradeliberalisationinBangladesh:ImplicationsofSPSmeasuresandeco
labeling for the exportoriented shrimp sector. Centre for Policy Dialogue, Dhaka,
Bangladesh.
DoF. 20052006. Annual report, Department of Fisheries, Ministry of Fisheries and
Livestock,Dhaka,Bangladesh.
DoF. 2008. Quality shrimp. Department of Fisheries. Ministry of Agriculture and Co
operatives,Bangkok,Thailand.
EC. 2007. Food traceability, DirectorateGeneral for Health and Consumer Protection,
EuropeanCommission,Brussels.
Gammage,S.,K.Swanburg,M.Khandkar,M.Z.Islam,M.Zobair,andA.M.Muzareba.
2006. A gendered analysis of the shrimp sector in Bangladesh. Gender Access to
TradeandExpansion,USAID,Bangladesh.
George,K.2007.FAOpartnershipworkingtoestablishguidelinesforcertificationof
farmerfish.FAONewsroom,MediaRelations,FAO,Rome.
Greenpeace.2008.Challengingtheaquacultureindustryonsustainability.Greenpeace
International,Amsterdan,TheNetherlands.
Gereffi, G. 1999. International trade and industrial upgrading in the apparel in
commoditychain.JournalofInternationalEconomics,Vol.48,No.1,Pp.3770.
Humphrey, J. 2005. Shaping value chains for development: Global value chains in
agribusiness. Trade Programme, Federal Ministry for Economic Cooperation and
Development,Bonn.
Humphrey,J.andH.Schmitz.2001.Governanceinglobalvaluechain.IDSBulletin,Vol.
32,No.3,Pp.1929.
IFPRI.2003.Foodsafetyinfoodsecurityandfoodtrade:2020visionforfood,agriculture
andtheenvironment.InternationalFoodPolicyResearchInstitute,Focus10,brief4
&5of17.
Infofish. 2008. Shrimp monthly market report (http://www.infofish.org/marketreports
/shrimp,accessedon2008.4.12).
Islam, M. S. 2008. From pond to plate: Towards a twindriven commodity chain in
Bangladeshshrimpaquaculture.FoodPolicy,Vol.33,Pp.209223.
Ito, S. 2007. Export competitiveness and propoor growth in the shrimp industry in
Bangladesh.EnterpriseDevelopmentandMicrofinance,Vol.18,No.2&3,Pp.
241254.
Ito, S. 2005. The distributional effects of compliance with food safety and agricultural
health standards on small producers in developing countries. Proceedings of the
JapanSocietyforInternationalDevelopment,Japan.

106

Jonker,T.H.,H.Ito,andH.Fujishima.2005.FoodsafetyandqualitystandardsinJapan:
Compliance of suppliers from developing countries. Agriculture and Rural
DevelopmentDiscussionPaper,theWorldBank,USA.
Joseph,G.2008.ExportsofshrimptoUSA,Japandecline.Businessstandard,afinancial
daily,March20,NewDelhi,India.
Kaplinsky, R. and M. Morris. 2001. A handbook for value chain research. University of
Sussex,UK.
Mathaba.2007.http://www.mathaba.net/news/thailand(accessedon2008.6.03).
Ponte, S. 2007. Bans, tests, and alchemy: Food safety regulation and the Uganda fish
exportindustry.AgricultureandHumanValues,Vol.24,No.2,Pp.179193.
Rouf, A. 2004. Bangladesh, Country report of Asian Productivity Organization, Tokyo,
Japan.
SWR. 2008. National marine fisheries service, Southwest Regional Office
(http://www.swr.ucsd.edu/fmd/sunee/shrimp,accessedon2008.5.21).
The Nation. 2008. New Australian inspections threaten Thai shrimp exports. Bangkoks
IndependentNewspaper,June03,Bangkok,Thailand.
Wyban,J.2007.Thailandsshrimprevolution,AQUACultureAsiaPacificMagazine,Vol.
15,No.4,Pp.1722.
Xinhua, Y. 2008. Comparative advantage analysis of shrimp production in Asia.
AquacultureAsiaMagazine,Vol.9,No.1,Pp.3336.
http://www.fao.org/decrep/008(accessedon2008.2.18).
http://www.fisheries.go.th(accessedon2008.4.10).
http://www.fishupdate.com/news/fullstory.php/aid/3558/Vietnam.
(accessedon2007.4.12).

http://www.siamcanadian.com(accessedon2008.6.4).
http://www.vasep.com.vn(accessedon2008.7.14).

107

108

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

SimulationofGrowthandYieldofMaizeunderWater
StressImposedduringCriticalGrowthPeriodsin
NakhonRatchasimaProvince,Thailand
KiattiyosThongsaga*andS.L.Ranamukhaarachchi*

Abstract
Waterstressisunpredictableinrainfedagriculture,andregularlyaffectsmaizeproductionin
Thailand. This study was conducted to determine the effect of 10day dry period occurring
duringcriticalgrowthstagesongrowthandyieldofmaize,andtosimulateandcomparethe
performanceusingCERESMaizemodelinDSSATversion4.0.2.0usingconcurrentweather
data.Twopopularmaizehybrids,namelySuwan4452andPacific224,weresubjectedtofour
10daydryperiods[viz.3645daysfromseeding(DFS),4655DFS,5665DFS,and6675
DFS]astreatmentstogetherwithacontrolgrownwithregularirrigationinasplitplotdesign
with four replicates. The experiment was conducted during the longest dry period, from
December2006toApril2007,undercontrolledirrigation.
Soil moisture content at 015cm and 1530cm depths reached near the lower limit ofreadily
available water (RAW) in six days, after the cessation of irrigation, which declined to near
permanentwiltingpoint(PWP),butdidnotreachPWPduringthe10dayperiod.Leafarea
index (LAI) (ranged from 4.3 to 4.6), plant height (ranged from 2.2 to 2.4m) and biomass
(ranged from 6.8 to 7.6 t/ha) at silking were not significantly affected by the 10day
suspension of irrigation, but various differences were observed in the plant height: Suwan
4452wastallerthanPacific224.Numberofgrainsperear,100grainweight,grainyieldand
the harvest index remained unaffected due to the 10day suspension of irrigation, but were
significantlygreaterinPacific224thanSuwan4452.
CERESMaizemodelshowedexcellentresultsinpredictingtimetosilking,LAIandbiomass
atsilking,100grainweight,grainyieldandtheharvestindex.However,somedeviationswere
found in Suwan variety for LAI and 100grain weight, and of Pacific 224 in the time to
silking. Comparisons showed that simulations were very much close, except for the cases
noted.Themodelsimulatedresponsessimilartotheactualvaluesinthisstudy.

1.0Introduction
Water stress due to erratic rainfall is a major constraint for maize production
under rainfed condition(Lafitte 2000). Significant yield reductions occur when

Agricultural Systems and Engineering Programme, School of Environment, Resources and

Development, Asian Institute of Technology, Pathumthani, Thailand. Corresponding Email:
ranamuka@ait.ac.th

109

maizeencountersdroughtconditionsduringcriticalgrowthperiods(Jampatong
andBalla2005).Maizeyieldisusuallyaffectedbywaterstressoccurringatany
stage (Grant et al. 1989), and drought occurring after planting and after
flowering causes higher yield losses compared to other stages (Ekasingh et al.
2004;KefeleandRanamukhaarachchi2004).
Corn is relatively tolerant to water stress during the vegetative stage, but very
sensitive during tasselling, silking and pollination, and moderately sensitive
duringgrainfillingperiod(Shaw1977;Dassetal.2001).However,yieldlossis
maximum when drought occurs during the flowering stage (Doorenbus and
Kassam1979).Thisisbecausereproductivedevelopmentatthetimeofflowering
is especially sensitive to drought stress (Zinselmeieret al. 1999), which inhibits
activities of both vacuolar and cellwallbound acid invertase activity during
kerneldevelopment(Zinselmeieretal.1995).Panitnoketal.(2005)observedthe
greatest effect on growth, development and grain yield of maize due to water
deficitattasseling,whereastheeffectwastheleastduringvegetativestage.
In Nakhon Ratchasima Province, where 26 per cent of Thailands maize is
grown, the risk of water stress is highest during reproductive and grain filling
periodsofearlyseasonmaize(MarchJuly),whileforlateseasonmaize(August
December), water deficit coincides with the vegetative phase (Thiraporn 1996).
Theareaofmaizeinthefirstseasonisapproximately20percent,and80percent
inthesecondseason.Inordertoavoidheavyyieldlosses,weeklyirrigationof40
mmofwaterhasbeenrecommendedtomaizegrowerswithirrigationfacilities
[PersonalcommunicationwithDr.Jampatong,theMaizebreeder&Agronomist
in the National Corn and Sorghum Research Centre (NCSRC), Pak Chong
District,NakhonRatchasimaProvince,NortheastofThailand].
Maize is widely grown in rainfed conditions, where water stress periods are
uncertain,supplementaryirrigationisnotpragmatic,andyielddeclinesareoften
occurred.Theonlyoptionleftforthefarmersistoadjustplantingtimefollowing
the historical rainfall probabilities, and soil storage of moisture is only an
expectation, capable of supporting field grown maize during prolonged dry
periods.However,themaximumlengthofadryperiodthatcouldaffectmaize
productionunderfieldconditionsisuncleartoboththefarmersandresearchers.
Therefore, it was hypothesised that 10day dry periods occurring at critical
growth periods would not lead to significant growth and yield reductions
comparedtomaizegrownundernonstressedconditions.

110

The effects of water stress have also been evaluated with crop simulation
modellinginrainfedmaize(Sinclair1994).Therefore,itwasalsohypothesised
that crop modelling in parallel to the fieldlevel performance under ongoing
weather conditions would provide valuable information similar to the actual
performance in field conditions. This would be guidance to the growers and
researchersinidentifyingmanagementpracticestominimisewaterstresseffects
andtoensuresatisfactorymaizeyields.
Thisstudywasconductedtodetermineandcomparethegrowthandtheyieldof
maizeunderirrigationinparallelwithsimulationsusingCERESMaizeModelof
the Decision Support Systems for Agricultural Technology Transfer Version
4.0.2.0(DSSATdevelopedbyIBSNAT).Thisstudymightprovidetheawareness
ofandopportunitiestodeterminethedegreeofyieldreductioncausedbywater
stressoccurringduringcriticalgrowthstagesandtodevelopandapplyeffective
management decisions including the selection of appropriate planting time in
ordertominimiseadverseyieldreductions.

2.0MaterialsandMethods
This experiment was conducted under irrigation during the dry season from
December2006toApril2007intheNationalCornandSorghumResearchCentre
(NCSRC) located in Pak Chong District, Nakhon Ratchasima Province of
NortheastofThailand(latitude14.5N,longitude101E,360metersabovemean
sealevel).

3.0ExperimentalTreatments
Thestudywasdesignedtoassesstheimpactofwaterstressbyimposing10day
dry periods during critical growth stages of maize starting from seedling
emergence,kneehighstage,tasselingandsilking,andphysiologicalmaturity.
The study was conducted as a 5 x 2 factorial experiment in a split plot design.
Four 10day dry periods imposed by suspending irrigation at four critical
growthperiods[viz.36to45(growthstagesV6toV9),46to55(V9toV14),56to
65(V14toVT)and66to75(VTtoR2)daysfromseeding(DFS)]andacontrol
(regularly irrigated at weekly interval) were tested for two popular maize
varieties (viz. Suwan 4452 and Pacific 224). Water control and varieties were
randomly assigned to main plots and subplots, respectively. There were four
replicatesbelow:

111

Note:

Experimentaltreatments(darklinesshowthetimingofimposing10daywaterstressby
suspendingofirrigation.)[Sseeding;SEseedlingemergence;KHKneehighstage;
VTTasseling;PMphysiologicalmaturity;Mmaturity]

4.0CulturalPracticesandVarieties
Theconventionallandpreparationwasadoptedusingadischarrow,andridges
andfurrowswereformedata75cmdistance.Eachmainplotwas16mlongand
4.5mwide,anddividedlengthwiseintotwosubplotsforthetwomaizevarieties.
Topreventlateralmovementofwater,aplasticsheetwasplacedverticallytoa
depthof50cmaroundeachmainplotbymakingadeepdrainandthencovering
fromeithersidewithsoils.TheperiodfromearlyDecembertoEarlyAprilofthe
following year is usually a dry period with no possibility of rainfall under
normal weather pattern. Therefore an irrigation system containing both
sprinklers and furrow irrigation was set up to satisfy the water requirement of
maize.Asbasaldressing,NandPwereappliedattherateof20and25kgha1,
respectively,bybandapplicationshortlybeforeseedingasperrecommendations
oftheNCSRC.Twoseedsperhillwereseededon01stDecember2006withinter
and intrarow spacing of 75cm and 20cm, respectively, manually by hand. The
followingday,allplotswereirrigatedwithsprinklersforaboutfourhourswhich
approximatelyprovided40mmwatertoeachplot.Againsprinklerirrigationwas
practised for five days after first irrigation, and thereafter continued at weekly
intervals until corn plants reached a kneehigh stage. From kneehigh stage
onwards furrow irrigation was applied to every plot for about two hours at
weekly intervals. Two weeks after seeding (WAS) the excess seedlings were
thinnedouttotheplantdensityof6.67plantsm2.Nitrogenwastopdressedat
115kgNha1,30daysaftertheseedlingemergence.Irrigationwassuspendedfor
10consecutivedayscommencingfrom36daysfromseeding(DFS)andrestored
after 10 days for the first treatment. In the second treatment, irrigation was
suspended during 4655 DFS, and for the third and fourth treatments during

112

5665 DFS and 6675 DFS, respectively. The dry periods coincided with critical
growth periods, i.e. 3645 DFS with V6 to V9 (i.e. kneehigh stage), 4655 DFS
with V9 to V14, 5665 DFS with V14 to VT (tasseling) and 6675 DFS with VT
(tasseling)toR2(grainfilling).Inallirrigationtreatments,furrowirrigationwas
given on the previous day prior to suspension of irrigation. In addition, a
control treatment was maintained without the suspension of irrigation during
thefullcropduration.Irrigationwasterminatedaftertheblacklayerformation
inmaizecobsinalltheplots,andthecropwasmanagedaspercropmanagement
guidelinesoftheNCSRC.

5.0ExperimentalObservations
5.1Soilmoisturecontent
Soilmoisturecontent(SMC)atthefieldcapacity(FC)andthepermanentwilting
point(PWP)weredeterminedwithin015cmand1530cmdepthsineachplotat
thebeginning.Thepressureplatewasusedat0.3and15barstodeterminethe
FC and the PWP, respectively. In addition, soil samples were taken at 2day
intervalsfromtheplotsafterirrigationwassuspendedfor10daystodetermine
soilmoisturedepletion.Soilaugurwasusedforsoilsampling,foursoilsamples
weretakenfromeachplotandeachreplicate,andgravimetricmethodwasused
to determine the SMC (Ryan et al. 2001). Fresh weight of soil samples was
recordedfirst,andthenovendriedat105Cuntilaconstantweightwasreached,
thedryweightwasrecorded,andfinally,thepercentageofSMCwascomputed.

5.2 Plantheightandleafarea
Bothplantheightandleafareawererecordedat50percentsilking(74daysafter
seeding DAS). Plant height was measured up to the tip of the tassel from 10
randomlyselectedplantsineachplot.Theleafareawascomputedbyrecording
the length and the width of the widest point of each leaf from the same ten
plantsineachplot.Theleafareawascomputedbymultiplyingtheleaflength,
widthand afactor (McKee1964) andleafareaindex (LAI) was then computed
(McKee1964).

5.3 Plantbiomass
Plant biomass was estimated at 50 per cent silking. Ten plants leaving two
outermost rows as border rows on either side and one meter from either end
were used for estimating the biomass. These plants were dissected into leaves
andstemparts,andfreshweightswererecorded.Thenthestemswerechopped,

113

and a subsample of approximately 1.0 kg was taken from both leaves and the
stemsfromeachtreatmentandreplicate,ovendrieduntilaconstantweightwas
reached and the weights were recorded. Using the weights of subsamples, the
bulk weights of leaf and stem were converted to a dry weight (Ogoshi et
al.1999).

5.4Yieldandyieldcomponents
Allmaizeplantsinthetwocentrerowsofeachsubplotleavingonemeterarea
from either end (sampling area of 6 m x 1.5 m = 9 m2) were harvested to
estimate the grain yield, yield components, and the total biomass. Ten plants
were randomly harvested at the ground level to determine the yield, 100grain
weight and stalk weight. These plants were separated for grain and nongrain
parts,freshweightofeachpartwasseparatelyrecorded,driedinanovenuntila
constant weight was reached, and then dry weights were recorded. The
remaining plants were also cut at ground level, separated for ear, grains and
stalk,andthefreshweightswererecorded.Thesefreshweightswereconverted
to dry weight using the data obtained from a 10plant subsample. The grain
yieldandthe100grainweightwerecomputedat15percentmoisturecontent.

