From Counting Houses, to Making Houses Count: Publicly Available Administrative Data on Subsidised Housing

This paper by Illana Melzer and Christoph Garbers from 71point4 explores publicly available delivery data on South Africa’s housing subsidy programme. It is an initial deliverable of a broader project that explores existing and proposed performance metrics for the State’s subsidy housing programme.

The primary indicator that is published on the housing subsidy programme is the number of housing units delivered. Nationally aggregated data on ‘housing delivery’ is published in various documents. This data is, in the main, derived from administrative entities that play a role in regulating construction or administering subsidies, including the National Home Builders Registration Council (NHBRC) and the National Department of Human Settlements, Water and Sanitation (NDHSWS). Yet, while the actual number of houses delivered is obviously critical in assessing the housing programme, there has been less of a focus on tracking how these housing assets perform over the long term. As a result, there are significant discrepancies within and between the published numbers derived from these various administrative data sets. The data published by CAHF on the total number of RDP properties, as well as new RDP properties registered (both drawn from the deeds registry using the proxy) differs significantly from data on housing completions published by the NDHSWS which differs in turn from the number of housing starts published by the NHBRC.

We would expect these numbers to differ to some degree, however large discrepancies – whether across data sources, or within the same data source over time –  can indicate weaknesses in underlying administrative processes.  Administrative weaknesses can have profound implications on the longer-term wealth-creating effects of the programme; delayed transfer undermines the potential value of the housing asset for households, negatively impacts on the functioning of property markets, inhibits bank lending and interferes with sound city or municipal management. More broadly, poor governance and neighbourhood service delivery by municipal government can also negatively impact on the extent of asset wealth realised by beneficiary households and local home owners. Addressing this, and making these entry-level property markets perform, is a potential next phase of our national housing programme, now approaching a quarter century since inception.

Our ability to address this, however, necessarily requires an understanding of where the problems are; and this requires that we identify government-subsidised properties with some degree of accuracy, monitor sales activity and price trends, and assess mortgage lending and performance on an ongoing basis. The paper unpacks this data problem, highlighting why it matters, and proposing steps to address the issue.