This document provides answers to some questions we think users may ask which are specific to the 2012-based subnational population projections.
Questions about projections in general may be answered by the general projections Frequently Asked Questions document, and specific methodological questions may be answered by the subnational population projections methodology document.
If you cannot find the answer to your question in these documents, please send your query to email@example.com or telephone +44 (0)1329 444652.
This document has been split into a number of sections. These are:
about subnational population projections data,
how the subnational population projections are produced,
understanding the projections, and
accuracy and improvements.
The latest subnational population projections for England, the 2012-based subnational projections, are located on the ONS website. Subnational population projections for other UK countries are the responsibility of the respective devolved administrations: for Scotland by National Records for Scotland (up to 2012-based), for Wales by the Statistical Directorate of the Welsh Government (up to 2011-based), and for Northern Ireland by the Northern Ireland Statistics and Research Agency (up to 2008-based).
A set of tables containing males, females and persons, by five year age groups, for all years, for regions, counties, local authorities, and clinical commissioning groups are available. We are also publishing summary components of change tables for regions, counties and local authorities. The data in these tables are rounded to the nearest 100 persons. For the first time, we are also publishing unrounded subnational population projection data by single year of age and components of change. This is part of the Open Data initiative to make more data available to users.
Subnational population projections data at single year of age is not as robust as projections data at greater levels of aggregation. However, making projections data available in this way will allow users to aggregate data to the age groups that they require, which may be different from the standard five year age groups. We advise that the data should be aggregated to five year age groups or larger and rounded to the nearest 100 people if quoted in any publication, presentation or documentation.
It was agreed at consultation that we would not publish subnational population projections by broad age group that used to be available in Table 4. Instead we have published the single year of age data (8.96 Mb ZIP) so that users can aggregate to their own age bands.
The formula for calculating subnational population projections for state pension ages is available on the subnational population projection release page. At the time of publication (29 May 2014), the current legislation is the Pensions Act 2014. The state pension ages for people will change over the projection period. Between 2012 and 2018, the state pension age will change from 65 years for men and 61 years for women, to 65 years for men and women. Between 2019 and 2020, the state pension age will change from 65 years to 66 years for both men and women. Between 2026 and 2046, the state pension age will increase in two stages from 66 years to 68 years for both men and women. Also published are the formulae for pension ages under the Pensions Act 2011.
Subnational population projections are available for regions (formerly government office regions), counties, local authorities and clinical commissioning groups in England.
The administrative geographic boundaries used were effective from 1 January 2012. The subnational population projections are not available on any other boundaries. The clinical commissioning group boundaries came into effect on 1 April 2013.
ONS does not produce subnational population projections for areas smaller than local authority. Some local authorities and other analysts do produce their own projections for specific smaller geographic areas.
As projections for clinical commissioning groups are not modelled directly but based on the projections created for local authorities, components of population change are not available for these areas. In many cases, clinical commissioning group areas are coterminous with local authorities or aggregations of local authorities, in which case population projections for these areas are calculated by aggregating the appropriate local authority projections. Where areas are not coterminous, clinical commissioning group projections are produced by apportioning local authority level projections, based on estimates of clinical commissioning group populations, by age and sex, for the base year of the projections.
ONS only produces one set of subnational population projections. At the consultation, some users said that they would like to see variant subnational population projections. Research into the feasibility of producing these variants, for example a variant based on longer local trends for internal migration, is under consideration
Subnational population projections are usually published every two years approximately 6 months after the national population projections. All planned future releases are announced on the release calendar. It is likely that the next set of subnational population projections will be published in the first half of 2016.
ONS do not currently produce subnational population projections by characteristics such as ethnicity or country of birth. This is because the methodology and data sources used do not currently support the production of projections at this additional level of detail.
The subnational population projections methodology is based on the internationally recognised cohort component method which takes the population at the start of the year and adjusts for births, deaths and migration to calculate the population at the end of the year. The births, deaths and migration figures used are calculated from rates derived from trends over the previous 5 years. The overall results are constrained to be consistent with the national population projections. For more detail about the methodology, please see the methodology document of this release.
The base year data used in these subnational population projections are the mid-2012 population estimates published on 26 June 2013.
