Impact of the coronavirus (COVID-19) pandemic on England and Wales Census 2021 outputs

The England and Wales Census 2021 was carried out during the coronavirus (COVID-19) pandemic and gives a unique snapshot of the population on 21 March 2021. It achieved high levels of response and provides high quality data about the population at that time. Lockdown restrictions meant that some people temporarily lived elsewhere, while some people's work and behaviours changed.

Census 2021 accurately reflects society during the pandemic which, for some topics, may be very different outside of that unique point in time. Following Census 2021, we have taken active steps to robustly estimate the population and account for students returning to study, rising immigration, and greater movement within the UK.

To support users, this article covers a range of resources detailing the pandemic's impact. This article signposts users to published information about the pandemic's impact on Census 2021 outputs, summarises topic-specific impacts, and highlights implications for users of these statistics.

Likelihood of impact

Some Census 2021 statistics are an accurate reflection at the time of the pandemic, but are not typical of the population during a non-pandemic time. For example, some people were more likely to live away from their usual home or to change their job.

There were challenges to accurately collecting these statistics because of the pandemic. For example, the form was not easily designed to identify people on furlough from work.

There are also challenges to estimating the population since 2021 using census data that reflect a very different time period. Users should be aware of these impacts when using data to inform decisions about the future.

We categorise which topic-based Census 2021 estimates are affected by the pandemic, including those most likely to be most affected, those that have some effects, or those that have little or no impact.

Topics most likely to have been affected by the pandemic are:

• geographic location of usual residents

• students

• international migration

• labour market

• travel to work

Topics with some pandemic-based effects are:

• living arrangements

• health

• unpaid care

Topics with little or no pandemic-based effects are:

• UK armed forces veterans

• ethnic group

• national identity

• language

• religion

• Welsh language

• sexual orientation

• gender identity

• education

• housing

We discuss each topic in the following sections, alongside signposting to relevant resources. We also discuss the impact on population estimates since Census 2021.

Why census data are important

Census data underpin a wide array of policy, research, and planning activities many years after Census Day. Local authorities and public bodies rely on these statistics for policy development, service provision, and funding allocation. Businesses use the data to understand customer bases and site locations. Voluntary organisations use the statistics to support community work and funding applications. Academics depend on census outputs for research.

The census data form the basis for mid-year population estimates. They also support the development of admin-based population estimates (ABPEs) by providing a benchmark against which to assess their quality.

Ensuring Census 2021 met trustworthiness, quality and value standards

The Office for Statistics Regulation (OSR) conducted a three-phase assessment of Census 2021 to ensure its outputs met the standards of trustworthiness, quality, and value set out in the Code of Practice for Statistics.

The OSR recognised our robust response to earlier requirements in June 2025 in their final phase report. However, they highlighted the need for more detailed guidance about pandemic impacts. This article aims to address that by helping users understand the weight that can be placed on affected statistics, signposting to other relevant sources of data, and explaining pandemic-related effects by topic and geography.

The coronavirus (COVID-19) pandemic brought new challenges to the operation of Census 2021. We discussed the potential impacts of the pandemic on the operational plans for Census 2021 and the design changes and contingency plans we made in response in our Coronavirus and the operational impact on Census 2021 report.

A high response rate is crucial to the success of the census. A 97% response rate was achieved across England and Wales, with at least 88% achieved in all local authorities.

Following Census Day, we undertook a rigorous and comprehensive quality assurance process, including comparing against the widest range of alternative and complementary data sources. This process was described in our Maximising the quality of Census 2021 population estimates methodology.

For the first time, we invited local authorities to review provisional census estimates, drawing on their local expertise to support our own quality assurance checks. We outline these quality considerations and the strengths and limitations of Census 2021 more generally in our Quality and methodology information (QMI) for Census 2021. Full details of the Census 2021 can be found in the Census 2021 general report.

Geographic location of usual residents

Main estimates and possible impacts

Census 2021 counted usual residents as those living in England and Wales on 21 March 2021 and who were intending to stay for a year or more.

