Transcript lsejun04 - Santa Fe Institute

Inequalities, inequalities…..
B. Milanovic, June-July 2004
1. Briefly: the three concepts
2. Concept 1 and Concept 2
inequality
3. What happened to global
inequality: new 1998 data
4. Openness and within-country
inequality
Sources
• Book: Worlds Apart: Global and
International Inequality, 1950-2000 coming
out in May 2005 (Princeton University
Press) ORDER IT NOW!!
• Website:
www.worldbank.org/research/inequality
• Email: [email protected]
1. Definitions
Different types of inequality
Individuals in:
Countries
World
Countries in:
The usual within-country
distributions
(e.g. inequality in the US is
greater than in Sweden)
-----
Global income distribution:
distribution of persons in the
world
(comparable prices)
Distribution of countries’
GDP per capita
(rich vs. poor countries; are
the poor countries catching up
or not;
the convergence literature)
(comparable prices)
How they differ?
Main source of
data
Unit of
observation
Welfare
concept
National
currency
conversion
Within-country
distribution
(inequality)
Concept 1:
unweighted
inter-national
inequality
National
accounts
Country
GDP or GNP
per capita
Concept 2:
weighted international
inequality
National
accounts
Country
(weighted by its
population)
GDP or GNP
per capita
Concept 3:
“true” world
inequality
Household
surveys
Individual
Mean per capita
disposable
income or
expenditures
Market exchange rate or PPP exchange rate (but
different PPP concepts used)
Ignored
Ignored
Included
2. Inequality
between
countries
Coverage: number of countries and share of world population
160
Number of countries or coverage of world population
140
Number of countries included
120
Coverage of world population (in %)
100
80
60
40
20
0
19
50
19
52
54
19
6
5
19
19
58
19
60
19
62
19
64
19
66
19
68
19
70
19
72
19
74
6
7
19
Year
19
78
19
80
19
82
19
84
19
86
19
88
19
90
92
19
19
94
19
96
19
98
About 140 countries included; about 6200 country/year GDPs
almost 100 percent of world population and world GDP (in
current dollars)
current countries projected backward (NEW)
 SIMA World Bank data used to get benchmark 1995 $PPP
GDP per capita; then these GDP per capita projected backward
and forward using countries’ real growth rates (78% of data
from WB sources; others mostly from national SYs; some from
PennWorld Tables, UN sources)
According to Concept 1, countries' performances have diverged over the last two
decades
Unweighted inter-national inequality, 1950 to 2000
0.560
0.540
0.520
Gini coefficient
0.500
0.480
World
0.460
0.440
0.420
0.400
World without
Africa
0.380
Year
And it is not only because Africa is falling behind
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
19
78
19
76
19
74
19
72
19
68
66
64
62
60
58
56
54
52
70
19
19
19
19
19
19
19
19
19
19
19
50
0.360
Downwardly mobile world
The Four Worlds defined
The Rich: All countries with the GDP per capita equal/greater than
the poorest WENAO
The Contenders: With GDP per capita at least 2/3 of the poorest
WENAO ( they can catch up within a generation)
The Third World: With GDP per capita between 1/3 and 2/3 of the
poorest WENAO
The Fourth World: With GDP per capita less than 1/3 of the poorest
WENAO
Overall upward and downward mobility
1960-78 and 1978-2000
1978-2000
1960-78
The border countries and their GDP per capita
levels (in $PPP, 1995 prices)
1960
1978
2000
Greece (13821)
Barbados
(13297)
Malaysia (9887)
Slovak (8595)
Egypt (4630)
Bulgaria (4313)
First to
second
Portugal (3205)
Croatia (3085)
Second to
third
Haiti (2139)
Malaysia (2120)
Portugal (7993)
Puerto Rico
(7662)
Armenia (5294)
Fiji (5156)
Third to
fourth
Nigeria (1080)
Madagascar
(1031)
Guyana (2728)
Cote d’Ivoire
(2649)
Why Concept 1 inequality
matters
• Are poor countries catching up as we would
expect from theory?
