Gini Index of Positive Adjusted Gross Income of all IRS Tax Returns, 1990-2008

Thomas Volscho (CUNY-College of Staten Island)

updated Feb 8, 2011

Year Gini

1990 0.529
1991 0.525
1992 0.532
1993 0.547
1994 0.532
1995 0.540
1996 0.552
1997 0.560
1998 0.566
1999 0.576
2000 0.584
2001 0.561
2002 0.550
2003 0.555
2004 0.571
2005 0.587
2006 0.594
2007 0.601
2008 0.584
http://scholar.library.csi.cuny.edu/~volschot/giniirs.html

The Distribution of Income

 Table 1 Distribution of U.S. Household Income in 2003
Group of Share of Lower Limit
Households Aggregate Income of Each Fifth
Poorest fifth 3.4%
How is income distributed
across households? Where Second fifth 8.7% $17,984
do you stand, in terms of Middle fifth 14.8% $34,000
outcomes of the
distributional process? Is Fourth fifth 23.4% $54,453
your family in the top, Richest fifth 49.8% $86,867
middle, or lower portion of
the income distribution? Richest 5% 21.4% $154,120
Source: U.S. Census Bureau, Historical Income
The U.S. Census Bureau Tables – Households, Tables H-1, H2.
has, for a number of
decades, published information on the distribution of incomes in the United States, as
shown in Table 1 for 2003. The Census Bureau measures incomes by summing up
households’ incomes from wages and salaries, rent, interest, and profits, and cash transfer
payments received from government agencies.

To understand what this table means, imagine dividing up U.S. households into five
equal-sized groups (called “quintiles”), with the poorest households all in one group, and
then the next poorest in the next group, and so on. The last group to be formed has the
richest one-fifth (or 20%) of households. The highest-income household in the poorest
group would, according to Table 1, have an income just short of $17,984. This group, the
poorest fifth, received 3.4% of all the household income in the country. The richest fifth,
those with incomes of $86,867 or more, received 49.8%—essentially half—of all the
income received in the United States

Suppose we look at just the top 5% of households by income. Households in this very top
group have annual incomes above $154,120. In 2003, this group—containing one-
twentieth of the total population—received just over one-fifth of the total income in the
country.

Measuring Inequality

Figure 1: Lorenz Curve for U.S. Household Income, 2003. A Lorenz curve is a way of
graphically portraying an income distribution. For example, point C indicates that the
poorest 60% of households received about 27% of total household income. If income
were perfectly equally distributed, the Lorenz curve would be a straight line from the
origin to point F. (Source: GDAE)
Economists frequently use a graph called the Lorenz curve—named after the statistician
who first developed the technique—to describe the pattern of inequality within an
economy. A Lorenz curve for household income in the United States, based on the data in
Table 1, is shown in Figure 1. To construct this curve, you first draw a rectangle, as
shown in the figure. The horizontal axis represents households, lined up from left to right
in order of increasing income. The vertical axis measures the cumulative percentage of
total income received by households up to a given income level.

In our example, the data shown in Table 1 are entered into the Lorenz curve in Figure 1
as follows. First, point A represents the fact that the lowest 20% of households received
3.4 % of total income. Point B indicates that the lowest 40% of households received 3.4%
+8.7% = 12.1% of total income; point C indicates that the lowest 60% of households
received 3.4%+8.7%+14.8% = 26.9% of total income; point D similarly shows the
income of the lowest 80%, and point E the income of the lowest 95%. The Lorenz curve
must start at the origin, at the lower left corner of the square (since 0% of households
have 0% of the total income) and end at point F in the upper right corner (since 100% of
households have 100% of the total income).

If income were distributed equally among all households, the Lorenz curve would be a
straight line connecting the origin and point F (the diagonal line in Figure 1). This line
thus represents a situation of maximum equality. At the other extreme, if one household
received all the income, then the Lorenz curve would hug the horizontal axis until all but
the very last household was accounted for and then shoot up to point F, just in front of the
right-hand-side vertical axis. Such a line would represent a situation of maximum
inequality.

Figure 2: The Gini Ratio, A/(A + B). The Gini ratio (or Gini coefficient) sums up the
income distribution in a single number: the ratio of the area A to the sum of the areas A
and B. If income were perfectly equally distributed, the Gini ratio would be equal to 0.
(Source: GDAE)

In all real situations, Lorenz curves for distributions of income will fall between these
extremes. Graphically, the curve will sag downward to some extent below the diagonal
— as in Figure 1. The more the curve sags, the greater is the extent of inequality in the
income distribution. This observation led an economist by the name of Corrado Gini to
introduce a numerical measure of inequality known as the Gini ratio, which is defined as
the ratio of the area between the Lorenz curve and the diagonal to the total area under the
diagonal line. Referring to areas A and B in Figure 2, the Gini ratio is A/(A+B). Clearly,
the Gini ratio can vary from 0 for perfect equality to 1 for complete inequality. The Gini
ratio for U.S. household income in 2003 was 0.464.