We took steps throughout all phases to ensure local estimates were as accurate as possible, as outlined in our Maximising the quality of Census 2021 population estimates methodology. However, lockdown restrictions meant some people lived in places they did not normally reside, such as urban residents relocating to rural areas or students staying at home instead of in term-time accommodation. International migration was also disrupted, with fewer arrivals and changes of migrants' length of stay, as described in our International migration research, progress update: November 2023 article. 

These factors likely influenced where people were usually resident. The Students subsection explains how we ensured that they were accurately enumerated. We explore the pandemic's impact on international migration and living arrangements, including second homes, in the following subsections.

Other data sources

There was a decrease in London's population over the first year of the coronavirus (COVID-19) pandemic, but it likely began growing again in 2021, according to the Greater London Authority's Population change in London during the pandemic report.

The fall was attributed to young adults leaving London during lockdown, higher mortality, and an increased movement of other age groups to surrounding regions. They point to evidence of many young adults returning to London during spring and summer 2021, following the recovery of hospitality and tourism. Several local authorities referenced similar trends during the census quality assurance process.

Implications for use

Census 2021 provides high-quality statistics on the usual residents at an unusual time. Estimates reflect how the pandemic influenced where people lived on Census Day, especially in urban areas. For some, these changes were temporary; for others, they were longer lasting.

Users should consider possible pandemic effects when using local estimates, particularly for urban areas or those with high population mobility, and factor these into any decisions made based on these data.

Students

Main estimates and possible impacts

The pandemic especially affected student populations, with many not living at their term-time addresses and some international students studying remotely.

We enhanced guidance, communication and data processing to ensure accurate enumeration of students, as outlined in our Students: Census 2021 article. We also transferred responses for students who completed the census at their home address but not their term-time address. We did this by checking for any non-responses at term-time addresses, which were recorded on home census forms. We identified over 90,000 students who would otherwise have been incorrectly enumerated at their non-term-time address.

There was a fall in the number and proportion of international students between the 2011 Census and Census 2021, according to our The international student population in England and Wales: Census 2021 article.

Other data sources

There was a marked increase in the number of students not living in term-time accommodation in 2020/2021 compared with previous years, according to the Higher Education Statistics Agency's (HESA's) The impact of the COVID-19 pandemic on 2020/2021 Student data report.

There were only 60% to 70% of University of Bristol students living in the city and some international students may never have come for the academic year, according to analysis by Bristol City Council and the University of Bristol. A student hall survey we conducted shortly after Census 2021 found that some halls were below full occupancy.

Enrolment numbers for international students increased over the decade to 2020/2021, in contrast to Census 2021 findings, according to HESA's Where do HE students come from data. However, it is possible many of these students were studying remotely from outside the UK because of travel restrictions during the pandemic.

There was a large increase in study visas for the year to March 2022. This reflects students starting or resuming a deferred course, or changing from distance to in-person learning after the pandemic-related restrictions on in-person attendance were lifted, according to data from the Home Office's Why do people come to the UK? To study release.

Implications for use

The census counts students at their term-time address. We are confident the estimates reflect the student population in March 2021. However, the pandemic affected where students were living and the size of the international student population in 2021. Caution should be taken when using these statistics to draw conclusions beyond the pandemic context.

International migration

Main estimates and possible impacts

Census 2021 gives a detailed picture of the 10 million non-UK born population, the majority of which (93%) arrived before pandemic-related travel restrictions. These restrictions from March 2020 led to a marked reduction in migration flows (people immigrating to and emigrating out of the UK). Census 2021 estimates showed fewer usual residents whose address one year before was outside the UK compared with 2011, according to our International migration, England and Wales: Census 2021 bulletin.

Wider changes to the immigration system, including those because of the Brexit transition period ending in December 2020, were also likely to have affected international migration patterns around Census Day. 

While immigration and emigration were affected during the pandemic, this only represents a small proportion of the total non-UK born population and non-UK passport holders. These estimates are therefore less affected by the pandemic.

Estimates of short-term residents were likely affected by the pandemic. Short-term residents are people who did not intend to reside in the UK for a total of 12 months or more. However, these estimates were not adjusted for non-response in 2021, so they are not comparable with 2011. This makes it difficult to assess the pandemic's impact on this group.

