Ginicoefficient

From Wikipedia, the free cncyclopedia

http:/cn.wikipedia.org/wiki/Gini cocfficicnt

Gini coefficient
100%

Perfect distnbution lne

sometines called 45 degree lune
Cumulative
share

of
income
CHni Index eaned

Lorenz cuve

The cunmulative share of people 100%

from lower incCOme

Graphical representation of the Gini coeficient

The Gini coefficient is a measure of inequa!ity of adistribution. It is defined as a
ratio with values between 0 and 1: the numerator is the area between the Lorenz curve
of the distribution and the uniform distribution line; the denominator is the
area under
the uniform distribution line. It was developed by the Italian statistician Corrado Gini
and published in his 1912 paper "Variabilitàe mutabilità" ("Variability and
Mutability"). The Gini index is the Gini coefficient expressed as a percentage, and is
equal to the Gini coefficient multiplied by 100. (The Gini coefficient is equal to half
of the relative mean difference.)
The Gini coefficient is often used to measure income inequality. Here, 0
corresponds to perfect income equality (i.e. everyone has the same income) and I
corresponds toperfect income inequality (i.e. one person has all the income, while
everyone else has zero income).
The Gini coefficient can also be used to mneasure wealth inequality. This use
requires that no one has a negative net wealth. It is also commonly used for the
measurement of discriminatory power of rating systems in the credit risk
management.

Caleulation
 The Gini
diagram. If the coefficient is as aratio of the areas on the Lorenz curve
area between defined
the line of perfect equality and Lorenz curve is A, and
the area under the
=0.5, the Gini Lorenz curve is B, then the Gini coefficient is
A/(A+B). Since A+B
coefficient, G=2A = 1-2B. If the
fünction Y=L(X), the value of Bcan be Lorenz curve is represented by the
foundwith integration and:
G=1-2
Ih some cases, this equation can be applied to calculate the Gini
coefficient
without direct reference to the Lorenz curve. For
example:
For a population with values y, i = | to n, that are indexed in non-decreasing
order (y, Syr):
1
G=-(n +1-2 (n+1-i)3,
For a discrete probability function f), where y, i = 1to n, are the points with
nonzero probabilities and which are indexed in increasing order (y, <y-i):

G=1 f(y)(S;-1 +S,)

where:

S, = D1 fy)and S% =0

For acumulative distribution function F() that is piecewise diferentiable, has

a mean 4, and is zero for all negative values of y:

G=1 -Fu)'dy
Since the Gini coefficient is half the relative mean difference, it can also be
calculated using formulas for the relative mean difference.

For a random sample Sconsisting of values y, i=lto n, that are indexed in

non-decreasing order (y,<y+), the statistic:

G(S)= (n +| - 2 - (2 1- i)y
 is aconsistent
estimator of the population Gini
unbiased. Like the relative meandifference, therecoefficient,
but is not, in general,
does not exist a sample statistic that
is in general an unbiased estimator of the
population Ginicoefficient.
intervals for the population Gini coefficient can be calculated using Confidence
bootstrap
techniques.
Sometimes the entire Lorenz curve is not known, and only values at certain
intervals are given. In that case, the Gini coefficient can be approximated by using
various techniques for interpolating the missing values of the Lorenz curve. If( Xk,
Yk) are the known points on the Lorenz curve, with the Xindexed in increasing
order ( Xk-|<Xk), so that:

X; is the cumulated proportion of the population variable, for k = 0.,.,.n, with

Xo =0, X, =1.
Y7 is the cumulated proportion of the income variable, for k =0,...n, with Yo
=0, Y, =1.

If the Lorenz curve is approximated on each interval as a line between consecutive

points, then the area B can be approximated with trapezoids and:

GË =1- ( - X;-1) (Y; +Y-)

obtained using
is the resulting approximation for G. More accurate results can be
other methods to approximate the area B, such as approximating the Lorenz curve
with a quadratic function across pairs of intervals,or building an appropriately
that matches the known
smooth approximation to the underlying distribution function
interval are also known,
data. If the population mean and boundary values for each
approximation.
these can also often be used to improve the accuracy of the
iividualu
 While most developed European nations tend to have Ginicoefficients between
0.24 and 0.36, the United States Gini coefficient is above 0.4, indicating that the
United States has greater inequality. Using the Gini can help quantify differences in
welfare and compensation policies and philosophies. However it should be borne in
mind that the Gini coefficient can be misleading when used to make political
comparisons between large and smallcountries (see criticisms section).

