A Housing Research Legacy: Remembering Alan Holmans
by Ken Gibb
I was in London today at an event put on at the LSE in honour of the life and work of Alan Holmans. Alan had a long and valuable career in housing first as a civil servant and then as a researcher mainly at Cambridge’s Centre for Housing and Planning Policy. Alan was originally an academic economist and lectured at Glasgow but he really made his mark in the Government Economic Service because of his interests in statistics, historic data time series, housing needs and demand research and especially the role of demography in supporting forecasts and projections. Alan was also a stalwart of the Housing Studies Association and was author or co-author of innumerable papers, reports and other research publications.
The event today was ably organised by Christine Whitehead and featured many colleagues who had worked with or knew him, several for more than three decades. There were quite a few stories and anecdotes but, as is often the case with these sorts of events, there was actually a lot substantively to get your teeth into from the many and varied contributions by speakers and from the floor.
More than anything Alan was an enthusiast and advocate of robust evidence supporting the development of policy, be it a local or regional needs or demand study or making sense of land requirements for new housebuilding to meet demand. Some in the room lamented the shift away in recent years from evidence and analysis in policy development. But Alan went further arguing that in some cases the evidence simply took you so far and then the politicians had to make choices.
A case in point is the handling of backlog or existing housing need. Should a needs assessment assume that policy intervention will attempt to ‘clear’ it in three, five or even ten years? At one level this is a resource or commitment question but the choice has a massive effect on the numbers. Recently, the new national affordable need estimates in Scotland chose five years – twice the implied commitment and spend than the ten years often used in local studies. The key thing is to be upfront about the judgement made and equally explicit about your assumptions.
At the heart of much of the debate this afternoon was the old controversy between economics and demography. Which is more important to understanding the housing market? Of course, it is really a false argument: both are important and matter and they are undeniably connected. Some of the most interesting points made concerned the economic elasticities that impact on household formation decisions I.e. Non-affordability may well reduce household growth below more unconstrained projections, just as jobs and relative wages impact on net migration. At the same time, understandable scepticism about economic forecasting can also be a sensible attitude with which to approach long range demographic projections. Alan was comparatively rare, as Glen Bramley pointed out, because he was an economist interested specifically in demography.
A recurring theme was the importance of uncertainty both for the researcher as an undeniable feature of the contemporary housing market but also for thinking about demographic trends e.g. the effects of ageing on the market, let alone the effects of the blizzard of often incoherent housing policies coming from Whitehall. To be clear, people were not talking about quantifiable risks but a series of unknown parameters and unknowable futures that can impact in quite different ways on the housing system. This was more Keynesian uncertainty than expected utility calculation. My colleague Geoff Meen warmed to this theme noting that market volatility made it hard to predict house prices in the short run of say a year hence, let alone contemplate the complexities of long term forecasting for housing planning purposes in the face of multiple uncertainties.
Several speakers noted the historical analysis conducted by Alan which chimed with a number of people in the room. Geoff argued that a longer view suggests that affordability can’t keep getting worse – at some point there will be a correction, a reversion to the mean, but this may not be a pleasant experience! At the same time, Alan’s historical housing supply data suggested that there is no long term trend in supply so making the kind of step change required i.e. sustaining 200,000 to even 300,000 completions, does seem fanciful. Geoff is leading a book (that I am also involved with) which explicitly ties historical urban development to housing economics today. It will be published in the new year.
Alan was fondly recalled by all who attended today. There was also recognition of the significance of the research he did, the rigour with which he did his work and the need for analysts and Governments to carry on the work and the spirit it was done in. In a way, this is (and should be) a continuing challenge for future housing research, research commissioners and the users of that research.