Inequality Outcomes vs Inequality Opportunitie Chile

7/28/2019 Inequality Outcomes vs Inequality Opportunitie Chile

1/18

Iul f ucm . Iul /Javier Nez, Andrea Tartakowsky 185

Eu Ecm. Vl. 34 - N 2, Dcmb 2007. Pg. 185-202

InequalIty of outcomes vs. InequalIty of

opportunItIes In a DevelopIng country.an exploratory analysIs for chIle1

desigualdad de resultados vs. desigualdad de

oportunidades en un pas en vas de desarrollo.

un anlisis exploratorio para chile

Javier Nez*Andrea Tartakowsky**

b

We apply a variation of Bourguignon, Melndez and Ferreira (2005) methodologyto examine the extent to which income inequality is associated with the inequal-ity of observed exogenous circumstances of origin that determine individualsopportunities to pursue their chosen life plans. We find that equalizing a set ofobserved circumstances of origin across individuals such as parents schooling,

parentss stability of employment, fathers age, household size and growing ina single vs a bi-parental household reduces income inequality, but in a smallmargin, in the range of 8 to 13 points of the Gini coefficient, about a 15-20 percent drop. These results are similar to those obtained by Bourguignon l.(2005) for Brazil, although the dataset and the set of observed circumstancesare partially different. These results suggest that a significant part of incomeinequality may be associated with unobserved heterogeneity across individualsunrelated to circumstances of origin, such as preferences, effort and sheer luck,transitory shocks and measurement errors in income. However, assessing therelative importance of these factors vs. the role of unobserved circumstancesremains as future research.

K : income inequality, equality of opportunities.

m

Aplicamos una variacin de la metodologa de Bourguignon, Melndez y Ferreira(2005) para examinar en qu medida la desigualdad de ingresos est relacio-nada con la desigualdad de circunstancias observadas de origen, exgenas alos individuos, que determinan las oportunidades de ste para realizar sus

* Dpm Ecm. U Chl.** Mpl, Chl.1 W gful J Bhm, D B, H Acu, Eb Pu, Rc

Mayer, Marcela Perticara and Leslie Miranda for all their very valuable and helpfulcmm. A uul, h uh pbl f ll .


2/18

Eu Ecm, Vl. 34 - N 2186

planes de vida elegidos. Encontramos que, al igualar un conjunto de circuns-tancias observadas de origen, como escolaridad y estabilidad de empleo de los

padres, edad del padre, tamao del hogar y crecer en un hogar conformadopor un solo padre vs. uno con ambos padres, reduce la desigualdad de ingreso,

aunque en un margen pequeo, en un rango de 8 a 13 puntos del coeficiente Gini,aproximadamente una baja de 15-20 por ciento. Estos resultados son similaresa los encontrados por Bourguignon et al. (2005) para Brasil, aunque la base dedatos y el conjunto de circunstancias observadas son parcialmente distintas.

Estos resultados sugieren que una gran parte de la desigualdad de ingresospodra estar asociada con heterogeneidad no observada entre individuos, norelacionada con las circunstancias de origen, como preferencias, esfuerzo ysuerte, shocks transitorios y errores de medicin en el ingreso. Sin embargo,evaluar la importancia relativa de estos factores vs. el rol de circunstancias noobservadas queda como futura investigacin.

Plb cl: desigualdad de ingresos, igualdad de oportunidades.

JEL Clfc:D31, D63.

I. Introduction

Th l b mg cm phlph bu h

the kind of economic inequality that public policies should aim to reduce.Whl m uh h h mpc f g h ulf ucm (pcll f cm), h h pp h publcplc hul pm m f ul f ppu c -ul.2 Th l ccp h h ul hul hml ppu puu h lf pl, hch u mplh h ppu hul b m b gu ccumch ul h hu h chc c, uch , f mpl,pl fml bcgu. Ac f ul ppu h guthat differences in economic outcomes partly reflect differences in aspectsunder the control of individuals, such as effort, responsibility, choices and so on.Accgl, ul f ppu c pul h publc plchul m ul h gu ccumc h hp ulppu puu h ch lf pl, h ccp h ulgll f ul f ucm h ul mg fm ul chcand preferences. With some variations, this has somewhat become the dominant f h f u h publc c, ugg f mpl

2 S f mpl Rm (1996), (1998) (2000) Dg (1981) f cp- f h f ul f ppu f ucm. Al, Am SCpbl ppch h mblc h h f ul f ppu, cb f mpl S (1999) Nubum S (2000). S Rm (1996)f cu f h m h f bu juc. S l Al, D Tll McCullch (2004) f cu ff u b Eup Amc ff f ul.


