Part One in PPR’s new policy series: “A new look at housing: what the census data can tell us about social housing provision and deprivation”

Grounded in its work in communities, when looking at social housing in the past PPR has for obvious reasons focused on housing need: the ever-growing cohort of people who for the most part aren’t in social homes, but very much wish they were. The Northern Ireland Housing Executive is responsible for data-keeping and reporting of these measures of social housing need, and we use its data to report regularly on the increasing numbers of households on the waiting list, in housing stress or with FDA homeless status; to count the number of children in those households; and to track the Housing Executive’s ‘residual need’ measure, which tells us about the shortfall in social homes.

The Housing Executive performs the same data-keeping and reporting role with regard to social housing building, via its online Social Housing Development Programme spreadsheet. Bizarrely, it has organised these ostensibly parallel systems in such a way that they use two different, non-matching geographies:  a public one for social housing building according to parliamentary constituency, but a bespoke, undisclosed, ‘for internal use only’ one for social housing need.

…the Housing Executive does not have a comparable dataset of already-existing social housing that is publicly available – so while rolling information on new builds, refurbishments and the like is publicly accessible, there is little detail, other than very general overall stock numbers, on existing social homes by area.

Anyone looking at the Social Housing Development Programme can easily determine where new social homes are being built by looking at the public electoral map. However, the Housing Executive does not have a comparable dataset of already-existing social housing that is publicly available – so while rolling information on new builds, refurbishments and the like is publicly accessible, there is little detail, other than very general overall stock numbers, on existing social homes by area.

Moreover, to record and report housing need, the Housing Executive uses its own mapping system, divided into ‘Common Landlord Areas’ and ‘Housing Need Areas’ – and its map is not in the public domain. Over the years it has repeatedly refused requests to share its map or to make it public.

Why does it matter? Well, In withholding its map, the Housing Executive ensures that its statistics on waiting list, housing stress and FDA numbers – as well as on allocations of social homes – are impossible to pin down geographically in any detail, and impossible to compare against each other (do all ‘Common Landlord Areas’ have the same numbers of households in them, for example?) Equally impossible is the potentially very useful task of tracking the extent to which new social homes are – or aren’t – built in the neighbourhoods where there is highest demand for them.

In the absence of the information that would permit place-specific analysis of the relationship between housing need and actual provision from the body responsible for data keeping on both, we turned to the 2021 census as a credible, albeit imperfect and partial proxy.

The census data does contain some information indicative of housing need, in the form of its measure of ‘housing deprivation’. Its data on household tenure gives a clear indication of the level of social housing tenancy (as opposed to private rental, private ownership etc) in any given area.  And – happy days! – the census uses, logically, the same geography for its deprivation and its tenure data. Even more importantly, both the data and the geography at which is recorded is publicly available: a win for transparency.

Next in the series: what can we learn from the census’ measures of housing need and housing provision?