Accredited official statistics

Technical report: an assessment of the impact of the COVID-19 pandemic on the HBAI statistics for FYE 2022

Updated 24 August 2023

What you should know

This document is designed to help users of the HBAI statistics understand and interpret the effect the coronavirus (COVID-19) pandemic continued to have on the sample data and published estimates for financial year ending (FYE) 2022. It will revisit the main areas of concern highlighted in the FYE 2021 technical report and assess the extent to which similar issues are a feature of the data in FYE 2022. It should be considered alongside analysis and commentary of the statistics.

As with FYE 2021, collection of the FYE 2022 Family Resources Survey (FRS) data was affected by the coronavirus (COVID-19) pandemic. Although government restrictions introduced in response to the pandemic were significantly eased over the course of the survey year, the change in survey mode from established face-to-face interviews to telephone interviews in response to the pandemic continued for the duration of the 2021 to 2022 survey year.

Ordinarily, such changes would not be made without thorough testing to examine the impact on survey response rates, data collection and reporting. Unfortunately, this testing was not possible during FYE 2021, given the pace at which it was necessary to make changes to maintain data collection. When analysing the FYE 2021 data, we found these changes resulted in the FRS sample having more outright owners and fewer renters than in FYE 2020. There was also a skew toward older respondents (aged 65 or over), fewer households with children, and fewer respondents educated to below degree level. The Family Resources Survey was not the only long-standing survey to be impacted by the pandemic in this way. The Office for National Statistics compared several of their prominent survey data collections and found similar changes in the distribution of household characteristics – more owner occupiers, fewer respondents from areas of deprivation, as well as an increase in older respondents, and those belonging to white ethnic groups.

Last year, following assessment of data quality, we took the decision to issue only a limited number of tables covering our headline estimates. For FYE 2022, we have undertaken extensive analysis of our low income measures across several dimensions and are content that levels of bias in the data resulting from the mode change are lower than FYE 2021 and are having less influence on the statistics. We have therefore returned to publishing the full suite of our statistics both in our supplementary tables and via the Stat-Xplore tool. However, there remain areas where caution is advised when making comparisons with previous years and when interpreting larger changes. This report will provide more context on this to assist users when drawing conclusions on the data and the trends.

Summary of the coronavirus (COVID-19) pandemic related changes to the FYE 2022 HBAI statistics

Change	Detail	Description
Methodology	Change to FRS/HBAI grossing regime	To address additional biases in the raw sample, we retained additional grossing controls introduced in FYE 2021 to weight the sample by level of educational attainment. This boosted numbers with education levels below degree level in younger age groups and brought it more in line with expectations. The grossing regime in FYE 2022 has also been adapted to control for the differential level of response seen through the year. This was necessary due to the introduction of FRS sample boost in England and Wales in October 2021. A quarterly grossing control was introduced for Northern Ireland.
Methodology	Inclusion of new coronavirus (COVID-19) pandemic related income sources	Any income received through the Coronavirus Job Retention Scheme (CJRS) or the Self-Employment Income Support Scheme (SEISS) during FYE 2022 is treated the same way as in FYE 2021. Receipt of the first 3 SEISS grants was treated as taxable income when calculating profits in FYE 2021 tax returns. Therefore, money received from the scheme will have been automatically included in income estimates for self-employed persons who reported their FYE 2021 profit data. Both schemes were closed on 30 September 2021.
Publication of breakdowns of headline statistics	In this release we have returned to publishing all breakdowns of the headline FYE 2022 HBAI statistics in our supplementary tables and in Stat-Xplore.	FYE 2021 unpublished estimates are marked as an [x] in our publication tables and are excluded from Stat Xplore. The change in measures will be reported compared to FYE 2021 where available, but FYE 2020 where it is not (at the headline level, this only affects estimates of low income for individuals in disabled families)
Methodology	All 3-year rolling averages calculated and published for any period including FYE 2021 will use 2 data points only.	This change affects our estimates of: a) the numbers and proportions of individuals in low income by region and ethnic group, b) mean and median household incomes by region (tables 2_5ts to 2_8ts), c) quintile medians and population means for family type and economic status groups (table 2_3ts_BHC and AHC).
Presentational Changes	As with last year, recorded changes in material deprivation measures are reported as not being directly comparable with both FYE 2021 and the period prior to the pandemic.	Several of the questions asked as part of the material deprivation measure remain affected by government restrictions introduced in response to the coronavirus (COVID-19) pandemic, and the gradual reopening of society in FYE 2022.

