On Assessing Pro-Poorness of Governments Programs: International Comparisons

On Assessing Pro-Poorness of Governments Programs: International Comparisons By N. Kakwani Director International Poverty Centre Brasilia Brazil Email: nanak.kakwani@undp-povertycentre.org

Different kinds of government programs Cash programs: Safety net in industrialized market economies Conditional cash transfers in Latin America In-kind programs: Food subsidies Food vouchers Public works Government services: Education : Schools Health : Hospitals Sanitation and clean water

What is pro-poor policy? A government policy can be said to be pro-poor if it benefits the poor more than the non-poor. It means that with a fixed cost to the government, a pro-poor policy should achieve greater poverty reduction compared to a counter-factual situation when everyone receives exactly the same benefit from the policy. Policy A will be more pro-poor than policy B if for a given cost, policy A leads to a greater poverty reduction than policy B..

Additive and decomposable class of poverty measures ( ) = z dx x f x z P 0 ) (, θ α = z x z P x) (z, 2 1 0 = = = α α α H e a Headcount ratio Poverty gap Severity of poverty

Pro-Poor Policy (PPP) Index b(x): benefits received by an individual with income x = 0 ) ( ) ( dx x f x b b = mean benefit per person % change in total poverty due to a universal targeting of b η b where is the percentage change in poverty if everyone receives one unit of currency (i.e. absolute elasticity) η = z dx x f x P b d 0 * ) ( θ θ θ = z dx x f x b x P d 0 ) ( ) ( 1 θ θ θ : % change in total poverty due to b(x) =

Pro-poor Policy Index 1 P λ = b( x) f ( x) dx bηθ x Example: (i) λ = 1.20 : a programme reduces poverty 20% more compared to a counterfactual universal targeting (ii) λ = 0.70 : a programme reduces poverty 30% less compared to a universal targeting

Setting a Benchmark for PPP index Imperfect targeting: worst scenario b ( x) = 0 if x< z b ( x) 0 if x z Perfect targeting: best scenario b( x) = k( z x) if x< z b ( x) = 0 if x z * Note: for incentives, k should be always < 1

PPP index under perfect targeting Single poverty line - poverty gap ratio: 1/H s - severity of poverty: 2 g g=poverty gap ratio s=severity of poverty index. Needs adjusted poverty lines λ m = 1 ηθ z b 0 P x ( z x) f ( x) dx

The pro-poor policy index for the whole country is the weighted average of the total-group PPP indices for the individual groups, with weight proportional to shares of benefits received by each group. PPP index by socioeconomic groups Within group PPP index λ k = 1 P x Total group PPP index λ * k = z bkη kθ k 0 z 1 P ηθ x bk 0 b( x) f b( x) f k k ( x) dx ( x) dx

Welfare programs : Thailand 2000 Table 1: Pro-Poor Policy Index for welfare programs in Thailand: 2000 Welfare Schemes Poverty gap ratio Severity of poverty Social pension for the elderly 1.68 1.54 Low-income medical cards 2.02 2.12 Health insurance cards 1.29 1.25 Free school lunches 2.02 2.06 Perfect targeting 6.77 10.31 Universal social pensions (for elderly over 65 years of age) 1.21 1.24

Welfare programs: Thailand 2000 Table 2: Pro-Poor Policy index by urban and rural areas: Thailand 2000 Total-Group PPP index Within-Group PPP index Welfare Schemes Urban Rural Urban Rural Poverty gap ratio Social pension for the elderly 1.13 1.76 4.41 1.31 Low-income medical cards 1.44 2.10 5.60 1.56 Health insurance cards 0.70 1.39 2.72 1.03 Free school lunches 0.81 2.21 3.15 1.64 Severity of poverty Social pension for the elderly 1.18 1.60 5.42 1.17 Low-income medical cards 1.34 2.23 6.18 1.63 Health insurance cards 0.61 1.36 2.83 0.99 Free school lunches 0.73 2.27 3.37 1.66

Russian welfare system Table 3: The Russian welfare system: 2002 Beneficiaries Per month cost (Rubles) in million % share in billion % share Old-age pension 26.32 49.08 38.74 82.79 Disability pension 3.19 5.96 3.61 7.71 Loss of breadwinner pension 1.64 3.05 1.27 2.72 Social pension 0.27 0.5 0.26 0.56 Care for children under 18 m 0.84 1.57 0.41 0.88 Children allowance 17.42 32.49 1.45 3.09 Unemployment benefits 0.45 0.84 0.31 0.65 Other benefits 0.95 1.77 0.2 0.42 Scholarship 2.55 4.76 0.55 1.17 All benefits 53.63 100 46.79 100 Total Population 143.32 37.41

