This report answers some of the questions users most frequently ask on national population projections in general as well as, more specifically, on the 2012-based release.
National population projections are prepared by the Office for National Statistics (ONS) on behalf of the National Statistician and the Registrars General for Scotland and Northern Ireland. They are produced every two years and provide projections of the future size and age structure of the population for the UK and its constituent countries. The national population projections are National Statistics, which means that they undergo regular quality assurance reviews and are produced free from political interference.
The national population projections are based on the latest available mid-year population estimate and a set of demographic assumptions about future fertility, mortality and migration based on analysis of trends and expert advice. They are produced using the internationally accepted cohort component methodology. This method accounts for changes which increase or decrease the population (births, deaths and net migration) and models the effect of these changes and the passage of time on the age structure of the population.
The national population projections are not forecasts and do not attempt to predict the impact that future government policies, changing economic circumstances or other factors (whether in the UK or overseas) might have on demographic behaviour. They simply provide the population levels and age structure that would result if the underlying assumptions about future fertility, mortality and migration were to be realised.
Projections are made of the usually resident population of the UK and its constituent countries, whatever their nationality. The usually resident population includes all long-term international migrants (people changing their country of usual residence for at least one year). However, the usually resident population does not include short-term migrants who come to or leave the UK for less than a year.
The assumptions about future levels of fertility, mortality and net migration are agreed in liaison with the devolved administrations (namely National Records of Scotland (NRS), the Northern Ireland Statistics and Research Agency (NISRA) and the Statistical Directorate of the Welsh Government (WG)), following consultation with key users of projections in each country and advice from an expert academic advisory panel. Details of the membership of the panel and minutes of meetings are published in the background and methods section of the release.
The main purpose of the national population projections is to provide an estimate of the future population of the UK (and its constituent countries) as a common framework for use in national planning in a number of different fields.
The national population projections are widely used across government for planning purposes. Examples include:
The Office for Budget Responsibility use the projections as a key input to their long-term fiscal projections published in the fiscal sustainability report.
The Department for Work and Pensions use the projections extensively to produce forecasts of expenditure for benefits and pensions and as a key input for analysis on policy areas such as extending working lives.
The Department for Education use the projections as the basis for their projections of future school pupil numbers.
The national population projections also provide the base for other products such as subnational population projections and household projections, which are widely used for resource allocation and planning.
Projections are uncertain and becoming increasingly so the further they are carried forward in time. It is vital that users of population projections, especially those with long-term planning horizons, take account of this uncertainty in their planning. In addition to the principal (main or central) projection, some variant projections are also published based on alternative, but still plausible, assumptions. These variant projections provide an indication of uncertainty by allowing users to consider the impact upon the population if future fertility, mortality and migration differ from the assumptions made for the principal projection. The publication of variant projections is an internationally recognised method for illustrating the uncertainty associated with population projections.
Also produced are a number of special scenario variants that allow users to consider what ifs. So, for example, the ‘No mortality improvement’ variant shows what the population would look like if mortality rates stayed at similar levels to those currently observed. This will be published in December. Some allow decomposition of the projections. For example, the ‘Natural Change’ variant looks at what would happen if net migration was zero at every age. By comparing this to the principal projection the impact of the migration assumptions can be assessed.
Projections are uncertain and become increasingly so the further they are carried forward in time. For this reason, analysis of the projection results mainly focuses upon the first 10 or 25 years of the projection period, which corresponds with the planning horizons of the majority of users of the projections, whilst recognising that uncertainty will be greater over a 25 year period. However, some key users require projections over a longer period for modelling purposes, and in accordance with the government’s transparency agenda, results for the 2010-based and 2012-based projections have been published for up to 100 years ahead. However, caution should be used when interpreting these longer-term projections as projections become increasingly uncertain the further into the future they go. Variant projections are also available to aid interpretation of the uncertainty.