3/18


b h f ul plc h f Th Wl B 2005Rp Eu Dlpm:

By equity we mean that individuals should have equal opportunities to

puu lf f h chg b p fm m p ucm, (p. 2)3

H, ll bu h hch ul f ucmreflects individual choices and preferences vs. the exogenous circumstancesh ul h. Th pp mpcl mp cbu hug f h lhp b h f ul f ucm. ppu. I pcul, h pp fll p h mhlglp b Bugug, Ml F (2005) u hl h f mp gu b ccumc f g pl plg h b cm ul Chl.

Studying the relationship between opportunities and outcomes is relevantf u . F, h pccl mplc f h phlphcl c- b ucm ppu ul b l gfc f h were closely associated. This would reinforce the idea of interpreting incomebu c l mu f ul f ucm, bu l a good representation of equality of opportunities. This would also suggest a moregfc l f h gu ccumc cmp h h l ful chc pfc pmg bh ul f ucm

opportunities in the long run. On the contrary, if exogenous circumstances playeda limited role in explaining inequality of outcomes, then this would have differentmplc pg h ch m p: ul ppuadvocates should expect a significant amount of income inequality to remain afterulg ppu, hl c f ul f ucm hul lh chg h m- h lg u- ul u m h plc ul m ccumc uch cc ul uc, h m l bu plc ul b .

W fll Bugug et al. (2005) pg , hch mp establish the effect of circumstances of origin and individual effort in the

determination of income inequality in Brazil.4 In their work, circumstancespl ubl l: h h c mpc g, c ffcon effort, that they take to be the schooling level. They define the former

3 On page 3 of the overview this view is reinforced in these passages: Three considerations

are important at the outset. First, while more even playing fields are likely to lead to lower

observed inequalities in educational attainment, health status and incomes, the policy aim

ul ucm Sc cc h ul f ppu mplthat public action should focus on the distributions of assets, economic opportunities, and

plcl c, h h cl ul cm.4 Bhm (2006) Ru-Tgl (2007) l m h l f chlg cm

ul, lhugh mplg fm ff h lp b Buguget al., which allows establishing and separating the direct and indirect effects of observed

circumstances on income inequality. However, their results are similar to the results found

h , h h bh u ugg lm l f chlg ucgcm ul.


4/18

Eu Ecm, Vl. 34 - N 2188

ffc h pl ffc f b ccumc g, hl ffc b h j ffc f h c c ffc f bccumc g. Ap fm ug ff f c-cumc, u ff fm h pc. Th f m

f p. Ou m m m h h l mp mu h ffc f b ccumc g, uh (m cmpl) u f h p f h cm bu ch ul ff. Hc, pf f h c ffc mpl h ffc f b ccumc h ll f chlg ( ff), l p h upl p f h cm bu mpl unknown combination of unobserved circumstances, individual effort, sheer luck pbl cm mum . Sc, p m lexploratory circumstance-equalizing benchmarks in order to shed some light onthe possible role of unobserved circumstances on the income distribution. In thef bchm, um m u h chlgll l flc ul ccumc- h b ub-,uch h ul m ff pl l h m fchlg. Th mu cmpug h cm bu f ulgchlg ll c ul. Th c ulg bchm cf gug mmum f 10 f chlg (hch cm-pleted at about age 16) under the assumption that a simulated value of schoolinglower than 10 would almost certainly reflect unobserved circumstances, but useh mul ll f chlg c b ul ccumc h

hgh h 10, f hch h l l f ub ccumc. ul ff h -ccumc fc c b pc bl. Y, f h ll fu ccumc ulg bchm l gfc f cm ul.

Th pp ucu fll: Th c c p h bcml h mpcl fc g f h fu b ccum-stances-equalizing benchmarks. The third section describes the data and the f ccumc mpl. Th fuh c p cu hul cmp hm h b Bugug et al. (2005), fll c f cclu.

II. Themodel

a. Bourguignon et al. (2005) Model

Fllg Bugug et al. (2005), mg h m m f ul g, pbl guh ff gup: hdeterminants that result from actions that people carry out along their lives,hch ll hm c h puc, h h b ccum-c u f ppl cl. Bugug et al. (2005) cll h f fm ff bl h c, ccumc bl. Thlhp b cm, ff ccumc c b cb W f C E

i i i= ( , ) , h ccumc C pcll clu f bl

f h ul ccmc g ff E hugh f humcpl bl.


