close× Call Us
close×
Title Political and Economic Targeting of Development Aid: A Field Experiment among Elected Officials in Malawi
Post date 11/02/2015
C1 Background and Explanation of Rationale Development interventions often assume funds and materials will be allocated in accordance with need. In practice, such efficient targeting outcomes are rarely achieved. Government officials are the middlemen in many development projects, allocating foreign aid to organizations, facilities, and individuals in their communities. However, these officials sometimes have political incentives that conflict with development goals, such as the need to cull favour with critical groups of voters and the desire to claim credit for visible development projects. This tension is particularly poignant in Malawi, where foreign development aid constitutes a large portion of the national budget and decentralization leaves district officials in charge of managing it.
C2 What are the hypotheses to be tested?
Baseline allocation
H1. Councillors will allocate more aid to schools with high need than with low need
H2. Councillors will allocate more aid to schools located in areas with higher support for the councillor in the last election
H3. Councillors will allocate more aid to schools located in areas with higher support for the incumbent president (Mutharika) in the last election.
H4. Councillors will allocate less aid to schools located in areas with higher support for opposition candidates in the last election.
H5. Councillors will allocate more aid to schools located in their home area.
H6. Councillors will allocate less aid to schools located in the opposition candidates’ home area.
H7. Councillors will allocate more aid to schools located in densely populated areas and near well-travelled areas like markets.
H8. Councillors will value less those goods that can only be used for schooling purposes (teacher supply kits) than those that can be used for other purposes (roofing, lamps).
H9. Councillors will be more to allocate those goods that can be used for non-schooling purposes (roofing, lamps) to political supporters than goods that can only be used for schooling (teacher supply kits).
H10. Councillors will be more to allocate those goods that can be used for non-schooling purposes (roofing, lamps) to less needy schools than goods that can only be used for schooling (teacher supply kits).
 
Political Support Information Treatment
When councillors receive information about the distribution of political support in their ward (relative to baseline):
H11. Councillors will be more likely to allocate to schools located in areas with higher support for the councillor in the last election.
H12. Councillors will be less likely to allocate to schools located in areas with higher support for opposition councillors in the last election.
H13. Councillors will be less likely to consult the VDC or the DC when deciding how to allocate to schools.
 
Economic Need Information Treatment
When councillors receive information about the distribution of political support in their ward (relative to baseline):
H14. Councillors will be more likely to allocate to schools in areas with high need than with low need.
H15. Councillors will be less likely to allocate to schools located in areas with higher support for the councillor in the last election. Transparency Treatment When councillors are aware that decisions will be announced to the ADC or citizens:
H16. Councillors will be more likely to allocate to schools with high need than low need.
H17. Councillors will be less likely to allocate to schools located in areas with higher support for the councillor in the last election.
H18. Councillors will be more likely to allocate to schools located in areas with higher support for opposition councillors in the last election.
 
Credit Claiming Treatment
When councillors are shown a picture of the item with their name and the name of the ward or the item itself:
H19. Councillors will be more likely to allocate to schools located in areas with more support for the councillor in the last election.
H20. Councillors will be less likely to allocate to schools located in areas with higher support for opposition councillors in the last election.
H21. Councillors will be more likely to allocate to schools located in areas where they have fewer political networks.
H22. Councillors will be more likely to value the ability to allocate the good.
 
Exposure Treatment
When councillors are shown information about the distribution of population and markets:
H23. Councillors will be more likely to allocate projects to highly populated areas, or near market towns.
 
Sub-group effects
H24. Female councillors will be more likely than male councillors to allocate goods to high need areas.
H25. Councillors who faced a competitive election in 2014; or who expect to face a competitive election, will be more likely to allocate projects to politically important or “swing” areas.
H26. Councillors who anticipate contesting the next election will be more likely to allocate projects to politically important or “swing” areas.
H27. Councillors who faced a competitive election in 2014; or who expect to face a competitive election, will be more likely to allocate projects to high need areas.
H28. Councillors who anticipate contesting the next election will be more likely to allocate projects to high need areas.
H29. Councillors who are declared members of the ruling party will be more likely to allocate to schools located in areas with higher support for the ruling party in the last election.
C3 How will these hypotheses be tested? * We plan to execute a unique intervention to experimentally evaluate how local government elected officials make decisions about targeting development aid in Malawi. In partnership with a local NGO, we will provide all elected councillors in Malawi with the opportunity to decide which school in her ward should receive development materials. In order to evaluate how officials make these decisions, we will randomly assign the availability of political support information, the availability of economic need information, the transparency of the official’s decisions, and the credit claiming opportunities embedded associated with the allocation decision. The randomly assigned treatments in this experiment are designed to be low-cost adjustments to the aid allocation process, so that this experiment can practically inform development interventions in Malawi and elsewhere.
C4 Country Malawi
C5 Scale (# of Units) 350
C6 Was a power analysis conducted prior to data collection? Yes
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Yes
C8 IRB Number London School of Economics and Political Science - 393
C9 Date of IRB Approval 25/06/15
C10 Will the intervention be implemented by the researcher or a third party? Researchers
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? not provided by authors
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? not provided by authors
C13 JEL Classification(s) F35, O1