6.0ExperimentalDesignandStatisticAnalysis
The data were analysed as per split plot design using SAS programme version
9.0.Growthandyielddataweresubjectedtoanalysisofvariance(AoV),andthe
meanswerecomparedwiththeFishersProtectedLeastSignificanceDifference
procedure(SteelandTorrie1980).

7.0ApplicationofCERESMaizeModel
The CERESMaize model of the Decision Support Systems for Agricultural
Technology Transfer (DSSAT) Version 4.0.2.0 software was used for simulating
growth and yield of maize under the same water stress levels. Model inputs
were weather data recorded during the same cropping season, which include
solarradiation,maximumandminimumairtemperatureandprecipitation,soil
physical and chemical characteristics including drainage, runoff, slope soil
classificationcontainingsoildepth,particlesizeanalysis,pH,%C,%Nandcation
exchange capacity (CEC), and crop management data, which include
management practices as per recommendation of the NCSRC. Crop residue of
1.3t/hawasusedasreceivedfromthepreviouscrop.Geneticcoefficientofmaize
(Hoogenboom 2004) as degree days (base 8oC) from emergence to the end of
juvenilephase(P1),fromsilkingtophysiologicalmaturity(P5),andforaleaftip

114

emergence (phyllochron interval, C d) (PHINT), photoperiod sensitivity

coefficient(01.0)(P2),potentialkernelnumber(G2),potentialkernelgrowthrate
mg/kerneld(G3)usedinthesimulationareshownintheTable1.
Table1:GeneticCoefficientsofSuwan4452andPacific224VarietiesofMaize

Geneticcoefficient

Suwan4452

Phenologicalcoefficient
343.9
Thermal time from seedling emergence to the end of the

juvenile phase (P1), (expressed in degree days above a

base temperature of 8C) during which the plant is not
responsivetochangesinphotoperiod.

Pacific224

346.9

Extent to which development (expressed as days) is

delayed(P2),foreachhourincreaseinphotoperiodabove
the longest photoperiod at which development proceeds
atamaximumrate(whichisconsideredtobe12.5hours).

Thermal time from silking to physiological maturity P5,

(expressed in degree days above a base temperature of
8C).

1149

1190

Phyllochron interval (PHINT), the interval in thermal

time (degree days) between successive leaf tip
appearances

48

48

640

640

Growthcoefficient
G2,Maximumpossiblenumberofkernelsperplant.
G3,Kernelfillingrateduringthelineargrainfillingstage
andunderoptimumconditions(mg/day).

Table2:SelectedSoilCharacteristicsatDifferentSoilLayersinResearch
PlotsattheNationalCornandSorghumResearchCentre,Pak
Chong,NakhonRatchasimaProvince,Thailand

Depth,
cm
015

OrganicC
%
2.84

TotalN
%
0.17

Avail.P
ppm
8.1

Exch.K
ppm
194

pH
(1:1water)
7.2

pH
(1:1KCl)
6.9

1530

2.75

0.16

7.5

138

7.3

6.9

3050

1.96

0.11

6.3

91

7.3

6.9

5075

0.78

0.13

3.2

139

7.3

7.0

75100

0.16

0.06

2.5

102

7.4

7.3

SoilsbelongedtothePakChongsoilseries(PC),andotherinformationincluded
taxonomy (veryfine, kaolinitic, Isohyperthemic Rhodic Kandiustox); texture

115

(clay53.8%; sand5.6%). Layerwise soil analysis was conducted at the

AgriculturalTechnologylaboratoryatAITtouseasinputstothemodel(Table
2).TheparametersincludedpH(1:1soil:waterand1:1)(McLean,1982);organic
C [WalkleyBlack method (Nelson and Sommers, 1982; FAO, 1974)]; total N
[Kjeldhalmethod(BremnerandMalvaney,1982)];availableP[(BrayIImethod)
(BrayandKurtz1945),exchangeableK(BarkerandSurh,1982)andCEC(Ryanet
al.,2001)](Table2).

8.0ModelValidation
The simulated values were compared with actual values using bias (Eq.1) and
rootmeansquareoferror(RMSE,Eq.2)assuggestedbyWillmott(1982).Grafet
al. (1991) used the standardised bias (R, Eq.3) and standardised mean square
error(V,Eq.4)asthemostappropriatetocomparesimulatedandactualdataon
maizegrowth,yieldandyieldcomponent.

wherenisthenumberoffieldobservations,SiandAcaresimulatedand
actual values, respectively. R and V are estimates for the overall error of the
model with regards to field data. R quantifies the models ability to reproduce
theactualgrowthpattern.Negativedeviationsfromsimulatedvalues(SiAc<0)
compensateforpositivedeviation(SiAc>0)andviceversa(Eq3).Alternatively,
V is a measure that reveals the models tendency to generally over or
underpredicttheyieldandotherparameters,whileRMSEshowedthereliability
ofthesimulateddata.

9.0ResultsandDiscussion
9.1 Weatherconditions
The weather data during the experimental period at Pak Chong Regional
Meteorological Station located at a distance of about one kilometre from the
experimentalsitearepresentedin(Figures1and2).Theweatherdatashowsolar

116

energy(SE)rangingfrom9to26(Mjm2d1),themaximumtemperature(Tmax)
from 25 to 37C, the minimum temperature (Tmin) from 12 to 24C, relative
humidity(RH)from26to70%,andevapotranspiration(ETc)ofmaizefrom2to
10 (mm/day). There was no rainfall prevailing during the experimental period
fromDecember2006toMarch2007
Figure1: Solar Energy, Mj m2 d1 (SE) and maximum (Tmax) and minimum
temperatureC(Tmin)duringtheexperimentalperiodfromDecember2006to
April2007
Tmin

Tmax

SE
40
35
30
25
20
15
10
5
0
01-Dec-06

31-Dec-06

30-Jan-07

01-Mar-07

31-Mar-07

30-Apr-07

RH (%)

Figure2: Evapotranspiration of the Plots (ETc), Rainfall (RF) and Relative Humidity
(RH)duringtheexperimentalperiodfromDecember2006toApril2007

9.2Soilmoisturecontent
Soilmoisturecontent(SMC)atfieldcapacity(FC)andpermanentwiltingpoint
(PWP) were 65.2 mm and 38 mm, respectively, whereas, the available soil
moisture(ASM)was27.2mmand18.1mm,atFCandPWPrespectivelyat015
cmdepthinalltreatments.Withthesuspensionofirrigationfor10daysduring

117

critical growth periods, the decreasing pattern of SMC on 2day interval is

shown in Figure 3 and 4. Results showed that the SMC in the 015 cm profile
reachednearthelowestlevelofreadilyavailablewater(RAW)insixdays,and
bythetenthdaytheSMCreachedthelowestlevelofRAWintreatmentswhen
irrigation was suspended during 3645, 4655, and 5665 DFS (Figure 3). When
irrigationwassuspendedduring6675DFS,theSMCreachedthelowestlevelof
RAWinsixdaysandinthenext4days(i.e.10thday)itreachedclosetothePWP.
ThepatternofdepletionoftheSMCduringthe10dayperiodofsuspensionof
irrigationatthesamegrowthstageswassimilarinthe1530cmprofiletoo,but
theSMCswerehigherinallplotscomparedto015cmdepth(Figure4).
Figure3:ActualandSimulatedSoilMoistureContent(SMC)during10day
suspensionofirrigationineachtreatmentat015cmdepth

Figure4:ActualandSimulatedSoilMoistureContent(SMC)during10day
suspensionofirrigationineachtreatmentat1530cmdepth

118

CERESMaize model overestimated SMCs at both soil profiles, and bias

(deviation ofsimulated valuefromactual value) was widerfor 1530cm profile
than 015cm profile. The only exception was the SMC in the treatment where
irrigation was suspended during 36 to 45 days from seeding at 015 cm soil
profile,whichhadthenarrowestbias(Figure3).
BothsimulatedandactualSMCsinalltreatmentsindicatethatthecropreached
RAWandwasslightlylowduringthe10dryperiodimposedinthisstudy,but
never reached the PWP. This indicates that there was no severe water stress
experienced by the maize crop. The only treatment that had the highest stress
was those receiving suspension of irrigation towards the latter part of the
growing season like 5665 and 6675 DFS. This greater reduction could be
attributed to increased LAI, which is considered as the evaporative surface of
plants. The results show that under the weather conditions that prevail during
theusualseveredryperiodfromDecembertoAprilinthePakChongDistrictof
Nakhorn Ratchasima Province, a 10day rainless period did not lead to severe
water stress in maize plants to cause significant crop losses. As per current
practice, the irrigation water is applied once a week to meet 40mm for maize
crop, which ensures that SMC would not reach below RAW in the field
condition.Therefore,itwouldbeworthexaminingthecropsperformanceunder
the depletion of water prior to determining the frequency of irrigation and
managementoptionsforalleviatingwaterstresseffectsinmaize.

9.3Growthperformance
9.3.1Timetosilking
There was a difference in the time taken for silking between the two varieties
(Figure5).ForthevarietySuwan4452,timeforsilkingrangedfrom73.5daysin
plots where irrigation was suspended during 5665 DFS to 74.5 days (mean) in
thecontrolplots(Figure.5a).WhereasforPacific224,silkingtimerangedfrom
72.5 days in plots that received a 10day dry period during 4655 DFS to 73.8
daysinthecontrolplots,andtheresthastaken73.0days(Figure.5b).Although
the difference between control and irrigation suspended treatments was not
wide enough, both varieties showed early silking under decreasing moisture
availabilitywhentheirrigationwassuspendedfor10days.
Ontheotherhand,themodelshowednosignificantdifferenceinpredictingthe
numbers of days for silking (Figure 5). For Suwan 4452, the bias ranged from
0.25 to 0.50 days with R values ranging from 0.006 to 0.006, and V ranging

119

0.00005to0.00014.ThedeviationforPacific224wasslightlyhigherwiththebias
ranging from 0.25 days for control plots to 1.50 days when irrigation was
suspended from 4655 DFS, and corresponding R values ranged from 0.003 to
0.021andVrangedfrom0.0003to0.0005.Thisindicatesthatthemodelcouldbe
helpfulinpredictingthetimingofsilkingofmaize.
Figure5:ActualandSimulatedTimetoSilkingofa)Suwan4452andb)Pacific224
VarietiesofMaizeasinfluencedbytimingofsuspensionofirrigation

Table3: LeafAreaIndex,PlantHeightandAboveground
Biomassat50%Silking

Treatment
Timeofwaterstress,DFS1/
3645
4655
5665
6675
Control
Lsd(p=0.05)
Variety
Suwan4452
Pacific224
Lsd(p=0.05)
CV%

LAIatsilking

4.40.4
4.40.4
4.60.4
4.30.4
4.50.4
ns

4.60.4
4.40.4
ns
7.23

1/DFSdaysfromseeding

120

Plantheightat
silking,m

2.320.2
2.240.1
2.280.2
2.300.2
2.360.2
ns

2.260.1b
2.340.2a
0.06
3.88

Biomassat50%
silking

7.40.7
6.80.4
7.30.7
7.50.6
7.60.7
ns

7.30.6
7.30.7
ns
20.69

9.3.2Plantheight
Tendaysuspensionofirrigationhadnosignificantaffectonplantheightwhich
rangedfrom2.24mwhenirrigationwassuspendedduring4655DFSto2.36min
thecontroltreatment(Table3).
However the maize variety hadasignificant effect on plant height, and Suwan
4452varietywassignificantlyshorter(2.26m)thanPacific224variety(2.34m).
Thedifferencewasonly8cmbetweenthetwovarietieswithaCVof3.88percent.
Theheightshadnonoticeablevariationwithineachvariety,exceptbetweenthe
two varieties. There was no simulation applied to predict plant height in
thestudy.
9.3.3Leafareaindex(LAI)
The LAI remained unaffected by both 10day suspension of irrigation and
variety (Appendix Table 1). The LAI ranged from 4.26 with suspension of
irrigationduring6675DFSto4.61during5665DFS,whilethecontrolplotshad
ameanLAIof4.45(Figure6).Thesevaluesvariedslightlyforthetwovarieties:
Suwan 4452 had a LAI ranging from 4.39 to 4.97 (Figure 6a), and Pacific 224
rangedfrom4.11to4.5(Figure6b).
Simulated LAI values were lower than the actual values for all irrigation
treatmentsandalsoforbothvarietieswiththebiasrangingfrom0.39to1.10for
Suwan4452andfrom4.11to4.56forPacific224.ForSuwan4452,thebiaswas
lowerfor3645and6675DFS,andR(0.09)andV(0.01)valuesweresupportive
ofthemodelperformanceandreliability(CV=7.23%)(Figure6a).IncasePacific,
thebiaswaslowestfor6675dayperiodofsuspensionofirrigation(0.11)with
R=0.03andV=0.01(Figure.6b).Whenirrigationwassuspendedduring36to45
days and 56 to 65 days in the two treatments, there was a prediction with
moderatebiaswithR=0.09andV=0.01.ThisshowsthatthesimulationsofLAI
by CERES were acceptable. The results also showed that there was a varietal
differenceinresponsetowatershortageperiods,andPaciificvarietyappearedto
perform slightly better than Suwan 4452 under shortterm water shortage. On
the other hand, LAI values were within the optimum of 4.0 to 5.0 for tropical
maize toensuresatisfactory yields. Duncan(1971) reported thatyieldincreases
ofmaizewereachievedwithaLAIof4.0butnoyieldincreasesoccurredwhen
theLAIincreasedupto4.7.AlthoughhigherLAIswerefoundathigherplant
population,thegrainyielddeclined(Daynardetal.,1971).Therefore,incurrent
study,LAIscouldbeconsideredsufficienttomaintainsatisfactorymaizeyields.

121

The results also indicate that different irrigation treatment did not significantly
affect LAI of all four stages. The simulations too were not outside the range,
althoughnegativevaluesrepresentedanunderestimationoftheLAIinthetwo
maize varieties compared to the actual values. These values provide valid
information to maize researchers and growers for examining the potential
variationinplantgrowthandperformance.
Figure6:ActualandSimulatedLeafAreaIndex(LAI)ofa)Suwan4452and
b)Pacific224VarietiesofMaizeasinfluencedbytimingof
suspensionofirrigation

9.3.4Plantbiomassatsilking
Biomass at 50 per cent silking was neither affected by the irrigation treatments
nor the variety (Appendix Table 1). The biomass ranged from 6.4 t/ha in plots
thatreceivednoirrigationduring3645DFSto7.5t/hainthosethatreceivedno
irrigationduring6675DFS.Thecontrolplotthatreceivedregularirrigationgave
thebiomassof7.3t/ha(Table3).Withrespecttothevarietaleffect,bothSuwan
4452andPacific224had7.3t/haofbiomassatsilking(Table3),whichindicates
thatbothvaritieshavethesamepotentialforbiomassproductionatsilking.
Simulated biomass values at 50 per cent silking were very close to the actual
values, with an exception in control treatment, which received irrigation at
regularintervals(Figure7).Themodelslightlyunderestimatedthebiomassat50
per cent silking compared to the actual values in the control treatment with a
biasof0.80,Rvalueof0.10andVvalueof0.03(Figure7aand7b).Biasofthe
biomass ranged from 0.04 in the treatment received no irrigation during 5665
DFSto0.80inthecontroltreatmentforSuwan4452withRrangingfrom0.05to
0.10 and V from 0.004 to 0.03 (Figure 7a), and for the Pacific 224 from 0.22

122

(overestimated)inthetreatmentreceivednoirrigationduring4655DFSto0.73
inthecontrol(Figure7b).Simulatedbiomassvaluesfortherestofthetreatments
rangedfrom0.04to0.10withRrangingfrom0.07to0.03andVfrom0.002to
0.01.VerylowRandVvaluesindicatethemodelsabilitytocloselyestimatethe
biomass at silking for both varieties of maize. In the meantime, RMSE values
were less than 1.0 for all treatments, except for the control of Suwan variety,
which indicate the reliability of the response. Simulated values, bias, R, and V
indicate that CERES could reasonably estimate plant biomass with the use of
ongoing weather data so that such results could be used as indicators of
performanceoftherainfedmaizecrop.