• Are similar policies producing the same effects or
not? (Rodrik: convergence of policies, divergence
of outcomes). Why?
• Migration issues
• Countries are not only interchangeable individuals
(random assortments of individuals); they are
cultures. Divergence in outcomes is elimination of
some cultures. Perhaps it’s good, perhaps not.
3. Moving to Concept 2: its
relevance and irrelevance
• Once we have Concepts 1 & 3, Concept 2 is
redundant.
• But we have imperfect grasp of Concept 3
inequality => Concept2 provides a check on
“true” inequality (its lower bound)
• We use it to approximate “true” inequality.
Think, at the limit, of each individual being
a country
Year
20
00
19
98
19
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
19
78
19
76
19
74
19
72
19
70
19
68
19
66
19
64
19
62
19
60
19
58
19
56
19
54
19
52
19
50
Gini index
The mother of all inequality disputes
0.700
Global inequality
0.600
Populationweighted
0.500
Unweighted
0.400
How are Concepts 2 and 3 related?
• In Gini terms:
n
n
1
i1 Gi pii   i
n
 y  y ) p p  L
j
i
i j
j i
• where Gi=individual country Gini, π=income share, yi =
country income, pi = popul. share, μ=overall mean income,
n = number of countries
• In Theil:
n
n
 p Ti   p ln
i
i 1
i
i 

yi
Inequality between population-weighted countries
According to Concept 2, there is convergence among countries…
0.780
0.740
0.700
Theil
0.660
0.620
0.580
Gini
0.540
19
50
19
52
19
54
19
56
19
58
19
60
19
62
19
64
19
66
19
68
19
70
19
72
19
74
19
76
19
78
19
80
19
82
19
84
19
86
19
88
19
90
19
92
19
94
19
96
19
98
20
00
0.500
20
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
00
98
96
94
92
90
88
86
84
82
80
78
76
74
72
70
68
66
64
62
60
58
56
54
52
50
Gini coefficient
...or maybe there is not
0.600
World
0.560
World without
China
0.520
0.480
World without India and China
0.440
0.400
Alternative Concept 2
calculations
• Alternative growth rates for China (official-World
Bank, Maddison, Penn World Tables)
• Breaking China, India, US, Indonesia and Brazil
into states/provinces (but redistribution within
nations)
• Breaking China into rural and urban parts
(Kanbur-Zhang data set)
• What PPP to use (Geary-Khamis, EKS, Afriat)
Implied China’s GDP per capita in different years
According to different sources
PWT 6.1
Maddison
World Bank
1952
568
627
262
1960
662
785
497
1966
773
879
534
1978
899
1142
754
1988
1703
2119
1676
1999
3319
3803
3867
2000
3642
na
4144
Concept 2 inequality for different
versions of China’s GDP per capita
0.5800
World Bank
0.5600
Penn World
Tables
Maddison
0.5200
0.5000
0.4800
98
96
94
92
90
88
86
84
82
80
00
20
19
19
19
19
19
19
19
19
19
78
Years
19
74
72
70
68
66
64
62
60
58
56
54
52
76
19
19
19
19
19
19
19
19
19
19
19
19
19
19
50
0.4600
19
Gini
0.5400
…and breaking China and India into their provinces/states.
Inter-national population-weighted inequality:
with China and India replaced by their provinces and states
65.0
60.0
Countries and regions of
China (R/U) and India
Countries and regions of China
and India
Gini index
55.0
Countries only
50.0
45.0
40.0
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Year
How much has Concept 2 inequality
changed (Gini points; 1985-00)?