Registrations of births and deaths collected by the civil registration system at the General Register Office are used to calculate fertility and mortality rates. Internal migration estimates and cross-border migration estimates are based on administrative data. International migration estimates are based on data from the International Passenger Survey, together with asylum seeker data from the Home Office and National Asylum Support Service. The mid-2012 population estimates and the revised back series of population estimates, which are used in the calculation of trends, reflect the results of the 2011 Census. For more information, please refer to the methodology document of this release.
Internal migration covers movement of people between different local areas within England. Cross-border migration covers movement of people between the constituent countries of the UK, that is between England and Wales/Scotland/Northern Ireland. International migration covers movements of people to/from a local authority from/to outside the UK according to the United Nations definition of an international long-term migrant. That is, someone who changes their country of usual residence for a period of at least a year, so that the country of destination effectively becomes the country of usual residence.
The 2012-based subnational population projections are consistent with the 2012-based national population projections published on 6 November 2013. Across all subnational areas, they sum to the national projection for England in terms of population (by single year of age and sex) and components of change (births, deaths, cross-border and international migration). For more detail, please refer to the methodology document of this release.
For information on the methodology used to constrain the subnational population projections to the national population projections and the impact that this may have on the subnational population projections, please see the methodology document of this release.
Armed forces personnel are treated as a ‘static population' such that they retain the same size and age-sex structure as the base year throughout the period of the subnational population projections. The 2012-based national population projections made an allowance in the short term assumptions of migration for planned returning armed forces personnel between the years 2014/15 and 2017/18. The 2012-based subnational population projections have distributed these armed forces across all local authorities according to the distribution of international migrants. Please refer to the methodology document of this release.
Assumptions about numbers of births, deaths and migration are based on trend data and constrained to the 2012-based national population projections. For more detail about the methodology, please refer to the methodology document of this release.
Subnational population projections are projections not forecasts and, as such, do not attempt to predict the impact that government policies, development aims, changing economic circumstances or other factors might have on demographic behaviour. The projections do not consider the capacity of an area to accommodate changes in its population, nor do they attempt to predict how any local level information may affect the projections. The projections simply indicate the population levels and age structure that would result if the assumptions and trends based on observed estimates were to be realised. A departure from the trend based approach is only considered where the calculated trend observed in one or more of the migration components is demonstrably biased in some way - possibly because of unusual patterns in the 5 year period on which the trends are based.
We would expect the population projections to be different, but larger differences are likely to be due to the fact that the trends used in the 2012-based subnational population projections are based on a historical population series rebased following the 2011 Census while the trends used in the 2011-based subnational population projections are based on an older population series that does not reflect the findings of the 2011 Census. Other reasons include:
changes in the population estimate, used as the base year in the projections, between mid-2011 and mid-2012,
changes in the trends (births, deaths and migration),
changes in assumptions for international migration at a national level.
The subnational population projections are used as an input into household projections, and are used by central and local government for planning education and health services provision. They are also used by local authorities for local planning or as a basis for producing their own projections. The subnational population projections are used to inform local level policy and planning. Other uses include, but are not limited to:
emergency service provision,
business development (including construction of new developments),
reviews of policy by officials, individuals and special interest groups,
survey design and development,
planning local events,
calculation of local rates, measures and indicators,
academic research, and
The following information is available as part of the release and will help you to understand the subnational population projections for your area:
Population pyramids are available which show how the population for an area is likely to change over time.
Components of population change (3.17 Mb Excel sheet) show whether births, deaths or migration are responsible for change over time.
The Statistical Bulletin includes examples of how the data described above is used to explain population change in a number of local authorities.
Yes. The mid-2013 population estimates will be based on the mid-2012 population estimates and will take account of the recorded number of births, deaths, and estimated migration in the year from mid-2012 to mid-2013. The population projections are based on the mid-2012 estimates that have been rolled forward to mid-2013 using components based on recent trends rather than current data that would be used in the production of the estimates.
As a result of inherent uncertainty of demographic behaviour, any set of projections will inevitably be proved wrong, to a greater or lesser extent, as a forecast of future demographic events or population structure. Projections are uncertain and become increasingly so, the further they are carried forward in time, particularly for smaller geographical areas. Care should be taken in interpreting these data, particularly when broken down by age and sex. The projections are more robust at greater levels of aggregation, either by age or by area, since more detailed levels mean smaller counts contributing to the projection process. Accuracy assessments, most notably the 2006-based accuracy report (234.6 Kb Pdf) , have been made comparing the early years of the projected period in previous sets of projections with corresponding mid-year estimates.