The latest set of national population projections is the 2012-based projections published by ONS on 6 November 2013. They are based on the estimated population at 30 June 2012 and are the first set to be based on results of the 2011 UK Census. The principal (central) projection is based on assumptions considered to best reflect demographic patterns at the time they were adopted. ONS also produce a number of variant population projections based on alternative, but still plausible, assumptions of future fertility, mortality and migration. Some special case scenario projections are also published. Nine variant projections are published alongside the principal projections on 6 November 2013, and a further seven variant projections will be published on 10 December 2013.
The latest national population projections are available from the 2012-based national projections release page. Detailed information on the assumptions underlying the projections and the methodology used to produce the projections can be found in the published reports. Information on seven further variants will be published on 10 December 2013, and the reference volume in Spring 2014. There are a number of interactive tools to help users find or view information:
An interactive table download tool - that enables users to find the reference table they need allowing them to select the country, variant and table content.
Interactive graphs - for the principal population projection and the associated components of population change to mid-2037. Users can select the country they wish to look at, there are some selected age groups to choose from (for the population charts).
Interactive population pyramids – that are animated graphically to show how the age and sex structure of the population is projected to change in the UK under different variants.
In accordance with the government’s transparency agenda, ONS has published more detailed tables showing projected population for users who required this for modelling. These tables contain population by single year of age to mid-2112 and can be found at the end of the reference tables numbered Z1 to Z6. However, caution should be used when interpreting these longer-term projections as projections become increasingly uncertain the further into the future they go. Variant projections are also available to aid interpretation of the uncertainty.
Yes. These projections are based on the 2012 mid-year population estimates which themselves are based on the 2011 Census population estimates. The demographic assumptions of fertility and mortality are based on updated fertility and mortality rates which have also been updated to reflect the revised population data series for England, Wales and Northern Ireland. The revised back series for Scotland was not available when the assumptions were set and have therefore not been incorporated. Analysis of the 2011 Census results has not highlighted any concerns with taking this approach.
Since the 2010-based projections were published the results of the 2011 Census have been released and used to rebase the population estimates series. There were around 500,000 more people estimated by the 2011 UK Census than had been previously estimated. At mid-2012 the estimated population of the UK was about 460,000 higher than that projected for mid-2012 in the 2010-based projections. Therefore the 2012-based projections start from a higher base. As growth is projected to be slower this results in there being little difference by mid-2022 in the projected total population.
ONS produces a principal (central) projection and also a number of variant projections based on alternative assumptions. The 2012-based principal population projection for the UK projects that the population will increase by 4.3 million over the next 10 years from 63.7 million at mid-2012 to 68.0 million at mid-2022, an annual average rate of growth of 0.65 per cent. It is projected that the UK population will be 73.3 million at mid-2037, a total increase of 9.6 million over the next 25 years. More detailed commentary is available in the Statistical Bulletin and the 2012-based National Population Projections Reference Volume (PP2) that will be published online in March/April 2014.
ONS produces a principal (central) projection and also a number of variant projections based on alternative assumptions. The 2012-based principal population projection for the UK assumes:
a long-term average completed family size of 1.89 children per woman;
life expectancy at birth in 2037 of 84.0 years for men and 87.3 years for women, with constant rates of mortality improvement assumed thereafter;
long-term annual net migration to the UK of +165,000 per year.
The variant projections illustrate future population change under alternative scenarios. The assumptions underlying these variant projections can be found in the statistical bulletin.
Of the 9.6 million projected increase in the UK population over the next 25 years, 57 per cent is projected natural increase (more births than deaths) and 43 per cent is projected net migration. However, future numbers of births and deaths are themselves partly dependent on future migration. Taking this into account, about 60 per cent of projected population growth between 2012 and 2037 is expected to be either directly or indirectly due to future net migration. A note considering the overall impact of assumed net migration on future population growth is available in the results summary.
