In Italy, the proportion of people with income below the poverty threshold is markedly higher in the south of the country, in both the Sud and Isole (Island) regions (together the Mezzogiorno), where the fi gure reaches some 34–35%, as compared to only 10–11% in the north of the country, refl ecting the relatively low level of income per head (almost 25% below the national average). Accordingly, around 63% of the population in Italy with poverty levels of income live in the Mezzogiorno, almost twice its share of the country’s total population.
In Hungary, the risk of poverty in Közép-Magyarország, the region where Budapest is situated, is much less (8%) than in the rest of the country, especially than in Alföld és Észak in the north and east, where the rate reaches 23%, refl ecting the much lower level of income per head in the latter.
Unlike in Hungary or other countries, the risk of poverty in Austria is (at 15%) highest in the region that contains the capital, Vienna — the eastern region (Ostös- terreich); but this region also includes Burgenland, an Objective 1 region in the past. Since this region has a higher level of income per head than the rest of the country, the relatively large proportion of people with income below the poverty line suggests that it also has a more unequal distribution of income.
In Poland, there is a relatively close association between the risk of poverty in the different broad regions and median income levels, with the lowest risk being in Południowy (which contains Silesia and the city of Krakow), where median income per head is highest. The risk is only slightly higher in Centralny, where Warsaw and Lodz are situated and where median income per head is marginally less. It is highest (25%) in Wschodni, in the agricultural east of the country, where income per head is lower than in the other NUTS 1 regions.
Similarly, in Finland, the proportion of the population below the poverty line is smallest (11%) in Etelä-Suomi in the south of the country, where Helsinki is located and where income per head is well above that in other regions. It is highest (17%
and 14%, respectively, though these proportions are not signifi cantly different from each other) in Itä-Suomi in the east of the country and Pohjois-Suomi in the north, which have the lowest levels of income per head.
Table 4.3: Measures of the degree of inequality of income distribution in EU regions, 2005 Measures of income
dispersion
95% confi dence interval for S80/S20
Risk of poverty (people with income
< 60% median)
Country, regions P90/P10 S80/S20 Lower Upper % total
Belgium 3.3 4.2 4.0 4.3 15
Bruxelles-Capitale 4.8 11.1 8.7 12.3 26
Vlaams Gewest 3.0 3.6 3.7 3.6 11
Région Wallonne 3.2 3.7 3.7 3.8 17
Czech Republic 2.8 3.5 3.5 3.6 10
Praha 3.4 4.2 3.9 4.4 5
Stredni Cechy 2.8 3.7 3.6 3.8 8
Jihozapad 2.5 2.8 2.8 2.8 6
Severozapad 3.0 4.1 3.7 4.4 16
Severovychod 2.7 3.2 3.1 3.3 9
Jihovychod 2.6 3.0 2.9 3.0 8
Stredni Morava 2.7 3.3 3.2 3.3 11
Moravskoslezsko 3.1 3.5 3.5 3.5 16
Germany 3.1 4.1 4.1 4.0 13
Baden-Württemberg 3.0 3.9 3.9 3.9 11
Bayern 3.2 4.3 4.3 4.3 13
Nordrhein-Westfalen 3.0 4.4 4.7 4.1 11
Hessen + Rheinland-Pfalz + Saarland 3.2 4.3 4.5 4.2 11
Berlin + Brandenburg + Mecklenburg- Vorpommern + Sachsen + Sachsen-
Anhalt + Thüringen 2.8 3.5 3.5 3.6 15
Bremen + Hamburg + Niedersachsen
+ Schleswig-Holstein 3.0 4.0 4.0 4.0 12
Greece 4.6 6.1 6.1 6.0 21
Voreia Elláda 4.5 5.6 5.6 5.7 25
Kentriki Elláda 4.8 6.5 6.6 6.5 28
Attiki 4.2 5.3 5.3 5.3 13
Nisia Aigaiou, Kriti 4.0 5.5 5.6 5.5 18
Spain 4.3 5.3 5.2 5.3 20
Noroeste 3.8 4.7 4.6 4.7 20
Noreste 3.8 4.4 4.4 4.4 11
Comunidad de Madrid 4.3 5.1 5.1 5.1 13
Centro (ES) 4.4 5.5 5.5 5.5 29
Este 3.9 4.6 4.6 4.6 14
Sur 4.5 5.6 5.6 5.5 29
Canarias 4.4 5.6 5.7 5.6 28
Measures of income dispersion
95% confi dence interval for S80/S20
Risk of poverty (people with income
< 60% median)
Country, regions P90/P10 S80/S20 Lower Upper % total
France 3.2 4.0 3.9 4.0 13
Île de France 3.7 4.5 4.5 4.4 11
Bassin Parisien 2.9 3.8 3.5 4.0 11
Nord-Pas-de-Calais 3.0 3.7 3.6 3.9 15
Est 2.8 3.4 3.4 3.4 14
Ouest 2.9 3.5 3.4 3.5 12
Sud-Ouest 3.4 4.0 3.9 4.0 16
Centre-Est 3.0 3.9 3.9 4.0 10
Méditerranée 3.5 4.1 4.0 4.1 20
Italy 4.3 5.5 5.5 5.5 20
Nord-Ovest 3.6 4.7 4.6 4.7 11
Nord-Est 3.4 4.4 4.4 4.4 10
Centro (IT) 3.9 4.9 4.8 4.9 14
Sud 4.7 5.9 6.0 5.9 34
Isole 4.9 6.1 6.2 6.1 35
Hungary 3.8 5.5 5.4 5.5 16
Közép-Magyarország 3.6 5.2 5.3 5.2 8
Dunántúl 3.4 4.7 4.6 4.9 13
Alföld és Észak 3.8 5.7 5.5 5.7 23
Austria 3.0 3.7 3.6 3.7 13
Ostösterreich 3.4 4.2 4.2 4.2 15
Südösterreich 2.8 3.2 3.2 3.2 13
Westösterreich 2.8 3.4 3.3 3.4 10
Poland 4.6 5.6 5.6 5.6 19
Centralny 5.5 6.9 6.8 6.9 17
Południowy 4.4 5.1 5.1 5.1 15
Wschodni 4.1 5.1 5.1 5.1 25
Północno-zachodni 4.0 4.9 4.8 4.9 19
Południowo-zachodni 5.2 6.0 6.0 6.1 19
Północny 4.5 5.3 5.3 5.3 21
Finland 3.0 3.6 3.6 3.7 13
Itä-Suomi 2.8 3.5 3.4 3.8 17
Etelä-Suomi + Åland 3.0 3.8 3.7 4.0 11
Länsi-Suomi 2.9 3.3 3.3 3.3 13
Pohjois-Suomi 2.7 3.3 3.3 3.4 14
Source: Eurostat, EU-SILC, 2006
In general, the measures of income dispersion support the implications for income distribution noted above. In particular, the degree of inequality of income distri- bution is much greater in the Brussels region than in either Vlaams Gewest or Wallonne, the S80/S20 ratio being around three times higher than that in the latter two regions (Table 4.3).