Figure7:ActualandSimulatedBiomassat50%Silkingofa)Suwan4452and
b) Pacific 224 Varieties of Maize as influenced by timing of
suspensionofirrigation

10.0GrainYieldandYieldComponents
10.1Numberofgrainspercob
Thenumberofgrainspercobwasneithersignificantlyaffectedbytheirrigation
treatments nor the varieties (Appendix Table 2). Number of grains per cob of
Suwan4452rangedfrom577intreatmentsexposedto10daydryperiodduring
3645 DFS to 611 in the control (Figure 8a), while in Pacific 224 grains per cob
rangedfrom572inthetreatmentwhereirrigationwassuspendedduring4655
DFSto600whennoirrigationwasduring5665DFS(Figure8b).Thecontrolhad
alowernumberofgrainsalthoughnotsignificantatp=0.05.Theresultsindicated
thattherewasnoadverseeffectof10daydryperiodatanyofthecriticalstages

123

of maize growth on number of grains per cob. As illustrated by Ritchie and

Harvey(1982),thepotentialkernelnumberisdeterminedbythevegetativestage
V12andcontinueduntilV17,whichcoincideswith3645daysfromseeding.The
periods before two weeks to and after two weeks from silking have also been
identified as critical to reduce the grain number per cob at harvest. These two
periods coincided with 4655 and 5665 DFS. Any stress occurring two weeks
priortosilkingadvancestasseldevelopment,leadstoearlypollensheddingand
atthesametimedelayssilkdevelopment,eventuallydeprivingpollinationand
fertilisation of ovules and developing to kernels (Ritchie and Harvey, 1982). In
theperiodfromsilking(R1stage)uptothedoughstage(R4),grainfillingoccurs,
assimilates mobilises into developing grain, and hence, water stress causes the
formation of unfilled grains. In present study, insignificant differences in the
grainnumberpercobbetweencontrolandstressimposedtreatmentsrevealthat
maize plants in the irrigation suspended treatments for ten days did not affect
kernel initiation, pollination, fertilisation as well as grain filling. Although soil
moistureinboth015cmand1530cmlayersreachedbetweenRAWandPWPfor
abouttwotofourdays(Figure3and4),thetwovarietieswereabletoovercome
the stress effects caused by the absence of irrigation for ten days without
compromisingthegrainnumberpercob,whichisthemostimportantparameter
thatcontributestotheyieldofindividualplants.

Figure8: ActualandSimulatedGrainNumberperCobofa)Suwan4452
and b) Pacific 224 Varieties as influenced by timing of
suspensionofirrigation

However,CERESMaizemodeltendstounderestimatethegrainnumberpercob
underallirrigationtreatmentsincludingthecontrol(Figure8aand8b).Thebias

124

was largeand negative for both varieties, whichindicates underestimation and

its magnitude. The predicted values ranged from 275 grains per cob with the
suspension of irrigation during 6675 DFS to 336 grains per cob in the control
of both varieties, and both R and V values indicated unreliability of the
simulatedvalues.

10.2100grainweight
Similar to the other parameters, 100grain weight remained unchanged due to
irrigation treatments which ranged from 34.0 g with water shortage that
occurredduring3645DFSto35.3gwhenirrigationwassuspendedfrom4655
DFS (Appendix Table 2 and Figure 9). Seeds of Pacific 224 were heavier (36.8
g/100 grains) than Suwan 4452 (32.2 g/100 grains). There was no significant
interactionbetweenthevarietyandtheirrigationtreatmentson100grainweight.
Thehundredgrainweightisdeterminedduringtheperiodfromsilking(R1)to
dough(R4)stages.UponpollinationandfertilisationduringR1,ovulesdevelop
into kernelsand startaccumulating starch(Ritchieand Harvey,1982). This is a
critical stage as water stress interferes with the translocation of assimilates.
Insignificantgrainweightsamongtreatmentsincludingthecontrolindicatethat
maizeplantsdidnothaveanunfavourableimpactfroma10daysuspensionof
water. Findings of the study of Kefele and Ranamukhaarachchi (2004) using
threemaizevarietiesgrowninEthiopiashowedthatvarietaldifferenceexistsfor
grainyield,andthelevelofwaterstressandtheleveloftolerancebythevariety
influence timing of tasseling and silking and its interval, which determine the
timing of pollination and fertilisation and the potential kernel number and the
rateofgrainfilling.
Themodelseemedtooverestimatethe100grainweightforbothvarietiesunder
imposed treatments, which ranged from 39.7g to 41.6g in plots that had
irrigationsuspendedduring4655DFSand3645DFS,respectively,whilecontrol
hadmoderatevalue(41.3g)forSuwan4452and41.4gand43.5g,respectivelyfor
Pacific 224 (Figure 9). The bias ranged from 5.93g to 10.12g for the same
treatments, respectively for Suwan 4452 variety, and 4.46g to 7.20g for the
suspensionofirrigationfrom4655daysand5665days,respectivelyforPacific
224 variety. R values were relatively higher for both varieties at all irrigation
treatments, but V values were lower for Pacific 224 variety. This means,
the reliability of the models ability to reproduce the observed values were
somewhatlower.

125

Figure 9: Actual and Simulated 100grain Weight of a) Suwan 4452 and b)

Pacific224VarietiesofMaizeasinfluencedbytimingofsuspension
ofirrigation

10.3Grainyield
Therewasasignificanteffectofvarietyongrainyieldofmaize,butthedifferent
irrigationtreatmenthadsimilareffectonmaizeyield(AppendixTable2).Grain
yieldwassignificantlyhigherinPacific224(10.4t/ha)thanSuwan4452(9.1t/ha)
Figure10:ActualandSimulatedGrainYieldofa)Suwan4452andb)Pacific
224 Varieties of Maize as influenced by timing of suspension of
irrigation

126

(Figure10).Grainyieldvariedfrom9.5t/hawhenthedryperiodwasimposed
from 3645 days from seeding to 10.1 t/ha with 5665 days from irrigation.
Controlhad9.6t/ha.
For Suwan 4452, the model overestimated the grain yield during all growth
stages tested, except on 6675 DFS, during which the opposite resulted. This
couldbeattributedtosomeunforeseenmoisturestressduringearlygrainfilling.
The bias was however low, except for the 10day dry periods imposed during
3645and6675DFS(Figure10).TheRandVvaluestoohadthesametrend.For
Pacific224,themodelunderestimatedgrainyield,exceptforthecontrol.Butthe
valueswereclosetotheactualandhencethebiaswasinaverynarrowrange.
BothRandVvalueswerealsolow,andthesevaluesindicatethesuitabilityofthe
CERESMaizemodelforsimulatingmaizeyieldsusingconcurrentweatherdata.

11.0HarvestIndex
The harvest index (HI) indicates the fraction of yield out of the total above
groundplantmaterialproduced(Holliday1960).ActualvaluesoftheHIranged
from 0.43 in the control and the treatment received 10day suspension of
irrigation during5665 days from seeding to 0.46for the rest of the treatments,
with a moderate value(0.44) for the treatment received from the suspensionof
irrigationduring3645dayperiodforSuwan4452,andforPacific224from0.52
with suspension of irrigation during 5665 DFS to 0.55 with suspension of
irrigationduring3645DFS(Figure11).ThisshowsthatthePacificvarietyhasa
higherHIthantheSuwanvariety,whichalsosupportsahighergrainproduction
abilityandhighergrainyields.
TheCERESModelestimatedthesameHIasobservedunderfieldconditionsfor
the Suwan 4452, receiving a suspension of irrigation during 4655 days from
seedingwithRMSEof0.03,slightlyunderestimatedforthetreatmentexposedto
10day suspension of irrigation during 6675 days from seeding and
overestimated for the rest of the treatments (Figure 11). The biasforHI ranged
from0.02to0.05withRrangingfrom0.05to0.12and0.007to0.04forSuwan
variety.ForPacific224,themodelunderestimatedHIsandthebiasrangedfrom
0.05 to 0.08, R from 0.09 to 0.15 and V from 0.01 to 0.03. For both varieties
RMSEvalueswerelessthan0.1,whichindicatesthereliabilityofthemodelfor
simulatingHIwithintherangesshowninFigure11.

127

Figure11: Actual and Simulated Harvest Index of a) Suwan 4452 and b)

Pacific 224 Varieties of Maize as influenced by timing of
suspensionofirrigation

12.0Discussion
Thecurrentstudyhadtwoimportantconsiderations.Thefirstisthefateofmaize
cropwhenthedryperiodoccursduringcriticalgrowthperiods.Criticalgrowth
periods have been recognised as the growth stages during which water and
other stresses could bring about unrecoverable crop damage and yield losses
(Eck, 1986; NeSmith and Ritchie, 1992; Zinselmeier et al. 1999; Kefale and
Ranamukhaarachchi,2004;Moseretal.2006).Rainfedmaiseusuallyexperiences
unexpected dry periods extendingup to varying durations. Itisimportant that
the SMC in the root zone does not reach the permanent wilting point (PWP),
thoughitmayreachalevelbelowreadilyavailablewater(RAW)sothatmaize
plantscouldoperatewithactivewateruptakemechanismstoacquirewaterfor
theirsurvivalandgrowth(Lambersetal.1998;Huang2000).Whenwaterstress
developsduringcriticalgrowthstages,itcouldhampersuchdevelopmentsand
eventuallylower the follow up growth andgrain yield through impairment on
yield components (Boonpradub 2000; Kefele and Ranamukhaarachchi 2004;
Panitnok et al. 2005; Moser et al. 2006). The ability of plants to survive under
suchlowwateravailabilitywilldependontherateofwaterreleasebythesoil,
thedepthofrootzoneandtheplantstolerancemechanisms(Huang2000).
TheresultsofthecurrentstudyshowedthattheSMCatboth015cmand1530
cm were above or near RAW after 8 days of withdrawn irrigation, but did not

128

reachPWPeveninthenext4days,withthe10daydryperiodoccurringduring
any stage of maize growth in this study (Figure 3 and 4). However, in the
treatmentthatimposeda10daydryperiodduring6675DFS,theSMCreached
belowRAWandwasyetabovePWPby8daysaftercessationofirrigation.Such
a decline in soil moisture content at this stage could be attributed to its higher
LAI compared to other stages, which provided greater transpiration. On the
otherhand,theperiodof66to76dayscoincideswiththeeffectivegrainfilling
period on maize grown in Nakhorn Ratchasima, and water stress during this
periodscouldinterferewiththemobilityofassimilatestodevelopinggrainsthus
reducinggrainsizeandfinalgrainyield(RichieandHarvey1982;Pandeyetal.
2000;Aschetal.2001).InpresentstudytheLAI,plantheight,biomass,number
ofgrainsperear,100grainweight,finalgrainyieldandtheHIinthetreatments,
whichreceiveda10daysuspensionofirrigationduringcriticalgrowthperiods,
didnotshowsignificantchangescomparedtothecontrol.Itwasalsointeresting
tonotethatNakhornRatchasimaProvinceofThailandhasseverelydryweather
andheavyblowingduringthegrowingperiodofthisstudy,butthisinformation
confirmedthata10daydryperiodcouldbetoleratedbythetwomaizevarieties
withoutsignificantyieldlosses,eventhoughtheSMCdeclinedtobetweenRAW
andPWPduringtheconcernedcriticalgrowthperiods.Previousstudiesofrain
fed maize grown during the dry season without supplementary irrigation in
IndonesiashowedthattheSMCevendidnotreachthePWPatasoildepthof30
cm and maize roots found even below a depth of 70cm, which supported
satisfactorygrowthandyieldofmaizecrop(AhadiyatandRanamukhaarachchi,
2007 and 2008). Grudloyma et al. (2005) also showed satisfactory maize yields
under drought conditions and low N stress. In contrast Moser et al., (2006)
reported that water stress during preanthesis period significantly reduced the
grainsperearand1000kernelweight,thusreducingthegrainyield.However,
MohrandSchopfer(1995)mentionedthattheplantshavespecialadaptationsto
tolerateagainstshorttermstressperiods.Thisisverymuchconnectedwiththe
length of water stress period and the variety for its ability to maintain water
uptake during periods of water shortage (Lafitte 2000; Huang 2000; Kefele and
Ranamukhaarachchi2004).
The second purpose was to examine whether concurrent modelling could be
usedtoeffectivelyguidethegrowersandresearcherstomakesuitabledecisions
and identify suitable practices related to the alleviation of water stress during
shorttermdryperiodsinordertoavoidlargeyieldreductions.Withtheuseof
concurrentweatherandmanagementdata,themodelgaveexcellentestimations
for the number of days to silking, the LAI and biomass at silking, 100grain

129

weight, grain yieldand the HI with low values ofbias,R andV. However,the
model underestimated time to take silking of the Pacific 224 for all treatments,
exceptforthecontrol,100grainweightofSuwan4452andthenumberofgrains
perearofbothSuwan4452andPacific224.Thesevariationscouldbeattributed
to approximated genetic coefficients employed in the model and the actual
ability of the plant to tolerate water stress with its survival mechanisms.
Gungula et al. (2003) used CERESMaize v3.5 to predict the number of days to
silking at high N rate (90 and 120 kg N ha1) in Nigeria and found close
deviations of simulated values from actual performance. However, there has
been some variation in the results of simulations: Asadi and Clemente (2003)
reported the overprediction of grain yield of maize with N treatments. The
currentstudyrevealedthepotentialofusingmodellingapproachforthebenefit
of yield forecasting. The results of the current study confirmed that
CERESMaize model can provide useful information of the effect of already
prevailingdryperiodonthegrowth,developmentandyieldofmaize.

13.0Conclusion
The10daysuspensionofirrigationoccurringduringcriticalgrowthperiodsdid
not adversely affect growth and yield of maize in the Nakhorn Ratchasima
Province.Thishelpstoconservethewaterfromthecurrentpracticeofirrigating
maize crop at weekly intervals. The CERESMaize model in DSSAT V 4.0 too
confirmed these results, except with slight deviations in a few parameters.
Therefore,theapplicationofcropsimulationwiththeCERESMaizemodelusing
concurrent weather data could provide information on the effect that the crop
could havealready had as a result of prevailing shortterm water stress, which
could signal agronomists, researchers and growers to determine and adopt
suitable yield enhancing and protecting management practices for the balance
periodofthecroptomitigatefurtheryieldreductions.

130

References

Ahadiyat,Y.R.andS.L.Ranamukhaarachchi.2007.Effectofsoiltillageandmaizegrass
intercroppingfollowedbygrassmanagementonsoilpropertiesandyieldofrainfed
maize.Int.J.Agri.Biol.,9:791799.
Ahadiyat, Y. R. and S. L. Ranamukhaarachchi. 2008. Effect of tillage and intercropping
withgrassonsoilpropertiesandyieldofrainfedmaize.Int.J.Agri.Biol.,10:133139.
Asadi, M. E. and R. S. Clemente. 2003. Evaluation of CERESMaize of DSSAT model to
simulate nitrate leaching, yield and soil moisture content under tropical conditions.
Food,Agriculture&Environment,1(3&4):270276.
Asch, F., M. N. Andersen, C. R. Jensen, and V. O. Mogensen. 2001. Ovary abscisic acid
concentration does not induce kernel abortion in fieldgrown maize subjected to
drought.EuropeanJournalofAgronomy,15:119129.
Barker,D.E.andN.H.Suhr.1982.Atomicabsorptionandflameemissionspectroscopy.
In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of soil analysis, Part 2,
Chemicalandmicrobiologicalproperties.Pp.1326.AmericanSocietyofAgronomy
Inc.,Madison,WI,USA.
Boonpradub, S. 2000. Drought responses and nitrogen partitioning in maize genotypes
under different soil moisture regimes. Doctoral dissertation No. 633.185912 S 693D,
ChiangMaiUniversity,2000.P.160.ChiangMai:ChiangMaiUniversity.
Bray,R.H.andL.T.Kurtz.1945.Determinationoftotal,organicandavailableformsof
phosphorusinsoils.SoilSci.,59:3945.
Bremner,J.M.andC.S.Mulvaney.1982.Totalnitrogen.In:A.L.Page,R.H.Miller,andD.R.
Keeny,(eds.)Methodsofsoilanalysis,Part2,Chemicalandmicrobiologicalproperties,2nd
edition,Number9intheseries.Pp.595624.AmericanSocietyofAgronomyandSoilScience
SocietyofAmerica,Inc.,Madison,Wisconsin.
Dass, S., M. Kumari, and P. Arora. 2001. Identification of drought enduring and
productivity traits and their use in combination breeding in maize (Zeamays L.). In
SustainablemaizeproductionsystemforNepalproceedingofmaizesymposium.Pp.113116.
Kathmadu,Nepal.
Daynard,T.B.,J.W.TannerandW.G.Duncan.1971.Durationofgrainfillingperiodand
itsrelationtograinyieldincorn,ZeamaysL.,CropScience,11:4548.
Doorenbus,J.andA.H.Kassam.1979.Yieldresponsetowater.FAOIrrigationandDrainage.
PaperNo.33Rome.Italy.
Duncan, W. G. 1971. Leaf angles. Leaf area and canopy photosynthesis. Crop Science,
11:482485.
Eck, H. V. 1986. Effect of water deficits on yield, yield component, and water use
efficiencyofirrigationcorn.AgronomyJournal,78:10351040.
Ekasingh,B.,P.Gypmantasiri,K.ThongngamandP.Grudloyma.2004.MaizeinThailand:
Production Systems, Constraints, and Research Priorities. Retrieved June, 2006, from
http://www.cimmyt.org/english/docs/maize_producsys/thailand.pdf
FAO. 1974. The Euphrates Pilot Irrigation Project. Method of soil analysis, Gadeb Soil
Laboratory(Alaboratorymanual).FoodandAgricultureOrganization,Rome,Italy.