The 2012-based principal projection assumes that levels of net migration to the UK will be +165,000 per year from 2018/19 onwards. Migration has shown a decrease in the last couple of years but it is not widely thought that this is the start of a downward trend, but more of a step change due to the introduction of specific policies to reduce migration. The assumptions have been set based on migration data to mid-2012. An allowance has been made in the short-term for the planned return of armed forces to the UK. Migration from accession countries has been modelled with other migration flows with a step intervention to account for the step change at the time of accession. No allowance has been made for future migration from countries gaining accession to the EU since their numbers are unknown.
There were a few drivers for changing the methodology used to set migration assumptions.
The previous methodology for determining long-term assumptions of future migration was introduced for the 1991-based national population projections with some later modifications made for the 1996-based NPPs. However migration patterns and data sources have changed since then so it was agreed following publication of the 2010-based NPPs that a review of the methodology would be carried out.
In the UK Statistics Authority assessment of population projections published on 20 April 2011, requirement 6 was to "Publish a plan outlining how the 2011 Census data will be used as the basis of reviewing and improving the methods used to produce the population estimates and related outputs (para 3.18)." In para 3.18 it stated "...across all the outputs, there is no systematic programme of review and some methods have not been reviewed for some time. For example, the last formal review of the methods underlying the population projections was in 1993". Therefore a review of the methods would be in accordance with the assessment requirements for national statistics status.
During production of the 2010-based NPPs there was a user need identified for setting the migration assumptions in terms of gross flows rather than as net migration. However it was not feasible to simply introduce this using the previous methods.
Details of the review and the new methods were published on 21 August 2013.
The latest data have not been modelled using the old methods and therefore it is not possible to say exactly what the assumptions might have been if the methods had not changed.
The migration assumptions setting process includes input from an academic expert panel, discussions with the Devolved Administrations (DAs), and extensive formal consultations held in each of the 4 countries of the UK.
ONS migration assumptions are not created based on where people are migrating to, or from, and they do not provide forecasts or projections of movements of people from particular countries to and from the UK.
ONS do publish estimates of long-term international migration that provides information on where people migrate to, or from. This can be found in the international migration section of the ONS website.
The population is projected to rise most quickly for the oldest age groups. The number of people aged 80 to 89 is projected to nearly double from 2.5 million to 4.5 million over the 25 years to mid-2037. The number of people aged 90 and over is projected to more than treble, from 0.5 million in mid-2012 to 1.7 million in mid-2037. The increases in State Pension age mean that the ratio of working age people to each person of state pensionable age is projected to increase slightly from 3.21 in mid-2012 to 3.39 in mid-2022 and then to decline to 2.74 in 2037.
In the short term the projected future population size is slightly higher than that in the previous (2010-based) projections. This is due to the projections starting from a higher base as the 2011 UK Census estimated 500,000 more people than in the rolled forward estimates. Population growth is projected to be lower in the 2012-based projections than the 2010-based, and therefore in the long term the 2012-based projections are lower. The lower growth will be due to differences in the assumptions of future fertility, mortality and migration.
The principal long-term fertility assumption has been increased in the 2012-based projections to 1.89 children per woman, from 1.84 in the 2010-based projections; in the short term levels of fertility are assumed to be lower than in the 2010-based projections. More information on the changes can be found in the fertility assumptions section.
In general the annual rates of mortality improvement in the longer-term have remained the same as in the 2010-based projections. More information can be found in the mortality assumptions section.
The 2012-based long-term assumption for net migration to the United Kingdom is lower at +165,000 each year, compared with +200,000 each year in the 2010-based projections. These changes reflect the most recent trends in international migration. New methods have been used to model migration trends in setting the migration assumptions for the 2012-based projections. More information can be found in the migration assumptions section.
The mortality rates for the first year of the projection, mid-2012 to mid-2013, are based on the best estimates that could be made in the autumn of 2013 of the numbers of deaths at each age in 2012-13. The provisional number of deaths registered in this period was higher than usual, resulting in the projected expectation of life at birth being lower than may be expected in the first year of the projection than in following years. This is not unusual when taken in context with normal year on year variation.
