Merz: The Distribution of Income of Self-employed, Entrepreneurs and Professions
indepth decomposition inequality analysis of the aggregated groups and of the single
professions based on an inequality generalized entropy decomposition approach.
The remainder of the study is organized as follows:
After discussing the specific problems and requirements when measuring income of the
self-employed and professions, the income data situation in Germany is briefly
described. Since tax statistics are of particular importance and suitability for our
distributional purposes the overall situation of income and tax revenues is sketched and
then our microdata base, the German Income Tax Statistic 1992 is characterized. The
results are discussed within two main chapters, one for the income distribution of
employees and self-employed (entrepreneurs and professions), the other for profession
with 14 single subgroups. Additionaly to the overall picture of predominant income
from different sources in Germany 1992, in each chapter the respective distribution of
individual net income, the redistributional effects and the decomposition of inequality
by the Theil index inequality shares are analyzed. One overall striking result is: the
occupational status as an employee, entrepreneur or as a profession with its connected
low between inequality share is by far not the overall driving factor to ‘explain’ the
overall income distribution and inequality picture of the re-unified Germany Germany;
it is the within group inequality which counts.
2 Measuring Income of Self-employed and Professions:
Problems and Requirements Encountered
There are a number of reasons why income distribution analyses are missing for the
self-employed. The reasons may be summarized as reporting and measurement together
with small sample problems which would bias the real picture with misleading results.
Traditional income analyses focus on income from dependent work only with the
argument that the self-employed are distorting the overall and their income distribution
picture, because many of those reporting zero, negative or very low incomes also
exhibit relatively high standard of living as measured by consumption and/or
expenditures. This argument has reinforced the widespread view that self-employed
people under-report their earnings. However, many of these assumptions about self-
employed earnings are untested, or could be circumvented by a proper and sound
microdatabase.
Problems with measuring self-employed earnings include the following2
?? Differential response rates
?? Time lags between the accounting and the survey periods
?? Concept of earnings and the measurement of profits
?? Treatment of taxes
?? Under-reporting of income
?? Definition of self-employment
?? Definition of professions
?? Small population and sample group sizes.
2 For a more detailled discussion within the UK data situation see e.g. Eardley and Corden 1994