Estimating long-run income inequality from mixed tabular data. Empirical evidence from Norway, 1875-2017

Most evidence on the long-run evolution of income inequality is restricted to top income shares. While this evidence is relevant and important for studying the concentration of economic power, it is incomplete as an informational basis for analysing inequality in the income distribution as a whole. This paper proposes a non-parametric approach for estimating inequality in the overall distribution of income on the basis of tabular data from different sources, some in a highly aggregated form. The proposed approach is applied to Norway, for which rich historical data exist. We find evidence of very high income inequality from the late nineteenth century until the eve of World War II, followed by a rapid equalization until the 1950s. Income inequality remained low during the post-war period but has increased steadily since the 1980s. Estimates of a measure of affluence demonstrate that overall inequality has largely been governed by changes in the top half of the distribution and in in the ratio between the mean incomes of the lower and upper halves of the population.