apjord vol-xix-1 july 09-1.pdf
TRANSCRIPT
C I R D A P The Centre on Integrated Rural Development for Asia and the Pacific (CIRDAP) is a regional, intergovernmental and autonomous institution. It was established in 1979 at the initiative of the countries of the Asia‐Pacific region and the Food and Agriculture Organisation (FAO) of the United Nations with support from other UN organisations and donor countries/agencies. The member countries of CIRDAP are Afghanistan, Bangladesh (host state), India, Indonesia, Iran, Lao PDR, Malaysia, Myanmar, Nepal, Pakistan, Philippines, Sri Lanka, Thailand and Vietnam. The main objectives of the Centre are to: (i) assist national action, (ii) promote regional cooperation, and (iii) act as a servicing institution for its member countries for promotion of integrated rural development through research, action research, training and information dissemination. Amelioration of rural poverty in the Asia‐Pacific region has been the prime concern of CIRDAP. 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; GO‐NGO collaboration; Gender issues; Governance issues; and Environmental concerns for sustainable rural development. Operating through designated Contact Ministries and Link Institutions in member countries, CIRDAP promotes technical cooperation among nations of the region. It plays a supplementary and reinforcing role in supporting and furthering the effectiveness of integrated rural development programmes in the Asia‐Pacific region.
EDITORIAL BOARD
Chairperson Durga P. Paudyal
Editor K. A. S. Dayananda
Members T. Abdullah, Development Consultant Q. K. Ahmad, Bangladesh Unnayan Parishad Momtaz Uddin Ahmed, Dhaka University Mohammed Farashuddin, East West University S. K. Singh, CIRDAP Hossein Shahbaz, CIRDAP
Editorial Assistant S. M. Saifuddin
ISSN 1018‐5291
Asia‐Pacific Journal of Rural Development
VOLUME XIX JULY 2009 NUMBER 1
CENTRE ON INTEGRATED RURAL DEVELOPMENT FOR
ASIA AND THE PACIFIC
ASIA‐PACIFIC JOURNAL OF RURAL DEVELOPMENT Volume XIX, July 2009, Number 1
ISSN 1018‐5291 © CIRDAP 2009
The views expressed in this publication are those of the author(s) and do not necessarily reflect those of CIRDAP
Published twice a year
Price including postage
Annual Subscription (Two Issues): Tk.300 for Bangladesh
US$20 for Asia‐Pacific Countries US$35 for Other Countries
Advertisements are invited from professional bodies/institutions
The following tariff charges will apply
a) Inner back cover: US$200 b) Full page text: US$100 c) Half page text: US$50
Published by Durga P. Paudyal, Director General, on behalf of
Centre on Integrated Rural Development for Asia and the Pacific (CIRDAP) Chameli House, 17 Topkhana Road, GPO Box 2883
Dhaka‐1000, Bangladesh
Printed by: Nizam Printers & Packages, Dhaka
ASIA‐PACIFIC JOURNAL OF RURAL DEVELOPMENT
Volume XIX July 2009 Number 1
CONTENTS Articles Page ICT and Empowerment of Rural and Deprived Women in Asia 1 D. A. Patil, A. M. Dhere, C. B. Pawar
Spatial Assessment of Potential for Groundwater 23 Development in Lower Chao Phraya River Basin Sudsaisin Kaewrueng, Mukand S. Babel, Ashim Das Gupta and Shahriar M. Wahid
Migration of Rural People to Urban Areas: 43 A Study of Three Upazilas in Mymensingh Gulsan Ara Parvin, Mamun Muntasir Rahman, Farhana Yasmin
Accessibility of Women to Productive Resources in 59 Farm‐Households of Kaduna State, Nigeria S. A. Rahman and F. A. Ajayi
Empirical Analysis on Rural Households’ Borrowing Behaviour: 67 The Case of Central Java Takashino Nina
Value Chains and Standards in Shrimp Export from Bangladesh and 89 Thailand to Japan: A Comparative Study on Safety Compliances Mohammad Taj Uddin
Simulation of Growth and Yield of Maize under Water Stress Imposed 109 during Critical Growth Periods in Nakhon Ratchasima Province, Thailand Kiattiyos Thongsaga and S. L. Ranamukhaarachchi
Decentralisation on Access to Rural Water Supply Services: 135 A Study of Six Villages in West Bengal, India Indranil De
Carp Seed Transport in Bangladesh 155 Mahmud Hasan
A V
1Wihi‘waww
*
*
*
sia‐Pacific Journal of Rural Development
ol. XIX, No. 1, July 2009
ICT and Empowerment of Rural and Deprived Women in Asia
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 ICT‐based initiatives in Asia in general and India in particular which are specifically designed for the empowerment of rural and deprived women. It shows how ICT‐based 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 technical challenges as well as policy suggestions to use ICTs proactively and effectively to promote gender equality and for the empowerment of rural and deprived women. The study used observations, focus group discussions and review of documents to collect secondary and primary data.
.0 Introduction omen constitute half of the world’s population, perform nearly two‐thirds of
ts work hours, receive one‐tenth of the world’s income and own less than one‐undredth of the world’s property. As per 2001Census, 23 per cent of women are n the work force, 94 per cent of all working women are in the informal sector. Women carry a disproportionately greater burden of work than men and since omen are responsible for a greater burden of work than men and since women re responsible for a greater share of non‐SNA (System of National Accounts) ork in case of economy, they enter labour market already overburdened with ork’ (United Nations 2005).
Corresponding Author. Assistant Professor, Centre for Equity, Social Justice and Human
Development (CESJHD) Yashada, Pune‐07, Maharashtra, India, E‐mail: [email protected]
* Lecturer, Inidra College of Commerce and Science, Pune‐33, Maharashatra, India. E‐mail: [email protected]
** Lecturer, Sangola College Sangola, Solapur, Maharashatra, India. E‐mail: [email protected]
In case of India, as according to 2001 Census, out of the total population of 1,027,015,247 the female population was 495,732,169 accounting for 48.2 persons of the total population. But, their participation in economically productive activities are often underestimated. For last two decades, women’s work participation rate increased from 22.73 per cent in 1991 to 28.6 per cent during 2001 (Census 2001).
In the patriarchal society, women have been subjected to exploitation, atrocities, male‐dominance and gender‐bias. 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 women’s empowerment. In this context ICTs are emerging as a key facilitator for empowering women, especially rural and deprived women. The present study is an attempt in this context to study the role of ICTs for the empowerment of rural and deprived women.
2.0 Background of Study
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 women’s ‘empowerment’ and a number of ICT projects that specifically target women have been established in several developed and underdeveloped countries. Before going to study the role of ICTs in women empowerment, is necessary to understand what is ICT.
2.1 What is ICT?
Information and Communication Technologies (ICTs) are a diverse set of technological tools and resources to create, disseminate, store, bring value‐addition and manage information. The ICT sector consists of segments as diverse as telecommunications, television and radio broadcasting, computer hardware, software and services and electronic media, for example, the internet and electronic mail (United Nations 2005:1).
2.2 Role of ICTs in gender empowerment
Empowerment of women in the context of knowledge societies entails building up the abilities and skills of women to gain insight into the issues affecting them and also building up their capacity to voice their concerns. In this context ICTs are emerging as a powerful tool for gender empowerment in many developing
2
countries. There has been a rapid growth in the ICT sector since the late 1980s and the use of ICT has dramatically expanded since the 1990s. According to the World Bank, tele‐density in India had reached 3.8 per cent of the population by 2001 (Jain 2006).
2.2.1 Examples from Asia
In Asia some of the ICT initiatives have been playing vital role for empowering rural and deprived women. For example, the Lumbini CMC project in Nepal was established in April 2004. The community multimedia centres CMC combining new technologies like computers, internet and specialised software applications with traditional media; it works mainly for education, health and adult literacy. Likewise the Information and Communication Technology Agency of Sri Lanka (ICTA) started a project to use technology to document violence against women and educating women on their legal rights with the help of web‐portal. However the e‐Homemakers project in Malaysia was chosen as a runner‐up in gender and 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 key gender ICT initiatives have also been presented in the next section.
2.2.2 Examples from India
India has initiated concrete policies to promote gender parity in information technology and a number of projects have been launched. For example, a project set up by the Indian Institute of Technology (IIT) Chennai where a total 80 per cent of the new kiosks setup are run by women, many of whom have had very little or no acquaintance with technology earlier (see the Chart). Women’s groups NGOs and other sections of the civil society are also playing a vital role in empowering women in the region (Vijayalaxmi and Bhavani 2006).
India shop, an e‐commerce website (2005) in Tamilnadu, has been designed to sell products made by rural women’s co‐operatives and NGOs. The Dhan Foundation (2004) and Swayam Krishi Sangam (2004) are using ICTs such as handheld devices and smart cards, to improve microfinance projects to empower poor women. The Self‐employed Women’s Association (SEWA 2004) has several ICT projects for women including community‐learning centres, a school of Science and Technology for self‐employed women and the Theliphone project, which provides mobile phones to women in the informal sector.i
3
Chart: Some Selected Gender Based Rural ICT Initiatives in India
Sl. No. Name Project
(Initiator) Area Target Group
Implementing Agency
Sponsoring Agency
Year of Starting
1. Community Radio‐Deccan Development Society
Deccan Develop‐ment Society
Medak District Andhra Pradesh
Dalit Women
DDS UNESCO
1998
2. Kudumbha‐shree
Govern‐ment of Kerala
Kerala (991 Panchayats and 58 municipa‐lities)
Family as a whole and women in particular
Govern‐ment of Kerala
SIRDI
2002
3. Mahiti Manthana
Mahila Samakhya, Karanataka and IT for change, Benglore
Three talukas (Humsur, Nanjansud and Periyapatn) in Mysore district
Self‐Help Group women
IT for change Karnataka
Mahila Smakhya Karnataka (MSK) 2005
4. Nabanna Change initiatives and UNESCO
Buduria Municipa‐lity North‐24 Paragans district‐West Bengal
Rural Women
Chagne initiatives, Kolkatta
UNESCO
2003
5. Project Enter Price‐Toe Hold
ASCENT, CLRI
Belgaum District, Karnataka
Rural artisans specially women
CLRI NLDP‐UNDP 1999‐
20002
6. Putting ICTs in the hands of the women
Datamation Foundation (Non‐profit Trust)
Rural Areas of Kanpur, Lacknow etc.
Rural Margina‐lised women
Datamation Foundation Charitable Trust
The World Bank Microsoft 2001
7. SEWA Self‐Employed Women’s Association
Ahmeda‐bad, Gujarat, Western Part of India
Informal Sector, Women
SEWA and Partner NGOs
SEWA
1977
8. Tamil Nadu Women in Agriculture (TANWA)
Ministry of Agriculture (GoI)
Tamil Nadu Rural Women
Directorate of Agriculture
DANIDA Govern‐ment of India
1994 2003
4
Sl. No. Name Project
(Initiator) Area Target Group
Implementing Agency
Sponsoring Agency
Year of Starting
9. PACS (Poorest Areas Civil Society) Programme Empowering the rural poor
Network of Entrepre‐neurship and Economic Develop‐ment (NEED)
Sitapur and Barabanki blocks of Uttar Pradesh
Rural population (Women and Children ) for health
PICs NEED
2002
10. Pravara Village IT Project (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. ICT Enabled Life Skill and Sexuality Education for Adolescent Girls
Centre for Women’s Develop‐ment and Research
Kancheepuram District, Chennai
Dalit Adolescent girls (for educatio)
Centre for Women’s Develop‐ment and Research
Centre for Women’s Develop‐ment and Research
2004
12. Computer on Wheels: Education Through Mobile Vans
Vidya Pratishthan’s Institute of Information Technology (VIIT)
Rural areas in all India
Rural Population
Vidya Pratish‐than’s Institute of Information Technology
State Govern‐ment
2000
13. Economic Empowerment of Minority Muslims in India
Datamation Foundation
Seelampur Zaffarabad
Suppressed minority women of the country
Damation foun‐dation
UNESCO
2001
14. Indira Soochna Shakti (ISS)
UNDP and Planning Commi‐ssion and Govt. of India
246 Village Counsils (Panchayat), headquarter villages of Chattisgarh
Students specifically girls and women
CHIPS Govern‐ment of Chattis‐garh and UNDP
2001
Source: http://ictrd
Self‐help groups of rural women in Andhra Pradesh have been so successful in marketing their products at home and abroad that the major MNCs (multi‐national corporations) want to use their selling skills.
Thus, with these facts in mind present paper looks at the avenues created by ICT enabled services and opportunity for women’s empowerment.
5
3.0 Review of Literature
There are very few studies which have been conducted to analyse the role of ICTs for women empowerment; some of the relevant studies are as follows:
CEEWA (Council for the Economic Empowerment of Women of Africa) (2005) in his study in Uganda found that as an enormous source of information, ICTs constitute a powerful learning tool that provides access to marketing information that can help women’s business succeed. ICTs like mobile telephony, can also offer direct and inexpensive means of communication for women’s organisations and enable them to share knowledge on a quick and collective basis. However, access to ICTs is restricted because of oppressive gender relations, social cultural 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 concept of work and workplace. New areas of employment such as networking, i.e. working from a distance, are becoming feasible with new technology.
Dalal (2006) argues that, while there is recognition of the potential of ICT as a tool for the promotion of gender equality and the empowerment of women, a ‘gender divide’ has also been identified reflected in the lower numbers of women accessing and using ICT compared with men. Unless this gender divide is specifically addressed, there is a risk that ICT may exacerbate existing inequalities between women and men and create new forms of inequality.
The study by the Vijalakshmi and Bhavani (2006) examines the nature and source of gender variations in internet use and how cultural factors contribute to such variations. The study provides insight into the gender variations in using the internet space and forming online relationships and online discourse. On the whole, the study supports the fact that the internet provides an opportunity for women to construct their identities to challenge certain traditional norms and reconstruct their lives.
The findings of Moser (1993) indicate that the assumption where access to ICT will automatically empower women economically or otherwise is questionable. On the contrary, women’s social position in society has not changed much.
4.0 Conceptual Framework
While studying the concept of empowerment it is necessary to understand this concept in a holistic manner, which will provide a concrete idea to the researcher to select specific indicators for the concept of empowerment.
6
4.1 Understanding empowerment: Dilemma of meaning and measurement
This section provides some key approaches of empowerment and also it gives a holistic definition of empowerment.
4.2 Empowerment and power
Many past studies, ‘empowerment’ is looked at in relation to how people understand power. It is also presented as a process through which women gain power over men. This specific understanding of empowerment is possible one of the reasons men resit the concept. In other cases, empowerment is presented as a process that enables women gain access to decision making and making choices about their own lives (Huyer and Sikoska 2003).
4.3 Feminist approach of empowerment
Feminist understanding of empowerment includes the idea of ‘the personal as political’ therefore calling for a consideration of empowerment as experienced by women within the public and private spheres (Rowlands 1997). Empowerment could also occur at a political social or economic level. Economic empowerment has 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 political empowerment involves the rights and abilities of people to participate as equals in decision‐making processes (Clement 1994).
4.4 Empowerment by development alternatives
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 (Moser 1993).
DAWN uses these strategies to address issues of inequality between women and men because it recognises that women’s empowerment is not a given thing and that women have to work collectively to overcome the structural inequalities (Longwe 1994).
4.5 Empowerment by Longwe’s framework
There is also the Longwe Women’s Empowerment Framework. This framework examines the extent to which projects lead to women’s empowerment.
7
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; while control is level at which there is a balance of power between women and men and neither has dominance. Access considers equality of access to resources and conscientisation is a level of awareness raising. For women to take appropriate action to close gender gaps or inequalities there must be recognition that their problems stem from inherent structural and institutional discrimination. Participation (mobilisation) is concerned with the extent to which women have been able to take part in decision‐making 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 multi‐dimensional, gradual process by which women become aware of their subordinate position and the power structures at play in their environment; then go ahead to do something to overcome these constraints. This process enables women to become self‐reliant and gain control over their own lives; and act on issues they consider important but without infringing upon the rights of others; this process may also lead to ‘collective’ change (Rowlands 1997 and Freire 1972). Thus, it entails developing the capacities of women to overcome social and institutional barriers and strengthening their participation in the economic, political and social processes so as to produce an overall improvement in their quality of life.
5.0 Research Questions
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 rural and deprived section women.
Women need ICT for the same reasons as men to access information of importance to their productive, reproductive and community roles and to obtain additional resources. Based on the above review of literature and conceptual framework, this study is trying to answer the following research questions:
• Can ICTs play an important role in women empowerment? • If yes, then is it empowering rural and deprived sections women? • If yes, what are challenges and obstacles? • How to overcome with these issues?
8
In this context this paper attempts to examine various ICT‐based experiments for the empowerment of rural and deprived women in different parts of India and Asia.
6.0 Methodology
6.1 Objectives
To understand the role of ICTs for women empowerment.
To study how ICTs are helpful for the empowerment of rural and deprived sections women.
To suggest policy implications for the effective implementation of further ICT‐based programmes.
6.2 Database
ICTs are playing an important role for the empowerment of rural and deprived women. We examine the contribution of these technologies in the process of empowerment at the Asian level for the periods‐1981‐82 to 2004‐05. The paper uses just released 2004‐2005 report of United Nations on Women and Beyond, UNDP (2001) Human Development Report and the proceedings of United Nations Economic and Social Commission for Asia and the Pacific (UNESCH P): with Special Reference to Gender and Development (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 one of the oldest ICT based project in Asia. However, discussions with rural ICT mangers of WWVP, Kolhapur, members of Women Association of SEWA, Gujarat and experts of NIRD, Hyderabad immensely helped us to understand socio‐technical hurdles and possible solutions for the effective implementation of ICT projects. The primary data for this study has been collected through personal interviews, focus group discussions and observations during the period July to September 2008.
7.0 Results and Discussion
7.1 ICTs and empowerment of rural and deprived sections of women: Some prominent initiatives
A number of ICT initiatives can be cited all across Asia that demonstrates how women can acquire ICT‐related skills and use the technology and at the same time have control over their use. In this section, efforts have been made to
9
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 non‐government organisations (NGOs) are providing special packages for them. Let us take an overview of some of the prominent ICT‐based initiatives pertaining to the empowerment of rural and deprived sections women.
7.1.1 ICT and women’s economic empowerment
In Asia, a number of credible models exist that could be replicated to address women’s economic issues through the use of ICTs.
7.1.1.1 Self‐Employed Women’s Association (SEWA)
It has been organising women in the informal sector in India since 1972.4 It was one of the first organisations globally to realise the potential of using ICTs for the productive growth of the informal sector. SEWA is establishing Technology Information Centres in 11 districts of Gujarat, India to provide computer awareness training and basic computer skills for their ‘barefoot managers’, built the capacity of women organisers and leaders and strengthen their members’ microenterprises. It now runs programmes that develop women’s abilities in the 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 co‐operatives 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, members of SEWA are able to access government schemes and tap into new markets. In the second phase, the centres will also support the education of girls (www.sewa.org).5
7.1.1.2 Grameen Phones, Bangladesh
It is the best known of the ICT‐enabled businesses with a high percentage of women owners/operators in Bangladesh.6
10
Grameen Banks Women Phone
In 1996, the Bank setup Grameen Phones, Bangladesh’s 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
7.1.1.3 Demonstrator application grant scheme, Malaysia
Among the 74 projects funded, a number were aimed to benefit women, and included women’s capacity building in the use of ICTs. For example, the project called T‐Centre for Tele‐networking and Telecommuting was designed to guide 200 participants mainly women and youths, to learn and acquire tele‐networking skills and to enable them to adapt tele‐working as a new mode of work. Another project that was funded under the DAGS scheme is the E‐Homemakers’ Project, the concept of working from home by providing resources and a platform for homemakers and home workers to tele‐trade and tele‐exchange. A Women’s 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 knowledge‐based economy. The project prepares them to work at home through other soft trainings and empowerment exercises (http://ehome‐makers‐net).
7.1.1.4 Smile (Savitri Marketing Institution for Ladies Empowerment)
It is a voluntary organisation in Pune. This project has increased literacy level of underprivileged women through the usage of ICT. Internet has also helped them market their various products like soft toys, candles, bags, utility items, etc.
11
Through Internet, there is greater awareness and exposures and market reach for the products.
7.1.1.5 Datamation Foundation
It started a project in 2003 in Seelampur area of Delhi for Muslim women. The project localised appropriate communication and information networks by setting up an ICT centre at a Chennai. This helped link resource‐poor women to the information and tolls for knowledge management. It also helped establish buyer‐seller linkages towards eradication of absolute poverty. It has established 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. The ICT centre has created self‐confidence in women, creatied awareness about their interest and helped them take collective decisions.
7.1.1.6 Observations from the above examples
• Women should be trained with business management/development skills: Women are culturally not exposed to establishing and running a business. In fact, in most countries, trade is generally a male domain as well and businesses can succeed or fail by whom you know and do business with. Hence, women will not only lack business management skills but also business development/marketing skills and a business network. Women should either be supported (metored but not made reliant) with the necessary expertise by a expert or be trained specifically on these skills. For almost all of the examples above, women who are enabled to run small businesses are almost always provided with a ‘captured market’ of peers and fellow villagers/community members, a small but still necessary to start outside the ‘captured’ or ‘family market’ which will provide sustainable path.
• Grameen’s development approach and system of accountability: Grameen’s 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 gender‐power 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
household duties and cheap labour. Without the power to work and earn a good income, their voices are silenced. Hence economic independency is very important facet for the empowerment of women. In this context, on the basis of the above initiatives, it is clear that ICTs are playing vital role for the empowerment of rural and deprived sections women8.
7.1.2 ICT and women’s social empowerment
Woman’s social disempowerment is often strongly linked to her isolation from information that she needs and this includes her ability, opportunity and space (both virtual and non‐virtual) to communicate in her own language with others 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. We will study some of the key initiatives in this regard.
‘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 women’s 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
7.1.2.1 Project, Nabanna
In India, new initiatives have been taken to put a web based information system to strategic use for the social empowerment of poor deprived women of Baduria, a rural region in North‐24 Paranganas District in the Indian State of West Bengal.
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 running for health and educational empowerment of rural and deprived women.
7.1.2.2 Sisu Samarakshak, Andhrapradesh
The programme aims at harnessing ICT to accelerate achievement of goals in women’s health children’s growth and development in the early years through enhanced community monitoring mechanisms and bridge the digital divide. This project is sponsored by Government of Andhra Pradesh and UNICEF. It is an ICT initiative to impart knowledge to illiterate communities on health, hygiene and sanitation using audio and culturally appropriate images. The project component includes information pages (Using audio, picture, video and touch tone screen), especially for poor illiterate and semi literate women (www.ict4d.org).
7.1.2.3 Yuva.com: ICT Education Training Programme
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 Observations from the above initiatives
• Technical and social limitations: As far as rural ICT initiatives are concern, it was found that due to women’s heavy workloads and multiple roles that
14
limit their available time to use the ICT centres; male attitudes towards women’s 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 women in their local languages; and their lack of disposable income for fee paying centres, are all gender‐based factors that contain the way of women empowerment through ICTs. Though the numbers of ICT initiatives are limited, this is a beginning of the silent revolution, especially in traditional agrarian countries like India, Sri Lanka, Bangladesh etc.
7.1.3 ICT and political empowerment of women
People around the world are using new technologies in unprecedented ways for networking, movement building, political participation and advocacy. ICTs have been applied as agents of change in enabling women to participate directly in politics and civic life.
7.1.3.1 SEWA’s initiative
In India, SEWA organises electronic discussions through the panchayati raj (village governance institutions). In these discussions, village women often pose questions that are answered promptly by a panel of experts. Through translation modules, responses go to the women in their vernacular language (Nath 2001).
7.1.3.2 Women to women video project, 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 Water and Sanitation Project that used technology to effectively address gender issues and include women in the design and planning processes. The women to women video project provided women with the opportunity to speak across the physical boundaries of the pudah (custom of seclusion), discuss what water and sanitation infrastructure could be afforded and participate in choosing the best water supply option for their households (www.wikibooks.com).
7.1.3.3 E‐Seva, Andhara Pradesh
A project that possibly captures the potential of ICTs in enabling women’s political and civic participation is the e‐seva (e‐services) project, which began in the district of West Godavari in Andhra Pradesh, India. The project uses ICTs to provide access to various Citizen‐to‐Citizen (C2C) and Citizen‐to‐Government
15
(C2G) services to the people living in rural areas. Under this project, web‐enabled rural kiosks termed e‐seva centres were established at the mandal (a sub‐district unit of administration) level. The unique thing about these centres is that they are run and managed by rural women as information leaders to help bridge the gender divide.
7.1.3.4 Observations from the above examples
• Focus on administrative goal rather than main goal: It is important to note that while e‐governance initiatives are promising, many focus only on administrative efficiency and goal and fail to enable the citizen’s full political participation in consultative and decision‐making bodies or mechanisms. This means that at best such efforts render women as consumers of information and not equal stakeholders in development.
• 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 heard and responded to with substantive affirmative action. Otherwise, even e‐government services will lose their relevance to women in the long run.
8.0 Conclusion and Policy Implications
It is imperative for a sustainable development in developing and over‐populated countries like India and other Asian countries, that women have access to education and appropriate need‐based technologies. ICT has been promoted as an vital tool in ensuring that marginalised groups, particularly women are included in the development of the global information society. While ICT alone 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 women have access to and fully participate in the area of ICT. Effective access to and use of ICT can improve rural women’s 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. According to findings of the ITU, there are many challenges to implement ICT programmes especially for the empowerment of rural and deprived women.
16
8.1 Challenges of ICT use for women’s empowerment
8.1.1 Technical problems
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 technical problems. Following are some of the key constraints:
– Insufficient regional specific information
– Inadequate internet connectivity and electricity
– Inadequate support from facilitator
– Lack of Infrastructure facility
– Lack of construction of content in local language
– Limited duration of projects (especially sponsored projects )
– Problem of monitoring and evaluation
– Lack of operators knowledge
– Lack of the involvement of other ICT channels (radio, television etc.)
– Lack of marketing management/development skills.
8.1.2 Social problems
While studying the problems for the effective implementation of ICT projects for women empowerment, it is also necessary to reduce social problems because social problems are deeply rooted in society. Some of the key social problems are as follows:
i. Orthodox and traditional attitude ii. Lack of enabling environment and awareness iii. Patriarchal dominated society iv. Lack of education and training (particularly rural and deprived women). v. Poverty and huge domestic and agricultural work. vi. Lack of healthy participation from all strata of the society. vii. Lack of need‐based approach. viii. Men hegemony
8.2 Strategies for effective use of ICTs for women empowerment
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, rural women associations and experts suggest following strategies for the effective utilisation of ICT for the empowerment of rural and deprived women.
8.2.1 Enabling environment for ICT
There is a need to create an information friendly environment, which supports and encourages strategies to promote women’s equal access to and opportunity to benefit from ICT projects as well as creating a regulation and policy environment which supports women’s use of ICTs.
8.2.2 Content development
Developing content which speaks to women’s concerns and reflects their local knowledge and which is of value for their daily lives, business enterprises and family responsibilities.
8.2.3 Need‐based development of software
To improve capacity of women managers/operators of information services to identify and supply the information needs of users based on the human capacity (literacy, numeracy) of beneficiaries and to utilise the information.
8.2.4 Bridging the divide
Strategic development of ICT to address socio‐economic dividers: age, gender, community, land holding and geography are essential.
8.2.5 Low‐cost training institutes
An emphasised by the advancement of women in 2002, the main barriers to women’s use of ICT continue to be lack of proper training especially among the rural and deprived women. In this context Schools, Panchayat offices and Women Self‐Help Groups (SHGs) should impart ICT training. Students can be effective ‘no cost trainers’ NSS/NCC students can play a key role.9
8.2.6 Marketing development and management skills
Now‐a‐days many ICT projects are not working effectively; marketing development and management skills, these two factors, are very important and most of the rural and deprived women are lacking in these skills. Hence these skills are necessary to be included in ICT training and modules.
18
8.2.7 Development of public access centres
One of the strategies adopted to increase access of remote, rural and marginalised groups to ICT is the development of public access centres, such as public phones, telecentres, libraries, information centres or cybercafés.
8.2.8 ICT centres can be part of existing institutions
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, panchayat and ICDS centres.
8.2.9 Venue and time of ICT centres
To ensure that women take full advantage of these ICT centres, it is important to 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), security issues and lack of transport.
8.2.10 Women to be involved in designing ICT policies
Equitable access to ICT technology and the autonomy to receive and produce the 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 women’s overall empowerment.
8.2.11 Participation of different agencies needed
For the effective implementation of ICT projects, no single agency can effectively deliver this critical input. Need for a proactive participation by the private sector, NGOs and other civil society organisations are being increasingly felt.
8.2.12 Small and medium rural enterprises
For the economic empowerment of rural and deprived women, the establishment of ‘Small and Medium Enterprises’ (SMEs), Small Scale Industries (SSIs), etc. can be encouraged.
8.2.13 Project planning
Gender concerns have to be addressed in initial stages of ICT projects to ensure 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 study undertaken by the World Bank in 2002 of 80 ICT projects concluded that 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 earliest phases of project design is imperative.
8.2.14 Funds for ICT for development
Gender perspectives should be incorporated into the allocation and implementation of funds to support ICT for development. Issues of importance include increasing the allotment of funds for women‐specific projects and the consideration of gender perspectives in all aspects of the management of such funds. Attention should be paid to ensuring that sufficient funds are directed to capacity building which is accessible to women and men.
8.2.15 Gender and ICT awards10
To create awareness and encourage ICT programmes and managers, it is necessary to give gender ICT awards. In this direction gender equality advocates have also initiated awards to highlight good practices. Gender equality advocates have also initiated awards to highlight good practices. The gender and ICT awards programmed was inaugurated during the WSIS in Geneva in December 2003. This awards programme was conceived to recognise gender and ICT initiatives globally and provide further impetus for mainstreaming gender perspectives in the field of ICT.
8.2.16 Research on gender equality and ICT
Research on gender equality and ICT gives vital contribution for the policy making. Continued research on gender equality and ICT issues at the national regional and global levels and documentation of good practices also contributes 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 importance of research to understand the grassroots level hurdles and provides proper direction to design effective ICT policies for the empowerment of rural and deprived women.
20
Notes:
1 Dedicated leaders in Maharashtra and Gujarat have contributed to the development of several successful co‐operatives; late Tatyasaheb Kore is one of them who developed Warana region through different co‐operatives. Warana Wired Village Project (WWVP) was implemented by Prime Ministers Special IT Task Force in 10th August, 1998, which is known as one of the oldest ICT project in Asia. The first author of this article did his Ph.D. on WWVP in 2007. 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, semi‐urban and urban environments. 3 DAWN is concept that arose out of southern Women’s 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 people’s needs. Although DAWN emerged out of an NGO forum in 1985, its philosophy reflects the dissatisfaction of many developing countries at the time with unfavourable terms of trade, protectionism and the conditionality of the Word Bank’s Structural Adjustment Programmes (SAPs). 4 SEWA is one of the oldest women organisations which were working in various parts of Gujarat, mainly for rural and deprived women. 5 As a union, SEWA’s current outreach stands at about 530,000 women members. For more information on SEWA’s, see http://www.sewa.org. 6 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 non‐payment of dowry for marriages. Grameen therefore interfered and challenged cultural norms and given power dynamics within the community and the family. 7 Twenty‐five per cent of telephone operators are men, a much larger percentage than their representation in the borrowing population as a whole (5%). As the Grameen Bank chooses the village operators, the male management of the Bank made a gender‐based decision in selection of a disproportionate number of male operators. 8 ‘Grameen Telecom’s Village Phone Programme: A Multi‐media Case Study’. Tele Commons Development Group. March 2000. See: http//:www.telecommons.com/villagephone/9 WENT was an Asia‐Pacific regional annual training for women in the use of ICTs, and was a project that was implemented for five years continuously. At the end of those five years, an award was funded by the United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP) as a form of acknowledgement to the most successful WENT graduate who applied her leanings from WENT for a selected community or within her own organisation. 10 The Gender and ICT Awards is a collaborative project between the APC WNSP and the Global Knowledge Partnership (GKP). The inaugural awards ceremony took place in Geneva in conjunction with WSIS in December 2003, where change initiatives accepted their award as the winner in the category of ‘Advocacy and Networking’. For more information on Nabanna, India and other Gender and ICT Award wining projects, visit http://www.genderawards.net
21
References
Adhiguru. 2004. Institutional Innovations for Using Information and Communication Technology. Policy Brief 18. National Centre for Agriculture Economics and Policy Research, New Delhi.