Other data sources

There was a reduction in long-term immigration between June 2020 and June 2021, in line with estimates from Census 2021, according to administrative data sources in our Long-term international migration, provisional: year ending December 2024 bulletin. There were also decreases in the number of visas granted and passenger arrivals in the year ending March 2021, according to the Home Office's Immigration statistics, year ending March 2021.

Non-UK born population level estimates in the year to June 2021 from our Annual Population Survey were broadly comparable, according to our Population of the UK by country of birth and nationality: year ending June 2021 bulletin. However, the overall lower estimate is attributed to survey quality issues.

Implications for use

Overall, we believe Census 2021 international migration estimates are an accurate reflection of the population at the time. Census 2021 provides a snapshot of the population in England and Wales at a time when travel patterns were heavily disrupted by the pandemic. Since then, rising immigration from non-EU nationals has further increased the size of the non-UK born population. This is an important consideration when using these estimates for decision making. Therefore, we advise users to avoid assuming that migration patterns seen in the 12 months to Census 2021 are representative of patterns seen in other years.

While estimates are comparable, our long-term international migration bulletins are still the official estimates for immigration, including for the Census 2021 period. Census 2021 remains our best estimate of the overall size and makeup of non-UK born and non-UK passport holders.

Economic activity status

Main estimates and possible impacts

The national lockdown and furlough schemes substantially affected the economic activity of England and Wales, as discussed in our Economic activity status, England and Wales: Census 2021 bulletin.

We gave extra guidance to help people on furlough answer the census questions about work, as described in our Labour market quality information for Census 2021 methodology. This guidance was written to reflect pandemic response requirements at the time, but was not tested in the same way as other guidance. Therefore, we cannot say how closely it was followed. Furloughed workers were advised to report themselves as employed and therefore, economically active. However, guidance on answering work-related questions may not have been consistently followed.

Other data sources

Analysis shows that Census 2021 estimated that there were one million fewer people aged 16 to 64 years in employment than the Labour Force Survey (LFS), according to our Comparing Census 2021 and LFS estimates of the labour market, England and Wales: 13 March 2023 article. This is likely because some furloughed individuals reported being out of work rather than employed. However, most respondents answered as expected, and other factors also contributed to the difference.

Implications for use

Because of the impact of the pandemic on the labour market topic, particularly around those on furlough, care should be taken when using Census 2021 data to draw conclusions about the labour market in other years. Official statistics on trends in employment patterns are published in our Labour market overview bulletins. We recognise the ongoing uncertainty in the estimates from the LFS.

Travel to work and workplace

Main estimates and possible impacts

Census 2021 data showed impacts on travel to work estimates, particularly an increase in homeworking, following government advice to work from home. We provided extra guidance to Census respondents affected by the pandemic, as described in our Travel to work quality information for Census 2021 methodology.

Respondents on furlough were asked to report how they usually travelled to work before their circumstances changed. Workplace address information was not collected for those working at home, including those following government advice. While many were still supported by government schemes, some may have reported travel based on their last time working or their behaviour on Census Day.

Other data sources

There were similar declines in travel to work estimates compared with pre-pandemic levels, which supported the Census findings, according to our Experimental travel to work data for 2021. These experimental models are based on assumptions, so they should not be used for decision making, as discussed in the Data Science Campus's Estimation of travel to work matrices blog post. However, these data help illustrate how the pandemic may have affected travel to work.

Longer term travel trends and insights into travel around the time of the Census 2021 are provided by the Department for Transport's (DFT's) Faster indicators of transport activity. Road traffic increased in 2021, compared with 2020, but remained below pre-pandemic levels, according to DFT's Road traffic estimates: Great Britain 2021 release (PDF, 1.1MB).

Implications for use

Census 2021 provides a snapshot during the pandemic. Given the impact of lockdown and furlough, Census 2021 results are unlikely to represent pre-pandemic or post-pandemic travel patterns or workplaces. Users should take care when using these data for planning and policy purposes.