Old-age pension Disability pensio Loss of breadwi Social pension Care for childre Children allowa Unemployment b Other benefits Scholarship All benefits Perfect targeting 6 5 4 3 2 1 0 PPP index for Russian welfare system The PPP index for Russian Welfare System Poverty gap Severity of poverty

Table 3: Pro-Poor Policy Index for health services in Vietnam: 1997-98, poverty gap ratio Total-Group PPP index Within-Group PPP index Health facilities Vietnam Urban Rural Urban Rural Poverty gap ratio Government hospitals 0.62 0.07 0.91 0.34 0.74 Commune health centres 1.17 0.27 1.23 1.38 1.00 Regional polyclinics 0.84 0.42 0.98 2.14 0.79 Eastern medicine facilities 0.96 0.04 1.15 0.21 0.94 Pharmacies 0.96 0.26 1.16 1.29 0.94 Private doctors 0.79 0.12 0.98 0.59 0.80 Health insurance 0.50 0.08 0.79 0.40 0.64 Perfect targeting 2.86

PPP index for education : Vietnam97-98 School Types Primary Lower Secondary Upper Secondary Poverty gap ratio Public 1.29 0.79 0.37 Semi Public 0.55 0.15 0.23 Sponsored 0.63 0.51 0.00 Severity of poverty Public 1.31 0.65 0.23 Semi Public 0.19 0.08 0.09 Sponsored 0.14 0.26 0.00 Universal Education Poverty gap ratio Severity of poverty Primary 1.28 1.33 Lower Secondary 1.08 1.06 Upper Secondary 0.91 0.85

3.0 2.5 2.0 1.5 1.0 0.5 0.0 Targeting children in Africa Figure 1: Pro-Poor Policy indices under universal transfers and perfect targeting (poverty gap ratio) Burkina Faso Cote d'ivoire Cameroon Ethiopia Ghana Guinea Gambia Kenya Madagascar Mozambique Malawi Nigeria Uganda Zambia Universal (rural children) Universal(all children) Perfect targeting Brundi

Brundi 98 Burkina Faso Cote d'voire 9 Camroon 96 Ethiopia 00 Ghana 98 Guinea 94 Gambia 98 Kenya 97 Madagascar 0 Mozambique Malawi 97 Nigeria 96 Uganda 99 Zambia 98 1.2 0.8 0.6 0.4 0.2 1 0 Targeting Unemployed in Africa Male Female Brundi 98 Burkina Faso 9 Cote d'voire 98 Camroon 96 Ethiopia 00 Ghana 98 Guinea 94 Gambia 98 Kenya 97 Madagascar 01 Mozambique 96 Malawi 97 Nigeria 96 Uganda 99 Zambia 98 1.5 0.5 1 0 Targeting unemployed in Africa Rural Urban

Zambia 98 Uganda 99 Cote d'ivoire 98 Cameroon 96 Ethiopia 00 Ghana 98 Guinea 94 Gambia 98 Kenya 97 Madagascar 01 Mozambique 96 Malawi 97 Nigeria 96 0.70 0.65 0.60 0.55 0.50 0.45 0.40 0.35 0.30 Food subsidies in Africa Poverty gap Severity of poverty Burundi 98 Burkina Faso 98

Burundi 98 Burkina Faso Cote d'ivoire Cameroon 96 Ethiopia 00 Ghana 98 Guinea 94 Gambia 98 Kenya 97 Madagascar 0 Mozambique Malawi 97 Nigeria 96 Uganda 99 Zambia 98 3 2.5 2 1.5 1 Food subsidies targeted at the poor PPP index for food subsidies targeted at the poor Poverty gap Severity of poverty

Children in Brazilsuffer more severe poverty than the rest of the population Table 1: Poverty in Brazil 2003 Percenatge Poverty Severity of poor Gap ratio of poverty The W hole Polulation Metropolitan 20.7 9.3 5.7 Urban 25.7 11.1 6.5 Rural 51.4 25.2 15.5 Brazil 27.9 12.6 7.6 Children 5-17 years old Metropolitan 31.8 14.4 8.7 Urban 37.0 16.3 9.6 Rural 66.5 34.4 21.7 Brazil 40.5 18.8 11.4

The poor have greater access to Bolsa-Escola than the Non-poor. Figure2: Percentage of children receiving benefits by poverty status 60 50 40 30 Poor Non-poor 20 10 0 5 6 7 8 9 10 11 12 13 14 15 16 17

