Policy and poverty in inland capture fisheries in Bangladesh.

The thesis's objective is to describe the relationship between inland capture fisheries management and the welfare of the poor m Bangladesh. Existing economic models of fisheries management do not address issues of income-distribution, concentrating instead upon rent-maximisation. The thesis therefore aims to develop new models of the distributional impact of management interventions, based upon a qualitative and quantitative understanding of the social, institutional and economic processes whereby fishery incomes are created and distributed. The study site is Hail Haor, a 13,600 ha floodplain in N.E. Bangladesh. Descriptions of its physical ecology, fishing technologies and fishing communities are given. There follows a detailed explanation of the social, institutional and economic transactions whereby fishing incomes are distributed between fishing labour, lessees, sub-lessees, government officials and the government. Estimates of yields, sales and employment levels are presented on a per hectare basis. Sales are divided between labour income, fishing costs and tolls. Tolls are in turn apportioned to the various groups of recipients. Fishermen's access to profitable fishing activities is shown to be a function of religion and social and economic status. Poor fishermen tend to operate with low-yield, labour-intensive gears, in shallow water and during the flood season. They therefore do not perceive themselves to benefit from policies that enhance the dry-season, capital-intensive fishery. Simulation and analytical models demonstrate how current policy interventions benefit rich people at the expense of poorer fisherfolk. They raise rents, which accrue to rich people, by reducing employment, upon which the poor depend. They also postpone the catch from the labour-intensive flood season fishery to the capital-intensive dry-season fishery. It is concluded that existing policy is regressive and that fisheries assessments should address the issue of income-distribution as well as that of rent-maximisation. Appropriate assessment methodologies are proposed