Whole
countries
Chiibus by
states + whole
countries
R/U for China
World Bank
data
-3.3
Maddison
data
-1.9
-3.8
-2.3
-3.3
-1.9
•.4
Distribution of lnGDPPP pc in 85 and 00
(WB numbers; states, R/U for China)
•0
•.1
•kdensity lngdp
•.2
•.3
2000
•6
•7
•8
•kdensity lngdp
•x
•9
•10
•kdensity lngdp
Gini: 60 (in 1985); 57 (in 2000)
•11
.3
.2
.1
0
kdensity lngdp
.4
.5
Distribution of lnGDPPPP per capita;
provinces/states and countries (1985, 2000)
6
7
8
9
10
x
kdensity lngdp
kdensity lngdp
11
Concept 2 between
1980 and 2000
Contributes to decline
(equilibrating factors)
• Inclusion of
provinces/states of
China, India, Brazil,
Indonesia, US (even if
many within
themselves are
diverging!) 0.5 point
Reverses decline
(disequilibrating
factors)
• Higher (old) income
level in China
(Maddison) 1.5 points
• Inclusion of
rural/urban break up
of China 0.5 points
Result: we shave off half of the Concept 2 decline
4. Global inequality
Number of income and expenditure-based surveys by region
Region
1988
Income Exp
Africa
3
11
Asia
9
10
Latin America and 18
1
the Caribbean
Eastern Europe and 27
0
former USSR
WENAO
23
0
Total
80
22
1993
Income
Exp
3
27
8
18
16
4
1998
Income
0
8
20
Exp
24
17
2
19
3
13
14
23
69
0
52
18
59
3
63
Note: “Expenditure” or “consumption” survey is used inter-changeably. Common sample: 86 countries.
Coverage: around 90% for
population, 96% for GDP
Full sample
Population
1988
1993
1998
GDP (in US$)
1988
1993 1998
Africa
Asia
E. Europe/FSU
LAC
WENAO
World
48.0
92.5
99.3
87.4
92.4
87.3
76.1
94.9
95.2
91.8
94.8
92.4
67.1
94.4
100
93.0
96.6
91.6
48.7
94.4
99.4
90.2
99.3
96.5
85.2
93.2
96.3
92.8
96.2
95.4
71.2
95.6
100
95.2
96.3
96.0
43.0
92.5
93.8
85.1
83.5
84.8
41.2
91.3
94.2
90.5
83.5
84.3
37.6
90.7
93.2
89.6
82.1
83.1
32.5
94.4
95.0
88.8
92.9
91.8
35.5
91.7
96.1
92.3
91.7
90.9
33.4
93.1
95.7
93.9
90.5
90.6
Common sample
Africa
Asia
E. Europe/FSU
LAC
WENAO
World
What does Concept 3 say?
World international dollar inequality in 1988 and 1993
(distribution of persons by $PPP and $ income per capita)
1988
Full sample
1993
1998
61.9
(1.8)
65.2
(1.8)
71.5
(5.8)
Gini index
Theil index
International
dollars
Gini index
Theil index
1988
Common sample
1993
1998
64.2
(1.9)
62.2
(1.8)
65.3
(1.6)
64.1
(1.9)
81.8
(6.1)
79.2
(6.3)
72.7
(5.6)
81.7
(5.5)
78.9
(6.6)
77.3
(1.3)
80.1
(1.2)
79.5
(1.4)
77.8
(1.4)
79.9
(1.6)
79.4
(1.5)
125.2
(7.1)
139.2
(7.5)
135.4
(8.3)
128.3
(8.1)
138.0
(9.3)
134.8
(8.7)
US Dollars
Note: Gini standard errors given between brackets.
Decomposition of global income inequality, 1988-1998
(common-sample countries; distribution of persons by
income/expenditure per capita)
Gini
1988
Gini
1993
Gini
1998
Theil
1988
Theil
1993
Theil
1998
10.6
(17)
11.1
(17)
11.0
(17)
20.3
(28)
22.8
(28)
23.2
(29)
Between-country inequality
51.6
(83)
54.2
(83)
53.1
(83)
52.4
(72)
58.9
(72)
55.7
(71)
Total world inequality
62.2
(100)
65.3
(100)
64.1
(100)
72.7
(100)
81.7
(100)
78.9
(100)
8.3
(11)
8.2
(10)
8.6
(11)
18.3
(14)
20.5
(15)
22.3
(17)
69.5
(89)
71.7
(90)
70.8
(89)
110.0
(86)
117.5
(85)
112.5
(83)
International dollars
Within-country inequality
US dollars
Within-country inequality
Between-country inequality
Total world inequality
77.8
79.9
79.4
128.3
138.0
134.8
(100)
(100)
(100)
(100)
(100)
(100)
100.0
Cumulative % of income
80.0
60.0
40.0
1988
1993
20.0
1998
0.0
5
10
15
20
25
30
35
40
45
50
55
60
Cumulative % of people
65
70
75
80
85
90
95
100
What explains the 1988-93 increase in inequality?
