A study by Alan Holmans and Merron Simpson has reviewed the latest population data and available information about moves into and away from local authority and housing association housing and vacancy rates there to see whether the reported signs of 'low demand' call into question published estimates of newly arising housing need for England in the region of 90,000 a year (JRF 1995 and Shelter 1998) and 60,000 to 100,000 (DoE 1995). The researchers also explored numerical evidence for Wales, Scotland and Northern Ireland and examined case study evidence. They concluded:

• Reanalysis of population and household data supports earlier estimates of housing need at the national level, but there is a strong regional divergence.

• More of the increase in the population was in the South of England and less in the North than was projected. The main reason is a rise in migration from the North (and to a lesser extent the West Midlands) from 1994 onwards.

• The England total of departures from local authority housing increased during the 1990s, from around 220,000 a year in 1991/92 and 1992/93 to around 260,000 a year in 1996/97 and 1997/98. This increase and the reported vacancy rates in local authority housing were concentrated in the North of England.

• Within the North of England there is much variation in local authority and housing association vacancy rates and departure rates from local authority housing. There is evidence of a partial association between high vacancy and departure rates and low house prices. Case studies demonstrate that low demand is a term used in various ways.

• Private house-building remains fairly strong in the North of England. Following the early 1990s slump, the increase in new dwellings started was as great proportionally in the North of England as in the South. Despite the reports of unsaleable older properties, the demand for new houses appears high in the North of England.

• Surpluses of unlettable and unsaleable houses are a local phenomenon and do not demonstrate an overall falling away of housing demand. If migration trends persist, more of the demand and need for new dwellings will be in the South of England where building costs and land prices are higher.

Concern about 'low demand' for local authority and housing association housing has grown since the early 1990s and has been vividly illustrated by reports of housing associations demolishing recently built homes (in Newcastle). Such manifestations of low demand, and less extreme symptoms such as rising vacancies and a more rapid turnover of tenants, have of late been interpreted as signifying a general fall in demand and need for housing. 

The recently published 1996-based household projections for England have revised the 1991-2016 projected increase in households from 4.4 million to 4.1 million (3.8 million if the time frame 1996-2021 is used). Studies based on the 4.4 million projection have produced estimates of newly arising housing need at around 90,000 a year (JRF 1995), 60-100,000 (Department of the Environment) and 85-90,000 (Shelter). Are these estimates still credible given the evidence of instances of demolition, rapid turnover of tenants etc?

Population and households
Population and households are the first place to look for evidence of a fundamental shift in demand and need for housing. For England as a whole there is no sign of this. Between 1991 and 1996, official population estimates show an increase of 539,000 in the population aged 16 and over, as compared with 437,000 in the population projection used to derive the household projection that was the basis for the housing need estimates. That the increase in the population aged 16 and over was 100,000 above the projection is due to net migration from outside the United Kingdom having run higher than assumed. The number of households is not as firmly estimated as is the population, but the most recent estimates suggest that the number of households in England at mid-1996 was almost identical with the figure given by the 1992-based official household projections from which the estimates of housing need were derived. At national level, there are no signs of any fundamental shift in the demography that underlies housing demand and need.

Below national level the picture is different. The increase in the population was more heavily concentrated in the South of England than had been projected. Inward migrants go disproportionately to London and the South East and during the 1990s migration from the North rose sharply. Table 1 shows the contrast between 1991 to 1993 when net migration from the North ran at fairly low levels and 1994 to 1997 (the most recent year for which figures are available) when it was much higher.

A possible explanation for the change in migration flows is that the economic recession of the early 1990s was more severe in the South of England than in the rest of the United Kingdom, but that as the economy came out of recession, the attractiveness of the South and the faster growth of job opportunities there reappeared. The increase in outward migration from the North of England and the West Midlands could well have contributed to a weakening demand for housing there, though not, of course, uniformly. The swing in Scotland from a considerable net gain of population by migration within the UK to a small net loss was in line with the increase in net outward migration from the North of England.

Local authorities' departure rates and vacancy rates
Vacancy rates for local authority and housing association housing are a widely used indicator of weakening demand. The number of new lettings, in circumstances where no significant numbers of new dwellings are being provided, is sometimes cited as evidence of low demand. But an increasing number of new tenants cannot really be called a symptom of low demand; what it shows is an increase in the supply of dwellings becoming available for reletting as a consequence of departures being taken up by new tenants (including some who had been local authority tenants before, had left and are coming back). 

