Evaluating the long-term impact of antipoverty interventions in Bangladesh: an overview

Programmes are rarely evaluated in terms of both short- and long-term impact. This paper provides an assessment of the long-term impact of three antipoverty interventions in Bangladesh on monetary and non-monetary measures of well-being. The interventions are about the introduction of new agricultural technologies, educational transfers, and microfinance.

The authors emphasise that evaluating the impact of antipoverty programmes must go beyond the household as the unit of analysis, to examine intra-household impacts. Impact indicators should also go beyond incomes and expenditures, because of the importance of inter-generational effects of child and maternal nutrition.

On the other hand, the paper notices that some interventions do well in terms of increasing per capita expenditures or assets, but do poorly in terms of improving nutritional status. Therefore, a genuine assessment of long-term impact of development interventions must involve trade-offs between objectives.

The results of the study indicate that:

• both government and civil society interventions have the potential to reduce poverty in Bangladesh
• targeting mechanisms and implementation modalities have important implications for the ability of interventions to reach the poor
• in the case of the use of NGOs and groups for service delivery, it is important not to idealise NGOs
• group-based approaches that involve women seem to have had favourable impacts on individual nutritional status, even if their impacts on monetary indicators appear low
• nevertheless, reaching the very poor may require using different approaches than the group-based approach
• in the case of the primary education stipend, better targeting would make it possible to give larger transfers to fewer beneficiaries
  

The authors conclude with a plea for systematic impact evaluations to guide the design and modification of antipoverty programmes.