Length of residence varies widely across the country and between different tenures. This has implications for the flagship Government policy to sell off high value local authority housing as it falls vacant. Unsurprisingly, the tenure with the longest length of residence is owner occupation and the shortest is private rented. Social rented, including local authority, housing falls somewhere in between and this is a consequence of the rights a social tenant has when they take up a tenancy.

Until 2010, all housing rented from local authorities and housing associations was let on an ‘assured tenancy’; that is, one that provides security of tenure for the lifetime of the original tenant as well as succession rights for relations of the tenant (provided they also live in the property). It therefore has some of the characteristics of owner occupation. The key exceptions are that a rent will always be payable, albeit at a low level, and the landlord will carry out repairs and maintenance.

This housing was also let on a rent that is dissociated from the market. Since 2002, there has been a migration to ‘Target Rents’, which are calculated with reference to local incomes as well as property values. The result is that, whilst in some areas there is very little difference between rents for social and private rented housing, in other areas market rents can be more than double Target Rents. For a three bedroom home in Kensington & Chelsea, market rent is more than five times Target Rent.

The introduction of Affordable Rent in 2010 brought new options to social housing providers: the ability to charge ‘Affordable Rent’ at up to 80% of market rent, subject to planning requirements, and the option to give fixed term tenancies. However, the vast majority of social housing is still occupied (and many new lettings are still agreed) under the old regime.

The rationale for moving from a social rented home therefore has special characteristics, influencing the turnover of social rented housing. Waiting lists are such that it can be difficult to secure a new social rented home and, even if that were possible, it may be let on a fixed term tenancy at Affordable Rent rather than the typically much lower Target Rent. This matters much less in more affordable parts of the country, where the difference between Target Rent and market rents is minimal and there is little cost in moving between social and private rented tenures. The opposite is true in less affordable areas, particularly London.

Charts!

Unusually for residential property, the social housing sector is blessed with lots of data. The Homes and Communities Agency (HCA) collects data as part of its regulatory role and releases the annual Statistical Data Return. The Department for Communities and Local Government runs the ‘Continuous Recording of Lettings and Sales in Social Housing in England’ (CORE) and publishes annual data tables. Combining this with Census data, the English Housing Survey and market data generates useful insights.

Of course, correlation does not reveal causation, as many people have said before. However, it can help understanding of a problem, particularly when there are clear links between the variables. This chart shows the relationship you can see by comparing the two maps on page 1, i.e. that social housing tends to turn over less in higher value areas. As explained above, this is the expected consequence of the rights conferred on a social housing tenant.

Averages, such as those in Fig 1, always hide lots of important detail. In this case, the English Housing Survey allows us to delve into the variation on a range of different measures. The most useful in the context of this note is variation by number of bedrooms – length of residence in social housing is longer in larger homes, more closely mirroring owner occupation than private rented homes.

The difference between market rents and target rents is much greater for larger properties. This is likely to be part of the explanation for the result shown in the chart above, that larger properties turnover less. Another contributing factor is that these properties are more likely to be occupied by family households, which are less mobile.

Consequences for funding the extension of Right to Buy

All of the above analysis is relevant to the Government policy to extend Right to Buy to housing association tenants. This will be funded by selling local authority housing over certain thresholds when they become vacant, although the proceeds will first have to fund replacement local authority homes. The thresholds appear likely to be set regionally and by number of bedrooms, and are set at the 70th percentile of the value of all homes in that area.

I can’t trace the origin of the current proposed thresholds, which were featured in John McDermott’s blog on FT.com, but those originally suggested in a paper by Policy Exchange come from analysis of English Housing Survey data on the market value of homes, carried out in 2008/9 and not repeated since.

Analysis of this data shows that approximately 10% of local authority housing in England is above the 70th percentile value for all housing. There were 2.2m local authority owned homes recorded by the 2011 Census, indicating that 220,000 would fall above the thresholds.

Looking at the data by region shows that the proportion of local authority housing that falls above the 70th percentile varies wildly across the country: from none in the East Midlands to 14% in London. The reason for this is shown in the chart below.

This chart shows how the difference in market value between social housing stock and all housing stock varies by region. With the exception of London, all regions show a similar pattern, with the 90th percentile social housing unit being worth between 47% and 57% of the value of the 90th percentile unit of all housing.

In London, the 90th percentile social housing unit is worth 67% of the 90th percentile unit of all housing.

The effect of this at local authority level is shown in the map below, using the 2014 distribution of market transaction values for each local authority. The market distributions have been adjusted on a regional basis depending on the difference in market value between social and all housing, shown in the chart below. This means that the total number of homes over the thresholds is only 78,000, 142,000 lower than the national level estimate would suggest.

What this means for local authority sales revenue depends on turnover. Using the turnover numbers shown in Fig 2, I have estimated that annual revenue could be in the region of £3.2bn, raised through the sale of 5,500 homes per annum, 9,500 lower than the Conservative Party estimate.

Of this £3.2bn, London is likely to contribute £2.5bn, while the South West may raise nothing at all. Within London, £320m per annum may be generated in Kensington & Chelsea, but nothing in Barking & Dagenham. This gives real importance to the question as to whether revenues will remain with local authorities or be redistributed around the country.

However, these numbers are likely to be a best case scenario because they don’t take account of residents moving between council homes. It is often the case that several tenants move in a chain between properties, giving rise to a number of lettings but only one vacant property. There is no data on the extent of this activity, but it could at least halve turnover, drastically reducing the potential revenue.

Taken from
Housing Market Notes:

• The value of land
• Approaching peak loneliness?
• Size matters
• Housing delivery
• Pension reform and the housing market
• Housing market activity
• Rental Britains
• First time buyer affordability
• Local house price indices
• Measuring housing need
• Magnetic London