Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Changes in Agriculture and the Economy in the Flood-prone Environment in Bangladesh, 1988 to 2000: Insights from a Repeat Survey of 16 Villages

M. Hossain
Head of Division
Social Science Division, IRRI, Los Banos, Philippines

M.L. Bose
Consultant
Social Science Division, IRRI, Los Banos, Philippines

A. Chowdhury
Director
Socio-consult, Bangladesh

Abstract/ Executive Summary:

Semi dwarf modern rice varieties are not suitable for the flood-prone environment. So, a general perception exists that the Green Revolution bypassed this ecosystem, which constitutes about 40% of the rice land in Bangladesh. This paper assesses the changes in agriculture and the economy in the flood-prone ecosystem by analyzing household-level data collected through a sample survey of villages conducted in 1988 and 2000. The findings show that, although farmers still grow mostly traditional varieties during the wet sea- son, the coverage of modern varieties during the dry season when flooding is not a problem has expanded from 43% in 1988 to 80% during 2000. In a large proportion of areas, farmers have abandoned the low-yielding, high-risk traditional rice varieties and shifted to high-yielding modern varieties during the dry season by investing in irrigation through installing shallow tube wells and power pumps. These changes in the cropping pattern have reduced crop. ping intensity from 174% to 143% of cultivated land, but increased the aver- age rice yield by 66% over the 12-year period. Technological progress and the fast increase in rice supply have contributed to a reduction in the unit cost of rice production, an increase in labor productivity, and a much slower increase in rice prices compared to the general cost of living index. These factors have benefited the rural and urban poor more than farm households. The share of rice farming in rural household income is small and declining, however, because of the small average holding size, which has declined from 0.85 to 0.76 ha over the period. The survey estimates a 3.4% annual increase in household income mostly arising from the expansion of nonfarm activities, which now account for 70% of household income compared with 44% in 1988. The incidence of poverty has declined from 65% to 54%, but the improvement in economic conditions has been more pronounced for households earning their livelihood from services and business than for households relying on farming or agricultural labor.

Keywords: Ecosystem, Flood-prone environment, Rice

Location:

Start Date: 00-00-1987

End Date: 00-00-2000

Program Area: Risk Management in Agriculture

Non-commodity: Flood

Objectives:

To assess the changes in the livelihood system of the people in the flood-prone ecosystem in Bangladesh over the last decade when Bangladesh experienced a rapid growth in rice production through technological progress.

Methodology:

The benchmark data for the study were drawn from a sample survey conducted in 1987-88 using a multistage random sampling method for the IRRI-sponsored project "Differential Impact of Modern Rice Technology in Favorable and Unfavorable Rice Production Environments" (David and Otsuka 1994). In the first stage, 64 unions, or village clusters, taking one union each from 64 districts of Bangladesh. In the second stage, data on land area, total population, and literacy rates were obtained for all villages of the selected unions from the district reports of the 1981 population census. Two villages were selected purposively for each union such that the population pressure and the literacy rate for the selected villages were similar to those for the selected unions. Thus, 128 villages were selected, with the first choice within each pair being the village most representative of the union. A census of all households in the first-choice village was undertaken to collect information on the ownership of land, adoption of modem rice varieties, and the major source of household income. Where the village community was found to be noncooperative for conducting the survey, the second choice village was included in the sample. Two sites were dropped at this stage because of the problem of logistics for conducting the survey. This nationally representative sample thus consisted of 62 villages from 57 districts. The census of the selected villages enumerated 9,874 households or 159 households per village. These households were used as the sample frame for the final selection of the sample for data gathering on the operation of the household economy. The households were classified into four land ownership groups: (1) functionally landless (up to 0.2 ha ofland), (2) small landowner (0.2 to 1.0 ha), (3) medium landowner (1.0 to 2.0 ha), and (4) large landowner (more than 2.0 ha). Each of the land ownership groups was further classified into two subgroups according to whether the household was engaged in tenancy cultivation or not. Twenty households were then selected from each village using the proportionate random sampling method so that each of the eight (4 x 2) strata was represented according to its weight. The total sample for the 1987 survey consisted of 1,245 households. The selected households were interviewed with a structured questionnaire to gather data on demographic characteristics of all household members, the use of all parcels of land owned and operated by the households, costs and returns of cultivation of major crops, purchase of inputs and marketing of products, ownership of nonland assets, employment of working members and earnings from nonfarm activities, and the perception of the respondent regarding the economic standing of the household in the village and the changes in its economic conditions. The findings from the 1987- 88 survey were published in Hossain et al (1994). A resurvey was conducted covering the financial year 1999-2000 for 16 of the 62 villages that belonged to the flood-prone ecosystem. The villages where more than 50% of the land was reported as being flooded at a depth of more than 50 cm during the peak of the monsoon season were identified as flood-prone. In the resurvey, the sample was drawn using the classification of households by the wealth ranking method rather than that based on land ownership and land tenure as used in the benchmark study. In consultation with the key informants using the participatory rural appraisal (PRA) methodology, the households in the village were classified into four groups: (1) rich, (2) solvent, (3) self-sufficient, and (4) poor. To ensure that all former sample households and their off-shoots were covered in the present survey, we decided to draw a sample of 30 households from the four groups using the stratified random sampling method. New samples were drawn for the cells that were underrepresented by the sample households selected during the 1987-88 survey. The sample consisted of 320 households for 1987-88 and 485 for the 1999-2000 financial year. Household income as measured in the study includes income received in cash as well as in kind. A money value was imputed to receipts in kind at prices prevailing in the survey villages. Household consumption of self-produced crops and their by-products and livestock, fishery, and forestry products is considered as income. The income from crop production activities is estimated as the value of the main product and the by-products net of the costs of seeds, fertilizer, pesticides, irrigation charge, payment of hired labor, and rental cost of draft machine power. Rent paid to the landowner and received from tenants was included in the calculation of income, but the cost of own land was not imputed on the grounds that the value of the land is not depreciated by its use. The interest on working capital employed in the activity was not imputed either and was considered as a cost. Changes in economic activities and socioeconomic conditions were assessed by comparing the mean values of the criterion variables for an average household for the two surveys. A limitation of the study is that it cannot assess the changes in the socioeconomic situation in the flood-prone ecosystem vis-a-vis the favorable irrigated ecosystem that faces no environmental constraints to the adoption of modem technologies .

