Econometric literature is notorious for its inconclusiveness. Even for seemingly straightforward issues such as the impact of minimum wage laws, you can always find econometric papers on both sides, and many more which just don’t come to any particular conclusion. If you asked an econometrician whether it was colder in winter than in summer, he would probably respond:
‘Based on my observations up to this point, it would seem that there is a relationship between the time of the year and the outside temperature which has the expected sign, and which is statistically significant at the 5% level. However, I cannot rule out the possibility that this is an idiosyncrasy of my data sample, my method of temperature measurement, or a failure to fully control for confounding factors. My conclusions are therefore suggestive and tentative, based on a small sample size and a limited number of control variables. More research on the subject is required.’
So, if there is a subject on which the econometric literature takes something resembling a clear stance, we should take it seriously. It probably means that what is known on the subject is watertight.
There is such a subject: planning restrictions substantially raise housing costs. See the literature review here, on pages 74-80. If we are interested in poverty, that should be our most obvious starting point. Planning restrictions in the UK are particularly severe, and rents represent a gigantic share of low-earners’ budgets. The graph below shows average expenditure on rent among households in rental accommodation by gross income decile, expressed as a percentage of average household expenditure in the respective decile. The figures are approximate, but they should not be too far from the truth.
They show that renters in the second decile have to reserve just under half of their budgets for rent, and renters in the bottom decile more than that. Do we need another argument for planning reform?
Rent of households in rental accommodation as a % of total household budget, by gross income decile
Author’s calculations, based on data from the ONS’s 2010 Living Costs and Food Survey
Politicians are, of course, aware of this. But if you are a politician, planning reform is an issue on which you cannot win. Before you can even say ‘development’, you already have the Campaign to Protect Rural England and likeminded groups at your throat. The next day you will see your name in a campaign brochure, with photos of happy families strolling through blooming meadows, and a text insinuating that you are out to destroy all this.
That might be tolerable if there were a strong counterbalance, i.e. if there were groups cheering you on, and explaining why what you do is right and necessary. But such groups are few and far between. Who would defend a pro-development planning reformer? The poverty lobby?
Forget it. During the last big planning debate in summer/autumn 2011, when the anti-development lobby was firing from all cannons, the poverty lobby was conspicuous by its absence. In the autumn newsletter of the Child Poverty Action Group, seven out of seven articles dealt with benefits.
It is a strange way of setting priorities. Housing-cost support systems have been extended massively already. Taken together, explicit subsidies (Housing Benefit) and implicit subsidies (social housing) now cover about two thirds of all households in rental accommodation. Unsubsidised private tenants are already a minority.
The rental sector by housing cost support instrument
Author’s calculations, based on data from the English Housing Survey Household Report 2010-11
The attempts to use Housing Benefit and social housing as substitutes for a functioning housing market have been taken far enough. We now have to make sure that housing becomes affordable under market conditions, and that requires supply-side reform, not benefits.
Kristian Niemietz is the author of Redefining the Poverty Debate: Why a War on Markets is No Substitute for a War on Poverty.
 The figures are not precise because the numerator consists of the average rent paid by renters, while the denominator consists of the average total expenditure of all households in the respective decile – i.e. not just renters, but also homeowners. That must bias the figures, although it is not clear in which direction.