RELATIVE TO WHAT?: CROSS-NATIONAL PICTURE OF EUROPEAN POVERTY MEASURED BY REGIONAL, NATIONAL AND EUROPEAN STANDARDS

ABSTRACT

In this paper three methods of relativizations are used: (1) we apply the conventional poverty approach: the poor are those whose income remains below 60 percent of the national equivalent disposable income, (2) we collapse nations together into one data pool and calculate a poverty line for the EU, and (3) we decompose nation states into smaller units representing the poorest and richest areas. Within-nation differences seem to be more pronounced than differences between nations. In the Nordic countries incomes between regions as well as between individuals are more evenly distributed and, consequently, the national means are more representative for the whole countries. Moreover, the Nordic cluster, together with central Europe, is robust against the method of comparison. The method affects the Mediterranean countries. The use of the European poverty line leads to poverty rates two to three times higher than analyses based on national data. The regional variation in these countries is the widest. For this reason, conclusions based on national means may be misleading and national means obscure more than they reveal. In societies with large socio-economic and regional variation in income, and consequently in consumption capacities, purchasing power parities implicitly assuming homogeneous consumption patterns over society may give a distorted picture of the price levels in the country in question.

Keywords:

• poverty
• income
• income comparisons
• Europe
• nation state
• regions

Notes

Earlier versions of the paper have been presented at the EU COST A15 meetings in Oslo, Spring 2002 Urbino, Autumn 2003 and Nantes, Spring 2004. The paper has also been discussed at the Danish National Social Research Institute, Autumn 2004. Our colleagues at the Swedish Institute for Social Research, University of Stockholm and Turku Center for Welfare Research have provided insightful comments. We want to heed our collective thanks to all commentators and the anonymous referees of European Societies. We thank Nicol Foulkes for improving our English.

²On the philosophical side, John Rawls (1972, 1999), for example, has touched upon these issues in his analyses on social justice.

³A good example is served e.g., by managers of multinational companies. They are comparing their salaries with earnings of their international colleagues living in richer countries not with the incomes of workers living in their own country. Interestingly enough, they do not apply the same strategy of comparison when it comes worker's wages that are usually contrasted against wage levels in poorer countries.

⁴We are thankful to Axel West Pedersen for this.

⁵Furthermore, it had been possible to go deeper and to try to develop regional purchasing power parties that would take regional price differences into consideration. Obviously, these more detailed analyses would have diminished regional differences somewhat but the main story would have been the same (cf. Siminski and Saunders 2004).

⁶The EURO transformations are derived from the LIS files.

⁷The shift from the previous 50 percent poverty line to the 60 percent line does not essentially change the rankings of countries. However, interesting changes do take place: poverty rates for the ‘low poverty countries’ are doubled, whereas poverty in ‘high poverty countries’ does not increase in the same pace. Consequently, the coefficient of variation between countries will diminish from .41 for 50 percent to .31 for 60 percent poverty lines. Thus, the increase of poverty line will squeeze differences between countries and display the poverty situation in European countries more similar. The higher the poverty threshold, the more homogenous the European countries seem to be no matter what kind of social policies they do apply.

⁸The interpretation of the Tukey-boxes is as follows: The upper boundary of the box is set at the 75 percentile and the bottom boundary represents the 25 percentile. Thus, half of the cases are within the box (or if the variation is small all cases can be within the box). The median values are indicated by the horizontal lines inside the boxes. The lines (‘whiskers’) drawn from the upper and lower edge of the percentile box represent cases that are not outliers, e.g., cases with values less than 1.5 box-lengths either form the upper or the lower boundary of the box. Cases deviating more than 1.5 lengths are classified as outliers and marked by circles (as Italy and Spain).

⁹Needless to say that of purely geographical reasons regional differences may be smaller in small countries than in ‘long countries’. But geography is not enough to explain within-country differences, cf. e.g., Italy and Sweden.