5/18


In order to estimate the model empirically, this relationship can be expressed l ml, fll:

(1) Ln W C E Ui i i i

( ) = + +

h cffc c U, h ul h clu hunobserved circumstance and effort variables, measurement error and variations

of the individuals measured income from their corresponding permanent income

ll. All h fc upp b p fm h clu -bles in Ci and Ei, to have zero mean and that to be identically and independentlybu c ul.

Th fmul h f c bbl, um complete additive separability between circumstances and efforts. For example, it

seems reasonable to expect that the circumstances that surrounded an individualug h chlh, ll h chcc f h huhl hp hum cpl mu h h fluc h hum cplccumul. Accgl, Bugug et al. (2005) pp h fllglhp h ff pl fuc f ccumc:

(2) E B C Vi i i

= +

h B cffc m V p -bbl ff -

m c. A uul, V upp h m b ...c ul.

Iucg u (2) (1) l,

(3) Ln W B C V Ui i i i

( ) ( )= + + +

Th ml pl (3) m gl h ml (1) c ll

h ccumc bl ffc ppl cm cl, ll -

cl hugh ffc h ff bl. I pcul, ml (1) h

mgl ffc f ccumc g mu l . Buguget al. (2005) cll h ffc h Pl Effc f b ccumc

g. O h h h, ml (3) h ffc f b ccumc

g + B

. Th cp h Tl Effc f b

ccumc g. N h h ffc clu h pl ffc f

ccumc g, , bu l c ffc f ccumc

g hugh ff, B. Th l ffc f b ccumc

g lg h h pl ffc fB>0, pc.

b. eff . hg

I pcc, Bugug et al. (2005) mpl chlg h mu

f ff E. H, cu h uc bl bh

controversial and misleading to refer to schooling as an effort variable, at least

in a country with known inequality of opportunities such as Chile. Accordingly,


6/18

Eu Ecm, Vl. 34 - N 2190

h pf plc ff E mpl b ul chlg ll S.

Given this new interpretation, equation (1) would simply indicate that wages are

fuc f hum cpl (.. chlg), ccumc f g, ll

term Ui

, which captures unobserved circumstances, sheer luck, effort at work,

deviations from permanent income, and possibly income measurement errors. In

addition, parameter would be more directly interpreted simply as the return to

schooling, while parameter B would reflect the effect of observed circumstances

f g h ccumul f chlg. F mpl, pm B c cp-

u p uc uc f h , h l f

cg -cg bl cu ug fc lcc

h chc f gg cc uc. I , pm

ul b p h c ffc f ccumc g, g

chlg ll, ll, h ffc f ccumc h u f g mu f chlg. F mpl, pm c cpu h ffc

of the quality of education (likely to be associated with circumstances), the role

f bl cu h huhl f g lb puc -

ings, access to social networks and even the possibility of class discrimination

h lb m.5 I cclu, h p ff

bbl bl, hch ul b cpu m V u (2).

c. The Partial and Total effects of observed Circumstances on

income inequality

The estimation of parameters , and B through an OLS estimation of

equations (1) and (2) allows performing two types of simulations of the distribu-

f cm f ulg gu b ccumc C. L WP

h mul cm bu c h h Pl Effc

described above, obtained after equalizing all the circumstance variables across

individuals in equation (1). Accordingly, the resulting income distribution would

flc ul ffc chlg h u U. M fmll,

h hphcl bu WP ul b fm h mul f hindividual incomes W

iP

using the following equation, and after estimating

u (1) b OLS:

(4) Ln W C S UiP

i i( ) = + +

h C h c f ppul m f h ccumc bl.

A l hphcl g bu WTc h h Tl

ffc f b ccumc g c b b b ulg

ll h b ccumc bl c ul u (3), festimating equation (1) and (2) by OLS. The income distribution WTwould thus

b b fm:

5 S f mpl N Gu (2004).


7/18


(5) Ln W B C V UiT

i( ) ( ) = + + +

where again C stands for the population means of the circumstance variables andh cffc b fm OLS m f u (1) (2).

Th cmp b h cul (b) bu W bu- WP flc h pl ffc f b ccumc h buf cm, hl h cmp b W WTp h ffc f hl ffc f b ccumc g, .. clug h ffc fb ccumc h ccumul f chlg. Bh mu fcm ul ll guhg h p f cm ul c- h h c fluc f b ccumc g, fm hp h cm fm h c ffc f h b ccumc hccumul f chlg.