Ceewa. 2005. Case Study of Rural ICT Initiative in Uganda, Study Report, Uganda. Census. 2001. Directorate Census, New Delhi. Clement, A. 1994. Computing at Work: Empowering Action by ‘low level users’. In
Communication of the ACM. 37, 1:53‐63. Dalal, P 2006. Use of ICT for Women Empowerment in India, Anonymous. Dhameja and Medury. 2002. Information and Communication Technology in the
Globalisation Era: The Socio‐Economic Concerns. Research Report, IGNU, New Delhi. Freire, P. 1972. Pedagogy of the Oppressed. London: Penguin Books. Huyers, S. and Sikuska, T. 2003. Overcoming the Gender Digital Divide: Understanding
ICT’s and their Potential for the Empowerment of Women. United Nations ‐ INSTRAW Virtual Seminar Series on Gender and ICTS. No. 1.
Jain, S. 2006. ICT’s and Women’s 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: Cross‐Disciplinary Perspectives. The World Bank. Washington DC.
Nath, V. 2001. Empowerment and Governance through Information and Communication Technologies: Women’s Perspective, Knownet Initiative, Anonymous.
Patil, D. A. 2006. Information and Communication Technology (ICT) and Rural Development: Lessons from Rural India. IASSI, Quarterly, Vol. 25, No.2.
Patil, D. A. 2006. New Communications Perspectives for Cooperatives and Rural Development. Journal of Communication Studies, Vol. 23.
Patil, D. A. 2007. A Study of Communication Networks and Development in Kolhapur District, Ph. D thesis (Sociology), Submitted to Shivaji University, Kolhapur.
Rowlands, J. 1997. Questioning Empowerment: Working With Women in Honduras, Oxford: Oxfam GB. UNDP. 2001. Human Development Report, Making New Technologies Work for Human Development. London: Oxford University Press.
United Nations. 2005. Women 2000 and Beyond, Division for the Advancement of Women’s Department of Economic and Social Affairs, United Nations.
Vijayalaxmi, P. and Durgabhavani, V. 2006. Internet Use, Indian Culture and Gender Variations, Media Asia, Vol. 33, No. 3‐4.
http://Ict4d.net.inhttp://ehomemakers.nethttp://sewa.orghttp://telecommons.comhttp://wikibooks.com
22
Asia‐Pacific Journal of Rural Development Vol. XIX, No. 1, July 2009
Spatial Assessment of Potential for Groundwater Development in Lower Chao Phraya River Basin
Sudsaisin Kaewrueng*, Mukand S. Babel**, Ashim Das Gupta** and Shahriar M. Wahid**
Abstract
In the Lower Chao Phraya River Basin of Thailand, development of the groundwater resources has played a vital role in meeting the basic water requirement for domestic purposes as well as for economic development through its use in the industrial and agricultural sectors. Amidst reports of falling groundwater levels, land subsidence and salt water intrusion, the potential for groundwater development has been investigated with full consideration to spatial distribution of current uses, exploitability and suitability, and demand for water. Owing to the nature of groundwater development and non‐availability of sufficient pertinent data, several analytical tools and techniques were used to estimate the current uses. While several districts have already reached the critical threshold of exploitation, considerable potential for development exists in one‐fourth of the 127 districts in the basin. A structured approach, based on zoning of potential areas, is recommended for future groundwater development.
1.0 Introduction
Groundwater is an invisible resource. Since it is widely distributed in a range of water‐producing geological structures (aquifers) and much less dependent on recent precipitation than surface sources, it can provide a uniquely reliable source of good‐quality and low‐cost water for domestic, industrial and agricultural purposes. Many Asian countries heavily depend on groundwater for domestic water supply and take advantage of the resource to facilitate economic activity. The rapid industrialisation and urbanisation taking place in many Asian countries, however, have intensified the stress placed on this precious resource. In response to the growing demand for water, groundwater has been depleted through excessive abstraction, and resultant land subsidence has been observed in some areas. Aquifer contamination with various types of pollutants including saltwater intrusion in coastal aquifers also poses serious challenge. The Lower Chao Phraya River basin in Thailand is a case in point.
* Lecturer, Kasetsart University, Bangkok, Thailand. E‐mail: [email protected] ** Asian Institute of Technology, Pathumthani, Thailand. Email: [email protected]; [email protected]; [email protected];
23
The basin, located in the lower central plain, holds the largest storage of groundwater in Thailand (Ramnarong and Wongsawadi 1998). Water is held in confined and unconfined aquifers. The largest of these aquifers – Bangkok Aquifer System – contains eight separate confined aquifers in the Upper Tertiary to Quaternary strata covering Bangkok and its six adjoining provinces: Ayutthaya, Pathum Thani, Nonthaburi, Nakhon Pathom, Samut Sakhon and Samut Prakarn. The ease of exploitation owing to the artesian condition of these aquifers and rapid spread of energised pumping technologies mean that groundwater development has been very rapid in the Lower Chao Phraya River Basin. Currently, it is one of the most vital and valuable resource for agriculture, municipal and rural water supply, and industry in the basin.
UNEP (2003) reported that outside the Bangkok aquifer system, agro‐wells are extensively installed in floodplains and in buried channels of the major rivers for 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 productivity of timely irrigation and security of application. Most agro‐wells are privately owned and often information on their numbers and amount of groundwater extraction is not available or inaccurate. Chulalongkorn University (2002) reported, based on field survey conducted in 1998, that an estimated 17,100 agro‐wells have helped the farmers to cope with erratic rainfall and long periods of dry spell and increase agricultural productivity. Growing three crops in a year or even seven crops in two years has become possible due to the use of these agro‐wells.
Groundwater is also extensively used by industries in Bangkok and surrounding areas because of self‐control over water supply and less‐than‐desired level of access to public piped water, unreliable or insufficient or low quality of piped water supply, and lower price of groundwater compared to that of piped water supply (DGR 2004). The use of groundwater for domestic supply is widespread 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 Provincial Waterworks Authority (PWA) provides water in the municipal areas of remaining 11 provinces in the basin area (Martin et al. 2004). Outside MWA and PWA service areas, piped water systems are managed by four government 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 management in Thailand.
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, over‐pumping has resulted in the falling of groundwater level up to 60m in Bangkok Metropolitan area and its vicinity. 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 environmental threats in the future (UNEP 2001).
Objectives of the Study: Considering the significance of sound groundwater management on social and economic fabric of the Lower Chao Phraya River Basin, this study is undertaken to spatially assess groundwater use potential in the basin.
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 development in the basin.
2.0 Methodology of Investigation
2.1 Study area
The study area covers 12 provinces in the Lower Chao Phraya River basin: Chai Nat, Sing Buri, Ang Thong, Lop Buri, Suphan Buri, Ayutthaya, Pathum Thani, Nonthaburi, Nakhon Pathom, Samut Sakhon, Samut Prakarn and Bangkok (Figure 1). The basin represents an area of Quaternary deposits of silt 15‐30m in depth overtopping the soft marine clays laid down when the area was a vast bay of the South China Sea about 6,000 to 8,000 years ago (Sinsakul 1997 reported in Kohnhorst et al. 2002).
25
Figure 1: The Study Area
The aquifer system is largely recharged by rainfall and seepage from rivers. In the lower central plain where Bangkok is situated, about half of the area is covered by thick marine clay, and it is estimated that only 5‐6 per cent of rainfall reaches the aquifer. Small sub‐basins that have been identified during petroleum
26
exploration in the central plain contain alluvial fans and fluviatile sediments that are also good aquifer systems (Martin et al. 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 records in 182 stations during 1952‐2002, it is observed that the average annual rainfall is 1,087mm and varies between 680 and 1,520 mm from year to year. The total rainfall is normally higher in the Bangkok area, and less in the areas located farther inland. About 87 per cent of the total annual rainfall occurs in the six months of the rainy season (May‐October). Often, September gets the highest rainfall in a year, while January gets the lowest.
The study is divided into two parts: Firstly estimation of current quantity and extent of groundwater exploitation, and secondly, assessment of groundwater development potential.
2.2 Estimation of groundwater exploitation
Estimation of groundwater exploitation in the study area is complicated by the different patterns of groundwater use and categories of installed wells as well as variation in corresponding available data type. Table 1 lists different types of registered (with Department of Groundwater Resources, DGR) and non‐registered wells used to abstract groundwater in 2002 along with the number of each kind of wells.
The private and public wells are classed based on ownership as well as well depth. Shallow wells are defined as those wells that penetrate up to 30m depth and deep wells penetrate beyond 30m. According to the Groundwater Act 1977, deep wells are required to be registered with permitted pumpage for a specified purpose (domestic, agricultural and industrial wells) while shallow wells need not be registered. The non‐registered wells constitute the majority (89.3%) of the total groundwater wells. Details of different types of wells used in three different use categories are summarised here:
− Domestic use: Normally, shallow wells are used for domestic water supply in rural areas. They are both private‐ and public‐owned. Though private deep wells or tube‐wells are reported in the National Rural Development (NRD2C) Database developed by the Community Development Department (CDD), Ministry of Interior (MOI), Thailand, often they are not registered with the DGR. The public deep wells developed by government agencies
27
Table 1: Number of Groundwater Wells in Study Area
28
Registered with DGR 1/ Non‐registered with DGR 2/ Category of well / utilisation Domestic Agricul‐
tural Indus‐trial Domestic Agricul‐
tural
Province Private deep well
Private deep well
Private deep well
Sub‐total
Private shallow well
Public shallow well
Private deep well
Public deep well (VPWS)
PWA 3/
PPWS 4/
Agro‐wells 5/
Sub‐total
Total number of wells
Percent‐age
Chai Nat 6 0 35 41 1,970 615 7673 1,649 1 0 ,260 16,168 16,209 18.10 Sing Buri 326 109 122 557 610 167 4394 738 8 1 745 8,663 9,220 10.30 Ang Thong 46 27 81 154 813 129 5744 630 11 0 ,782 10,109 10,263 11.46 Lop Buri 14 18 136 168 4,057 695 2362 1,825 16 0 ,854 11,809 11,977 13.38 Suphan Buri 98 51 206 355 6,057 1,235 6553 1,875 17 0 ,821 19,558 19,913 22.24 Ayutthaya 392 30 492 914 2,489 251 371 927 50 15 24 4,327 5,241 5.85 Pathum Thani 804 28 789 1,621 890 22 146 341 57 37 2 1,565 3,186 3.56 Nothaburi 252 4 230 486 174 2 12 33 0 2 5 238 724 0.81 Nakhon Pathom 144 25 468 637 1,073 168 1849 1,681 30 15 941 5,757 6,394 7.14 Samut Sakhon 404 13 1,261 1,678 146 27 178 366 44 4 96 861 2,539 2.84 Samut Prakarn 444 3 1,142 1,589 38 15 172 553 0 2 9 829 2,418 2.70 Bangkok 643 51 684 1,378 0 0 0 72 0 10 0 82 1,460 1.63 Total 3,573 359 5,646 9,578 18,317 3,326 29,454 10,690 234 86 17,859 79,966 89,544 100.00
Percentage 3.99 0.40 6.31 10.70 20.46 3.71 32.89 11.94 0.26 0.10 19.94 89.30 100.00 Remark: PWA is Provincial Waterworks Authority which provides piped water for domestic and industrial purposes in municipal areas
PPWS is private piped water system VPWS is village piped water system
Source: 1/ data obtained from DGR 2/ data obtained from NRD2C database 3/ data obtained from PWA 4/ data obtained from Department of Water Resources (DWR) 5/ data obtained from Royal Irrigation Department (RID) and integrated with NRD2C database
(PWD, DOH and ARD are 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 the MONRE.
− Agricultural use: Most of the agro‐wells are shallow type and serve extensively as source of supplemental and small scale irrigation.
− Industrial use: There are two types of groundwater use for industrial water supply: one is groundwater extraction by private deep wells owned by individual factories and the other is extraction by MWA and PWA to supply water through their piped supply systems.
With much difference in data availability of various categories of groundwater wells and uses, different types of estimation tools and techniques are adopted to estimate exploitation for different types of wells/uses. These are explained here.
Groundwater use by registered wells: The DGR provides registration of private deep wells in the study area. During the registration process, the well owners specify purpose (domestic, agricultural and industrial) and extent of proposed groundwater use. Based on these specifications, the DGR registers the wells for permitted pumpage rates. Available data on the number of wells and corresponding permitted pumpage rate can be used to calculate the groundwater 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 groundwater use by registered wells.
To estimate GPCs, actual and permitted pumpage data from 697 domestic, 73 agricultural and 4,434 industrial wells in seven provinces were obtained from the DGR and analysed. The overall sample size was taken within significant confidence level at 99 per cent (standard error of 0.01). The annual average GPCs were found to be 0.859, 0.669 and 0.655 for domestic, agricultural and industrial wells respectively with an overall average of 0.691 (Table 2). This fact indicates that the actual pumpage in the domestic water used sector is highest but remains less than the permitted pumpage.
29
Table 2: Estimated Groundwater Pumpage Coefficients (GPC)
Domestic wells Agricultural wells Industrial wells Province
No. of wells GPC No. of wells GPC No. of wells GPC
Ayutthaya 120 0.75 10 0.732 455 0.678 Pathum Thani 117 0.943 9 0.722 619 0.534 Nonthaburi 40 0.515 1 0.11 225 0.72 Nakhon Pathom 139 1.005 25 1.189 738 0.627 Samut Sakhon 1 0.972 ‐ ‐ 1,065 0.82 Samut Prakarn 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
The estimated GPCs compare well with earlier study by the Japan International Cooperation Agency, JICA (1995) which has estimated overall GPC of 0.67 for Bangkok and its vicinity areas.
Groundwater use by non‐registered wells: Non‐registered wells include private non‐registered shallow, deep and agro‐wells as well as wells managed by organisations other than DGR. Data on these wells are limited. Several techniques were used to estimate groundwater use by these wells depending on type of available data. Details of these techniques and their application are provided here:
− Domestic use through private/public shallow or deep non‐registered wells: These are used to supply domestic water in rural areas. Groundwater use by these wells was estimated by multiplying the pumping rate (0.71m3/day/well) proposed by Chulalongkorn University (2002) and number of such wells.
− Domestic use through VPWS and PPWS: Most rural inhabitants are supplied by the public deep wells through VPWS. Others use PPWS supplied by private deep or shallow wells in their communities. Groundwater exploited by VPWS or PPWS is computed by analysing data on number and average 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 were provided by the Department of Water Resources (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 offices to estimate groundwater abstraction by them.
− 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 records on groundwater pumpage and surface water supply as wells as the total amount of piped water supplied to their clienteles. The ratios of ground 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 taken as industrial water use from groundwater sources.
− Agricultural use: To determine agricultural use of groundwater, field surveys were carried out to estimate pumping hours and rates from agro‐wells in rain‐fed and irrigated paddy farms which dominate agriculture in the study area. Pumping rates were estimated for 30 agro‐wells during the wet and dry seasons of 2004‐2005. Additionally, 19 and 14 farmers in rain‐fed and irrigated areas respectively were interviewed in August 2004 to validate estimation of pumping hours and rate. This information along with data on number of agro‐wells was used to estimate groundwater use through agro‐wells.
2.3 Assessment of groundwater development potential
Groundwater use varies considerably over space in the study area. In the lower part of the study area extensive use of groundwater is reported and has been attributed for various environmental problems, such as lowering of groundwater 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, defining critical zones, licensing and metering of groundwater wells, 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, and two variables related to future water demand of the domestic sector. These are explained here.
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 re‐aggregated at district level within Geographical Information System (GIS).
Sustainable yield or safe yield is the maximum amount of groundwater that can be continuously withdrawn without critically lowering the groundwater level or 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 unimportant aquifers, the natural recharge amounts to 5, 3 and 2 per cent of the annual rainfall respectively (Ramnarong and Wongsawadi 1998). The hydrogeological map was originally created in digital format in GIS; it was overlain and intersected with the administrative district boundaries of the study 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 rainfall for each pixel was used to estimate the potential aquifer yield of the pixel polygon. Sustainable aquifer yield of a district was then computed based on proportional aerial contribution of individual polygons yields to the district based on the following equation:
YT = a1y1 + a2y2 + ... + anyn where, YT = total district sustainable aquifer yield (m3/y); an = area within the district containing a particular hydrogeological unit (km2); and yn = sustainable aquifer yield of a selected pixel (m3/y/km2) based on the natural recharge from rainfall.
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 use potential in the district is considered to be very good while if its value exceed 0.90 the district has the poorest potential for development. The UTY ranges of 0.25‐0.50 and 0.51‐0.90 are considered to indicate good and moderate potential for development. A UTY map was prepared by overlaying groundwater use and yield maps. However, the uncertainty in aquifer systems has not been evaluated and the physical properties of field well studies are thereby not used.
Table 3: Guidelines for Assessing Potential for Groundwater Development Based on Use to Yield Ratio
Use to yield ratio Potential for development
< 0.25 Very good 0.25 ‐ 0.50 Good 0.51 ‐ 0.90 Moderate > 0.90 Poor
Source: Wehmann et al. 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. Considering the fact that groundwater is the major source of domestic water, it is argued that districts ‐ where inhabitants have less access to domestic water ‐ are better candidates for groundwater development. It is proposed that districts where less than 50 per cent inhabitants have access to domestic water is graded as very good potential areas for groundwater development while districts with 80 per cent inhabitants having access to domestic water should be graded as poor for groundwater development (Table 4).
Table 4: Guidelines for Assessing Potential for Groundwater Development Based on Demand for Development Arising from Lack of Access to Domestic Water
Percentage of population with access to
domestic water Potential for development
<50 % Very good 50‐65 % Good 65‐80 % 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. Corresponding data for the urban areas were collected from MWA and 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 study area were extracted from NRD2C database.
Table 5: Guidelines for Assessing Potential for Groundwater Development Based on Demand for Development Arising from Lack of Sufficient Domestic Water
Percentage of village/district with more than 10% households getting insufficient water
Potential for development
>10 % Very good 7.5‐10 % Good 5‐7.5 % Moderate <5 % Poor
Furthermore, it is argued that even if the inhabitants have access to domestic water, they may not have sufficient quantity to meet basic domestic needs. 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 village lack sufficient water, it is considered water insufficient village while if less than 10 per cent HH in a village lack sufficient water it is termed water sufficient village (CDD 2003). Since district is taken as the lowest spatial management unit in this study, village data were scaled up to district level based on the percentage of water insufficient village in a district (Table 5).
Table 6: Guidelines for Assessing Potential for Groundwater Development Based on Exploitability
Extraction depth (m) Well yield (m3/h)
<50 50‐150 >150 >10 Very good Very good Good 5‐10 Very good Good Moderate 2‐5 Good Moderate Poor <2 Moderate Poor Poor
34
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 quantitatively assess exploitation potential are shown in Table 6.
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 during the analysis.
Table 7: Guidelines for Assessing Potential for Groundwater Development Based on Suitability of Groundwater
Total dissolved solids (TDS) (mg/L) Specific capacity
(m3/h/m) <500 500‐1,500 >1,500 >3.5 Very good Very good Good
2.0‐3.5 Very good Good Moderate 0.5‐2.0 Good Moderate Poor <0.5 Moderate Poor Poor
Overall groundwater development potential: For quantitative assessment of potential for groundwater development, a numerical value is assigned based on the level of potential as very good (4), good (3), moderate (2) and poor (1) for all five indicators and then they were aggregated to provide an overall pictures of districts. All five indicators were given equal weights during aggregation to avoid biasness of decision. It is proposed to identify four classes of overall potential for groundwater development: very good, good, moderate and poor. If a district secured overall score between 16 and 20, it is considered to have very good potential for groundwater development. Likewise, scores of 12‐16, 8‐12 and less than 8 were taken as indications of good, moderate and poor potential of groundwater development respectively. Resulting overall groundwater potential was mapped to visually aid management decision.
3.0 Results and Discussion
3.1 Groundwater exploitation
The total groundwater exploitation was estimated to be 3.625 million m3/d or 1,323 million m3 in 2002. Of this, 1.432 million m3/d was abstracted through deep
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 (Table 8).
Agriculture represents the largest user of groundwater. It consumed 1.767 million m3/d (48.75%) through combined use of private deep‐wells and agro‐wells. Agro‐wells represent the preferred mode of abstraction in agriculture. The average pumping rate of agro‐wells was estimated to be 385 litres/minute for wells in the rain‐fed area while it is 800 litres/minute in the irrigated areas. On average agro‐wells in the rain‐fed area were operated for 1,110 and 1,450 hours during the wet and dry seasons respectively. On the other hand agro‐wells in the irrigated area were operated for 285 and 385 hours during the wet and dry season respectively. Field survey data revealed that about 65 per cent of installed agro‐wells in the irrigated area remains operational during the wet season. Industrial and domestic sectors consumed 0.969 and 0.888 million m3/d of groundwater annually representing 26.74 per cent and 24.51 per cent of the total groundwater use respectively (Table 8).
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 service providers. Among service providers, MWA and PWA distributed 74,778 m3/d (2.06%) and 55,255 m3/d (1.52%) of groundwater to domestic and industrial sectors respectively during 2002. Non‐registered private deep/shallow and public shallow wells together abstracted an estimated 38,004 m3/d (1.05%) of groundwater for domestic purpose. VPWS and PPWS supplied 262,099 and 4,256 m3/d of groundwater respectively for domestic use representing 7.35 per cent of the total abstraction.
Among provinces within the study area, groundwater use varied from 2.13 to 13.55 per cent. Nonthaburi Province consumed lowest groundwater while Suphan Buri province the highest. Provinces in the upper part of the study area are also high groundwater users (Figure 2 and Table 1).
On the other hand, in the lower part of the study area higher amount of groundwater is abstracted for domestic and industrial purposes. Pathum Thani and Samut Sakhon are the largest groundwater users for domestic and industrial purposes, respectively.
36
Table 8: Estimated Groundwater Use (m3/d) in 2002
Registered with DGR Non‐registered with DGR
Domestic Agricul‐tural
Indus‐trial
Province Domestic Agri‐
cultural Indus‐trial
Sub‐total
Private shallow well
Public shallow well
Private deep well
Public deep well (VPWS)
PWA1/PPWS
MWA1/
Sub‐total
Agro‐wells
PWA & MWA
Sub‐total
Total volume of use
Percent‐age
Chai Nat 430 0 2,869 3,299 1,399 437 6,063 15,795 0 0 ‐ 23,694 406,181 0 429,875 433,174 11.95 Sing Buri 1,362 728 6,277 8,367 433 119 3,740 11,822 8,287 486 ‐
‐ ‐ ‐ ‐ ‐
‐
‐
‐ ‐
24,887 226,718 3,603 255,208 263,575 7.27 Ang Thong 133 264 2,180 2,577 577 92 4,384 15,682 4326 0 25,061 245,213 1,354 271,628 274,205 7.57 Lop Buri 508 2127 17420 20,055 2,880 493 2,139 27,688 2,514 0 35,714 322,098 1,046 358,858 378,913 10.45 Suphan Buri 1,641 590 3,736 5,967 4,300 877 4,818 44,521 11,150 0 65,666 411,754 7,745 485,165 491,132 13.55 Ayutthaya 38,862 1,498 124,102 164,462 1,767 178 147 38,492 5,351 402 46,337 20,665 2,738 69,740 234,202 6.46 Pathum Thani 204,181 891 159,947 365,019 632 16 28 19,805 1,159 400 22,040 5,922 1,241 29,203 394,222 10.88 Nothaburi 31,302 47 29,397 60,746 124 1 0 14,736 400 03/ 15,261 1,224 0 16,485 77,231 2.13 Nakhon Pathom 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
Samut Sakhon 58,945 169 246,456 305,570 104 19 39 10,597 23,047 ‐
835 34,641 13,631 25,502 73,774 379,344 10.47 Samut Prakarn 53,740 23 147,586 201,349 27 11 5 19,456 402 03/ 19,901 7,246 0 27,147 228,496 6.3 Bangkok 93,439 1,250 107,533 202,222 0 0 0 1,800 495 7,6033/ 9,898 0 3,630 13,528 215,750 5.94 (m3/d) 509,155 8,655 914,199 1,432,009 13,005 2,362 22,637 262,099 67,175 4,256 7,603 379,137 1,758,532 55,255 2,192,924 3,624,933 100.00
(106 m3/y) 185.8 3.2 333.7 522.7 4.7 0.9 8.3 95.7 24.5 1.6 2.8 138.4 641.9 20.2 800.4 1,323.1Percentage 14.05 0.24 25.22 39.50 0.36 0.07 0.62 7.23 1.85 0.12 0.21 10.46 48.51 1.52 60.50 100.00
Note: MWA is Metropolitan Waterworks Authority PWA is Provincial Waterworks Authority
PPWS is private piped water system VPWS is village piped water system or public deep wells. 1/ is the whole abstraction identified to be for domestic purposes 2/ is the whole abstraction identified to be for business and industrial uses 3/ stands for MWA supply in Bangkok, Nonthaburi and Samutprakarn provinces
37
37
Figure 2: Spatial Distribution of Groundwater Well Installation
3.2 Groundwater development potential
Groundwater development potential was assessed considering aquifer yield, its exploitability, use and quality of supply, and domestic water demand. Yield is taken as equal to natural recharge. Occurrence of alluvial aquifer (Qfd) through majority of the study area generally indicates condition for high recharge rate (Figure 1).
A natural recharge of 4.825 million m3/d, or about 7.9 per cent of average annual rainfall of 1,087 mm, was estimated. In general, groundwater exploitability and suitability are good in most of the districts. Of the 127 districts in the study area, 118 districts have exploitability exceeding 5m3/h within 150m from the ground surface. Groundwater suitability, in terms of TDS and Specific Capacity, shows that 122 districts have Specific Capacity exceeding 2m3/h/m while still maintaining TDS below 1,500 mg/L, the threshold for freshwater.
38
Figure 3: Groundwater Development Potential Based on (a) Use to Yield (UTY) Ratio, (b) Demand for Water due to insufficient domestic water supply, (c) Groundwater Exploitability, and (d) Overall Potential for Future Development
(a) (b)
(c) (d)
39
Table 9: Number of Districts in Provinces Differentiated Based on Potential for Groundwater Development
Number of districts based on Province No. of dis‐tricts
Potential for development Use to
yield ratio
Lack of access to domestic water source
Lack of sufficient amount of domestic water
Exploi‐tability
Suita‐bility
Integrated Potential
Very good 1 ‐ 4 4 6 ‐ Good ‐ 1 ‐ ‐ 2 4
Moderate ‐ ‐ 2 4 ‐ 4
Chai Nat
8
Poor 7 7 2 ‐ ‐ ‐ Very good ‐ ‐ ‐ 6 6 ‐
Good 1 ‐ ‐ ‐ ‐ 1 Moderate ‐ ‐ ‐ ‐ ‐ 5
Sing Buri
6
Poor 5 6 6 ‐ ‐ ‐ Very good ‐ 1 ‐ 6 7 ‐
Good 3 ‐ 1 1 ‐ 4 Moderate ‐ 1 ‐ ‐ ‐ 3
Ang Thong
7
Poor 4 5 6 ‐ ‐ ‐ Very good 4 1 5 5 10 3
Good 3 ‐ ‐ 5 1 7 Moderate 1 5 1 1 ‐ 1
Lop Buri
11
Poor 3 5 5 ‐ ‐ ‐ Very good 4 1 4 8 10 4
Good 2 2 2 2 ‐ 6 Moderate 1 1 3 ‐ ‐ ‐
Suphan Buri
10
Poor 3 6 1 ‐ ‐ ‐ Very good 5 ‐ 3 11 16 1
Good 5 ‐ 2 5 ‐ 14 Moderate 4 ‐ 2 ‐ ‐ 1
Ayutth‐aya
16
Poor 2 16 9 ‐ ‐ ‐ Very good ‐ ‐ 1 4 6 1
Good 3 ‐ ‐ 3 1 3 Moderate 1 3 2 ‐ ‐ 3
Pathum Thani
7
Poor 3 4 4 ‐ ‐ ‐ Very good 1 ‐ ‐ ‐ 5 ‐
Good 2 ‐ 2 6 1 4 Moderate 3 ‐ ‐ ‐ ‐ 2
Non‐thaburi
6
Poor ‐ 6 4 ‐ ‐ ‐ Very good 3 ‐ ‐ 3 7 ‐
Good 1 ‐ 1 4 ‐ 6 Moderate 1 1 1 ‐ ‐ 1
Nakhon Pathom
7
Poor 2 6 5 ‐ ‐ ‐ Very good ‐ 1 1 1 2 ‐
Good ‐ ‐ 1 2 1 3 Moderate 1 ‐ ‐ ‐ ‐ ‐
Samuth Sakhon
3
Poor 2 2 1 ‐ ‐ ‐ Very good ‐ ‐ 5 1 6 ‐
Good 2 ‐ ‐ 4 ‐ 5 Moderate ‐ 2 ‐ 1 ‐ 1
Samuth Prakan
6
Poor 4 4 1 ‐ ‐ ‐ Very good 23 ‐ ‐ 2 3 ‐
Good 5 ‐ ‐ 35 32 22 Moderate 4 ‐ ‐ 3 3 17
Bang‐kok
40
Poor 8 40 40 ‐ 2 1
40
However, the estimated overall Use to Yield (UTY) ratio (around 0.75) is a pointer for opting strategic management regime in the future. Comparatively higher UTY values (UTY > 0.9) in the lower (east of Bangkok and west of Samut Sakhon, Nakhon Pathom) and middle of the upper parts (Pathum Thani, Ang Thong, Suphan Buri, Sing Buri and Chai Nat) of the study area indicate higher exploitation by industrial wells and/or agrowells (Figure 3 (a)).
Inhabitants in seven districts, out of 127 in the study area, have poor and moderate access to domestic water indicating that these districts hold very good or good potential for groundwater development (Table 9).
In 32 districts inhabitants have insufficient supply of domestic water indicating that they will have demand for groundwater development in the future. Overall, considering all the indicators, 87 districts have good or very good potential for groundwater development (Figure 3 (d) for location). However, a close scrutiny reveals that in 16 districts of these 87 districts UTY ratio is very high imposing restriction on further groundwater extraction.
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 Lower Chao Phraya Basin. Analysis and results indicated that an estimated 1,323 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 population in the rural areas are dependent on groundwater for domestic use.
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 indicator‐based evaluation scheme was applied to spatially differentiate the potential for development. Results point out 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 cent of districts have very good potential for further development.
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 recommended to be estimated based on the hydrogeology of the basin.
References CDD (Community Development Department). 2003. Guideline for National Rural
Development Database (NRD2C) questionnaire. Community Development Department, Ministry of Interior, Bangkok (in Thai). Pp. 71.
Chulalongkorn University. 2002. Groundwater management in upper part of the lower Chao Phraya River basin. Final Main Report (in Thai). Pp. 431.
DGR (Department of Groundwater Resources). 2002. Effects of groundwater recharge on land subsidence and groundwater quality: Mathematical model study, Draft Final Main Report (in Thai). Department of Groundwater Resources, Ministry of Natural Resources and Environment, Thailand.
DGR (Department of Groundwater Resources). 2004. Effect of groundwater over pumping mitigation: Mathematical model study, Final Report. Department of Groundwater Resources, Ministry of Natural Resources and Environment, Thailand.
JICA (Japan International Cooperation Agency). 1995. The study on management of groundwater and land subsidence in the Bangkok Metropolitan area and its vicinity, final main report. Pp. 376.
Kohnhorst, A., L. Allan, P. Pokethitiyoke, and S. Anyapo. 2002. Groundwater Arsenic in central Thailand. In: Sustainable Environmental Sanitation and Water Services. Preprints of 28th WEDC Conference Kolkata (Calcutta) India. Pp. 123‐125.
Martin, R., R. Savage, and R. Pyvis. 2004. A study to develop a strategy to expand rural water infrastructure within the restructured water and wastewater sector in Thailand, Ministry of Finance, Draft Final Report. Public‐Private Infrastructure Advisory Facility (PPIAF). Pp. 105.