These changes will be discussed in more detail below. The FRS Background Information and Methodology and HBAI Quality and Methodology Information Report should also be consulted for more details on changes to the grossing methodology and on the treatment of income sources introduced during the pandemic.

Publication of FYE 2021 estimates

In FYE 2021, we did as much as we could to address biases and make the weighted FRS sample as representative as possible. However, we were unable to validate several of the HBAI statistics we produce for subnational populations (for example, by region, ethnic grouping, and economic status).

In many instances, we were unable to identify how much the observed change in the estimates was influenced by the change in mode and/or change in methods of data collection. This process was made more difficult given that FYE 2021 included unprecedented levels of real change (e.g., changes in household circumstances resulting from economic changes or policies introduced during the coronavirus (COVID-19) pandemic). Statisticians in DWP took the decision not to publish any of the HBAI results below headline level, where we were satisfied the trends and results were more robust.

Following consideration, we have maintained our decision to not release the FYE 2021 data publicly. The HBAI dataset underpinning the headline estimates remains available for expert users and researchers in the UK Data Service, and we recommend consulting the FYE 2021 technical report for more guidance on use and interpretation of sub-national estimates from that year.

Where FYE 2021 data is not published, we make a direct comparison of changes compared to FYE 2020 in our publication and would advise users to do likewise, alongside commentary on longer-term trends. Where FYE 2021 is available, such as for headline measures, we make a direct comparison between FYE 2022 and FYE 2021, while also drawing on changes since the pre-pandemic period and the longer-term trends in the measures.

The main impacts of the coronavirus (COVID-19) pandemic on the FYE 2022 HBAI statistics

Impact on achieved sample size and response rates

The circumstances surrounding the coronavirus (COVID-19) pandemic resulted in a smaller achieved FRS sample size than pre-pandemic, with over 16,000 households in the FYE 2022 sample (down from around 20,000 in a usual year). This was an improvement on FYE 2021 where the achieved sample size was around 10,000 households.

Users should note that the FYE 2022 sample was unbalanced across the 2 halves of the survey year, with an achieved sample of 6,000 households for the period from April to September, and a 10,000 achieved sample for the period from October to March. This reflects the introduction of the pre-planned boost to the size of the FRS issued sample in England and Wales from 1st October 2021. We accounted for this step change by introducing a biannual grossing control for Great Britain so there were equal numbers of private households from each half of the survey year in the weighted sample. In Northern Ireland, changes to the approach of contacting respondents in July 2021 to include use of knock to nudge (KtN) meant that the achieved sample increased markedly partway through the year. This is not normally a feature of the FRS achieved sample, with response normally spread relatively equally over each twelve-month run of fieldwork. We introduced a quarterly household grossing control to balance their sample across the year.

Response rates for the survey were still below the levels achieved pre-pandemic. In the period to FYE 2020, average response rates across the UK were around 50%. The overall response rate for the FRS in FYE 2022 was 26%. In Great Britain knock-to-nudge (KtN) remained available throughout the year but was not as effective as Northern Ireland in eliciting additional responses. The response rate in Great Britain was more negatively affected by continuing challenging fieldwork conditions for both interviewers and potential respondents. No new initiatives were introduced in-year to further improve response rates. For the period October 2021 to March 2022, addresses sampled per Primary Sampling Unit (PSU) varied by region because the England and Wales boost was targeted at certain areas. Therefore, the impact of the boost on sample sizes was not uniform across regions. Further details on the sample design, response rates and fieldwork in FYE 2022 is provided in the FRS Background Information and Methodology report.

As the reliability of the results of sample surveys, including the Family Resources Survey, is positively related to the unweighted sample size, the smaller sample in FYE 2022 may have contributed to some of the differences observed this year. The precision of any estimates from sample surveys is provided through the availability of confidence intervals, which are used to assess whether apparent differences between 2 figures are statistically significant or not. This year, the width of the confidence intervals around our estimates are wider than in FYE 2020 but are not as wide as FYE 2021. We have again published details of the confidence intervals around our main estimates in our accompanying uncertainty tables. We would encourage users to refer to these when interpreting any changes compared to FYE 2021, where data is available to do so.

