Significant Statistical Uncertainty over Share of High Net Worth Households

DIW Weekly Report 14/15 / 2015, S. 210-219

Christian Westermeier, Markus M. Grabka

get_appDownload (PDF  227 KB)

get_appGesamtausgabe/ Whole Issue (PDF  416 KB)


The analyses of wealth inequality based on survey data usually suffer from undercoverage of the upper percentiles of the very wealthy. Yet given this group’s substantial share of total net worth, it is of particular relevance. As no tax data are available in Germany, the largest fortunes can only be simulated using “rich lists.” For example, combining the Forbes list, with its approximately 50 German US dollar billionaires, with survey data results in an increased aggregate total net worth for all households in Germany in 2012 of between one-third and 50 percent, depending on the scenario. Moreover, the share of the richest one percent of the population (about 400,000 households) rises from approximately one-fifth to one-third. After reassessment, the richest ten percent of the population’s share of total net worth is estimated to be between 64 and 74 percent, depending on the scenario. These reassessments are characterized by a high degree of uncertainty which eventually can only be reduced by improving the base data.

Markus M. Grabka

Direktorium SOEP und kommissarische Bereichsleitung Wissenstransfer in the German Socio-Economic Panel study Department

JEL-Classification: D31;I31
Keywords: Wealth Inequality, pareto distribution, SOEP
Frei zugängliche Version: (econstor)