The Gini ratio for the U.S. is higher than that of all other industrialized countries,
signifying that the U.S. has a greater degree of income inequality. The Gini for Canada,
for example, is about 0.32, while the United Kingdom has a Gini of 0.36, Germany about
0.30, and Japan and Sweden both about 0.25. Countries with more unequal distributions
of income than the U.S. tend to be less industrialized countries, like Brazil (0.59) and
Nigeria (0.51).

Perhaps, you might object, something is wrong with the measure of income we are using.
Shouldn’t the effect of tax and transfer programs be more fully included? The U.S.
Census Bureau has experimented with at least 15 different definitions of personal
income, each of which includes a different way of accounting for income, taxes and
transfers. In one definition, for example, it subtracts the value of government transfers
and adds in the value of health insurance fringe benefits paid by businesses for their
(often middle-class or higher) employees and the value of net capital gains (these are
usually earned by the relatively wealthy). Under this definition, the Gini ratio, not
surprisingly, rises to over 0.5, showing greater inequality. The share of the bottom fifth
drops considerably, while the share of the top fifth rises. Another measure adjusts for the
effects of the tax system. This causes some change at the top, but little at the bottom.
When they further add in the effects of both cash and non-cash government transfer
programs (such as food stamps), the Gini ratio drops down closer to 0.4.

Government tax and transfer policies—and especially the transfer side—have significant
effects on the U.S. household income distribution. Even with the most thorough
accounting for transfer aid to low-income households, however, the income of the top
fifth of the population is still roughly ten times that of the bottom fifth.

Some important goods and services are obtained, of course, without the use of cash
income. Many families prefer to produce at least some services (such as child care and
cooking) for themselves. In addition, many of the things we enjoy, such as pleasant parks,
safe roads, or clean air add to our well-being without requiring payments out of our cash
income. If we were to look at the distribution of well-being rather than just the
distribution of income, we would need to take account of these non-income sources of
important goods and services. No such comprehensive study has been done. Some of
these goods may contribute to lessening inequality – for example, everyone, rich or poor,
can enjoy a public park or use a public library. Evidence suggests, however, than at least
in some cases the distribution of such non-purchased goods may accentuate, rather than
lessen, measures of inequality. Proponents of “environmental justice” for example, point
out that polluting industries and toxic waste disposal sites tend to be disproportionately
located near poor and minority communities.

Income Inequality Over Time

Figure 3: Income Shares of the Richest and Poorest Households, 1968-2003. Inequality in
the United States has been increasing since 1968. The share of the richest households in
aggregate income rose from about 17% to about 22%, while the share of the poorest 20%
of the population fell from about 4% to about 3.5%. (Source: GDAE)

The U.S. household income distribution has been recorded every year since 1967. A
similar but not quite identical measure, the family income distribution, has been recorded
since 1947. These data show that inequality was gradually decreasing—that is, income
was becoming more equally distributed—until 1968. In that year, the Gini ratio for
household income was 0.388, the lowest (most equal) on record in the United States.
Since 1968 the Gini ratio has increased in almost every year.

Figure 3 shows what has happened in recent decades at the very top and the very bottom
of the income distribution. The general trend has been for a larger share of income to go
to the very richest households (from about 17% in 1968 to about 22% in 2003), while the
share going to the bottom (and, not shown in this figure, the middle) quintile(s) has
gradually fallen.

Why has income inequality been increasing in the United States over this period? One
point economists agree on is that some of the increase in inequality has been due to
changing demographic characteristics of the U.S. population.

Increases in the proportion of the population that is aged, and increases in single
parenthood, have tended to drive down incomes at the low end. People too old to work
and people in single-parent households (where paid work and caring activities compete
for a limited resource—the adult’s time) often lack economic resources. About 18% of
U.S. children live in poor families. Meanwhile, the entry of women into the labor force in
increasing numbers has helped boost the incomes of married-couple households at the
top. Demographic change, however, is only part of the story and cannot explain the
whole pattern of increasing inequality. Economists continue to debate the relative
importance of at least three other explanations. (Note that all three explanations propose
reasons why the poor have become poorer or more numerous, while the third one also
addresses why the rich have gotten richer.)

