Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created...

26
Gini index and Robin Hood index Rafal Kucharski

Transcript of Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created...

Page 1: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Gini index and Robin Hood index

Rafał Kucharski

Page 2: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Notation

I Let N = {1, 2, . . . , n} be a set of individuals.I Incomes: x = (x1, . . . , xn) ∈ Rn

+\{0}.I Value xi denotes the income of the agent i .I For x ∈ Rn let x∗ denote the vector obtained form x by sorting its

element in non-decreasing way:

x∗1 ≤ · · · ≤ x∗n .

I The group of permutations on N: Sn = {σ : N → N : σ is bijective}.

Page 3: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Gini indexI Discussed in 1876 by F. R. Helmert (geodesist, discoverer of the

chi-squared distribution?),I „Rediscovered” by Corriado Gini in 1912.I For integrable random variable X :

G (X ) =E|X − Y |2E(X )

,

where Y is an independent copy of X .I Discrete equivalent:

G (x) =

∑ni ,j=1 |xi − xj |2n∑n

i=1 xi, x ∈ Rn.

I Another abstract formula:

G (X ) = 1− E(min(X ,Y ))

E(X ),

where Y is an independent copy of X .

Page 4: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

I The attractiveness of the Gini index to many economists is that ithas an intuitive geometric interpretation: it can be defined as theratio of the area that lies between the line of perfect equality and theLorenz curve, over the total area under the line of equality.

0 0.2 0.4 0.6 0.8 10

0.2

0.4

0.6

0.8

1

Page 5: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Lorentz Curve

I Lorenz curve:{(F (x), L(F (x))) : x ∈ R},

where F is the cdf of X and

L(F (x)) =E(X · 1(X ≤ x))

E(X )=

∫ x−∞ t dF (t)

E(X ).

Page 6: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Gini index – Sen formula

I Sen [9] defined the Gini index as a function G : Rn+\{0} → R such

that

G (x) =1n

[n + 1− 2

∑ni=1(n + 1− i)x∗i∑n

i=1 x∗i

]. (1)

I For this discrete version, 0 ≤ G (x) ≤ n−1n for x ∈ Rn

+\{0}.I Equivalent formula for non-increasingly ordered x∗:

G (x) =1n

[n + 1− 2

∑ni=1 ix∗i∑ni=1 x∗i

]. (2)

Page 7: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Axioms

In Plata-Pérez et. al. [2015], the following properties were defined:

1. The inequality index I is said to be scale-independent if

I (λx) = I (x) for all x ∈ Rn+\{0} and λ > 0.

2. The inequality index I is symmetric if and only if

I (xσ) = I (x) for every σ ∈ Sn and x ∈ Rn+\{0}.

3. The inequality index I is comonotone separable if

I (βx + (1− β)y) = βI (x) + (1− β)I (y)

for every β ∈ [0, 1], every x , y ∈ Rn+\{0} that are comonotone and∑n

i=1 xi =∑n

i=1 yi . (We say that x and y are comonotone if(xi − xj)(yi − yj) ≥ 0, for all i , j ∈ N.)

Page 8: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Axioms cont.

4. The inequality index I is said to be standardized if I (vk+1) =k

n,

where

vk+1 =

(0, . . . , 0︸ ︷︷ ︸

k

,1

n − k, . . . ,

1n − k︸ ︷︷ ︸

n−k

), k = 0, . . . , n − 1,

The idea behind this axiom is such that:

I (v1) = I ( 1n , . . . ,

1n ) = 0, I (vn) = I (0, . . . , 0, 1) = n−1

n ,

and the differences

I (v1)− I (v2) = I( 1n , . . . ,

1n

)− I(0, 1

n−1 , . . . ,1

n−1

),

...

I (vn−1)− I (vn) = I(0, . . . , 0, 1

2 ,12

)− I(0, . . . , 0, 1

)are all equal.

Page 9: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Characterization

Theorem (Plata-Pérez et. al. 2015)Let I : Rn

+\{0} → R. Then I equals the Gini index given by

G (x) =1n

[n + 1− 2

∑ni=1(n + 1− i)x∗i∑n

i=1 x∗i

],

if and only if it satisfies the axioms of scale independence, symmetry,comonotone separability and standardization.

Page 10: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Proof idea

If I : Rn+\{0} → R is any index satisfying the four axioms, then using the

symmetry and scale independence axioms we have:

I (x) = I (x∗) = I

(x∗/

n∑j=1

xj

), x ∈ Rn

+\{0}.

The key idea is to show that Axioms 4 and 3 characterize the Gini indexon the convex subset

K =

{y ∈ Rn

+ :n∑

i=1

yi = 1 and y1 ≤ y2 · · · ≤ yn

}.