131

Graf, B., M. Dingkuhn, F. Schnier and V. Coronel. 1991. A Simulation Model for the
DynamicsofRiceGrowthandDevelopment:III.ValidationoftheModelwithHigh
YieldingVarieties.AgriculturalSystems,36:329349
Grant,R.F.,B.S.Jackson.J.R.KiniryandG.F.Arkin.1989.Waterdeficittimingeffectson
yieldcomponentinmaize.AgronomyJournal,81:6165.
Gungula, D. T., J. G. Kiling and A. O. Togun. 2003. CERESMaize prediction of maize
phenology under nitrogenstressed condition in Nigeria. Agronomy Journal,
95:892899.
Grudloyma, P., N. Kamlar and S. Prasitwatanaseri. 2005. Performance of promising
tropical late yellow maize hybrids under drought and low nitrogen conditions. In
Proceedings of international conference on maize adaptation marginal environments 25th
anniversaryofthecooperationbetweenKasetsartUniversityandSwissFederalInstituteof
Technology/.Pp.112116.Bangkok:AsksornSiamPrinting.
Holliday,R.1960.Plantpopulationandcropyield.FieldCropAbstracts,13:116.
Huang, B. 2000. Role of root morphological and physiological characteristics in drought
resistanceofplant.InWilkinson,R.E.(Eds),Plantenvironmentinteractions.Pp.3964.
NewYork:MarcelDekker,Inc.
Hoogenboom, G. 2004. Genetic CoefficientsCERESmaize/sorghum/millet. In Hussain,
S.S.andMudasser,M.(Eds.),SouthAsiaregionaltrainingworkshoponCropsimulation
modelingattheMultipleCroppingCenter,ChiangMaiUniversity,Thailand.Pp.255269.
ChiangMai,Thailand:GCISC.
Jampatong, S. and C. Balla. 2005. Improving drought tolerance at flowering in maize:
potential selection tools. In Proceedings of international conference on maize adaptation
marginalenvironments25thanniversaryofthecooperationbetweenKasetsartUniversityand
SwissFederalInstituteofTechnology.Pp.99104.Bangkok:AsksornSiamPrinting.
Kefale, D. and S. L. Ranamukhaarachchi. 2004. Response of maize varieties to drought
stressatdifferentphonologicalstagesinEthiopia.Trop.Sci.J.,44:4449.
Lafitte, H. R. 2000. Abiotic stresses affecting maize. In Tropical maize improvement and
production,Rome:FAOplantproductionandprotectionseriesNo.28.
Lambers, H., F.S. Chapin and T. L. Pons. 1998.Plant water relations. Chapter 3inPlant
physiologicalecology.Pp.154229.NewYork:Springer.
McKee,G.W.1964.Acoefficientforcomputingleafareainhybridcorn.AgronomyJournal,
56:204241.
McLean,E.O.1982.SoilpHandlimerequirement.InPage,A.L.,Miller,R.H.,Keeney,
D.R. (eds.), Methods of soilanalysis, Part 2, Chemical and microbiological properties.
Pp.199224.AmericanSocietyofAgronomyInc.Madison,WI,USA,
Mohr, H. and P. Schopfer. 1995. Physiology of stress resistance, Chapter 32 in Plant
Physiology. (translated by Gudrun and Lawlor, D.W.). Pp. 539566. Publ. Springer
Verlag,Berlin,NewYork.
Moser, S. B., B. Feil, S. Jampatong and P. Stamp. 2006. Effects of preanthesis drought,
nitrogen fertilizer rate, and variety on grain yield, yield components, and harvest
indexoftropicalmaize.AgriculturalWaterManagement,81:4158
NeSmith, D. S. and J. R. Ritchie. 1992. Shortand longterm response of corn to pre
anthesissoilwaterdeficit.AgronomyJournal,84:107113.

132

Nelson,D.W.andL.E.Sommers.1982.Totalcarbon,organiccarbon,andorganicmatter.
In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of soil analysis, Part 2,
Chemical and microbiological properties. Pp. 539579. American Society of
AgronomyInc.,Madison,WI,USA.
Ogoshi,R.M.,Jr.,B.M.Cagauan,andG.Y.Tsuji.1999.Fieldandlaboratorymethodsfor
the collection of the minimum data set. Vol. 48 in Hoogenboom, G., Wilkens, P.W.
andTsuji,G.Y.(Eds.)DSSATversion3vol.4(Pp.218286).Honolulu,Hawaii;Univ.of
Hawaii.
Pandey, R. K., J. W. Maranville and A. Admou1. 2000. Deficit irrigation and nitrogen
effects on maize in a Sahelian environment I. Grain yield and yield components.
AgriculturalWateManagement,46:113.
Panitnok, K., S. Tubngeon, S. Techapinyawat, T. Somwang, S. Limaroon, and N.
Udomprasert, 2005. Effect of water deficit on yield of three maize cultivars. In
Proceedings of international conference on maize adaptation marginal environments 25th
anniversary of the cooperation between Kasetsart University and Swiss Federal Institute of
Technology.Pp.140144.Bangkok:AsksornSiamPrinting.
Ritchie,S.W.andJ.J.Hanway.1982.Howacornplantdevelops.SpecialReportNo.48.P.
21. Iowa State University of Sci. and Technology Cooperative Extension Service,
Ames,Iowa,(RevisededitionFeb.1982),.
Ryan, J., G. Estefan, and A. Rashid, 2001. Soil and plant analysis laboratory manual.
Aleppo,Syria:ICARDA,172pages.
Shaw, R. H. 1977. Climate requirement. In Sparague, G. F. (eds.), Corn and corn
improvement.Pp.591623.Agron.Monograph,Madison:ASA.
Sinclair, T. R. 1994. Limits to crop yield? In Boote, K. J., Bennett, J. M., Sinclair, T. R. &
Paulsen, G. M. (eds.), Physiologyanddeterminationofcropyield. Madison:ASA, CSSA
andSSSA.Pp.509532.
Steel, R. G. D. and J. H. Torrie. 1980. Principles and procedures of statistics, A biometrical
approach.P.633.NewYork:McGrawHillBookCompany.
Thiraporn,R.1996.Maize:Production,use,ProblemanalysisandTechnologytransferto
farmers (in Thai Language). P. 274. publ. Dansuttakhanpim Co. Ltd, Bangkok,
Thailand.
Walkley,A.1947.Acriticalexaminationofrapidmethodfordeterminingorganiccarbon
insoils:Effectsofvariationsindigestionconditionsandoforganicsoilconstituents.
SoilSci.63:251263.
Willmott,C.J.1982.Somecommentsontheevaluationofmodelperformance.Bull.Am.
Meteorol.Soc,63:13091313.
Zinselmeier,C.M.,J.LauerandandJ.S.Boyer.1995.Reversingdroughtinducedlossesin
grainyieldmaintainsembryogrowthinmaize.CropSci.35:13901400.
Zinselmeier,C.,B.R.Jeong,andJ.S.Boyer.1999.Starchandthecontrolofkernelnumber
inmaizeatlowwaterpotentials.PlantPhysiologist,121:2535.

133

Appendix

AppendixTable1: Source of Variation, Degrees of Freedom and Mean

squaresobtained fromanalysis ofvariance forleaf area
index (LAI), plant height and biomass at 50 per cent
silking

Sourceof
variation1

DF

W
Blocks
Error(a)
V
VxW
Error(b)
CV%

4
3
12
1
4
15

LAIatsilking
0.15
0.41*3
0.14
0.36
0.20
0.10
7.23

MS2
Plantheight
158.85
1388.46***
145.05
705.60**
177.0
80.13
3.88

Biomassat
50%silking
8.05
0.53
5.26
0.17
2.46
5.64
10.20

Note: 1WTimeofwaterstress(daysfromseeding);VVariety

2MSMeansumofsquares

3*,**and***indicatethelevelofsignificanceofatp=0.05,0.01and0.001,respectively

Appendix Table 2: Source of Variation, Degrees of Freedom and Mean

Squares obtained from analysis of variance for yield
componentsandyieldandharvestindex(HI)ofmaize

Sourceof
variation1
W
Blocks
Error(a)
V
VxW
Error(b)
CV%

MS2

DF

4
3
12
1
4
15

Grainsperear,
no.
769.81
1134.00
201.62
759.68
430.46
653.57
4.35

100grain
weight
2.26
7.82*3
2.46
215.80***
1.87
1.65
3.73

Grainyield

HI

0.57
3.55**
0.65
17.07***
0.32
0.38
6.33

0.42
3.88*
1.79
33.65***
0.21
0.002
10.15

Note: 1WTimeofwaterstress(daysfromseeding);VVariety

2MSMeansumofsquares

3*,**and***indicatethelevelofsignificanceofatp=0.05,0.01and0.001,respectively

134

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

DecentralisationonAccesstoRuralWaterSupply
Services:AStudyofSixVillagesinWestBengal,India
Indranil De*

Abstract
AsurveyofsixvillagesofBirbhumDistrictofWestBengalinIndiaillustratesinteras
well as intra village inequality in access to water supply. The number of sources per
household is higher in most prosperous villages. Access to water supply is better in
villageswhereliteracyrateishigher.Moreover,educationlevelofthehouseholdalsohas
positive impact on access, considering the intravillage allocation of water supply.
Externalaidforcapitalinvestmentinwatersupplyhasalsobeenfoundtohaveincreased
access.Theredressalofgrievancesregardingwatersupplyisalsobetterinvillageswhere
literacy rate is higher. The study concludes that decentralised delivery mechanism in
provisionofwatersupplyservicesincreasesdisparityinaccessinthepresenceofsocio
economicdisparity.

1.0Introduction
Provision of safe water supply is an integral part of global programme of
poverty alleviation. It not only has direct impact on human development but
alsohasindirectimpactoneconomicgrowth.Thecurrentstateofaccesstowater
supply services in the developing countries is awful. About two of every ten
peopleinthedevelopingworldwerewithoutaccesstosafewaterin2000(World
Bank 2004). Decentralisation in delivery of water supply has been initiated to
improve the access in countries like Ghana, South Africa, Malawi, Uganda,
ZimbabweandEthiopia.Theexperiencesofthesecountrieswiththisrespectare
mixed.WhileinGhanaandSouthAfricadecentralisationhasledtosubstantial
increaseinthecoverage,butinMalawi,Uganda,ZimbabweandEthiopiaithas
made no significant impact (WSP 2004; Slaymaker and Newborne 2004;
Makumbe 1996; Mtisi and Nicol 2003; WSP 2003). In India decentralisation in
delivery of basic services to the rural households has been accentuated by the
73rd Amendment of Indian Constitution, 1993. This amendment has made
provisions for devolving powers and responsibilities upon the local
governments, called as Panchayats or Panchayati Raj Institutions (PRIs), for
implementation of schemes to deliver essential services as drinking water
supply,sanitation,streetlightingandroad.

*BusinessAnalyst,HSBCTechnologyandServices,India.

135

The participation of localgovernment in provision of basic servicesisexpected

toincreaseefficiencyofservicedelivery(Tiebout1956;Oates1972,1977).Thisis
becauselocalgovernmentoperatesmorecloselytopeoplethananyotherhigher
levelofgovernment.Therefore,localgovernmentswouldbeabletoidentifythe
needs and preferences of the communities. Instead of local government, if line
agenciesareentrustedwiththeworkofimplementingprogrammesofprovision
of basic services like water supply and sanitation then it would end up in
primarily engineering solution (Slaymaker and Newborne 2004; WSP 2004).
However, the provision of these services by local government in developing
countries may be hampered due to low capacity of local governments,
corruption, elite capture and political influence (Bardhan 2002; Bardhan and
Mukherjee 2000; Asthana 2003; Slaymaker and Newborne 2004; Mtisi and
Nicol2003).
IncountrieslikeMalawi,UgandaandZimbabwe,ithasalsobeenobservedthat
provision of water services (investment and maintenance) are significantly
greater in already well served area, dominated by powerful politicians than in
poorlyservedareas(SlaymakerandNewborne2004;MtisiandNicol2003).The
existenceofelitecaptureindistributionofwatersupplyhasalsobeenobserved
in India. Water supply sources have been installed in close proximity and
according to convenience of the rural elite (GOM 1998). Besley et. al (2005) has
found that for high spillover public good, residential proximity to elected
representativesmattersinprovision.Localgovernmentshavebeenfoundtobe
moreactiveinthemainvillages,neglectingthehamlets.Therearealsoinstances
thatefficiencyofwatersupplyservicesinIndiaislessinadecentralisedsystem
thaninacentralisedsystem(Asthana2003).Inthisbackdrop,thepresentpaper
hasexaminedthevariationsinaccesstowatersupplyandefficiencyofredressal
by the PRIs with respect to the grievances related to water supply. These
variations may occur due to divergence in their socioeconomic characteristics
andinstitutionalfactors.
WestBengalstatehasbeenselectedforthepresentstudyasithasalonghistory
of decentralisation. It is among a handful of other Indian states, which has
initiateddecentralisationevenbeforethe73rdConstitutionalAmendment.There
are three tiers of local governments in West Bengal. The highest tier is Zilla
Parishad (ZP) at the district level, middle tier is Panchayat Samiti (PS) at the
block level and the lowest tier is Gram Panchyat (GP) at the village level. The
responsibility of rural water supply has been delegated to the Panchayats way
backin1973.ToquotefromtheBengal.Act.XVof1939..thedutiesofaGram

136

Panchayatshallbetoprovidewithintheareaunderitsjurisdictionforsupply
of drinking water and cleansing and disinfecting the sources of supply and
storageofwater.
ThePublicHealthEngineeringDepartment(PHED)isthelineagencyforpublic
provision of rural water supply in West Bengal. The water supply schemes are
delivered through two types of structure. One is the vertical structure, which
consists of the staff of the PHED at various administrative level. The other
structureishorizontalstructure,whichconsistsofGP,PSandZP.Proposalsof
water supply are prepared by Village Committee at the village level. These
proposals are passed from bottom to top through GP, PS, ZP and at last to the
State Government. The schemes are implemented by the PHED through close
coordination with the local governments1. Location of public stand posts and
tubewells are selected by elected representatives of GPs. The operation and
maintenance(O&M)ofwatersupplyschemesmaybedoneeitherbythePHED
or PRIs or by both. The responsibility of O&M of large piped water supply
schemesismainlydevolvedtothePHED.BothPHEDandPRIsareinvolvedin
O&M of small and medium piped water supply schemes. GPs are primarily
responsibleformaintenanceofspotsourcessuchastubewellandhandpump.
Overall, the local governments are intertwined with the line department at
variousadministrativelevelsforprovisionofwatersupply.Nevertheless,GPis
thefirstcontactpointforthepeopletomakeanycomplainregardingproblems
ofaccess,availabilityandqualityofwater.

2.0ObjectivesofStudy
Thegeneralobjectiveofthestudyistoanalysetheimpactofdecentralisationon
watersupplyservicesinsixvillagesofBirbhumDistrictinWestBengal.
Thespecificobjectivesofthestudyareasfollows:
To focus on the role of GP in access to water supply services of the
communities;
To explore the efficiency of the GP in redressal of different kinds of
problemsofwatersupply
Tosuggestpolicyimplicationsarisingoutofthestudy
1

TheWestBengalmodelofdevolutioncouldbeapplaudednotsomuchforthetransferofalarge
number of important specific responsibilities in the rural development arena exclusively to the
PRIs, but for the consultative mechanisms built into the system whereby decisionmaking
powersintheformofadministrativesanctionofthe schemesproposedandtobeimplemented
bythelinedepartmentoftheStategovernmentatthedistrictandsubdistrictlevelsarevestedon
theStandingCommitteeoftheZillaParishads(SubrahmanyamandChoudhury2002).

137

3.0SurveyMethodology
The methodology adopted in the study regarding selection of study area and
studysamplehasbeenillustratedinthissection.Moreover,thissectionhasalso
described the specifications of the models used to explain the determinants of
accesstowatersupplyandredressalofwatersupplyproblems.

