Low interest rates have no doubt pushed up house prices, but the root of the housing crisis in our most successful cities is supply
It was interesting to see the economics commentator Simon Wren-Lewis pick up on a theory pushed recently by Ian Mulheirn of Oxford Economics: that the issue of unaffordable housing in the UK results not from a lack of housing supply, but the availability of cheap credit. While there is undoubtedly some truth in this theory, it doesn’t tell the whole story.
Yes, lower interest rates make borrowing more affordable. And just like most other goods and services, if you lower the price of something, people will be prepared to pay for it. In the housing market, cheaper borrowing means people have got a bit more money to play with, allowing them to bid a bit more for a house, and so pushing up its price (assuming the supply of houses doesn’t increase). This likely explains why house prices have more than doubled in every English and Welsh city in the last 20 years.
Yet if this was the only driver of house price growth, then we’d expect to have seen similar house price growth across the country – but we have not. The map below shows changes in house prices across UK cities since 2009 (when the Bank of England base rate was cut to the historic low of 0.5 per cent). It shows that there is a very clear geography to house price changes – despite a much more uncertain housing market and pretty shocking income growth after the financial crisis, cities in the Greater South East have seen very large increases. However, further north, rises have been much more modest.
Cambridge leads the list of rising house prices, which were 76 per cent higher in 2017 than eight years earlier. Meanwhile Burnley is at the other end of the scale, with an increase of just 2 per cent. In real terms this means that houses in the city are cheaper today than in 2009, despite historically low interest rates. (This data is for whole dwellings. If it was sale price per square metre, the divergences would likely have been even wider.)
Source: HM Land Registry data © Crown copyright and database right 2017. This data is licensed under the Open Government Licence v3.0.
In theory, this geographical divergence could result from the fact that housing in some cities are seen as a better investment than in others. In London for example, the prestige of owning in the capital may be a bigger draw, with property there seen as a luxury good. This would mean that we would expect to see lots of empty homes in the city, as investors buy them purely as an asset.
But again the available data does not back this up. The map below shows the share of properties that are empty across the country. There are fewer empty properties (defined as being empty for six months or longer, identified through council tax records) in our least affordable cities. On this measure, there are not swathes of empty houses being used only to store wealth.
Source: Ministry of Housing, Communities & Local Government
Data available for England only
A further argument put forward by Mulheirn is that the number of new houses being built has outstripped the number of new households formed in recent years. Once more, the geography of these patterns is very important. In 2001 (the earliest data available), there was an average of 2.37 people living in a property in London, compared to an English average of 2.33. Fast forwarding to 2016 shows that while the English average remained at 2.33, it had risen to 2.51 in the capital.
If more people are living in a single dwelling – for example through flat sharing – this again points to a shortage of houses in particular parts of the country.
Even if the number of people in per dwelling in London was to fall to the British average, an extra 360,000 homes would be needed in the capital. And that’s before any considerations of how high prices in the capital may have deterred people from elsewhere moving there.
All these factors point to the fact that there is no single national housing market – instead we have a series of separate housing markets in different places across the country. This makes comparisons of national house prices with national supply and almost meaningless (building more homes in Burnley, the most affordable city, does nothing for Bristol, one of the least affordable). Digging beneath the national level shows that housing markets across the country face very different challenges.
The greatest economic challenge facing our most successful cities such as Reading and London is unaffordable housing. Future interest rate rises may slow the pace of increase of house prices seen in these cities recent years. But they don’t address the underlying challenge of an undersupply of housing in them.
In any market, increases in demand without any increases in supply mean that prices increase. The same applies to housing. If we want our most successful cities to continue to prosper, then we need to build more homes in and around them.
We use house prices here as data on rents (which removes the asset value wrapped up in house prices) is lacking at a city level. The data that is available at a regional level shows a consistent story to house price data shown in this blog.
Leave a comment
Both Paul Swinney and Ian Mulheirn generalising about housing need and supply from too narrow a perspective: both miss the inter-generational aspect. I have analysed the 2014 household projections from this angle, with interesting results:
The projected net increase in the number of households in England is 211,000 pa over the period 2011-2031. This is made up of three components:
• 363,000 pa new households formed by those under 25 in 2011 (under 45 in 2031);
• 53,300 pa additional household formed by those 25-65 in 2011 (45-85 in 2031);
• 206,000 pa fewer household formed by those 65+ in 2011 (85+ in 2031);.
The overwhelming majority of newly forming households (87%) are young. Most cannot afford to buy new homes (even ‘Starter homes’ and with assistance from ‘Help to Buy’). Meanwhile very few social rented homes are being built and the existing stock continues to be sold off. The consequence is that newly-forming households depend mainly on the cheaper end of the existing stock: the average price paid by first time buyers (£226k) is well below the general level. There is thus a serious misalignment in terms of price between housing needs and what is being built, and this extends also to numbers, tenures, and locations.
The requirement for a minimum of 5 years’ worth of land to be ‘readily available’, and for local plans to look at least 10 years ahead, means a national pipeline of land for between 1 and 2 million homes must be identified. This is a minimum, since Planning Policy Guidance also encourages additions to OAN for a wide range of local factors. Since local planning policies generally prioritise use of urban brownfield land, the additional land to provide for OAN is necessarily mostly in the form of greenfield sites .
New permissions have exceeded starts by some 50,000 pa for the last decade, and land available in planning terms (allocated in Local Plans, or with permission) for some 900,000 homes is already held by major builders. However, in spite of this, actual output has been well below projected needs since 2007. The conclusion must be that such needs exceed effective demand (purchasers or tenants able to pay market prices). In essence the planning policy for housing provision, continued by the present HWP, requires a surplus of planned provision of land above the rate at which it can viably be developed.
The effect of this is to focus development on the most profitable locations – typically higher-priced greenfield sites. These seldom if ever cover the extra costs they impose on infrastructure and services. Housing policies effectively incentivise dispersed new development, diverting limited public resources and attention from renewal of infrastructure and services within existing settlements.
This is particularly damaging to the housing choices available to new and lower-income households, who depend on buying or renting existing entry-level homes. The result is a vicious circle involving the whole housing stock. Housing policy focused on new development and transport policy focused on supporting new housing lead to more travel, more car dependency and a poorer range of housing choices for newly-forming households. Churn accounts for 90% of housing choices acts as a multiplier, in a vicious circle.