Can We Really Rely on Income Distribution Statistics? Some Issues in the Swedish Data

Authors: Anders Björklund
Series: Issue: 7 2020
Themes:
Download Publication (PDF)

The Swedish Income Distribution Statistics have shown rising gaps in disposable income since the early 1980s. Several reports have shown that capital income is an important driver behind this development. I identify several weaknesses in the measurement of capital income in these statistics. One weakness is that realised capital gains, which generally are included in Swedish reports on income distribution but not in cross-national ones, are not adjusted for inflation from 1991 and onwards but were so until 1990. A considerable part of the capital gains included in data is thus compensation for inflation and not real gains. Further, there is a considerable rise in income shifting from earned income to capital income in closely held corporations and a decline of income shifting from capital income to pension income. New modes of saving, which are taxed according to a standard revenue principle, will most likely create severe problems for the Statistics in the near future but have not done so yet. A final section of the paper argues that the statistics do not account for the rising prevalence of shared residence for children of separated parents. Therefore, the statistics underestimate the economic standard of children with separated parents.