3.1Selectionofstudyarea
Birbhum District of West Bengal has been selected as survey area in the study
duetocharacteristicsofruralwatersupplyandsocioeconomicindicatorsofthe
district. The survey has been conducted during OctoberNovember 2006. The
district has been selected as it captures water supply schemes funded by both
Government of India and foreign aid. Moreover, the responsibility of O&M of
pipedwatersupplyhasbeenbornebybothPHEDandthePRIsinthedistrict2.
Moreover,itisoneofthesocioeconomicallybackwarddistrictsofWestBengal.
The socioeconomic characteristics of Birbhum districts visvis that of West
BengalhasbeenpresentedinTable1.
Table1:SocioeconomicCharacteristicsofRuralAreasinBirbhumand
West Bengal in 2001
(Figure in Percentage)

Indicators
Literacyrate
SC&STpopulation
BPLfamilies
Householdshavingnolatrine
Householdsusingtap

Householdsusinghandpump&
tubewell
Householdsusingwell

Birbhum
60.55
38.96
44.02
88.36

WestBengal
64.06
36.16
36.68
73.07

6.58

7.02

80.91

79.97

11.7

11.41

Source:CensusofIndia2001

3.2Selectionofstudysample
Thevillageshavebeenchosenfollowingatwostagestratifiedsamplingmethod.
In the first stage three GPs have been selected from three different blocks
according to the characteristics of water supply systems3. Three different GPs
havebeenselectedaccordingto(1)governmentinvestmentinwatersupplyand
ForotherwatersupplysystemstheresponsibilityofmaintenanceisgenerallywiththePRIsin
WestBengal
3
One GP has been selected from each block since institutional structure of delivery of water
supplyserviceisalmostsamewithinablock.
2

138

O&M of piped water supply by PHED, (2) government investment in water

supplyandO&MofpipedwatersupplybyPRIs,and(3)externalaidforcapital
investment and O&M by both PHED and PS. At the second stage, the most
prosperousvillage(MPV)andtherelativelylessprosperousvillage(RLPV)have
been chosen to compare and contrast delivery of water supply services across
thesevillages(Chart1).Literacyrateofthevillageshasbeenusedasproxyfor
prosperity. Furthermore, within a village, a combination of cluster and
systematicsamplingofhouseholdshavebeenconducted.
Literacyhasalsobeenusedinthestudyasaproxyforcapabilitytobargainand
efficiencyoftheGP.Literacymovementhaspositivespillovereffectsintermsof
enlarging the stake of large numbers of poor people in the system and
strengthening the institutions of local democracy (Bardhan 2002). Education
increases bargaining capability of the poor in intraas well as intervillage
allocation. It also increases the efficiency in the functioning of the GP as it
increasesconsciousnessofthepeopleregardingtherolesandresponsibilitiesof
thelocalgovernment.PRIsbecomemoreaccountabletolocalpeopleifeducation
levelofhouseholdsorliteracyrateofthejurisdictionishigher.
Chart1:TwoStageStratifiedRandomSamplinginBirbhumDistrict

Source:FieldSurvey2006

139

The sample sizes in each village and their socioeconomic characteristics have
been presented in Appendix Table A1 and Table A2 respectively. Politically all
the GPs are dominated by Left Alliance. In Illambazar and Pourandarpur GPs,
the representation of Left Alliance is 100 per cent and in Ruppur GP the Left
Alliancerepresentsmorethan90percentoftheseatsaccordingtoWestBengal
PanchayatElection,2003.

3.3Variablesandspecificationsofmodels
Regressionmodelshavebeenusedinthestudytofindoutthedeterminantsof
accesstowatersupplyandchancesofredressalofwatersupplyproblems.
Different socioeconomic variables, source type, participation in GP meetings
and village specific characteristics have been considered as determining factor/
independent variables in the regression analysis. The socioeconomic variables
considered are education level of the respondent, caste of the household and
level of deprivation of the household. The number of years spend in formal
educationbyanyrespondenthasbeenconsideredasproxyforeducationlevelof
the household. The caste of the household is either Scheduled Caste (SC)/
ScheduledTribe(ST)orgeneral.Totakeintoaccounttheeconomicstatusofthe
households,anindexofdeprivationhasbeenconstructed.Themethodologyof
constructionoftheindexhasbeendiscussedlaterinthissection.Thefunctionof
accesstowatersupplycanbespecifiedas:
W=a+bE+cC+dD+eS+fM+gV+eii=1,2mhouseholds

WhereW=accesstowatersupply

E= educationleveloftherespondent,C=casteofthehousehold,D=economic
deprivationofthehousehold,S=sourceofwatersupply,M=attendanceatGP
meetingsbyanymemberofthehouseholdduringlastoneyear,andV=village
characteristics.
Thefunctionforredressalofproblemhasbeenspecifiedas:
R=a+bE+cC+dD+eM+fP+gV+eii=1,2mhouseholds

WhereR=redressalofproblemandP=typeofwatersupplyproblem

Theeducationleveloftherespondenthasbeendenotedaseducation,casteof
the household has been represented as caste, level of deprivation of the
householdhasbeenindicatedasdeprivation,attendanceofanymemberofthe
householdinGPmeetingduringlastoneyearhasbeenexpressedasmeeting,

140

typeofwatersupplyproblemreportedtotheGPduringlastoneyearhasbeen
denotedbyproblemandthetypeofwatersupplysource(eitherstandpostor
other)hasbeenrepresentedassource..
Itisexpectedthateducationleveloftherespondent,casteanddeprivationofthe
householdareimportantcomponentsofbargainingstrengthforaccesstowater
supply.Itispresumedthataccesstowatersupplyincreaseswiththeeducation
level of the respondent. If the household is from lower caste rather than from
higher caste then access is expected to decline. If deprivation of household
increases then access to water supply is expected to decline. Attendance in GP
meetings by any member of the household is expected to increases the
bargaining strength of the household in getting greater access to water supply.
Accesstowatersupplymayalsodependonthetypeofsource,sinceavailability
timeofwaterforpipedwatersupplyislimited.
ThechancesofredressalofwatersupplyproblemreportedtotheGPisexpected
tobehighifeducationleveloftherespondentincreases,casteofthehouseholdis
higheranddeprivationlevelofthehouseholdislower.Highereducation,caste
and economic status of the household is expected to lead to greater bargaining
power in the GP. It is also expected that if any member of the household has
attended GP meeting during last one year then chances of redressal would be
higherduetogreaterproximityofhouseholdtotheGPofficebearers.Moreover,
it is also expected that if the GP functions efficiently, then chances of redressal
increase. The efficiency of the GP increases with increase in the literacy rate of
theGP.
Totakeintoaccounttheeconomicstatusandlivingconditionofthehouseholds
an index of deprivation has been calculated considering assets owned, type of
house and electrification of the households. Assets owned include land, radio,
TV, telephone, bicycle, scooter and car. If households do not own these assets
then score 1 isassignedand 0 otherwise. If the house is pucca, then household
getsascore0and1otherwise.Ifthehouseholdhaselectricityathomethenthe
householdwillgetascore0and1otherwise.Allthesescoreshavebeenaddedto
construct the index of deprivation. The index ranges from 0 to 9. The index of
deprivationcanbewrittenas:
D=Sii=1,2,9

WhereSi=1,ifdonotownland,radio,TV,telephone,bicycle,scooterandcar
ifhouseisnotpucca
andifdonothaveelectricityathome
0,otherwise

141

The access to water supply has been categorisedinto three different categories,
such as low, medium and high. Access is low if both source is more than 50
metersawayandhavetowaitformorethan15minutestocollectwater;accessif
mediumifeithersourceislessthan50metersawayorhavetowaitlessthan15
minutestocollectwater;andaccessishighifbothsourceislessthan50meters
awayandhavetowaitlessthan15minutestocollectwater.
Themeasureofaccesstowatersupplyisanordinalcategoricaldata.Toestimate
thefunctionsofaccesstowatersupply,anOrderedProbitModelhasbeenused.
The education level of the respondent, caste of the household and level of
deprivation of the household are highly correlated. To avoid multicollinearity
problem between them in regression, three separate models have been
constructed. In Model I, education level of the respondent has been used. In
ModelIIandModelII,casteofthehouseholdandlevelofdeprivationofthe
household have been used respectively. The function of redressal of water
supplyproblemisabinarydiscretechoicemodel.Ithasbeenestimatedthrough
Probitmethod.Inthiscasetoo,multicollinearityproblemhasbeentakencareof
byconstructingthreedifferentmodelsasmentionedearlier.

4.0ResultsandDiscussions
Theresultsarisingoutofthestudyarediscussedbelow:

4.1Accesstowatersupply
Therearebothpublicandprivatesourcesofwatersupplyinthesurveyvillages.
Thepublicwatersupplysourcesareingenerallocatedoutofthepremises.Only
publictapwatersupplyprovideshouseholdconnectioninsidethepremises.The
outsidethepremisespublicsourcesarestandpost,deeptubewell,handpump,
masonry well and other sources such as tank, pond etc. The inhouse water
sourcesareinhousetap,handpump,andwell.Ithasbeenobservedthat78per
cent of the sample households collect water from outside the premises public
watersupplysources(Table2).Percentageofhouseholdscollectingwaterfrom
inhouse tap is 9 per cent and inhouse other sources (handpump/ tubewell/
well)is13percent.
Withintheoutsidethepremisespublicwatersupplysources,40percentofthe
sample households collect water from deep tubewell, 32 per cent from
standpost,and26percentfromhandpump.Therestofthehouseholdscollect
waterfrommasonarywellandothertypeofsystemslikegovernmentreservoir.
PipewaterserviceisnotavailableinRuppur(RLPVofRuppurGP)and

142


Table2:PercentageDistributionofHouseholdsbySourcesofWaterSupply

Illambazar

Ilambazar Baruipur

(MPV)

(RLPV)

Ruppur
Surul

Ruppur

(MPV)

(RLPV)

Total

Total

Tapaipur

(MPV)

(RLPV)

AllGram

Total
l

Panchayat

Total

11.4

6.3

36.7

18

3.4

9.3

Total

22.9

3.4

14.1

10

3.2

6.6

24.2

17.2

12.6

Standpost

78.3

42.9

58.8

52.9

19.1

29.2

15.2

31.9

17.4

35.7

27.5

35.3

93.3

72.3

29.2

9.1

19.6

39.6

17.9

9.8

5.9

6.7

6.4

41.7

90.9

65.2

26.4

5.9

2.1

0.7

Deep
Tubewell
C

Handpump
Masonary
Well
Other

4.3

3.6

3.9

1.4

Total

65.7

96.6

79.7

53.3

96.8

75.4

72.7

88

79.3

78.1

Note:

A=InhouseTapConnection

B=InhouseHandpump/DeepTubewell/MasonaryWell

C=OutofthePremisesPublicWaterSupplySources

Source:FieldSurvey2006

Pourandarpur
Purandarpur

Tapaipur Village (RLPV of Pourandarpur GP). The percentage of households

taking water from standpost is more than 50 per cent in Ilambazar, Baruipur
and Surul Village. Inhouse public tap water source is only provided in Surul
Village.InallotherGPsinhousetapconnectionsareprivate.
Theremainingpartofthestudyhasonlyconsideredoutsidethepremisespublic
water supply sources as the majority of the sample households collect water
from these sources. It has been found that considering all types of sources,
householdspersourceisaround38.Incaseofstandpost,itisalmostthesame.It
is highestincase of tubewellataround41 households per source (Table3). In
Illambazar andRuppurGP,numberofhouseholdspersourceishigherinRLPV
than in MPV. The more prosperous villages are better served presumably
because they have greater bargaining strength in the GP. However, the reverse
holdstrueinPournadarpurGP,wherethenumberof householdspersourceis
lowerintheRLPV(Tapaipur)thaninMPV(Purandarpur).Asthepercentageof
Scheduled Caste (SC) and Scheduled Tribe (ST) population is very high in
Tapaipur Village (Appendix Table A2), therefore distribution of earmarked
resource for SC and ST inhabited areas is also higher4. The average number of
householdspersourceislowestinRuppurGP.Thismay beduetohighcapital
investmentinexternallyaidedprojects.
Table3: AverageNumberofHouseholdsperOutsidethePremises

PublicWaterSupplySource

Illambazar

Ilam Barui
bazar
Standpost

35.56 26.67

Handpump
Other
Allsources

Total

Surul

Rup
pur

Pourandarpur
Total

Puran Tapai
darpur

pur

AllGP

32

26.89

26.89

81.43

67.14

21.67

31.75

29.97

47.86

32.5 44.44 41.39

40

40

50

31

37.33

44.7

27.35 33.13 34.37

20

20

40

34.3 48.39

22

42.04

26

31.7

29.64

56.33

Total

79.5

Deeptubewell 36.25
Masonarywell

pur

Ruppur

81.43 38.52

27.82 42.7

22
38.2

Source:FieldSurvey2006

TheStateorUnionTerritoriesarerequiredtoearmarkandutiliseatleast25percentoftheARWSP
(Accelerated Rural Water Supply Programme) funds for drinking water supply to the SCs and
anotherminimum10percentfortheSTs(GoI2008).

144

4.2Distanceofsource
The distanceof outside the premises public water supplysources is less than
50metersformorethan50percentofthehouseholds.Distanceofsourcesfrom
theresidenceofthehouseholdismoreinRLPVthaninMPV.Incaseofallthe
GPs,percentageofrespondentsreportingdistanceofsourcelessthan50meters
ishigherandmorethan150metersislowerinMPVsthaninRLPVs(Table4).It
maybe due to relative abundanceofwatersupplysourcesinMPV as compared
toRLPV.TherelativeabundanceofwatersupplysourcesinMPVascomparedto
RLPVcanbeexplainedbygreaterbargainingcapabilityofresidentsofMPVsfor
resources as compared to RLPVs in the GP. As MPVs have higher literacy rate
than RLPVs, therefore their bargaining capability may be higher in the MPVs
thaninRLPVs.
Table4: PercentageDistributionofHouseholdsbyDistanceof

OutsidethePremisesPublicWaterSupplySources

Illambazar

Ilam Barui

<50m

bazar

pur

56.5

53.6

Ruppur

Pourandarpur
Puran Tapaipu

Total

Surul

Ruppur

Total

darpur

54.9

64.7

40

48.9

70.8

36.4

Total
54.3

AllGP
52.75

50100m 34.8

7.1

19.6

11.8

26.7

21.3

20.8

31.8

26.1

22.23

100150m 4.3

28.6

17.6

17.6

13.3

14.9

4.2

18.2

10.9

14.58

150300m 4.3

7.1

5.9

5.9

6.7

6.4

4.2

13.6

8.7

6.96

300500m

3.6

13.3

8.5

3.48

100

100

100

100

100

100

100

100

100

100.00

Total

Source:FieldSurvey2006

4.3.Waitingtimetocollectwater
Ithasbeenobserved thatmorethan30per centofthehouseholds inthe sample
have to wait for more than 30 minutes for water from outside the premises
public water supply source. The waiting time to collect water is found to be
generally larger in MPV than in RLPV of a GP taking all type of sources into
consideration. Lower percentage of households has waited more than half an
hourinRLPVthaninMPVinalltheGP(TableA.3inAppendix).Thetimespent
tocollectwaterdependsontheavailabilitytimeofwatersupplyandnumberof
householdspersource.Thewaitingtimeisingenerallargerincaseoftapwater
supply compared to other sources. Amongst the households collecting water

145

fromstandpost,thewaitingtimeformorethan60percentofthemismorethan
halfanhour.Largerwaitingtimeforstandpostmaybeattributabletolimited
availability time of water supply. For most of the households in Baruipur
availability time is 35 hours/day; in Surul it is 13 hours/day, and in
Purandarpuritis35hours/day.Duetolessernumberofhouseholdspersource
inRuppurGP,thepercentageofhouseholdswhodonothavetowaitishighest.

