Leveraging digital footprints data for accurate estimation of the residential housing stock in the United Kingdom, 1997–2022

Abstract

This paper explores the use of digital footprints data to improve on the limited information available on the size and distribution of the residential housing stock in the United Kingdom over the past two decades. We use a subset of a large dataset consisting of the names and addresses of more than one billion individuals dating back to 1997 (LCRs: Linked Consumer Registers) to calibrate property lifecycle information within an authoritative geolocated address and property dataset. We validate the results of our novel calibration method against official estimates by the UK government covering the same period. We further show that our calibration method, based on individual level data, captures a much larger share of the residential housing stock since 1997 than would have been possible with other available data sources. We argue that these housing stock estimates are fundamental to the development of a comprehensive digital footprints data infrastructure that has the potential to significantly enhance population, migration, and other social statistics in the UK. More generally, this study highlights the benefits of using digital footprints data in refining, calibrating, and extending existing datasets.

Keywords

• Digital footprints data
• residential housing stock
• address matching
• uncertainty