First, international trade has been increasing. Competition from imports has eliminated
many industrial jobs that formerly fell in the middle of the U.S. income distribution. If
middle-income industrial jobs are replaced by lower-income service and retail jobs,
inequality will increase.
Second, new technologies such as computers and biotechnology have become more
important, increasing the incomes of skilled workers who understand and use the new
techniques and equipment, while leaving behind the less-skilled workers who remain in
low-technology occupations.

Finally, unions have grown weaker and government policy has become markedly less
supportive of unions and low-wage workers, while the compensation given to top
executives and board members of large corporations has skyrocketed. According to
studies done by Business Week, in 1980 chief executive officers (CEOs) of large U.S.
corporations earned an average of 42 times the amount earned by the average hourly
worker. In 1990, they earned 85 times as much. In 2000, they earned 531 times as much.

In short, along with demographic change, global competition, technology, or changes in

government and business policies—or some combination of these factors—may account
for the rise in inequality within the U.S.

Human Development Report 2009

LDemographic trends
Old age dependency ratio

199 201
HDI RankCountry
0 0
1 Norway 25.2 22.7
2 Australia 16.8 20.7
3 Iceland 16.5 17.4
4 Canada 16.6 20.3
5 Ireland 18.5 16.7
6 Netherlands 18.6 22.9
7 Sweden 27.7 28.1
8 France 21.6 26.2
9 Switzerland 21.3 25.5
10 Japan 17.2 35.1
11 Luxembourg 19.4 20.5
12 Finland 19.9 25.9
13 United States 18.7 19.4
14 Austria 22.1 25.9
15 Spain 20.5 25.3
16 Denmark 23.2 25.6
17 Belgium 22.3 26.4
18 Italy 22.2 31.3
19 Liechtenstein .. ..
20 New Zealand 16.9 19.4
21 United Kingdom 24.1 25.1
22 Germany 21.7 30.9
23 Singapore 7.7 13.8
24 Hong Kong, China (SAR) 12.1 17.0
25 Greece 20.4 27.2
26 Korea (Republic of) 7.2 15.2
27 Israel 15.2 16.4
28 Andorra .. ..
29 Slovenia 16.3 23.5
30 Brunei Darussalam 4.3 4.9
31 Kuwait 1.9 3.2
32 Cyprus 17.3 19.0
33 Qatar 1.6 1.3
34 Portugal 20.3 26.7
35 United Arab Emirates 1.8 1.3
36 Czech Republic 19.0 21.6
37 Barbados 15.1 14.4
38 Malta 15.8 21.2
39 Bahrain 3.4 3.1
40 Estonia 17.5 25.2
41 Poland 15.5 18.8
42 Slovakia 16.0 16.9
43 Hungary 20.1 23.8
44 Chile 9.6 13.5
45 Croatia 16.6 25.6
46 Lithuania 16.4 23.7
47 Antigua and Barbuda .. ..
48 Latvia 17.7 25.4
49 Argentina 15.3 16.6
50 Uruguay 18.7 21.8
51 Cuba 12.7 17.5
52 Bahamas 7.0 10.3
53 Mexico 7.6 10.0
54 Costa Rica 8.4 9.5
55 Libyan Arab Jamahiriya 4.7 6.6
56 Oman 3.6 4.7
57 Seychelles .. ..
58 Venezuela (Bolivarian Republic of) 6.4 8.7
59 Saudi Arabia 4.1 4.6
60 Panama 8.4 10.4
61 Bulgaria 19.7 25.5
62 Saint Kitts and Nevis .. ..
63 Romania 15.8 21.3
64 Trinidad and Tobago 9.2 9.5
65 Montenegro 12.7 18.8
66 Malaysia 6.2 7.3
67 Serbia 14.3 21.1
68 Belarus 16.1 18.6
69 Saint Lucia 13.4 10.1
 70 Albania 8.6 14.4
71 Russian Federation 15.1 17.9
72 The former Yugoslav Republic of Macedonia11.2 16.9
73 Dominica .. ..
