What is it about?

This article deals with personal insolvency in Germany. The results presented, and their level of detail is novel, as previously, due to limited data availability for the German and European context, little was known about the demographics and spatial distribution of persons affected by insolvencies. These new insights and statistics were derived by utilizing a large administrative database in combination with web scraping and text-mining techniques. The results and techniques presented here show new opportunities for official statistics.

Featured Image

Why is it important?

A task of national statistical agencies (NSI) is to provide information that is of societal importance. A relevant societal phenomenon is personal insolvency, as in recent decades, there has been a substantial increase in those rates. This article presents an application in the area of personal insolvency that unlocks the full potential of administrative data in combination with data science techniques. Since the use of new data sources and sound data science methodologies are two central points on many research agendas of NSIs, this article demonstrates opportunities for NSIs to develop innovative statistical products that are helpful for policy and social research.


This article shows the potential of Official Statistics. Official Statistics have exclusive access to data which, combined with sound methodologies, enable results that only Official Statistics can produce. Such information can play a part in, for example, facilitating targeted and evidence-based political decisions or in answering social and economic questions. It is important to remember that these data sets often already exist; they simply need to be analyzed. We hope this contribution will highlight opportunities for Official Statistics and foster innovation.

Jonas Klingwort
Centraal Bureau voor de Statistiek

Read the Original

This page is a summary of: Spatial and demographic distributions of personal insolvency: An opportunity for official statistics, Statistical Journal of the IAOS, December 2023, IOS Press,
DOI: 10.3233/sji-230072.
You can read the full text:



The following have contributed to this page