Identifying housing vacancy using data on registered addresses and domestic consumption

Housing vacancy is a significant issue in developed countries’ decaying and densely populated cities. Comparisons are made between the number of ‘vacant housing’ and ‘homeless people’ stressing the existence of inequalities in access to housing. The reasons for addressing vacancy are manifold, ranging from mitigating urban blight to mobilising latent resources in tight markets. Little attention is paid to vacancy in municipal housing strategies. Still, mapping vacant units appears to be complex and resource-demanding, likely discouraging municipalities from planning further operations against vacancy. Given the lack of methodological support in the literature, this paper discusses how to identify housing vacancy units. Through a case study in Wallonia (Belgium), this paper highlights the benefits of combining visual surveys and processing data provided by utilities and registered addresses. Our results suggest that housing vacancy is underestimated through official statistics and that data mining would help mitigate the administrative burden related to identification and help to prioritise operations designed to reduce housing vacancy.