National Statistical

The latest ONS mid-year estimates suggest the UK population has reached 67 million for the first time. But the year to June 2020 saw considerably slower growth than recent decades as the effects of the pandemic took hold. Increased deaths and restrictions on international travel clearly affected the national population level, while the first lockdown limited internal migration within the UK.  Here Rich Pereira explains how best to interpret the new data and how the ONS has been addressing the data collection issues the pandemic imposed.     

The 12 months to June 2020 can be broken into two clear parts. The first eight months, when births, deaths and migration patterns were similar to trends seen in recent years, and the four months from March, when the first wave of coronavirus hit.

For the UK as a whole, the year to mid-2020 saw the highest number of deaths since the year to mid-1986. The major factor in this increase is the coronavirus (COVID-19) pandemic which accounted for more than 55,000 deaths in the year to mid-2020 and were recorded in the last part of the period.

During the period March to June the UK was in lockdown for much of the time and these restrictions had direct impacts on population change at both the national and subnational level.

Take, for example, internal migration. Cast your mind back to that period and house sales plummeted as moving became nigh on impossible.

Not only that, following the temporary closure of universities across England & Wales large numbers of students left their place of study and returned to their parental home or other residence.  However, if they were unable or did not register with a new GP after making this move, then their move would not be captured by the administrative data sources used to compile internal migration. It is therefore unsurprising that in the year to mid-2020 there were estimated to be around 11% fewer internal moves than in the previous year.

Increased uncertainty

Our population estimates are built forwards from the 2011 Census, adjusting each year for births, deaths, internal and international migration. Now 10 years after the previous census, and before we publish Census 2021 results, the challenge in producing the best possible estimates is at its greatest.

To add to the complexity, we’ve had to do things differently this year, due to delays in birth registrations and changes in the collection of statistics on migration since the beginning of the pandemic.

Ordinarily we’d be able to use the population projections to provide a reliable indication of the ‘current population’ but the latest projections do not (and could not) take account of the impact of the pandemic. Whilst we will be producing 2020-based national population projections, because of the increased uncertainty and complexity caused by the pandemic, we will not be producing 2020-based subnational population and household projections, as we explain here. Instead we plan to produce 2021-based subnational population and household projections. We’re aiming to release these in 2023 to allow us to better understand the effects of the (COVID-19) pandemic and to enable us to use results from Census 2021 as the baseline moving forwards.

Results from Census 2021, the first of which are due out in March next year, will ultimately provide the most accurate picture of the population for the coming years.

So, what do we know now?

The pandemic has highlighted the importance of providing more timely population estimates. Whilst today’s release contains our most up to date official estimates, they are 12 months old and we have some more up-to-date information on population change. Deaths data, for example, are readily available on a weekly basis on the ONS, NRS and NISRA websites.

These provide a very accurate indication of the number of deaths registered in the year to June 2021.

Births data are published weekly by NRS and NISRA and a set of provisional birth data for England and Wales has just been published up to end March 2021.

We also know that, as restrictions have eased, house sales data shows an increased demand in people buying property in the countryside, for example. This could suggest increased movement in people leaving towns and cities for a life in the countryside.

However, we don’t have all the data required to provide a complete picture of population change to the present day, particularly at a subnational level.

International migration is much more difficult to estimate as our traditional measure, Long-Term International Migration underpinned by the International Passenger Survey, was disrupted due to the pandemic.

We’ve previously stated our intentions to measure international migration using administrative sources, including Home Office visa data. This work remains challenging due to coverage gaps in the data, for example. We’ve announced today that in the future, our work on administrative based migration estimates will be supported by statistical modelling, which we’ve developed over the past year to help plug those gaps.

Modelling of international migration allows us to produce timely population estimates and is consistent with our wider ambitions to transform population and social statistics. We’ll have further updates on this later in the year.

Population estimates are also used to ensure our surveys remain representative of the overall population, and to do that we ‘reweight’ our surveys like the Labour Force Survey (LFS) with the latest population data. Our work on this is continuing. We are planning to publish reweighted LFS estimates next month and will also provide a more robust estimate of the size of the non-UK born population later this year.

Coming up

Next month we are also planning to provide information about how we will meet the recommendations of the recent review of population estimates and projections, which was carried out by the Office for Statistics Regulation.

We will continue to provide regular updates and are always keen to receive feedback. Please contact us at pop.info@ons.gov.uk with any comments.