This release includes unrounded subnational population projection data by single years of age, allowing users to aggregate the data as they wish. However, users should note that subnational population projections are more robust at greater levels of aggregation. We advise that the data should not be aggregated to less than 5 year age groups and rounded to the nearest 100 people if quoted in any publication, presentation or documentation.
Population estimates for the Isles of Scilly and to a lesser extent the City of London, are considered to be less reliable given the small population in these areas. Certain individual populations (specific age groups in specific local authorities) may also be very small. The population projections employ the same basic methodology across the country, so these areas are treated consistently. However projections for these areas should be treated with particular caution.
The intercensal population estimates between mid-2002 and mid-2011 were rebased so that they were in line with the 2011 Census. Since trend data from these years and the mid-2012 population estimates (also based on the 2011 Census) are used in the production of the subnational population projections, their improved accuracy will be reflected in the projections.
Earlier improvements made to the population estimates and some of the components of change as a result of the Migration Statistics Improvement Programme will also be reflected in the subnational population projections. Small changes have been made to the subnational population projection processes to incorporate these improvements.
ONS are aware of the delay observed in some young people when they re-register with a doctor after changing their address. This is likely to have an impact on the internal migration estimates and the population of working age groups in some areas. We have not made an adjustment to the 2012-based subnational population projections to address this issue because it has not been possible to develop a transparent method that can be applied across all areas within our timescales. However, ONS is working on a project to make improvements to the methodology to estimate internal migration and this work will help us to develop a more comprehensive adjustment that will address this issue. For further information on this issue, please see the methodology document of this release.
Following the 2011 Census, the rolled forward population estimates were rebased so that the population estimates for mid-2011 were in line with the 2011 Census. After making allowances for methodology changes made during the decade, the remaining difference is referred to as Unattributable Population Change (UPC). Research has found that UPC is likely to be due to either sampling variability and/or other methodology issues that may have occurred in the following:
international migration estimates,
census estimates (both 2001 and 2011), or
internal migration estimates (at subnational level only).
At consultation, we proposed that no adjustment was to be made to the 2012-based subnational population projections for Unattributable Population Change (UPC). This is because our research concluded that an adjustment would only be made if it could be demonstrated that UPC measures a bias in the trend data that will continue in the future. In addition, it is also unclear what proportion of the UPC is due to sampling error in the 2001 Census, adjustments made to the population estimates after the 2001 Census, sampling error in the 2011 Census or error in the intercensal components - so methodologically, this would be difficult to adjust for.
The UPC is unlikely to be seen in continuing subnational trends because:
if it is due to either the 2001 Census or 2011 Census, then the components of population change will be unaffected, and
if it is due to international migration, it is likely that the biggest impacts will be seen earlier in the decade and will have less of an impact in the later years, because of improvements introduced to migration estimates in the majority of these years.
Therefore, ONS has concluded that no adjustment should be made. Further information on UPC can be found in the consultation documentation.
While ONS recognises that Unattributable Population Change (UPC) can have a significant impact at some age and sex groups in some local authorities, it was concluded that there is insufficient evidence that an adjustment for UPC should be made to the subnational population projections. This is because we do not have sufficient evidence that the UPC measures a bias in the trend data that will continue in the future.
A consultation was held from 20 January 2014 to 28 February 2014 to obtain comments on the 2012-based subnational population projections to improve the quality and acceptability of the subnational population projections when published. The consultation was conducted in line with the UK Statistics Authority Code of Practice. The consultation was held via consultation papers published on the ONS website and consultation meetings. The meetings were held in London, Birmingham and Manchester where local authority representatives were able to see the provisional migration assumptions and projections data for their area. This allowed users to see the impact of the proposed methods and enable more informed discussions and feedback about the data sources and proposed methodology. Previous consultations on the subnational population projections have been managed via correspondence so this is the first time that users have attended a meeting to review the data for their area. The new approach ensured that data confidentiality was maintained.
The consultation documentation is available to download from the ONS website. This now includes a report documenting the responses to the consultation, published as part of this release.
Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: firstname.lastname@example.org
These National Statistics are produced to high professional standards and released according to the arrangements approved by the UK Statistics Authority.