Result Based Transfers Marcos C. Holanda Universidade Federal do Ceará [email protected] Result Based Transfer • Fiscal transfers are the oxygen that keep political unions alive. • Fiscal Transfers: – First generation “The King” will – Second generation Output based – Third generation Result Based • RBT is the big “carrot” ( stick, carrot and sermons) of RBM - Result Based Management. Result Based Transfer RBT are a powerful tool to promote the key message of RBM: What matters is not what you did but what you delivered. Besides that it change the incentive: FROM : If a get worse I get more TO : If a get better I get more Result Based Transfer • An example of RBT is a New Law, signed by the Governor of the State of Ceará on December 17, 2007, that regulate the way the State distribute part of its VAT revenue among its municipalities. • In Ceara, as in the rest of Brazil, we have: 75% to State ICMS (VAT) 25% to Municipalities 75% → Defined by Federal Law (Econômic Activity) 25% → Defined by State Law (Selected Variables) *ICMS is the denomination for the State’s Value Added Tax . Law ICMS Law • The old law: 5% → Municipality Population 25% → Defined by State Law (Selected Variables) 12,5% → Expenditure on Education 7,5% → Equally distributed • The new law 25% → Defined by State Law 18% → Education (performance of students on standardized exams) 5% → Health (Infant Mortality Rate) (Selected Variables) 2% → Enviroment (Appropriate Waste Disposal System) ICMS Law • Education: Indicators considered in the coeficient: • Student approval rate in first five grades of elementary school • Average grades of Second Year Students in reading exams • Average grades of Fifth Year Students in math and portuguese Formula: Level Advance Approv i Scorei Scorei Coeficienti 0,2 0,8 0,4 0,6 Approv Score Scorei i i i i i * Score is adjusted for Standard Deviation and the proportion of students that were tested Health: Indicator considered in the coeficient: Infant Mortality Rate (IMR) Formula: Level Coeficienti 0,5 Advance DMi DMi 0,5 DMi DMi i i DMi 100 IMR i Controls for potencial “gaming” behavior Groups of students Municipality A Municipality B Results Only some groups receive higher attention, which can raise the average of the municipalities in detriment of the majority of students. All groups receive attention, reducing the dispersion between the results of students. Average Grade = 200 Standard Deviation = 50 AG 0,5 SD 8 Average Grade = 150 Standard Deviation = 30 AG 0,5 SD 10 Controls for potencial “gaming” behavior Students enrolled (NE) Students Assessed (NA) Results Average Grade = 200 Municipality A NA/NE = 0 ,50 AG NA N E 100 Average Grade = 150 NA/NE = 1 Municipality B AG NA N E 150 Convergence Municipality Share 0.4 0.35 0.3 % 0.25 COEF. 2008 0.2 COEF. 2009 0.15 COEF. 2010 0.1 0.05 0 0 20 40 60 80 100 Municipality 120 140 160 180 200 Robin Hood Effect 200.00% % Var. mun. Coeficiente 150.00% 100.00% 50.00% 0.00% 0.00000 0.05000 0.10000 0.15000 0.20000 -50.00% -100.00% Mun. Coeficient 0.25000 0.30000 0.35000 0.40000 Robin Hood Effect 350 300 250 % Var. IQE 200 150 100 50 0 0 5000 10000 15000 -50 -100 Per-capita GDP 20000 25000 Gainers & Losers 2008 - 2009 - 2010 1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126 131 136 141 146 151 156 161 166 171 176 181 Gainers & Losers 200% Net G-L in the share. 150% 100% 50% -50% -100% Net G-L 0% Race to the Top Net Gain 2008-2010 Top 20 Abaiara 145,75% Pedra Branca 97,91% Milhã 87,85% Ibaretama 86,08% Brejo Santo 85,81% Mauriti 85,55% Campos Sales 74,58% Deputado Irapuan Pinheiro 70,45% Penaforte 69,44% Poranga 68,17% Salitre 67,12% Senador Sá 66,77% Ererê 63,18% Aiuaba 63,09% Piquet Carneiro 62,43% Catunda 60,84% Morrinhos 60,65% Granjeiro 60,21% Ipaumirim 53,15% Pacujá 50,16% Race to the Bottom Net Loss 2008-2010 Botton 20 Guaramiranga -64,69% Acarape -61,51% Aratuba -44,65% Groaíras -43,98% Itaiçaba -40,36% Independência -37,11% Irauçuba -36,03% Aracati -34,77% Juazeiro do Norte -31,33% Icapuí -31,06% São Benedito -30,85% Cruz -30,25% Fortim -29,97% Umirim -28,51% Mucambo -28,47% Capistrano -28,10% Ubajara -27,67% Parambu -27,61% Ibicuitinga -25,15% Acopiara -24,76% Convergence in Scores 80 % Var Score 60 40 20 0 120 140 160 180 200 220 240 -20 -40 Score MAT 70 60 % Var. Score 50 40 30 20 10 0 -10 120 -20 130 140 150 160 170 Score PORT 180 190 200 210 Divergence in Standard Deviation 80 % Var SD MAT 60 40 20 0 -20 20 25 30 35 40 45 50 55 60 65 -40 -60 SD MAT 50 40 % Var SD PORT 30 20 10 0 -10 20 25 30 35 40 -20 -30 SD PORT 45 50 55 Conclusions • The distribution of resources became more homogeneous and smaller municipalities gained a chance to increase their revenue. • In education, there have been improvements in students performances, but at a cost of greater inequality. • Some municipalities behaved strategically and took advantage of the new law for significant gains in their budgets. Conclusions • To reduce the risk of “poverty trap” in some municipalities the state has an important role in “explain” the New Law to society and municipal governments. • It seems that the main goal of the law is being achieved. As each municipality compete with each other’s to obtain better performances they all advance and get close in a higher level of performance. References • Marcos Costa Holanda, Marcelo Ponte Barbosa, Leandro Oliveira Costa, Metodologia de Cálculo da Nova Lei do ICMS Municipal, NOTA TÉCNICA IPECE n º 33. http://www.ipece.CE.gov.br/publicacoes/notas_tecnicas/NT_33. PDF • Marcos Costa Holanda, Marcelo Ponte Barbosa, Leandro de Oliveira Costa, Memória de Cálculo dos Coeficientes de Distribuição do ICMS Municipal 2009, NOTA TÉCNICA IPECE n. º 32. http://www.ipece.CE.gov.br/publicacoes/notas_tecnicas/NT_32. PDF THANKS!!! [email protected]