5.0RedressalofGrievances
The households report to the local government about the problems of water
supplyservice.Duringlastoneyear,44percentofthehouseholdshavereported
anyproblemregardingwatersupplytotheGP.Amongthesereportedproblems,
39percenthavebeenredressed(Table5).InRuppurandPourandarpurGP,the
percentageofgrievancesredressedismuchhigherinMPVthaninRLPV.Onthe
other hand, it is higher in RLPV than in MPV in Illambazar GP. The type of
problems reported differs from one village to another. Moreover, redressal of
grievanceswouldalsodependonthetypeofproblem.
Table5: Percentage of Households Reported Water Supply Problems and
PercentageofProblemRedressed

Illambazar

Ruppur

Pourandarpur

Ilambazar Baruipur Total Surul Ruppur Total

AllGP

Purandarpur TapaipurTotal

Reported

51.4

55.2

53.1 26.7

45.2

36.1

51.5

32

43.1

44.26

Redressed

22.2

41.2

31.4 62.5

28.6

40.9

52.9

25

44

38.56

Source:FieldSurvey2006

The types of water problem reported have been classified in three different
categories. They are (a) nonavailability of water from the source or the source
requiresrepairing,(b)qualityofwateravailableisdissatisfactoryand(c)lackof
adequatewatersupplysources.Thefirstandsecondtypesofproblemarisedue
tolackofmaintenance.Thethirdtypeofproblemoccursduetolackofadequate
provisionofwatersupplysources.InIllambazarGP,80percentoftheproblems
reported are regarding maintenance (Table 6). Around 29 per cent of the
problems are against quality of water in Illambazar GP. On the contrary, in
Ruppur and Pourandarpur GP, most of the problems are regarding access to
watersupplyduetoinadequateprovision.InallthethreeGPs,householdshave
approachedtheGPmorefrequentlyinRLPVthaninMPVforsettingupofnew
source.Itisjusttheoppositeincaseofmaintenance.

146

Table6: Percentage Distribution of Households by Type of Water Supply

ProblemReportedinGramPanchayat

Illambazar

Ruppur

Problem Ilambazar Baruipur Total

NV

Pourandarpur

Surul Ruppur Total

All

Purandarpur TapaipurTotal

GP

50

52.9

51.4

75

28.6

45.5

41.2

37.5

40

45.82

38.9

17.6

28.6

10.00

11.1

29.4

20

25

71.4

54.5

58.8

62.5

60

44.18

Total

100

100

100

100

100

100

100

100

100

100

Note: NV=Nonavailabilityofwaterfromthesource/sourcerequiresrepairing,

Q=Qualityofwateravailableisdissatisfactory

L=Lackofadequatewatersupplysources
Source:FieldSurvey2006

Theefficiencyoflocalgovernmentinredressalofgrievancescanbeexpressedas
water supply problem redressed as a percentage of problems reported. The
problemsregardingnonavailabilityofwaterandrepairinghavebeenredressed
in much higher percentage (68%) of cases than the problems relating toquality
(10%)andlackofadequatewatersupplysources(12%).Thismaybeduetolack
of capacity of the GP to resolve quality problems. Moreover, GP is not directly
responsibleforinstallationofnewsourcesofwatersupply.Timetakentoredress
these two problems is also longer than that of nonavailability of water and
repairing.WithintheGPs,the problemsregarding nonavailabilityofwater and
repairinghavebeenredressedin90percentofthecasesinPourandarpurGP,80
per cent of the cases inRuppurand only50 per cent of the cases in Illambazar
GP(Table7).Inaddition,theproblemsregardinginstallationofnewsourcehave
beenredressedin13percentofcasesinPourandarpur,8percentofthecasesin
Ruppur GP and 14 per cent of the cases in Illambazar GP. Taking into account
redressal of both these problems, the evidences suggest that efficiency of
redressal of grievances is best in Pourandarpur GP. This might have happened
as Pourandarpur GP has highest literacy rate among all the GPs. As discussed
earlier,literacystrengthenslocaldemocracyandincreasesefficiencyoftheGP.

147

Table7:PercentageofWaterSupplyProblemRedressedbyTypeofProblem

Illambazar

Problem Ilambazar

Ruppur

Pourandarpur

Baruipur Total Surul Ruppur Total Purandarpur Tapaipur Total

AllGP

NV

33.3

66.7

50

66.7

100

80

100

66.7

90

68.4

14.3

10

10.0

20

14.3

50

8.3

20

13.3

11.8

Note: NV=Nonavailabilityofwaterfromthesource/sourcerequiresrepairing,

Q=Qualityofwateravailableisunsatisfactory

L=Lackofadequatewatersupplysources
Source:FieldSurvey2006

6.0 RegressionResults
Inthissection,thedeterminantsofaccesstowatersupplyandredressalofwater
supplyproblemsbytheGPhasbeenanalysed.

6.1Accesstowatersupply
The maximum likelihood estimates of determinants of access to outside the
premises public water supply source have been illustrated in Table 8. The
estimates reveal that in Model I, education turns out to be significant and
positive.Inallthethreemodelsthesignificantvariablesaresourceandvillage
dummiesofSurul,RuppurandPurandarpur.Thesignofsourceisnegativeand
thesignofvillagedummiesarepositive.Allthemodelsarewellfitted.
The results imply that access to water supply increases with higher education
level of the respondent. This is presumably because higher education increases
bargainingstrengthofthepeopleintheGramPanchayatforsettingupsources
neartheirpremises.Ithasalsobeenfoundthatifthesourceisstandpostrather
thananyothersource,thenaccesstowatersupplydeclines.Moreover,accessto
water supply is higher in Surul, Rupur and Purandarpur as compared to
Ilambazar Village. Access is better in Surul and Ruppur, may be due to higher
capital investment in these villages due to foreign aid. Access is better in
PurandarpurascomparedtoIlambazarmaybeduetohigherliteracyrateofthe
village.Moreover,PourandarpurGPalsohasaveryhighliteracyrate,whichhas
probably increased the bargaining strength of the GP in the upper level of
governmentforfundstoinvestinwatersupply.Anindividualmaybeilliterate
butwouldgetgreateraccessinPurandarpurduetohighercollectivebargaining
strengthofthevillageaswellastheGP.

148

Table8:MaximumLikelihoodEstimatesofDeterminantsof
AccesstoWaterSupply

Variables

ModelI

ModelII

ModelIII

0.04*
(1.78)

Caste(SC/ST=1.Others=0)

0.01
(0.03)

Deprivation

0.85***
(3.18)
0.31
(1.49)
0.44
(1.24)
1.81***
(4.33)
1.38***
(3.33)
0.79**
(2.12)
0.49
(1.13)

0.85***
(3.17)
0.22
(1.10)
0.32
(0.91)
1.69***
(4.08)
1.21***
(2.95)
0.75**
(1.97)
0.24
(0.57)

0.40
(0.73)
0.83***
(3.11)
0.24
(1.18)
0.40
(1.08)
1.72***
(4.19)
1.28***
(3.10)
0.78**
(2.08)
0.33
(0.78)

0.14

0.24

0.43

Education

Source(standpost=1,others=0)
Meeting(attend=1,donotattend=0)
Baruipur
Surul
Ruppur
Purandarpur
Tapaipur
Cutpoint1
Cutpoint2

1.41

1.01

0.83

129.13

130.74

130.47

Chisquare

55.92

52.71

53.25

Prob>chi2

0.00

0.00

0.00

Numberofobservations

144

144

144

Loglikelihood

Note: Numbersintheparenthesisarezstatistics

*Significantat10%Level,**Significantat5%Leveland***Significantat1%Level

6.2 Redressalofwatersupplyproblem
The maximum likelihood estimates of determinants of redressal of grievance
havebeenillustratedinTable9.Theestimatesrevealthatinallthreemodelsthe
variables problem and village dummy of Purandarpur are significant. The
coefficientsofthesetwovariablesarepositive.

149

Table9:MaximumLikelihoodEstimatesofDeterminantsofRedressalof
GrievanceRegardingWaterSupplyProblem

Variables

ModelI
0.01#
(0.86)

Education

ModelII

ModelIII

Caste(SC/ST=1.Others=0)

0.08
(0.64)

Deprivation

Meeting(attend=1,donotattend=0)

0.03
(0.30)

0.04
(0.30)

7.72e06
(0.00)
0.03
(0.25)

Problem(Nonavailabilityofwaterfrom
source/sourcerequiresrepairing=1,
others=0)

0.60***
(4.84)

0.60***
(4.86)

0.63***
(5.09)

0.15
(0.71)
0.37
(1.61)
0.17
(0.71)
0.47**
(2.29)
0.01
(0.04)

0.22
(1.14)
0.34
(1.46)
0.20
(0.87)
0.47**
(2.26)
0.07
(0.28)

0.23
(1.10)
0.35
(1.51)
0.25
(1.03)
0.51**
(2.47)
0.25
(0.85)

Loglikelihood

34.91

35.09

34.31

Chisquare

38.92

38.57

39.17

Prob>chi2

0.00

0.00

0.00

82

82

82

Baruipur
Surul
Ruppur
Purandarpur
Tapaipur

Numberofobservations
Note: Numbersintheparenthesisarezstatistics

#Marginaleffect,notthecoefficients

*Significantat10%Level,**Significantat5%Leveland***Significantat1%Level

The results illustrate that probability of redressal of grievance increases by at

least 60 per cent if the problem is nonavailability of water from sources or
sources requireing repairing. It has also been observed that probability of
redressal increases by at least 47 per cent if the problem is reported from
Purandarpur village rather than from Ilambazar Village (Table 9). This may be
duetomuchefficientfunctioningofGPinPourandarpurforredressalofwater
supplyproblemsastheliteracyrateishighestintheGP.

150

7.0 SummaryandConclusion
Accesstowatersupplyhasbeenmeasuredbydistanceofthesourceandwaiting
time to collect water. Distance of outside the premises public water supply
sourcefromtheresidenceofthehouseholdishigherinRLPVthaninMPV.This
may be due to higher bargaining capability of the residents of MPV than the
RLPVforresourcesintheGP.Waitingtimetocollectwaterisgenerallyhigherin
MPVthaninRLPVofaGP.Itmaybeduetolimitedavailabilitytimeof water
supplyfromstandpost,whichismainlyprevalentinMPV. WithintheGPs,the
number of households per source is lowest in Ruppur GP. This may be due to
higherpercapitacapitalinvestmentwith thehelpofforeign aid.Accessis also
higherinPurandarpurVillage,whichhasthehighestliteracyrateamongall the
villages.ThehighliteracyratesofPurandarpur Village aswell asPourandarpur
GPhaveprobablyincreasedbargainingstrengthofthevillage intheupperlevel
of government for funds to invest in water supply. In view of interhousehold
access to water supply sources in the villages, access to water supply of the
household has also increased with education level. In a nutshell, the study
demonstrates that education has impact on both interand intrajurisdictional
allocation.
The problems of access and quality of water supply are reported to the Gram
Panchayat for redressal. It has been observed that around 44 per cent of the
households have approached GP for redressal of grievance regarding water
supply. Among these, 39 per cent of the grievances have been redressed. The
grievances reported to the GP are mostly related to the lack of adequate water
supplysourcesinRLPVandregardingmaintenanceinMPV. Ithasbeenfound
that 68 per cent of the grievances related to nonavailability of water from the
sources or sources requireing repairing have been redressed. The redressals of
othertypesofproblemshavebeensubstantiallyless.ThismaybesincetheGPis
more directlyresponsibletosolvethefirsttypeofproblemascomparedtoother
problems.TheefficiencyofGPsinredressalhasbeenobservedtobesignificantly
higherinPurandarpurVillage,presumablyduetohighliteracyrateoftheGP.
Thestudyrevealsthataccesstowatersupplyservicesandefficiencyofthelocal
government in redressal of grievances regarding water supply have significant
relationwitheducationlevelofthecommunities.Moreover,numberofsources
per household is higher in the most prosperous villages. The prevalence of
socioeconomic disparity in rural India has led to such inequity in provision of
water supply services. Thus the evidences suggest that decentralised delivery
mechanism in water supply would increase disparity in access where
socioeconomicinequityiswellpervasive.Centralinterventioninplanningand

151

allocationatthevillagelevelwouldprobablyreducetheinequity.Thisisevident
from higher provision of water supply sources in a relatively less prosperous
villageofPourandarpurGP.Furthermore,externalaidcouldalsobesupportive
forprovisionincaseofpaucityofgovernmentgrants.

References
Asthana,AnandN.2003.DecentralisationandSupplyEfficiency:TheCaseofRuralWater
Supply in Central India. The Journal of Development Studies, Vol. 39, No. 4,
Pp.148159.
Bardhan, Pranab. 2002. Decentralization of Governance and Development. Journal of
EconomicPerspectives,16(4),185206.
Bardhan, Pranab, and Dilip Mookherjee. 2000. Capture and Governnence at Local and
NationalLevels.AmericanEconomicReview,90(2),135139.
Besley, Tim, Rohini Pande, Lupin Rahman and Vijayendra Rao. 2005. The Politics of
Public Good Provision: Evidence from Indian Local Government, Journal of the
EuropeanEconomicAssociation,Volume223,2004.
GOI.2008.AnnualReport200708.MinistryofRuralDevelopment
GOM, Water Supply and Sanitation Department. 1998. India Second Maharastra Rural
WaterSupplySanitationProject,Tribal(IndigenousPeople)DevelopmentPlan,No.
IPP47.
Makumbe, J. 1996. ParticipatoryDevelopment:TheCaseofZimbabwe. Harare: University of
ZimbabwePublications.
Mtisi, Sobona and Alan Nicol. 2003. Water Points and Water Policies: Decentralisation
and Community Management in Sangwe Communal Area, Zimbabwe. Sustainable
LivelihoodinSouthernAfrica,ResearchPaper15.
Oates,Wallace.1972.FiscalFederalism.NewYork:HarcourtBraceJovanovich.
Oates, Wallace. 1977. An Economists perspective on fiscal federalism. In The Political
EconomyofFiscalFederalism,ed.OatesLexington:LexingtonBooks.
Slaymaker, Tom and Peter Newborne. 2004. Implication of Water Supply & Sanitation
ProgrammesunderPRSPs,ODI,WaterPolicyProgramme.
Subrahmanyam,K.SivaandR.C.Choudhury.2002.FunctionalandFinancialDevolutionon
PanchayatsinIndia.NationalInstituteofRuralDevelopment.
Tiebout,C.1956.Apuretheoryoflocalexpenditures.JournalofPoliticalEconomy,Vol.64,
P.416424.
WorldBank.2004.WorldDevelopmentReport2004:MakingServicesWorkforPoorPeople.
WSP.2003.GovernanceandFinancingofWaterSupplyandSanitationinEthiopia,Kenya
andSouthAfrica:ACrossCountrySynthesis,SectorFinanceWorkingPapers:No.5,
September.
WSP. 2004. Rural Water Supply and Sanitation in Africa: Global Learning Process on
ScalingUpPovertyReduction.ShanghaiConference,May2527.

152

Appendix
TableA1:SizeofSampleDrawnfromDifferentVillages

GP

GP

GP

Illambazar

Ruppur

Pourandarpur

Total

Ilambazar

35

35

Baruipur

29

29

30

30

Ruppur

31

31

Purandarpur

33

33

Surul

Tapaipur
Total

25

25

64

61

58

183

Source:FieldSurvey2006

TableA2:PercentageofLiterate,ScheduledCaste(SC)andScheduledTribe
(ST)inGramPanchayats

Literate

RuppurGP

PourandarpurGP

Illambazar

55.72

80.20

55.00

SC

26.31

44.91

26.73

ST

21.81

16.17

19.32

SC&ST

48.12

61.08

46.05

Surul

Purandarpur

Ilambazar

Literate

61.38

81.40

62.90

SC

27.53

48.27

33.44

ST

10.67

6.24

6.18

SC&ST

38.20

54.51

39.62

Ruppur

Tapaipur

Baruipur

46.03

55.00

58.39

Literate
SC

21.00

10.47

13.36

ST

21.24

73.45

0.00

SC&ST

42.23

83.91

13.36

Source:BolpurSriniketan,SuriIIandIllambazarPanchayatSamiti

153

TableA3: PercentageDistributionofHouseholdsbyWaitingTimetoCollect
WaterfromOutsidethePremisesPublicWaterSupplySource

Illambazar

Ruppur
Pourandarpur
AllGP
Rup
Puran
Ilambazar Baruipur Total Surul
Tapaipur Total
Total
darpur
pur
A
0
8.3
3.3
11.1

11.1
0

0
4.32
B
0
25
10
11.1

11.1 14.3

14.3 10.87
C
11.1
8.3
10
11.1

11.1
0

0
8.69
D
16.7
16.7
16.7 22.2

22.2
0

0
15.23
E
44.4
33.3
40
11.1

11.1 28.6

28.6 32.61
F
27.8
8.3
20
33.3

33.3 57.1

57.1 28.25
Total
100
100
100
100

100 100

100
100
A
50
0
14.3 16.7 28.6 26.5
0
0
0
19.32
B
0
10
7.1
33.3 14.3 17.6 42.9
50
44.4 19.25
C
0
10
7.1
50 35.7 38.2 57.1
50
55.6 33.31
D
0
40
28.6
0
3.6 2.9
0
0
0
8.75
E
25
0
7.1
0
0
0
0
0
0
1.74
F
25
40
35.7
0
17.9 14.7
0
0
0
17.54
Total
100
100
100
100 100 100 100
100
100
100
A
0
20
20
0
0
0
0
10
6.7
7.92
B
0
0
0
0
50 33.3
20
35
30
26.31
C
0
40
40
100 50 66.7
50
35
40
42.11
D
0
20
20
0
0
0
10
10
10
10.53
E
0
0
0
0
0
0
10
5
6.7
5.29
F
0
20
20
0
0
0
10
5
6.7
7.92
Total
0
100
100
100 100 100 100
100
100
100
A

100
0
100

100

Stand
post

Deep
Tube
well

Hand
pump

Maso
naryWell
Other

C
F
Total

0
100
100

100
0
100

50
50
100

50
50
100

All
Sources

A
B
C
D
E
F
Total

8.7
0
8.7
13
39.1
30.4
100

7.1
14.3
17.9
25
14.3
21.4
100

7.8
7.8
13.7
19.6
25.5
25.5
100

17.6
17.6
29.4
11.8
5.9
17.6
100

26.7
16.7
36.7
3.3
0
16.7
100

23.4
17
34
6.4
2.1
17
100

0
20.8
41.7
4.2
12.5
20.8
100

9.1
36.4
36.4
9.1
4.5
4.5
100

4.3
28.3
39.1
6.5
8.7
13
100

11.8
17.4
28.4
11.1
12.5
18.7
100

Note: A=Donothavetowait,B=Lessthan5minutes,C=515minutes,D=1530minutes,E=30
minutes1hour,F=Morethan1hour
Source:FieldSurvey2006

154

AsiaPacificJournalofRuralDevelopment
Vol.XIX,No.1,July2009

CarpSeedTransportinBangladesh
MahmudHasan*

Abstract
Theaimofthisstudywastoexplorethecurrentpracticesofcarpseedtransportmethods,
types of traders and markets, loading density and duration, methods of confinement,
mediums used in transport, chemicals/drugs used during transport and trading in
Bangladesh. Data have been drawn through structured and semistructured surveys.
Thiswouldprovidethebasisforfurtherdevelopmentofcarpseedtransportmethodsand
tradingsystemsinthecountry.