74 Grenada 14.8 10.6
75 Brazil 7.4 10.2
76 Bosnia and Herzegovina 8.8 19.6
77 Colombia 7.2 8.6
78 Peru 6.9 9.3
79 Turkey 6.8 8.8
80 Ecuador 7.4 10.6
81 Mauritius 7.1 10.7
82 Kazakhstan 9.3 10.0
83 Lebanon 8.8 10.8
84 Armenia 8.8 16.1
85 Ukraine 18.3 22.1
86 Azerbaijan 6.9 9.5
87 Thailand 7.1 10.9
88 Iran (Islamic Republic of) 6.2 6.8
89 Georgia 14.1 20.7
90 Dominican Republic 6.6 9.8
91 Saint Vincent and the Grenadines 11.0 10.0
92 China 8.3 11.4
93 Belize 7.4 6.7
94 Samoa 7.1 8.6
95 Maldives 5.2 6.4
96 Jordan 6.3 5.9
97 Suriname 7.6 9.9
98 Tunisia 8.0 9.6
99 Tonga 8.0 10.3
100 Jamaica 12.5 12.2
101 Paraguay 7.4 8.4
102 Sri Lanka 8.9 11.4
103 Gabon 10.6 7.2
104 Algeria 6.8 6.8
105 Philippines 5.8 6.9
106 El Salvador 8.6 12.0
107 Syrian Arab Republic 5.4 5.2
108 Fiji 5.3 7.7
109 Turkmenistan 6.8 6.2
110 Occupied Palestinian Territories 6.8 5.5
111 Indonesia 6.3 9.0
112 Honduras 6.6 7.3
113 Bolivia 6.8 8.0
114 Guyana 7.8 9.5
115 Mongolia 7.4 5.8
116 Viet Nam 8.4 9.3
117 Moldova 13.0 15.4
118 Equatorial Guinea 7.6 5.1
119 Uzbekistan 7.3 6.6
120 Kyrgyzstan 8.7 7.7
121 Cape Verde 9.0 6.8
122 Guatemala 6.6 8.2
123 Egypt 6.9 7.3
124 Nicaragua 6.2 7.5
125 Botswana 5.0 6.1
126 Vanuatu 6.8 5.7
127 Tajikistan 7.2 6.0
128 Namibia 6.3 6.1
129 South Africa 5.5 7.1
130 Morocco 6.8 8.1
131 Sao Tome and Principe 8.9 6.9
132 Bhutan 6.1 7.5
133 Lao People's Democratic Republic 6.7 6.1
134 India 6.6 7.7
135 Solomon Islands 5.8 5.4
136 Congo 7.2 6.8
137 Cambodia 5.2 5.6
138 Myanmar 8.4 8.1
139 Comoros 5.9 5.2
140 Yemen 4.2 4.4
141 Pakistan 7.0 6.9
142 Swaziland 5.5 5.9
143 Angola 5.2 4.7
144 Nepal 5.9 6.8
145 Madagascar 6.1 5.6
146 Bangladesh 5.6 6.1
147 Kenya 5.6 4.8
148 Papua New Guinea 3.9 4.3
149 Haiti 7.2 7.3
150 Sudan 5.7 6.4
151 Tanzania (United Republic of) 5.2 6.0
152 Ghana 5.7 6.3
153 Cameroon 7.0 6.4
154 Mauritania 5.2 4.6
155 Djibouti 4.5 5.4
156 Lesotho 8.5 8.4
157 Uganda 5.5 5.2
158 Nigeria 5.7 5.8
159 Togo 6.1 6.3
160 Malawi 5.3 6.1
161 Benin 7.0 6.1
 162 Timor-Leste 3.5 5.8
163 Côte d'Ivoire 5.2 7.0
164 Zambia 5.4 6.0
165 Eritrea 5.1 4.5
166 Senegal 4.9 4.4
167 Rwanda 5.4 4.5
168 Gambia 5.0 5.2
169 Liberia 5.7 5.7
170 Guinea 6.2 6.1
171 Ethiopia 5.5 6.0
172 Mozambique 6.4 6.2
173 Guinea-Bissau 6.5 6.4
174 Burundi 6.0 4.7
175 Chad 6.7 5.5
176 Congo (Democratic Republic of the) 5.5 5.2
177 Burkina Faso 5.1 3.9
178 Mali 5.4 4.3
179 Central African Republic 7.5 6.9
180 Sierra Leone 5.1 3.4
181 Afghanistan 4.5 4.3
182 Niger 4.1 4.1
Other UN member states
Iraq 6.6 5.8
Kiribati .. ..
Korea (Democratic People's Rep. of) 6.8 14.2
Marshall Islands .. ..
Micronesia (Federated States of) 6.8 6.1
Monaco .. ..
Nauru .. ..
Palau .. ..
San Marino .. ..
Somalia 5.6 5.2
Tuvalu .. ..
Zimbabwe 5.8 7.3

Sources :
UN (2009e). "World Population Prospects: The 2008 Revision". New York: Department
of Social and Economic Affairs.
Symbols :
.. Data not available
Greater (or less) than zero but small enough to be rounded off to zero at the displayed
(.)
number of decimal points
< Less than
- Not applicable
T Total
http://hdrstats.undp.org/en/indicators/147.html