Ramnarong, V. and S. Wongsawadi 1998. Potential of groundwater in Thailand, Journal of Hydrologist Assembly Vol. 2, Pp 240‐276 (in Thai).
Sinsakul, S. 1997. Country Report: Late Quaternary Geology of the Lower Central Plain, Thailand. Inter. Symp. on Quaternary Environmental Change in the Asia and Western Pacific Region, Oct. 14‐17, 1997, U. of Tokyo, Tokyo, Japan.
UNEP (United Nations Environment Programme). 2001. Bangkok state of the environment, Environmental Quality Management and Control Division, Office of the Permanent Secretary for the Bangkok Metropolitan Administration. BMA. Pp. 104.
UNEP (United Nations Environment Programme). 2003. Groundwater and its susceptibility to degradation: a global assessment of the problems and options for management, Div. of Early Warning and Assessment, UNEP, Kenya. Pp. 138.
Wehmann, H. A., S. V. Sinclair, and T. P. Bryant. 2003. An analysis of groundwater use to aquifer potential yield in Illinois, Groundwater Section, Illinois State Water Survey. Pp. 30.
42
1
Aspltapal2ohps
Bgr
∗
Asia‐Pacific Journal of Rural Development
Vol. XIX, No. 1, July 2009
Migration of Rural People to Urban Areas: A Study of Three Upazilas in Mymensingh
Gulsan Ara Parvin∗, Mamun Muntasir Rahman*, Farhana Yasmin*
Abstract
This study is an approach of that facts finding, which would create a basis for further policy formulation to convert migration as a catalyst of development. Through empirical studies it intends to seek answer of two queries that are ‘who migrate from rural to urban areas and what are the differences between migrant and non‐migrant households living in the place of origin i.e. rural areas. It has been found that the overwhelming majorities (60%) are migrating within the age16‐35 years to seek job, and they are primarily migrating to Dhaka city. It has been also noticed that there is wide difference between migrant and non‐migrant families in different socio‐economic indicators. Families having migrant members are in a better position in housing condition, income and land ownership.
.0 Introduction
ccording to scholars, definition and classification of migration is not easy and traight forward. In the traditional definition, migration means movement of eople from one place to another, temporarily or permanently in search of better ife, livelihood or to avoid threat of life and livelihood (Siddiqui 2005). Whatever he definition is migration is one of the crucial phenomena in both developed nd developing countries though the nature and facts are different in different laces. In Asia among the developing countries Bangladesh bears significant ttention in the issue of migration, which is recognised as an important ivelihood strategy for the people of Bangladesh (Nazem 1998 and Siddiqui 005). Being one of the members of Least Developed Countries (LDCs), low level f economic growth, lack of employment generation, rapid population growth, igh population density, wide rural‐urban disparities, frequent natural disaster, olitical unrest and corruption have made the people of Bangladesh bound to earch for a new place for livelihood.
oth in the matter of internal and international migration, Bangladesh attain, reat concern of policymakers and researchers. Internal migration in the form of ural to urban migration (4.5%) is primarily responsible for the rapid urban
Urban and Regional Planning Department, BUET, Dhaka. E‐mail: [email protected]; [email protected]; [email protected]
43
growth, which is estimated at 3.2 per cent within year 1991 to 2001 (BBS 2003 and Nazem 1998). Study conducted by Skeldon (2005) has claimed that in Bangladesh 40 per cent of rural households send adult members to seek work in town and in some areas this figure is more than 80 per cent. On one side rural‐urban 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 2005 and Daily Ittefaq, 27 June 2007).
Therefore, due to being key feature of national socio‐economic structure, migration issues should be diagnosed properly and adequately. At this stage a number of scholars dominated by Rita Afser and Tasneem Siddiqui have made great contributions. But still it is noticed that like many other LDCs Bangladesh has poor data base and researches related to various dimensions of migration (Afsar 2000). Therefore, though migration issue is well addressed by some legendary scholars, it is not old yet and migration still comprises wide interest of researchers.
2.0 Objective of Study
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 Bangladesh is one of such countries (Skeldon 2005). It is advised that in the light of the changes sweeping through rural Bangladesh, the link between migration and development is needed to be re‐examined (Afsar 2005). Before the initiative of exploring this link it is necessary to be familiar with the forms and facts of rural migration. In this context, this study aims to investigate the issues like who migrate, where they migrate, for what purpose and what are the differences between migrant and non‐migrant families in the place of origin. This study is an approach of these facts finding, which would create a basis for further policy formulation to convert migration as a catalyst of development.
3.0 Methods of Investigation
Being concern to the limitation of time and resources this paper delimits its scope of investigation in the three upazilas (administrative area below the district level) of Mymensingh, which is one of the largest districts of Bangladesh, is not well addressed in migration‐related studies. Therefore, this study has concentrated to 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
three upazilas named Muktagacha, Haluaghat and Gafargao have been selected as study areas from three sides of the district. These upazilas have been selected 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, one has large number of families having migrant members, another has medium concentrations of migrant families and third one has low concentration of migrant families. After finalising the selection of upazilas 180 households (60 from each upazila) have been selected randomly for questionnaire survey. No statistical method has been followed to determine sample size
Through empirical studies using field investigation and questionnaire survey it intends to seek answer of two broad questions. These queries are:
i) Who migrate from rural areas; and
ii) What are the differences between migrant and non‐migrant families?
To examine the first query this paper examines various forms of migrants. Here, forms indicate issues such as the ratio of migrant and non‐migrant 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 socio‐economic issues vary between migrant family (families having at least one migrant member) and non‐migrant family. At this stage it would be tried to depict comparative picture of migrant and non‐migrant households in the case of various socio‐economic issues such as family type and size, housing condition, income and occupation, land ownership and access to various services and facilities. It should be noted that here survey has been conducted at the place of origin, where family members of the migrants live. Hereunder, the following section analysis based on primary data has been presented.
4.0 Results and Discussions
4.1 Who migrate from rural areas?
In Bangladesh there are already some researches that have investigated various characteristics of migrants and have given the answer of the question‐who migrate from rural areas. But there are little attempts to seek this answer for all 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 research are different in some cases of
45
their socio‐economic and physical condition though there are similarities in many aspects. In order to present the overall condition of Mymensingh District, it has been tried to select three different upazilas having different settings. Though socio‐economic, physical and even political context of a locality is very important in occurrence of migration from an area here the characteristics of migrants have not been presented locality wise.
4.1.1 Extent of migration in study area
In order to know the extent of migration in the locality, it has been tried to identify the number of households having at least one migrant member and the 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) member in the family is termed as Non‐Migrant Household.
Table 1: Distribution of Household According to Migration
Migration Number of household Percentage Household having migrant member 71 39 Household having no migrant members 109 61 Total 180 100
Source: Field Survey, January, 2007
At the introduction section it is already mentioned that in Bangladesh 40 per cent of rural households have adult members who have migrated to towns for livelihood. This study area is not exception from that. The table demonstrates that 39 per cent of the households have migrant members. It should be noted that there are total 108 migrant members migrated from 71 households. It denotes that average 1.5 persons are migrating from each household.
Figure 1 presents interesting result by showing variation in the number of migrant and non‐migrant 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. Due to the natural disaster people of this area have higher tendency to migrate for seeking job. It is interesting that though they are vulnerable in natural disaster, they are not migrating with full family rather one or 2 members are migrating for their secured livelihood.
46
Figure 1: Migrant and Non-Migrant Households in Different Upazilas
15%63%
60%85%
37%
40%0%
20% 40% 60% 80%
100% 120%
Names of Studied Upazilas
% of Non- Migrants Households
% of Migrant Households
Perc
enta
ge
Haluaghat Muktagacha Gafargao
Note: Migrant family means the family has at least one migrant member in the family Non‐Migrant Family means the family which has no migrant member Source: Field Survey, January, 2007
Unlike Gafargao, Muktagacja has the lowest number of migrant households, which is only 15 per cent of the total households. Muktagacha is a well served upazila of Mymensingh. It is one of the oldest upazilas consists of 1 municipality (town) and 10 unions (rural areas). This upazila has 4 college and 40 high schools which is exceptionally better that many upazilas of Bangladesh. 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 facilities have significantly reduced the out‐migration from this area.
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 non‐Muslim community and these people migrate for better livelihood and education.
During examining the extent of migration from the study area it has been clearly noticed that social and economic potentialities of an area primarily determine the area’s level of out migration. Next hereunder, the characteristics of migrant members have been presented.
47
4.1.2 Characteristics of migrant
During identifying the characteristics of migrant their age, education level, time period of migration, purposes and destination of migration have been examined. Following sections presents migrants characteristics.
Table 2: Distribution of Migrant According to Their Age and Education Level
Age group Percentage Literacy Percentage
Less than 15 1.85 Illiterate 12.03
16 to 25 26.85 Signature 0.92
26 to 35 34.26 Primary 5.55
36 to 45 18.52 Madrasha 0.92
46 to 55 12.04 Secondary 21.3
56 to 65 3.70 Higher Secondary 23.15
Above 65 2.78 Graduate 28.7
Post‐Graduate 7.4
Total
100.00
Total 100
Note: Total migrant members are 108 and all are male members Source: Field Survey, January, 2007
4.1.2.1 Age and literacy level of the migrants
The age group analysis of the out‐migrated 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, 16‐35 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, usually incurred by this age class.
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
search for job to earn a better livelihood. This also depicts the lack of sufficient higher level educational institutions (especially college and tertiary institutes) and insufficient service industry for the educated mass.
Trends of Migration
e Period
Fig.2 No. of Migrated Person in Different TimFigure 2: Number of Migrated Persons in Different Time Period55
13
28
831
0102030405060
1981
– 19
85
1986
– 19
90
1991
– 19
95
1996
– 20
00
2001
- 200
5
Above
2006
Time Period
No.
of M
igra
nt (p
erso
n)
Source: Field survey, 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 very low in the decade 1980‐1990 when only around 4 per cent of total present day migration was taken place (total 108 persons). A sudden boom in shift of local inhabitants to different location outside their locality took place since 1996 (28 persons i.e. 26% of total) and it reached to the peak in 2005 when more than half of the total migrant members (51%) migrated from the locality. Rapid flourish of garment industries, lather processing industries, shrimp culture since 1990 to 2000 and later real estate business in Dhaka city might have contributed 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 there is no clear explanation of sharp fall in the flow of migration in year 2006 it is thought that the political unrest and frequent labour unrest in garment industries in the whole year of 2006 are the responsible factors behind such fall.
4.1.2.2 Purpose and destination of migrants
It is known that people migrate from one place to another for seeking a better livelihood or to escape from hazard, risk or such kind of threats of life. Earlier it 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 variations among the three upazilas in the case of the purposes of migration.
Table 3: Purposes of Migration from Rural to Urban Areas
Purposes Numb of migrated persons Percentage erServices 77 71 Job in workshops or factory
r
h husband
108 100
4 4 Shop keeping 2 2 Rickshaw pulle 1 1 Business 6 6 Education 16 15 Migrate wit 2 2 Total S ield survey, January, 2007
migration here people are primarily migrating for
and
ource: F
Like the normal feature ofseeking better livelihood. More than 80 per cent of the people have migrated for different economic activities. The main reason forcing people to leave their dwellings are dominated by pull factors like formal service sector employment. According to the migrants, they are mostly migrating for services. Here it should be noted that this service sector includes jobs in different garments industries, factories, real estate companies and different business offices. Majority of these migrants, have secondary and higher secondary level of education. So, they are not able to have a higher position in the service sectors. But those who have graduation level of education have migrated for government services, which are mostly teaching. In the locality seeking better or higher education is identified as another important purpose of migration (15%). Gafargao and Haluaghat which are relatively deprived in higher educational facilities accommodate more number of migrants who have migrated for education purpose. Now it is needed to know where these migrants, go to seek their livelihood and education.
Since social and economic opportunities coupled with other servicesamenities are available in the larger towns and metropolis, most of the cases destinations of the migrants are cities, especially Dhaka. The survey findings also shows that almost half of the total out‐migration (around 48 per cent of total migrant and 54 per cent of migrant within the country) from Mymensingh region has been to Dhaka, the capital of Bangladesh and the hub of major trade, commerce and industry. Moreover, it is also interesting to notice that since
50
Mymensingh Town is the nearest urban centre and offers 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 out‐migration has been within the district of Mymensingh, especially district sadar (centre). Some other internal migration is also observed though it is negligible. Table 4: Destinations of Migrants
Destination adesh)
Number of migrated people Destination Number of (Outside
(Inside BanglB angladesh)
migrated people
Dhaka 52 (54% of migration inside the country)
Dubai 4
Mymensingh 21 Saudi Arab 4 Muktagacha 3 Italy 1 Char Algi 1 F rance 1 Chittagong 4 USA 1 Valuka 2 Narayanganj 2 Ghorashal 1 Gazipur 4 Shalotia 1 Khulna 1 Sylhet 1 Kishorganj 1 Norail 1 Tangail 2 Total 97 Total 11 S survey, January, 2007
able 4 reveals that only 10 per cent of total migrant members (108) have
ource: Field
Tmigrated to outside the country and the place is dominated by Middle East. In fact, Bangladesh exports contact labour mostly to Middle East and South Eastern countries. Therefore, the people having relatively low level of education (secondary and higher secondary level) try to seek job as a labour in Dubai Saudi Arabia and such other countries. In contrast, people having a better level of
51
education and have any reference person to European countries and USA try to migrate there.
From the above sections various characteristics of migrant have been known. Now in the following section it has been tried to present a comparative picture of migrant and non‐migrant families on the basis of different socio‐economic conditions.
4.2 What are the differences between migrant and non‐migrant families?
In order to assess the differences between migrant and non‐migrant households, 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 been considered.
4.2.1 Family type and size of migrant and non‐migrant households
In this part family type and size of migrant and non‐migrant households have been compared and it has been tried to explore the differences.
Table 5: Family Type and Size of Migrant and Non‐migrant Households
Family type Non‐migrant household Migrant household No. of
household Percentage of
total No. of household Percentage of
total Single family 88 81 42 59 Joint family 21 19 29 41
Total 109 100 71 100 Non‐migrant household Migrant household
Family size No. of household
Percentage of total
No. of household
Percentage of total
2‐4 47 43 12 17 5‐9 51 48 42 59 10‐14 5 4 10 14 15‐19 3 3 6 8 20 + 3 3 1 1 Total 109 100.00 71 100.00
Source: Field survey, 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 amenities as well as economic insolvency that provoke the joint family (usually comprising greater family members) members to leave their dwellings to meet up larger family demand. It is interesting to find from the above table that family size is an influential catalyst behind migration. Probably the increasing economic demand of the large families have compelled their adult population (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 5‐14 members. In contrast, overwhelming concentration of non‐migrant households (81%) are in the single family type and their family size varies within 2‐9 members. The percentage of migrant households having more than 10 members (23%) is more than double than that of non‐migrant households (10%). In addition with the higher family demand, another responsible factor of having higher migrant members in joint‐large families is having alternative male person to take care rest of the family members in the place of origin.
4.2.2 Housing condition of migrant and non‐migrant households
Housing condition of the migrant and non‐migrant households has been compared on the basis of major structural condition of their living place. Hereunder the Table 6 presents the comparative picture.
Table 6: Housing Condition of Migrant and Non‐Migrant Households
Non‐migrant household Migrant household Housing condition
No. of household
Percentage of total
No. of household Percentage of total
Kutcha 82 75 26 36.62 Semi‐pucca 24 22 34 47.89 Pucca 3 3 11 15.49 Total 109 100.00 71 100.00 Source: Field survey, January, 2007
It is interesting to observe that more than twice of the dwellings of non‐migrant families (75.23%) are temporary in nature (kutcha structures) than that for
53
migrant families. In contrast, ownership of pucca (concrete and brick built) dwellings is three times higher among the migrant households than that of non‐migrant households. All these reflect the change in living standard and affluence of the migrant families. Here, it should be added that the income level of the migrant households is also higher than that of non‐migrant households and it is presented in the following discussion.
4.2.3 Income and occupation of migrant and non‐migrant households
Here, under this analysis the monthly income and the main occupation of both migrant and non‐migrant households have been identified and compared.
Table 7: Income and Occupation of Migrant and Non‐migrant Households
Non‐migrant household Migrant household Income (in Tk.)
No. of household Percentage of total
No. of household Percentage of total
1001‐2000 13 12 9 13 2001‐3000 30 28 11 15 3001‐5000 39 36 17 24 5001‐10000 22 20 23 32 >10000 5 4 11 15 Total 109 100.00 71 100.00
Non‐migrant household Migrant household Occupation types No. of household Percentage of
total No. of household Percentage of
total Agriculture (including fisheries and livestock)
54 50
33
46
Service 19 17 22 31 Business 16 15 8 7 Day labour 15 14 3 4 Rickshaw puller
3 3 3
4
Artisan 2 2 2 3 Total 109 100 71 100 Source: Field survey, January, 2007
The income level of the families in the study areas, are higher for the household having migrated people. As observed, the share of low‐income families (Taka 1001‐ Taka 3000 per month) is more among the non‐migrant families (40%) compared to migrated ones (28%). On the other hand, in case of a higher income
54
level (Taka 5001‐Taka 10000 per month), the migrated families (32%) clearly dominate over non‐migrated ones (20%). In addition, the families with migrated members have as much as 3 times share (15%) to non‐migrated families among the highest income families (> Taka 10000). All these demonstrate the affluence of the families with migrated members and show the financial inducement for migration.
In the case of occupation pattern of migrant and non‐migrant households, it was 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 concentration of migrant households is higher (almost double than that of non‐migrant). In the previous section it has been also mentioned that people of the area are mostly migrating for the purpose of services. Another important fact in the case of occupation patter is that the dependency on day labour is three times higher among the non‐migrant households comparing with migrant households. It denotes that migrant households are relatively less engaged in low profile economic activities and their income level also support this fact.
4.2.4 Land ownership and access to services and facilities
Half of the both migrant and non‐migrant households are dependent on agriculture and the average land ownership of the migrant and non‐migrant households is also almost same in amount, which are 2.98 and 3.15 acres respectively. It should be mentioned that both migrant and non‐migrant 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 acre per household in Gafargao, whereas 1.8 acre and 1.2 acre in Haluaghat and Muktagacha respectively.
Table 8: Average Amount of Land Ownership of Migrant and Non‐Migrant Households
Study area Assets (in acre) of migrant households
Assets (in acre) non‐migrant households
Haluaghat 1.543 2.195 Muktagacha 2.294 0.109 Char Algi 5.12 6.43 Average 2.986 3.151
Source: Field survey, January, 2007
55
4.2.5 Access to services and facilities
Here, services and facilities include only water supply and sanitation facilities, electricity connection at home and access to credit facility and institutional support (mostly training facilities and agricultural extension services).
Table 9: Access to Credit Facility and Institutional Supports
Issues Non‐migrant family (total no.109) Migrant family (total no. 71) Yes (%) No (%) Total (%) Yes (%) No (%) Total (%) Access to credit facility 56 44 100 44 56 100 Access to institutional supports
96 4 100 87 13 100
Note: Here Institutional supports means support for training and skill development, motivation, agricultural extension services, etc.
Source: Field survey, January, 2007
Before conducting field investigation it was assumed that the migrant households would have better access to various services and facilities due to their better income. But in the issue of water supply and sanitation condition and in the electricity connection no noticeable difference has been identified between migrant and non‐migrant households. It has been found that average 70 per cent 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 drinking water and hygienic toilet facility. Motivational activities and campaign programme of both the Government and Non‐Government 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 non‐migrant households, overwhelming majority have agreed that they have access to institutional supports. But the support is delimited within the agriculture extension services mostly.
56
In the above sections it has been tried to seek the answer of two questions that are who the migrants are and what are the difference between migrant and non‐migrant households in the place of origin, which is rural Bangladesh. Based on all the analysis and outcome following conclusion can be drawn.
5.0 Conclusion
By conducting field level intensive study, this research not only examines few principal characteristics of migrants but also compares migrant and non‐migrant families of three upazilas of Mymensingh District. Study reveals that since 1990s out‐migration from the region has been increased rapidly and similar with the national average 39 per cent of the families of the area have average 1.5 person migrant members. These people have migrated primarily for their livelihood and in some cases for better education. Interestingly, Dhaka is the destination of more than half of these migrants (54%). Comparative analysis between migrant and non‐migrant families shows that dependency on the tertiary occupation (services) is higher among the migrant families, comparing with non‐migrant families and in the higher income level migrant households have more concentration than non‐migrant households. No remarkable difference between migrant and non‐migrant households has been noticed in the aspects of the access to electricity, water supply and sanitation.
As one of the most important observations of this study, it is noticed that in the case of income, housing condition and land ownership migrant families are in a better position. With the use of remittance the migrant families are improving their living standard. It is said that migration is shaping Bangladeshi society and remittances have been causing silent economic revolution in Bangladesh (Siddiqui 2003 and Aziz 2001, cited in IOM 2005). Therefore migration should be 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 convert migration (both rural to urban and international) as a catalyst of national economic development.
This study does not give any attention to the problems, potentials and consequences of migrations. But it is true that migration induces huge problems 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
minimise and cope with the induced problems and on the other hand to accrue the potentials for enhancing development pace. For such kind of diagnosis this 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 adverse impacts in the economy and society.
References
Afsar, Rita. 2000. Rural‐Urban Migration in Bangladesh‐Causes, Consequences and Challenges. Dhaka: The University Press Limited.
BBS. 2003. Population Census 2001: National Report Provisional, Bangladesh Bureau of Statistics. Government of People’s Republic of Bangladesh, Dhaka.
IOM. 2005. Dynamics of Remittance Utilisation in Bangladesh. IOM Research Series No. 18. International Organisation for Migration, 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: University Press Limited.
Skeldon, Ronals. 2005. Migration and Migration Policy in Asia: A Synthesis of Selected Cases. In Tasneem Siddiqui (ed.), Migration and Development‐Pro‐Poor Policy Choices. Dhaka: University Press Limited,
Siddiqui, Tasneem. 2005. Introduction. In Tasneem Siddiqui (ed.), Migration and Development‐Pro‐Poor Policy Choices. Dhaka: University Press Limited.
Siddiqui, Tasneem. 2005. International Migration as a Livelihood Strategy of the Poor. In Tasneem Siddiqui (ed.), Migration and Development‐Pro‐Poor Policy Choices. Dhaka: University Press Limited.
58
1IsstspH
TMcahaW1a *
Asia‐Pacific Journal of Rural Development Vol. XIX, No. 1, July 2009
Accessibility of Women to Productive Resources in
Farm‐Households of Kaduna State, Nigeria S. A. Rahman* and F. 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. A multi‐stage random sampling technique was used in the selection of a sample of 240 farm households. Ordinary least squares (OLS) regression model was used in analysing the data that were collected through interview schedule. It was observed that 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. The variables included in the regression model explained 51 and 58 per cent of variation in the rate of women accessibility to farm resources in northern and southern parts of Kaduna State respectively.
.0 Introduction n Nigeria women are not homogenous because of the enormous variety of ocio‐economic and cultural situations in which they may find themselves. The pecifics of traditional gender‐based subordination would, therefore, vary across he country. In spite of the heterogeneity, however, Nigeria women’s ubordination, has typically limited their access to and control over such roductive resources as land, labour, credits and other farm inputs (Rahman and aruna 1999).
he doctrine that women are not expected to work on farms, especially in uslim communities of northern Nigeria, has met with considerable empirical hallenges. According to Saito (1992), by 1990, women in northern Nigeria ccounted for 22 per cent of the farm labour, either on their own farms or as ired labour. In the other parts of the country men make major farm decisions, lthough women decide on what crops to grow on their own fields (Okorji 1991). omen make decisions on daily management of farms and households (Saito 995). Gabriel (1991) reports a 16‐hours working day for African farming women t certain times of the year, while Whatmore (1991) pointed out that no mater the Keffi Faculty of Agriculture (Shabu‐Lafia Campus), Nasarawa State University, Nasarawa State, Nigeria. Corresponding E‐mail: [email protected]
59
extent of women’s agricultural activities, there is little variation in the extent to which domestic labour is shared by other members of the households.
The overall feature of Nigerian women’s status is essentially that of marginalisation, which is best explained within the context of farm‐household 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; Mgbada 2000; Rahman et al. 2004).
Commercialisation of land has overlooked women’s cultivation rights. According to Quisumbing (1994), there has been a great disparity between women and men in the size of landholdings, due to population pressure. There are also constraints on women labour time, as they cannot call on the labour of other household members in the way the men can (Malena 1994). Women interest and involvement in farm decision making and production have a lot of implications for their access to and control over resources (Morna 1990; Rahman and Alamu 2003).
The general aim of this paper is to examine the nature and extent of women accessibility to productive resources for farm production. Specifically, the paper:
• determines the rate of women involvement in farm decision making.
• assesses the accessibility of women to productive resources.
• identifies factors determining the accessibility of women to productive resources.
2.0 Methodology
A survey was conducted on farm households in two geographical regions of Kaduna State (northern and southern parts). The two regions were chosen to allow for comparison between women in Muslim communities (in northern part of Kaduna State) and women in Christian communities (in the southern part of the state), especially in the aspect of women’s 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.
Multi‐stage random sampling techniques were used in the selection of a sample 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 selection of sample households from the selected villages. In each of the twenty‐four (24) selected villages, ten farm households were randomly selected giving a total of 240 sampled farm‐households.
The survey data include detailed module on household size, farm size, farm inputs, farm decisions, women’s socio‐economic characteristics and their accessibility to farm resources. Analysis of the data was done using descriptive statistics such as mean and percentage; and by using multiple regression models.
The ordinary least squares (OLS) regression model was used in analysing the factors that determine the level of accessibility of women to productive resources (land, labour, fertiliser, credits and other farm inputs). This model is specified in form of double logarithmic function as follows:
Log RA=logβ0 + β1log FD +β2 log VC + β3 log CP + β4 NC + β5 log ES + β6 log AG + β7 log IL + log U
RA = Rate of accessibility of women to productive resources (%)
FD = Rate of Involvement in Farm Decisions (%)
VC = Value of Contribution to farm production by the women (N)
CP = Cooperative participation (years)
NC = Present Number of Children by the woman.
ES = Educational status (years)
AG = Age of the woman (years)
IL = Level of Income of the woman (N)
β0 = Constant term
β0 ‐ β7 = Regression
U = Error term
3.0 Results and Discussion
3.1 Women’s involvement in farm decision making
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 decision‐making power than women in northern part of Kaduna. Women in
61
northern part of Kaduna State were involved in farm decision making at the rate of 25.07 per cent while their counterpart in the southern part of the state rated 56.09 per cent (Table 1). Considering the two regions together, the rate observed was 40.58 per cent. In northern part of Kaduna State, women had below 50 per cent rate of involvement in all the identified aspects of farm decisions. In southern part, women had above 50 per cent involvement rate in decisions with 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 decisions and responsibilities taken by women are related to the farm tasks they perform.
Table 1: Women Involvement in Farm Decisions in Northern and Southern Parts of Kaduna State
Rate of involvement (%) Decisions with regards to: Northern Kaduna Southern Kaduna Average
Selection of enterprises 14.40 49.17 31.79 Size of enterprises 12.50 46.04 29.27 Inputs procurement 11.46 44.17 27.82 Inputs allocation 13.54 47.92 30.73 Harvesting of produce 20.00 50.42 35.21 Transporting of produce 18.13 52.08 35.11 Selling of produce 28.54 64.38 46.46 Consuming of produce 48.13 62.08 55.11 Processing of produce 48.75 74.17 61.46 Storing of produce 35.21 70.42 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 (very much involved, much involved, moderately involved, less involved and not involved). The levels were scored 4,3,2,1 and 0, respectively; and considered for the past five years to give a maximum score of 20.
Source: Field Survey Data, 2003
The enormous variety of socio‐economic and cultural situations in the State might have influence the rate of women’s involvement in the farm decision making. The northern part of the State is a Hausa‐Muslim‐dominated zone while the southern part is non‐Hausa Christian‐dominated. The socio‐economic life styles of women in the Hausa‐Muslim communities are strictly guided by Islamic principles which favour less works for women on farms. This is not the
62
case with other non‐Hausa Muslim women who are freely engaged in both agricultural and non‐agricultural activities. In the non‐Hausa communities women own farms and play a more active role in farm activities.
3.2 Women accessibility to productive resources
The fact that women generally do not control a lot of money in the households, thus, their access to productive resources is usually limited. Given the superior 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 southern part it was 19.73 per cent; while the overall was 11.58 per cent as shown in Table 2.
Table 2: Women Accessibility to Farm Productive Resources in Northern and Southern Parts of Kaduna State
Accessibility rate (%) Productive resources
Northern Kaduna Southern Kaduna Average Farm land 3.17 32.31 17.74 Labour 5.56 21.15 13.36 Fertiliser 2.74 15.38 9.06 Credit 2.38 10.58 6.48 Other farm inputs 3.25 19.23 11.24 Average 3.42 19.73 11.58
Note: Other farm inputs include seeds (local or improved) and chemicals (herbicides or/and insecticides)
Source: Field survey data, 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 women’s poor financial status and poor access to credit facilities do not allow them to acquire farm inputs. This limits the scale of their farm production. Women may be willing to adopt new farming practices or buy equipment and other agricultural inputs, but they have no power to decide without the husband’s approval. Moreover the husbands control the family finances in the majority of cases, which make situation even more difficult for the women.
3.3 Determinants of women’s accessibility to farm resources
Improvements in the rate of women accessibility to productive resources require examining the socio‐economic constraints that hinder the accessibility to the
63
resources. The variables included in the regression model explained 51, 58 and 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 gender‐based subordination which varies across the state.
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 resources in both the northern and southern part of Kaduna State.
Table 3: Factors Determining Women Accessibility to Farm Resources in Northern and Southern Kaduna State
Estimated coefficient Factors Northern Kaduna Southern Kaduna Both regions
0.512 0.693 0.417 Constant (1.053) (1.241) (1.382) 0.286 0.432* 0.379 FD (1.008) (4.126) (1.121) 0.116* 0.134* 0.234* VC (2.983) (3.817) (3.004) 0.098 0.141 0.105* CP (1.122) (1.003) (2.576) 0.064* 0.043 0.024 NC (2.818) (1.131) (1.500) 0.080 0.111 0.098 ES (1.103) (1.254) (1.052) 0.034 0.017 0.014 AG (1.313) (1.248) (1.066) 0.161 0.432* 0.274* IL (1.401) (4.243) (2.998)
R2 0.51 0.58 0.63.
Note: *Significant at 5 per cent level Figures in parentheses are t‐values
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 the State together, it was observed from the regression that contribution to farm production, co‐operative participation and income were identified to be significantly related to
64
the rate of women accessibility to productive resources in Kaduna State (Table 3). Participation of the women in co‐operatives could be a source of improvement in their socio‐economic status.
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 their labour productivity. This is as a result of their limited access to and control over such productive resources as land, labour, credits and fertilisers. This study has revealed that in Kaduna State, women were less involved in farm decision‐making, and they had low accessibility to farm resources. Income and co‐operative participation of women were identified as the significant factors that positively influenced accessibility of women to farm resources.
This study, therefore, recommends the followings:
• Women oriented policies and programmes should be introduced to improve women’s 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 formation of co‐operatives.
• Developing and reinforcing research and information gathering exercises on village community‐level socio‐cultural financial conditions that affect the rural women. This will enable improvement in the traditional gender‐based subordination in various communities for good welfare of women.
65
References
Gabriel, T. 1991. The Human Factor in Rural Development. Belhaven Press: London. Ingawa, S. A. 1991. Welcome Address at the National Workshop for Women in
Agriculture, held in FACU Headquarters, Sheda, Abuja, Nigeria. 31st August‐2nd September.
Malena, C. 1994. Gender issues in integrated pest management in African agriculture. NRI Socio‐Economic Series 5. Natural Resources Institute. Chatham, U.K:
Mgbada, J. U. 2000. Production of staple crops by rural women in Enugu and Ebonyi States: lessons for enhancing poverty alleviation programmes. In Olowu, T.A. (ed.). Agricultural extension and poverty alleviation in Nigeria Proceeding of the 6th Annual National Extension Society of Nigeria. Pp. 10‐12.
