Introduc)on to CEQ: The Commitment to Equity

TrainingWorkshopontheCommitmenttoEquityMethodology

CEQIns;tuteandTheMinistryofFinanceAccra

February7-10,2017

Outline of the Workshop

•  Day1:•  Introduc;ontoCEQAssessments•  Illustra;onwithcountryresults,includingGhana•  Introduc;ontoStata

•  Day2:•  Introduc;ontotheLivingStandardsSurveydata•  Construc;ngCEQincomeconceptsandtheircomponents

•  Day3:•  Construc;ngCEQincomeconceptsandtheircomponents•  Themasterworkbook•  Genera;ngandinterpre;ngresults

•  Day4:•  Cross-checkingresults•  Policysimula;ons

•  Agendaavailableatc:\CEQ_Ghana_training\presenta;ons\CEQ-MOF_Training_Agenda_Feb_2017.docx

Workshop Materials

•  TheCEQHandbook•  Lus;g,Nora,editor,CommitmenttoEquityHandbook.AGuidetoEs6ma6ngtheImpactof

FiscalPolicyonInequalityandPoverty,CEQIns;tute,TulaneUniversity,forthcoming.hYp://www.commitmentoequity.org/publica;ons/handbook.php

•  PreliminaryStataskills•  CarolinaPopula;onCenterStataIntroduc;on:hYp://www.cpc.unc.edu/research/tools/data_analysis/statatutorial/index.html•  GermánRodriguez’StataIntroduc;on:hYp://data.princeton.edu/stata/•  ChristopherBaum’sStataIntroduc;on:hYp://fmwww.bc.edu/GStat/docs/StataIntro.pdf•  Stata’sdocumenta;on(alsoavailableonyourpcaspdf’swhenyouinstallStata):hYp://www.stata.com/features/documenta;on/

•  Introduc;ontotheGhanaLivingStandardsSurvey,round5•  c:\CEQ_Ghana_training\GLSS_5_docs\*.pdf

Commitment to Equity Ins)tute (CEQ Ins)tute)

Objec&ve:Tomeasuretheimpactoffiscalpolicyoninequalityandpovertyincountriesacrosstheworld

•  Research-basedpolicytools•  CEQDataCenteronFiscalRedistribu;on•  CEQAdvisoryandTrainingServices•  BridgestoPolicy

Ø TwograntsfromBill&MelindaGatesFounda;onfor2014-2020

Ø www.commitmentoequity.org4

CEQ Ins)tute: Core Staff

• Director:NoraLus;g• DirectorofPolicyArea:LudovicoFeoli• AssociateDirectors:MaynorCabrera,JonJellema,EstuardoMoranandStephenYounger• DataCenterDirectors:SeanHigginsandSandraMar;nez•  Communica&onsDirector:CarlosMar;ndelCampo• MasterdataCoordinator:IsraelMar;nezInaddi;on:• AdvisoryBoard• NonresidentResearchAssociates(morethan40worldwide)

5

6www.commitmentoequity.org

Commitment to Equity Ins)tute

• Workingoncloseto40countries;coversaroundtwothirdsoftheworldpopula;on•  Collabora;veeffortsandpartnershipswithmul;pleorganiza;ons:ADB,AfDB,CAF,ERF,IDB,IMF,ICEFI,OECD,Oxfam,UNDP,WorldBank•  U;lizedbygovernments•  Publica;ons:Handbook,WorkingPaperseries,scholarlypublica;onsinpeer-reviewedjournals,bookchapters,editedvolume(inprogress),blogsandpolicybriefs• Websitewww.commitmentoequity.org

7

CEQ Assessment •  Purpose:

•  Provideastandardizedanalysisoftheextenttowhichtaxesandsocialexpendituresalterthedistribu;onofincome

•  Includeseffectsonpovertyandinequality•  Allowsforpolicysimula;onsofproposedreforms

•  Standardiza;onallowsforcross-countrycomparisons

•  Tools•  CEQhandbook•  Themasterworkbookofresults(MWB)

•  CEQStatacommands•  CEQcross-checkingprotocolsforqualitycontrol

8

CEQ Assessment – Conceptual Framework •  Governmenttaxa;onandsocialspendinginfluencethedistribu;onofincomeinmanyways•  Directeffectsondisposableincomes(e.g.PAYE,LEAP)•  Indirecteffectsonpurchasingpower–pricechanges(e.g.VAT,excises,petrolandelectricitysubsidies)

•  Provisionoffreeorsubsidizedservices(e.g.educa;onandhealth)

•  ThegeneralapproachofCEQistodescribehoweachofthesepolicieschangesthedistribu;onofincome,inasystema;cway

