Fiscal Policy, Inequality and the Poor in the Developing World Nora Lustig Tulane University Nonresident Fellow CGD and IAD National Treasury and World Bank Workshop Fiscal Policy and Redistribution in an Unequal Society Pretoria, South Africa, November 5, 2014 When using material in this ppt please cite as: • Lustig, Nora. 2014. “Fiscal Policy, Inequality and the Poor in the Developing World. Round 1.” CEQ Working Paper No. 23, Center for Inter-American Policy and Research and Department of Economics, Tulane University and InterAmerican Dialogue, forthcoming. www.commitmentoequity.org 2 CEQ Teams (Year of Survey; C=consumption & I=income)(MWB Version) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 10 11 12 13 14 15 16 17 Argentina (2009, I): Nora Lustig and Carola Pessino (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3 Armenia (2011; I): Stephen Younger and Artsvi Khachatryan (May 31, 2014; paper) Bolivia (2009; I): Veronica Paz Arauco, George Gray-Molina, Wilson Jimenez and Ernesto Yañez (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3 Brazil (2009; I): Sean Higgins and Claudiney Pereira (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3 Chile (2009, I): Jaime Ruiz-Tagle and Dante Contreras (Oct. 25, 2014) Colombia (2010, I): Marcela Melendez, Nora Lustig and Valentina Martinez (May 2014) Costa Rica (2010; I): Pablo Sauma and Juan Diego Trejos (February 2014; paper) El Salvador (2011; I): Margarita Beneke, Nora Lustig and Jose Andres Oliva (March 11, 2014) Ethiopia (2010/11; C): Ruth Hill, EyasuTsehaye, Tassew Woldehanna (Sept. 28, 2014) Guatemala (2011; I): Maynor Cabrera, Nora Lustig and Hilcias E. Moran (August 27, 2014) Indonesia (2012; C) : Jon Jellema and Matthew Wai-Poi (Sept. 9, 2014) Jordan (2010; C) : Morad Abdel-Halim, Shamma Adeeb Alam, Yusuf Mansur, Umar Serajuddin, Paolo Verme (May 16, 2014) Mexico (2010; I): John Scott (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3 Peru (2009; I): Miguel Jaramillo (CEQ Web Dec 2013) Public Finance Review, May 2014, Volume 42, Issue 3 South Africa (2010; I): Ingrid Woolard, Precious Zikhali, Mashekwa Maboshe, Jon Jellema (Aug. 25, 2014) Sri Lanka (2009/10; C): Nisha Arunatilake, Gabriela Inchauste and Nora Lustig (April 8, 2014; paper) United States (2011; I): Sean Higgins, Nora Lustig, Whitney Ruble and Timothy Smeeding (paper Oct. 2014) Uruguay (2009; I): Marisa Bucheli, Nora Lustig, Maximo Rossi and Florencia Amabile (CEQ Web Dec 2013) 3 Public Finance Review, May 2014, Volume 42, Issue 3 Handbook • Lustig, Nora and Sean Higgins. 2013. Commitment to Equity Assessment (CEQ): Estimating the Incidence of Social Spending, Subsidies and Taxes. Handbook. CEQ Working Paper No. 1, Center for Inter-American Policy and Research and Department of Economics, Tulane University and InterAmerican Dialogue, September. 4 Commitment to Equity Assessments (CEQ) • Accounting Approach: no behavioral, no general equilibrium effects and no intertemporal effects • Point-in-time • Mainly average incidence; a few cases with marginal incidence • Comprehensive standard fiscal incidence analysis of current systems • Harmonized definitions and methodological approaches to facilitate cross-country comparisons • Uses income/consumption per capita as the welfare indicator • Tax shifting assumptions are the standard ones • Allocators vary => full transparency in the method used for each category, tax shifting assumptions, tax evasion • Secondary sources are used to a minimum 5 • Handbook (Lustig and Higgins, 2013) 6 Basic elements of standard fiscal incidence • Before taxes and transfers income of unit h, or Ih • Taxes Ti – personal income taxes; contributions to social security – consumption and production taxes and subsidies • Transfers Ri – social spending: cash & near-cash transfers; in-kind transfers (education and health) – consumption and production (agriculture) subsidies • “Allocators” of tax i and transfer j to unit h, or Sih , Sjh (the share of tax i borne or transfer j received by unit h) => Incidence • Post-taxes and transfers income of unit h (Yh) 7 • Post-taxes and transfers income of unit h (Yh) is: Taxes & Transfers Post-fisc Income Yh = Ih - ∑i TiSih + ∑j RjSjh Pre-fisc Income Incidence of Taxes & Transfers 8 MARKET INCOME PLUS DIRECT TRANSFERS MINUS DIRECT TAXES GROSS INCOME NET MARKET INCOME MINUS DIRECT TAXES PLUS DIRECT TRANSFERS DISPOSABLE INCOME MINUS NET INDIRECT TAXES Construction of Income Concepts POST-FISCAL INCOME PLUS MONETIZED VALUE OF PUBLIC SERVICES: EDUCATION & HEALTH FINAL INCOME 9 Methods to Construct Income Concepts • Direct Identification Method • Imputation Method – Direct (Education and Health) – Simulation (Direct and Indirect Taxes) • Inference Method • Alternate Survey • Secondary Sources Method 10 Reconciling the Two Economies: Survey Data vs. Administrative Accts. • What to do when totals in Survey do not match administrative accounts? • Should imputed values be scaled-down or the rest of the concepts scaled up? 11 Contributory Pensions • Are they a government transfer or deferred consumption and hence part of market income? – No consensus – Results, especially for poverty, are extremely sensitive => Do it both ways 12 Importance of Comprehensive Analysis • Obvious reason: to capture the full effect of the net fiscal system • More subtle reason: partial assessments of progressivity and regressivity can be misleading => a regressive tax can be equalizing and re-inforce the equalizing impact of transfers 13 Lambert’s Conundrum 1 2 3 4 Original income x 10 20 30 40 Tax Liability t(x) 6 9 12 15 Benefit level b(x) 21 14 7 0 Post-benefit income 31 34 37 40 Final income 25 25 25 25 Source: Lambert, 2001, Table 11.1, P. 278 Total 100 42 42 142 100 14 Lambert’s Conundrum • The Reynolds-Smolensky (R-S) index for taxes in this example is equal to -0.0517, highlighting their regressivity. • Yet, the R-S for the net fiscal system is 0.25, higher than the R-S for benefits equal to 0.1972. • If taxes are regressive vis-à-vis the original income but progressive with respect to the less unequally distributed post-transfers (and subsidies) income, => regressive taxes exert an equalizing effect over an above the effect of progressive transfers. • Two renowned studies found this type of result in the US and the UK. 15 Results Redistribution and Inequality Reduction 16 Redistribution in the rich and developing countries Change in Gini: Disposable vs. Market (in GINI points) Ireland Belgium United Kingdom Finland Luxembourg France Germany Netherlands Austria Portugal Denmark Slovenia Sweden Spain South Africa(2010) Czech Republic Malta US(2011) Hungary Italy Slovakia Latvia Romania Estonia Cyprus Lithuania Greece Brazil(2009) Chile(2009) Uruguay(2009) Bulgaria Poland Mexico(2010) Costa Rica(2010) Peru(2009) Bolivia(2009) El Salvador (2011) Colombia(2010) -0.05 Guatemala(2010) 0.00 -0.10 -0.15 -0.20 -0.25 -0.30 Sources: EUROMOD for EU,Higgins et al. (2014) for US and for CEQ countries see Lustig (2014) and references at the end. Note: in these calculations contributory pensions are part of market income and NOT treated as a government transfer. 