Morna, C. I. 1990. Women farmers emerge from the shadows. African farmers, No. 3 April. Okorji, E. C. 1991. A comparative study of the role of women in traditional and modern
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, World Bank, Washington, D.C., U.S.A.
Rahman, S. A. and I. M. Haruna. 1999. Determinants of women’s economic contribution 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, Ahmadu Bello University, Zaria, Nigeria. 3rd to 4th November.
Rahman, S. A. and J. F. Alamu. 2003. Estimating the level of women interest in agriculture: The application of Logit Regression model. Nigerian Journal of Scientific research, 4(1): 45‐49. Vol. 4, No. 1, Pp. 45‐49.
Rahman, S. A., J. Gabriel, and N. D. Marcus. 2004. Gender differentials in labour contribution and productivity in farm production: Empirical evidence from Kaduna state of Nigeria. Paper presented at the National Conference on The Family, held at New Theatre complex, Benue State University, Makurdi, Nigeria. 1st ‐5th March.
Saito, K. A. 1992. Raising the productivity of women farmers in sub‐saharan Africa: Overview Report vol. 1. Population and Human Resources Development Department, World Bank, Washington DC., USA.
Saito, K. A. 1995. Raising the productivity of women farmers in sub‐saharan Africa; Agricultural and Environmental challenges. In Srivastava, J.P. and Alderman, H. (Eds.) Proceedings of the 13th Agricultural sector symposium. The World Bank. Pp. 147 – 152.
Whatmore, S. 1991. Women in agriculture. Journal of Rural Studies, Vol. 7, No. 1 & 2 (special issue).
66
A V
1
Alatamwi
Tbdvb
*
1
sia‐Pacific Journal of Rural Development ol. XIX, No. 1, July 2009
Empirical Analysis on Rural Households’ Borrowing Behaviour: The Case of Central Java
Takashino Nina*
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) villagers in the research area have access to the formal financial institutions; (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 self‐help saving group (arisan in Javanese) more frequently than from formal institutions or friends and relatives.
.0 Introduction
ccess to credit is essential to improve the living standard of rural villagers in ow‐income countri.1 However, formal financial institutions (banks, co‐operatives nd so on) cannot always achieve high repayment rate successfully, and hence, he policymakers need to explore a better design for such institutions (Zeller nd Meyer 2002). Knowledge about rural households’ borrowing, such as how uch financial needs they face and from what sources they borrow money, ould provide much helpful insight to designing an effective rural financial
nstitution.
he aim of this paper is to empirically examine the determinants of households’ orrowing demand and how borrowers choose the lending source by using the ata set on rural villagers’ financial transactions in Central Java. Same as the illagers in other low‐income countries, rural households in Indonesia can orrow money from two kinds of lender, formal and informal. Financial
Center for Experimental Research in Social Sciences, Hokkaido University, Japan. E-mail:
[email protected] 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 (Petrick 2005).
institutions under government’s control, such as banks and co‐operatives, are called formal lenders. Their transactions involve a written contract, filling documents and proposal of collateral. On the other hand, traditional self‐help saving group (arisan in Javanese), moneylenders, and villagers who provide loan to friends and relatives are called informal lenders. Their transactions are out of the governmental regulation and usually based on an oral contract without taking collateral.
Existing literatures propose two different views on the role of each sector in low‐income countries. While some studies treat informal sectors as secondary sources 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 cheaper than formal (Kochar 1997). This study will show that Javanese villagers chose a lender depending on the situation, and hence, the latter view is applicable for the case of the research site.
Some studies describe borrowing behaviour observed in rural areas of Indonesia (Hamada 2006; Mizuno 1999: 199‐226)2, but empirical data on borrowing demand and sector choice of rural household is limited. Another important contribution of this paper is to fill this gap and to show under what conditions villagers would choose borrowing from informal sectors.
The main findings are summarised as follows: (1) villagers in the research area have access to the formal financial institutions; (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 relationship of trust; (4) the poor tends to borrow money from arisan more frequently than from formal institutions or friends and relatives.
The organisation of this paper is as follows: After describing an overview of sample hamlets and socio‐economic conditions of sample households in the second section, the data on borrowing behaviour and asset management are summarised in the third section. Analytical frameworks on the empirical study
2 Hamada (2006) points out rural villagers are familiar with informal lenders and more closely
related with them than with formal institutions. Mizuno (1999: pp.199‐226) compares the role of formal and informal finance on the providing loans to small textile enterprises in rural West Java. Mizuno found that the transaction cost of formal finance is high because it requires borrowers to submit collateral and hence informal finance absorbs borrowing demands.
68
and estimation results are discussed in the fourth section. Finally, summary and conclusions are given in the fifth section.
2.0 Overview of Sample Hamlets
2.1 Geographical conditions
To investigate households’ borrowing behaviour, this paper employs a dataset compiled from a sample household survey conducted from 2001 to 2005 by the researchers of JSPS‐DGHE Core University Programme. About 120 households in four hamlets in the rural area of Yogyakarta province were surveyed as samples3 (Table 1). Hamlet A and B are located on a hillside away from Yogjakarta city. On the other hand, hamlet C and D are close to Yogyakarta city. (Distance to the city is 40km, 32,4km from hamlet A, B and 10km from both C and D.) Table 1: Basic Data of Sample Hamlets
Hamlets A B C D Area condition hilly hilly lowland lowland Distance from city (km) 40 32.4 10 10
Irrigation system rain‐fed traditional /rain‐fed
traditional technical
Number of population 622 641 604 430 Number of household 115 130 163 97 Average number of household member 5.41 4.93 3.71 4.43 Total area of land (ha) 129.5 191.5 44.5 41.3 Paddy field ‐ 22.1 23.0 31.0 Dry field 127.0 144.2 ‐ ‐ Compound 2.5 25.2 17.1 6.8 Others ‐ ‐ 4.4 3.5 Average size of farming land (ha) 1.10 1.28 0.14 0.32
Source : Subejo and Iwamoto, Noriaki, 2003
Geographical differences between hilly area and low land are affected household’s economy mainly in two aspects. One is water supply and another is opportunities to get off‐farm job. During the dry season people in hamlet A and B suffer from a water shortage and a decrease in income because almost all of them are farmers. Water condition of hamlet B is slightly better than A. Hamlet C
3 Details of the survey are given in Iwamoto, Hartono and Fukui. (2003) and Subejo and
Iwamoto (2003).
69
is semi‐irrigated while D is fully irrigated. Some farmers in Hamlet D raise catfish and it yields high profit with high risk. Since hamlet C and D are located near to the city, villagers have more opportunities to be engaged in off‐farm job.
2.2 Households’ risk and investment needs
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. Seeing from higher level of schooling in hamlet C and D than those in A and B, financial needs for investment on education might be more in the hamlets in the lowland area (C and D). Table 2: Characteristics of Sample Households A B C D Household characteristics Age of household head 48.2 49.9 59.6 54.6 Family member 4.6 4.3 4.6 4.0 Number of labourers (man/hh) 3.1 3.0 3.2 2.6 Ratio of graduates1) Elementary school (%) 64.2 58.3 52.9 46.6 Junior high school (%) 32.5 34.3 19.6 11.8 High school (%) 3.4 5.0 22.5 32.4 University (%) 0.0 2.5 5.0 9.3
Household income (Rp/hh) 5,447,712 4,873,281 8,847,360 15,247,219 Agriculture (Rp/hh) 3,400,448 3,123,402 1,897,347 7,828,038 Off‐farm job (Rp/hh) 1,513,794 1,347,316 6,394,178 6,351,973 Land rent (Rp/hh) 9,216 500 115,594 136,649 Remittance (Rp/hh) 524,255 402,063 440,241 930,558 Ratio of Agric. income (%) 62.4 64.1 21.4 51.3
Number of sample households (households/ 3 years) 85 80 87 77
Note: 1) Graduate ratio per family labourers 2) Exchange rate at Aug. 2003 was 1 US‐dollar = 8, 500 rupiahs Source: JSPS‐DGHE Core University Programme 2001‐2003 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, financial needs for consumption smoothing or investment are also different depending on each household’s characteristics. Villagers earn their income from agriculture, off‐farm job, land rent, and remittance.
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 rain‐fed 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, farmers in the area face much income fluctuation and risk. On the other hand, farmers in lowland area can well utilise irrigate water and achieve comparatively high yields of paddy. Even though they can mitigate risk by using irrigation, many share‐cropping contracts are still observed there. Reason might be the farmers want to share risks caused by crop diseases and rats attack (the damage is specifically observed in lowland area). Another possible reason is that most of the sharecropping contracts are based on personal relationship among friends and relatives (Fukui, Hartono, and Iwamoto 2002).
Many households raise livestock such as cows, goats, and chickens in all sample hamlets. Particularly in hamlet D, some households raise catfish or other fishes 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 favourite food of rats, crop damage caused by rats is especially severe in hamlet D (Iwamoto, Hartono and Fukui 2003).
Earnings from off‐farm employment provide villagers stable income and mitigate income fluctuation. Some villagers are employed in government sector, 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 villagers can commute to the centre of the city. This circumstance may have an impact on demand for investment on schooling because high school enrolment is 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 part‐time workers, they can deal with any income drop by earning income ex‐post.
Another comparatively stable income source is land rent. But only small numbers of landowners are included in sample households. While almost all the farmers in hilly area cultivate their own land, a typical farmer in lowland area (hamlet C and D) employs tenant contract. Therefore, some villagers in hamlet C and D earn income from land rent (Table 2 and 3).
71
Remittance takes two forms: One is permanent annual (or monthly) transfer at constant amount and another is casually sending to help sharp drop in income or unexpected consumption needs. Table 2 shows that some households would get remittance and it might have decreased borrowing demand. Table 3: Land Ownership and Tenancy A B C D
Paddy
Owned (square meter) 5,056 2,539 410 1,433
Leased in (square meter) 0 0 1,945 1,423
Total (square meter) 5,056 2,539 2,355 2,857
Other crops
Owned (square meter) 16,696 3,162 174 268
Leased in (square meter) 1,023 375 0 0
Total (square meter) 17,719 3,310 174 268
Home garden (square meter) 2,590 1,387 523 562
Sample (household) 31 30 30 28
Source: JSPS‐DGHE Core University Programme 2003
3.0 Households’ Borrowing Behaviour
3.1 Overview on credit market
Table 4 describes the history of sample households’ borrowing for 3 years from September 2000 to August 2003. Households can borrow money from two kinds of lenders, namely formal and informal ones. Financial institutions under government’s control, such as banks and cooperatives, are called formal lenders. Their transactions involve written contract, filling documents and requiring proposal of collateral. On the other hand, traditional self‐help saving group (or arisan in Java), moneylenders, and villagers who provide loan to friends and relatives are called informal lenders. Their transactions are out of governmental regulations and usually based on oral contract without taking collateral.
Sorting all borrowings by the lender’s type, Table 5 clarifies different characteristics in their contracts. Villagers borrowed money from various formal lenders. Accessible banks are BRI‐UD (Bank Rakyat Indonesia Unit Desa), Bank Kredit Desa (BKD), BUKOPIN, BMT, and other BPRs (Bank Perkreditan Rakyat).
72
Some villagers may also have access to farmers’ co‐operatives or KUD (Koperasi Unit Desa) or employees’ co‐operatives. 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.
Table 4: Number of Credit Transactions (Unit: No. of households) A B C D Total 2001 a. Formal 5 1 2 8 16 b. Informal 2 9 8 6 25 c. Both 0 0 1 3 4 d. No 26 20 24 22 92 Total (a+b‐c+d) 33 30 35 34 137
2002 a. Formal 5 0 3 8 16 b. Informal 3 6 7 11 27 c. Both 0 0 1 2 3 d. No 25 24 24 16 89 Total (a+b‐c+d) 33 30 35 37 135
2003 a. Formal 5 1 4 8 18 b. Informal 5 16 18 8 47 c. Both 1 1 2 2 6 d. No 24 14 13 19 70 Total (a+b‐c+d) 35 32 37 37 141
Source: JSPS‐DGHE Core University Programme 2001‐2003 Assets used as collateral include certificate of land, motorbike, bicycle, sewing machine, certificate of income, personal recommendation, car, TV, and so on. Adding to these traditional types of loans, many formal institutions provide loan programme for the poor and small enterprises. Takesra Kukesra, Tabungan Keluarga Sejahtera (Takesra) dan Kredit Usaha Keluarga Sejahtera (Kukesra) are some of the programmes nationally conducted for the poverty alleviation between 1994 and 2001.
When villagers borrow money from friends and relatives, they do not need to prepare any documents or collateral because their personal trusts enable them to have a contract orally. Besides, lenders usually allow the borrowers to
73
reschedule their loans when they cannot pay. Because the amount of loan cannot get over the lenders personal asset holdings, it is typically smaller than banks. Comparing loans from friends and those from relatives, villagers tend to borrow more money from relatives than friends based on stronger social tie or trust. 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, some of lenders live outside the hamlets.
Table 5: Credit Transactions Sorted by Lender Type
No. Average amount Interest rate Term Collateral
(Rp.) (%/month) (month) (%)
Formal
Bank 48 3,000,000 2.33 14.2 90 BRI 23 3,526,087 2.07 18.0 91.3 BPD 1 6,000,000 2.13 36.0 100.0 BUKP 4 1,600,000 1.89 14.5 50.0 BMT 7 4,171,429 2.95 10.9 85.7 BUKOPIN 8 475,000 2.79 5.3 100.0 BPR 2 2,250,000 2.25 5.0 100.0 Other 3 4,333,333 2.33 16.0 100.0
Co‐operative 13 1,319,231 1.85 12.5 23.1 KUD 3 633,333 2.50 10.0 33.3 Other 10 1,525,000 1.65 13.2 20.0
Programme 7 121,429 2.17 12.3 0 Takesra 4 112,500 2.30 9.5 0.0 Other 3 133,333 1.99 16.0 0.0
Informal Relatives 22 1,730,000 0.00 . * 4.5 Friends 33 569,697 0.50 . * 6.1 Arisan 99 109,414 3.92 . * 3.0
Total 222 1,034,649 2.50 23.4 Note: * Informal loans are often rescheduled Source: JSPS‐DGHE Core University Programme 2001‐2003 Self‐help 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 lottery with some fees. At the same time, they collect savings from the members
74
and pool them in order to provide loan to any member facing financial needs. In the same manner, with the loans between friends and relatives, the transactions 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 types mentioned above because the amount of collected saving is very small.
Although some of the villagers have a little information regarding informal money lenders around sample hamlets, no sample household has ever borrowed money from them. Only 31 households out of 118 have any information on the 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 tended to lend money as the second job.
3.2 Motivations of borrowing
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 ex‐post dealing with drop in income or unexpected expenditure. Sample households tend to borrow money for the ex‐post consumption smoothing, such as daily consumption, medical expenditures and so on rather than for the ex‐ante borrowing for buying durable goods.
When villagers borrow money for consumption, they seem to choose borrowings 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 amount is big, loans from friends and relatives are actively chosen. If loan is used for unprofitable purposes and the need is small, they tend to borrow from arisan. When loan purpose has characteristics both as investment and consumption, like loan for education and housing, formal and informal loans are used. Table 6 shows that some villagers borrow money from informal source even though they use it for investment and it suggest two possible situations, one is lower effective cost of informal loans and the second one is constrained access to cheap formal loans.
75
Table 6: Credit Transactions Sorted by Purpose
No. of transactions Average
Informal Formal amount of loan
Relative Friend Arisan Coop. Bank Programme Total
(Rp.)
Consumption Daily 2 8 29 1 4 2 46 154,935 Medical 5 3 2 1 1 0 12 2,470,000 Social 1 1 8 0 0 0 10 59,500 Durable 1 4 2 1 1 0 9 1,208,889 Debt 0 0 0 0 2 0 2 2,950,000 Help 0 0 1 0 0 0 1 95,000
Investment Education 7 5 19 4 6 1 42 568,571 Housing 2 2 3 2 6 0 15 2,385,000 Agriculture 1 7 17 0 3 2 30 236,333 Fish farm 1 0 1 2 6 0 10 1,780,000 Trading 0 2 4 2 7 1 16 945,313 Other 2 1 11 1 8 0 23 2,186,739
Other 0 0 2 0 3 1 6 4,248,333 Total 22 33 99 14 47 7 222 1,034,649
Source: JSPS‐DGHE Core University Programme 2001‐2003 3.3 Saving and asset
Withdrawing asset holdings is alternative device to deal with financial needs. Households having less own asset rely on borrowing facing a financial needs. The villagers can accumulate their wealth in various forms, such as land, durable goods like TV, motorbikes or sewing machines, financial asset (gold or deposit), livestock, trees, etc. There is a slight difference between the two study areas (table 7). Livestock and trees play important role as asset in hilly area, whereas, many villagers own motorbikes and often use them as collateral for bank loan.
Some villagers may prefer borrowing to withdrawing, while others may not. The amount of loan as such as purpose of needs might determine the choice. Figure 1 illustrates the households’ answers to the question ‘how do you deal with various amounts of financial needs’. Villagers choose the most favourable way to deal with each range of the financial needs from the alternatives including
76
borrowing from relatives, non‐relative friends, arisan, formal lender, and using saving or asset4.
Table 7: Asset Holdings (Unit: Rupiah) A B C D Agricultural tools 267,815 334,377 134,100 205,393 Livestock Cow 4,142,458 4,984,434 1,605,000 96,429 Goat 600,597 380,645 176,667 189,286 Chicken 87,024 90,485 137,167 127,429 Other 0 2,708 20,733 21,196
Trees 3,620,000 3,460,000 674,167 836,897 Durable goods TV 403,241 553,333 428,167 528,213 Bicycle 3,190 12,667 270,370 285,594 Motorbike 655,172 1,914,167 4,769,167 3,941,829 Car 0 0 0 1,551,724
Financial asset Gold 376,690 344,367 654,583 286,552 Deposit 476,207 425,633 317,600 127,241
Other 17,241 80,000 50,000 523,703 Total 10,649,636 12,582,816 9,237,720 8,721,486 Sample size 31 30 30 29 Source: JSPS‐DGHE Core University Programme 2003
The figure shows when facing small amount of needs (less than 0.5 million rupiah), many villagers want to and are able to satisfy the needs using their own 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 from formal loan because of its expensive transaction cost.
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 relatives or formal lender and also selling fixed assets as goat, trees, motorbike, and TV.
4 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, large livestock like cow.
77
Figure 1: Sector Choice Strategies
0
10
20
30
40
50
60
100,000 1,000,000 10,000,000 100,000,000
Am ount of Needs
No. of
Ans
wers
Fam ily Friend Arisan Form al Saving Asset
If the villagers need to prepare money over 1 million rupiahs, formal loan and selling assets play important role to meet up the needs. Furthermore, the figure suggests choosing more asset selling than formal loan because of higher transaction cost (including risks to loose collateral if default). Villagers tend to avoid selling land or house, and only two households choose it. The facts imply their desire to retain their land and that they do use land as collateral only if they are sure to repay the loan.
78
Table 8: Coping Strategies for Big Financial Needs
Financial needs Coping strategy (%) Amount No. Asset Borrowing Wedding 6,525,000 8 100 25 Disease 6,338,846 13 85 62 Funeral 1,670,000 5 100 100 Education 3,133,889 18 89 28 Housing 16,555,556 9 89 33
Note: Total percentages of Asset and Borrowing are over 100 because some households use both of them
Source: JSPS‐DGHE Core University Programme 2001‐2003
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, and investment on schooling or housing. They can raise fund through deposit, livestock sale for a foreseeable event like wedding or investment, and it covers most part of the expenditure at low rate of borrowing in such occasions. However, because many of the villagers cannot cover unexpected event like diseases or funerals by their own assets, they need to borrow money.
4.0 Empirical Analysis on Choice of Borrowing
4.1 Analytical framework
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 loan demand as D*, and the amount of loan the household actually borrowed as D, Tobit estimation method is applied to estimate loan demand as follows:
(1) D* =β1’X1 + u1
D = D* if u1 > ‐β1’X1
= 0 otherwise
X1 is a vector of explanatory variables. It includes variables related to households’ financial needs, such as age and sex of the household head, amount of asset holding, event dummy, and location dummy (Table 9 summarises the
79
definitions of various explanatory variables). β1 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 because the number of positive amount of borrowing is too small to make panel estimation.
Table 9: Independent Variables for Loan Demand Estimation
Definition Ave. S.D. Min. Max. Household head Age Age of the household head 53.62 12.83 19.0 80.0
Sex Sex of the household head: Male=0, Female =1
0.06 0.25 0.0 1.0
Land Owned land area (1,000 m2) 23,621.49 117,930.37 0.0 1,577,562.0 Asset Asset holdings (million rupias) 9.70 8.42 0.1 42.4 Labour and education Labour No. of labour (15‐70 years old) 3.05 1.27 0.0 7.0 High school Ratio of high school graduated 0.10 0.18 0.0 0.8 University Ratio of univ. graduated 0.03 0.10 0.0 0.6 House = 1 if constructed house 0.03 0.16 0.0 1.0 Illness = 1 if suffered from illness 0.07 0.30 0.0 3.0 Student No. of students 0.48 0.73 0.0 4.0
Distrust Self report evaluation on ‘neighbours take advantage’
2.31 1.49 1.0 5.0
Location Hamlet A = 1 for hamlet A 0.26 0.44 0.0 1.0 Hamlet B = 1 for hamlet B 0.25 0.43 0.0 1.0 Hamlet C = 1 for hamlet C 0.27 0.44 0.0 1.0 Year 2001 = 1 for the year of 2001 0.28 0.45 0.0 1.0 2002 = 1 for year 2002 0.37 0.48 0.0 1.0
Source: JSPS‐DGHE Core University Programme 2001‐2003
Explanatory variables are selected on the basis of discussion in previous sections. 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 borrowing demand to invest, and at the same time, enable villagers to deal with 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, coefficient on the dummy for hamlet A, B and C (with lower productivity than hamlet D) are expected to be negative.
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 that observed borrowing reflects the equilibrium of loan demand and supply in the study locations.
Table 10: Possibility of Borrowing Money from Bank
Reason why ‘No’
Yes No Neither No collateral Low profitability
Been rejected loan offer 0 115 0 0 0 Would be rejected if offered
4 110 1 2 2
Source: JSPS‐DGHE Core University Programme 2004
Using the results of the first estimation on borrowing demand, it is as examined how the borrowers choose the lending source. Suppose the borrowers face unordered three choices of borrowing: (1) borrowing from friends and relatives, (2) arisan, and (3) formal lender. And the probabilities of borrowing from each of the above lenders are denoted as P1, P2 and P3 respectively. Let X2 be the vector of the characteristics variables which determines the borrower’s choice. Then multinomial logit estimation method is employed to estimate Pj (j = 1, 2, 3) as follows (Maddala 1983: 35):
5 For the case of India, it is assumed that not all borrowers can borrow as much money as they want. Therefore, existing study estimates both demand and supply functions at the same time (Bell, Srinivasan, and Udry 1997, and Kochar 1997). 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 households in Vietnam.
81
(2) P1 = exp(β21’ X2) / [1+exp(β21’X2) + exp(β22’X2)] P2 = exp(β22’ X2) / [1+exp(β21’X2) + exp(β22’X2)]
P3 = 1 / [1+exp(β21’X2) + exp(β22’X2)]
Households without any borrowing experience were discarded from the sample. Details of the explanatory variables are shown in Table 9. Estimated value of D* or borrowing demand of the household is also used as an explanatory variable. Because the coefficients represent the favourability of borrowing from each category compared to borrowing from formal loan, it would change depending on the cost of borrowing.
Table 11: Degree of Distrust to Neighbours
Self‐report agreement to ‘Neighbours take advantage of you’.
(5) Strongly agree (4�Agree (3)Neither (2)Disagree (1) Strongly
disagree Total
No. of answers 30 99 3 54 171 357
Share (%) 8.4 27.7 0.8 15.1 47.9 100.0
Source: JSPS‐DGHE Core University Programme 2001‐2003 The cost of borrowing from friends and relatives would involve not only material interest payment but also psychological factor. As shown in the previous section, villagers usually do not require interest payment for the loans to friends and relatives in the research fields. If the choice 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 alternative determinant of borrowing from friends and relative is a psychological factor, such as trust relationship between them. To get the index of the degree of distrust, villagers rated their agreeableness on a five‐point scale (1 = strongly disagree to, 5 = Strongly agree) to the statement ‘Neighbours take advantage of me’ (Table 11).
Expected sign of the coefficient for this variable on the probability of borrowing from friends and relatives would be negative because those who distrust neighbours would not prefer borrowing from them. Adding to this effect, social norm which encourages villagers to help widows would be other determinants (expected to have positive impact for the household with female household head).
82
For the case of arisan loan, the household without much asset might borrow money easier than the rich villagers because this loan is originally set up to help the poor. From the interview, villagers told that when there are many applicants for loan, they would give loan to a member who has urgent needs. This implies a negative impact of the asset holdings on the possibility of borrowing from arisan.
Loan size would be another determinant. When villagers borrow money from the formal lending source such as bank, they have to pay the initial cost. It means that the average cost of borrowing becomes higher if they borrow small amount of loan. Therefore, given the average cost of another choice to be constant, borrowing from formal lender would be more favourable when the borrowing amount is increased.
4.2 Estimation on borrowing amount
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 female head and others on borrowing demand.
Another significant factor shown in the result is the level of asset holdings. From the view point of the demand, the positive coefficient suggests that the more assets the villagers have, the more their chances to invest. From the point of loan supply, it might reflect more availability of getting loans. As an alternative factor affecting investment opportunities, the number of student is positively associated with the amount of borrowing, reflecting the villagers’ high expectation on the return to schooling. On the other hand, land holding has no 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 loan is not the sole choice for the villages: Villagers can utilise their own asset to deal with this kind of big consumption needs as shown in previous section.
The impact of the location dummies is estimated based on that of the hamlet D. The coefficients of the dummies for hamlet A, B, and C are negative showing low demand of borrowing in theses hamlets compared with hamlet D where the agricultural profitability is high with the well organised modern irrigation.
83
Table 12: Borrowing Function
Coefficient p‐value Characteristics of household head Age ‐487.4 0.226 Age squared 4.4 *** 0.006 Female 1283.0 0.290
Land ‐27.7 0.228 Asset 169.2 * 0.094 Asset squared ‐3.5 0.253 Human capital Labour 143.2 0.541 High school 1105.1 0.516 University 6110.8 ** 0.024
Student 806.1 ** 0.047 House 863.3 0.589 Illness 1263.8 0.153 Location dummy Hamlet A ‐2304.7 ** 0.064 Hamlet B ‐2845.6 *** 0.002 Hamlet C ‐2230.7 *** 0.007
Year dummy 2001 ‐1344.3 0.627 2002 ‐2055.5 *** 0.002
Constant 11870.7 *** 0.006 Σ 4172.5 *** 0.000 Sample size 325 No. of positive observations 135 Likelihood ‐1403.4 Note: *, **, *** denotes significance of 10%, 5%, 1% respectively
Source: Author’s estimation
4.3 Choice of lending source
The result of estimation for equation 2 shows the determinants of borrowers’ choice on lending source (Table 13). In the table, first row shows the impacts of 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. The turning point (or maximum point of the impact) is around 55 years 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 effect is also shown for the case of arisan borrowing with weak significance (in second column).
Table 13: Choice Behaviour of Lending Source
Friends and relatives Arisan Coefficient P‐value Coefficient P‐value Household head Age 0.331 * 0.079 0.097 0.604
Age squared ‐0.003 * 0.064 ‐0.001 0.722 Female 2.636 * 0.094 1.823 0.211 Distrust ‐0.632 *** 0.004 ‐0.200 0.315 Borrowing (estimates)
‐0.378 0.130 ‐0.524 ** 0.050
Land ‐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 Location dummy Hamlet A 0.896 0.441 ‐0.927 * 0.551 Hamlet B 2.445 ** 0.033 3.170 ** 0.011 Hamlet C ‐0.336 0.710 0.967 0.209
Year dummy 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 Sample size 130 No. of positive obs. 40 74 LR statistics 86.6 *** Pseudo R‐square 0.55 Likelihood ‐98.5 Note: *, **, *** denotes significance of 10%, 5%, 1% respectively
The coefficients are estimated based on the borrowing from formal lender Source: Author’s estimation
85
The results show that the probability of borrowing from friends and relatives increased when the borrowers trust neighbours more (the coefficient on distrust was positive). In other words, those who distrust neighbours prefer the loan from formal financial institutions. Similar tendency is shown for the case of borrowing from arisan in the second column. These results implied interest rate is not the only factor that affects villagers’ choice but social relationship also has an impact on their behaviour.
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. As mentioned above, because they have to pay the initial cost for the formal loan application, the cost of borrowing from formal institution will become less for the bigger amount of loan.
For the impact of landholding, a villager with larger land tends to choose formal lender more than any other sources. Negative coefficient of asset holdings on arisan borrowing means poor households more likely to choose borrowing from arisan rather than from bank. Positive association between formal borrowing and employee status might happen because employees with stable salary can apply loan with easier process of evaluation (but the significance of the impact is weak).
5.0 Summary and Conclusions
The major findings are summarised as follows: 1) villagers in the study area have 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) the poor tend to borrow money from arisan more frequently than from formal institutions or friends and relatives.
Empirical analysis shows that the villagers can borrow money from any source if they want. However, the villagers do not actively borrow for the purpose of productive investment. Possible explanation is that most of the villagers do not have investment opportunities from which they can expect enough returns. Particularly, it is revealed that such tendency is observed for the household with fewer asset holdings. Therefore, training to manage small business or introduction of commercial crops would raise their productivity and activate the villagers’ borrowing.
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: 557‐585.
Duong, B. P. and Y. Izumida. 2002. Rural Development Finance in Vietnam: A Microeconometric Analysis of Household Surveys. World Development, 30: 319‐335.
Fukui, S., I. S. Hartono, and N. Iwamoto. 2002. Risk and Rice Farming Intensification in Rural Java. The Japanese Journal of Rural Economics, 4: 32‐43.
Hoff, K. and J. E. Stiglitz. 1990. Imperfect Information and Rural Credit Markets‐‐Puzzles and Policy Perspectives, World Bank Economic Review, 4: 235‐50.
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 Agricultural and Resource Economics. Working Paper Series No. 03‐H‐002
Jaffee, D., J. Stiglitz. 1990. Credit rationing. In Friedman, B. M., Hahn, F. H. (eds.), Handbook of Monetary Economics 2, Elsevier, Amsterdam: 838‐888.
Kochar, A. 1997. An empirical investigation of rationing constraints in rural credit markets in India. Journal of Development Economics, 57: 339‐371.
J. Leith, C. Porter, SMERU Institute, P. Warr. 2003. Indonesia Rice Tariff, Poverty and Social Impact Analysis. March, http://www.odi.org.uk/prspsynthesis/Indonesia_Final_PSIA.doc
Maddala. 1983. Limited Dependent and Qualitative Variables in Econometrics. Cambridge: Cambridge University Press.
Petrick, M. 2005. Empirical measurement of credit rationing in agriculture: a methodological survey. Agricultural Economics, 33: 191‐203.
Zeller, M., R. L. Meyer (eds.). 2002. The Triangle of Microfinance. Financial Sustainability, Outreach, and Impact. Baltimore: Johns Hopkins University Press.
Subejo and Noriaki Iwamoto. 2003. Labor Institutions in Rural Java: A Case Study in Yogyakarta Province. Working Paper Series No. 03‐H‐01. Department of Agriculture and Resource Economics, the University of Tokyo.
87
Asia‐Pacific Journal of Rural Development Vol. XIX, No. 1, July 2009
Value Chains and Standards in Shrimp Export from Bangladesh and Thailand to Japan: A Comparative
Study on Safety Compliances Mohammad Taj Uddin*
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 with different stakeholders in Bangladesh, Thailand and Japan were analysed and used as the basis of the paper. The results reveal that the competent authority monitors the hygiene and sanitation condition of the buyer‐driven value chain activities, whereas processors‐cum‐exporters implement HACCP procedures about 85 per cent to 90 per cent 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 used as traceability tools. Both the countries have already started working for certification, eco‐labelling and sustainable seafood trade. However, it is recommended to ensure chain traceability together with improved management to increase products’ competitiveness.