Living arrangements

Main estimates and possible impacts

There were some effects on living arrangements because of the coronavirus (COVID-19) pandemic according to our People's living arrangements in England and Wales: Census 2021 article. For example, more adults were living with their parents and fewer households changed address in 2021, compared with 2011. The number of people with second addresses for work or study also declined, likely reflecting restrictions and remote working or learning. 

Other data sources

The increase in adults living with their parents may reflect the increasing average age at life milestones, such as moving out, according to Labour Force Survey (LFS) data in our Milestones: journeying into adulthood article. However, the pandemic may have contributed to some adults remaining or returning to a parental home. Further LFS data on young adults living with their parents is discussed in our Families and households in the UK: 2024 bulletin.

Implications for use

Overall, we believe Census 2021 estimates on living arrangements accurately reflect of the population at the time. However, some variables such as adults living with their parents, second addresses, and household mobility may have been affected by the pandemic. The potential for pandemic effects should be considered when using these variables in decision making.

Health

Main estimates and possible impacts

Self-reported health improved overall in 2021, compared with 2011, which was prompted by improvements among older people according to our General health by age, sex and deprivation, England and Wales: Census 2021 article. This may reflect pandemic-related factors, such as a more positive self-perception of health. Or excess deaths and COVID-19 related deaths of disabled people may have reduced the proportion of people who were likely to rate their health as "bad" or "very bad". Conversely, self-reported health among younger people declined, corresponding with increased reported disability.

Other data sources

There was an increase in self-reported health for adults in 2021, compared with other years, according to NHS England's Health Survey for England 2022 tables. However, data collection methods differed that year.

In contrast, there was a downward trend in the proportion of adults reporting good or very good health according to our Public opinions and social trends, Great Britain: personal well-being and loneliness dataset. As data collection began in March 2020, this reflects only a 12-month trend. Our Opinions and Lifestyle Survey (OPN) also showed an increase in depressive symptoms among young people, aligning with the decline in self-reported health in this age group in Census 2021 estimates.

Implications for use

The pandemic likely influenced how some people reported their health. Care should be taken when using census data to draw conclusions about changes in self-reported health. As evidenced by our OPN data, the decline among young people is more likely to represent a real-world trend. These factors should be considered when making decisions based on self-reported health data.

Unpaid carers

Main estimates and possible impacts

The level of unpaid care fell from 11.4% in 2011 to 9.0% in 2021 according to our Unpaid care, England and Wales: Census 2021 bulletin. Possible explanations for this include government guidance on reducing travel and household mixing and increased deaths in the older population. Changes in the wording of the question from the 2011 Census may also have affected how many people identified as carers.

Other data sources

Several other sources collect information on unpaid or informal care. The Department for Work and Pensions' Family Resources Survey also recorded a decline over the decade, while our Labour Force Survey showed levels were broadly stable. These sources are not directly comparable because of differences in question wording and target audience, but both highlight uncertainty in the data and the pandemic's impact.

Implications for use

Lockdown restrictions likely limited some people's ability to provide care, which may also explain why most of the decline was among those giving only a few hours. Carers UK advises caution when comparing census data with other sources, noting they are collected for different purposes using different questions. Users should consider multiple sources of data, particularly when assessing how unpaid care has changed since the pandemic.

Impact on UK statistics

Main estimates and possible impacts of the pandemic

While censuses have previously been on the same day across the UK, Scotland's census was delayed to 2022 because of the pandemic. We adjusted Scottish data to produce our UK-wide Census statistics: 2021. We published how we produced UK-level census-based statistics, their strengths, limitations and potential uses, and guidance on alternative approaches to combining and comparing data from the most recent UK censuses in our Quality and methods guide for census-based statistics UK: 2021.

Implications for use

Depending on their need for UK or Great Britain totals, or comparisons between countries, users may use the tables in our census-based statistics UK release. Alternatively, users may create their own aggregations of published census data as described, with awareness of the respective strengths and limitations.

We also recommend that users comparing local areas in Scotland with the rest of the UK refer to the Scotland's Census 2022 data, England and Wales Census 2021 data, and Northern Ireland Census 2021 data, respectively.