Key changes in inter-country terms between 1988 and 1993 (in Gini points)
Japan
Germany
France
UK
Subtotal
China (urban)
India (urban)
Total
•
•
•
India(rural)
+0.18
+0.15
+0.10
+0.06
+0.49
+0.18
+0.67
China(rural)
+0.11
+0.12
+0.08
+0.03
+0.34
+0.19
+0.02
+0.55
Slow growth of rural incomes in populous Asian countries compared to rich OECD
countries.
The pulling ahead of urban China vis-à-vis rural China and India. The urban-rural
ratio increased by a half in China and it went up in India too. The mean rank of
population in urban China increased from 53rd percentile to 62nd while the mean ranks of
populations in rural India and China stayed within 1 and 2 percentile of 1988.
The “hollowing out” of the world’s middle class—a problem with Latin America and
Eastern Europe since respectively the early 1980’s and early 1990’s; eg. the mean
income rank of Russian population decreased from 80th to 73rd percentile.
And what explains the 1993-98 decrease in inequality?
Key changes in inter-country terms between 1993 and 1998 (in Gini points)
Japan
Germany
France
USA
UK
Subtotal
China (urban)
India (urban)
Subtotal
India(rural) China(rural)
-0.07
-0.19
-0.07
-0.15
-0.08
-0.14
-0.11
-0.06
-0.11
-0.28
-0.71
+0.23
+0.16
+0.02
+0.04
+0.25
+0.20
-0.99
+0.45
The three factors that all worked toward increasing inequality between 1988 and
1993, behaved very differently over the next five-year period.
•One of them (rising income distance between rural and urban areas in China and
India), continued almost unabated.
•Another—income distance between rural India and China and the West—
reversed, contributing to inequality decrease.
•And the third, the crisis in transition countries, moderated, and basically no
longer affected world inequality very much.
The key determinants of global inequality
Interaction between
1. the rich countries of the West,
2. urban incomes in China and India
3. rural incomes in these two countries
The ratio between (2) and (3) has been rising, and is unlikely to moderate. Moreover, while
China and India are the most important examples of the trend, the urban-rural gap is rising in
several other Asian countries (Bangladesh, Indonesia, Thailand).
But as (2) catches on (1), world inequality is reduced.
The crucial “swing” factor then becomes the ratio between (3) and (1): what happens to rural
incomes in China and India vs. incomes of the rich world. If the former catch up, world
inequality goes down; if they do not, world inequality tends to rise.
SENSITIVITY ANALYSIS FOR
CONCEPT 3
The three building blocks for the calculations…
● national distributions available from Household
surveys,
● mean incomes again available from Household surveys
or from National accounts (GDP per capita) and
● PPP exchange rates.
…and problems with each of them
How Concept 3 changes with various
definitions: what we expect
GDP Survey income
Whole countries
A
B
Broken R/U
--
C
A>B because HS/NA decreases in GDP (Little doubt that
HS data should be used in distribution analysis, as it is
done for individual countries; but people have recently
used GDP per capita)
C>B because of increasing rural/urban disparity
Would conclusions change if we used GDPs per capita instead of HS means?
Yes to some extent. The increase in 1988-93 is less sharp.
World income inequality in 1988, 1993 and 1998
(common-sample countries; $PPP)
Gini
(1) HS mean based
(2) GDP based
Difference (1)-(2)
Theil
(3) HS mean based
(4) GDP based
Difference (3)-(4)
1988
1993
1998
62.3
(2.1)
64.1
(2.1)
-1.9
65.3
(2.1)
65.5
(2.3)
+0.2
64.1
(2.5)
63.5
(2.4)
+0.6
72.3
(6.2)
78.2
(6.8)
-5.9
81.6
(6.5)
83.0
(7.2)
-1.4
78.8
(8.1)
77.0
(8.1)
+1.8
IMPORTANT NOTE: Relatively high HS/NA ratios for China and India in 1988
(compared to later years); explain relatively low global inequality in 1988
Problems with the use of GDP
• Property incomes chronically underreported in HS
are allocated to all (but the poor hardly receive
any of them!)