1.0 Introduction
Shrimp as a fisheries commodity represents a great importance in the international trade (Xinhua 2008). It’s 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 South East Asian countries, act as the main shrimp suppliers of the world.
Shrimp is considered significant to both Bangladesh and Thai economy. In Bangladesh, the production of shrimp by aquaculture method is almost an export‐oriented activity producing an average output of 53,000 tons and earning US$530 million, which is some 8.0 per cent of the country’s total exports in the 2007‐08 fiscal year (July 2007 to June 2008). Shrimp is the second largest foreign exchange earner after garments. The sector employs nearly 1.3 million rural
* Associate Professor, Department of Agricultural Economics, Bangladesh Agricultural University,
Mymensingh, Bangladesh. E‐mail: [email protected]
89
people in forward and backward linkage activities like hatching, nursing, producing, harvesting, processing, exporting and other logistic support services. It has the highest potential for further development, enough to culture shrimps 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 world’s major importers of shrimp from Bangladesh.
In Thailand, the contribution of shrimp farming sector has been very impressive. Thailand has emerged as one of the largest seafood suppliers to global market and at present, the sector accounts for 30 per cent of the world trade in shrimps. Thailand’s shrimp production was about 600,000 tons and export value was US$2.28 billion in 2007 (www.fisheries.go.th). The major markets for shrimp are the USA, Japan, EU, Canada and South Korea.
Now‐a‐days, import countries have developed standard criteria and a full certification system for the shrimp production line, from farm to table. In view of 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 quality‐oriented 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 production procedures (Rouf 2004).
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 quality management and systemic performance objective (Ponte 2007). 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 indeed very high. Any disruption in this market was bound to have severe and important implications for this export‐oriented sector of the country, and negative multiplier impact for the national economy (CPD 2006).
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, this study deals with‐Japan’s shrimp imports and its trends; and measures of different stakeholders in shrimp industry in Bangladesh and Thailand in order to comply with the safety standards of Japanese importers and buyers.
2.0 Materials and Methods
Data and information for this study were collected from both primary and secondary sources. In Bangladesh, primary data were collected in 2007 from General Managers, Executive Directors of processing plants and shrimp farmers in Khulna District following random sampling. Semi‐structured interviews were 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 semi‐structured 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, one representative from a fish import company and a number of shrimp buyers were interviewed in 2008. Besides, secondary data and information from various organisations as well as from published and unpublished sources of government agencies and trade organisations in Bangladesh, Thailand and Japan in the type of documents, reports, handouts, notifications, etc. having relevance with this study were also consulted. In addition to these, follow‐up telephonic discussions were conducted with other knowledgeable personnel. The data and information from all these field surveys, interviews, communications and discussions were summarised which were analysed and used to write this paper.
3.0 Results and Discussion
3.1 Seafood safety standards and measures implemented by different stakeholders
As food safety is receiving increased attention in the international markets, it is essential to maintain appropriate quality of the exported food items through proper quality control and monitoring measures of respective public‐private
91
stakeholders. IFPRI (2003) mentioned ‘food safety is affected by the decisions of producers, processors, distributors, food service operators, and consumers, as well as by government regulations’.
In the seafood industry, as in other food industries, the food safety management system known as HACCP (Hazard Analysis and Critical Control Point) has come to be widely adopted in the US, EU countries, Canada and, to a limited degree, Japan (Ito 2007). During the 1990s, developed countries made a strong shift toward requiring the HACCP approach to assuring food safety. Under HACCP, companies are responsible for analysing how hazards such as food‐borne pathogens may enter the product, establishing effective control points for those hazards, and monitoring and updating the system to assure high levels of food safety (IFPRI 2003). It is designed to prevent hazard through inspection at multiple points in production processes, representing a departure from point inspection of end‐products (Cato 1998).
In response to the increasingly emerging food safety requirements, the government and the private sector are jointly implementing a series measures to protect shrimp industry in Bangladesh. As a member of Codex Alimentarius Commission (which states that countries should promote responsible fish trade in such manner and environment that is hygienically acceptable, safe and also meets quality requirements) together with the agreements like SPS (Sanitary and Phytosanitary, to apply measures necessary to protect human, animal and plant 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, to prevent the use of unjustified technical barriers to trade and it includes 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 export‐oriented shrimp processing plants in March 1998, the government has strengthened the Fish Inspection and Quality Control wing of the Department of Fisheries (DoF) to provide statutory support to the processing plants, to monitor processing and quality assurance activities and to fulfil the HACCP requirements (DoF 2005‐06).
Bangladesh shrimp industry practices standard in compliance of HACCP to produce safe and quality products. It is evident that Bangladesh shrimp industry
92
implements HACCP procedure about 85 per cent in all stages of production, distribution, processing and export of shrimp in order to supply safe and quality products to the importers and consumers. However, the shrimp industry was found to face a number of problems including shortage of quality seed and feed, illegal grabbing of government land by non‐farmers, imposition of turn over tax, natural hazards, disorganised production practices, harmful effects on coastal environment and soil fertility, etc.
Thai Ministry of Agriculture and Co‐operatives through the Department of Fisheries (DoF) implemented a voluntary HACCP fish inspection programme in 1991 for the first time, which was finally become mandatory in fishery establishments in 1996 (www.fao.org/decrep/008). Now all processors under DoF 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 factories to upgrade their operations, develop better plans, and handle fish more systematically.
3.2 Value chains of shrimp in Bangladesh and Thailand
Shrimp’s value chain involves a combination of different actors (located along different points of the shrimp industry) with their institutional arrangements and actions from brood shrimp collection to end use. The value chain, 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 disposal after use.
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 marketed to overseas consumers through a number of channels that is from farm to village traders, purchasing agents, depot owners and final delivery to the processing plants (Flow Chart 1). All the steps of value‐added activities are strictly controlled under GMP (Good Manufacturing Practices) and SSOP (Standard Sanitation Operating Procedure). Competent authority like FIQC (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), United Kingdom based Lloyd’s and Denmark based Baltic Control
93
perform inspection, testing, certification and verification services following the Codex guidelines, code of practice, standards and directives of European Union Food Law, HACCP regulations and requirements of other import countries. To ensure safety and quality of the shrimp, Department of Fisheries monitors the hygiene and sanitation conditions of the food through quality management and process‐oriented supervision throughout the shrimp value chain.
Flow Chart 1: Value Chain of Shrimp, Bangladesh
Export
Products inspected by competent authority (i.e., FIQC-Fish Inspection and Quality Control) and certified by third party certification agence certification agency
by ship or air
Servicing Centre
Village Traders Depot OwnersPurchasing Agents
by refrigerated van
Shrimp Farm/ Gher
Processing Plant
Negotiated and verified as per import order instruction by the local agency of the import country
by rickshaw/van/pick-up/boat
Facilities monitored by DoF
by rickshaw/van/pick-up/boat
Monitored by DoF
Nursery by air/rickshaw/van
by rickshaw or van
Hatchery
Monitored by DoF
by a
ir
Monitored by DoF
by rickshaw or van
Brood Checking Centre Checked and certified by Department of Fisheries
by local boat Collected/caught from the sea by trawler and
monitored by marine surveillance check post Brood Shrimp
Source: Adopted from BFFEA, Bangladesh and Field Survey 2007
94
S
IiHceatthsapw
Uoe
Flow Chart 2: Value Chain of Shrimp, ThailandFlow Chart 2: Value Chain of Shrimp, Thailand
Brood Stock
Hatchery & Nursery
Farming
Collector/Pre-processor
Processor
GAP, inspected, sampling fry, feeds
FMD
GAP, inspected, sampling shrimp/fish, feeds
MD
GMP, inspected
MD
GMP/HACCP, inspected, sampling finished products
Feed, drug manufacturers, distributors, suppliers
Licensed, registered, inspected etc.
GAP: Good Aquaculture Practices
GMP: Good Manufacturing Practices
HACCP: Hazard Analysis and Critical Control Point
MD : Movement Document, FMD: Fry Movement Document
Distribution/Export
Domestic consumersForeign buyers
Fish market
Product
Farm
Table
Control of Department of Fisheries
Source: Department of Fisheries, Thailand and field survey 2008
ource: Department of Fisheries, Thailand and Field Survey 2008
n Thailand, it is compulsory that all seafood processors and exporters must mplement CoC (Code of Conduct), GMP (Good Manufacturing Practices) and ACCP to produce quality shrimp. Thai Department of Fisheries (DoF), as ompetent authority for fishery products exported from Thailand, monitors the stablishments and their products regularly, administers several programmes nd activities to ensure that seafood safety is maintained properly through farm‐o‐table approach (Flow Chart 2). To ensure credible inspection and controls hroughout the value chain, competent authority covers all relevant aspects of ygiene, public health and, also animal health. In order to sustain the Thai hrimp industry and to maintain consumer confidence in product safety, it pplies standard practices to all sectors of the value chain from primary roduction to the market place and to apply a single standard for products hether the products are destined for domestic or overseas markets.
nder CoC guidelines, Thai shrimp industry applies best practices at each step f the value chain to meet consumer expectations for quality, food safety, nvironmental 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 certifies hatcheries, farms, distributors and processing plants in order to ensure that stakeholders follow the standardised practices. Guideline on GAP emphasises production of shrimp which is fresh, clean, free of therapeutic agents and chemicals and non‐contaminated 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 well‐organised, fully integrated shrimp industry (Wyban 2007).
3.3 Global value chain governance
The global value chain approach places particular emphasis on the coordination 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). Governance occurs when one firm follows parameters set and enforced (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 Produced‐product design and specification, (ii) how it is to be produced‐process specifications, and (iii) how much is to be produced and when production scheduling and logistics (Humphrey and Schmitz 2001).
Global farm (i.e., import countries/companies or supermarkets) negotiates with the local farm (i.e., processors‐cum‐exporters/suppliers) on product price, product quality, reliability of delivery, speed of response, product design and product performance/specification standards. Fulfilling such demand specification by local farms in export countries through the collective initiatives by the actors of value chain make local shrimp products more competitive in the market.
Normally, local agents of import countries (companies or supermarkets) 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 specifications are met based on consideration such as quality, design and branding together with safety and standards. The Japanese importers demand the processors‐cum‐exporters to produce and export following their
96
specifications. However, these refer to an important governing role of the buyers 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 buyer‐driven commodity chain, Islam (2008) concluded that shrimp commodity chain is neither buyer‐driven nor producer‐driven, but a twin‐driven commodity chain in which the wealthy buyers control supply network, while a third‐party certifier and some environmental groups define the regulatory aspects of production, codification, certification, and to some extent management aspects.
3.4 Japan’s shrimp imports and its trend
Shrimp is the largest import group‐both in volume and in value‐in the category of imported fisheries products to Japan. Followed by USA, Japan is the second largest shrimp buyer in the world. About 70 per cent of the imported shrimp go 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 Japan’s total shrimp imports volume have been fluctuating with declining trend 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 are increasing and they are consuming reduced amount of shrimp’. As shown in
Table 1: Japan’s Shrimp Imports by Product in Different Years (in Metric Tons)
Product form 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/in brine 1,977 2,351 2,008 2,035 1,648
Cooked & smoked 453 618 422 414 324
Sushi (with rice) 92 341 263 204 144
Live 293 383 271 184 167
Fresh/chilled 19 33 19 7 0.4
Total quantity 283,317 301,608 294,658 301,078 275,589
Source: Data compiled from Infofish 2008 (http://www.infofish.org/marketreports/shrimp)
97
the Table 1, Japan’s 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 lowest by Japan since 2003. Japan’s imports in the form of raw/frozen shrimp are declining since 2005 and at a record low in 2007. There is slightly a positive trend in imports of the prepared/preserved and cooked/frozen shrimps in 2006.
3.4.1 Market shares of different countries in Japan
Frozen shrimp imports from the major export countries (mostly located in Asia) to Japan are shown in Table 2. Vietnam, Indonesia, India followed by Thailand (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 Japan’s shrimp imports also dropped mainly on account of the presence of an anti‐biotic 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 227,202 million yen in 2006.
Imports of frozen shrimp are still dominated in Japan by Vietnam followed by Indonesia since 2004. There are higher imports of raw frozen shrimp (shell‐on 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 vannamei shrimp (more than 90 per cent of the country’s total shrimp production) in Asia after China. The Nation (2008) cited, however, white shrimp has a lower export price than black‐tiger shrimp, which will directly affect the country’s total 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 (Mathaba 2007).
Japan has long been the second largest (consuming 33 per cent of Vietnam’s seafood export) market for Vietnam, after the USA. Both Thai and Japanese firms have invested heavily in Vietnam fishery projects, where shrimp and catfish are the two major export species (www.siamcanadian.com). Vietnam gained US$ 746 million in 2007 in exports to Japan, 21.1 per cent of the country’s total seafood
98
export value (www.vasep.com.vn). Vietnam’s shrimp exports to Japan have recently been increased by 20‐30 per cent (www.fishupdate.com). The Ministry of Fisheries in Vietnam is taking measures to increase the safety of Vietnamese aquatic products in order to meet increasing demand from foreign markets. One of the measures is to develop aquaculture communities throughout the country, whose production processes, from the selection of breeds, feed and medicines to processing and export, will be put under strict supervision.
Table 2: Japan’s Frozen Shrimp Imports in Different Years (in Metric Tons)
Country of origin 2003 2004 2005 2006 2007
Vietnam
47,626 (20.4)
55,506 (23.0)
54,573 (23.5)
51,133 (22.2)
40,041 (19.3)
Indonesia
52,367 (22.5)
48,623 (20.1)
45,574 (19.6)
43,665 (19.0)
37,080 (17.9)
India
28,191 (12.1)
31,572 (13.1)
26,309 (11.3)
28,546 (12.4)
27,025 (13.0)
Thailand
16,803 (7.2)
17,192 (7.1)
18,398 (7.9)
20,097 (8.7)
26,380 (12.7)
China
20,494 (8.8)
22,609 (9.4)
24,092 (10.4)
22,810 (9.9)
23,997 (11.6)
Russia
8,641 (3.7)
8,911 (3.7)
10,382 (4.5)
9,518 (4.1)
8,903 (4.3)
Myanmar
5,377 (2.3)
7,630 (3.2)
7,519 (3.2)
8,847 (3.8)
8,021 (3.9)
Canada
8,938 (3.8)
8,469 (3.5)
8,054 (3.5)
8,665 (3.8)
7,554 (3.6)
Greenland
9,180 (3.9)
7,683 (3.2)
7,527 (3.2)
6,788 (3.0)
5,427 (2.6)
Bangladesh
3,004 (1.3)
3,415 (1.4)
3,194 (1.4)
4,001 (1.7)
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)
Total quantity (MT) Total value (Mil. Yen)
23,3251 (100.0) 225,409
241,445 (100.0) 216,820
232,443 (100.0) 213,858
229,952 (100.0) 227,202
207,257 (100.0) 204,661
Note: Figures in parentheses indicate percentages
Source of data assembling: SWR 2008 (http://www.swr.ucsd.edu/fmd /sunee/shrimp)
99
3.5 Value added products and buyers’ requirements
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 added products have been rising. Value addition always focuses on market and consumer expectations. Among seafood, shrimp exhibits the widest range and highest degree of value addition. In consideration of health aspect, every food items have been diversifying in developed countries of the world, where caloric counts, dietary plans, nutritional facts, serving size on seafood packs and recipies in customers language are useful addition to value added products. Inclusion of these will definitely increase the presence of producer’s products on supermarkets and other retail outlets of potential markets (BFFEA 2007).
Aside from meeting the sanitary and phytosanitary requirements of an import country, all parties in the supply chain must meet increasing consumer demand for quality and safety assurance of produce. They must also meet additional requirements of importers and retailers in terms of compliance with the relevant standards of the ISO (International Standard‐Setting 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 as freshness, quality, visual perfection, taste, stability of supply, and low prices. All stakeholders in the value chain must ensure and maintain these standards as 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 processing plants in Bangladesh and Thailand are involved in processing value added products following stringent sizing, weight, cleanness, colour grading and general quality standards. The quality of such value‐added products exported from both the countries is also being appreciated for maintaining taste and meeting consumers’ preferences as well.
3.6 Traceability requirements in shrimp export
Traceability of food at present emerges as a vital area of competition and a crucial demand trend across the world market. Under EU law, traceability means
100
the ability to track any food, feed, food‐producing animal or substance that will 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 ensuring traceability, exporters should maintain transparent and credible records of the traded products that he receives from the farmers or suppliers.
Both Bangladesh and Thai shrimp industries have meantime taken a good number of steps to implement this issue through internal traceability (i.e., recording information on raw materials, intermediate and final products within a fish plant) and external traceability (i.e., making documentation through market chain from fry collection to consumer plate) in order to comply with the requirements of the import countries. Processor will guarantee to the buyers that 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 relevant hygiene condition.
In Thailand, ‘movement document (MD)’ is issued by Department of Fisheries at hatcheries, farms through to processing plant to cover all cultured or imported species including ornamental fish, whereas ‘fry movement document (FMD)’ is practised for delivering fry to buyer in order to trace back to origin when drug 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 pathogenic bacteria and no chemical drugs are used during rearing; the shrimps are harvested, preserved and transported in a hygienic manner. Besides, in the processing plants of both Bangladesh and Thailand, all inner carton or polybag and master carton are labelled and printed with all necessary information and specifications such as, approval number, FDA (Food and Drug Administration) registration, place/country of origin, processors name or ID, lot number, date/method of production, etc. for the foreign buyers so that they could be able to trace the origin of each product.
3.7 Certification, eco‐labelling and sustainable seafood trade
In recent years, the landings of fish from the world’s 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, certification and eco‐labelling of seafood provides assurance to meet the needs of future generations.
It may be pointed out that eco‐labelling 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 certification criteria, the potential benefits to fisheries sector will be much higher. However, the potential usefulness of eco‐labelling schemes may be constrained due to the difficulties faced by the low‐income countries in terms of compliance, and developing necessary organisational capacity. Though eco‐labelling is considered to be an informal international environment‐promoting tool and in most cases these are voluntary in nature, its introduction in shrimp exports from Bangladesh needs to be actively considered since many developed countries have already adopted this schemes. The inability or unwillingness of Bangladesh to do this may lead to weakening of its competitive strength and erosion of its global market share (CPD 2006).
Certification of a seafood product indicates if it was produced in a sustainable, healthy, socially responsible and environmentally‐friendly way. It should address four main areas: food safety and quality, social impacts of fish farming on local communities, environmental issues and economic feasibility (George 2007).
Governments in both countries strongly prohibit shrimp fry collectors not to catch fry from the sea to control stock status. Strict supervision and monitoring system are needed to protect mother shrimps at the sea and there should be a clear‐cut law regarding catching of mother shrimp. To maintain biodiversity, trawlers fishing in the sea are using turtle and other non‐targeted species excluding device to reduce unwanted catching and to prevent damage to other species. Governments can make an impact through policy‐making; fish farmers can use sustainable techniques in fish harvesting so that future catches and harvest can be assured, and processors‐cum‐exporters have a responsibility to purchase fish from the sustainable sources. Finally, ecosystem based fishery management approach should be encouraged for sustainable harvesting to ensure fish availability for today and also for future catches and harvests.
For internationally certified organic aquaculture in Bangladesh, SIPPO (Swiss Import Promotion Programme) took a pilot project in 2004 at the district of
102
Satkhira, in South West of Bangladesh and next to the national mangrove forest. Farmers produce shrimp by using traditional methods based on the best utilisations of the sun, the rich waters of the nearby Sundarban and large space of land under a unique programme of reforestation, without use of any ingredients such as compound feed, chemicals, fertilisers. The shrimp grows naturally just with natural food coming from this fantastic natural food reservoir. With such initiatives, Bangladesh would be able to earn substantially by exporting shrimp to maintain competitiveness in the global market (BFFEA 2007).
Thai Department of Fisheries has developed specific projects to enhance the shrimp culture sector by improving various aspects such as environmental quality, shrimp product safety, and trust marketing. One of the projects named ‘seawater irrigation systems for marine shrimp culture’ that produces appropriate quality seawater for shrimp culture, collects effluent from shrimp ponds, and treats the effluent before discharge into natural water. The objectives of the systems are to produce sustainable shrimp culture, maintain environment and balance ecological system (DoF 2008).
3.8 Comparative features of both Bangladesh and Thailand shrimp industries
Thailand emerges as the world’s leading exporter of shrimp with 30 per cent of the market share. As the buying countries demand that the food must be produced, processed and handled with necessary hygiene maintenance, 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 finished products up to export are strictly controlled under GMP and SSOP.
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 farm‐gate price, where farmers have less bargaining power.
There are four types of shrimp being exported from Bangladesh, i.e., sea water black tiger shrimp (Penaeus monodon‐63% of total export) and fresh water prawn (Macrobrachium rosenbergii‐19% of total export), sea water brown/grey shrimp (Metapeneaus monoceros‐10% of total export), and sea water white/pink shrimp (Penaeus indicus‐8% of total export). On the contrary, white shrimp (Penaeus
103
vannamei‐ more than 90% of its export) and black tiger shrimp (less than 10% of its total export) from aquaculture are the major fishery commodities in Thailand. Both the countries follow traditional extensive method for shrimp aquaculture.
Thailand has four up to date and well‐equipped laboratories for testing and detection of bacteria, virus, antibiotics and heavy metals. They carefully examine the shrimp products prior to shipment to assure optimum quality in accordance 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 used to send samples to Thailand and Singapore for the required tests, which are now possible in Bangladesh.
In comparison to 42 large scale 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, Lloyd’s and Baltic Control, as international independent company, perform comprehensive inspection and superintendence both at the company and at shipping ports as per instruction of the import countries to ensure food safety.
While Bangladesh did not explore the possibilities of export diversification and still it remains happy with the export of frozen shrimp only, Thailand is continuing exports with its diversified range of product‐ frozen, canned, dried and salted shrimps.
Traceability has become an important factor in food safety which also helps to define the responsibilities of food operators. Thailand shrimp industry ensures it through both ‘movement document’ and ‘fry movement document’ whereas Bangladesh starts to execute it by ‘shrimp suppliers certificate’ and ‘farmers and region code’.
4.0 Conclusions
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 value‐added products is increasing. Therefore, countries with stringent quality assurance programmes would also be able to sell more value‐added products to Japan. Bangladesh has a possibility to expand its shrimp trade with Japan establishing value innovation through export diversification of products and meeting food‐safety standards properly.
Bangladesh should invest on ways to improve the present status of shrimp trade and make this avenue into the real business sharing and experiencing the knowledge from Thailand. There is a bright prospect of shrimp culture in Bangladesh recovering the illegally occupied government land and distributing those to the real shrimp farmers and utiliseing the full capacity of the processing plants in order to augment its production and export volume. In fact Bangladesh 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, ensuring chain traceability with effective farm registration procedure as well as strict supervision and monitoring system at every stage in the export chain are essential to satisfy the safety standards and requirements of developed country consumers.
Recently, Thailand has started to implement a new approach, i.e., ‘shrimp cluster’ integrating together all related supporting industries such as ice producer, cold storage, animal feed, packaging, transport, institution/association etc., to maintain products’ supply smooth with better quality and low cost to the market. It is also recommended to start the feasibility of such approach in Bangladesh to improve management, meet quality standards and increase products’ competitiveness.
105
References BFFEA. 2007. Shrimp & fish news, Newsletter of Bangladesh Frozen Foods Exporters
Association, March‐June edition, Dhaka, Bangladesh. Cato, J. C. 1998. Economics of hazard analysis and critical control point (HACCP)
programmes, FAO Fisheries Department, FAO, Rome. CPD. 2006. Fish trade liberalisation in Bangladesh: Implications of SPS measures and eco‐
labeling for the export‐oriented shrimp sector. Centre for Policy Dialogue, Dhaka, Bangladesh.
DoF. 2005‐2006. 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, Directorate‐General for Health and Consumer Protection, European Commission, Brussels.
Gammage, S., K. Swanburg, M. Khandkar, M. Z. Islam, M. Zobair, and A. M. Muzareba. 2006. A gendered analysis of the shrimp sector in Bangladesh. Gender Access to Trade and Expansion, USAID, Bangladesh.
George, K. 2007. FAO partnership working to establish guidelines for certification of farmer fish. FAO Newsroom, Media Relations, FAO, Rome. Greenpeace. 2008. Challenging the aquaculture industry on sustainability. Greenpeace International, Amsterdan, The Netherlands. Gereffi, G. 1999. International trade and industrial upgrading in the apparel in
commodity chain. Journal of International Economics, Vol. 48, No. 1, Pp. 37‐70. 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. and H. Schmitz. 2001. Governance in global value chain. IDS Bulletin, Vol. 32, No. 3, Pp. 19‐29.
IFPRI. 2003. Food safety in food security and food trade: 2020 vision for food, agriculture and the environment. International Food Policy Research Institute, Focus 10, brief 4 & 5 of 17.
Infofish. 2008. Shrimp monthly market report (http://www.infofish.org/marketreports /shrimp, accessed on 2008.4.12).
Islam, M. S. 2008. From pond to plate: Towards a twin‐driven commodity chain in Bangladesh shrimp aquaculture. Food Policy, Vol. 33, Pp. 209‐223.
Ito, S. 2007. Export competitiveness and pro‐poor growth in the shrimp industry in Bangladesh. Enterprise Development and Microfinance, Vol. 18, No. 2&3, Pp. 241‐254.
Ito, S. 2005. The distributional effects of compliance with food safety and agricultural health standards on small producers in developing countries. Proceedings of the Japan Society for International Development, Japan.
106
Jonker, T. H., H. Ito, and H. Fujishima. 2005. Food safety and quality standards in Japan: Compliance of suppliers from developing countries. Agriculture and Rural Development Discussion Paper, the World Bank, USA.
Joseph, G. 2008. Exports of shrimp to USA, Japan decline. Business standard, a financial daily, March 20, New Delhi, 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 (accessed on 2008.6.03). Ponte, S. 2007. Bans, tests, and alchemy: Food safety regulation and the Uganda fish
export industry. Agriculture and Human Values, Vol. 24, No. 2, Pp. 179‐193. 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, accessed on 2008.5.21). The Nation. 2008. New Australian inspections threaten Thai shrimp exports. Bangkok’s
Independent Newspaper, June 03, Bangkok, Thailand. Wyban, J. 2007. Thailand’s shrimp revolution, AQUA Culture Asia Pacific Magazine, Vol.
15, No. 4, Pp. 17‐22. Xinhua, Y. 2008. Comparative advantage analysis of shrimp production in Asia.
Aquaculture Asia Magazine, Vol. 9, No.1, Pp. 33‐36. http:// www.fao.org/decrep/008 (accessed on 2008.2.18). http://www.fisheries.go.th (accessed on 2008.4.10). http://www.fishupdate.com/news/fullstory.php/aid/3558/Vietnam.
(accessed on 2007.4.12). http:// www.siamcanadian.com (accessed on 2008.6.4). http://www.vasep.com.vn (accessed on 2008.7.14).
107
108 108
Asia‐Pacific Journal of Rural Development Vol. XIX, No. 1, July 2009
Simulation of Growth and Yield of Maize under Water Stress Imposed during Critical Growth Periods in
Nakhon Ratchasima Province, Thailand Kiattiyos Thongsaga* and S.L. Ranamukhaarachchi*
Abstract
Water stress is unpredictable in rain‐fed agriculture, and regularly affects maize production in Thailand. This study was conducted to determine the effect of 10‐day dry period occurring during critical growth stages on growth and yield of maize, and to simulate and compare the performance using CERES‐Maize model in DSSAT version 4.0.2.0 using concurrent weather data. Two popular maize hybrids, namely Suwan 4452 and Pacific 224, were subjected to four 10‐day dry periods [viz. 36‐45 days from seeding (DFS), 46‐55 DFS, 56‐65 DFS, and 66‐75 DFS] as treatments together with a control grown with regular irrigation in a split plot design with four replicates. The experiment was conducted during the longest dry period, from December 2006 to April 2007, under controlled irrigation.
Soil moisture content at 0‐15cm and 15‐30cm depths reached near the lower limit of readily available water (RAW) in six days, after the cessation of irrigation, which declined to near permanent wilting point (PWP), but did not reach PWP during the 10‐day period. Leaf area 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 10‐day suspension of irrigation, but various differences were observed in the plant height: Suwan 4452 was taller than Pacific 224. Number of grains per ear, 100‐grain weight, grain yield and the harvest index remained unaffected due to the 10‐day suspension of irrigation, but were significantly greater in Pacific 224 than Suwan 4452.
CERES‐Maize model showed excellent results in predicting time to silking, LAI and biomass at silking, 100‐grain weight, grain yield and the harvest index. However, some deviations were found in Suwan variety for LAI and 100‐grain weight, and of Pacific 224 in the time to silking. Comparisons showed that simulations were very much close, except for the cases noted. The model simulated responses similar to the actual values in this study.
1.0 Introduction Water stress due to erratic rainfall is a major constraint for maize production under rain‐fed 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 E‐mail: [email protected]
109
maize encounters drought conditions during critical growth periods (Jampatong and Balla 2005). Maize yield is usually affected by water stress occurring at any 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; Kefele and Ranamukhaarachchi 2004).
Corn is relatively tolerant to water stress during the vegetative stage, but very sensitive during tasselling, silking and pollination, and moderately sensitive during grain filling period (Shaw 1977; Dass et al. 2001). However, yield loss is maximum when drought occurs during the flowering stage (Doorenbus and Kassam 1979). This is because reproductive development at the time of flowering is especially sensitive to drought stress (Zinselmeier et al. 1999), which inhibits activities of both vacuolar and cell‐wall‐bound acid invertase activity during kernel development (Zinselmeier et al. 1995). Panitnok et al. (2005) observed the greatest effect on growth, development and grain yield of maize due to water deficit at tasseling, whereas the effect was the least during vegetative stage.
In Nakhon Ratchasima Province, where 26 per cent of Thailand’s maize is grown, the risk of water stress is highest during reproductive and grain filling periods of early season maize (March‐July), while for late season maize (August‐December), water deficit coincides with the vegetative phase (Thiraporn 1996). The area of maize in the first season is approximately 20 per cent, and 80 per cent in the second season. In order to avoid heavy yield losses, weekly irrigation of 40 mm of water has been recommended to maize growers with irrigation facilities [Personal communication with Dr. Jampatong, the Maize breeder & Agronomist in the National Corn and Sorghum Research Centre (NCSRC), Pak Chong District, Nakhon Ratchasima Province, Northeast of Thailand].
Maize is widely grown in rain‐fed conditions, where water stress periods are uncertain, supplementary irrigation is not pragmatic, and yield declines are often occurred. The only option left for the farmers is to adjust planting time following the historical rainfall probabilities, and soil storage of moisture is only an expectation, capable of supporting field grown maize during prolonged dry periods. However, the maximum length of a dry period that could affect maize production under field conditions is unclear to both the farmers and researchers. Therefore, it was hypothesised that 10‐day dry periods occurring at critical growth periods would not lead to significant growth and yield reductions compared to maize grown under non‐stressed conditions.
110
The effects of water stress have also been evaluated with crop simulation modelling in rain‐fed maize (Sinclair 1994). Therefore, it was also hypothesised that crop modelling in parallel to the field‐level 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 researchers in identifying management practices to minimise water stress effects and to ensure satisfactory maize yields.
This study was conducted to determine and compare the growth and the yield of maize under irrigation in parallel with simulations using CERES‐Maize Model of the Decision Support Systems for Agricultural Technology Transfer Version 4.0.2.0 (DSSAT‐ developed by IBSNAT). This study might provide the awareness of and opportunities to determine the degree of yield reduction caused by water stress occurring during critical growth stages and to develop and apply effective management decisions including the selection of appropriate planting time in order to minimise adverse yield reductions.
2.0 Materials and Methods
This experiment was conducted under irrigation during the dry season from December 2006 to April 2007 in the National Corn and Sorghum Research Centre (NCSRC) located in Pak Chong District, Nakhon Ratchasima Province of Northeast of Thailand (latitude 14.5°N, longitude 101°E, 360 meters above mean sea level).
3.0 Experimental Treatments
The study was designed to assess the impact of water stress by imposing 10‐day dry periods during critical growth stages of maize starting from seedling emergence, knee‐high stage, tasseling and silking, and physiological maturity.