Instead, estimated departure rates are calculated. Where there are no significant numbers of new dwellings or demolitions, departures in a given year are equal to lettings to new tenants plus the increase in vacancies (or minus the decrease) between the beginning and the end of the year. The departure rate is the number of departures as a percentage of the number of tenants at the beginning of the year. Departures are derived from lettings to new tenants as reported by local authorities in their Housing Investment Programme (HIP) returns to the DoE (now Department of the Environment, Transport and the Regions) and vacant dwellings likewise. Comparison with other sources of data indicates that the lettings to new tenants are overstated, probably by some authorities including what are only technically new lettings, for instance, to a sole tenant in place of a joint tenancy.

Evidence from the 1996 English House Condition Survey (EHCS) points to understatement in the total number of vacant dwellings, though whether the degree of understatement has altered through time and how it varies between authorities is not known. The point is that the data used to analyse changes in departure rates and vacancies and make comparisons between authorities are mixed in quality. They have to be used, however, for lack of an alternative. Sample sizes confine the Survey of English Housing (SEH) to analysis at national level.

Estimated departures are shown in Table 2 for the English regions grouped as in Table 1 and Wales. The Scottish figures are not really comparable owing to differences in coverage.

Notwithstanding the caveats about data quality, it is clear that the course of departures year by year was very different in the North of England, and to a lesser extent the Midlands, from what it was in the South of England outside London. The figures for London are more problematic. Departures increased in the North of England at around the same time as outward migration (Table 1).

A similar contrast between North and South is shown in vacancy rates in Table 3. Not all years are shown for reasons of space; but it is useful to show the figures for individual regions because there are differences within groups as well as similarities.

In absolute terms, the vacancy rates in the North of England are not high, even with allowance for a degree of understatement. But the changes through time have been very different from the South of England outside London.

There is considerable variation within the North of England in departure rates and vacancy rates. Examples of high vacancy rates (in 1997) include 5.2 per cent in Bradford, 4.2 per cent in Salford, 7.4 per cent in Liverpool, 4.4 per cent in Newcastle and outside the metropolitan districts, 7.1 per cent in Blackburn and 5.6 per cent in Burnley. But in the North of England there were also 31 local authorities with reported vacancy rates of 2.0 per cent or less (out of 82, excluding authorities that had transferred their entire stock). An attempt was made to see whether high vacancy and departure rates went with low house prices. There were signs of this, but with enough exceptions to prevent it from being a hard and fast finding.

Important for showing that there has been no general collapse of demand for housing in the North of England is that as house-building for private owners recovered from the slump of the early 1990s, the increase in the number of new dwellings started (which respond more quickly to a strengthening of activity than do completions) was as strong proportionately in the North as in the South of England. 1992 was the bottom year of the slump. Between then and 1997 the number of starts rose by 46 per cent in the North and 42 per cent in the South outside London. If there had been a general weakening of housing demand in the North, the upturn from the slump would have been concentrated in the South, with a much smaller increase in the North, if that. That is not what is observed. The reported collapse of the housing market in parts of Newcastle with houses left empty and unsaleable is thus not part of a more general slump in demand in the North for houses for owner-occupation or rented out.

Conclusion
The conclusion to which the evidence points is that due to migration, probably stemming mainly from job opportunities, there has been some shift in demand for housing from the North of England to the South. This shift has not brought about any general collapse of housing demand in the North of England, but it has led to increased departures and higher vacancies in local authority housing and has produced local surpluses in the least popular localities. None of this calls into question the national projections or the estimate of housing need derived from them. More of the need will arise in the South, however, where building costs and land prices are higher and objections to new house building stronger.

About the study
The comparison estimates referred to are published in Housing demand and need in England 1991-2011 by Alan Holmans, Joseph Rowntree Foundation 1995, and How many homes will we need? by Alan Holmans, Nicola Morrison and Christine Whitehead, Shelter 1998. The DoE estimates were submitted to the House of Commons Environment Committee in 1995. This study reviewed data from the Survey of English Housing, local authorities' Housing Investment Programme returns and information from the Office for National Statistics about changes in the population nationally and by region and about internal migration.

Click on the 'order report' icon in the left margin to order online.