Income Distribution in European Countries
Опубликовано на портале: 13-12-2003
Luxembourg Income Study Working Paper Series. 1995. No. 121.
Empirical facts are treacherous objects. The subject of income distribution is littered with "facts" that have ceased to hold or which proved on closer examination to be mere statistical artefacts rather than genuine economic regularities (remember Keynes' constancy of the share of labour?). The value of empirical generalisations has more often been found in the theoretical process used to explain them than in the empirical observation itself. It can be argued that the - very substantial - contribution of Kuznets' Presidential Address (1955) lay in his analytical framework rather than in the celebrated Kuznets curve indicating that inequality first rises and then falls as a country develops. This paper starts therefore from a position of modesty as to what can be achieved by a summary of the empirical evidence. Not the least of the reasons for this are the difficulties in making such comparisons, and it is with their limitations that the paper begins in Section 1.
The second difficulty in writing a paper on empirical facts is that these have many dimensions. Here we have chosen to concentrate on the comparison of income inequality across countries, and across time. Sections 2 and 3 summarise evidence for fourteen of the European countries about the extent of income inequality in the 1980s, drawing largely on data from the Luxembourg Income Study (LIS). (The country not covered in these sections is Denmark.) Do all distributions have broadly the same shape? Can one identify distinct groupings of countries with different degrees of inequality?
Sections 4 and 5 are concerned with the changes over recent years. Is there a world-wide trend towards greater inequality? In addressing this question, use is made not only of the LIS data but also of the findings of individual national studies in different European countries (now including Denmark). These studies, carried out by government statisticians or academic researchers, are not typically designed to be comparable with those in other countries, but are an important complement to the LIS data, providing a point of comparison and typically allowing a longer time series to be examined.
Even within the field delimited above, any compilation of empirical facts is bound to be selective, and there is a high probability that the tables and graphs in this paper fail to answer the particular questions in the reader's mind. It is in fact our belief that this method of dissemination should be supplemented by one which takes advantage of modern micro-technology. We should move to a situation where subsets of variables from full micro datasets are made available in a form where the reader can choose, within limits, the method of presentation. This however requires a change in the rules of official statistical agencies, and for the present we can only apologise for the straitjacket imposed on the micro-data.
Finally, it should be noted that the title refers to income inequality in European countries, not to income inequality in Europe. It is a challenging project to produce income distribution estimates for Europe as a whole, but that is the subject of another paper.
The Commission launches several procedures involving labour law against Italy, Ireland, Greece and the Netherlands (Комиссия запускает процедуры трудового права против Италии, Ирландии, Греции и Нидерландов)
IMF ,Working Paper. 2006. No. 06/296.