The study was conducted as a 5 x 2 factorial experiment in a split plot design. Four 10‐day dry periods imposed by suspending irrigation at four critical growth periods [viz. 36 to 45 (growth stages V6 to V9), 46 to 55 (V9 to V14), 56 to 65 (V14 to VT) and 66 to 75 (VT to R2) days from seeding (DFS)] and a control (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 replicates below:
111
Note : Experimental treatments (dark lines show the timing of imposing 10‐day water stress by
suspending of irrigation.) [S – seeding; SE – seedling emergence; KH – Knee‐high stage;
VT‐ Tasseling; PM – physiological maturity; M – maturity]
4.0 Cultural Practices and Varieties
The conventional land preparation was adopted using a disc harrow, and ridges and furrows were formed at a 75cm distance. Each main plot was 16m long and 4.5m wide, and divided length‐wise into two subplots for the two maize varieties. To prevent lateral movement of water, a plastic sheet was placed vertically to a depth of 50cm around each main plot by making a deep drain and then covering from either side with soils. The period from early December to Early April of the 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. As basal dressing, N and P were applied at the rate of 20 and 25 kg ha‐1, respectively, by band application shortly before seeding as per recommendations of the NCSRC. Two seeds per hill were seeded on 01st December 2006 with inter‐ and intra‐row spacing of 75cm and 20cm, respectively, manually by hand. The following day, all plots were irrigated with sprinklers for about four hours which approximately provided 40mm water to each plot. Again sprinkler irrigation was practised for five days after first irrigation, and thereafter continued at weekly intervals until corn plants reached a knee‐high stage. From knee‐high 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 thinned out to the plant density of 6.67 plants m‐2. Nitrogen was top dressed at 115kg N ha‐1, 30 days after the seedling emergence. Irrigation was suspended for 10 consecutive days commencing from 36 days from seeding (DFS) and restored after 10 days for the first treatment. In the second treatment, irrigation was suspended during 46‐55 DFS, and for the third and fourth treatments during
112
56‐65 DFS and 66‐75 DFS, respectively. The dry periods coincided with critical growth periods, i.e. 36‐45 DFS with V6 to V9 (i.e. knee‐high stage), 46‐55 DFS with V9 to V14, 56‐65 DFS with V14 to VT (tasseling) and 66‐75 DFS with VT (tasseling) to R2 (grain filling). In all irrigation treatments, furrow irrigation was given on the previous day prior to suspension of irrigation. In addition, a control treatment was maintained without the suspension of irrigation during the full crop duration. Irrigation was terminated after the black layer formation in maize cobs in all the plots, and the crop was managed as per crop management guidelines of the NCSRC.
5.0 Experimental Observations
5.1 Soil moisture content
Soil moisture content (SMC) at the field capacity (FC) and the permanent wilting point (PWP) were determined within 0‐15 cm and 15‐30cm depths in each plot at the beginning. The pressure plate was used at ‐0.3 and ‐15 bars to determine the FC and the PWP, respectively. In addition, soil samples were taken at 2‐day intervals from the plots after irrigation was suspended for 10 days to determine soil moisture depletion. Soil augur was used for soil sampling, four soil samples were taken from each plot and each replicate, and gravimetric method was used to determine the SMC (Ryan et al. 2001). Fresh weight of soil samples was recorded first, and then oven‐dried at 105ºC until a constant weight was reached, the dry weight was recorded, and finally, the percentage of SMC was computed.
5.2 Plant height and leaf area
Both plant height and leaf area were recorded at 50 per cent silking (74 days after seeding ‐ DAS). Plant height was measured up to the tip of the tassel from 10 randomly selected plants in each plot. The leaf area was computed by recording the length and the width of the widest point of each leaf from the same ten plants in each plot. The leaf area was computed by multiplying the leaf length, width and a factor (McKee 1964) and leaf area index (LAI) was then computed (McKee 1964).
5.3 Plant biomass
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 and stem parts, and fresh weights were recorded. Then the stems were chopped,
113
and a sub‐sample of approximately 1.0 kg was taken from both leaves and the stems from each treatment and replicate, oven‐dried until a constant weight was reached and the weights were recorded. Using the weights of sub‐samples, the bulk weights of leaf and stem were converted to a dry weight (Ogoshi et al. 1999).
5.4 Yield and yield components
All maize plants in the two centre rows of each subplot leaving one meter area 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, 100‐grain weight and stalk weight. These plants were separated for grain and non‐grain parts, fresh weight of each part was separately recorded, dried in an oven until a 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, and the fresh weights were recorded. These fresh weights were converted to dry weight using the data obtained from a 10‐plant sub‐sample. The grain yield and the 100‐grain weight were computed at 15 per cent moisture content.
6.0 Experimental Design and Statistic Analysis
The data were analysed as per split plot design using SAS programme version 9.0. Growth and yield data were subjected to analysis of variance (AoV), and the means were compared with the Fisher’s Protected Least Significance Difference procedure (Steel and Torrie 1980).
7.0 Application of CERES‐Maize Model
The CERES‐Maize 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 solar radiation, maximum and minimum air temperature and precipitation, soil physical and chemical characteristics including drainage, runoff, slope soil classification containing soil depth, particle size analysis, pH, %C, %N and cation exchange capacity (CEC), and crop management data, which include management practices as per recommendation of the NCSRC. Crop residue of 1.3 t/ha was used as received from the previous crop. Genetic coefficient of maize (Hoogenboom 2004) as degree days (base 8oC) from emergence to the end of juvenile phase (P1), from silking to physiological maturity (P5), and for a leaf tip
114
emergence (phyllochron interval, °C d) (PHINT), photoperiod sensitivity coefficient (0‐1.0) (P2), potential kernel number (G2), potential kernel growth rate mg/kernel d (G3) used in the simulation are shown in the Table 1.
Table 1: Genetic Coefficients of Suwan 4452 and Pacific 224 Varieties of Maize
Genetic coefficient Suwan 4452 Pacific 224 Phenological coefficient Thermal time from seedling emergence to the end of the juvenile phase (P1), (expressed in degree days above a base temperature of 8ºC) during which the plant is not responsive to changes in photoperiod.
343.9
346.9
Extent to which development (expressed as days) is delayed (P2), for each hour increase in photoperiod above the longest photoperiod at which development proceeds at a maximum rate (which is considered to be 12.5 hours).
0
0
Thermal time from silking to physiological maturity P5, (expressed in degree days above a base temperature of 8ºC).
1149 1190
Phyllochron interval (PHINT), the interval in thermal time (degree days) between successive leaf tip appearances
48 48
Growth coefficient
G2, Maximum possible number of kernels per plant. 640 640
G3, Kernel filling rate during the linear grain filling stage nd under optimum conditions (mg/day).
8 8 a
Table 2: Selected Soil Characteristics at Different Soil Layers in Research Plots at the National Corn and Sorghum Research Centre, Pak Chong, Nakhon Ratchasima Province, Thailand
Depth, cm
Organic C %
Total N %
Avail. P ppm
Exch. K ppm
pH (1:1 water)
pH (1:1 KCl)
0‐15 2.84 0.17 8.1 194 7.2 6.9 15‐30 2.75 0.16 7.5 138 7.3 6.9 30‐50 1.96 0.11 6.3 91 7.3 6.9 50‐75 0.78 0.13 3.2 139 7.3 7.0 75‐100 0.16 0.06 2.5 102 7.4 7.3 Soils belonged to the Pak Chong soil series (PC), and other information included taxonomy (very‐fine, kaolinitic, Isohyperthemic Rhodic Kandiustox); texture
115
(clay‐53.8%; sand–5.6%). Layer‐wise soil analysis was conducted at the Agricultural Technology laboratory at AIT to use as inputs to the model (Table 2). The parameters included pH (1:1 soil: water and 1:1) (McLean, 1982); organic C [Walkley‐Black method (Nelson and Sommers, 1982; FAO, 1974)]; total N [Kjeldhal method (Bremner and Malvaney, 1982)]; available P [(Bray II method) (Bray and Kurtz 1945), exchangeable K (Barker and Surh, 1982) and CEC (Ryan et al., 2001)] (Table 2).
8.0 Model Validation
The simulated values were compared with actual values using bias (Eq.1) and root mean square of error (RMSE, Eq.2) as suggested by Willmott (1982). Graf et al. (1991) used the standardised bias (R, Eq.3) and standardised mean square error (V, Eq. 4) as the most appropriate to compare simulated and actual data on maize growth, yield and yield component.
where n is the number of field observations, Si and Ac are simulated and actual values, respectively. R and V are estimates for the overall error of the model with regards to field data. R quantifies the model’s ability to reproduce the actual growth pattern. Negative deviations from simulated values (Si‐Ac<0) compensate for positive deviation (Si‐Ac>0) and vice versa (Eq 3). Alternatively, V is a measure that reveals the model’s tendency to generally over‐ or under‐predict the yield and other parameters, while RMSE showed the reliability of the simulated data.
9.0 Results and Discussion
9.1 Weather conditions
The weather data during the experimental period at Pak Chong Regional Meteorological Station located at a distance of about one kilometre from the experimental site are presented in (Figures 1 and 2). The weather data show solar
116
energy (SE) ranging from 9 to 26 (Mj m‐2 d‐1), the maximum temperature (Tmax) from 25 to 37ºC, the minimum temperature (Tmin) from 12 to 24ºC, relative humidity (RH) from 26 to 70%, and evapotranspiration (ETc) of maize from 2 to 10 (mm/day). There was no rainfall prevailing during the experimental period from December 2006 to March 2007
Figure 1: Solar Energy, Mj m‐2 d‐1 (SE) and maximum (Tmax) and minimum temperature ºC (Tmin) during the experimental period from December 2006 to April 2007
0
5
10
15
20
25
30
35
40
01-Dec-06 31-Dec-06 30-Jan-07 01-Mar-07 31-Mar-07 30-Apr-07
SE Tmax Tmin
Figure 2: Evapotranspiration of the Plots (ETc), Rainfall (RF) and Relative Humidity (RH) during the experimental period from December 2006 to April 2007
117
RH
(%)
9.2 Soil moisture content
Soil moisture content (SMC) at field capacity (FC) and permanent wilting point (PWP) were 65.2 mm and 38 mm, respectively, whereas, the available soil moisture (ASM) was 27.2 mm and 18.1 mm, at FC and PWP respectively at 0‐15 cm depth in all treatments. With the suspension of irrigation for 10 days during
critical growth periods, the decreasing pattern of SMC on 2‐day interval is shown in Figure 3 and 4. Results showed that the SMC in the 0‐15 cm profile reached near the lowest level of readily available water (RAW) in six days, and by the tenth day the SMC reached the lowest level of RAW in treatments when irrigation was suspended during 36‐45, 46‐55, and 56‐65 DFS (Figure 3). When irrigation was suspended during 66‐75 DFS, the SMC reached the lowest level of RAW in six days and in the next 4 days (i.e. 10th day) it reached close to the PWP. The pattern of depletion of the SMC during the 10‐day period of suspension of irrigation at the same growth stages was similar in the 15‐30cm profile too, but the SMCs were higher in all plots compared to 0‐15cm depth (Figure 4).
Figure 3: Actual and Simulated Soil Moisture Content (SMC) during 10‐day suspension of irrigation in each treatment at 0‐15 cm depth
Figure 4: Actual and Simulated Soil Moisture Content (SMC) during 10‐day suspension of irrigation in each treatment at 15‐30 cm depth
118
CERES‐Maize model overestimated SMCs at both soil profiles, and bias (deviation of simulated value from actual value) was wider for 15‐30cm profile than 0‐15cm profile. The only exception was the SMC in the treatment where irrigation was suspended during 36 to 45 days from seeding at 0‐15 cm soil profile, which had the narrowest bias (Figure 3).
Both simulated and actual SMCs in all treatments indicate that the crop reached RAW and was slightly low during the 10‐dry period imposed in this study, 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 56‐65 and 66‐75 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 the usual severe dry period from December to April in the Pak Chong District of Nakhorn Ratchasima Province, a 10‐day 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, it would be worth examining the crop’s performance under the depletion of water prior to determining the frequency of irrigation and management options for alleviating water stress effects in maize.
9.3 Growth performance
9.3.1 Time to silking
There was a difference in the time taken for silking between the two varieties (Figure 5). For the variety Suwan 4452, time for silking ranged from 73.5 days in plots where irrigation was suspended during 56‐65 DFS to 74.5 days (mean) in the control plots (Figure. 5a). Where as for Pacific 224, silking time ranged from 72.5 days in plots that received a 10‐day dry period during 46‐55 DFS to 73.8 days in the control plots, and the rest has taken 73.0 days (Figure. 5b). Although the difference between control and irrigation suspended treatments was not wide enough, both varieties showed early silking under decreasing moisture availability when the irrigation was suspended for 10 days.
On the other hand, the model showed no significant difference in predicting the 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
0.00005 to 0.00014. The deviation for Pacific 224 was slightly higher with the bias ranging from 0.25 days for control plots to 1.50 days when irrigation was suspended from 46‐55 DFS, and corresponding R values ranged from 0.003 to 0.021 and V ranged from 0.0003 to 0.0005. This indicates that the model could be helpful in predicting the timing of silking of maize.
Figure 5: Actual and Simulated Time to Silking of a) Suwan 4452 and b) Pacific 224 Varieties of Maize as influenced by timing of suspension of irrigation
Table 3: Leaf Area Index, Plant Height and Above‐ground Biomass at 50% Silking
Treatment LAI at silking Plant height at
silking, m Biomass at 50%
silking Time of water stress, DFS 1/ 36‐45 4.4 ±0.4 2.32 ±0.2 7.4 ± 0.7 46‐55 4.4 ±0.4 2.24 ±0.1 6.8 ± 0.4 56‐65 4.6 ± 0.4 2.28 ±0.2 7.3 ± 0.7 66‐75 4.3 ± 0.4 2.30 ±0.2 7.5 ± 0.6 Control 4.5 ± 0.4 2.36 ±0.2 7.6 ± 0.7 Lsd (p=0.05) ns ns ns Variety Suwan 4452 4.6 ± 0.4 2.26 ± 0.1 b 7.3 ± 0.6 Pacific 224 4.4 ± 0.4 2.34 ± 0.2 a 7.3 ± 0.7 Lsd (p=0.05) ns 0.06 ns CV% 7.23 3.88 20.69 1/ DFS ‐ days from seeding
120
9.3.2 Plant height
Ten‐day suspension of irrigation had no significant affect on plant height which ranged from 2.24m when irrigation was suspended during 46‐55 DFS to 2.36m in the control treatment (Table 3).
However the maize variety had a significant effect on plant height, and Suwan 4452 variety was significantly shorter (2.26 m) than Pacific 224 variety (2.34 m). The difference was only 8cm between the two varieties with a CV of 3.88 per cent. The heights had no noticeable variation within each variety, except between the two varieties. There was no simulation applied to predict plant height in the study.
9.3.3 Leaf area index (LAI)
The LAI remained unaffected by both 10‐day suspension of irrigation and variety (Appendix Table 1). The LAI ranged from 4.26 with suspension of irrigation during 66‐75 DFS to 4.61 during 56‐65 DFS, while the control plots had a mean LAI of 4.45 (Figure 6). These values varied slightly for the two varieties: Suwan 4452 had a LAI ranging from 4.39 to 4.97 (Figure 6a), and Pacific 224 ranged from 4.11 to 4.5 (Figure 6b).
Simulated LAI values were lower than the actual values for all irrigation treatments and also for both varieties with the bias ranging from ‐0.39 to ‐1.10 for Suwan 4452 and from 4.11 to 4.56 for Pacific 224. For Suwan 4452, the bias was lower for 36‐45 and 66‐75 DFS, and R (0.09) and V (0.01) values were supportive of the model performance and reliability (CV = 7.23%) (Figure 6a). In case Pacific, the bias was lowest for 66‐75 day period of suspension of irrigation (‐0.11) with R= ‐0.03 and V= 0.01 (Figure. 6b). When irrigation was suspended during 36 to 45 days and 56 to 65 days in the two treatments, there was a prediction with moderate bias with R= ‐0.09 and V= 0.01. This shows that the simulations of LAI by CERES were acceptable. The results also showed that there was a varietal difference in response to water shortage periods, and Paciific variety appeared to perform slightly better than Suwan 4452 under short‐term water shortage. On the other hand, LAI values were within the optimum of 4.0 to 5.0 for tropical maize to ensure satisfactory yields. Duncan (1971) reported that yield increases of maize were achieved with a LAI of 4.0 but no yield increases occurred when the LAI increased up to 4.7. Although higher LAIs were found at higher plant population, the grain yield declined (Daynard et al., 1971). Therefore, in current study, LAIs could be considered sufficient to maintain satisfactory maize yields.
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, although negative values represented an underestimation of the LAI in the two maize varieties compared to the actual values. These values provide valid information to maize researchers and growers for examining the potential variation in plant growth and performance.
Figure 6: Actual and Simulated Leaf Area Index (LAI) of a) Suwan 4452 and b) Pacific 224 Varieties of Maize as influenced by timing of suspension of irrigation
9.3.4 Plant biomass at silking
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 that received no irrigation during 36‐45 DFS to 7.5 t/ha in those that received no irrigation during 66‐75 DFS. The control plot that received regular irrigation gave the biomass of 7.3 t/ha (Table 3). With respect to the varietal effect, both Suwan 4452 and Pacific 224 had 7.3 t/ha of biomass at silking (Table 3), which indicates that both varities have the same potential for biomass production at silking.
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 regular intervals (Figure 7). The model slightly underestimated the biomass at 50 per cent silking compared to the actual values in the control treatment with a bias of ‐0.80, R value of ‐0.10 and V value of 0.03 (Figure 7a and 7b). Bias of the biomass ranged from ‐0.04 in the treatment received no irrigation during 56‐65 DFS to ‐0.80 in the control treatment for Suwan 4452 with R ranging from ‐0.05 to 0.10 and V from 0.004 to 0.03 (Figure 7a), and for the Pacific 224 from 0.22
122
(overestimated) in the treatment received no irrigation during 46‐55 DFS to ‐0.73 in the control (Figure 7b). Simulated biomass values for the rest of the treatments ranged from ‐0.04 to ‐0.10 with R ranging from ‐0.07 to 0.03 and V from 0.002 to 0.01. Very low R and V values indicate the model’s ability to closely estimate the 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 performance of the rain‐fed maize crop. Figure 7: Actual and Simulated Biomass at 50% Silking of a) Suwan 4452 and
b) Pacific 224 Varieties of Maize as influenced by timing of suspension of irrigation
10.0 Grain Yield and Yield Components 10.1 Number of grains per cob
The number of grains per cob was neither significantly affected by the irrigation treatments nor the varieties (Appendix Table 2). Number of grains per cob of Suwan 4452 ranged from 577 in treatments exposed to 10‐day dry period during 36‐45 DFS to 611 in the control (Figure 8a), while in Pacific 224 grains per cob ranged from 572 in the treatment where irrigation was suspended during 46‐55 DFS to 600 when no irrigation was during 56‐65 DFS (Figure 8b). The control had a lower number of grains although not significant at p=0.05. The results indicated that there was no adverse effect of 10‐day dry period at any of the critical stages
123
of maize growth on number of grains per cob. As illustrated by Ritchie and Harvey (1982), the potential kernel number is determined by the vegetative stage V12 and continued until V17, which coincides with 36‐45 days from seeding. 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 46‐55 and 56‐65 DFS. Any stress occurring two weeks prior to silking advances tassel development, leads to early pollen shedding and at the same time delays silk development, eventually depriving pollination and fertilisation of ovules and developing to kernels (Ritchie and Harvey, 1982). In the period from silking (R1 stage) up to the dough stage (R4), grain filling occurs, assimilates mobilises into developing grain, and hence, water stress causes the formation of unfilled grains. In present study, insignificant differences in the grain number per cob between control and stress imposed treatments reveal that maize plants in the irrigation suspended treatments for ten days did not affect kernel initiation, pollination, fertilisation as well as grain filling. Although soil moisture in both 0‐15cm and 15‐30cm layers reached between RAW and PWP for about two to four days (Figure 3 and 4), the two varieties were able to overcome the stress effects caused by the absence of irrigation for ten days without compromising the grain number per cob, which is the most important parameter that contributes to the yield of individual plants.
Figure 8: Actual and Simulated Grain Number per Cob of a) Suwan 4452 and b) Pacific 224 Varieties as influenced by timing of suspension of irrigation
However, CERES‐Maize model tends to underestimate the grain number per cob under all irrigation treatments including the control (Figure 8a and 8b). The bias
124
was large and negative for both varieties, which indicates underestimation and its magnitude. The predicted values ranged from 275 grains per cob with the suspension of irrigation during 66‐75 DFS to 336 grains per cob in the control of both varieties, and both R and V values indicated unreliability of the simulated values.
10.2 100‐grain weight
Similar to the other parameters, 100‐grain weight remained unchanged due to irrigation treatments which ranged from 34.0 g with water shortage that occurred during 36‐45 DFS to 35.3 g when irrigation was suspended from 46‐55 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 interaction between the variety and the irrigation treatments on 100‐grain weight. The hundred grain weight is determined during the period from silking (R1) to dough (R4) stages. Upon pollination and fertilisation during R1, ovules develop into kernels and start accumulating starch (Ritchie and Harvey, 1982). This is a critical stage as water stress interferes with the translocation of assimilates. Insignificant grain weights among treatments including the control indicate that maize plants did not have an unfavourable impact from a 10‐day suspension of water. Findings of the study of Kefele and Ranamukhaarachchi (2004) using three maize varieties grown in Ethiopia showed that varietal difference exists for grain yield, and the level of water stress and the level of tolerance by the variety influence timing of tasseling and silking and its interval, which determine the timing of pollination and fertilisation and the potential kernel number and the rate of grain filling.
The model seemed to overestimate the 100‐grain weight for both varieties under imposed treatments, which ranged from 39.7g to 41.6g in plots that had irrigation suspended during 46‐55 DFS and 36‐45 DFS, respectively, while control had moderate value (41.3g) for Suwan 4452 and 41.4g and 43.5g, respectively for 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 suspension of irrigation from 46‐55 days and 56‐65 days, respectively for Pacific 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 model’s ability to reproduce the observed values were somewhat lower.
125
Figure 9: Actual and Simulated 100‐grain Weight of a) Suwan 4452 and b) Pacific 224 Varieties of Maize as influenced by timing of suspension of irrigation
10.3 Grain yield
There was a significant effect of variety on grain yield of maize, but the different irrigation treatment had similar effect on maize yield (Appendix Table 2). Grain yield was significantly higher in Pacific 224 (10.4 t/ha) than Suwan 4452 (9.1 t/ha)
Figure 10: Actual and Simulated Grain Yield of a) Suwan 4452 and b) Pacific 224 Varieties of Maize as influenced by timing of suspension of irrigation
126
(Figure 10). Grain yield varied from 9.5 t/ha when the dry period was imposed from 36‐45 days from seeding to 10.1 t/ha with 56‐65 days from irrigation. Control had 9.6 t/ha.
For Suwan 4452, the model overestimated the grain yield during all growth stages tested, except on 66‐75 DFS, during which the opposite resulted. This could be attributed to some unforeseen moisture stress during early grain filling. The bias was however low, except for the 10‐day dry periods imposed during 36‐45 and 66‐75 DFS (Figure 10). The R and V values too had the same trend. For Pacific 224, the model underestimated grain yield, except for the control. But the values were close to the actual and hence the bias was in a very narrow range. Both R and V values were also low, and these values indicate the suitability of the CERES‐Maize model for simulating maize yields using concurrent weather data. 11.0 Harvest Index
The harvest index (HI) indicates the fraction of yield out of the total above‐ground plant material produced (Holliday 1960). Actual values of the HI ranged from 0.43 in the control and the treatment received 10‐day suspension of irrigation during 56‐65 days from seeding to 0.46 for the rest of the treatments, with a moderate value (0.44) for the treatment received from the suspension of irrigation during 36‐45 day period for Suwan 4452, and for Pacific 224 from 0.52 with suspension of irrigation during 56‐65 DFS to 0.55 with suspension of irrigation during 36‐45 DFS (Figure 11). This shows that the Pacific variety has a higher HI than the Suwan variety, which also supports a higher grain production ability and higher grain yields.
The CERES Model estimated the same HI as observed under field conditions for the Suwan 4452, receiving a suspension of irrigation during 46‐55 days from seeding with RMSE of 0.03, slightly underestimated for the treatment exposed to 10‐day suspension of irrigation during 66‐75 days from seeding and overestimated for the rest of the treatments (Figure 11). The bias for HI ranged from ‐0.02 to 0.05 with R ranging from ‐0.05 to 0.12 and 0.007 to 0.04 for Suwan variety. For Pacific 224, the model underestimated HIs and the bias ranged from ‐0.05 to ‐0.08, R from ‐0.09 to ‐0.15 and V from 0.01 to 0.03. For both varieties RMSE values were less than 0.1, which indicates the reliability of the model for simulating HI within the ranges shown in Figure 11.
127
Figure 11: Actual and Simulated Harvest Index of a) Suwan 4452 and b) Pacific 224 Varieties of Maize as influenced by timing of suspension of irrigation
12.0 Discussion
The current study had two important considerations. The first is the fate of maize crop when the dry period occurs during critical growth periods. Critical growth 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; Moser et al. 2006). Rain‐fed maise usually experiences unexpected dry periods extending up to varying durations. It is important that the SMC in the root zone does not reach the permanent wilting point (PWP), though it may reach a level below readily available water (RAW) so that maize plants could operate with active water uptake mechanisms to acquire water for their survival and growth (Lambers et al. 1998; Huang 2000). When water stress develops during critical growth stages, it could hamper such developments and eventually lower the follow up growth and grain 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 such low water availability will depend on the rate of water release by the soil, the depth of root zone and the plant’s tolerance mechanisms (Huang 2000).
The results of the current study showed that the SMC at both 0‐15 cm and 15‐30 cm were above or near RAW after 8 days of withdrawn irrigation, but did not
128
reach PWP even in the next 4 days, with the 10‐day dry period occurring during any stage of maize growth in this study (Figure 3 and 4). However, in the treatment that imposed a 10‐day dry period during 66‐75 DFS, the SMC reached below RAW and was yet above PWP by 8 days after cessation of irrigation. 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 other hand, the period of 66 to 76 days coincides with the effective grain filling period on maize grown in Nakhorn Ratchasima, and water stress during this periods could interfere with the mobility of assimilates to developing grains thus reducing grain size and final grain yield (Richie and Harvey 1982; Pandey et al. 2000; Asch et al. 2001). In present study the LAI, plant height, biomass, number of grains per ear, 100‐grain weight, final grain yield and the HI in the treatments, which received a 10‐day suspension of irrigation during critical growth periods, did not show significant changes compared to the control. It was also interesting to note that Nakhorn Ratchasima Province of Thailand has severely dry weather and heavy blowing during the growing period of this study, but this information confirmed that a 10‐day dry period could be tolerated by the two maize varieties without significant yield losses, even though the SMC declined to between RAW and PWP during the concerned critical growth periods. Previous studies of rain‐fed maize grown during the dry season without supplementary irrigation in Indonesia showed that the SMC even did not reach the PWP at a soil depth of 30 cm and maize roots found even below a depth of 70cm, which supported satisfactory growth and yield of maize crop (Ahadiyat and Ranamukhaarachchi, 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 pre‐anthesis period significantly reduced the grains per ear and 1000‐kernel weight, thus reducing the grain yield. However, Mohr and Schopfer (1995) mentioned that the plants have special adaptations to tolerate against short‐term stress periods. This is very much connected with the 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 Ranamukhaarachchi 2004).
The second purpose was to examine whether concurrent modelling could be used to effectively guide the growers and researchers to make suitable decisions and identify suitable practices related to the alleviation of water stress during short‐term dry periods in order to avoid large yield reductions. With the use of concurrent weather and management data, the model gave excellent estimations for the number of days to silking, the LAI and biomass at silking, 100‐grain
129
weight, grain yield and the HI with low values of bias, R and V. However, the model underestimated time to take silking of the Pacific 224 for all treatments, except for the control, 100‐grain weight of Suwan 4452 and the number of grains per ear of both Suwan 4452 and Pacific 224. These variations could be attributed 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 CERES‐Maize v3.5 to predict the number of days to silking at high N rate (90 and 120 kg N ha‐1) 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 over‐prediction of grain yield of maize with N treatments. The current study revealed the potential of using modelling approach for the benefit of yield forecasting. The results of the current study confirmed that CERES‐Maize model can provide useful information of the effect of already prevailing dry period on the growth, development and yield of maize.
13.0 Conclusion
The 10‐day suspension of irrigation occurring during critical growth periods did not adversely affect growth and yield of maize in the Nakhorn Ratchasima Province. This helps to conserve the water from the current practice of irrigating maize crop at weekly intervals. The CERES‐Maize model in DSSAT V 4.0 too confirmed these results, except with slight deviations in a few parameters. Therefore, the application of crop simulation with the CERES‐Maize model using concurrent weather data could provide information on the effect that the crop could have already had as a result of prevailing short‐term water stress, which could signal agronomists, researchers and growers to determine and adopt suitable yield enhancing and protecting management practices for the balance period of the crop to mitigate further yield reductions.
130
References Ahadiyat, Y. R. and S. L. Ranamukhaarachchi. 2007. Effect of soil tillage and maize‐grass
intercropping followed by grass management on soil properties and yield of rainfed maize. Int. J. Agri. Biol., 9:791‐799.
Ahadiyat, Y. R. and S. L. Ranamukhaarachchi. 2008. Effect of tillage and intercropping with grass on soil properties and yield of rainfed maize. Int. J. Agri. Biol., 10:133‐139.
Asadi, M. E. and R. S. Clemente. 2003. Evaluation of CERES‐Maize of DSSAT model to simulate nitrate leaching, yield and soil moisture content under tropical conditions. Food, Agriculture & Environment, 1(3&4): 270‐276.
Asch, F., M. N. Andersen, C. R. Jensen, and V. O. Mogensen. 2001. Ovary abscisic acid concentration does not induce kernel abortion in field‐grown maize subjected to drought. European Journal of Agronomy, 15:119‐129.
Barker, D. E. and N. H. Suhr. 1982. Atomic absorption and flame emission spectroscopy. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of soil analysis, Part 2, Chemical and microbiological properties. Pp. 13‐26. American Society of Agronomy 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, Chiang Mai University, 2000. P. 160. Chiang Mai: Chiang Mai University.
Bray, R. H. and L. T. Kurtz. 1945. Determination of total, organic and available forms of phosphorus in soils. Soil Sci., 59:39‐45.
Bremner, J. M. and C. S. Mulvaney. 1982. Total nitrogen. In: A.L. Page, R.H. Miller, and D.R. Keeny, (eds.) Methods of soil analysis, Part 2, Chemical and microbiological properties, 2nd edition, Number 9 in the series. Pp. 595‐624. American Society of Agronomy and Soil Science Society of America, 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 (Zea mays L.). In Sustainable maize production system for Nepal proceeding of maize symposium. Pp. 113‐116. Kathmadu, Nepal.
Daynard, T. B., J. W. Tanner and W. G. Duncan. 1971. Duration of grain filling period and its relation to grain yield in corn, Zea mays L., Crop Science, 11:45‐48.
Doorenbus, J. and A. H. Kassam. 1979. Yield response to water. FAO Irrigation and Drainage. Paper No. 33 Rome. Italy.
Duncan, W. G. 1971. Leaf angles. Leaf area and canopy photosynthesis. Crop Science, 11:482‐485.
Eck, H. V. 1986. Effect of water deficits on yield, yield component, and water use efficiency of irrigation corn. Agronomy Journal, 78:1035‐1040.
Ekasingh, B., P. Gypmantasiri, K. Thong‐ngam and P. Grudloyma. 2004. Maize in Thailand: 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 (A laboratory manual). Food and Agriculture Organization, Rome, Italy.
131
Graf, B., M. Dingkuhn, F. Schnier and V. Coronel. 1991. A Simulation Model for the Dynamics of Rice Growth and Development: III. Validation of the Model with High‐Yielding Varieties. Agricultural Systems, 36:329‐349
Grant, R. F., B. S. Jackson. J. R. Kiniry and G. F. Arkin. 1989. Water deficit timing effects on yield component in maize. Agronomy Journal, 81:61‐65.
