jats:p This paper seeks to predict the impact of future housing supply on the affordability of residential space in the United Kingdom, using quantitative model-based simulation methods. Our spatially disaggregated analysis focuses on the greater South East region, approximately within 1.5 hours commuting time from Central London. A dynamic spatial panel model is applied to account for observed temporal variations in property prices and housing affordability across districts. The dynamic structure of this model allows us to assess the scale and extent of knock-on effects of local supply shocks in one district on other districts in the region. These complex spatial effects have been largely ignored in local or regional housing market forecasting models to date. Applying this model, we are able to demonstrate that local house prices and affordability are not only determined by the underlying supply and demand conditions in the market in question, but also depend crucially on conditions in neighbouring housing markets whose properties can be considered close substitutes within a larger regional housing market. We also show that increasing housing supply in the most critical areas has little impact on (both local and regional) affordability, even if wages do not change in response to an increase in employment. </jats:p>