Page 11: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood index

I also known as:I Hoover index (Edgar Malone Hoover jr., 1936),I Schutz (1951) [7]I Ricci index, (Umberto Ricci 1879–1946)I Pietra ratio, (G. Pietra)I Lindahl index (Erik Robert Lindahl 1891–1960),

I The share of total income which would have to be transferred fromhouseholds above the mean to those below the mean to achieve astate of perfect equality in the distribution of incomes:

H(x) =

∑{i :xi≥x} (xi − x)∑n

i=1 xi.

I Higher values indicate more inequality and that more redistribution isneeded to achieve income equality [6].

Page 12: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood index – alternative formulas

H(x) =

∑{i :xi≥x} (xi − x)∑n

i=1 xi.

I Since∑{i : xi≥x} (xi − x) = −

∑{i : xi≤x} (xi − x):

H(x) =

∑ni=1 |xi − x |2∑n

i=1 xi. (∗)

I Dividing nominator and denominator by n:

H(x) =1n

∑ni=1 |xi − x |2x

=MAD(x)

2x. (∗∗)

I Dividing by the total income inside the absolute value in (∗):

H(x) =12

n∑i=1

∣∣∣∣ xi∑ni=1 xi

− 1n

∣∣∣∣ . (∗ ∗ ∗)

Page 13: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood index – in L1 space

I Let us think about Rn as the normed space L1:

‖x‖1 =n∑

i=1

|xi |.

I For x ∈ Rn+\{0} we have xi ≥ 0, i = 1, . . . , n, hence

‖x‖1 =n∑

i=1

xi .

I Denoting by 1n = ( 1

n , . . . ,1n ) ∈ Rn, we can rewrite (∗ ∗ ∗) as:

H(x) =12

∥∥∥∥ x

‖x‖1− 1

n

∥∥∥∥1.

I We can exploit the properties of the L1 norm!

Page 14: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood index and Lorenz curve

I For integrable random variables (referring to (∗∗)):

H(X ) =E|X − E(X )|

2E(X )=

12E∣∣∣∣ X

E(X )− 1∣∣∣∣ .

I We can easily show that (L is a Lorenz curve)

E|X − E(X )| = 2E(X )[F (EX )− L(F (EX ))] (?)

henceH(X ) = F (EX )− L(F (EX ))

(vertical distance between the Lorenz curve and the line of perfectequality at F (EX )).

Page 15: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Proof of (?)

Let us denote µ = E(X ).

E|X − µ| =

∫ ∞−∞|x − µ| dF (x) =

=

∫ µ

−∞(µ− x) dF (x) +

∫ ∞µ

(x − µ) dF (x) =

= µ

∫ µ

−∞dF (x)−

∫ µ

−∞x dF (x) +

∫ ∞µ

x dF (x)− µ∫ ∞µ

dF (x) =

= µF (µ)−∫ µ

−∞x dF (x) + (µ−

∫ µ

−∞x dF (x))− µ(1− F (µ)) =

= 2µF (µ)− 2∫ µ

−∞x dF (x) = 2µF (µ)− 2µL(F (µ)).

since L(F (t)) =∫ t−∞ x dF (x)/E(X ).

Page 16: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood index and Lorenz curve

I Assuming F and L are differentiable respectively at t0 and F (t0),F ′(t0) = f (t0), the slope of tangent at (F (t0), L(F (t0))) is equal:

limt→t0

L(F (t))− L(F (t0))

F (t)− F (t0)=

= limt→t0

1E(X )

∫ t−∞ x dF (x)− 1

E(X )

∫ t0−∞ x dF (x)∫ t

−∞ dF (x)−∫ t0−∞ dF (x)

=

=1

E(X )limt→t0

∫ tt0x dF (x)∫ t

t0dF (x)

=1

E(X )· t0f (t0)

f (t0)=

t0E(X )

.

I The slope of the tangent of the Lorenz curve equals 1 (tangent isparallel to the line of perfect equality) for t0 = E(X ).

I Lorenz curve is convex ⇒ slope is increasing ⇒ vertical distancebetween the Lorenz curve and the line of perfect equality is greatestat F (E(X )).

Page 17: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood index and Lorenz curve

Robin Hood index can be graphically represented as the maximum verticaldistance between the Lorenz curve and the 45-degree line that representsperfect equality of incomes.

