This paper first reviews the performanceof the current targeting system and recommends improvementsto the existing PMT formula. One implementation issue weidentified is the inconsistency in how monetary andnonmonetary values are treated when calculating the PMTscore. The practice has been to calculate the score based onmonetary variables (income, social benefits package, andutility expenditures) at current nominal prices, which haveinevitably inflated the predicted consumption. As the PMTformula was estimated at 2013 prices, the recommendation tocorrectly implement the current PMT formula would be todeflate all monetary variables to 2013 prices. Second, tofurther improve the poverty targeting accuracy, the paperpresents the updated PMT model estimation and discusses thebenefits and cost of transitioning to the new PMT formula.The empirical analysis is based on the 2018 HIES collectedby the National Statistics Office of Georgia (Geostat), thelatest available to the team at the time of the analysis.The remainder of this note is organized as follows. Sectiontwo looks at the performance of the current PMT model toidentify key areas of possible improvements. Section threedevelops alternative estimates and proposes a new PMTformula, comparing its theoretical performance with that ofthe current one. This is followed by a discussion of theimplementation of the formula in the following years toavoid problems encountered with the current approach.Section four goes further to make a more radical proposal toconsider moving to an ‘hybrid’ approach combining an incometest with the PMT, highlighting the advantages of theapproach and the steps needed for its development andimplementation. A final section provides a summary of keyrecommendations for the immediate, short, and medium term.
PDF
download
878987797
028-85220240
