Fiscal Policy, Inequality and the Poor in the Developing World Nora Lus<g Tulane University Nonresident Fellow CGD and IAD Na<onal Treasury and World Bank Workshop Fiscal Policy and Redistribu3on in an Unequal Society Pretoria, South Africa, November 5, 2014 When using material in this ppt please cite as: •  Lus<g, 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 Inter-­‐
American Dialogue, forthcoming. www.commitmentoequity.org 2 CEQ Teams (Year of Survey; C=consumpBon & I=income)(MWB Version) 1. 
2. 
3. 
4. 
5. 
6. 
7. 
8. 
9. 
10. 
10 
11 
12 
13 
14 
15 
16 
17 
ArgenBna (2009, I): Nora Lus<g 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 LusBg and ValenBna MarBnez (May 2014) Costa Rica (2010; I): Pablo Sauma and Juan Diego Trejos (February 2014; paper) El Salvador (2011; I): Margarita Beneke, Nora Lus<g 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 Lus<g and Hilcias E. Moran (August 27, 2014) Indonesia (2012; C) : Jon Jellema and Mafhew 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 Scof (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 Aruna<lake, Gabriela Inchauste and Nora Lus<g (April 8, 2014; paper) United States (2011; I): Sean Higgins, Nora Lus<g, Whitney Ruble and Timothy Smeeding (paper Oct. 2014) Uruguay (2009; I): Marisa Bucheli, Nora Lus<g, Maximo Rossi and Florencia Amabile (CEQ Web Dec 2013) 3 Public Finance Review, May 2014, Volume 42, Issue 3 Handbook •  Lus<g, Nora and Sean Higgins. 2013. Commitment to Equity Assessment (CEQ): Es3ma3ng 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 Inter-­‐American Dialogue, September. 4 Commitment to Equity Assessments (CEQ) •  Accoun<ng Approach: no behavioral, no general equilibrium effects and no intertemporal effects •  Point-­‐in-­‐<me •  Mainly average incidence; a few cases with marginal incidence •  Comprehensive standard fiscal incidence analysis of current systems •  Harmonized defini<ons and methodological approaches to facilitate cross-­‐country comparisons •  Uses income/consump<on per capita as the welfare indicator •  Tax shining assump<ons are the standard ones •  Allocators vary => full transparency in the method used for each category, tax shining assump<ons, tax evasion •  Secondary sources are used to a minimum 5 •  Handbook (Lus<g 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; contribu<ons to social security –  consump<on and produc<on taxes and subsidies •  Transfers Ri –  social spending: cash & near-­‐cash transfers; in-­‐kind transfers (educa<on and health) –  consump<on and produc<on (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 Construc<on of Income Concepts POST-­‐FISCAL INCOME PLUS MONETIZED VALUE OF PUBLIC SERVICES: EDUCATION & HEALTH FINAL INCOME 9 Methods to Construct Income Concepts •  Direct Iden+fica+on Method •  Imputa+on Method –  Direct (Educa+on and Health) –  Simula+on (Direct and Indirect Taxes) •  Inference Method •  Alternate Survey •  Secondary Sources Method 10 Reconciling the Two Economies: Survey Data vs. AdministraBve Accts. •  What to do when totals in Survey do not match administra<ve 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 consump<on and hence part of market income? –  No consensus –  Results, especially for poverty, are extremely sensi<ve => Do it both ways 12 Importance of Comprehensive Analysis •  Obvious reason: to capture the full effect of the net fiscal system •  More subtle reason: par<al 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, highligh<ng 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 RedistribuBon and Inequality ReducBon 16 RedistribuBon in the rich and developing countries Change'in'Gini:'Disposable'vs.'Market'
