DCLG has published interim household projections for England using data from the 2011 Census. The last full set was the 2008-based projections. The projections are ‘interim’ because not all of the Census data required to produce a comprehensive new set of projections are yet available. A key limitation is that the interim projections only cover the 10 years 2011-2021. Full projections would cover 25 years – e.g. the 2008-based covered the period 2008-2033. The next full set of household projections will be published following Office for National Statistics (ONS) production of the 2012-based sub-national population projections, currently planned for spring 2014.
Projections have been produced for England and for local authority areas. DCLG no longer publishes regional data.
‘A household is defined as one person living alone, or a group of people (not necessarily related) living at the same address with common housekeeping – that is, sharing either a living room or sitting room or at least one meal a day.’
The number of households is essentially the household population multiplied by the appropriate household representative rate. A household representative is defined as ‘the individual that is taken to represent that household. This is usually taken as the eldest male within the household.’ The household representative rate is the probability (between 0 and 1) of anyone in a particular demographic group (by age, sex and relationship status) classified as being a household representative.
The projections are based on 75 population groups – i.e. 75 representative rates. The household representative rate for the whole population (i.e. total households/total population) was 0.424 in 2011.
Projected household growth in England from 2011-21 is put at 221,000 per year, 24,900 per year lower than the annual growth rate for the comparable 10-year period in the 2008-based projections (245,000). This represents a 10%, or 2.2 million increase over the 10 years. (The widely quoted 232,000 households per year was for the 25-year period 2008-2033 in the 2008-based projections.)
Population growth accounts for 98% of the projected growth in households in the latest projections. Changes in household formation, by contrast, account for only 3% of projected household growth, a lower relative contribution compared to previous projections reflecting decreased household formation rates. (The discrepancy of 1% is down to the impact of interactions between the two sources of change.) By comparison, population change accounted for 85% of projected household growth in the 2008-based projections. In 221 out of 326 local authorities, population change accounts for 90-100% of projected household change.
The average household size is projected to fall from 2.36 in 2011 to 2.33 in 2021. These compare with 2.31 and 2.23 respectively in the 2008-based projections.
28% of projected growth is one-person households, followed by ‘couple and no other adult’ (i.e. couples with or without dependent children) households which account for another 27%. Two thirds of the total projected increase comes from households without any dependent children.
The under-55 age groups (15-24, 25-34, 35-44, 45-54) are all projected to grow at the same rate or slower than the overall household average (10%), while the age groups over 55 (55-64, 65-74, 75-84, 85+) are all projected to grow more rapidly. Overall, households over 65 account for 54% of total household growth.
It is very striking that household growth 2011-2021 is expected to be sharply lower for the 25-34 and 35-44 age groups than in the 2008-based projections.
Nearly half (46%) of local authorities are projected to have household growth of 5-10%. In 30% of authorities projected growth is 10-15%.
The majority (86%) of local authorities – 281 out of 326 – have a projected decrease in average household size from 2011-2021, implying that the population will increase more rapidly than household numbers.
A revised household projections methodology – projections have been produced periodically since the early 1930s – first used in the 2008-based projections has been used for the interim 2011-based projections.
They are projections, not forecasts. I.e. they project forward trends observed in past data, but do not attempt to predict the possible influences of future policies or economic or social changes. Because they are based on past data, the projections implicitly incorporate the influence of past policies and economic and social factors.
The household projections are based on the 2011-based interim population projections (produced by the ONS) at national and local levels, plus projections of household formation derived from Censuses and the Labour Force Survey (LFS). The ‘institutional population’ (those living in nursing homes, halls of residence, military barracks and prisons) is deducted from the total population to derive the household population upon which the household projections are based.
Unfortunately it is difficult understanding the derivation of the latest projections because different data sources and time periods have been used for different steps in the projections, and the explanations for these are scattered across a number of documents published at different times.
The national and sub-national population projections assume a continuation of the estimated trends in fertility, mortality and migration used in the 2010-based projections (i.e. they were derived before Census data were available, and therefore will require modifying for the next set of full projections), but applied to the higher 2011 Census base year population.
It is notable that the 2011 Census population estimates for England are 452,000 higher than the previously published 2010-based projection for 2011. By 2021 the interim 2011-based projection shows 667,000 more people than the 2010-based projections.
Fertility and mortality rates are based on trends over the five years 2005-2010. Migration is assumed to occur evenly throughout the projection period. The assumed level of annual net migration to England is 172,500, which is 15,500 higher than for the 2008-based household projections.
The marital status projections, a key influence on household formation, are those used in the 2008-based population projections, but applied to the interim 2011-based population projections. Marital status projections are only available at national level. Therefore sub-national projections are derived by applying national/local differentials from the 2011 Census to projected marital status factors.
The 2008-based projections used data from the 1971, 1981, 1991 and 2001 Censuses to project household representative rates by demographic group. Some limited results from the 2011 Census were also used for the 2011-based household projections, as were age-band data from recent years of the LFS. LFS data have suggested that there have been some steep falls in household representative rates for the younger age groups since the 2001 Census. 2011 Census results reinforce these trends observed in the LFS.
The treatment of students in halls of residence, which is important in deriving the ‘institutional population’, is discussed in the methodology report for the 2008-based household projections:
Whether private-sector halls should be included as part of an educational establishment (including Halls of residence) definition is a grey area. As private sector halls are included in housing supply monitoring, it suggests that they should be excluded from the institutional population definition for students to avoid double counting…This issue should, however be revisited after the 2011 Census.
It seems highly likely that the undersupply of housing over the last two decades has had a negative impact on the ability of people form households. This is most likely to have been felt among younger people. The economic crisis of the last five years also seems likely to have had a negative impact on household formation. Data from the LFS, reinforced by data from the 2011 Census, confirm that there have been steep falls in household formation among the younger age groups (25-34, 35-44). As the DCLG report on the 2011-based projections observes:
Household representative rates for younger age groups (35-39 and below) tend to be flat or declining over time – largely reflecting the trends between 2001 and 2011 and in some cases (20-24, 25-29 and 30-34) declining rates between 1991 and 2011.
The male and female under 30’s singles household representative rates tend to be lower than the others and reflect a higher degree of living at home and sharing.
While housing undersupply will not have been the only influence on household formation, it seems likely to have been a significant factor. Other ONS data show a big increase in the number of young people living at home with their parents.
Because the household projections project past trends forward, any dampening impact on household formation over the last 20 years, and especially over the last five years (as reflected in the LFS data used in the projections), will be projected forward over the 10 years of the projections. Therefore the local planning process tends to become a vicious circle: planning has constrained housing supply for the last two decades, which has suppressed household formation, which is reflected in the household projections, which are a key input into local authority planning for future housing supply.
The ageing of the population, and the much bigger increase in households in the older age groups than among the younger age groups, have important implications for the housing market. The propensity of households to move is inversely correlated with age – i.e. the younger age groups have relatively high propensities to move in any year, while the oldest age groups have low propensities to move.
Sources: documents and tables
DCLG. Household Interim Projections, 2011 to 2021, England. Housing Statistics Release. 9th April 2013 - click here
DCLG. Updating DCLG household projections to a 2011 base; Methodology Report. 9th April 2013 - click here
To unsubscribe, simply click the following link: Unsubscribe
Please do not reply to this email. It has been sent from an email address that does not accept incoming emails. To contact us please email email@example.com