A very significant data gap in the housing area has been filled as the CSO has published its own estimate of housing completions in Ireland. Policymakers should now focus their attention on resolving the issue of undersupply, which according to the new data, is even bigger than was suspected, without having the distraction of not fully knowing what the level of that undersupply actually is.
The new information now needs to be followed up with a refocusing on old policy actions – and the introduction of new ones. Opening up sites through the provision of services and transport infrastructure, so that builders can get on with the process of building homes, was the motivation behind the Local Infrastructure Housing Activation Fund (LIHAF), but its implementation has been held up by bureaucratic processes. This must be streamlined, but its size, in the context of the level of infrastructural deficits will need to be reviewed. The use of state-owned land for housing through the new National Development Regeneration Agency needs to be accelerated.
Up to now, responsibility for collating data on housing supply in Ireland was the sole responsibility of the Department of Housing, Planning and Local Government (DHPLG). For the past fifty years or so, it used information from the ESB on electricity connections to gauge the extent of new supply. There have always been doubts about the accuracy of the data in this area. Indeed, I remember attending a meeting of stakeholders in the mid-2000s, where there was broad consensus that there were very large gaps in our knowledge base around construction in general and housing in particular.
Goodbody has been among the commentators questioning the accuracy of the "official" statistics coming from the DHPLG. In the vacuum, we used Building Energy Ratings (BER) as an indicator of new build, even though we knew that data had its own shortcomings. In an opinion piece for the Irish Times in April 2017, we argued that the responsibility for housing statistics should rest with the independent national statistics body, the Central Statistics Office (CSO). Having a situation whereby a department is responsible for both setting housing targets and producing the information that underlies those targets creates some dangerous incentives to delay or obfuscate for political reasons. That is not in the interests of proper public debate on such an important topic.
The motivation for addressing these issues rose exponentially over recent years as housing became the biggest domestic policy concern for the government. The current Minister for Housing Eoghan Murphy has been keen on removing the distraction of questionable data. The decision to hand the responsibility to the CSO by the Housing Analytics Group, which was set up to recommend appropriate actions to address the issue, is a welcome one. In the area of housing, it has a strong track record with its monthly publication of prices and transactions, an area that also had shortcomings until the CSO took it over in 2016.
True to form, the CSO has taken a thorough and methodological approach to its new responsibility. It recognises that the data are not perfect and may require further refinements in the future. However, we believe it now provides as definitive gauge as possible.
As we suspected, the data tell us that housing shortages are indeed bigger than the old data series had suggested. In 2017, 14,446 new housing units were completed across the country, 25 per cent below the previous estimate. This figure excludes three types in particular that formed part of the previous data that suggested there were 19,271 homes connected. The most important was reconnections, partly a reflection of the increase in banks taking possession of properties. The second most important factor is so-called “ghost” estates, units that were completed but moth-balled for years during the crash. Finally, the CSO has removed units that were not houses at all, but instead are out-houses or sheds.
Over the period from 2011 to 2017, 53,578 new dwellings have been completed in Ireland, more than 30,000 less than previously thought. The completion of ghost estates has represented somewhat of an easy win in terms of adding to the available housing stock over recent years. Fortunately, these are, by and large, no longer an issue. As these estates would have already had issues such as planning, roads, infrastructure and services already in place, they would have represented relatively easy “wins” in terms of delivery.
Rebuilding Ireland, the government’s flagship plan for the sector, set a target for the delivery of 25,000 units per annum by 2021, but faster population growth, booming rents and prices, and rising levels of homelessness suggests that this target is far too conservative. A target of 35,000 is much more realistic.
Housing output is accelerating, but if the run-rate of the first quarter continues, less than 20,000 housing units will be completed in 2018. The clarity that the new data provides means that there is no hiding place for the government. We all know the goal is to increase supply. The policies needed to achieve the targets now need to pursued with renewed vigour.
Dermot O’Leary is chief economist with Goodbody