Affordable housing has become a major issue for all Australian and New Zealand cities. A key contributing factor is inadequate land supply. The New Zealand government introduced a National Policy Statement on Urban Development Capacity which seeks to improve housing affordability by increasing land supply. The policy requires the production of various ‘evidences’ that, it is argued, are necessary to improved decision making leading to improved land supply. Specifically, the policy requires the production of a housing needs assessment that specifies demand for housing by price points, locations and typologies, three, ten and thirty years into the future. It also requires the regular monitoring and reporting of housing statistics.  The expectation is that these evidential pillars will drive local governments’ land use zoning decisions – a supply side response. The national policy is decidedly technocratic and is built on a rational planning model under the auspices of ‘evidence’ with a significant reliance on computer modelling. This paper will use this national policy as a case study to examine how this very specific idea of ‘evidence’ was constituted, is being applied and whether we are witnessing a new evidence-based rational planning model.