Heterogeneity of Inter-regional Efficiency of Education Public Spending in Fragile State: Evidence from Chad
Date
2021-05-22
Authors
Mallaye, Douzounet
Gadom, Gadom Djal
Journal Title
Journal ISSN
Volume Title
Publisher
AERC
Abstract
This study analyses the inter-regional efficiency of education public spending in Chad,
which is characterized by a long history of armed conflict. Specifically, we investigate
the institutional, political, and security dimensions of fragility influencing the interregional education public spending efficiency for the period 2007-2016. The data used
are from the Ministry of National Education and the National Institute of Statistics,
Economics and Demographic Studies (INSEED) and cover 23 regions of Chad. The
education public sector efficiency scores is estimated in the first stage using the Data
Envelopment Analysis (DEA) approach, while in the second stage, the panel data Tobit
regression is used to evaluate the determinants of the efficiency of the education
public spending. The mean efficiency score is 0.96 in primary education and 0.90
in secondary education. This means that the better management of the education
sector may allow reducing the amount of public education spending by 4% in primary
education sector and 10% in the secondary education sector, given the actual level of
the quality and the quantity of education. With regard to these results, the primary
sector is more efficient than the secondary sector. The Tobit estimates suggests
that efficiency is negatively associated to head of region reshuffle, armed conflicts
sequence, and the number of parliament member, which capture, respectively, the
institutional, security, and political dimensions of fragility. Based on our findings, it
is recommended to the government to reduce the central government instability in
general, and specifically the recurrent turnover of the head of region, these being the
main determinants of public spending inefficiency in Chad.
Description
Keywords
Public spending , Efficiency , Fragile state , DEA , Panel data Tobit , Chad.