Household Food Aid Targeting: the case of Antsokia-Gemiza wereda.
No Thumbnail Available
Date
2001-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
This study aims at assess ing th e targeting efficiency in Antsokia Ge miza Wereda, in
North Shewa. The study ba sed its obj ectives on th e hypothesis of household targeting
prac tice s are not efficient in Antsokia due to the failure to exclude the non-eligible
beneficiaries at the expense of eligible non-beneficiaries. Th e specific objectives are (i) to
examine different targeting methods used to select beneficiary households of relief, (ii) to examine
criteria used to set a threshold among beneficiary households, (iii) to estimate the targeting
accuracy (error of inclusion and error of exc lusion) against its intended eligib le groups, (iv) to
compare resource given to households with the 'idea/' ration (error of under supply and error of
over supply) and finally (v) to assess major factors that affect targeting efficiency.
Information gathered by empl oying a co mbination of prim ary and secondary data; th e
study applied both qualitative and quantitative methods. Th e secondary data ga th ered
from previously published materials wh ile the primary data based its information on the
200 relief distribution have been co llected from the sampled households h eads,
interviews, and observa tio ns. Th e household survey app lied economic vulnerability
indicators to identify household's 'eligibility' or 'ine ligibility' for relief aid. Th e major
indices, i.e. livestock ownership, land holding, dependency ratio, off-farm in come, crop
production and oxen own ership, were convert ed into a standard scale th e 'Z-score '.
The study noticed that the current targeting mechanism in Antsokia wereda has
proficient procedures, de sp ite actual practices impeded its e fficiency. It applied a
co mbined administrative-community targeting method, the combina tion is advantageous
to complement each other. Th e administrative targetin g decided th a t the eligibility
indicator would be asse t ownership . However, the community targe ting found oxen
ownership indicator easier to screen out eligible beneficiary households. The wereda
implemented its selection factor (oxen own ership) with 14% inclusion and exclusion
errors. Both pair of errors, inclusion/exclusion and over/under supply targeting errors are
witnessed in th e relief targeting. The findin gs show that there was 32.5% error of incidence (using
economic vulnerability indica tor) and 91.3% error of benefit. Th e major factors which determined
the efficiency of targeting are: (a) DPPC approved below half of the wereda's relief request , (b)
oxen ownership selection factor genuinely denied some of the food insecure and included th e
food secure households, (c) the community targeting had been affected by favoritism and
discrimination (d) the co mmunity refused to participate in the selection and (e) the beneficiary
households did not get all their ration.
Identification of the exact problems, tied with th e efficiency of targeting system s, is used
to suggest practical and affordable methodologies for future intervention in the wereda
and other we redas with similar features. Th e study recommended (a) to balance Ple dge
and delivery (b) to construct additional warehouse to th e wereda (c) to extend th e
se le ction factor from oxen ownership to related economic statu s indicators (d) to
encourage th e community to participate in targ eting and (e)to strictly f ollow th e intended
household/individual ration.