In addition, the degree of inequality is also greater in the Mediterranean region of France than in Nord-Pas-de-Calais, though less so than between the two Belgian regions. It is less, however, than in Ile de France, which has the widest dispersion of income in the country, as does Praha in the Czech Republic, despite the rela- tively low risk-of-poverty rates in both cases.
This, however, is not a universal feature of capital city regions. Even though it is also evident in Poland (the S80/S20 ratio being 7 in Centralny, as against around 5 in most other regions) and in Etelä-Suomi in Finland, it is not evident in Hungary, Spain or Greece.
What is evident, on the other hand, as implied by the above, is the general absence of a relationship between the degree of income inequality and the risk-of-poverty rate. In a number of countries, therefore, the region with the widest dispersion of income has the smallest proportion of people with income below the poverty line
— the Czech Republic, France, Poland and Finland — while in others the region with the widest dispersion also has the largest share of the population at risk of poverty (Belgium, Spain and Italy).
Concluding remarks
The estimates of disposable household income at the regional level that can be obtained from the EU-SILC data, although far from complete, reveal interesting differences in the risk of poverty between regions within countries. These differ- ences are not, in all countries, in line with differences in average household income levels and still less with those in GDP per head, which is a commonly used indicator of regional incomes but which can be affected signifi cantly by income transfers into and out of regions and, in some cases much more importantly, by both inward and outward commuting.
The data on household disposable income compiled by the EU-SILC seem to show a similar pattern of average income levels across regions in many (but not all) countries and, moreover, in some countries show a markedly different level of income relative to the EU average.
The data also show signifi cant differences in the distribution of income between regions of particular countries, Belgium being the main example, but also Italy.
Such differences are, in some cases, refl ected in variations across regions in the risk-of-poverty rate (again, such as in Belgium and Italy), whereas in others they are not, such as in the Czech Republic, France or Finland.
Terry Ward
1The usual means of measuring people’s standard of living and their vulnerability to poverty, deprivation and social exclusion is by reference to the disposable income that they have access to. The earlier chapters of this report, therefore, focused on both the distribution of income within Member States (and, more widely, across the EU), as an indicator of the degree of inequality in purchasing power, and the relative number of people with income at levels that put them at risk of poverty.
The underlying assumption throughout the analysis was that income was a suitable measure of these aspects. There are a number of reasons why this might not be the case.
As noted above, the fact that the indicator used to identify the risk of poverty in the above chapters is a relative measure, expressed in relation to median incomes, means that it takes no account of the level of the median itself and what this is capable of purchasing. In addition to this, however, there are a number of poten- tial problems with the use of income itself, as conventionally measured, as an indicator of purchasing power, or the command over resources, which is the ultimate concern of social policy-makers.
In particular, income as such takes no account of accumulated savings and wealth
— except in the form of the interest they generate — which can equally be used to purchase goods and services. Moreover, since it is generally measured on an annual basis, the fi gure reported for any 12-month period does not take account of year-to-year fl uctuations in the amount received, which might be substantial for some of the self-employed especially, and, accordingly, is not necessarily a good guide to long-term — or ‘permanent’ — income and, therefore, the command over resources. Equally, it does not incorporate any income in kind in the form, for example, of free or subsidised goods and services, or food and other goods produced for one’s own consumption.
While these factors are reasons why income may understate purchasing power, there are also reasons why it might overstate it. For example, the income received in a given year might be much higher than that received in previous years and, accordingly, might be accompanied by the need to service signifi cant debts, so reducing the amount that can be used for purchases. Alternatively, it might be that people have unusually high expenses or living costs that they need to meet because of their personal or family circumstances — such as having a disability, a
1 With the assistance of Mayya Hristova and Fadila Sanoussi, Applica.
need to care for a relative, or high housing costs — which reduces their effective purchasing power once these commitments have been met.
We examine the information on the extent of material deprivation, which can be used to supplement the data on disposable income in order to gain additional insight into the purchasing power of households and individuals across the EU.