UK Housing Market
Yes, there is a shortage

Some evidence suggests that housing shortages have been largely eliminated in Britain: in the past quarter century, the proportion of households with six or more people has fallen by a factor of three, and there has been a dramatic fall in the proportion of young couples who begin marriage by sharing accommodation with their parents. There is, however, a widespread belief that a housing shortage exists. At a lunchtime meeting on 18 December, Research Fellow Mike Murphy discussed this apparent contradiction. The changing age structure of the population, together with the cultural influences on household formation and dissolution, have resulted in a large growth in the numbers of households for whom the conventional three-bedroom family house is unsuitable. Murphy also discussed how demographic behaviour, such as household formation and childbearing, is influenced by the nature of housing provision, particularly across the owner- occupied and local authority sectors. Local authority tenancy is associated with larger family sizes and higher rates of unemployment, Murphy claimed, even after accounting for the influence of other factors such as social class.

Michael Murphy lectures in Population Studies at the London School of Economics and Political Science and is a Research Fellow in CEPR's Human Resources since 1900 programme. He has written on the demographic and social policy aspects of household change and housing tenure as well as on family formation and marriage trends. He spoke at one of a series of lunchtime meetings organized jointly with the British Society for Population Studies (BSPS); this talk was sponsored by the Population Investigation Committee.

Murphy began by discussing the aggregate supply and demand for housing. In 1981, the number of dwellings in Great Britain was 21.2 million, compared with an estimated 19.5 million households. This apparent surplus is, however, misleading. In England alone, two million of these houses were either unfit for habitation or vacant because of delays in the process of moving house or households with second homes. There are also indications of substantial housing stress, Murphy noted. House price inflation is high, and the numbers of homeless people accepted by local authorities in England and Wales rose from 75,000 to 88,000 in 1981-4. Moreover, these national statistics take no account of the substantial geographical mismatch between the housing stock and the locations which are more desirable or offer better employment opportunities.

Murphy noted three demographic trends which have contributed to the rise in the number of households over the past quarter century. The first is the age structure of the population: the numbers above pensionable age increased from 7.6 to 10 million between 1961 and 1986. In the same period, the proportion of the elderly choosing to live independently also increased, a phenomenon which CEPR Research Fellow Richard Wall has discovered in many developed countries. The second factor is the tendency for young people to live away from their parents, either alone or sharing with others as a prelude or alternative to marriage (see the report elsewhere in this Bulletin of the lunchtime talk given by Kathleen Kiernan). The number of young married couples not forming an independent household fell by around 75% from 1961-81. But young people still leave their parents' homes relatively late, compared to other countries; Kiernan has ascribed this to the fact that the private rented sector only comprises 9% of the total UK housing market. The third factor is the increase in marital breakdown. David Eversley has suggested that the breakdown of each married couple household leads to the formation, on average, of 1.5 new households in the medium term. The rise in divorce rates in the period 1961-81 thus accounted for an extra 200,000 households, Murphy estimated.

The changing composition of the population is leading to shifts in the demand for different types of housing. The absolute numbers of married couple households will change little, according to Department of the Environment forecasts. As a proportion of all households, however, married couples will decline from 74% in 1971 to around 57% in 2001. The expected increase in households will continue to come from groups who tend to be relatively mobile or poor and to form smaller households. The conventional three-bedroom family house and the most common form of tenure, owner-occupation, are often inappropriate for them. Recent trends in housing tenure, such as the growth in owner-occupation, thus run counter to the housing needs of these groups, Murphy argued.

Murphy's research showed a strong relationship between early childbearing and local authority tenancy. Women who bear children outside marriage are five times more likely to be local authority tenants than to be owner-occupiers. Conversely, women who marry and begin childbearing when older are three times less likely to be local authority tenants than to be owner-occupiers. Murphy estimated that owner-occupiers tend to marry about one year later and to have their first child about two years later than local authority tenants. Some small-scale studies, e.g. by Clare Wallace, have shown that local authority allocation procedures can encourage some young unemployed women to bring forward childbearing, in order to obtain an acceptable home. In contrast, the need for house buyers to save for a deposit and high mortgage payments often requires two incomes and thereby encourages postponement of childbearing, especially among less well-off owner-occupiers.

This pattern is reinforced as childbearing continues. Murphy had found that movement between the major housing tenure types has been relatively rare once childbearing starts. As a result, the local authority sector has a substantially higher proportion of larger families, even though it contains a smaller (and shrinking) proportion of larger dwellings than the owner-occupied sector.

Housing and family formation patterns influence and are in turn influenced by employment experience. In his joint work with Oriel Sullivan, Murphy had found that unemployment rates among men under the age of 30 in the local authority sector were 2.5 times greater than in the owner-occupied sector; many unemployed people are forced out of owner-occupation. Other demographic variables also strongly affect access to housing. Marital breakdown is, apart from unemployment, the major reason for movement of families from the owner-occupied to the local authority sector. There is also evidence that the relative risk of marital breakdown is about 30% higher in the local authority sector, even after controlling for other influences.

These findings confirm that changes in housing demand are not simply a response to demographic changes, Murphy concluded. The pattern of housing supply has a strong long-term impact on demographic behaviour in all age groups, from young adults' decisions concerning childbearing to the influence of sheltered housing provision on the ability of the elderly to live independently. In Britain, access to different forms of housing tenure in particular has substantial long-term demographic effects.

Much of the discussion after the talk focused on the complex relationship between unemployment and housing tenure. The view that local authority tenancy significantly reduces labour mobility was not supported by the evidence, Murphy insisted. Regional variations in house prices made it ludicrous to suppose that unemployed owner-occupiers in the North-East could sell up and buy houses in the Home Counties. The polarization between owner-occupation and local authority tenancy, should not be exaggerated: less well-off home-owners and local authority tenants exhibited very similar characteristics, according to Murphy.

Murphy was also asked about his observation that single women had some incentive to bear children in order to qualify for local authority housing. He referred to in-depth interviews carried out by Bea Campbell and Paul Willis which indicated that many factors motivated the decision to have children. Although local authority allocation procedures must have an influence, it was quite impossible to tell whether it was numerically significant.