Gungula, D. T., J. G. Kiling and A. O. Togun. 2003. CERES‐Maize prediction of maize phenology under nitrogen‐stressed condition in Nigeria. Agronomy Journal, 95:892‐899.
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 anniversary of the cooperation between Kasetsart University and Swiss Federal Institute of Technology/. Pp. 112‐ 116. Bangkok: Asksorn Siam Printing.
Holliday, R. 1960. Plant population and crop yield. Field Crop Abstracts, 13:1‐16. Huang, B. 2000. Role of root morphological and physiological characteristics in drought
resistance of plant. In Wilkinson, R.E. (Eds), Plant‐environment interactions. Pp. 39‐64. New York: Marcel Dekker, Inc.
Hoogenboom, G. 2004. Genetic Coefficients‐CERES‐maize/sorghum/millet. In Hussain, S.S. and Mudasser, M. (Eds.), South Asia regional training workshop on “Crop simulation modeling” at the Multiple Cropping Center, Chiang Mai University, Thailand. Pp. 255‐269. Chiang Mai, 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 marginal environments 25th anniversary of the cooperation between Kasetsart University and Swiss Federal Institute of Technology. Pp. 99‐104. Bangkok: Asksorn Siam Printing.
Kefale, D. and S. L. Ranamukhaarachchi. 2004. Response of maize varieties to drought stress at different phonological stages in Ethiopia. Trop. Sci. J., 44:44‐49.
Lafitte, H. R. 2000. Abiotic stresses affecting maize. In Tropical maize improvement and production, Rome: FAO plant production and protection series No. 28.
Lambers, H., F. S. Chapin and T. L. Pons. 1998. Plant water relations. Chapter 3 in Plant physiological ecology. Pp.154‐229. New York: Springer.
McKee, G. W. 1964. A coefficient for computing leaf area in hybrid corn. Agronomy Journal, 56: 204‐241.
McLean, E. O. 1982. Soil pH and lime requirement. In Page, A. L., Miller, R. H., Keeney, D.R. (eds.), Methods of soil analysis, Part 2, Chemical and microbiological properties. Pp. 199‐224. American Society of Agronomy Inc. 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. 539‐566. Publ. Springer‐Verlag, Berlin, New York.
Moser, S. B., B. Feil, S. Jampatong and P. Stamp. 2006. Effects of pre‐anthesis drought, nitrogen fertilizer rate, and variety on grain yield, yield components, and harvest index of tropical maize. Agricultural Water Management, 81:41–58
NeSmith, D. S. and J. R. Ritchie. 1992. Short‐and long‐term response of corn to pre‐anthesis soil water deficit. Agronomy Journal, 84:107‐113.
132
Nelson, D. W. and L. E. Sommers. 1982. Total carbon, organic carbon, and organic matter. In: Page, A.L., Miller, R.H., Keeney, D.R. (Eds.), Methods of soil analysis, Part 2, Chemical and microbiological properties. Pp. 539‐579. American Society of Agronomy Inc., Madison, WI, USA.
Ogoshi, R. M., Jr., B. M. Cagauan, and G. Y. Tsuji. 1999. Field and laboratory methods for the collection of the minimum data set. Vol. 4‐8 in Hoogenboom, G., Wilkens, P.W. and Tsuji, G.Y. (Eds.) DSSAT version 3 vol. 4 (Pp. 218‐286). 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. Agricultural Wate Management, 46:1‐13.
Panitnok, K., S. Tubngeon, S. Techapinyawat, T. Somwang, S. Lim‐aroon, 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. 140‐144. Bangkok: Asksorn Siam Printing.
Ritchie, S. W. and J. J. Hanway. 1982. How a corn plant develops. Special Report No. 48. P. 21. Iowa State University of Sci. and Technology Cooperative Extension Service, Ames, Iowa, (Revised edition Feb. 1982),.
Ryan, J., G. Estefan, and A. Rashid, 2001. Soil and plant analysis laboratory manual. Aleppo, Syria: ICARDA, 172 pages.
Shaw, R. H. 1977. Climate requirement. In Sparague, G. F. (eds.), Corn and corn improvement. Pp. 591‐623. 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.), Physiology and determination of crop yield. Madison: ASA, CSSA and SSSA. Pp. 509‐532.
Steel, R. G. D. and J. H. Torrie. 1980. Principles and procedures of statistics, A biometrical approach. P. 633. New York: McGraw‐Hill Book Company.
Thiraporn, R. 1996. Maize: Production, use, Problem analysis and Technology transfer to farmers (in Thai Language). P. 274. publ. Dansuttakhanpim Co. Ltd, Bangkok, Thailand.
Walkley, A. 1947. A critical examination of rapid method for determining organic carbon in soils: Effects of variations in digestion conditions and of organic soil constituents. Soil Sci. 63:251‐263.
Willmott, C. J. 1982. Some comments on the evaluation of model performance. Bull. Am. Meteorol. Soc, 63:1309‐1313.
Zinselmeier, C. M., J. Lauerand and J. S. Boyer. 1995. Reversing drought induced losses in grain yield maintains embryo growth in maize. Crop Sci. 35:1390‐1400.
Zinselmeier, C., B. R. Jeong, and J. S. Boyer. 1999. Starch and the control of kernel number in maize at low water potentials. Plant Physiologist, 121:25‐35.
133
Appendix
Appendix Table 1: Source of Variation, Degrees of Freedom and Mean squares obtained from analysis of variance for leaf area index (LAI), plant height and biomass at 50 per cent silking
MS2 Source of
variation1 DF
LAI at silking Plant height Biomass at 50% silking
W 4 0.15 158.85 8.05 Blocks 3 0.41* 3 1388.46*** 0.53 Error (a) 12 0.14 145.05 5.26 V 1 0.36 705.60** 0.17 V x W 4 0.20 177.0 2.46 Error (b) 15 0.10 80.13 5.64 CV% 7.23 3.88 10.20
Note: 1 W – Time of water stress (days from seeding); V – Variety 2 MS – Mean sum of squares 3 *, ** and *** indicate the level of significance of at p=0.05, 0.01 and 0.001, respectively
Appendix Table 2: Source of Variation, Degrees of Freedom and Mean Squares obtained from analysis of variance for yield components and yield and harvest index (HI) of maize
MS2 Source of
variation1 DF
Grains per ear, no.
100‐grain weight
Grain yield HI
W 4 769.81 2.26 0.57 0.42 Blocks 3 1134.00 7.82* 3 3.55** 3.88* Error (a) 12 201.62 2.46 0.65 1.79 V 1 759.68 215.80*** 17.07*** 33.65*** V x W 4 430.46 1.87 0.32 0.21 Error (b) 15 653.57 1.65 0.38 0.002 CV% 4.35 3.73 6.33 10.15
Note: 1 W – Time of water stress (days from seeding); V – Variety 2 MS – Mean sum of squares 3 *, ** and *** indicate the level of significance of at p=0.05, 0.01 and 0.001, respectively
134
Asia‐Pacific Journal of Rural Development Vol. XIX, No. 1, July 2009
Decentralisation on Access to Rural Water Supply Services: A Study of Six Villages in West Bengal, India
Indranil De*
Abstract
A survey of six villages of Birbhum District of West Bengal in India illustrates inter as 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 villages where literacy rate is higher. Moreover, education level of the household also has positive impact on access, considering the intra‐village allocation of water supply. External aid for capital investment in water supply has also been found to have increased access. The redressal of grievances regarding water supply is also better in villages where literacy rate is higher. The study concludes that decentralised delivery mechanism in provision of water supply services increases disparity in access in the presence of socio‐economic disparity.
1.0 Introduction
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 also has indirect impact on economic growth. The current state of access to water supply services in the developing countries is awful. About two of every ten people in the developing world were without access to safe water in 2000 (World Bank 2004). Decentralisation in delivery of water supply has been initiated to improve the access in countries like Ghana, South Africa, Malawi, Uganda, Zimbabwe and Ethiopia. The experiences of these countries with this respect are mixed. While in Ghana and South Africa decentralisation has led to substantial increase in the coverage, but in Malawi, Uganda, Zimbabwe and Ethiopia it has 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, street lighting and road.
* Business Analyst, HSBC Technology and Services, India.
135
The participation of local government in provision of basic services is expected to increase efficiency of service delivery (Tiebout 1956; Oates 1972, 1977). This is because local government operates more closely to people than any other higher level of government. Therefore, local governments would be able to identify the needs and preferences of the communities. Instead of local government, if line agencies are entrusted with the work of implementing programmes of provision 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 Nicol 2003).
In countries like Malawi, Uganda and Zimbabwe, it has also been observed that provision of water services (investment and maintenance) are significantly greater in already well served area, dominated by powerful politicians than in poorly served areas (Slaymaker and Newborne 2004; Mtisi and Nicol 2003). The existence of elite capture in distribution of water supply has also been observed 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 spill‐over public good, residential proximity to elected representatives matters in provision. Local governments have been found to be more active in the main villages, neglecting the hamlets. There are also instances that efficiency of water supply services in India is less in a decentralised system than in a centralised system (Asthana 2003). In this backdrop, the present paper has examined the variations in access to water supply and efficiency of redressal by the PRIs with respect to the grievances related to water supply. These variations may occur due to divergence in their socio‐economic characteristics and institutional factors.
West Bengal state has been selected for the present study as it has a long history of decentralisation. It is among a handful of other Indian states, which has initiated decentralisation even before the 73rd Constitutional Amendment. 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 back in 1973. To quote from the Bengal. Act. XV of 1939 ‘.. the duties of a Gram
136
Panchayat shall be to provide within the area under its jurisdiction for‐ … supply of drinking water and cleansing and disinfecting the sources of supply and storage of water’.
The Public Health Engineering Department (PHED) is the line agency for public 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 structure is horizontal structure, which consists of GP, PS and ZP. Proposals of 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 tube‐wells are selected by elected representatives of GPs. The operation and maintenance (O&M) of water supply schemes may be done either by the PHED or PRIs or by both. The responsibility of O&M of large piped water supply schemes is mainly devolved to the PHED. Both PHED and PRIs are involved in O&M of small and medium piped water supply schemes. GPs are primarily responsible for maintenance of spot sources such as tube‐well and hand‐pump. Overall, the local governments are intertwined with the line department at various administrative levels for provision of water supply. Nevertheless, GP is the first contact point for the people to make any complain regarding problems of access, availability and quality of water.
2.0 Objectives of Study
The general objective of the study is to analyse the impact of decentralisation on water supply services in six villages of Birbhum District in West Bengal.
The specific objectives of the study are as follows: • 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
problems of water supply • To suggest policy implications arising out of the study
1 The West Bengal model of devolution could be applauded not so much for the transfer of a large
number of important specific responsibilities in the rural development arena exclusively to the PRIs, but for the ‘consultative mechanisms’ built into the system whereby ‘decision‐making’ powers in the form of ‘administrative sanction’ of the schemes proposed and to be implemented by the line department of the State government at the district and sub‐district levels are vested on the Standing Committee of the Zilla Parishads (Subrahmanyam and Choudhury 2002).
137
3.0 Survey Methodology
The methodology adopted in the study regarding selection of study area and study sample has been illustrated in this section. Moreover, this section has also described the specifications of the models used to explain the determinants of access to water supply and redressal of water supply problems.
3.1 Selection of study area
Birbhum District of West Bengal has been selected as survey area in the study due to characteristics of rural water supply and socio‐economic indicators of the district. The survey has been conducted during October‐November 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 piped water supply has been borne by both PHED and the PRIs in the district2. Moreover, it is one of the socio‐economically backward districts of West Bengal. The socio‐economic characteristics of Birbhum districts vis‐à‐vis that of West Bengal has been presented in Table 1.
Table 1: Socio‐economic Characteristics of Rural Areas in Birbhum and West Bengal in 2001
(Figure in Percentage) Indicators Birbhum West Bengal Literacy rate 60.55 64.06 SC & ST population 38.96 36.16 BPL families 44.02 36.68 Households having no latrine 88.36 73.07 Households using tap 6.58 7.02 Households using handpump & tube‐well 80.91 79.97
Households using well 11.7 11.41
S ource: Census of India 2001
3.2 Selection of study sample
The villages have been chosen following a two‐stage stratified sampling method. In the first stage three GPs have been selected from three different blocks according to the characteristics of water supply systems3. Three different GPs have been selected according to (1) government investment in water supply and 2 For other water supply systems the responsibility of maintenance is generally with the PRIs in
West Bengal 3 One GP has been selected from each block since institutional structure of delivery of water
supply service is almost same within a block.
138
O&M of piped water supply by PHED, (2) government investment in water supply and O&M of piped water supply by PRIs, and (3) external aid for capital investment and O&M by both PHED and PS. At the second stage, the most prosperous village (MPV) and the relatively less prosperous village (RLPV) have been chosen to compare and contrast delivery of water supply services across these villages (Chart 1). Literacy rate of the villages has been used as proxy for prosperity. Furthermore, within a village, a combination of cluster and systematic sampling of households have been conducted.
Literacy has also been used in the study as a proxy for capability to bargain and efficiency of the GP. Literacy movement has positive spill‐over effects in terms of 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 intra‐as well as inter‐village allocation. It also increases the efficiency in the functioning of the GP as it increases consciousness of the people regarding the roles and responsibilities of the local government. PRIs become more accountable to local people if education level of households or literacy rate of the jurisdiction is higher.
Chart 1: Two‐Stage Stratified Random Sampling in Birbhum District
Source: Field Survey 2006
139
The sample sizes in each village and their socio‐economic characteristics have
of models
to find out the determinants of
gs
i = 1, 2 …m households
= = caste of the household, D = economic
redressal of problem has been specified as:
seholds
lem
he ste of
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 Alliance represents more than 90 per cent of the seats according to West Bengal Panchayat Election, 2003.
3.3 Variables and specifications
Regression models have been used in the studyaccess to water supply and chances of redressal of water supply problems.
Different socio‐economic variables, source type, participation in GP meetinand village specific characteristics have been considered as determining factor/ independent variables in the regression analysis. The socio‐economic 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 education by any respondent has been considered as proxy for education level of the household. The caste of the household is either Scheduled Caste (SC)/ Scheduled Tribe (ST) or general. To take into account the economic status of the households, an index of deprivation has been constructed. The methodology of construction of the index has been discussed later in this section. The function of access to water supply can be specified as:
W = a + b E + c C + d D + e S + f M + g V + ei
Where W = access to water supply
E education level of the respondent, Cdeprivation of the household, S = source of water supply, M = attendance at GP meetings by any member of the household during last one year, and V = village characteristics.
The function for
R = a + b E + c C + d D + e M + f P + gV + ei i = 1, 2 …m hou
Where R = redressal of problem and P = type of water supply prob
T education level of the respondent has been denoted as ‘education’, cathe household has been represented as ‘caste’, level of deprivation of the household has been indicated as ‘deprivation’, attendance of any member of the household in GP meeting during last one year has been expressed as ‘meeting’,
140
type of water supply problem reported to the GP during last one year has been denoted by ‘problem’ and the type of water supply source (either stand‐post or other) has been represented as ‘source. .
It is expected that education level of the respondent, caste and deprivation of the
reported to the GP is expected
to account the economic status and living condition of the households
not own land, radio, TV, telephone, bicycle, scooter and car
tricity at home
household are important components of bargaining strength for access to water supply. It is presumed that access to water supply increases with the education 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. Access to water supply may also depend on the type of source, since availability time of water for piped water supply is limited.
The chances of redressal of water supply problemto be high if education level of the respondent increases, caste of the household is higher and deprivation level of the household is lower. Higher education, 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 higher due to greater proximity of household to the GP office bearers. 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 the GP.
To take inan 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 is assigned and 0 otherwise. If the house is pucca, then household gets a score 0 and 1 otherwise. If the household has electricity at home then the household will get a score 0 and 1 otherwise. All these scores have been added to construct the index of deprivation. The index ranges from 0 to 9. The index of deprivation can be written as:
D = ∑ Si i = 1, 2, …9 Where Si = 1, if do if house is not pucca and if do not have elec 0, otherwise
141
The access to water supply has been categorised into three different categories,
The measure of access to water supply is an ordinal categorical data. To estimate
c
The results arising out of the study are discussed below:
4.1 Access to water supply
There are both public and private sources of water supply in the survey villages.
Within the ‘outside the premises public water supply sources’, 40 per cent of the
Pipe water service is not available in Ruppur (RLPV of Ruppur GP) and
such as low, medium and high. Access is low if both source is more than 50 meters away and have to wait for more than 15 minutes to collect water; access if medium if either source is less than 50 meters away or have to wait less than 15 minutes to collect water; and access is high if both source is less than 50 meters away and have to wait less than 15 minutes to collect water.
the functions of access to water supply, an Ordered Probit Model has been used. The education level of the respondent, caste of the household and level of deprivation of the household are highly correlated. To avoid multi‐collinearity problem between them in regression, three separate models have been constructed. In Model I, ‘edu ation’ level of the respondent has been used. In Model II and Model II, ‘caste’ of the household and level of ‘deprivation’ of the household have been used respectively. The function of redressal of water supply problem is a binary discrete choice model. It has been estimated through Probit method. In this case too, multi‐collinearity problem has been taken care of by constructing three different models as mentioned earlier.
4.0 Results and Discussions
The public water supply sources are in general located out of the premises. Only public tap water supply provides household connection inside the premises. The outside the premises public sources are stand‐post, deep tube‐well, hand‐pump, masonry well and ‘other’ sources such as tank, pond etc. The in‐house water sources are in‐house tap, hand‐pump, and well. It has been observed that 78 per cent of the sample households collect water from ‘outside the premises public water supply sources’ (Table 2). Percentage of households collecting water from in‐house tap is 9 per cent and in‐house ‘other’ sources (hand‐pump/ tube‐well/ well) is 13 per cent.
sample households collect water from deep tube‐well, 32 per cent from stand‐post, and 26 per cent from hand‐pump. The rest of the households collect water from masonary well and other type of systems like government reservoir.
142
ble 2: Percentage Distribution of Households by Sources of Water Supply
e : A = In‐house Tap Connection
B = In‐house Hand‐pump/ Deep Tube‐well/ Masonary Well
C = Out of the Premises Public Water Supply Sources
Field Survey 2006
Illambazar Ruppur Pourandarpur Ilambazar Baruipur Surul Ruppur Purandarpur Tapaipur
(MPV) (RLPV)
Total (MPV) (RLPV)
Total (MPV) (RLPV)
Total l
All GPanch
A Total 11.4 0 6.3 36.7 0 18 3 4 3.4 9.
ram ayat
3 B Total 22.9 3.4 14.1 10 3.2 6.6 24.2 8 17.2 12
Standpost 78.3 42.9 58.8 52.9 0 19.1 29.2 0 15.2 31.6 .9
Ta
Not
Source:
Deep Tubewell
17.4 35.7 27.5 35.3 93.3 72.3 29.2 9.1 19.6 39
Handpump 0 17.9 9.8 5.9 6.7 6.4 41.7 90.9 65.2 26
Masonary Well
0 0 0 5.9 0 2.1 0 0 0 0.
Other 4.3 3.6 3.9 0 0 0 0 0 0 1.
C
Total 65.7 96.6 79.7 53.3 96.8 75.4 72.7 88 79.3 78
.6
.4
7
4 .1
144
Tapaipur Village (RLPV of Pourandarpur GP). The percentage of households taking water from ‐p m tha 50 per cen Ilambazar, Baruipur and Surul Vil ‐h w r r provided in Surul Village. In all o Ps in‐house tap connections are private.
The remaining p of the st y as ly considered ‘ou de the premises public water supply so jo of collect water from these sou es. It has been found that considering all types of sources, households urce is around 38 ca stand‐po it is almost the same. It is highest in of ‐well at ar nd households source (Table 3). In Illambazar Ruppur GP, number of households per source is higher in RLPV than in MP more prosperous villages are better served presumably because they have str th the However, the reverse holds true in Pourn rpu h e th umber of per source is lower the R (Tapaipur) than in MPV (Purandarpur). As the percentage of Scheduled Cast ) a T e po tion is very high in Tapaipur Villag App l ), distribution of earmarked resource for SC inhabited areas is also higher4. The average number of households urce is be due to high capital invest r aided projects.
Ta Average u r e Premises Public u ’
a R ur Poura
Source: Survey 2006
standge. Inther G
oste p
is ublic
ore tap
n ate
t ins onlyla ous sou ce i
art urc
udhe
hma
onrity
tsihouseholdses’ as t the sample
rc so
per caseandV.
tube
. In ou
se of 41
st, per
The
LPV
greaada
ter bargr GP,
ain wing er
enge n
in GP. households
in
ment
ble 3:
e (SCe (
ndend
Scix
heduTab
lede A2
rib there
(ST)fore
pula
anr soxte
d ST
nally pe in e
lowest in Ruppur GP. This may
NumerberSup
of plyHo So
seholdsrce
pe ‘Outside th Wat
Illamb zar upp ndarpur
Ilam‐ Barui‐
Total Surul Rup‐ r
Total Puran‐
Total
All GP bazar pur pu darpur
Tapaipur
‐
Standpost 2 81.43 38.52 35.56 26.67 32 6.89 26.89 81.43Deep ‐well 79.5 67.14 21.67 31.75 29.97 47.86 44.44 41.39 Handpump 40 50 31 37.33 44.7 33.13 34.37 Masonary well Other 22 22 All sou 42.04 26 31.7 29.64 56.33 42.7 38.2
4 Th or Union Territories are required to earmark and utilise at least cent ARWSP
(A ed Rural Water Supply Programme) funds for drinking water to SCs and an r minimum 10 per cent STs (GoI 2008)
tube
rces
e Statecceleratothe
Field
36.2540
4
40
32.527.35
27.82
20 20
34.3 48.39
for the .
25 per supply
of the the
4.2 Distance of source
The distance of ‘outside the premises public water supply sources’ is ‘less than 50 meters’ for more than 50 per cent of the households. Distance of sources from the residence of the household is more in RLPV than in MPV. In case of all the
GPs, percentage of respondents reporting distance of source ‘less than 50 meters’ is higher and ‘more than 150 meters’ is lower in MPVs than in RLPVs (Table 4). It may be due to relative abundance of water supply sources in MPV as compared to RLPV. The relative abundance of water supply sources in MPV as compared to RLPV can be explained by greater bargaining capability of residents of MPVs for 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 than in RLPVs.
Table 4: Percentage Distribution of Households by Distance of ‘Outside the Premises Public Water Supply Sources’
Illambazar Ruppur Pourandarpur
Ilam‐
bazar Barui‐ pur
Total Surul Ruppur Total Puran‐ darpur
Tapaipur
Total All GP
<50 48.9 70.8 36.4 54.3 52.75 m 56.5 53.6 54.9 64.7 40 50 .1 22.23 10 150m 4. 14.9 4.2 18.2 10.9 14.58 150‐300m 4.3 7.1 5.9 6.7 4.2 6.96 300‐500m 0 3.6 0 13.3 0 0 Total 100 0 1
‐ 100m 34.8 7.1 19.6 11.8 26.7 21.3 20.8 31.8 260‐ 3 28.6 17.6 17.6 13.3
5.9 2
6.4 8.5
13.6 8.70 3.48
100.00 100 100 100 100 10 00 100 100 Source: Field Surv 6
l at
ved at more than per nt o e household in th samp to wait m tha 0 minutes fo wate from utsid the remi
ater e ing e lle t o to than in RLPV of a GP taking all type of sources into
onsideration. Lower percentage of households has waited ‘more than half an our’ in RLPV than in MPV in all the GP (Table A.3 in Appendix). The time spent
to collect water depends on the availability time of water supply and number of
s
ey 200
4.3. Waiting time to co lect w er
It has been obser th 30 ce f th s e lehave for ore n 3 r r ‘o e p ses public w supply source’. Th wait tim to co ct wa er is f und be generally larger in MPVch
households per source. The waiting time is in general larger in case of tap water upply compared to other sources. Amongst the households collecting water
145
from stand‐post, the waiting time for more than 60 per cent of them is ‘more than
ievances redressed is much higher in MPV than in RLPV. On the . The type of
roblems er, redressal of grievances would also depend on the type of problem.
Table 5 of Households Reported Wa p oblems c b e
I za Pou rpur
half an hour’. Larger waiting time for stand‐post may be attributable to limited availability time of water supply. For most of the households in Baruipur availability time is 3‐5 hours/day; in Surul it is 1‐3 hours/day, and in Purandarpur it is 3‐5 hours/ day. Due to lesser number of households per source in Ruppur GP, the percentage of households who ‘do not have to wait’ is highest.
5.0 Redressal of Grievances
The households report to the local government about the problems of water supply service. During last one year, 44 per cent of the households have reported any problem regarding water supply to the GP. Among these reported problems, 39 per cent have been redressed (Table 5). In Ruppur and Pourandarpur GP, the percentage of grother hand, it is higher in RLPV than in MPV in Illambazar GPp reported differs from one village to another. Moreov
: PerPercentageentage
ter Su ply Pr and of Pro lem R dressed
llamba r Ruppur randa
Ilamb zar ul Purandarpur l P
a Baruipur Total Sur Ruppur Total Tapaipur TotaAll G
R .4 55. 53.1 6.7 .2 5 .1 6 eported 51 2 2 45 36.1 51. 32 43 44.2
R 2.2 41.2 31.4 6 .9 25 44 6 edressed 2 62.5 28. 40.9 52 38.5
Source: Fie ve
he types of water problem reported have been classified in three different ty of water from the source or the source
requires repairing, (b) quality of water available is dissatisfactory and (c) lack of
ld Sur y 2006 Tcategories. They are (a) non‐availabili
adequate water supply sources. The first and second types of problem arise due to lack of maintenance. The third type of problem occurs due to lack of adequate provision of water supply sources. In Illambazar GP, 80 per cent of the problems 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 water supply due to inadequate provision. In all the three GPs, households have approached the GP more frequently in RLPV than in MPV for setting‐up of new source. It is just the opposite in case of maintenance.
146
Table 6: Percentage Distribution of Households by Type of Water Supply Problem Reported in Gram Panchayat
Illambazar Ruppur Pourandarpur
Pro
l blem Ilambazar Baruipur Total Surul Ruppur Total Purandarpur Tapaipur Total
AlGP
NV 50 52.9 51.4 75 28.6 45.5 41.2 37.5 40 45.82
Q 0 0 0 0 0 10.00
L 8
Tot 0
38.9 17.6 28.6 0
11.1 29.4 20 25 71.4 54.5 58.8 62.5 60 44.1
al 100 100 100 100 100 100 100 100 100 10
Note : NV = Non‐availability of water from the source/ source requires repairing, Q = Quality of water available is dissatisfactory L = Lack of adequate water supply sources Source: Field Survey 2006
The efficiency of local government in redressal of grievances can be expressed as re
in much an the problems relating to quality (10%) nd lac e water sup (12%). Th e to lack of cap . cresponsib ur supply. ess
two blem al n th at n‐a bility wain t Ps p m ar non labilit wa an
redressed in 90 per cent of the cases in Pourandarpur GP, 80 l nt
water supply problem dressed as a percentage of problems reported. The problems regarding non‐availability of water and repairing have been redressed
higher percentage (68%) of cases th a k of adequat ply sources is may be duacity of th
le for inste GP to
allatioresolvn of new
e qua so
lity probces of
lemswater
Moreover, G Tim
P is noe taken to
t dire redr
tly
these pro s is so lo ger an th of no vaila of ter and repairing. With he G , the roble s reg ding ‐avai y of ter d repairing have been per cent of the cases in Ruppur and on y 50 per ce of the cases in Illambazar GP (Table 7). In addition, the problems regarding installation of new source have been redressed in 13 per cent of cases in Pourandarpur, 8 per cent of the cases in 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, literacy strengthens local democracy and increases efficiency of the GP.
147
Table 7: Percentage of Water Supply Problem Redressed by Type of Problem
Illambazar Ruppur
Pourandarpur
Problem azar r Total Surul Total Purandar Total All GP
Ilamb Baruipu Ruppur pur Tapaipur
NV 33.3 66.7 50 666.7 100 80 100 66.7 90 8.4 Q L
14.3 13.3
0 20
1014.3
50
20
0
10.0 11.8 0 0 8.3
Note: N ‐ bil wat from sour source requires repairing, Q ality ater ilab un cto
L ck of uate er ly s s urce: Field Survey
.0 R
eterminants of access to water supply and redressal of water problems by the GP has been analysed.
s robably increased the bargaining strength of the GP in the upper level of overnment for funds to invest in water supply. An individual may be illiterate ut would get greater access in Purandarpur due to higher collective bargaining trength of the village as well as the GP.
V = Non availa ity of er the ce/ = Qu of w ava le is satisfa ry
So
= La
adeq 2006
wat supp ource
6 egression Results
In this section, the dsupply
6.1 Access to water supply
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. In all the three models the significant variables are ‘source’ and village dummies of Surul, Ruppur and Purandarpur. The sign of ‘source’ is negative and the sign of village dummies are positive. All the models are well fitted.
The results imply that access to water supply increases with higher education level of the respondent. This is presumably because higher education increases bargaining strength of the people in the Gram Panchayat for setting up sources near their premises. It has also been found that if the source is stand‐post rather than any other source, then access to water supply declines. Moreover, access to 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 Purandarpur as compared to Ilambazar may be due to higher literacy rate of the village. Moreover, Pourandarpur GP also has a very high literacy rate, which hapgbs
148
Table 8: Maximum Likelihood Estimates of Determinants of Access to Water Supply
I IVariables Model Model II Model II
Educat0.04* (1.78)
ion
Caste (SC/ST=1. Others=0) ‐0.01
epri 40
(‐0.73)
hers = 0) 0.85*** (‐3.18)
‐0.85*** (‐3.17)
‐0.83*** (‐3.11)
= 0) 0.31 (1.49)
0.22 (1.10)
0.24 (1.18)
4) (0.91) (1.08) 1.81*** (4.33)
1.69*** (4.08)
1.72*** (4.19)
0.43
(‐0.03) ‐0.
D vation
‐Source (stand‐post = 1, ot
Meeting (attend =1, do not attend
Baruipur 0.44 (1.2
0.32 0.40
Surul
Ruppur 1.38*** (3.33)
1.21*** (2.95)
1.28*** (3.10)
Purandarpur 0.79** (2.12)
0.75** (1.97)
0.78** (2.08)
Tapaipur 0.49 (1.13)
0.24 (0.57)
0.33 (0.78)
Cutpoint 1 0.14 ‐0.24 ‐Cutpoint 2 1.41 1.01 0.83 Log likelihood ‐129.13 ‐130.74 ‐130.47 Chi square 55.92 52.71 53.25 Prob>chi2 0.00 0.00 0.00 Number of observations 144 144 144
Note: Numbers in the parenthesis are z statistics * Significant at 10% Level, ** Significant at 5% Level and *** Significant at 1% Level 6.2 Redressal of water supply problem
The maximum likelihood estimates of determinants of redressal of grievance have been illustrated in Table 9. The estimates reveal that in all three models the variables ‘problem’ and village dummy of Purandarpur are significant. The coefficients of these two variables are positive.
149
Table 9: Maximum Likelihood Estimates of Determinants of Redressal of ter Supply Problem
Mo M
Grievance Regarding Wa
Variables del I odel II Model III
Education ‐0.0(‐0.8 1#
6)
Caste (SC/ST=1. Others=0) 0(0
7
0) 0.(0. (
Problem (Non‐availability of water from
others=0)
0.60*(4.8
0.6(4
0
0.15(0.7
0.(1
0.(1.6
0(1
0.(0.7
0(0
0.47(2.2
0.(2
‐0.0(‐0.0
0(0
d ‐34.9 ‐35 ‐38.9 380.
bservations 82
.08
.64)
Deprivation .72e‐06(0.00)
Meeting (attend =1, do not attend = 03 30)
0.04 0.30)
0.03 (0.25)
source/ source requires repairing = 1, ** 4)
0*** .86)
.63***(5.09)
Baruipur 1)
22 .14)
0.23 (1.10)
Surul 37 1)
.34
.46) 0.35 (1.51)
Ruppur 17 1)
.20
.87) 0.25 (1.03)
Purandarpur ** 9)
47** .26) .
0.51** (2.47)
Tapaipur 1 4)
07 28) .