(in'GINI'points)'
Ireland
Belgium
United Kingdom
Finland
Luxembourg
France
Netherlands
Germany
Austria
Portugal
Denmark
Slovenia
Sweden
South Africa(2010)
Spain
US(2011)
Czech Republic
Hungary
Malta
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 Lus<g (2014) and references at the end. Note: in these calcula<ons contributory pensions are part of market income and NOT treated as a government transfer. 17 RedistribuBon 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)--
!0.04%
Bolivia(2009)--
!0.03%
El-Salvador-(2011)--
!0.02%
Colombia(2010)-
!0.01%
Guatemala(2010)--
0.00%
!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"CoefAicients
0.9"
16000"
0.8"
14000"
0.7"
12000"
0.6"
10000"
0.5"
8000"
0.4"
6000"
0.3"
4000"
0.2"
0.1"
2000"
0.0"
0"
SA"
Bra" Col"
Chi"
(2010)"(2009)"(2010)"(2009)"
Gini"of"Market"Income"
Mex"
(2010)"
Gini"of"Disposable"Income"
Gini"of"PostKAiscal"
GNI"per"capita"(2005"PPP)"
20 Lindert’s (2006) historical result is also found in cross secBon: Higher GDP/capita, more redistribuBon 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$
0.01$
COL$
0.00$
0$
2000$
SLV$
BOL$
4000$
GTM$
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:'Post0fiscal'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$
0.30$
Sputh$Africa$
URY$
MEX$
SLV$
0.35$
0.40$
0.45$
CRI$
PER$
BOL$
CHL$
BRA$
COL$
GTM$
0.50$
0.55$
0.60$
Mkt$Income$Gini
0.65$
0.70$
0.75$
0.80$
22 Results RedistribuBon and Poverty ReducBon 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)
ri
ca
(2
01
09
20
ut
h
Af
az
il(
20
Br
So
Ur
ug
ua
y(
(2
ia
)
)
09
9)
00
)
10
20
Bo
ic
ex
M
liv
o(
(2
or
ad
lv
Sa
El
01
1)
9)
00
(2
0)
Ch
ile
01
)
(2
10
ca
Ri
a
st
Co
te
m
al
a(
20
00
(2
ru
ua
20
a(
bi
m
Pe
G
!6%%
Co
lo
!4%%
10
!2%%
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 Change&in&Headcount&RaHo&($2.5&PPP/Day)&
(in%percentage%points)
6%%
4%%
2%%
0%%
a(
20
Br
10
az
)&
Ur il(20
ug
09
ua
)&&
y(
20
09
)&&
Bo
liv
ia
(2
M
ex 009
)&&
ico
(2
01
0)
El
&&
&Sa
lv
ad
or
&(2
0
Ch 11)
&&
ile
Co
(
st
20
a&
09
Ri
)&
ca
Gu
(2
at
em 010
)&&
al
a(
20
10
Pe
)&&
ru
(2
00
9)
&&
Co
lo
m
bi
a(
20
10
)&
!2%%
ric
!4%%
So
ut
h&
Af
!6%%
!8%%
!10%%
!12%%
!14%%
Disposable%vs.%Market%
Post!fiscal%vs.%Market%
25 Changes#in#Headcount#Ratio:#Post@Aiscal#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"
6000"
Brazil 0.3"
4000"
0.2"
Headcount"of"Market"Income"" Headcount"of"Disposable"Income" Headcount"of"PostCDiscal"
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'Contribu2on'of'Net'Indirect'
Taxes'
Change'in'Headcount'Ra0o'($2.5'PPP/Day):'Marginal'Contribu0on'from'Net'Indirect'Taxes'''
(in%percentage%points)%%
Gu
a
8%%
6%%
4%%
2%%
0.045%
0.026%
0.036%
0%%
&2%%
0.006%
&0.015%
&0.007%
0.006%
&0.012%
&0.016%
&4%%
0.012%
0.006%
&0.002%
&0.024%
&0.019%
&0.028%
&6%%
&8%%
&0.072%
ContribuUon%from%Net%Indirect%Taxes%
Change%in%Headcount:%Post&Fiscal%vs%Mkt%
Bo
l
Ind
Eth
i op
te
m
ala
(2
ivi
01
a(
0)
2
00
%%
In
do
9)
%%
ne
S r s i a( 2
i %L
an 012
k
) %%
Co a(2
00
lo
m
9)
bi
%
a
El
%Sa (20
1
lva
0)
Pe dor %
%( 2
ru
(2
01
00
1)
Et
9)
%%
hi
%
%
op
ia(
Co
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
u t 9)%
h%
A
ia(2
011
)%
on e
si a
(20
12)
Sri%
%%
Lan
ka(
200
9)%
Per
u(2
009
)%
Co l
om
bia
(20
10)
%
Gu
ate
ma
la(2
010
Co s
) %%
ta%R
ica
(20
10)
Chi
%%
le(2
009
)%
E l %S
alv
ado
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)%
S ou
th%
Afr
ica
(20
10)
%%
(in%GINI%points)
0.01%
0.00%
!0.01%
!0.02%
!0.03%
!0.04%
!0.05%
!0.06%
!0.07%
!0.08%
!0.09%
ContribuUon%from%Net%Indirect%Tax%
Gini%Change:%Post!fiscal%vs%Mkt%
27 Thank you! 28 

PREACHING PLAN - SOUTH DURBAN CIRCUIT

NumMoto IDCategoria NombreApellido FechaNac

www.baNdeNbeurS.Nl