•  SeenextslideforCEQ“incomeconcepts”

9

MARKETINCOME

DISPOSABLEINCOME

PLUSDIRECTTRANSFERSMINUSDIRECTTAXES

PLUSINDIRECTSUBSIDIESMINUSINDIRECTTAXES

CONSUMABLEINCOME

PLUSMONETIZEDVALUEOFPUBLICSERVICES:EDUCATION&HEALTH

FINALINCOME

CEQ Assessment:IncomeConcepts

10

HigginsandLus;g.“AAlloca;ngTaxesandTransfers,Construc;ng Income Concepts, and Comple;ngSec;on C of CEQ Master Workbook” in Lus;g(editor) Commitment to Equity Handbook. A Guideto Es6ma6ng the Impact of Fiscal Policy onInequalityandPoverty,TulaneUniversity,Fall2016.

CEQ Assessment – Results for Ghana

• BasedonaCEQstudyusingGLSS-6data(Younger,Osei-Assibey,andOppong,2016)•  ForeachCEQincomeconcept,wecalculateGinicoefficientsandFGTpovertymeasures•  Foreachsocialexpenditureandtax,wecalculateconcentra;oncoefficients

11

Taxes Expenditures

Direct Taxes Direct Transfers

PAYE LEAP (simulated)

Presump)ve taxes (informal) School feeding program

Presump)ve taxes (formal) Pensions*

Indirect Taxes Indirect Transfers

VAT Electricity subsidies

Import du)es Fer)lizer subsidies

Cocoa du)es Kerosene cross-subsidy

Excises

Petroleum products In-Kind Benefits

Beverages Public schooling (various levels)

Tobacco products Public health services, inpa)ent

Communica)ons services Public health services, outpa)ent

What’s Included in the Study?

BasicResults

povertyline: GH₵1314peryearGH₵792peryear

GiniHeadcount

index PovertyGapHeadcount

index

MarketIncome+Pensions 0.437 0.240 0.078 0.083

GrossIncome 0.436 0.238 0.076 0.081

DisposableIncome 0.424 0.242 0.078 0.084

Disp.Income+IndirectSubsidies 0.424 0.235 0.075 0.080

Disp.Income-IndirectTaxes 0.423 0.271 0.089 0.099

ConsumableIncome 0.423 0.261 0.085 0.094

Cons.Income+In-KindEduca;on 0.409 0.201 0.057 0.053

FinalIncome 0.402 0.186 0.051 0.046

AnExampleSimula;on Simula;on

Changein: (1) (2) (3) (4)

ExtremePoverty

Headcount

DisposableIncome -0.013 -0.007

ConsumableIncome 0.004 0.004 -0.011 -0.003

FinalIncome 0.001 0.001 -0.007 -0.003

PovertyHeadcount

DisposableIncome -0.022 -0.009

ConsumableIncome 0.009 0.005 -0.013 0.000

FinalIncome 0.008 0.005 -0.015 -0.002

PovertyGapDisposableIncome -0.010 -0.005

ConsumableIncome 0.003 0.002 -0.008 -0.002

FinalIncome 0.002 0.001 -0.006 -0.002

GiniDisposableIncome -0.009 -0.004

ConsumableIncome -0.001 0.000 -0.010 -0.005

FinalIncome -0.001 0.000 -0.009 -0.005

Budgetarysavings(shareofGDP): 0.014 0.007 0.000 0.008

CEQ Assessment – Results for 25 Countries

•  Twolow-incomecountries:Ethiopia(Hilletal.,2016)andTanzania(Youngeretal.,2016)

•  Ninelowermiddle-incomecountries:Armenia(YoungerandKhachatryan,2016),Bolivia(Paz-Araucoetal.,2014),ElSalvador(Beneke,Lus;gandOliva,2014),Georgia(CanchoandBondarenko,2016),Ghana(Youngeretal.,2015),Guatemala(Cabrera,Lus;gandMoran,2015),Honduras(CastañedaandEspino,2015),Indonesia(Awaretal.,2016),andSriLanka(Aruna;lakeetal.,2016)

•  Elevenuppermiddle-incomecountries:Brazil(HigginsandPereira,2014),Colombia(Lus;gandMelendez,2016),CostaRica(SaumaandTrejos,2014),DominicanRepublic(Aristy-Escuderetal.,2016),Ecuador(Llerenaetal.,2015),Jordan(Alametal.,2016),Mexico(ScoY,2014),Peru(Jaramillo,2014),Russia(Lopez-Calvaetal.,2016),SouthAfrica(Inchausteetal.,2016),andTunisia(Shimelesetal.,2016)

•  Twohigh-incomecountries:Chile(Mar;nez-Aguilaretal.,2016),andUruguay(Buchelietal.,2014).