The ONS published an analysis of the past accuracy of national population projections in 2007 in Population Trends 128. This analysis considered the 1955-based to 2004-based projections and compared these projections with the latest estimates of the UK population up to mid-2005. The analysis found that the mean absolute error of the projected total UK population 20 years ahead was about 2.5 per cent overall (when considering 1955-based to 1985-based projections), and lower than 2 per cent when just the most recent (1975-based to 1985-based) projections were considered. This would correspond to around 1.4 to 1.8 million people (2.0 to 2.5 per cent mean absolute error calculated on the 2012-based principal projection for 2032). The largest differences between projected and actual populations were found to be for the youngest and oldest ages, while projections of the working age population were found to be comparatively accurate.
ONS do not publish confidence intervals around the projections. The projections are inherently uncertain and become more so the further they are carried forward in time, particularly for smaller geographical areas. Variant projections are produced to provide an indication of uncertainty by allowing users to consider the impact upon the population if future fertility, mortality and migration differ from the assumptions made for the principal projection.
The 1955-based national population projections, produced prior to the 1960s baby-boom, projected the lowest future population sizes of any official projection. The projected population for 1995 (40 years ahead) was 53 million, some 5 million lower than the population estimate for mid-1995. In contrast, the 1965-based projections, produced at the height of the 1960s baby-boom, projected the highest future population sizes - the projected population for 2001 (36 years ahead) was 75 million, some 16 million higher than the population estimate for mid-2001. This illustrates the difficulty in projecting the population during periods of demographic change, and the importance of bearing in mind the uncertainty of projections, particularly over longer periods of time.
According to the 2012-based principal projection, the UK population will reach 70 million in 2027, in the same year as projected by the 2010-based projections. In comparison the 2008-based projections indicated that it would reach this level in early 2029, and the 2006-based projections in early 2028. A population level of 70 million is of no special demographic significance, although some customers of population statistics will inevitably be interested in when and how fast the population might reach numerical milestones such as 50 million, 60 million and 70 million. The UK population is estimated to have reached 50 million in 1948 and 60 million by mid- 2005. It is projected that the population will rise from 60 to 70 million over a period of 22 years, compared to the 57 years over which it rose from 50 to 60 million.
Under the principal projection assumptions, the size of the UK population is projected to continue increasing over the projection period. However projections are uncertain and become increasingly so the further they are carried forward. The low population variant projection indicates that if all the assumptions for fertility, mortality and net migration were set at lower levels than that assumed for the principal projection, the population would start to decline from mid-2043 onwards.
The release on the 6 November 2013 covered the main (principal) 2012-based projections and nine key variant projections based on alternative, but still plausible, assumptions. On 10 December 2013, ONS will release a further seven variant projections that illustrate additional alternative scenarios to supplement the projections released in November. This December release will form part of the overall release of the 2012-based national population projections, the main results of which are described in the Statistical Bulletin and News Release published on 6 November.
We carried out analysis on how often users have accessed the tables from 2010-based variant projections which showed that some variants had not been accessed by users. We then consulted with users on whether these variants were required and as a result we have not produced the high medium-term dependency, low medium-term dependency, stationary and ‘no mortality improvement & zero net migration’ variants.
The 2012-based national population projection release includes projections for Great Britain, England & Wales, England, Wales, Scotland and Northern Ireland as well as the UK. Projections for subnational areas for England produced by the ONS are published typically about 6 months after the national projections. Subnational projections for Wales, Scotland and Northern Ireland are produced by the WG, NRS and NISRA respectively.
The latest Eurostat projections are still EUROPOP10. A discussion about these was included with the 2010-based projections.
The next set of European projections ‘EUROPOP2013’ will be 2013-based and are planned for dissemination by Eurostat by end of March 2014.
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: media.relations@ons.gsi.gov.uk
These National Statistics are produced to high professional standards and released according to the arrangements approved by the UK Statistics Authority.