Other topics and characteristics

Main estimates and possible impacts

In addition to topics already discussed in this article, statistics and data were published on a range of characteristics and other topics from Census 2021, including:

• our Ethnic group, national identity, language, and religion releases

• our Welsh language, Wales: Census 2021 bulletin

• our Sexual orientation and gender identity releases

• our UK armed forces veterans releases

• our Education, England and Wales: Census 2021 bulletin

• our Housing: Census 2021 in England and Wales releases

The pandemic will have had little or no impact on these topics. There are a few exceptions - mainly when those topics that were affected by the pandemic are closely linked to these other topics. For example, the housing topic includes the number of bedrooms and central heating, which were unaffected by the pandemic. However, the number of second homes used by students did go down over the decade, which could be a result of pandemic restrictions.

Implications for use

Statistics for these characteristics and topics are unlikely to have pandemic effects that affect decision making, with minor caveats in areas with large student populations. There are other known limitations with Census 2021 data that are unrelated to the pandemic, notably concerns about the quality of gender identity estimates. Additionally, the social landscape has changed considerably since Census 2021, meaning users need to consider how well these topics still reflect current realities.

Mid-year population estimates

Main estimates and possible impacts

Census 2021 is the base for official England and Wales Mid-year population estimates (MYEs). They are produced by adjusting the previous year's population for births, deaths, and both international and internal migration, as described in our Population estimates for England and Wales mid-2024: methods guide. They are re-based every 10 years using census data. The 2021 MYEs used Census 2021 as the base.

We are confident that our methods ensure that our estimates do not have systemic pandemic effects in the years since 2021. However, though most restrictions had lifted by mid-2021, some population patterns may still have been unusual. We put additional processes in place to ensure that internal migration estimates between Census Day and the mid-year point were robust.

Urban local authorities and those with high student populations may have seen different patterns of internal and international migration since the pandemic than would usually be the case. There may be increased uncertainty in these estimates. Quality indicators published in our Population estimates: quality information dataset show where there is greater uncertainty for components of the mid-year estimates for some local authorities.

We are developing population statistics that make greater use of administrative data. Admin-based population estimates (ABPEs) use independent population stocks for each year and are described in our Understanding mid-year ABPEs for local authorities in England and Wales methodology. ABPEs use statistical modelling to reduce the increase in uncertainty that occurs as we move further away from a census year, as described in our Dynamic population model, improvements to data sources and methodology article.

Unlike the mid-year estimates, Census 2021 data are not used directly in the ABPEs. However, the 2021 MYEs, which are based on 2021 Census data, are used to adjust the initial population stock estimates for known coverage issues. The modelling approach provides estimates of uncertainty in the ABPEs, which are published as credible intervals. The latest estimates are available in our ABPEs for local authorities in England and Wales dataset. The ABPEs do not currently replace accredited official mid-year population estimates or international migration estimates and should not be used for policy or decision making.

Other data sources

There was variation between rolled forward 2011 estimates and Census 2021-based mid-year estimates across local authorities, as shown by our Reconciliation of mid-year population estimates with Census 2021 at local authority level article. There were no consistent patterns that indicated issues specific to any particular type of local authority. For instance, differences varied even among three local authorities with large student populations.

For mid-2024, ABPEs are very similar to our accredited official MYEs as shown by comparisons between the MYEs and ABPEs published in our latest Population estimates bulletin. The total ABPE for England and Wales was 0.1% lower than the official MYE. From mid-2022 to mid-2024, the accredited official MYE was within the credible intervals of the ABPE for all except two local authorities. We used case studies to analyse differences in the population estimates for particular types of local authority, such as those with large student or communal establishment populations, or more mobile populations.

Implications for use

Overall, we believe that mid-year population estimates based on Census 2021 data are robust and there are no systemic issues with their use because of impacts of the pandemic. Mid-year population estimates are subject to statistical uncertainty and published quality indicators should be used to inform decisions.

Population estimates for local authorities with more mobile populations or higher numbers of students may have a higher level of uncertainty. Data analysts and policy makers within local authorities will be the most familiar with data sources for their area, and the strengths and limitations of those data sources. These users should use official population estimates alongside local intelligence to inform decisions.