Empty homes: mapping the extent and value of low-use domestic property in England and Wales

We combine economics, housing theory and data science to gain a greater understanding of low-use properties in England and Wales. We collect a unique dataset of domestic properties unoccupied by a permanent resident from 112 local authorities via freedom of information requests. The dataset covers 23 million residents and 340,000 low-use properties (3.4% of all properties). We find that the distribution is very skewed, with 5% of the lower super output areas (our smallest geographic unit) containing 29% of all low-use properties. We estimate the value of low-use properties in the dataset to be £123 billion and that an empty homes tax of 1% would generate the equivalent to 11% of the current council tax (local government tax). We use logistic regression to identify local authorities with high numbers of low-use properties (72% accuracy), local authorities where low-use properties are more expensive than ordinary homes (77% accuracy), and local authorities where both those conditions are true (79% accuracy). The coefficients of the models indicate that low-use property tends to be found in the most and least affordable areas and that the probability of low-use property being more expensive than a regular home increases as affordability decreases and tourism increases. We estimate that 39–47% of the population in England and Wales live in an area where low-use property is more expensive than property occupied by a full-time resident. We conclude that as the areas with the least affordable housing also tend to have the highest demand for low-use property, it may be appropriate to reduce demand via measures such as an empty homes tax rather than increasing housing supply.