0.25 0.85) (
Log likelihoo 1 .09 34.31 Chi square 2 .57 39.17 Prob>chi2 00 0.00 0.00 Number of o 82 82 Note: Numb # Mar
ers in the parenthesis are z statistics the coefficients vel, ** Significant at 5% Lev *** Signific 1% Level
he r s by at lability of water from sources or
sources requireing repairing. It has also been observed that probability of
the GP.
ginal effect, not * Significant at 10% Le el and ant at
T esults illustrate that probability of redressal of grievance increaseleast 60 per cent if the problem is non‐avai
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 due to much efficient functioning of GP in Pourandarpur for redressal of water supply problems as the literacy rate is highest in
150
7.0 Summary and Conclusion
water supply has been measured by distance ofAccess to the source and waiting collect water. Distance of ‘out premi ic wa ly
the residence of the househo gher in n in is ue to higher bargaining capab the reside s of MPV than the
GP. Waiting time to collect wat nerally higher in GP. It may be due limited av lity time of water
stand‐post, which is mainly prevalent in MPV Within th he households per source is lowest in Ruppur GP. This may to
nt with of for aid. Acce also has ghest lit te amo the rand Village as Pour ur
aining s of the v in the up vel nt for funds to invest in wat pply. In v inter‐household
water supply sources in the villages, access ter supp the hold has also increased with educ level. In utshell, t udy
that education has impact th inter ra‐juri nal
s of access and quality of w upply ar rted to am Panchayat for redressal. It has been obs that aro 4 per ce the
ds have approached GP for redressal of grie regardi ater supply. Among these, 39 per cent of th ances ha n redre e
orted to the GP are mostly d to the f adequ ter rces in RLPV and regarding m ance in M It has be nd r cent of the grievances related on‐availab of water the
uireing repairing ha been redressed. The redressals of antially less. This may be since the GP is
ore first type of problem as compared to other ificantly
igher in Purandarpur Village, presumably due to high literacy rate of the GP.
ity in access where ocio‐economic inequity is well pervasive. Central intervention in planning and
time to source’ from
side thei
ses publRLPV tha
ter suppMPV. Thld is h
may be d ility of ntRLPV for resources in theMPV than in RLPV of a
er is ge to ailabi
supply fromnumber of
.
e GPs, tbe due
higher per capita capital investme the hithe help eign ss is
higher in Purandarpur Village, whichvillages. The high literacy rates of PuGP have probably increased barg
eracy ra ng all arpur as well andarptrength illage per le
of governmeaccess to
er su iew ofto wa ly of
housedemonstrates
ation a n he st on bo ‐and int sdictio
allocation.
The problem ater s e repo the Grerved und 4 nt of
househol vance nssed. Thg w
e griev ve beegrievances rep relate lack o ate wasupply sou ainten PV. en fouthat 68 pe to n ility fromsources or sources req ve other types of problems have been substm directly responsible to solve theproblems. The efficiency of GPs in redressal has been observed to be signh
The study reveals that access to water supply services and efficiency of the local government in redressal of grievances regarding water supply have significant relation with education level of the communities. Moreover, number of sources per household is higher in the most prosperous villages. The prevalence of socio‐economic 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 dispars
151
allocation at the village level would probably reduce the inequity. This is evident
upply Sanitation Project, Tribal (Indigenous People) Development Plan, No.
,
from higher provision of water supply sources in a relatively less prosperous village of Pourandarpur GP. Furthermore, external aid could also be supportive for provision in case of paucity of government grants.
References
Asthana, Anand N. 2003. Decentralisation and Supply Efficiency: The Case of Rural Water Supply in Central India. The Journal of Development Studies, Vol. 39, No. 4, Pp. 148‐159.
Bardhan, Pranab. 2002. Decentralization of Governance and Development. Journal of Economic Perspectives, 16(4), 185‐206.
Bardhan, Pranab, and Dilip Mookherjee. 2000. Capture and Governnence at Local and National Levels. American Economic Review, 90(2), 135‐139.
Besley, Tim, Rohini Pande, Lupin Rahman and Vijayendra Rao. 2005. The Politics of Public Good Provision: Evidence from Indian Local Government, Journal of the European Economic Association, Volume 2‐23, 2004.
GOI. 2008. Annual Report 2007‐08. Ministry of Rural Development GOM, Water Supply and Sanitation Department. 1998. India‐ Second Maharastra Rural
Water SIPP47.
Makumbe, J. 1996. Participatory Development: The Case of Zimbabwe. Harare: University of Zimbabwe Publications.
Mtisi, Sobona and Alan Nicol. 2003. Water Points and Water Policies: Decentralisation and Community Management in Sangwe Communal Area, Zimbabwe. Sustainable Livelihood in Southern Africa, Research Paper 15.
Oates, Wallace. 1972. Fiscal Federalism. New York: Harcourt Brace Jovanovich. Oates, Wallace. 1977. An Economist’s perspective on fiscal federalism. In The Political
Economy of Fiscal Federalism, ed. Oates Lexington: Lexington Books. Slaymaker, Tom and Peter Newborne. 2004. Implication of Water Supply & Sanitation
Programmes under PRSPs, ODI, Water Policy Programme. Subrahmanyam, K. Siva and R. C. Choudhury. 2002. Functional and Financial Devolution on
Panchayats in India. National Institute of Rural Development. Tiebout, C. 1956. A pure theory of local expenditures. Journal of Political Economy, Vol. 64
P. 416‐424. World Bank. 2004. World Development Report 2004: Making Services Work for Poor People. WSP. 2003. Governance and Financing of Water Supply and Sanitation in Ethiopia, Kenya
and South Africa: A Cross Country Synthesis, Sector Finance Working Papers: No. 5, September.
WSP. 2004. Rural Water Supply and Sanitation in Africa: Global Learning Process on Scaling Up Poverty Reduction. Shanghai Conference, May 25‐27.
152
Appendix Table A1: Size of Sample Drawn from Different Villages
GP GP GP Illambazar Ruppur Pourandarpur Total
Ilambazar 35 35 Baruipur 29 29
Rup
Tap 25 25
Surul 30 30 pur 31 31
Purandarpur 33 33 aipur
Total 64 61 58 183 So
Tab
urce: Field Survey 2006
le A2: Percentage of Literate, Scheduled Caste (SC) and Scheduled Tribe
(ST) in Gram Panchayats
Ruppur GP Pourandarpur GP Illambazar Lite 55.72 80.20 55.00 rate 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 0
Lite 58.39
ST 21.24 73.45 0.00
Literate 61.38 81.40 62.9SC 27.53 48.27 33.44
10.67 6.24 ST 6.18SC & ST 38.20 54.51 39.62
Ruppur Tapaipur Baruipur rate 46.03 55.00
SC 21.00 10.47 13.36
SC & ST 42.23 83.91 13.36 Source: Bolpur‐Sriniketan, Suri‐II and Illambazar Panchayat Samiti
153
Table A3: Percentage Distribution of Households by Waiting Timr Supply Source’
Illambazar Ruppur Pourandarpur
e to Collect Water from ‘Outside the Premises Public Wate
Ilambazar Tota
Ruppur
Baruipur l Surul‐Total
Puran‐darpur
TapaipurAll GP
Total
A 0 3.3 11.1 0 4.32 8.3 11.1 0 B 0 10 1.1 11.1 14.3 1 10.87 C 11.1 3 10 1 11.1 0 8.69 16.7 16.7 2 22.2 0 15.23
44.4 40 11 28.6 32.61 27.8 3 20 3.3 33. 57.1 5 28.25
Stpo
Total 100 100 0 100 100 100 0 14.3 16.7 28.6 26.5 0 0 0 19.32
25 1 4.3 8. 11. 0
Dand‐
16.7 22. 0 E
st 33.3 11.1 .1 28.6
F 8. 3 3 7.1 100 10 100
A 50 B 0 10 7.1 33.3 14.3 17.6 42.9 50 44.4 19.25
31 5
E 0 0 0 0 0 0 1.74 F 25 40 35.7 17.9 14.7 0 0 0 17.54
DTwell
Total 100 100 0 10 100 0 20 20 0 0 0 10 7.92
C 0 10 7.1 50 35.7 38.2 57.1 50 55.6 33.D 0 40 28.6 0 3.6 2.9 0 0 0 8.7
25 0 7.1
eep ube‐
0 100 100 100 100 10 0 100
A 0 6.7B 0 0 0 0 50 20 35 26.31 C 0 40 40 100 50 50 35 42.11 0 20 20 0 0 10 10 10.53
0 0 0 0 0 10 5 5.29 F 0 2 20 0 10 5 7.92
Hapu
tal 0 10 100 100 100 100 100 Mana ell
A 100 0 100
33.3 30 66.7 40
DE
mp
0 0
106.7
0 0 0 6.7
nd‐
To 0 100 100 100so‐ ry W
100
C 0 10 50 50 0 F 100 50 Other
tal 100 1 100 8.7 7.1 7.8 17.6 26.7 0 9.1 11.8
0 500
ToA
0 100 23.4 4.3
B 0 14 17.6 16. 20.8 36.4 17.4 C 8.7 17 29.4 36. 41.7 36.4 1 28.4
13 25 11.8 3.3 4.2 9.1 11.1 12.5 4.5 8.7 12.5
F 30.4 21.4 25.5 17.6 16.7 17 20.8 4.5 13 18.7
AllSo
Total 100 100 100 100 100 100 100 100 100 100
.3 7.8 7 17 28.3
.9 13.7 7 34 39.D
19.6 6.4 6.5
E 39.1 14.3 25.5 5.9 0 2.1urces
Note: A= Do not have to wait, B= Less than 5 min s, C= 5‐15 minutes, D= 15‐30 minutes, E= 30 minutes‐ 1 hour, F= More than 1 hour
Source: Field Survey 2006
ute
154
Asia‐Pacific Journal of Rural Development Vol. XIX, No. 1, July 2009
Carp Seed Transport in Bangladesh
Mahmud Hasan*
Abstract
The aim of this study was to explore the current practices of carp seed transport methods, 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 semi‐structured surveys. This would provide the basis for further development of carp seed transport methods and trading systems in the country.
1.0 Introduction
Fish fry/fingerling traders use their ‘Indigenous Technical Knowledge’ (ITK) to transport carp seed between and within regions of the country to the fish farmers by land transport. Traders load carp seed as much as they can per unit volume of water. They use different transport mediums to transport fish seed, for example, pond, well and a mixture of the two. Typically the traders agitate the transport mediums during transport by hand for oxygenation. Time required for selling carp seed in the markets depends on the availability of the traders.
Hand agitation, means of transport, transport mediums, loading density and duration and distance cover to transport carp seed vary between regions within the country. The transport techniques and measures taken by the traders for agitation and conditioning of the fry/fingerlings are also different between regions. These different handling techniques may lead to physical injury, water quality and health problems mediating stress and high mortality. Hasan and Bart (2006) estimated the loss (6200 mt) resulting from transport mortality.
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. The effects of vigorous agitation on the physical injury such as scale loss, injury and fin bifurcation has been demonstrated by Ahmad (1954). Nabi et al. (1983) has described the effects of transport medium on the mortality of
* Associate Professor, Department of Fisheries, University of Dhaka, Bangladesh. E‐mail:
155
mixed carp fry. He found that while river water resulted in 24.00 ± 5.96 per cent 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 (0.25 g) using well water and 15 per cent using pond water 3 h after closed transport with compressed oxygen. However, information relating to carp seed transport by the small‐scale fry/fingerling traders is lacking. Thus there is a need to explore the baseline information of carp seed transport related activities to determine the problems associated with this part of aquaculture in Bangladesh. However, baseline information across the country related to carp fingerling transport has never been reported.
The overall objective of this study was to review the current carp seed transport method in Bangladesh. The study also attempted to identify the current loading density, transport mediums used to transport fish seed, transport duration and distance, and transport mortality; to determine which fish is the most sensitive to transport mortality as well as some of the common practices being used to reduce stress and improve survival during and after transport.
2.0 Study Method
Purposive sampling technique was used to collect data using semi‐structured survey and structured HH survey as the primary tools. This study was undertaken to identify the major issues relating to carp seed transport situation in Bangladesh. Some 13 carp seed markets were surveyed based on the intensity of fish seed trading activities. The survey covered Jessore, Zhenaidaha, Natore, Bogra, Dinajpur, Rajshahi, Mymensingh, Laxmipur and Comilla districts. Carp seed traders, nursers and fish farmers were sampled.
2.1 Land classification
The sampled HH were classified into six land classes following IFAD (2002) to denote their socio‐economic status in which the basis was land. Land owned (ha) = cultivable land + pond/gher + fallow land excluding the homestead; total land owned (ha) = land owned + homestead land. The land classes are: absolute landless = no cultivable land (excluding homestead); landless = land owned 0‐≤ 0.2; marginal = land owned >0.2‐≤ 0.6; small = >0.6‐≤ 1.0; medium = >1.0‐≤ 2.0; and large = >2.0 ha.
156
2.2 Data analysis
Data generated through semi‐structured survey were manually analysed while structured HH survey data analysed using statistical software SPSS version 10.0. Only the descriptive statistics was used to present the data as weighted mean.
3.0 Findings of Surveys
3.1 Carp seed trading
Although carp seed include several stages of life history, here in the present study only the term ‘fingerling’ is used to mean fish seed. Carp seed trading is a large and an important component of aquaculture industry in Bangladesh since farming practices initiate from this input. Usually the large hatchery owners produce hatchlings upon getting order from the fingerling (seed) growers. Small hatchery owners produce both fry and fingerlings. The nursers raise fingerling in earthen pond and sell to the seed traders via middlemen in the fish seed markets. The middlemen transport the carp seed over 0.5‐2 h and confine into hapa in holding pond water before sale. The traders transport and distribute carp seed across the country to the end users.
Indian major carps e.g., catla, rohu and mrigal, Cirrhinus cirrhosus were found to be in highest demand with 85 per cent of the respondents of food fish farmers. This could be due to people’s familiarity with the indigenous fishes. The exotic carps e.g., silver carp, grass carp, Ctenopharyngodon idella, common carp, Cyprinus carpio and silver barb, Puntinus gonionotus had lower demand.
3.2 Carp seed transport
Fertilised eggs, fry and fingerling are usually transported and distributed across 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 (20‐60 L) and metal/plastic drum (200 L) by hand agitation (water splashing). This traditional transport of fingerling often leads to total/mass mortality. This increased the traders risk leading to the loss of capital 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 seed to be 6200 mt in Bangladesh.
Aluminium transport vessels are traditionally called ‘Patil’ or ‘Hari’ (Lewis et al., 1996). Large Patils (40‐60 L) are usually used for long haul and small Patils (20‐30 L) for short hauls. Three different distribution channels were obvious in this
157
survey (Figure 1). In the first channel, the hatchery produced hatchlings are sold to the fingerling growers. The growers again sell to the seed traders who finally sell to the grow‐out farmers. The second channel includes production of fry and fingerlings at the hatchery which is sold to the farmers or traders without involving market. In the third channel, the fry are sold to the nursers/fingerling growers. They intern grow them to fingerling size and sell to the traders in the carp seed markets. In this channel there is no farm gate selling. Figure 1: Carp Seed Distribution Channels Surveyed in Carp Seed Markets of
Bangladesh Fingerling growers Farmgate selling Traders/farmers
Middlemen Traders/farmers Nursers
Market selling Traders/farmers
Hatchery
Market
Nursery
Grow-out sites
3.3 Types of traders
Carp seed traders can be classified into two types based on the quantity of seed they transport, i) ‘Mohajon’ or large‐scale traders, they transport >30 kg seed with the help of hired day labour; and ii) ‘Paikers or Patil oala’ are small‐scale traders. The paikers carry one pair of patils hung from a bamboo stick with the help of a gear made of jute fibres on the trader’s shoulders. Mohajans sell carp seed as far as 500 km from the source requiring as much as 10‐12 h, for example, from Bogra to Lalmonirhat or from Bogra or Natore District through Parbotipur railway station to Ulipur Sub‐district of Kurigram District takes approximately 15 h. The number of intermediaries tended to increase with increasing market channels for a single haul.
158
3.4 Carp seed markets Carp seed trading is an old enterprise as some traders were found trading carp seed over last 40 years. In the past, they transported seed from the wild stock to 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. After surveying more than 20 districts and sub‐districts as well as a 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 were found in peri‐urban Jessore town alone. Some other carp seed markets are Borobazar and Kaliganj in Zhenaidaha District mostly based on a big hatchery‐cum‐nursery pond.
Carp seed markets can be classified into three groups depending on the sources of seed sold in the markets: (i) primary market: where carp seed are sold after collection from many different farms; (ii) secondary market, where the carp seed are sold upon bringing out from other markets by intermediaries. Mohajons obtain carp seed mainly from the primary market and then haul to the secondary 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 subdistricts of Kurigram District.
Selling of carp seed in an organised market was found to have several advantages. For example, traders/farmers can buy target species of required quantity and competitive prices, greater options for the buyers to select desired species and quantity, guarantee in daily supply, less travel time to buy seed and money, easy entrance, higher chances of getting assistances from fellow traders during transport, less chance of loosing cash, and higher chances of getting compensation from the middlemen. However, late comers may not get the seed of desired quantity and species. Middlemen and their assistants (day labourer) tend to give less (both in weight and number). The disadvantages of these handling and transport related activities increase the stress and result mortality.
3.5 Carp seed trading in Jessore area
Middlemen typically lease in ponds on a yearly basis with advance payment. Roadside ponds are expensive and high in demand because of ease movement. Pond owners themselves may even be the middlemen. Short haul paikers/traders buy and transport seed covering distance 20‐150 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 traders initially transport carp seed from Jessore to their own area. They confine the seed into hapa in pond water. After one or two day of stocking, they transport these seed long distance. These traders also hire ponds in those areas to hold seed prior to selling. Usually, the traders sell fish seed within a day. In most cases they are booked early and if seeds are not completely sold, they hold them in the pond water until selling.
People holding carp seed in the ponds along the selling area are called middlemen or ‘commission agents’. They obtain 10 per cent commission on the amount sold. They contact the fingerling growers on their own. Middlemen have 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. Middlemen with the permission of the nursers sometimes, compensate in cash or in kind if the traders claim for withdrawal because of poor quality seed.
3.5.1 Water supply
In Jessore carp seed market, 10 tube‐wells and one shallow tube‐well were found to supply clean subsurface water. More than 120 women were found selling well water to the traders. Each earns nearly 2 US$ a day during the peak season (May to July) and 1 US$ during the lean season. Water retailers charge 20 taka for filling a 200 L drum and 5‐8 taka for an aluminium vessel (1 US$ = 60 Taka). Usually, the traders transfer carp seed with small amount of pond water from the pond to the transport container.
3.6 Confinement
Carp fingerlings are held in hapa in pond water before selling usually for 2‐6 hours. Middlemen confine 100‐200 kg fingerling into a small earthen pond. The confinement density varies with species and size. Common carp fingerlings are confined in high densities while silver barb and silver crap are confined at low density. Seed are confined in hapa into pond and hatchery effluent water in Jessore area, into pond water at Adomdighi in Bogra, Rajshahi and Durgapur in Comilla, and into drum with well water at Parbotipur railway station of Dinajpur 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. Silver carp and rohu were found as the most hyperactive fish. The fingerlings are released back into holding pond after transport from the nursery to the seed market. The next morning, the same seed are seined and held in hapa in pond.
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 morning until sale. Confinement density varies between systems. Typical drum confinement with well water is 125 g L‐1 whereas rohu and catla are confined at 20 kg m‐2 and silver carp at 18 kg m‐2 in hapa in pond. Confinement mortality usually varies between 5 and 20 per cent depending on the quality of water, density, duration, size and species.
3.7 Carp seed trading in Northwest (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 sub‐districts of Bogra District followed by Rajshahi District (NFEP 20). Trading of carp seed usually starts from late February and continues until July particularly at Adomdighi as the nursers produce only the first crop. On the other hand at Kahalu, carp seed are traded across the seasons starting from February to September. Over‐wintered 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, Adomdighi has become the main source of catfish seed supply.
The seed market at Parbotipur railway station is equipped with the supply of underground water form the local municipality. There is also an organising office for the local traders. Usually carp seed from Natore District and Santahar railway station of Adomdighi Sub‐district of Bogra District are transported to Parbotipur railway station. Some seed are sold to the small‐scale traders and most are sold and/or further transported to Rangpur, Dinajpur, Lalmonirhat, Nilfamari and Kurigram districts. Kakina Bazaar in Hatibandha sub‐district of Lalmonirhat District has become popular for carp seed trading since 2002. Mohazons (large‐scale traders) transport seed using plastic drums mostly from Bogra and Natore districts to the local small‐scale traders. Some small‐scale traders also transport carp seed from Bogra and Natore districts to Kakina bazaar. Unlike Jessore there are no middlemen in Kakina carp seed market.
161
3.8 Carp seed trading in Mymensingh region
Most carp seed in Mymensingh are sold at the farm gate. The traders transport 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, Ishwarganj and Gaforgaon railway stations. However, unreliability of train was found to be a big constraint if any derailment that frequently occurs, disrupt the channels and income of the traders.
3.9 Transport mediums and water exchange
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 followed by well water (37%) while other 18 per cent used pond water.
Most traders change water during transport if the water becomes turbid. The type of transport water to be changed depends on the availability. However, the usual water change interval varies from 2‐3 h or 3‐4 h. The transport of silver barb fry need frequent change, for example, 1‐2 h interval because they are more sensitive to transport stress. Changing of transport water also depends on the transport means. For example, if seed are transported by train, it may not always be possible to change water when needed.
3.10 Transport distance
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 50‐150 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 farmers’ pond by on foot, bus, rickshaw van or bicycle. Most traders (29%) were found to cover distance of nearly 50 km and 16 per cent covered 50‐100 km (Table 1). However, distances 150‐200 and 200‐250 km are covered by 13 per cent of traders.
162
Table 1: Intervals of Distances (km) Covered by Carp Seed Traders during Transport Sampled in 11 Districts of Bangladesh
Interval of Distance (km) Traders’ Percentage
up to 50 29.4 50‐ 100 15.7 100‐150 13.3 150‐200 13.3 200‐250 9.3 250‐300 7.4 300‐350 3.1 350‐400 4.9
Above 400 3.6 Total 100.0
3.11 Transport duration and loading density
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 depend on the prior order. If the traders do not have specific order before hand, they peddle around to sell to the food‐fish farmers.
Traders load as high as 500 g L‐1 for short distance (0.5‐1 h) while below 200 g L‐1 for short hauls (Table 2).
Table 2: Loading Density of Carp Fry/Fingerlings L‐1 by Size in Traditional Open Method by Hand Agitation Surveyed in Bangladesh
Size of fish Loading density (g L‐1) Respondents’ percentage
Small <1 ‘ 1 to 200 9 Fingerling>1’ and <3’ 2 to 300 29 ’’ 3 to 400 24 ’’ 4 to 500 11 ’’ 500 27
3.12 Transport mortality
Immediate mortality of carp seed resulting from handling and transport vary widely depending on species, size, duration and distance, condition of the seed
163
and transport mediums. The immediate mortality usually varies from 5‐25 per cent. Occasionally, even the whole batch can die. Sometimes the food‐fish growers claim for compensation which could be due to delayed mortality. However, the reason for lower recapture is unknown. Hasan and Bart (2007) have demonstrated that transport may cause even 50 per cent delayed mortality.
Common carp was found to be the hardiest fish among carps during confinement and transport in the present survey. Although transport mortality is common, some fishes are more sensitive than others. However, according to the traders, silver carp was the most sensitive to transport mortality followed by rohu, mrigal, catla, Catla catla and silver barb, Puntius gonionotus (Table 3). Silver carp, rohu, mrigal, catla and silver barb are more hyperactive than other fishes. As a result, these fishes are susceptible to diseases and sensitive to mortality.
Table 3: Fish Mortality by Species according to Carp Seed Traders Surveyed in Carp Seed Markets in Bangladesh
Species Traders’ percentage Silver carp 50
Rohu 29
Mrigal 11
Others (Rajputi/Catla) 10
Total 100 3.13 Duration of carp seed trading (days, months and years)
Most traders (87%) transport carp seed for 5‐7 days a week and some (11%) transport for 5‐6 days. Similarly, most traders (49%) transport carp seed for 3‐6 months, one‐quarter transport for 6‐9 months and some traders (18%) transport for 9‐12 months a year. The duration of trading was found to vary among the traders. Nearly one‐third traders traded carp seed during the previous 5‐10 years and 21 per cent for 10‐15 years (Table 4). One‐fifth traders were found trading carp seed for the previous 5‐10 years and 14 per cent for 15‐20 years. Traders trading carp seed in excess of 15‐20 years were few.
164
Table 4: Distribution (%) of Carp Seed Traders by Years of Trading Carp Seed Surveyed in 11 Districts of Bangladesh
Duration (y) Traders’ percentage 1‐5 19.6 5‐10 33.3 10‐15 21.0 15‐20 13.5 20‐25 5.5 25‐30 4.1 30‐35 2.0 35‐40 0.9 Total 100.0
3.14 Quantity of seed transport
The quantity seed transported by the traders was found to vary widely (Table 5 & 6). Nearly 41 per cent traders transported 5‐10 kg seed a day while 31 per cent transported only 5 kg. Further 14 per cent traders were found to transport 10‐15 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 highest quantity (18 kg) while large land class HH transported the lowest (8 kg).
Table 5: Amount (kg) of Carp Seed/Trader/Day Transported by Traders Sampled in 11 Districts in Bangladesh
Quantity (kg) Traders’ percentage
up to 5 31.1 5‐10 41.3 10‐15 12.6 15‐20 1.4
above 20 13.7 Total 100.0
3.15 Reasons for trading carp seed as a profession The reason for adoption of carp seed trading as a profession was also explored. Most traders (97%) responded that they trade carp seed because of higher income compared to daily labour wage. Further 37 per cent trade seed because they can earn enough to support their families.
165
Table 6: Quantity (kg)/Day/Trader of Carp Seed Transported by District and Land Class of Carp Seed Traders Sampled in 11 Districts in Bangladesh
Land class
District Absolutel landless Landless Marginal Small Medium Large
Mean
Chuadanga 12.31 21.50 13.00 0.00 0.00 0.00 15.11 Norail 10.67 11.75 16.67 0 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 Use of additives during transport
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 District were adding something in the water. Interestingly, sampled traders of all 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, vitamin C (ascorbic acid) and B2 (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. The average dose of salt per vessel (50‐60 L) was nearly 10 g while the number of pack of ORS was 1.28. The quantity of antibiotic tablet, for example, renamycin per vessel (50‐60 L) varied from 1 to 5. The highest quantity of antibiotic (4‐5 tablets per 50‐60 L vessel) was used by the traders of Khulna District.
166
3.17 Problems/constraints encountered by carp seed traders
Almost all traders reported immediate transport mortality and compensation due to delayed mortality claimed by the food‐fish farmers as problems to carp seed trading. Credit selling was found as another constraint since recovery is always less and sometimes no recovery at all. Most traders (85%) reported that they lack capital which is the main constraints in transporting fish seed. Further 67 per cent traders complained lack of good quality seed and 57 per cent claimed high mortality. Nearly half the traders (44%) think poor transport technique is one of the main problems for seed trading. Nearly 35 per cent traders indicated that they lack basic training skills.
3.18 Policy implications
3.18.1 Development aspects: Market development
There is a need for new carp seed markets to reduce the transport distance, transformation of the current earthen holding ponds into flow‐through‐tank 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 replacing the current earthen holding ponds into flow‐through‐tank system.
3.18.2 Technical aspects: Problems need to be resolved
Negative impact resulting from handling and transport related activities on the quality deterioration of fish seed, causes of immediate and delayed mortality, effects of stress resulting from handling and transport related activities on the survival, stress responses, disease susceptibility and growth of fish and technical solution to reduce mortality as management strategy.
167
References Ahmad, N. 1954. Mortality of fish during transport and cultural operation: Causes and
prevention. Pakistan Journal of Science, 6(2): 1‐5. Chowdhury, A. Z. M. N. I. 1997. Effect of prolonged transportation of carp fingerlings on its
growth and survival in cultural pond. M.Sc. Thesis. Bangladesh Agricultural University, Mymensingh.
Hasan, M. and A. N. Bart. 2006. Carp seed traders in Bangladesh: sources of livelihoods and vulnerability resulting from fish seed mortality. Asia‐Pacific Journal of Rural Development, 16(2): 97‐120.
Hasan, M. and A. N. Bart. 2007. Effects of capture, loading density and transport stresson the mortality, physiological responses, bacterial density and growth of rohu Labeo rohita fingerlings. Fish Physiology and Biochemistry, 33(3):241–248.
IFAD. 2002. Assessment of Rural Poverty: Asia and the Pacific. P. 30. Lewis, D. J., G. D. Wood and R. Gregory. 1996. Trading the silver seed. Dhaka: University
Press limited. P. 199. Nabi, R., M. A. Hossain and S. Rahman. 1983. On the use of tube‐well and pond water for
transporting carp fry. University Journal of Zoology Rajshahi University, 2: 25‐28. NFEP (Northwest Fisheries Extension Project), Second Phase, Paper number 20. Fry and
fingerling trade at Parbotipur railway station, 1989, and 1992 to 1996. Sampson, D. R. T. and D. J. Macintosh. 1986. Transportation of live carp fry in sealed polythene
bags. Aquaculture, 54: 123-127.
168
Asia‐Pacific Journal of Rural Development NOTE TO CONTRIBUTORS
1.0 Aims
The Asia‐Pacific Journal of Rural Development (APJORD) is an interdisciplinary journal devoted to issues and discussions on rural development primarily in the Asia‐Pacific region. APJORD focuses on poverty issues and rural transformation, keeping in view the four programme priorities of CIRDAP, namely (1) Agrarian Development; (2) Institutional/ Infrastructural Development; (3) Resource Development including Human Resources; and (4) Employment.
2.0 Submission of Articles
Articles are accepted for publication provided they are submitted to APJORD only. If there is more than one author, a statement to the effect that the article has 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 editorʹs decision regarding articles submitted for publication is final. Copyright rests with CIRDAP.
Authors are entitled to three complimentary copies of the issue in which their articles appear. Articles written between 4,000‐5,000 words are preferred. Both the hard copy and soft copy of the write‐up must have to be submitted to the editor. The soft copy either may be given in floppy diskette/pen drive/CD or can be sent as an attachment to the e‐mail address: [email protected].
Typescripts should be submitted in duplicate, on A4 size paper, typewritten on one side, in double space, with margins of at least 2.54 cm. (one inch). Statistical tables, illustrations and charts should be submitted on separate sheets, and their positions should be indicated in the text.
3.0 Style and Presentation
APJORD mostly follows the Chicago Manual of Style for preparing articles. Authors should, as far as possible, adhere to the following basic requirements of style:
i. The name and institutional affiliation of the author should appear as a footnote on the first page together with acknowledgement, if any.
ii. An abstract of approximately 100 words should precede the text on the first page. It should be titled ‘Abstract’ indented, italicised and centred.
iii. British and ‘s’ spellings should be used. Numbers from zero to ten should be spelled out, and numerals should be used for all other numbers.
iv. Where references are made without any comment, they should be inserted at 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; Razia 1999: 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 appearing as footnotes or listed sequentially at the end of the article.
vi. The reference list should appear after the Notes. For books give author’s surname, other name(s) and/initial(s), year of publication, title of publication, place of publication, and publisher, e.g.
Syed, Shireen Kamal. 1992. Women in Post‐harvest Loss Prevention. Dhaka: Centre on Integrated Rural Development for Asia and the Pacific.
Tamanga, A., N Shrestha., K. Sharma, and S. Kathi. 1994. Case Studies of Women with Abortion Experiences. Kathmandu: IDS.
Acharya, Meena. 1994. Political Participation of Women in Nepal. In Nelson, Barbara and Chowdhury, and Norma edited Women and Politics Worldwide. London: Yale University Press.
For Journal articles, give the author’s 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? Asia‐Pacific Journal of Rural Development, Vol. II, No. 2, December, Pp.1‐14.
171
ASIA-PACIFIC JOURNAL OF RURAL DEVELOPMENT ISSN 1018-5291
O R D E R F O R M
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: [email protected]
Visit our website at
http://www.cirdap.org.sg
172
170 170
171