17 Redistribution in Middle and Low Income Countries: CEQ 16 Change in Gini: Disposable vs. Market (in GINI points) South Africa(2010) Chile(2009) Brazil(2009) Uruguay(2009) Mexico(2010) Costa Rica(2010) Peru(2009) Bolivia(2009) -0.03 El Salvador (2011) -0.02 Colombia(2010) -0.01 Guatemala(2010) 0.00 -0.04 -0.05 -0.06 -0.07 -0.08 -0.09 18 The impact of direct taxes and transfers on inequality (Gini coefficient): CEQ 16 Changes in Gini: Disposable VS Market Income 0.8 16000 14000 0.7 12000 10000 0.6 8000 0.5 6000 4000 0.4 2000 0.3 0 Mexico(2010) Chile(2009) Colombia(2010) Brazil(2009) South Africa(2010) Gini of Disposable Income Gini of Market Income GNI per capita (2005 PPP) 19 The impact of net indirect taxes on inequality (Gini coefficient): CEQ 16 Changes in Gini Coef icients 0.9 16000 0.8 14000 0.7 12000 0.6 10000 0.5 8000 0.4 6000 0.3 4000 0.2 2000 0.1 0.0 0 SA Bra Col Chi (2010) (2009) (2010) (2009) Gini of Market Income Mex (2010) Gini of Disposable Income Gini of Post- iscal GNI per capita (2005 PPP) 20 Lindert’s (2006) historical result is also found in cross section: Higher GDP/capita, more redistribution Change in Gini: Post-fiscal vs. Market (decline in Gini points shown in posi ve quadrant) 0.09 South Africa 0.08 0.07 0.06 0.05 CHL 0.04 BRA 0.03 MEX CRI 0.02 URY PER SLV 0.01 COL 0.00 0 2000 BOL GTM 4000 6000 8000 10000 GNI/Capita (2005 PPP) 12000 14000 16000 21 However, no Robin Hood Paradox And results do not depend on South Africa Change in Gini points: Post-fiscal vs. Market (decline in Gini points shown in posi ve quadrant) 0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0.00 Sputh Africa URY SLV 0.30 0.35 0.40 0.45 MEX CHL BRA CRI PER BOL COL GTM 0.50 0.55 0.60 Mkt Income Gini 0.65 0.70 0.75 0.80 22 Results Redistribution and Poverty Reduction 23 Direct Transfers (net of direct taxes) reduce poverty (except in Ethiopia): CEQ 16 Change in Headcount Ratio ($2.5 PPP/Day): Disposable vs. Market Income (in percentage points) 2% 0) ) 01 ri ca (2 20 So ut h Af az i l( 20 Br Ur ug ua y( (2 ia 09 ) 09 9) 00 ) 10 20 Bo ic ex M li v o( (2 or ad lv Sa El 01 1) 9) 00 (2 0) Ch ile 01 ) (2 10 ca 20 Co st a Ri a( al m ua te Pe ru (2 00 10 G -6% Co lo m bi a( 20 -2% -4% 9) ) 0% -8% -10% -12% -14% 24 Indirect Taxes increase poverty over and above market income poverty in six out of the CEQ 16 countries 25 Changes in Headcount Ratio: Post- iscal vs. Market Income (Poverty Line: US$2.50ppp/day) 0.9 16000 0.8 14000 0.7 0.6 0.5 12000 South Africa 10000 8000 0.4 0.3 6000 Brazil 4000 0.2 Headcount of Market Income Headcount of Disposable Income Headcount of Post- iscal Chi(2009) Mex(2010) Bra (2009) 0 Peru (2009) 0.0 Col (2010) 2000 SA (2010) 0.1 GNI per capita (2005 PPP) 26 Note that Net Indirect Taxes can be equalizing and yet poverty increasing: Ethiopia Change in Gini: Marginal Contribu on of Net Indirect Taxes Change in Headcount Ra o ($2.5 PPP/Day): Marginal Contribu on from Net Indirect Taxes (in percentage points) Gu at Ind Eth iop ia em al Bo liv a(20 ia( 10 20 ) In 09 do ) ne Sr sia(2 iL an 012 k ) Co a(2 00 lo m 9) bi a El Sa (201 lv a 0) Pe d or ru (2 (2 01 00 1) Et 9) hi op i Co a(20 st a R 11) ic Jo rd a(20 an 10 (2 01 ) M ex 0 ico ) ( 2 Ar m 010 en ) i Ur a(20 ug 11 ua ) y ( Br az 200 9) il( 2 Ch 009 ) ile (2 00 So ut 9) h A (20 11) one sia (20 12) Sr i Lan ka( 200 9) Per u(2 009 ) Col om b ia (20 10) Gu ate ma la(2 010 Cos ) ta R ica (20 10) Chi le(2 009 ) El S alv ad o r (2 011 Jor ) dan (20 10) Me xico (20 10) Bol ivia (20 09) Arm eni a( 2 011 ) Uru gua y(2 009 ) Bra zil( 200 9) Sou th Afr ica (20 10) (in GINI points) 8% 0.01 0.00 -0.01 -0.02 -0.03 -0.04 -0.05 -0.06 -0.07 -0.08 -0.09 6% 4% 2% 0.045 0.026 0.036 0% -2% 0.006 -0.007 -0.015 0.012 0.006 0.006 -0.002 -0.012 -0.016 -0.024 -4% -0.019 -0.028 -6% -8% -0.072 Contribu on from Net Indirect Taxes Change in Headcount: Post-Fiscal vs Mkt Contribu on from Net Indirect Tax Gini Change: Post-fiscal vs Mkt 27 Thank you! 28