The Office for National Statistics has been working at pace to produce more timely and reliable population estimates. By using multiple data sources to reduce reliance on a 10-yearly census, we are proposing a new system which would be responsive to change and provide decision makers with the evidence to put the right services in the right places. Jen Woolford updates on our progress to date, explains how the new data should be used and looks ahead to this summer’s public consultation.
We know that our historical census and mid-year estimate approach doesn’t meet the full range of user needs. Therefore, we are beginning the transition to a new system that is less reliant on a census. Our initial focus is population and migration estimates at local authority level that are similar to those users get currently – but using different methods which mean the estimates will be available more quickly and more stable over time.
Our new dynamic population model (DPM) will deliver timely, coherent and accurate statistics, providing new insights to support a better understanding of the population of England and Wales; critical for effective decision making to improve people’s lives.
Today’s admin-based population estimates (ABPEs), which is the output from the DPM, give a timely indicator of population change. They show that the population of England and Wales has increased by 1.0% between 2021 and 2022, to 60.2million. The largest growth has been seen in some London authorities, which is in line with our expectations as people returned to cities following the pandemic.
Experimental statistics
These estimates are classified as experimental statistics with plans in place to move them to National Statistics over the next year, signalling our confidence in moving to a new transformed system. For the first time, we’ve been able to take the statistics down to lower layer super output area for five local authorities, highlighting the ongoing development of the system.
However, while we continue to talk to users, improve our methods and go through this transition, this year the admin-based population estimates should only be used as an indicator of population change. We are publishing them to illustrate our methods development and seek feedback from our users. Our final official mid-year population estimates (MYEs) for 2022 will be published in September, and as we always do after a census, we are revising our historic migration statistics to rebase population estimates and will also publish these in September.
We will then aim to use improved methods to produce provisional mid-2023 estimates via the DPM at the end of this year. This would be six months earlier than we are able to using our existing method.
A flexible system
A significant advantage of the DPM is its flexibility. While it uses administrative data sources as stocks, flows and indicators of demographic change each year, it can incorporate other data sources as and when they become available. This could include sources relating to local areas or particular population groups, or sources that represent the total population.
The model can also adapt to quality issues in our underlying data, drawing strength from a range of sources and balancing information available from population stocks and flows based on their respective measures of precision.
This is demonstrated by our updated ABPEs having been produced by the DPM despite quality issues in some of the data used for the internal migration component of the estimates. This is a significant benefit compared to our current official mid-year estimates, which have been delayed due to this data issue.
Our progress so far
Our national and local population and migration statistics need to be high quality, producing estimates by age, sex and for local authorities for England and Wales, including patterns of internal and international migration, as well as housing stock.
We have demonstrated our ability to produce more timely and frequent population and migration statistics that can sustain a better level of quality over time than the census-based estimates.
We are now routinely producing official statistics on international migration that are derived from administrative data. In May 2023 we published estimates for 2022 using Home Office data to measure long-term migration of non-EU visa holders as well as, for the first time under our new methods, asylum seekers.
The flexibility in our modelling combined with new innovative methods shows how we can adapt to measure the population using different definitions. We’ve produced five case studies – for Manchester, Blackpool, Cambridge, Ceredigion and Newham – so that users can see how our proofs of concept can be used to inform detailed insights into their local areas. We have also delivered research showing progress towards priority user needs on daytime population insights. We are publishing all of this research to get feedback which we will use to continue to learn and develop our methods.
Providing more detail
We also need to deliver high quality estimates of the size and distribution of population sub-groups down to local level, covering protected characteristics, such as disability and ethnicity, and other subgroups, including armed forces veterans and refugees. In this report we summarise our research to date.
For several topic areas, we have demonstrated the potential for producing small area statistics more often and more quickly than is currently possible. Here we describe our next steps for these topics and those characteristics we have not yet explored.
Our Refugee Integrated Outcomes study illustrates the potential for delivering new insights into the outcomes experienced by refugees. This proof of concept demonstrates a significant step forwards in creating new longitudinal data assets (such as the Longitudinal Population Dataset, which links data over time). This would support research to improve people’s lives through understanding how outcomes are affected by life experiences, while ensuring privacy and confidentiality.
We have also expanded our research to produce admin-based census-related outputs on travel to work, armed forces veterans and health indicators.
Taken together, this progress gives us confidence in our ability to transition to our proposed new system which will be set out in the population and migration statistics consultation beginning on 29 June.
Consultation
For the next four months, the ONS will be gathering feedback on how far the proposals (including the research published this week) meet user needs and where it should be prioritising future research, particularly on those topics where further research is needed.
Responses to this consultation will inform a recommendation by the National Statistician, as set out in the 2018 Census White Paper, on how the ONS should produce statistics about the population in future.
The consultation will launch at this link on Thursday and will be live until October 26.
Jen Woolford is Director Population Statistics at the ONS