Source of Information: Rice research and development in the flood-prone ecosystem. Proceedings of the international workshop on flood-prone rice systems held in Gazipur, Bangladesh, 9-11 January 2001. Los Banos (Philippines): International Rice Research Institute, P. 15-31.

Article Link (Online Journal):

Funding Source:

1.

Budget:

Total Budget (Tk.) :

Achievement/Findings:

Although the availability of land for agricultural production has been declining for rural households, the base of land productivity has increased because of the rapid expansion in the coverage of irrigation and the adoption of modern rice varieties during the dry season. In the wet season, farmers continue to grow traditional rice varieties because modern varieties are unsuitable for the flood-prone environment, but the area under these varieties has declined substantially over time, with farmers growing modern rice varieties during the dry season and leaving the land fallow during the wet season. Nearly 31 % of the land was under the rice-fallow system in 1987- 88, but this increased to 46% in 1999-2000. This led to a significant reduction in cropping intensity. A major change in the livelihood system has been a reduction in the dependence of land-poor households on the agricultural labor market. The proportion of workers reporting agricultural wage employment as the principal occupation was almost halved from 1988 to 2000, as a result of increased employment opportunities available in the rural nonfarm sector. The higher incidence of tenancy and the introduction of mechanized land preparation using power tillers were two other significant changes during the period. Although the increase in yield for individual crop varieties has been moderate, the overall increase in yield in rice cultivation has been substantial because of the reallocation of land from low-yielding aus and deep water aman rice to high-yielding boro rice. The increase in productivity in rice cultivation, however, has not been translated into higher farm income because of the much slower increase in paddy prices compared with the wage rate and fertilizer prices. The nominal wage rate has in- creased almost on a par with the consumer price index, but, because of the slow increase in the nominal price of paddy, the entitlement of staple food for land-poor households has improved substantially. Rice accounts for a small share of household income because of the very small farm size and rice's unfavorable terms of trade. Per capita income has grown at 3.4% per year, entirely because of a rapid expansion of income from nonagricultural sources-trade and business, transport operations, services, and remittances. This transformation from agriculture to a nonfarm-based rural economy has been facilitated by an improvement in human capital and the development of rural infrastructure, particularly rural roads. As a result of the rapid growth in the nonagricultural economy, there has been a moderate improvement in the poverty situation, although in absolute terms the level of poverty remains very high. The improvement has been more pronounced for households that derive income from services and trade as the major source, and only marginal for households engaged in fanning or daily wage earning. This study indicates that improvement in rice technology can make only a limited direct contribution to an increase in household income. The role of research for rice improvement should be seen as increasing rice supplies on a par with demand and reducing the unit cost of production so that the price of this dominant food staple could be kept within affordable limits for the rural and urban poor. Since a large proportion of land in the flood-prone environment remains single- cropped, options to introduce a double-rice cropping system should be explored by developing shorter-duration and cold-tolerant aman and boro rice varieties. The incorporation of cold tolerance in modern boro varieties may allow the early harvesting of boro and incorporation of traditional deep water aman in the cropping system. Areas in which the floodwater recedes early may benefit from the availability of short- duration arnan varieties that could be transplanted in late September and harvested in late December, still allowing time to grow boro rice during the dry season. Since farmers keep the seed from their own harvest for the large number of traditional aman varieties grown during the wet season, some yield gains could be made and biodiversity preserved from pure-line selections of the popular varieties and their distribution through the seed market.

Knowledge Management: Report/Proceedings