. w m-qzg bhmk

H, lm f h mhlg cb b h p fh cm ul b f ulg b ccumc mll b cu b ffc ub ccumc. I pcul, cb gu h ub ccumc c pl p f h chlg h c h b ccumc, V. I h c-, h ccumc-ulg pp f Buguget al. (2005) cb b, ml h pl l ffc, pfm

two additional equalizing benchmarks of the effect of circumstances on incomedistribution to explore the possible role of unobserved circumstances. First,assume an extreme situation where all schooling acquired by an individual werefully determined by his circumstances of origin, either observed or unobserved.O ph m mpl, um h h l f ff m h ccumul f chlg. Th u ul b ul gh m V = 0 (hch clu ub ff) f ll ul. I hc, chlg ul c ul l u h ffc f c-cumc, ff. Th ul mulg ul cm breplacing C

i

by the population mean circumstancesC

and Vi

= 0 in equation (3),or equivalently, replacing Ci and Si by C and the population mean schooling Sin equation (1), respectively.6 More formally, the simulated income distributionf ulg b ccumc chlg, WES, ul b fm h mul ul g fm:

(6) Ln W B C UiES

i( ) ( ) = + +

Hence, in this case the only source of variation in the simulated incomebu ul fm m Ui u (1).

7-8

6 N h mg u (2) b OLS l K= BC.7 N h, h m U c clu h c ffc f ub ccumc

g.8 H, h g g clu pl pc p

bl, hch h uc f h mul cm.


8/18

Eu Ecm, Vl. 34 - N 2192

Th c l pl ulg bchm h c uarises from the observation that individuals cannot be made responsible for theirhum cpl ccumul h l f h lf ccl, bu h cgubl b m pl pbl f l h lf ccl, f m

g hhl. L S BC V i i' = + h mul chlg f ul If ulg b ccumc u (2). I h c, lll f mul chlg ll

S

i' , ppg u f chl l

g, c b p lc f ff, bu h ul f ubccumc c V

i. H, f m g hhl, h lu

f mul chlg Si' ll pumbl flc cmb f ff

ub ccumc. Alhugh m m bu f pcfc ghhl f hch ul c b m pl pbl f h c-cumul f chlg, mu b mmb h h hpp de facto other spheres such as penal responsibility, and in the gain of rights such as voting g, ug h . F mul pup, mplm hbchm b gug 10 f chlg (ch bug 16), mpl h mul lu f chlg S

i' h g

h 10. M fmll, h mul cm bu f gug10 f chlg, Wi

GS, fm:

(7) Ln W C S UiGS

i i( ) ''= + +

h,

(8) Si'' = 10 f S C Vi

' = + 10 , S S C V i i i'' ' = = + f S C Vi

' = + > 10.

Alhugh h hhl ml b, clm h pl - h hcmg mplc h c ffc, ml h f ug g um b pl pbl f h chlg.9

Finally, let be an operator that computes an income inequality coef-fc fm g bu f cm W. I pcul, h pp cmpu h G h Thl cffc. Th, g h ffc h uc f h b h mul ul cm

under each of the four circumstance-equalizing benchmarks, it can be expectedh (W) > (WP) > (WT) > (WGS) > (WES).

III. Data

This work employs data from the Survey of Employment and Unemploymentin Greater Santiago, undertaken in June of 2004 by the Department of Economicsof the University of Chile. In addition to the regular economic and laborm u, u u h u bmu f h ul ccumc f g. Th clu huhlchcc ug fc uch huhl , f h g gl

9 Using alternative age thresholds in the range of 14 to 18 years of age yielded only margin-ll ff ul h h p bl f g hhl 16.


9/18


. b-pl huhl, fh mh chlg, p g bh , fh mh pcp h lb m, fuc ffh mh mplm, p b h . Th mpl h l f 11.007 b, cpg 3.060 hm. Th

lc mpl p c-c f h ppul f GSg, hch hl bu h f h cu ppul.

I lc u c h pcp f m h lb m, h fcu l . Th mpl f lm g h g fm 15 b. Th g gemploy sons of 15 years of age and above, but the schooling regressions employ f 23 f g b. Th blh pbllc pblm, m ul ug h 23 uc- full h lb m, l bcu ug m h ch h lg-u mu f chlg. Th bf umpl ul h h p p cm eliminated, as well as those who did not report sufficient information abouth chcc f h p. Fll, c ul gb 30 72 hu p .