1.0Introduction
Fish fry/fingerling traders use their Indigenous Technical Knowledge (ITK) to
transportcarpseedbetweenandwithinregionsofthecountrytothefishfarmers
bylandtransport.Tradersloadcarpseedasmuchastheycanperunitvolumeof
water.Theyusedifferenttransportmediumstotransportfishseed,forexample,
pond,wellandamixtureofthetwo.Typicallythetradersagitatethetransport
mediums during transport by hand for oxygenation. Time required for selling
carpseedinthemarketsdependsontheavailabilityofthetraders.
Hand agitation, means of transport, transport mediums, loading density and
durationanddistancecovertotransportcarpseedvarybetweenregionswithin
the country. The transport techniques and measures taken by the traders for
agitation and conditioning of the fry/fingerlings are also different between
regions.Thesedifferenthandlingtechniquesmayleadtophysicalinjury,water
qualityandhealthproblemsmediatingstressandhighmortality.HasanandBart
(2006)estimatedtheloss(6200mt)resultingfromtransportmortality.
While Lewis et al. (1996) described the silver seed trading in the Northwest
Bangladesh, Chowdhury (1997) reported the effects of transport on the
production of rohu, Labeo rohita fingerling after train transport from Jessore to
Parbotipur.Theeffectsofvigorousagitationonthephysicalinjurysuchasscale
loss,injuryandfinbifurcationhasbeendemonstratedbyAhmad(1954).Nabiet
al. (1983) has described the effects of transport medium on the mortality of

Associate Professor, Department of Fisheries, University of Dhaka, Bangladesh. Email:

mhasan@univdhaka.edu

155

mixedcarpfry.Hefoundthatwhileriverwaterresultedin24.005.96percent
overall mortality of mixed carp fry in a simulation laboratory experiment, the
well and pond water had 11.08 3.91 and 15.50 4.95 per cent. Samson and
Macintosh (1986) observed 6.45 per cent mortality in unconditioned silver carp
fry(0.25g)usingwellwaterand15percentusingpondwater3hafterclosed
transport with compressed oxygen. However,informationrelating to carp seed
transportbythesmallscalefry/fingerlingtradersislacking.Thusthereisaneed
to explore the baseline information of carp seed transport related activities to
determinetheproblemsassociatedwiththispartofaquacultureinBangladesh.
However, baseline information across the country related to carp fingerling
transporthasneverbeenreported.
Theoverallobjectiveofthisstudywastoreviewthecurrentcarpseedtransport
methodinBangladesh.Thestudyalsoattemptedtoidentifythecurrentloading
density, transport mediums used to transport fishseed, transport duration and
distance,andtransportmortality;todeterminewhichfishisthemostsensitiveto
transport mortality as well as some of the common practices being used to
reducestressandimprovesurvivalduringandaftertransport.

2.0StudyMethod
Purposive sampling technique was used to collect data using semistructured
survey and structured HH survey as the primary tools. This study was
undertakentoidentifythemajorissuesrelatingtocarpseedtransportsituation
inBangladesh.Some13carpseedmarketsweresurveyedbasedontheintensity
of fish seed trading activities. The survey covered Jessore, Zhenaidaha, Natore,
Bogra, Dinajpur, Rajshahi, Mymensingh, Laxmipur and Comilla districts. Carp
seedtraders,nursersandfishfarmersweresampled.

2.1Landclassification
The sampled HH were classified into six land classes following IFAD (2002) to
denotetheirsocioeconomicstatusinwhichthebasiswasland.Landowned(ha)
=cultivableland+pond/gher+fallowlandexcludingthehomestead;totalland
owned (ha) = land owned + homestead land. The land classes are: absolute
landless=nocultivableland(excludinghomestead);landless=landowned0
0.2;marginal=landowned>0.20.6;small=>0.61.0;medium=>1.02.0;and
large=>2.0ha.

156

2.2Dataanalysis
Data generated through semistructured survey were manually analysed while
structuredHHsurveydataanalysedusingstatisticalsoftwareSPSSversion10.0.
Onlythedescriptivestatisticswasusedtopresentthedataasweightedmean.

3.0FindingsofSurveys
3.1Carpseedtrading
Although carp seed include several stages of life history, here in the present
studyonlythetermfingerlingisusedtomeanfishseed.Carpseedtradingisa
largeandanimportantcomponentofaquacultureindustryinBangladeshsince
farming practices initiate from this input. Usually the large hatchery owners
producehatchlingsupongettingorderfromthefingerling(seed)growers.Small
hatcheryownersproducebothfryandfingerlings.Thenursersraisefingerlingin
earthenpondandselltotheseedtradersviamiddlemeninthefishseedmarkets.
The middlemen transport the carp seed over 0.52 h and confine into hapa in
holding pond water before sale. The traders transport and distribute carp seed
acrossthecountrytotheendusers.
Indianmajorcarpse.g.,catla,rohuandmrigal,Cirrhinuscirrhosuswerefoundto
beinhighestdemandwith85percentoftherespondentsoffoodfishfarmers.
Thiscouldbeduetopeoplesfamiliaritywiththeindigenousfishes.Theexotic
carpse.g.,silvercarp,grasscarp,Ctenopharyngodonidella,commoncarp,Cyprinus
carpioandsilverbarb,Puntinusgonionotushadlowerdemand.

3.2Carpseedtransport
Fertilisedeggs,fryandfingerlingareusuallytransportedanddistributedacross
the country. Fry is transported in plastic bag filled with well water and
compressed oxygen for safe transport while fingerling are transported in open
aluminium vessel (2060 L) and metal/plastic drum (200 L) by hand agitation
(water splashing). This traditional transport of fingerling often leads to
total/massmortality.Thisincreasedthetradersriskleadingtothelossofcapital
for several times a season. Hasan and Bart (2006) have been estimated that the
annual transport mortality resulting from mass and delayed mortality of carps
seedtobe6200mtinBangladesh.
AluminiumtransportvesselsaretraditionallycalledPatilorHari(Lewisetal.,
1996).LargePatils(4060L)areusuallyusedforlonghaulandsmallPatils(2030
L) for short hauls. Three different distribution channels were obvious in this

157

survey(Figure1).Inthefirstchannel,thehatcheryproducedhatchlingsaresold
tothefingerlinggrowers.Thegrowersagainselltotheseedtraderswhofinally
selltothegrowoutfarmers.Thesecondchannelincludesproductionoffryand
fingerlings at the hatchery which is sold to the farmers or traders without
involvingmarket.Inthethirdchannel,thefryaresoldtothenursers/fingerling
growers.Theyinterngrowthemtofingerlingsizeandselltothetradersinthe
carpseedmarkets.Inthischannelthereisnofarmgateselling.

Figure1: CarpSeedDistributionChannelsSurveyedinCarpSeedMarketsof
Bangladesh

Fingerlinggrowers

Farmgateselling

Nursery

Traders/farmers

Grow-out

sites
Hatchery

Middlemen

Traders/farmers
Market

Nursers

Marketselling

Traders/farmers

3.3Typesoftraders
Carpseedtraderscanbeclassifiedintotwotypesbasedonthequantityofseed
theytransport,i)Mohajonorlargescaletraders,theytransport>30kgseedwith
thehelpofhireddaylabour;andii)PaikersorPatiloalaaresmallscaletraders.
Thepaikerscarryonepairofpatilshungfromabamboostickwiththehelpofa
gearmadeofjutefibresonthetradersshoulders.Mohajanssellcarpseedasfar
as500kmfromthesourcerequiringasmuchas1012h,forexample,fromBogra
to Lalmonirhat or from Bogra or Natore District through Parbotipur railway
stationtoUlipurSubdistrictofKurigramDistricttakesapproximately15h.The
numberofintermediariestendedtoincreasewithincreasingmarketchannelsfor
asinglehaul.

158

3.4Carpseedmarkets
Carpseedtradingisanoldenterpriseassometraderswerefoundtradingcarp
seedoverlast40years.Inthepast,theytransportedseedfromthewildstockto
the farmers using earthen vessel or boat. However, with the introduction of
artificial breeding technique in the 80s, the private hatchery industries have
expanded as a profitable business. Consequently rather dynamic markets have
developed.Aftersurveyingmorethan20districtsandsubdistrictsaswellasa
number of large market areas, Jessore was found to be the biggest carp seed
market in the country based on the number of holding ponds. Over 400 ponds
werefoundinperiurbanJessoretownalone.Someothercarpseedmarketsare
Borobazar and Kaliganj in Zhenaidaha District mostly based on a big
hatcherycumnurserypond.
Carpseedmarketscanbeclassifiedintothreegroupsdependingonthesources
of seed sold in the markets: (i) primary market: where carp seed are sold after
collectionfrommanydifferentfarms;(ii)secondarymarket,wherethecarpseed
are sold upon bringing out from other markets by intermediaries. Mohajons
obtaincarpseedmainlyfromtheprimarymarketandthenhaultothesecondary
market; and (iii) in the tertiary market, the level of intermediary is one step
increased found at Kakina of Lalmonirhat district and Ulipur and Nageshwari
subdistrictsofKurigramDistrict.
Selling of carp seed in an organised market was found to have several
advantages. For example, traders/farmers can buy target species of required
quantityandcompetitiveprices,greateroptionsforthebuyerstoselectdesired
speciesandquantity,guaranteeindailysupply,lesstraveltimetobuyseedand
money,easyentrance,higherchancesofgettingassistancesfromfellowtraders
during transport, less chance of loosing cash, and higher chances of getting
compensationfromthemiddlemen.However,latecomersmaynotgettheseed
of desired quantity and species. Middlemen and their assistants (day labourer)
tend to give less (both in weight and number). The disadvantages of these
handlingandtransportrelatedactivitiesincreasethestressandresultmortality.

3.5CarpseedtradinginJessorearea
Middlemen typically lease in ponds on a yearly basis with advance payment.
Roadsidepondsareexpensiveandhighindemandbecauseofeasemovement.
Pondownersthemselvesmayevenbethemiddlemen.Shorthaulpaikers/traders
buy and transport seed covering distance 20150 km by truck. Traders of the

159

same location hire a truck in group. Upon arriving in the location, the traders
transport seed to the end users on foot over few hours to a day. Long haul
tradersinitiallytransportcarpseedfromJessoretotheirownarea.Theyconfine
the seed into hapa in pond water. After one or two day of stocking, they
transporttheseseedlongdistance.Thesetradersalsohirepondsinthoseareasto
holdseedpriortoselling.Usually,thetraderssellfishseedwithinaday.Inmost
casestheyarebookedearlyandifseedsarenotcompletelysold,theyholdthem
inthepondwateruntilselling.
People holding carp seed in the ponds along the selling area are called
middlemenorcommissionagents.Theyobtain10percentcommissiononthe
amountsold.Theycontactthefingerlinggrowersontheirown.Middlemenhave
their own staffs for transport, weighing and taking care of seed including the
weighing gears. They are also responsible for all transactions with the
buyers/traders. Fingerling growers observe weighing, pricing, bargaining, and
selling fish seed but usually do not play any active role on these activities.
Middlemenwiththepermissionofthenurserssometimes,compensateincashor
inkindifthetradersclaimforwithdrawalbecauseofpoorqualityseed.
3.5.1Watersupply
InJessorecarpseedmarket,10tubewellsandoneshallowtubewellwerefound
tosupplycleansubsurfacewater.Morethan120womenwerefoundsellingwell
watertothetraders.Eachearnsnearly2US$adayduringthepeakseason(May
to July) and 1 US$ during the lean season. Water retailers charge 20 taka for
filling a 200 L drum and 58 taka for an aluminium vessel (1 US$ = 60 Taka).
Usually,thetraderstransfercarpseedwithsmallamountofpondwaterfromthe
pondtothetransportcontainer.

3.6Confinement
Carp fingerlings are held in hapa in pond water before selling usually for 26
hours.Middlemenconfine100200kgfingerlingintoasmallearthenpond.The
confinementdensityvarieswithspeciesandsize.Commoncarpfingerlingsare
confinedinhighdensitieswhilesilverbarbandsilvercrapareconfinedatlow
density. Seed are confined in hapa into pond and hatchery effluent water in
Jessorearea,intopondwateratAdomdighiinBogra,RajshahiandDurgapurin
Comilla,andintodrumwithwellwateratParbotipurrailwaystationofDinajpur
District. In the night when the concentration of dissolved oxygen depleted, the
middlemen use water pump for spreading water over into pond that enhance

160

oxygenation.Silvercarpandrohuwerefoundasthemosthyperactivefish.The
fingerlingsarereleasedbackintoholdingpondaftertransportfromthenursery
totheseedmarket.Thenextmorning,thesameseedareseinedandheldinhapa
inpond.
At Parbotipur railway station in Dinajpur District carp seed are confined in a
drum with well water. The drum water is agitated from midnight to the next
morninguntilsale.Confinementdensityvariesbetweensystems.Typicaldrum
confinementwithwellwateris125gL1whereasrohuandcatlaareconfinedat
20 kg m2 and silver carp at 18 kg m2 in hapa in pond. Confinement mortality
usually varies between 5 and 20 per cent depending on the quality of water,
density,duration,sizeandspecies.

3.7CarpseedtradinginNorthwest(NW)Bangladesh
Bogra has become the main source of carp seed supply in the NW. Nearly 72
per cent seed are produced at Adomdighi and Kahalu subdistricts of Bogra
District followed by Rajshahi District (NFEP 20). Trading of carp seed usually
startsfromlateFebruaryandcontinuesuntilJulyparticularlyatAdomdighias
thenursersproduceonlythefirstcrop.OntheotherhandatKahalu,carpseed
are traded across the seasons starting from February to September.
Overwintered carp seed are usually transported and distributed from
Adomdighi and Kahalu of Bogra district to Rangpur, Lalmonirhat, Kurigram,
Nilfamari, Naogaon and Gaibandha districts. Both carp and catfish seed are
distributed from Raninagar of Naogaon District to the Southeastern part of the
country.However,currently,Adomdighihasbecomethemainsourceofcatfish
seedsupply.
The seed market at Parbotipur railway station is equipped with the supply of
underground water form the local municipality. There is also an organising
officeforthelocaltraders.UsuallycarpseedfromNatoreDistrictandSantahar
railway station of Adomdighi Subdistrict of Bogra District are transported to
Parbotipur railway station. Some seed are sold to the smallscale traders and
most are sold and/or further transported to Rangpur, Dinajpur, Lalmonirhat,
Nilfamari and Kurigram districts. Kakina Bazaar in Hatibandha subdistrict of
Lalmonirhat District has become popular for carp seed trading since 2002.
Mohazons (largescale traders) transport seed using plastic drums mostly from
Bogra and Natore districts to the local smallscale traders. Some smallscale
tradersalsotransportcarpseedfromBograandNatoredistrictstoKakinabazaar.
UnlikeJessoretherearenomiddlemeninKakinacarpseedmarket.