0 0.2 0.8 10

0.2

0.4

0.6

0.8

1

H(X )

F (E(X ))

Page 18: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Corollary: Robin Hood ≤ Gini

DA0.2 0.8

0.2

0.4

0.6

0.8

1 C

F

B

E

I Let q = F (E(X )) = |AE |. We have H = |BF |, |BE | = q − H.I |AEB| = 1

2q(q − H) = 12(q2 − Hq),

I |BCDE | = (1− q)1+(q−H)2 = 1

2(1− q2 − H + Hq),I∫ 10 L(t) dt ≤ |AEB|+ |BCDE | = 1−H

2 ,

I G =12−

∫ 10 L(t) dt

12

= 1− 2∫ 10 L(t) dt ≥ 1− 2 · 1−H

2 = H.

I By-product: H = 1− 2(|AEB|+ |BCDE |) = 2|ABC |.

Page 19: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Robin Hood vs Gini: min and max difference

0 0.2 0.4 0.6 0.8 10

0.2

0.4

0.6

0.8

1max: H = 0.5, G = 0.75

0 0.2 0.4 0.6 0.8 10

0.2

0.4

0.6

0.8

1min: H = 0.5, G = 0.5

Page 20: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Attempt to characterize Robin Hood index

I Robin Hood index fulfils:I scale invariance,I symmetry,I standardization.

I Moreover (int-standardization):

H(x) =zeros(x)

n, for x ∈ Jn,

where zeros(x) = #{i : xi = 0} and

Jn = {(x1, . . . , xn) : xi = ki/n, ki ∈ N, i = 1, . . . , n,∑n

i=1 ki = n}.

Page 21: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Attempt to characterize Robin Hood index,cont.

I We say that x , y ∈ Rn+\{0} are co-mean if

(xi − x)(yi − y) ≥ 0, i = 1, . . . , n.

I Axiom 3b: Robin Hood index is co-mean separable:

I (βx + (1− β)y) = βI (x) + (1− β)I (y)

for every β ∈ [0, 1], every x , y ∈ Rn+\{0} that are co-mean and∑n

i=1 xi =∑n

i=1 yi .

HypothesisFunction I : Rn

+\{0} → R equals the Robin Hood index given by (∗ ∗ ∗) ifand only if it satisfies the axioms of scale independence, symmetry,co-mean separability and int-standardization.

Page 22: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Real dataBased on: OECD Income Distribution Database (IDD),http://www.oecd.org/social/OECD2016-Inequality-Update-Figures.xlsx

H = 0.745876G − 0.005591.

Page 23: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Estimation from decilesFrom [5]: The Robin Hood index may also be calculated by:

I summing the percentage of income for each tenth of an incomedistribution where the percentage exceeds 10%,

I and subtracting from this the product of the number of tenths thatmeet this criterion times 10%.

Example

Decile 1 2 3 4 5Percentage of total income 1.08 2.48 4.13 5.74 7.33

Decile 6 7 8 9 10Percentage of total income 8.97 10.83 13.09 16.41 29.93

In this case four deciles (7-10) exceed 10%, so the Robin Hood index

= (10.83% + 13.09% + 16.41% + 29.93%)− (4× 10%) =

= 70.26%− 40% = 30.26%.

Page 24: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Thank you for your attention.

Page 25: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Atkinson A.B., Micklewright J., Economic Transformation in EasternEurope and the Distribution of Income, Cambridge University Press,1992.

Dniestrzański P., Gini Index and Angle Measure as Special Cases of aWider Family of Measurements, preprint, (2015).

Helmert, F. R. Die Genauigkeit der Formel von Peters zurBerechnung des wahrscheinlichen Fehlers directer Beobachtungengleicher Genauigkeit. Astronom. Nachr. (1876) 88, 113–132.

Hoover E. M., The Measurement of Industrial Localization, Review ofEconomics and Statistics (1936) 18, No. 162–71.

Kennedy B.P., Kawachi I., Prothrow-Stith D., Income distributionand mortality: cross sectional ecological study of the Robin Hoodindex in the United States, BMJ 1996; 312:1004, doi:https://doi.org/10.1136/bmj.312.7037.1004,

Inequality Measurement, Development Issues No. 2, 21 October 2015,http://www.un.org/en/development/desa/policy/wess//wess_dev_issues/dsp_policy_02.pdf,

Page 26: Gini index and Robin Hood index · Gini index and Robin Hood index Author: Rafał Kucharski Created Date: 1/17/2018 9:05:43 AM ...

Marshall A. W., Olkin I., Arnold B. C., Inequalities: Theory ofMajorization and Its Applications, 2nd edition, SpringerScience+Business Media, 2011.

Plata-Pérez L., Sánchez-Pérez J., Sánchez-Sánchez F., An elementarycharacterization of the Gini index, Math. Soc. Sci. 74 (2015) 79–83.

Sen A., On Economic Inequality, 1973, Clarendon Press, Oxford.