Background Image
Table of Contents Table of Contents
Previous Page  86 / 145 Next Page
Information
Show Menu
Previous Page 86 / 145 Next Page
Page Background

THE STATE OF THE EUROPEAN UNION

86

higher than the decrease in the inequality be-

tween countries.

If the Europe-wide inequality were measured

by forming the EU quintiles with approximately

100 million people each through different ways

and subsequently calculating the ratio between

the income of the richest and the poorest quin-

tile, we would obtain other S80/S20 values, de-

pending on the approach.

1. If we added up complete countries (or parts

of them) so as to form the EU quintiles, we

wouldn’t take into account the inequality

within each one of those countries.

2. If we added up regions so as to form the

EU quintiles, we wouldn’t take into ac-

count the inequality within each region.

3. ����������������������������������������

If we added up the poorest and the rich-

est quintiles of every country, in the same

way the official Eurostat value does, we

wouldn’t take into account the inequality

between countries.

4. Our approach (

Tables 8

and

9; Chart 2

)

takes both dimensions into consideration

(1 and 3).

The following

Table 10

shows an overview

of the different approaches and their resulting

values and modifications. In the 1

st

and the 4

th

approach, values are shown in purchasing pow-

er parities (PPP) and exchange rates (

).

The Europe-wide inequality has decreased in

all approaches except in the official (methodi-

cally wrong) Eurostat value, in which the grow-

ing internal inequality of the states controls the

trend because catch-up processes are not taken

into account.

Chart 3.

Development of the dispersion (standard deviation) of the regional income per head (in €).

Source:

Eurostat and author’s calculations.

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

13000

12500

12000

11500

11000

10500

10000

9500

9000