161

3.8CarpseedtradinginMymensinghregion
MostcarpseedinMymensingharesoldatthefarmgate.Thetraderstransport
carp seed from Gouripur, Ishwarganj and other parts of Mymensingh District
down to low land areas for example, Netrokona and Kishorganj districts. The
traders usually collect seed from different ponds and gather at Gouripur,
IshwarganjandGaforgaonrailwaystations.However,unreliabilityoftrainwas
foundtobeabigconstraintifanyderailmentthatfrequentlyoccurs,disruptthe
channelsandincomeofthetraders.

3.9Transportmediumsandwaterexchange
Carp seed traders use different transport mediums, for example, pond water,
well water and a mixture of the two depending on the availability. However,
most traders (45%) were found to use the mixture of well and pond water
followedbywellwater(37%)whileother18percentusedpondwater.
Most traders change water during transport if the water becomes turbid. The
typeoftransportwatertobechangeddependsontheavailability.However,the
usual water change interval varies from 23 h or 34 h. The transport of silver
barbfryneedfrequentchange,forexample,12hintervalbecausetheyaremore
sensitive to transport stress. Changing of transport water also depends on the
transportmeans.Forexample,ifseedaretransportedbytrain,itmaynotalways
bepossibletochangewaterwhenneeded.

3.10Transportdistance
Transport distance usually varies from few to several hundred kilometres, for
example, from Naogaon to Chittagong District. Short haul traders usually
transport seed around by covering 50150 km (Table 1). Long haul transport is
not a continuous journey. First, the traders transport seed for a short haul and
confine into pond water in hapa. Next day the traders transport seed to the
farmerspondbyonfoot,bus,rickshawvanorbicycle.Mosttraders(29%)were
found to cover distance of nearly 50 km and 16 per cent covered 50100 km
(Table1).However,distances150200and200250kmarecoveredby13percent
oftraders.

162

Table1: Intervals of Distances (km) Covered by Carp Seed Traders during

TransportSampledin11DistrictsofBangladesh

IntervalofDistance(km)

TradersPercentage

upto50

29.4

50100

15.7

100150

13.3

150200

13.3

200250

9.3

250300

7.4

300350

3.1

350400

4.9

Above400

3.6

Total

100.0

3.11Transportdurationandloadingdensity
Transport duration depends usually on the demand of fish farmers, where the
traders come from and availability of transport means. Traders from various
parts of the country travel the nearest seed production areas/markets, and
transport around their localities. The duration and distance to transport also
dependonthepriororder.Ifthetradersdonothavespecificorderbeforehand,
theypeddlearoundtoselltothefoodfishfarmers.
Tradersloadashighas500gL1forshortdistance(0.51h)whilebelow200gL1
forshorthauls(Table2).
Table2: Loading Density of Carp Fry/Fingerlings L1 by Size in Traditional
OpenMethodbyHandAgitationSurveyedinBangladesh

Sizeoffish

Loadingdensity(gL1)

Respondentspercentage

Small<1

1to200

Fingerling>1and<3

2to300

29

3to400

24

4to500

11

500

27

3.12Transportmortality
Immediate mortality of carp seed resulting from handling and transport vary
widelydependingonspecies,size,durationanddistance,conditionoftheseed

163

and transport mediums. The immediate mortality usually varies from 525 per
cent. Occasionally, even the whole batch can die. Sometimes the foodfish
growers claim for compensation which could be due to delayed mortality.
However, the reason for lower recapture is unknown. Hasan and Bart (2007)
havedemonstratedthattransportmaycauseeven50percentdelayedmortality.
Common carp was found to be the hardiest fish among carps during
confinementandtransportinthepresentsurvey.Althoughtransportmortalityis
common,somefishesaremoresensitivethanothers.However,accordingtothe
traders, silver carp was the most sensitive to transport mortality followed by
rohu,mrigal,catla,Catlacatlaandsilverbarb,Puntiusgonionotus(Table3).Silver
carp,rohu,mrigal,catlaandsilverbarbaremorehyperactivethanotherfishes.
Asaresult,thesefishesaresusceptibletodiseasesandsensitivetomortality.
Table3: FishMortalitybySpeciesaccordingtoCarpSeedTradersSurveyed
inCarpSeedMarketsinBangladesh

Species
Traderspercentage
Silvercarp

50

Rohu

29

Mrigal

11

Others(Rajputi/Catla)

10

Total

100

3.13Durationofcarpseedtrading(days,monthsandyears)
Most traders (87%) transport carp seed for 57 days a week and some (11%)
transportfor56days.Similarly,mosttraders(49%)transportcarpseedfor36
months,onequartertransportfor69monthsandsometraders(18%)transport
for 912 months a year. The duration of trading was found to vary among the
traders.Nearlyonethirdtraderstradedcarpseedduringtheprevious510years
and 21 per cent for 1015 years (Table 4). Onefifth traders were found trading
carp seed for the previous 510 years and 14 per cent for 1520 years. Traders
tradingcarpseedinexcessof1520yearswerefew.

164

Table4: Distribution (%) of Carp Seed Traders by Years of Trading Carp

SeedSurveyedin11DistrictsofBangladesh

Duration(y)

Traderspercentage

15
510
1015
1520
2025
2530
3035
3540
Total

19.6
33.3
21.0
13.5
5.5
4.1
2.0
0.9
100.0

3.14Quantityofseedtransport
Thequantityseedtransportedbythetraderswasfoundtovarywidely(Table5
&6).Nearly41percenttraderstransported510kgseedadaywhile31percent
transportedonly5kg.Further14percenttraderswerefoundtotransport1015
kg seed a day. The average quantity of seed transported by the traders was
nearly 11 kg a day. Medium land class traders were found to transport the
highestquantity(18kg)whilelargelandclassHHtransportedthelowest(8kg).
Table 5: Amount (kg) of Carp Seed/Trader/Day Transported by Traders
Sampledin11DistrictsinBangladesh

Quantity(kg)

Traderspercentage

upto5
510
1015
1520
above20
Total

31.1
41.3
12.6
1.4
13.7
100.0

3.15 Reasonsfortradingcarpseedasaprofession

Thereasonforadoptionofcarpseedtradingasaprofessionwasalsoexplored.
Most traders (97%) responded that they trade carp seed because of higher
income compared todaily labour wage. Further37per cent trade seed because
theycanearnenoughtosupporttheirfamilies.

165

Table6: Quantity (kg)/Day/Trader of Carp Seed Transported by District and

LandClassofCarpSeedTradersSampledin11DistrictsinBangladesh

Landclass
District
Chuadanga

Absolutel
landless

Landless

Marginal

Small

Medium

Large

12.31

21.50

13.00

0.00

0.00

0.00

Mean
15.11

Norail

10.67

11.75

16.67

0.00

0.00

11.84

Jessore

11.47

10.22

11.62

10.00

0.00

0.00

11.25

Kurigram

5.28

7.05

8.71

7.00

0.00

0.00

6.18

Lalmonirhat

6.33

9.89

4.25

0.00

2.00

0.00

6.74

Khulna

14.36

10.56

10.35

8.00

22.50

5.00

13.03

Bogra

13.14

9.89

7.54

6.00

0.00

0.00

11.07

Naogaon

7.65

5.70

6.00

14.50

15.00

5.00

7.81

Satkhira

9.88

9.20

12.00

15.00

12.00

0.00

10.18

Bagerhat

15.11

14.00

12.33

15.00

10.00

14.50

14.05

Mymensingh

8.46

7.71

12.00

9.00

0.00

0.00

8.82

Mean

10.31

9.84

10.20

11.92

18.11

8.17

10.46

3.16 Useofadditivesduringtransport
Nearly 78 per cent traders were found to use different chemicals during
transport. The use of chemicals was found to vary between land classes with
wide geographical variations. While all the traders of Naogaon District were
found to use some form of chemicals, only 36 per cent traders from Jessore
Districtwereaddingsomethinginthewater.Interestingly,sampledtradersofall
eleven districts were found to use ORS (oral rehydration saline), a product of
ICDDR,B for human consumption as life saving antidote. Renamycin, an
antibiotic was found to be used by all traders except the traders of Norail and
Jessore districts. Traders from 6 of 11 districts were found to use mineral salt.
Moreover,vitaminC(ascorbicacid)andB2(riboflavin),doxin(antibiotic),betel
leaf juice, agricultural potassium, sugar for human consumption and poultry
eggs were also used to reduce mortality and maintain water quality during
transport.Theaveragedoseofsaltpervessel(5060L)wasnearly10gwhilethe
numberofpackofORSwas1.28.Thequantityofantibiotictablet,forexample,
renamycin per vessel (5060 L) varied from 1 to 5. The highest quantity of
antibiotic (45 tablets per 5060 L vessel) was used by the traders of Khulna
District.

166

3.17Problems/constraintsencounteredbycarpseedtraders
Almost all traders reported immediate transport mortality and compensation
due to delayed mortality claimed by the foodfish farmers as problems to carp
seed trading. Credit selling was found as another constraint since recovery is
alwaysless and sometimes no recovery atall. Most traders (85%)reported that
theylackcapitalwhichisthemainconstraintsintransportingfishseed.Further
67percenttraderscomplainedlackofgoodqualityseedand57percentclaimed
high mortality. Nearly half the traders (44%) think poor transport technique is
oneofthemainproblemsforseedtrading.Nearly35percenttradersindicated
thattheylackbasictrainingskills.

3.18Policyimplications
3.18.1Developmentaspects:Marketdevelopment

There is a need for new carp seed markets to reduce the transport distance,
transformation of the current earthen holding ponds into flowthroughtank
systems for confinement/holding carp seed in the market before selling. The
infrastructure and other market facilities are also necessary to improve. NGOs
engaged in fisheries can be involved in developing the carp seed markets by
replacingthecurrentearthenholdingpondsintoflowthroughtanksystem.
3.18.2Technicalaspects:Problemsneedtoberesolved
Negativeimpactresultingfromhandlingandtransportrelatedactivitiesonthe
quality deterioration of fish seed, causes of immediate and delayed mortality,
effects of stress resulting from handling and transport related activities on the
survival,stressresponses,diseasesusceptibilityandgrowthoffishandtechnical
solutiontoreducemortalityasmanagementstrategy.

167

References

Ahmad, N. 1954. Mortality of fish during transport and cultural operation: Causes and
prevention.PakistanJournalofScience,6(2):15.
Chowdhury, A. Z. M. N. I. 1997. Effectofprolongedtransportationofcarpfingerlingsonits
growthandsurvivalinculturalpond.M.Sc.Thesis.BangladeshAgriculturalUniversity,
Mymensingh.
Hasan,M.andA.N.Bart.2006.CarpseedtradersinBangladesh:sourcesoflivelihoods
and vulnerability resulting from fish seed mortality. AsiaPacific Journal of Rural
Development,16(2):97120.
Hasan,M.andA.N.Bart.2007.Effectsofcapture,loadingdensityandtransportstresson
the mortality, physiological responses, bacterial density and growth of rohu Labeo
rohitafingerlings.FishPhysiologyandBiochemistry,33(3):241248.
IFAD.2002.AssessmentofRuralPoverty:AsiaandthePacific.P.30.
Lewis, D. J., G. D. Wood and R. Gregory. 1996. Tradingthesilverseed. Dhaka: University
Presslimited.P.199.
Nabi,R.,M.A.HossainandS.Rahman.1983.Ontheuseoftubewellandpondwaterfor
transportingcarpfry.UniversityJournalofZoologyRajshahiUniversity,2:2528.
NFEP(NorthwestFisheriesExtensionProject),SecondPhase,Papernumber20.Fryand
fingerlingtradeatParbotipurrailwaystation,1989,and1992to1996.
Sampson, D. R. T. and D. J. Macintosh. 1986. Transportation of live carp fry in sealed polythene
bags. Aquaculture, 54: 123-127.

168

AsiaPacificJournalofRuralDevelopment
NOTETOCONTRIBUTORS
1.0 Aims
The AsiaPacific Journal of Rural Development (APJORD) is an interdisciplinary
journaldevotedtoissuesanddiscussionsonruraldevelopmentprimarilyinthe
AsiaPacificregion.APJORDfocusesonpovertyissuesandruraltransformation,
keepinginviewthefourprogrammeprioritiesofCIRDAP,namely(1)Agrarian
Development; (2) Institutional/ Infrastructural Development; (3) Resource
DevelopmentincludingHumanResources;and(4)Employment.
2.0 SubmissionofArticles
Articles are accepted for publication provided they are submitted to APJORD
only.Ifthereismorethanoneauthor,astatementtotheeffectthatthearticlehas
been approved by all the authors must accompany it. It is the responsibility of
the author(s) to obtain permission for the use of previously published or
unpublished materials including tables and figures. All articles are subject to
peer review and may be revised if necessary. The editors decision regarding
articlessubmittedforpublicationisfinal.CopyrightrestswithCIRDAP.
Authors are entitled to three complimentary copies of the issue in which their
articles appear. Articles written between 4,0005,000 words are preferred. Both
the hard copy and soft copy of the writeup must have to be submitted to the
editor.Thesoftcopyeithermaybegiveninfloppydiskette/pendrive/CDorcan
besentasanattachmenttotheemailaddress:infocom@cirdap.org.
Typescriptsshouldbesubmittedinduplicate,onA4sizepaper,typewrittenon
oneside,indoublespace,withmarginsofatleast2.54cm.(oneinch).Statistical
tables,illustrationsandchartsshouldbesubmittedonseparatesheets,andtheir
positionsshouldbeindicatedinthetext.
3.0 StyleandPresentation
APJORD mostly follows the Chicago Manual of Style for preparing articles.
Authorsshould,asfaraspossible,adheretothefollowingbasicrequirementsof
style:
i.

The name and institutional affiliation of the author should appear as a

footnoteonthefirstpagetogetherwithacknowledgement,ifany.

ii. Anabstractofapproximately100wordsshouldprecedethetextonthefirst
page.ItshouldbetitledAbstractindented,italicisedandcentred.
iii. Britishandsspellingsshouldbeused.Numbersfromzerototenshouldbe
spelledout,andnumeralsshouldbeusedforallothernumbers.
iv. Wherereferencesaremadewithoutanycomment,theyshouldbeinsertedat
the appropriate places in the body of the text in parenthesis giving the
surname of the author(s) and year of publication of the work under
reference, e.g. (Shama 1915: 41), (Mayco and Santalines 1991: 18), (Jupitar
1985:21;Razia1999:66).
v. Where references are made with detailed notes/comments, they should
appear as Notes, numbered sequentially throughout the article by
superscripting in raised Arabic numbers and the corresponding notes
appearingasfootnotesorlistedsequentiallyattheendofthearticle.
vi. The reference list should appear after the Notes. For books give authors
surname,othername(s)and/initial(s),yearofpublication,titleofpublication,
placeofpublication,andpublisher,e.g.
Syed, Shireen Kamal. 1992. Women in Postharvest Loss Prevention. Dhaka:
CentreonIntegratedRuralDevelopmentforAsiaandthePacific.
Tamanga, A., N Shrestha., K. Sharma, and S. Kathi. 1994. Case Studies of
WomenwithAbortionExperiences.Kathmandu:IDS.
Acharya,Meena.1994.PoliticalParticipationofWomeninNepal.InNelson,
Barbara and Chowdhury, and Norma edited Women and Politics Worldwide.
London:YaleUniversityPress.
For Journal articles, give the authors surname, other name(s) and/or initial(s),
year of publication, title of article, volume and issue number, and inclusive
pagination.e.g.
Quarantelli, E.L. 1992. Can and Should Social Science Disaster Research
Knowledge
and Findings from Developed Societies be Applied in Developing
Societies?
AsiaPacificJournalofRuralDevelopment,Vol.II,No.2,December,Pp.114.

171

ASIA-PACIFIC JOURNAL OF RURAL DEVELOPMENT

ISSN 1018-5291

ORDER FORM
Asia-Pacific Journal of Rural Development, Six-Monthly, First issue published in July 1991
Please enter my/our subscription to Asia-Pacific Journal of Rural Development beginning with

Vol. _____

No. ______

Name :
Designation :
Organisation :
Address :

City :

Postcode :

Country:

E-mail :

Signature:

Date:

Annual Subscription Rates for two issues (including postage): US$20 for Asia-Pacific
countries excluding Bangladesh. Tk.300.00 for Bangladesh. US$35 for all other countries.
Please send Order Form to:
Editor
Asia-Pacific Journal of Rural Development (APJORD)
Chameli House, 17 Topkhana Road, GPO Box 2883, Dhaka-1000, Bangladesh
Fax: 880-2-9562035 E-mail: infocom@cirdap.org
Visit our website at
http://www.cirdap.org.sg

172

170

171