• So are other parts of GDP: undisbursed profits,
build-up of inventiories (all corporate income that
is not distributed to HHs)
• Underreporting and undersurveying is strong
among the rich; but the entire difference btw. GDP
and HBS mean is allocated to all (in proportion to
their income)
• Distribution of social transfers (H&E) is often prorich too
• Basically, the problem is that the gap
between GDP and household mean is
caused by underestimating top incomes
• But that gap is allocated ACROSS the entire
income distribution
And what happens if we do not change PPPs?
That is, use CPIs to convert all incomes, express them all in domestic prices of 1988
and then use the 1988 PPPs.
A milder increase between 1988 and 1993, and practically unchanged between 1998
and 1993 (or even continued increased if measured by the Theil index)
Global inequality calculated using 1988 PPPs and incomes
expressed in 1988 domestic prices (full-sample countries)
Gini
Theil
Standard errors between brackets.
1988
61.9
(1.9)
1993
63.2
(1.9)
1998
63.1
(2.0)
71.5
(5.7)
75.7
(6.1)
76.7
(6.7)
My preferred measure (Gini with R/U split +
HS data) with 95% confidence interval
72
68
64
60
56
52
1988
1993
1998
Everyone agrees global inequality is high--but
recent trend is a matter of dispute
72
Dowrick-Akmal
70
68
Bhalla
Dikhanov-Ward
Bourgignon-Morrison
66
Chotiapanich-Val.-Rao
Sala-i-Martin
Milanovic
64
Sutcliffe
62
60
58
56
19
70
19
71
19
72
19
73
19
74
19
75
19
76
19
77
19
78
19
79
19
80
19
81
19
82
19
83
19
84
19
85
19
86
19
87
19
88
19
89
19
90
19
91
19
92
19
93
19
94
19
95
19
96
19
97
19
98
19
99
20
00
54
Intuitively, what is a Gini of 64-66; how big is it?
Is it “grotesquely high” as UNDP Human Development Report says?
5 p ercen t ($ P P P )
1 0 p ercen t ($ P P P )
5 p ercen t in $
1 0 p ercen t in $
5 to p co u n tries (G D P
p er cap ita)
1 0 to p co u n tries
(G D P p e r cap ita )
T o p sh are in
to tal w o rld
in co m e
(in % )
3 3 .3
5 0 .0
B o tto m sh a re
in to tal w o rld
in co m e (in
%)
0 .3
0 .8
R atio
1 0 0 to 1
6 3 to 1
4 5 .0
6 7 .5
0 .1 5?
0 .4 5
3 0 0 to 1?
1 5 0 to 1
3 1 8 50
570
5 6 to 1
2 8 0 66
660
4 2 to 1
Or is it that we don’t know what “optimal inequality” is
(as an economist said).
"Inequality transition"?
• Lucas and Firebaugh view: global
inequality has peaked; Why?
• Permanent effects of industrial revolution
• Policy convergence => Income convergence
• Historically, Concept 2 inequality drove
global inequality since IR; for the last 30+
years has been on the decline; then Concept
3 must follow.
But...
• Technological revolution continues (not
only one discrete big bang…), differences
may be accentuated (speed of tech.
inventions = > speed of dissemination)
• Policy convegence did not result in income
convergence
• And all hangs on the break in Concept 2
trend which depends on one country and
one particular set of growth numbers for it.
