One of the most interesting current debates on housing is one that to many commentators had long been settled – whether we actually need to build more homes to solve the housing crisis. Only today, Jonathan Eley had a long read piece in the FT that is well worth reading.
Last year’s Redfern Review, commissioned by Labour, came up with a conclusion that surprised many – building more homes would do little to improve housing affordability or boost home ownership. This has provoked a debate between Ian Mulheirn of Oxford Economics, whose analysis underpins the Redfern Review, and, it seems, almost everyone else (see the critical comments in that Guardian piece from both Shelter and Aldermore). This is an important debate because it forces everyone to consider what the problem really is, and to recognise the non-housing factors that have contributed to the crisis.
The basis of his argument is outlined in this post on Ian’s blog in which he responds to criticism from John Kay. In short, the housing problem is one of distribution not supply. Housebuilding is keeping up with demand, which we know this from looking at the rate of housing completions vs household formation (166,000 vs 199,000 since 1996). Data on rents reveals that they have remained stable, indicating there is no unmet demand (or at least not at a level that should concern policy-makers). There have been a number of responses on the dependency of household formation rates on housing affordability, which I won’t repeat here. Sam Bowman’s blog explains the case.
A couple of points have been overlooked. First, rents are limited by incomes because housing is the major household expense for private renters. We may have a severe housing crisis but it won’t perfectly translate into rent increases because there is a limit to what people can afford (to take this to one extreme, average rents could not reach 101% of average incomes because people simply could not pay). This may explain the shrinking size of rented homes – it’s the only way landlords can generate higher returns. And unlike house prices, you can’t speculate on rents or accumulate capital by signing a tenancy agreement.
Second, rents may have increased at a much faster rate than the data shows. Rents are the first indicator of unmet demand as people unable to buy are forced to (continue to) rent. If rents have stayed broadly stable or, as he argues, they have fallen in relation to incomes, there can be no unmet demand. But he also admits that the size of rental properties has fallen. As we still don’t build homes for market rent at any great scale, the vast majority of new tenancies are taken up in existing rather than new homes. If rented homes are shrinking in size, perhaps it’s because landlords are converting and dividing existing homes into ever smaller units (circa 40,000 of the 217,000 new homes added last year were from conversions). If the average 2 bed flat has not become more expensive to rent but has shrunk in size by 20%, rents would have increased.
But what if for the sake of argument we accept that we are building the right number of homes?
In this case, building more of the same will not solve the problem, because those currently priced out due to non-housing factors will continue to be priced out. However, even in this case I believe that we do need to build more.
We need to build the right types of home in the right places, but we tend to get a debate focused on a) overall numbers (hence a constant focus on national targets) and b) location (partly this is because of the vociferous backers of planning reform and, in particular, relaxation of green belt restrictions).
Numbers and location are vital, of course, but so too is housing tenure (which is Ian’s point). Talk of building up to 300,000 homes a year only makes sense if the gap between homes built now – 187,000 last year – and that target is filled with what is currently not being built at scale. That is, affordable homes for rent. Building tens of thousands more homes for sale would only help those in or close to being in a position to buy.
Even if rents in the private sector are static or falling slightly, the proportion of 25-34 year olds renting privately has nearly doubled in the last decade. For those people, and older generations who have been unable to buy, we should be concerned that 28% of private rented homes fail to meet the Decent Homes standard (1), the rising churn in the sector which brings with it increased instability and the high costs of moving, and those shrinking sizes. And if we want to support home ownership, we should want to increase the supply of rented housing that is affordable enough to allow tenants to save to buy.
So even if we accept that there is not a shortage of housing numbers, there is a shortage of affordable rented homes, and we still need to do something about that.
It’s hard to change the distribution of housing if you are trying to shift from market to sub-market housing (primarily because it requires a lot of government funding). The first option would be to take action to convert existing homes into social housing or impose much heavier regulation on private landlords. This could mean councils buying up private stock and renting them out as social homes. It’s hard to see how this is a deliverable or affordable solution. I don’t want to get into the debate on rent caps and PRS regulation but there are legitimate concerns that landlords would exit the market and the supply of rented homes would fall. Again, it’s not obviously a solution to the problem.
The second option would be to build more affordable rented homes, in much the same way that homes were built in the postwar period. Private sector development would likely continue as it did then – demand for homes for sale would not be affected, and indeed may increase as more people can gradually save to buy – and so the overall rate of housebuilding would increase. In my view, therefore, moving past the ‘housing supply’ question to think about the distributional problem of housing still leads you back to a need to build more homes.
(1) Although this proportion has fallen from 47% in 2006.