IV. ResulTs

a. Schooling and earnings regressions

Tbl 1 p OLS g f h chlg m, u-tion (2) of the model.10 Table 1 indicates that various observed circumstances ofg h gfc ffc h ccumul f chlg. I pcul,parental schooling has a large and statistically significant effect on the sons ac-cumul f chlg, cl 0.28 f chlg p l the mean parental schooling.11 However, this effect is stronger for older sons, asthe interactive variable of parental schooling and sons age indicates.12 This mayb cuc f cg ucl mbl hughu m. Tbl 1l c h h fh g l h ffc h chlg,

lhugh h ffc cc.13 Th ul b c h h fhpc h cc ffc fh g, hch u m lmh uc h uc ll-bg. Fll, Tbl 1l c h huhl bg gl p huhl

10 We performed regressions with robust standard errors for both the schooling and earningsg, bu l ml ul h p h.

11 W mpl m pl chlg cl pblm c hincluding fathers and mothers schooling separately in the regressions. In the few cases

where schooling was reported for one parent only, we employed this figure as the meanpl chlg, u h ump h h h p m ll h h .

12 However, this is consistent with the evidence of higher intergenerational educationalmbl ug ch Chl p N M (2007).

13 W mpl fh g hugh m c h cul h b b fm hmbcu h cul ll h cbu uc h huhl.


10/18

Eu Ecm, Vl. 34 - N 2194

TABLE 1SCHOOLING DETERMINANTS1)

VblSpcfc2)

1 23)

Ag 0.0688111*** 0.0737303***

[0.0174026] [0.0156892]

Pl Schlg

M pl chlg 0.2949843*** 0.2685192***

[0.0811011] [0.0748456]

Mh/fh chlg ffc 0.0517466

[0.043312]M pl chlg * g 0.0037926** 0.0042125***

[0.001713] [0.0015728]

Fh Lf Ccl Vbl

Fh g h h 15 0.0331722** 0.0268721**

[0.0150761] [0.0139196]

Fh g h h 15 - u 0.0000371** 0.0000289**

[0.0000153] [0.0000139]

Chlh Huhl Chcc

Huhl 0.0879002* 0.094932**

[0.0471903] [0.0427122]

B-pl huhl umm 0.9095529** 0.6558078**

[0.3780006] [0.3145342]

Pl mplm bl umm 0.8409475 0.9744392**

[0.5414537] [0.4819123]

C 9.871226*** 10.48943***

[1.193715] [1.071267]

Smpl S 773 867

R-u 0.370 0.3757

Aju R-u 0.3626 0.3699

1) Dp bl f chlg.2) OLS estimates standard errors in brackets; *= significant at the 10% prob. level; **= significant h 5% pb. ll; ***= gfc h 1% pb. ll.

3) Spcfc u mul.


11/18


TABLE 2EARNINGS EqUATIONS1)

VblSpcfc2)

1 2 33)

Schlg Ru

Pm Euc 0.0197612 0.0339435

[0.0281318] [0.0236105]

Scu Euc 0.0676964* 0.0502285

[0.0389149] [0.0347507]Pm ScuEuc

0.063888***

[0.0113342]

T Euc 0.1182382*** 0.1238904*** 0.1475077***

[0.0255809] [0.0242499] [0.0179333]

Epc Vbl

Pl Epc 0.0161548* 0.0245867*** 0.0231317***

[0.0083877] [0.0055474] [0.0054598]

Pl Epc - u 0.0001197 0.0001951** 0.0001668*

[0.0001127] [0.0000995] [0.0000976]

Pl Schlg

M pl chlg 0.0013671 0.029537*** 0.0295607***

[0.0203902] [0.0061033] [0.0061083]M pl chlg* g 0.000685

[0.0004757]

Fh Lf Ccl Vbl

Fh g h h 15 0.0037693

[0.0027396]Fh g h h 15- u

0.0000037

[0.00000272]Chlh HuhlChcc

Huhl 0.0057045

[0.009183]

Bpl huhl umm 0.0547253

[0.066334]Parental employment instabilityumm

0.1085018

[0.0986114]

C 11.21416*** 11.03948*** 10.87809***

[0.3191213] [0.1822356] [0.1441385]

Smpl S 595 667 667

R-u 0.5577 0.5321 0.5306

Aju R-u 0.5486 0.5279 0.5271

1) Dp bl lg f g.2) OLS estimates standard errors in brackets; *= significant at the 10% prob. level; **= significant

h 5% pb. ll; ***= gfc h 1% pb. ll.3) Spcfc u mul.