Bolsa-Escola does not necessarily favor the girls in terms of access. Figure1: Percentage of children receiving benefits by sex 40 35 30 25 20 15 Boys Girls 10 5 0 5 6 7 8 9 10 11 12 13 14 15 16 17

Bolsa-Escola has large exclusion error Exclusion error: probability of not receiving benefits when poor Inclusion error: probability of receiving benefits when non-poor 50 Figure 3: Exclusion and inclusion errors 40 30 20 Exclusion Inclusion 10 0 5 6 7 8 9 10 11 12 13 14 15 16 17 Total

Bolsa-Escola reduces the gap between poor and non-poor in school attendance Percentage of children attending school who are non-beneficiery 110 100 90 80 Poor non-beneficiery Non-poor non-beneficiery 70 60 50 5 6 7 8 9 10 11 12 13 14 15 16 17 Percentage of children attending school who receive benefits 105 100 95 90 85 80 Poor beneficiery Non-poor beneficiery 75 70 65 5 6 7 8 9 10 11 12 13 14 15 16 17

Bolsa-Escola has greater impact on school attendance among the poor than the non-poor. Percentage of poor children attending school 110 100 90 80 Poor beneficiery Poor non-beneficiery 70 60 50 5 6 7 8 9 10 11 12 13 14 15 16 17 Percentage of non-poor attending school 105 100 95 90 Non-poor beneficiery Non-poor non-beneficiery 85 80 75 5 6 7 8 9 10 11 12 13 14 15 16 17

Bolsa-Escola is generally pro-poor Table 2: PPP index for beneficiaries PPP index for beneficiaries Universal targeing 5-17 years Gap Severity Gap Severity Metropolitan 2.82 3.02 1.54 1.56 Urban 2.39 2.49 1.44 1.48 Rural 1.47 1.59 1.29 1.36 Brazil 2.35 2.52 1.46 1.50 Table 3: PPP index for applicants PPP index for beneficiaries Gap Severity Metropolitan 2.93 3.19 Urban 2.21 2.34 Rural 1.38 1.45 Brazil 2.17 2.30

Bolsa-Escola is generally pro-poor Table 2: PPP index for beneficiaries PPP index for beneficiaries Universal targeing 5-17 years Gap Severity Gap Severity Metropolitan 2.82 3.02 1.54 1.56 Urban 2.39 2.49 1.44 1.48 Rural 1.47 1.59 1.29 1.36 Brazil 2.35 2.52 1.46 1.50 Table 3: PPP index for applicants PPP index for beneficiaries Gap Severity Metropolitan 2.93 3.19 Urban 2.21 2.34 Rural 1.38 1.45 Brazil 2.17 2.30

Targeting efficiency is higher in rural areas Table 4: PPP and TE Indices (5-17 years). PPP index TE Index Gap Severity Gap Severity Actual beneficiaries Metropolitan 1.83 1.93 0.58 0.61 Urban 1.66 1.69 0.61 0.62 Rural 1.14 1.16 0.76 0.77 Brazil 1.62 1.68 0.65 0.68 Applicants Metropolitan 1.90 2.04 0.60 0.65 Urban 1.54 1.58 0.57 0.59 Rural 1.07 1.06 0.71 0.70 Brazil 1.49 1.53 0.60 0.62

Santa Catar São Paulo Rio de Janei Rio Grande d Paraná Rondônia Distrito Fede Mato Grosso Goiás Espírito Sant Minas Gerais Mato Grosso Roraima Acre Amapá Pará Amazonas Tocantins Sergipe Rio Grande d Paraíba Pernambuco Ceará Bahia Piauí Maranhão Alagoas 0.9 0.8 0.7 0.6 0.5 0.4 0.3 The poorer states generally have greater targeting efficiency Fig: Targeting Efficiency of Bolsa-Escola

Inequity in Brazilian education system is very high Table 5: Pro-poor index by levels of grades G rade P overty S everity of gap poverty P rim a ry - re g u la r 1.5 3 1.5 7 P rim ary-regular public 1.68 1.73 P rim ary-regular private 0.27 0.23 S econdary - regular 0.73 0.63 S econdary-regular public 0.83 0.72 S econdary-regular private 0.10 0.09 A dult prim ary 1.09 1.04 A dult secondary 0.52 0.44 Tertiary 0.07 0.07 Tertiary public 0.12 0.10 Tertiary private 0.05 0.06 A dult literacy 1.73 1.90 Child care 1.08 1.14 P re-s c hool 1.46 1.56 P re-vestibular (pre-tertiary) 0.19 0.15 P ost-graduation 0.00 0.00