Context

Background

In recent years, ONS has been seeking ways to improve its population estimates for England and Wales by making more use of administrative data in its methods. These efforts have resulted in the admin-based population estimates (ABPEs). ONS has published a series of articles demonstrating its development of the ABPEs (labelled official statistics in development) using the dynamic population model (DPM), a Bayesian statistical modelling approach. ONS had been working towards replacing the current traditional cohort component method used to produce national and subnational population estimates by age, sex and local authority for England and Wales with the ABPEs. ONS’s prioritisation and focus had been to stabilise and improve the new methods of the ABPEs so that users could be confident when the change in methods occurred.

ONS has since made the decision not to transition from the current traditional cohort component method used to compile its annual MYEs to admin-based population estimates (ABPEs) as its official estimate of the population in England and Wales. This is a significant shift in ONS’s direction for the future development of population estimates. ONS now plans to focus its activity on a programme of continuous improvement to the current methods to produce the mid-year estimates.

The findings presented in this report are based on a due diligence check of the mid-2024 national and subnational mid-year estimates for England and Wales. We conducted desk research and spoke to a range of population statistics users to help inform our judgements and findings. The report is also informed by insights from the Office for Statistics Regulation (OSR)’s three previous regulatory projects on ONS’s population statistics. With its publication, we consider our 2021 review of population estimates and projections, the 2024 assessment of the ABPEs in England and Wales and the 2025 assessment of the 2021 Census in England and Wales to be closed.

User views sought on the ONS’s development of the ABPEs are reported in our letter dated 13 March, confirming the closure of the ABPE assessment project.

Mid-year estimates

Measuring the size of the population accurately is essential to understanding different aspects of our lives and communities. It is also inherently challenging. ONS publishes annual mid-year estimates (MYEs) of the population of England, as well as annual population estimates for Wales. This includes statistics at the following geographic breakdowns:

• England and Wales combined
• England and Wales at country level
• Regional
• Local authority

The annual MYEs are produced from data on four aspects of the population – namely stock, births, deaths, and migration:

1. Stock (the size of the population on a given day) is taken from the census. Estimates are rolled forward to 30 June (mid-year), and for consecutive years between censuses.
2. Birth data are obtained from birth registrations. ONS publishes birth statistics for England and Wales. ONS also publishes UK birth figures.
3. Death data are obtained from death registrations in constituent countries, similar to birth registrations.
4. Migration data include estimates of both international and internal migration. An international migrant is defined as a person who changes their country of usual residence for a period of at least one year. Internal migration estimates account for the movement of people within England and Wales and to, or from, the rest of the UK (cross-border flows).

Finally, adjustments are made to account for special population groups that are not captured by the internal or international migration estimates, such as prisoners and armed forces.

ONS also collates data from National Records of Scotland and Northern Ireland Statistics and Research Agency to produce population estimates for the UK. Estimates for each of the UK constituent countries are compiled using a common methodological approach, with the aim to be as consistent as possible.

The cohort component method

To produce the MYEs, ONS takes data from the most-recent census and rolls them forward to 30 June (mid-year) to determine the population stock. ONS then updates the MYEs for population change (also known as population flow) using births, deaths, and migration data. This method is referred to as the cohort component method and accounts for a full year’s population change between 1 July and the subsequent 30 June. However, in census years, the MYEs instead account for the change between the day the census was conducted (for example, 21 March 2021) and mid-year (30 June 2021), a period of three months.

Between censuses, the estimates tend to drift and their quality deteriorates. This drift can be caused, in principle, by an error in the population baseline from the preceding census (census base) and by the accumulation of error in population flow data until they can be rebased again using data from the next census. So, the more time that has passed since the last census took place, the less accurate the estimates. For example, in November 2023, ONS rebased its MYEs (2012–2021) once data from Census 2021 for England and Wales became available and revised the back series of components of population change, as depicted in the charts below. In England, this led to an increase in rebased estimates for females and a decrease for males compared to the rolled-forward MYEs. In Wales, rebased estimates decreased the number of both males and females compared to the rolled-forward estimates.

Figure 1. Comparison between rolled-forward mid-year population estimates and rebased back series by sex, England, 2012–2021 (Figure 5).

Figure 2. Comparison between rolled-forward mid-year population estimates and rebased back series by sex, Wales, 2012–2021 (Figure 8).

Source: Figures 5 and 8 – ONS website.

This rebasing is standard practice and conducted by ONS as part of the current process to estimate the size of the population, with benchmarking against the census every 10 years. Most of the revisions made have been for net international migration flows as a result of improved methods and data over time.

The value of the dynamic population model (DPM)

The extensive research and progress made to produce administrative-based population estimates using the Dynamic Population Model (DPM) offers a wealth of insight and opportunity for ONS to harness in its continuous improvement plans for the MYEs. As a Bayesian statistical model, the DPM is designed to estimate components of population change, incorporating prior knowledge and estimates of stocks, flows, and rates and their uncertainties using a Bayesian approach. As a statistical model, it updates prior beliefs using new evidence and as an approach can account for associated statistical uncertainty more effectively than the current MYE methods. Most uncertainty is associated with migration. Methods for international migration estimates are currently under development, and changes to these methods lead to revisions.

One of the strengths of the DPM is that it should, in principle, avoid the existence of unattributable population change (UPC) due to the way it balances the stocks and flows. UPC refers to the remaining population change that can be seen between the census-based population estimates and the rolled-forward mid-year estimates that cannot be explained by any of the components of change. Continued improvements to the methods used to produce migration estimates should also help to ensure greater accuracy and coherence in subnational estimates, especially for local authorities with high population churn, large student populations, or complex migration patterns.

The ONS article Evaluating the accuracy of the admin-based population estimates for England and Wales, published on 4 March, provides evidence that ABPEs between 2011 and 2021 show less bias than MYEs at the subnational level, although MYEs perform better nationally. This reinforces the potential value of DPM-based methods, particularly for LAs experiencing high population turnover, large student populations, or significant migration flows.

Although ONS has now decided not to move to new methods for the 2026 MYE publication, there is significant value in ensuring that insight from the DPM and ABPE development programme is not lost.