12/18

Eu Ecm, Vl. 34 - N 2196

decrease sons schooling, as well as having parents with unstable employment. Inparticular, growing in a single-headed household decreases schooling in almost , hg p h ubl mplm h ml ffc.14 Icclu, Tbl 1 c h b ccumc f g Chl

have a significant effect in reproducing inequality through their impacts onh ccumul f hum cpl. W mpl pcfc 2 f Tbl 1 c u h cm mul c h h ccumc-ulgbchm cb b.

Tbl 2 h h ul f OLS g f h g. Allpcfc h h ffc f chlg pl pc g. I , Tbl 2 c h m pl chlg hl ccumc bl h h bu cll gfc ffc g. Th ugg h h mg ccumc bl mpl h m ffc g cl, lhugh h hughtheir indirect effect on the accumulation of schooling, as shown in Table 1.Tbl 2 c h uc Chl h h h h hgh- pcu u, h g f 20 p c p . I , Tbl 2h h f m pl chlg g bu 3p c.15 W mpl pcfc 3 f mulg ul cm b h fu ccumc-ulg bchm.

b. m m b ff

Using specification 2 of Table 1 and specification 3 of Table 2, we perform thefu ccumc-ulg bchm cb b cmpuh ulg mul cm bu cffc. Tbl 3 p ll fhm f h G h Thl cffc. W l h ul f h f p bm ul. All ul c p f ll g, ll f h 14 ch.

Tbl 3 p G cffc f h cul (b) ul (0.5)h mh l h h G cffc f G Sg -, hch u 0.52-0.55.16 Th pbl ul f hg lmpl ul f g 23 b. H, h fcu f h

blh h l f b ccumc h cm bu,h mll gc m f cc. Tbl 1 l c h hG cffc f h ug ch l, hch m b cucf hg m ml g pfl l h lf ccl.

Table 3 indicates that the Partial and Total Effects of observed circumstances cm ul pl l bu 7 12 p f h G cffc,pg h g ch, l p h cffc f bu 15

14 This variable is a dummy variable equal to one if either both parents had unstable employ-m, h c f gl p huhl, h huhl h h ublmplm.

15 Th c h h fg p b B, C M (1999),h p cll gfc cffc f bu 0.02 0.01 f h fh h mh chlg h g, pcl.

16 S f mpl F, P, F Wl (2003).


13/18


TABLE 3EFFECTS OF EqUALIING CIRCUMSTANCES ON LABOR INCOME INEqUALITY

GREATER SANTIAGO, CHILEG Thl cffc, Tp-bm qul R

A. G Cffc Ag 23-36 Ag 37-50 Ag 51-65 Ag 23-65

Ob Iul (W) 0,454 0,518 0,527 0,503

Smul Iul

Pl Effc (WP) 0,395 0,464 0,410 0,433

Tl Effc (WT) 0,343 0,428 0,408 0,42010 chlg

gu (WSG

)

0,335 0,419 0,397 0,412

Eul Schlg (WES) 0,274 0,320 0,307 0,314

B. Thl Cffc Ag 23-36 Ag 37-50 Ag 51-65 Ag 23-65

Ob Iul (W) 0,387 0,502 0,512 0,477

Smul Iul

Pl Effc (WP) 0,278 0,375 0,293 0,331

Tl Effc (WT) 0,206 0,308 0,287 0,30710 chlg

gu (WSG)0,197 0,296 0,270 0,295

Eul Schlg (WES) 0,131 0,180 0,158 0,171

C. q5/q1 R Ag 23-36 Ag 37-50 Ag 51-65 Ag 23-65

Ob Iul (W) 8,386 13,273 14,002 11,465

Smul Iul

Pl Effc (WP) 6,513 9,933 7,766 8,133

Tl Effc (WT) 5,271 8,421 7,509 7,83010 chlg

gu (WSG)5,023 8,053 7,361 7,548

Eul Schlg (WES) 3,951 5,070 4,625 4,796

25 p c. Th mul f h Thl cffc h ml p,although more accentuated. This indicates that part of the observed incomeinequality in Chile is associated with inequalities associated with the set ofcircumstances of origin employed in this work. However, these results alsougg h f ulg h f b ccumc mpl h , gfc mu f cm ul m. Ah g-nificant feature of the results in Table 1 is that the Partial and the Total effects ofb ccumc l ml cm ul, lhugh f p


14/18

Eu Ecm, Vl. 34 - N 2198

l h c f h G cffc f h Tl ffc, pc. Thul c h, l Chl, h l f b ccumc hccumul fyears f chlg h gfc bu lm ffc hincome distribution.17 Note also that a large part of the Total effect is associated

l h h Pl ffc, h , h c ffc f b ccumcon earnings. This suggest that variables such as the quality of education, and thecu f bl c h fml bcgu pl humcpl pcul ( ugg h g g Tbl 2) m pl mp l hpg cm ul Chl.

