Bangladesh: Disasters and Public Finance

Working and discussion papers
November 2002
Charlotte Benson and Edward Clay

1. Bangladesh is one of the most disaster-prone countries in the world. Physical hazards that could potentially cause considerable loss of life and catastrophic physical damage and disruption to society and the national economy include exceptionally widespread riverine flooding, severe tropical cyclones and associated coastal storm surges, drought and earthquakes. In addition, rapid on-set flash flooding, tornadoes and riverbank erosion are frequent causes of more localized, but nevertheless intense human suffering and losses.
2. Because of Bangladesh’s large, densely settled population, low income and widespread poverty, the impacts of disasters have been the focus of considerable international attention and a substantial body of investigations from environmental, social and wider economic perspectives. However, this study is the first systematic investigation into the implications of disasters on the public finances in Bangladesh, apart from assessments of short-term impacts of individual events such as the extreme floods of 1998.
3. Major disasters have had massive human and social impac ts: official estimates are that 139,000 people were killed during the 1991 cyclone, whilst 31 million were directly affected by the 1998 floods. These extreme events also have clearly demonstrable negative impacts on the Bangladesh economy. However, the relative severity of these economic effects of disaster shocks has considerably diminished since the 1960s and highly vulnerable post-conflict 1970s. This is largely due to structural change in the economy:
•agriculture still suffers severe losses, but secto ral output has become more resilient with the spread of
dry season irrigation;
•non-agricultural sectors are more important and less severely affected than agriculture was in the past;
•the integration of food markets, linked with liberalization of internal and external trade, is a factor reducing price effects of shocks.
There have been improvements in response by government, donors and civil society including:
•the rapid short term increase in food safety net operations;
•better targeted relief and rehabilitation by NGOs;
•the refinancing ( in 1998/9) of micro-finance.
There have also possibly been improvements in mitigation, such as protection of urban areas and some infrastructure.
4. The 1998 floods were handled better, than earlier major disasters by government and NGOs, with substantial and mostly timely support by the international community. Improved and timely assessment of losses facilitated these responses. Some newer critical areas of increasing vulnerability were exposed, including the garment industry’s production and exports and the microfinance sector. Many continuing weaknesses in disaster management were underscored.
5. The public finance effects of major disasters are complex and impacts are difficult to isolate and quantify. The immediate disruptive effects of a major disaster on both central government spending and revenue collection have been progressively contained. Instead in the 1990s some short term increases in recurrent spending began to buffer the economy-wide effects of disaster shocks. Disasters appear to largely involve reallocations of expenditure to relief from the Annual Development Plan (ADP) and reallocations within categories of expenditure.
The prevailing view is that reallocations are made largely on an ‘equal pain principle’, cutting expenditure uniformly across the board without regard to negative longer term effects on development activity.
6. There is little additional aid in the medium term (2 to 3 years), as many donors largely respond to a crisis by reallocating and bringing forward commitments. There is also apparently reallocation between categories from project to food aid and commodity (short term BoP/budgetary support) and between channels from direct aid to government to NGOs. These effects merit further investigation. Some donors felt they were able to shift funds from non-performing projects to meet rapid short-term requirements. But concerns were expressed that some have resisted allocations away from their bilateral priorities, without regard for wider implications.
7. There are intrinsic difficulties in examining issues of public finance, especially where there are widely recognized problems of governance that may affect the quality of data and the transparency of the budgetary process. Some reorganization of public finances reporting, such as integration of the food account into the budget proper would contribute to transparency and possibly improved budgetary practice.
8. If the response to disasters largely involves reallocations of public expenditure and development assistance, then the effects will require examination of spending and activity at a project level, both to detect negative effects and to identify ways in which these can be minimized in the future.
9. Disaster mitigation, especially the effectiveness of river and coastal flood protection has been weakened by inadequate recurrent expenditure on operations and maintenance (O&M). The decentralization of responsibility for O&M to local government and community level and cost recovery from protected communities need to be seriously explored.
10. The practicality of newer forms of spreading the financial costs of disasters merits further careful consideration. The sensitivity of micro credit to disasters is another indication of the need to incorporate insurance into all formal lending to the rural sector. At a national level the scale of reallocations by government and short term borrowing at high cost after the 1998 floods suggest the need for risk spreading, including disaster-earmarked contingency reserves and, possibly, market based instruments for risk spreading.
11. The complexity of the physical environment and concerns about the changing risks resulting from climatic change and human intervention in Bangladesh and neighbouring states justify continued investigation into hazards and improved risk assessment. Better risk assessment is a necessary condition for developing newer forms of risk spreading mechanisms. The limitations of flood forecasting again exposed in 2000 and weakness in seismic monitoring illustrate the need for ensuring adequate investment in hazard risk assessment in Bangladesh and strengthening regional cooperation.

Read the research