Correlation with per-capitaGDP

Poor countries (those with low per-capita GDP) have Ginicoefficients that fall
over the whole range from low (0.25) to high (0.71 ), while rich countries have
generally low Gini coefficient (under 0.40).

Advantages as a measure of inequality

The Ginicoefficient's main advantage is that it is ameasure of inequality by
means of a ratio analysis, rather than a variable unrepresentative of most of the
population,such as per capitaincome or gross domestic product.

Itcan be used to compare income distributions across different population

sectors as wellas countries, for example the Ginicoefficient for urban areas
differs from that of rural areas in many countries (though the United States'
urban and rural Ginicoefficients are nearly identical).

It is sufficiently simple that it can be compared across countries and be easily

interpreted. GDP statistics are often criticised as they do not represent changes
for the whole population; theGini coefficient demonstrates how income has
changed for poor and rich. If the Gini coefficient is rising as well as GDP,
poverty may not be improving for the majority of the population.

The Ginicoefficient can be used to indicate how the distribution of income

has changed withina country over a period of time, thus it ispossible to see if
inequality is increasing or decreasing.

The Gini coefficient satisfies four important principles:

Anonymity: it does not matter who the high and low earners are.
lhaliig ohe
A

ranftud
ncoe hdividal.
a
 Scale
independence:
tne economy,
the Gini
the way it is coefficient does not consider the size of
country on average. measured, or whether it is a rich or po0r
Population independence: it does not matter howlarge the population
of the
country is.
Iransfer principle: if income (less than the difference), is
mom arich person to apoor person the transferred
resulting distribution is more
equal.

Disadvantages as a measure of inequality

The Ginicoefficient measured for a large economically diverse country will
generally result ina much higher coefficient than each of its regions has
individually. For this reason the scores calculated for individual countries
within the EUare difficult to compare with the score of the entire US.

difficult because
Comparing income distributions among countries may be
countries give benefits in the
benefits systems may differ. For example, some
which may not be counted as
form of money while others give food stamps, Gini
therefore not taken into account in the
income in the Lorenz curve and
coefficient.
instead of
results when applied to individuals
The measure willgive different
households. When different populations
are not measured with consistent
meaningful.
definitions, comparison is not
inequality if richer
Lorenz curve may understate the actual amount of
The income
more efficiently than lower
households are able to use income be the
view, measured inequality may
households. From another point of incomes.
efficient use of household
result of more or less
errors in the data.The
statistics, there will be systematic and random
As for all less accurate.
Gini coefficient decreases as the data become
meaningof the compare
collect data differently,making it difficult to
Also, countries may
statistics between countries.

still have very

and Ginicoeflicients can
Economies with similar incomes curvescan have
This is because the Lorenz
different income distributions. coeflicient. As an extreme
yield the same Gini
different shapes and yet still
 example, an economy where half the households have no income, and the
other half share income equally has a Gini coefficient of ½; but an economy
with complete incomeequality, except for one wealthy household that has half
the total income, also has a Gini coefficient of 2.

Too often only the Gini coefficient is quoted without describing the
proportions of the quantiles used for measurement. As with other inequality
coefficients, the Gini coefficient is influenced by the granularity of the
neasurements. For example, five 20% quantiles (low granularity) will y ield a
taken from
lower Ginicoefficient than twenty 5% quantiles (high granularity )
the same distribution.

additionally to or in competition with the Gini

Asone result of this criticism,
measures are frequently used (e.g. the Atkinson and Theil indices).
coefficient entropy players
attempt to compare the distribution of resources by intelligent
These measures distribution, which would occur if
maximum entropy random
in the market with a closed systemfollowing the laws
non-intelligent particles in a
these players acted like
of statistical physics.