Regarding the additional circumstance-equalizing benchmarks describedb, Tbl 3 c h gug 10 f chlg l mlcm ul h Tl ffc. Fll, h m u f ul-izing schooling (at the population mean schooling level) decrease the Ginicffc 10 p fm h G c h h Tl ffc.E hugh h ulg c m m m, fc h h ll gfc mu f cm ul ul p uh ccumc.

I mp h h ul p Tbl 3 ml hresults obtained by Bourguignon et al. (2005), although they employ a dif-f ( much lg) pll ff f ccumc(lhugh pl chlg l mpl). I h u, h Pl Tl ffc l fl ml blu ppl p h G Thl cffc h p b. Th fc fc h gl

f h c, ml h, lhugh cm ul c h ul f f b ccumc, ulg hccumc f g h h lm ffc ucg cm bu-. Th u ugg h h mg cm bu cwith as combination of unobserved circumstances, individual effort andpfc, hc cm, h luc, ll mum h cm bl.

It would be interesting to assess the role played by unobserved circumstances mg h upl p f h cm bu f ulgb ccumc. Alhugh h b h pup f h ,

we provide some comments on this issue. First, using data from twins, Behrmanand Rosenzweig (2004) suggest that the role of unobserved circumstances (fixedfml bcgu) h ffpg pfmc c b lg cmph h l f b ccumc uch pl chlg.This would indicate that a substantial part of the income inequality obtained afterulg b ccumc m b c h ubccumc. H, l l u, Bhm Rg(2002) l ugg h ml chlg m p m mpub fc c h fml bcgu. Th c ulugg h h b ccumc mpl h ll capture the effect of important unobserved circumstances associated with family

17 Th c h h hgh g ll f f chlg l l -ul chlg h Chl hb cmp h h f h g. Sf mpl F, P, F Wl (2003).


15/18


bcgu. I pl h pbl, m u f l of observed circumstances that was gathered in 2006 for 400 sons sampled fromthe sons employed in this work. In this survey we asked them to report informa-tion about additional observed circumstances, in particular their parents interest

h pg chl, c p . publc chl, cc p ug fc, p g g ll, gg urban vs. a rural environment, parents ethnicity (amerindian vs. non-amerindianbackground), and access to pre-school education during infancy. We do notemploy these additional circumstance variables in this work because the sample uc. H, App 1 p ul h cl c- f h l ccumc bl h m pl chlg,hch h m ccumc bl mpl h . Th c App 1 c h m pl chlg c h ch f h l ccumc bl. Th ugg h h bl ll cpu f ff l ccumc f g f hul h mpl.

v. onclusions

Th pp h ppl f h mhlg lp hpg b Bugug et al. (2005) l h hchcm ul Chl c h f gfc ul

f gu b ccumc f g, ml pl chlg,vulnerability of the household heads occupation, fathers age, family size andbg gl . b-pl huhl. Th mhlg llto examine the direct effect of these observed circumstances on earningscl chlg (h pl ffc), ll h l lh h ccumc pl mg ul chlg ll,hch jl l h l ffc f b ccumc g.W f h f ulg ul b ccumc h mvalues of the population, the resulting standard income distribution indicatorsbcm m gl, cg h p f cm ul flc

ul f ccumc. H lg mu f cm ul c h ul h ccumc. I pcul, fulg b ccumc, h G cffc c 7 12p 8 13 p u h Pl h Tl ffc, pcl,pg ppml p f 10-20 15-20 p c ch c.Th ul ml h b b Bugug et al. (2005) fBrazil, although they employ a partially different set of circumstance variables lg .

W pl h pbl l f ub ccumc b lpg l bchm h Bugug et al. (2005). Eulgul chlg h ppul m flc m uh ll chlg um p ccumc, h b ub fuh c cm ul, bu ll gfc mupersists. In addition, guaranteeing all individuals 10 years of schooling (achievedat about 16 years of age) under the argument that having less schooling may reflect ub ccumc l ml ul h Tl ffc.


16/18

Eu Ecm, Vl. 34 - N 2200

Th ul f h ugg h, f h c f ulg ccum-stances is a close approximation of the notion of equality of opportunities, thencm ul c m ul flc h g f ulf ppu, cm ul m l flc h pc uch

individual effort, preferences, choices, sheer luck and possibly transitoryhc cm cm mum .

