Page 121 - Contributed Paper Session (CPS) - Volume 7

CPS2043 Adnan Dawood K. B. et al.
3. Result
We can describe the whole process of the Enhanced PMTF in the following
four steps:
i. Applying PMTF1; is the predicted expenditure above CP_1, if yes then
the household belongs to HWL, then household is not poor, if no
continue.
ii. Applying PMTF2, Is the predicted expenditure above CP_2, if yes then
the household belongs to MWL; apply PMTF3 to predict the annual per
capita expenditure and decide whether the household is poor or not, if
no continue (below CP_2).
iii. Applying PMTF4, Is the predicted expenditure above CP_3, if yes then
the household belongs to LMWL; apply PMTF5 to predict the annual
per capita expenditure and decide whether the household is poor or
not, if no continue (below CP_3).
iv. Applying PMTF6 for the BWL to predict the annual per capita
expenditure and decide whether the household is poor or not.

This new mechanism on using the PMTF is enhancing the efficiency of
targeting the poor households, where the leakage rate and the under‐
coverage rate are decreased significantly. Especially the decrease in the under‐
coverage rate, because of the sixth PMTF which covers the poorest of poor is
passing through the origin. Hereafter, a table explains how the under‐coverage
rate and the Leakage rate are calculated.

4. Discussion and Conclusion
Using the PMTF is one of the best from the six targeting methods available
and most accurate method. The PMTF method does not force us to ask about
income or expenditure, especially in "poor countries" where no official
administrative data on households/ individuals income. The enhancement on
the efficiency of the PMTF proposed in this paper; by applying more than one
formula for targeting the poorest of the poor households, increase the
importance and the credibility of the PMTF in targeting poor households, and
guarantee that the assistance goes to those who deserve it.
The suggested number of cutoff points and the number of equations;
which serves in increasing the efficiency of the PMTF, are changeable based
on poverty line, poverty incidence in the society, and the available number of
cases required to run the linear regression are available. In this paper we have
suggested three equations for classifying the households in four categories;
HWL, MWL, LMWL, and the BWL through three cutoff points on the scale of
annual per capita income/ expenditure. The number of cutoff points can be
decreased to two points, then; three equations to classify the households in
three categories and get two final equations for the final PMTF.

108 | I S I W S C 2 0 1 9