1. Redistributive Impact and Efficiency of Mexico's Fiscal System John Scott, CIDE

2. Policy Context Over the last two decades a series of policy reforms have contributed to increase the redistributive impact of social spending: – reallocation of public spending to social programs (from 25% of primary spending in the 1980s to 49% in the 2000s), – introduction of innovative and effectively targeted transfer programs, including the Progresa/Oportunidades CCT, and – a general reorientation of basic education services, health services for the uninsured and food subsidies from urban to poor rural areas. A second wave of reforms over the last decade increased public spending on non-contributive social protection programs benefiting the poor (though with the collateral effect of subsidizing informality), including basic pension programs (Adultos Mayores, 70 y más) and health insurance (Seguro Popular ). The most progressive spending categories at present include most direct transfers and food programs, pre-school and primary education services, and health services for the uninsured

3. Results However, the redistributive impact of Mexico's fiscal system still falls well below the country's redistributive potential as measured by its per capita income level, the quality (and redistributive effectiveness) of its best social programs, and the achievements of comparable countries in the region. Mexico's per capita income is similar to Argentina's and higher than Brazil's, but social spending as a share of GDP (10%) is less than half of Argentina's and 60% of Brazil's. The redistributive effectiveness of this spending in Mexico (1.39) is lower than Argentina (2.12) and similar to Brazil (1.37): the redistributive impact of social spending in Mexico (14.4%) is 10 and 6 percentage points lower than Argentina's and Brazil's, respectively.

4. Results Mexico's direct transfer programs are just as effective in reducing extreme poverty (2.5 USD PPP) as Argentina's, and more than twice as effective as Brazil's. However, Mexico spends only 1% of GDP on such transfers, while Argentina and Brazil spend 3.7% and 4.2%, respectively, so while these transfers reduce extreme poverty by 58% and 28.5% in the latter countries, they do so by only 15% in Mexico. The failure of Mexico's fiscal system to achieve its redistributive potential is explained by three principal characteristics of this system: – a progressive but unproductive tax system (non-oil tax revenues: 11.6% of GDP in 2010), – low spending on direct transfers, and – a significant share of fiscal resources allocated to instruments with limited redistributive effectiveness, including energy subsidies and subsidies to contributive social security.

5. Results A wide range of concentration coefficients across as well as within the principal policy dimensions analyzed (education, health, social security, income support, food programs) reveals significant opportunities to improve the system's overall redistributive impact. The persistence of ineffective redistributive instruments absorbing large shares of fiscal resources may be explained by two principal factors: – capture of benefits by organized interest groups (agricultural subsidies, social security subsidies for public sector workers). – economic costs to access public benefits in a context of high (market) income inequality: explicit or implicit user fees, social security contributions, labor opportunity costs of attending public higher education, purchasing power for subsidized goods and services. Population dispersion acts as a third barrier contributing to both income and fiscal inequality.

6. Table 2 Redistributive Effects and Effectiveness of the Fiscal System: Gini Coefficient and Extreme Poverty Headcount Index Gini Poverty Headcount Index 2.5 USD PPP LBM 2008201020082010 Nat. UrbanRuralNat. UrbanRural Nat. UrbanRural Market income0.52930.51070.48060.519712.2%12.6%5.6%35.6%18.3%12.9%36.1% Net Mkt Income0.51710.49750.46660.512312.4%12.6%5.7%35.8%18.4%13.0%36.2% Disposable Income0.50940.48760.46220.481010.8%10.7%4.8%30.2%16.4%12.0%30.9% Post-fiscal Income0.50020.48090.45900.476110.2%10.7%4.8%30.1%16.9%12.6%30.9% Final Income0.44810.42940.41490.3930 -.- Change with respect to Market Income Net Mkt Income-2.3%-2.6%-2.9%-1.4%1.4%0.6%1.2%0.3%0.5%0.9%0.1% Disposable Income-3.8%-4.5%-3.8%-7.4%-11.2%-14.9%-14.6%-15.1%-10.3%-6.8%-14.5% Post-fiscal Income-5.5%-5.8%-4.5%-8.4%-16.2%-15.1%-14.5%-15.5%-7.8%-2.2%-14.3% Final Income-15.3%-15.9%-13.7%-24.4% Effectiveness a Direct transfers1.992.05 16.6716.04 11.23 Total transfers1.421.39 -.- Source: Author's estimates using data from ENIGH 2008, 2010, and Cuenta Pública 2008, 2010. a. Effectiveness is defined as the change in net market Gini associated with transfer divided by transfer share in GDP.

8. *Procampo, 70 y Más, programas estatales y DF de Adultos Mayores, PET, Becas escolares, PAL, Liconsa, Diconsa, Desayunos Escolares y otros programes sociales menores. The distribution of direct transfers and non- contributive social security benefits is pro-poor…

9. …but these transfers are dwarfed by energy subsidies and subsidies to contributory social security concentrated on higher income households through

10. Table 3 Incidence of Taxes, Monetary and In-kind Transfers, and Indirect Subsidies by Decile Nationally (with Respect to Market Income): 2010 Deciles Net benefitsTaxesBenefits TotalMonetary a TotalDirect b Indirect c,d Indirect Subsidies Direct TransfersIn-kind Transfers Benchmark : informal consumption 1137.932.1-6.4-0.1-6.37.231.4105.8 254.08.1-5.1-0.4-4.84.68.645.9 335.32.6-6.1-0.7-5.43.65.132.8 425.10.0-6.3-0.8-5.43.42.825.1 517.5-2.8-7.8-1.6-6.23.01.920.3 612.1-4.2-8.2-2.1-6.12.71.316.3 76.5-5.7-8.9-2.9-62.40.912.2 81.6-7.5-10.3-4-6.32.20.69.0 9-3.2-9.4-11.6-5.4-6.21.70.46.3 10-10.5-12.7-14-7.9-61.00.22.2 All3.1-7.7-11-5-62.01.310.8 Sensitivity analysis: no informality d 1128.122.4-16.1-16.0 249.33.4-9.8-9.4 332.3-0.4-9.1-8.5 422.9-2.2-8.5-7.6 516.3-4.0-9.0-7.4 611.2-5.1-9.1-7.0 75.9-6.3-9.5-6.6 81.5-7.6-10.4-6.4 9-3.0-9.2-11.3-6.0 10-9.7-11.9-13.1-5.2 All2.8-8.0-11.3-6.3 Source: Author's estimates using data from ENIGH 2008, Cuenta Pública 2008, SHCP (2010, 2012). a. Net monetary benefits are direct transfers and indirect subsidies net of dierct and indirect taxes. b. Direct taxes include contributions to social security. c. Benchmark: all rural consumption and informal urban consumption (informal markets) is assumed not to pay indirect taxes. VAT and excise taxes are imputed based on the SHCP (2012) decile estimates using this definition of informality. d. Indirect tax sensitivity analysis: all consumers are assumed to pay indirect taxes. VAT, excise taxes, and two special taxes on cars (Tenencia, ISAN) are imputed separately based on the SHCP (2012) decile estimates without assuming informality.