These results also suggest some implications for public policy, which depend h ch ml p h ul-f-ucm . ul-f-p-pu b. If cm ul m, h h ul ugg hachieving this aim-even in the long run-would require more than just equalizingccumc uch cc ucl ppu, cucl bu plc ll b . O h h h,c f ul f ppu mu b ccp h blhgul ppu ll c h gfc mu f cmul. H, lhugh h ul m ugg, m ch b m cclu ul. I pcul, m c m h ffc f hg lg f b ccumc h hmpl h .

VI. RefeRences

Al, Alb, D Tll, Rfl Rb McCullch (2004). Iul

Hpp: A Eup Amc Dff? Journal ofPublic Economics 88 (9-10), pp. 2009-42.

Bhm, J. M. Rg (2002). D cg m chlgraise the schooling of the next generation?,American Economic Review92:1, pp. 323-334.

Bhm, J. M. Rg (2004). Ru Bhgh,Review ofEconomics and Statistics, 86: 2, pp. 586-601.

Behrman, J. (2006). How Much Might Human Capital Policies Affect EarningsInequalities and Poverty?, IDB-University of Chile Workshop on IncomeIul, Dcmb 2006.

Bourguignon, Franois, Ferreira, Francisco and Marta Menndez (2005).Iul f Oppu Bl? Wl B. Whg, D.C.

Bravo, D., Contreras, D. y P. Medrano (1999). Measurement error, unobservablesand skill bias in estimating the return to education in Chile, Departmentf Ecmc U Chl.

De Ferranti, D., Perry, G., Ferreira F., and M. Walton (2003).Inequality in LatinAmerica and the Caribbean. Breaking with History? Th IlBank for Reconstruction and Development/The World Bank, WashingtonDC.

Dworkin, R. (1981). What is Equality? Part 2: Equality of Resources, Philosophyand Public Affairs, 10 (3): 283-345.

N, J. R. Gu (2004). Cl cm mcc h lb m: c fm Chl.Estudios de Economa, 31: 2,pp. 113-132.

N, J. L. M (2007). Rc fg gl cmand educational mobility in Chile. IDB-University of Chile Workshop on


17/18


Icm Iul, Dcmb 2006, Dcum Tbj N 244,Dpm f Ecmc, U Chl.

Nussbaum, M. and A. Sen (2000). La Calidad de Vida, Fondo de CulturaEcmc, Cu Mc.

Rl, Jh (1971).A Theory of Justice. Cmbg, MA: H UP.

Ruiz-Tagle, J. (2007). Forecasting Income Inequality, forthcoming inEstudiosde Economa.

Roemer, John E. (1996). Theories of Distributive Justice. Cambridge, MA:H U P.

Rm, Jh E. (1998). Equality of Opportunity, Cmbg, MA: HU P.

Roemer, John E. (2000). Equality of Opportunity, inMeritocracy and EconomicInequality, Kh A, Smul Bl S Duluf, ,Pc U P, N J.

S, A. (1999).Development as Freedom, Kpf, N Y.World Bank (2005).Equity and Development, World Development Report 2006,

Th Wl B Of U P, N Y.


18/18

Eu Ecm, Vl. 34 - N 2202

APPENDIx 1

MEAN DIFFERENCE T-TEST FOR MEAN PARENTAL SCHOOLING,

BY VARIOUS CIRCUMSTANCE VARIABLES

P pg

chl

M Pl

SchlgS. D. P(T > )

Hgh 8.78 4.205.19 1.00

Mum & L 6.48 3.99

Schl TpM Pl

SchlgS. D. P(T > )

P (Pcul Pg) 9.97 4.056.63 1.00

Publc (Mucpl &

P. Subc)6.72 3.90

Acc p ug

fc

M Pl

SchlgS. D. P(T > )

Y 7.94 4.144.40 1.00

N 4.98 4.08

Rul . Ub

Em

M Pl

SchlgS. D. P(T > )

Ub 7.89 4.173.59 0.998

Rul 5.43 4.15

P g & g

ll

M Pl

SchlgS. D. P(T > )

Whu ffcul 8.32 3.987.83 1.00

Wh ffcul 3.67 3.46

P hcM Pl

SchlgS. D. P(T > )

A l p Am 5.29 3.97

2.48 0.007N f Am

Bcgu7.74 4.21

Acc p-chl

uc

M Pl

SchlgS. D. P(T > )

Y 9.15 4.145.02 1.00

N 6.79 4.09

Inequality Outcomes vs Inequality Opportunitie Chile

Documents

Transcript of Inequality Outcomes vs Inequality Opportunitie Chile