•  Oneunclassified:Argen&na(Rossignolo,2016)

15

16

0.25

0.35

0.45

0.55

0.65

0.75

0.85

Marketincomepluspensions Disposableincome Consumableincome Finalincome

GiniCoefficient

Argen&na(2012) Armenia(2011) Bolivia(2009)Brazil(2009) Chile(2013) Colombia(2010)CostaRica(2010) DominicanRepublic(2013) Ecuador(2011)ElSalvador(2011) Ethiopia(2011) Georgia(2013)Ghana(2013) Guatemala(2011) Honduras(2011)Indonesia(2012) Jordan(2010) Mexico(2010)Peru(2009) Russia(2010) SouthAfrica(2010)SriLanka(2010) Tanzania(2011) Tunisia(2010)Uruguay(2009)

Marketincome(pluscontributorypensions)

Disposableincome Consumableincome

Finalincome

Fiscal Policy and Inequality – Contributory pensions as deferred income

Source:Lus;g(2016)

Redistribu)ve effect (Change in Gini points: market income plus pensions and market income to disposable income, circa 2010)

17Source:Lus;g(2016)

0.000.100.200.300.400.500.600.700.800.90

-0.20-0.18-0.16-0.14-0.12-0.10-0.08-0.06-0.04-0.020.00

Red

istribu&

veeffe

ct-m

arketincom

etodispo

sable

(rankedbyredistribu&veeffect(leghandscale);Ginicoefficientsrighthandscale)

Contributorypensionsasdeferredincome Contributorypensionsasdirecttransfer

Ginimarketincomepluspensions Ginimarketincome

More social spending, more redistribu)on

18Source:Lus;g(2016)

ARG

ARM BOL

BRA

CHLCOL

CRI

DOMECU

SLV

GHAGTM HND

IND JOR

MEX

PER

RUS

ZAF

LKA

TZA

TUN

URYy=0.9253x***-0.0287*

(7.05)(-1.99)R²=0.6933

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0% 5% 10% 15% 20% 25%

Redistrib

u&veeffe

ct

Socialspending

ETH

19

More unequal, more redistribu)on Consistent with Meltzer-Richard Median Voter Theorem - No “Robin Hood Paradox”

Source:Lus;g(2016)

ARG

ARM

BOL

BRACHI

COL

CRIDOMECU

SLVETH

GEO

GHA GTM

HNDINDJOR

MEX

PER

RUS

ZAF

LKA

TZATUN

URY

y=0.0952x*-0.0184(1.76)(-0.70)

R²=0.1183

0.00

0.04

0.08

0.12

0.16

0.20

0.20 0.35 0.50 0.65 0.80

Red

istribu&

veeffe

ct

Ginimarketincomepluspensions

In sum…

•  InNOcountry,inequalityincreasesasaresultoftaxes,subsidiesandsocialspending

Ø Fiscalpolicyisalwaysequalizing

Ø Themoreunequal,themorefiscalredistribu;on

20

•  Fiscalpolicycanbeequalizingbutpovertyincreasing(intermsofthepoor’sabilitytoconsumeprivategoodsandservices):Ø 1.25/dayline:Ethiopia,Ghana,Guatemala,Tanzania

Ø 2.50/dayline:Armenia,Bolivia,Ethiopia,Ghana,Guatemala,Honduras,SriLanka,Tanzania

Ø 4/dayline:alloftheaboveplusArgen;na,Brazil,CostaRicaandTunisia

•  Thisworrisomeresultstemsmainlyfromconsump;ontaxes

21

Poverty Impact

22Source:Lus;g(2016)

Fiscal Policy and Poverty Reduc)on (Change in Headcount Ratio from Market to Consumable Income (Poverty line $1.25 / day

2005 ppp; Contributory Pensions as Deferred Income; in %)

-0.2%-1.7%-3.3%-8.3%-10.3%-11.5%-16.1%-16.5%-18.0%-18.5%

-24.9%-31.8%-33.9%-34.6%

-44.9%-45.5%-45.9%-50.6%-56.4%

-69.6%-69.8%-70.7%

-88.2%-97.0%

-36.2%

17.8%13.3%

4.2%2.4%

-12.7%

-8.8%-2.3%

-14.1%-14.1%

-24.6%

-7.5%-15.7%

-35.0%

-53.8%

-37.8%

-22.7%

-53.8%

-36.2%-42.7%

-76.4%

-66.2%

-54.2%

-82.0%-82.6%

-29.4%

-120%

-100%

-80%

-60%

-40%

-20%

0%

20%

(rankedbypovertyreduc&onin%;povertyline$1.252005PPP/day)