Poor, middle class and rich in the world
According to three inequality concepts
(1)
(2)
(3)
Concept 1
Concept 2
Concept 3
Number of countries Percent
of Percentage of world
population living in population
with
countries
with income being…
average
income
being…
Poor (below mean income
79
70.1
77.4
of Brazil)
Middle class
28
13.9
6.7
Rich (above mean income
29
16.0
15.9
of Portugal)
Out of which WENAO
21
13.0
10.0
Total
136
100
100
Note: Columns (1) and (2) based on GDP per capita. Column (3) based on data from household surveys (full sample; 122
countries in 1998). Brazil and Portugal always included in the higher group (respectively middle-income and rich).
Poor people in poor countries? How many are they? Almost 4 billion.
Rich people in rich countries? About 700 million.
Poor people in rich countries; rich people in poor countries?
About a hundred million each.
Brings us to almost 5 billion people? So, where is the middle?
Correspondence between poor countries and poor people in the
world (in million people; 1998; household survey data)
Persons
Poor people
Countries
Middle-income
Rich people
people
Total
population
Poor countries
Middle-income
countries
Rich
3879
210
96
4185
189
92
35
115
52
707
277
913
Total population
4160
360
855
5375
Note: Full sample countries (122 countries). Poor below mean income of Brazil, or social assistance
eligibility in the West (about $PPP 10 per capita per day.
5. Openness and within-country
inequality
Decile shares: deciles from 321 surveys, 1988-1998 (SURE
and GMM/IV estimates; openness and govt exp.
instrumented)
Equations (for 10 deciles and more than 100 countries, 1988, 1993, 1998)
•j
yij
 i  iOPENj  i mj  i3(OPENj * mj)  i 4 DFIj  i5 DFIj * mj)  i6 FDj  i7 DEMj  eij
mj
yij = income of i-th decile of j-th country in constant PPP dollars,
mj = mean household survey income of j-th country
OPEN = (exports and imports)/GDP
DFI = (direct foreign investment)/GDP
FD = financial depth = (M2/GDP)
DEM = democracy
RINT = real rate of interest
EXP = government expenditures/GDP
lnINF=rate of inflation
A look at the data: decile shares
(unweighted)
First
Second
Third
Fourth
Fifth
Sixth
Seventh
Eighth
Ninth
Tenth
Total
Number of
countries
Decile ratio
1988
0.307
0.441
0.539
0.635
0.736
0.855
1.000
1.201
1.541
2.745
1
95
8.9
All countries
1993
0.235
0.375
0.476
0.571
0.677
0.804
0.959
1.182
1.566
3.156
1
113
13.4
1998
0.233
0.380
0.482
0.581
0.686
0.810
0.962
1.181
1.552
3.138
1
113
13.5
Panel (common sample countries)
1988
1993
1998
0.303
0.244
0.233
0.437
0.391
0.387
0.535
0.495
0.491
0.631
0.593
0.590
0.733
0.701
0.697
0.853
0.831
0.821
1.000
0.984
0.972
1.202
1.207
1.188
1.548
1.580
1.553
2.757
2.973
3.068
1
1
1
82
82
82
9.1
12.2
13.2
The openness variable (trade/GDP in %)
Year
Number of
countries
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
124
125
128
129
130
132
130
140
150
152
154
155
155
153
152
149
Average share of
openness
(in percent; all
countries)
72.8
67.5
68.1
69.1
73.5
76.5
75.0
75.9
76.8
79.7
82.8
83.9
84.9
86.8
85.3
91.9
Minimum
(in %)
13 (Lao)
10 (Iran)
10(Sudan)
15(Sudan)
13(Sudan)
14(Brazil)
15(Brazil)
16(Tajik)
16(Japan)
16(Haiti)
16(Brazil)
15 (Brazil)
17 (Brazil)
17 (Brazil)
19(Japan)
20(Japan)
Maximum
(in %)
Average share of
openness
(in percent;
common sample)
317 (S’pore)
70.7
308 (S’pore)
66.0
341(S’pore)
67.0
375(S’pore)
69.5
362(S’pore)
71.6
539(Suriname)
73.3
399(Suriname)
72.6
385(Suriname)
70.1
326(S’pore)
72.6
331(S’pore)
75.7
339(S’pore)
80.2
328(S’pore)
83.1
317(S’pore)
85.0
457(Eq. Guinea)
86.0
313(S’pore)
85.1
341(S’pore)
94.0
Results
• At very low level of GDP per capita, openness
reduces the share of the bottom 7 deciles,
increases the share of the top two
• The negative effect of openness is lessened for
richer countries as the interaction term between
openness and mean income is positive
• Turning point. At income level of Spain and
Israel, there is a reversal: openness increases
relative share of the bottom and middle deciles;
the share of the top declines.