Market income plus pensions to disposable income Market income plus pensions to consumable income

Fiscal Impoverishment

Analyzingtheimpactontradi;onalpovertyindicatorscanbemisleading

Ø Fiscalsystemscanshowareduc;oninpovertyandyetasubstan;alshareofthepoorcouldhavebeenimpoverishedbythecombinedeffectoftaxesandtransfers

23

HigginsandLus;g(2016)Canapoverty-reducingandprogressivetaxandtransfersystemhurtthepoor?JournalofDevelopmentEconomics122,63-75,2016

Fiscal Impoverishment (Market to Consumable Income)

24HigginsandLus;g(2016)

Country(surveyyear)

Marketincomeplus

pensionsPoverty

headcount(%)

Changeinpovertyheadcoun

t(p.p.)

Market

incomepluspensionsinequality(Gini)

Reynolds-Smolensky

Changeininequality(▲Gini)

Fiscally

impoverishedas%of

popula&on

Fiscally

Impoverishedas%

ofconsumableincomepoor

PanelA:Upper-middleincomecountries,usingapovertylineof$2.52005PPPperday

Brazil(2009) 16.8 -0.8 57.5 4.6 -3.5 5.6 34.9

Chile(2013) 2.8 -1.4 49.4 3.2 -3.0 0.3 19.2

Ecuador(2011) 10.8 -3.8 47.8 3.5 -3.3 0.2 3.2

Mexico(2012) 13.3 -1.2 54.4 3.8 -2.5 4.0 32.7

Peru(2011) 13.8 -0.2 45.9 0.9 -0.8 3.2 23.8

Russia(2010) 4.3 -1.3 39.7 3.9 -2.6 1.1 34.4

SouthAfrica(2010) 49.3 -5.2 77.1 8.3 -7.7 5.9 13.3

Tunisia(2010) 7.8 -0.1 44.7 8.0 -6.9 3.0 38.5

Brazil(2009) 16.8 -0.8 57.5 4.6 -3.5 5.6 34.9

Chile(2013) 2.8 -1.4 49.4 3.2 -3.0 0.3 19.2

Country(surveyyear)

Marketincomeplus

pensionsPoverty

headcount(%)

Changeinpovertyheadcoun

t(p.p.)

Marketincomeplus

pensionsinequality(Gini)

Reynolds-Smolensky

Changeininequality(▲Gini)

Fiscally

impoverishedas%of

popula&on

Fiscally

Impoverishedas%

ofconsumableincomepoor

PanelB:Lower-middleincomecountries,usingapovertylineof$1.252005PPPperday

Armenia(2011) 21.4 -9.6 47.4 12.9 -9.3 6.2 52.3

Bolivia(2009) 10.9 -0.5 50.3 0.6 -0.3 6.6 63.2

DominicanRepublic(2013) 6.8 -0.9 50.2 2.2 -2.2 1.0 16.3

ElSalvador(2011) 4.3 -0.7 44.0 2.2 -2.1 1.0 27.0

Ethiopia(2011) 31.9 2.3 32.2 2.3 -2.0 28.5 83.2

Ghana(2013) 6.0 0.7 43.7 1.6 -1.4 5.1 76.6

Guatemala(2010) 12.0 -0.8 49.0 1.4 -1.2 7.0 62.2

Indonesia(2012) 12.0 -1.5 39.8 1.1 -0.8 4.1 39.2

SriLanka(2010) 5.0 -0.7 37.1 1.3 -1.1 1.6 36.4

Tanzania(2011) 43.7 7.9 38.2 4.1 -3.8 50.9 98.6

25

Fiscal Impoverishment (Market to Consumable Income)

HigginsandLus;g(2016)

Main Messages of CEQ Work to Date

1.  Thepopularviewofmanypoliciesisozenwrong2.  Analyzingthetaxsidewithoutthespendingside,or

viceversa,canbemisleading3.  Analyzingtheimpactoninequalityonlycanbe

misleading4.  Analyzingtheimpactontradi;onalpoverty

indicatorscanbemisleading5.  Richercountriesredistributemore6.  Poorercountriesrelymostlyonsubsidizedservices7. Moreunequalcountriesredistributemore

26

Partnerships and Collabora)on

• Prepara;onofCEQAssessmentsinfullorcomponents• QualitycontrolofCEQAssessments•  Trainingworkshops• Advisoryservicesforstaffandgovernments• Cost-sharingarrangementsvarydependingonthecontribu;onsofpartneringorganiza;on

27