Average level of income and change in
income share by decile
15
second decile
Gain/loss of decile income
10
5
top decile
bottom decile
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000 11000 12000 13000 14000 15000 16000 17000 18000
-5
-10
Mean income from HS (in $PPP)
•Calculated at average decile shares (bottom 2.6%; second 4%, top 30%),
Δtrade=0.2.
Further results
• Higher real rate of interest reduces the share of
bottom 8 deciles, increases the share of the top two
• How strong the effect: each percentage point
increases income of the rich by 1/3 of a percent
• Democracy increases the share of the middle
income deciles (4th to 8th), reduces the share of the
top
• Inflation reduces the shares of bottom and middle
deciles
• Government expenditures raise the shares of the
bottom groups. A 10% increase in G raises the
share of the bottom decile by 0.2 percent (1/10 of
its income)
• Financial deepening raises the share of the bottom
groups
• Only the middle income countries behave as we
would expect from neoclassical theory (in poor
countries, openness Gini, in rich it Gini)
Why Dollar & Kraay do not find the
effect of openness on inequality?
• Largely, because they use a wrong measure:
trade/GDP in PPP terms
• It is a wrong measure for poor countries if
we want to study inequality
• Chinese barbers' output is assessed at
international prices, but they are paid
CHINESE wages (not internatonal)
• So for inequality, the latter matters!
Measurement of openness
USA
.25
.3
Three measures of
openness:
.2
1) trade/GDP in current $
.1
.15
2) trade/GDP in constant $
1975
1980
1985
1990
1995
2000
year
(xpo+mpo)/gdp
trade/GDP in constant prices from PWT6.1
(xpo+mpo)/GDPPP w ith xpo,mpo in 95 prices
3) trade/GDP in constant
$PPP
But in poor countries, hardly any increase in
openness according to the third (PPP) measure
India
.05
.1
.1
.2
.15
.3
.2
.4
.25
.5
.3
China
1980
1985
1990
year
1995
2000
(xpo+mpo)/gdp
trade/GDP in constant prices from PWT6.1
(xpo+mpo)/GDPPP w ith x po,mpo in 95 prices
1980
1985
1990
year
1995
2000
(xpo+mpo)/gdp
trade/GDP in constant prices from PWT6.1
(xpo+mpo)/GDPPP w ith xpo,mpo in 95 prices
Wrong measure for the study of inequality!
To wrap up…moving towards an integrated
approach to growth and distribution?
• Connection between endogeneous or neo-classical
growth theories; convergence literature; Quah’s
“twin peaks”; transition matrices; global
inequality…
• If we know that determine growth and withinnational distribution => we have distribution of
countries by their Ypc and distribution of people
• Global distribution of national incomes (Concept
2) or people (Concept 3) is directly derived
But we should not expect a “grand theory” and
let’s end up with Tocqueville’s warning…
I hate...these absolute systems which make all the events in
history depend on primary causes, linking one to another by an
inevitable chain, and which, so to speak, take out people from
the general history of mankind. I find them narrow in their
pretended grandeur, and false under their guise of mathematical
truths. I believe, whatever the view of the writers who have
invented these sublime theories to nourish their own vanity and
to facilitate their work, that many of the important historical facts
cannot be explained but by the accidental circumstances, and
that many others remain inexplicable. And that finally, chance, or
rather that mixing of the secondary causes, which we thus call,
since we do not know how to tell them apart, explain a lot of
what we see on the world stage. But I strongly believe that
chance does not do anything which is not prepared in advance.
The existing reality, nature of the institutions, state of mind of
people, customs, are the raw materials with which chance
constructs the facts which surprise and awe us.