Impact on sample composition

This year, we were encouraged to see that there were fewer differences observed in the FYE 2022 sample composition compared to the pre-pandemic period. However, there remain changes in composition of the raw sample which suggests some continuation of a mode effect, albeit a less pronounced one.

As we highlighted in last year’s report, it is not possible to control for all observed bias in the sample, and some of the observed bias relates to unobservable or unmeasured attributes. The move to telephone interviewing is likely to have changed the profile of non-responders to the FRS. Those who provided their telephone contact details may have had certain attributes (higher civic engagement or more available time) which we cannot directly adjust for when weighting the sample.

An assessment of the main changes we have identified in the composition of the raw FYE 2022 sample are detailed below.

Age

Figure 1 compares the age profile of the FRS raw sample for the last 3 survey years. This shows there were proportionately fewer pensioners sampled (particularly aged 70-74) and a higher proportion of younger participants (particularly in the 25 to 40-year-old age bracket) between FYE 2021 and FYE 2022. However, pensioners as a group were still overrepresented in the sample compared to FYE 2020 and the representation of younger adults (aged 16-24) remains below pre-pandemic levels.

Figure 1: Age profile of the achieved FRS sample by 5 year age band, FYE 2020 to FYE 2022

Housing tenure

Figure 2 details the tenure split of households in the survey over the last 3 survey years. In FYE 2021 there was a notable increase in the percentage of owner-occupied households, and a corresponding decrease in renters. This year, the achieved sample was more representative, but as with the age profile, has still not returned to pre-pandemic proportions.

Figure 2: Tenure split of households in the achieved FRS sample, FYE 2020 to FYE 2022

The FRS and HBAI weighting regimes bring the age and tenure profile in line with the UK population, so at a grossed level the proportion of households who rent or own their home are line with population estimates. However, the concern in FYE 2021 was that the characteristics of those people sampled within each group were different to those sampled in previous years - the non-response bias had changed.

As part of quality assuring this year’s HBAI outputs we looked closely at the sample across dimensions we do not control or adjust for, to assess how it compared to the period prior to FYE 2021. We identified several areas where the sample had changed compared to before the pandemic, and below we outline the extent to which we believe these changes are affecting the FYE 2022 estimates.

Education

Last year, closer examination of the initial weighted sample revealed it contained a disproportionate number of working-age respondents who had been educated to at least degree level. It was important to adjust for this bias because income levels are strongly correlated with the level of education achieved. Therefore, additional grossing controls were introduced to rebalance the educational levels of those in the sample.

When preparing the FYE 2022 estimates, we again found that the weighted sample included a disproportionate number of respondents with education levels at or above degree level, and too few below degree level. Therefore, we maintained the use of educational grossing controls using annual growth in the degree population measured in the Annual Population Survey (APS). Figure 3 breaks down the proportion of working-age adults in the grossed FRS population by their level of educational attainment, before and after the addition of the educational grossing control.

Figure 3: Educational attainment levels in the FRS population, FYE 2019 to FYE 2022

We found that the inclusion of the grossing controls had less effect on the HBAI headline estimates than in FYE 2021. Introduction of the controls reduced median income by £1.30 per week in FYE 2022 compared with £8.10 per week in FYE 2021. The split in the non-degree population between those with no qualifications and those with qualifications below degree level was also closer to pre-pandemic levels than in FYE 2021.

More information on the educational grossing changes introduced since FYE 2021 can be found in the HBAI Quality and Methodology Information Report.

Disability

Disabled people are identified in the FRS as those who report any physical or mental health condition(s) or illness(es) that last or are expected to last 12 months or more, and which limit their ability to carry out day-to-day activities a little, or a lot.

In FYE 2022, disability prevalence in the FRS population has recorded further growth. Compared to FYE 2020, there has been a 3 percentage point increase in the proportion of the working-age and child population who report a disability, rising to 22% and 11% respectively. Over the same period, there has been a 2 percentage point reduction in the proportion of pensioners reporting a disability, but the proportion remains consistent with the 5-year survey period up to and including FYE 2019. Further contextual information on disability in the UK population can be found in the FRS report.

Last year, we took the decision not to publish our estimates of the numbers and proportions of individuals living in a disabled family who are in a low-income household. This reflected concerns we had about the representativeness of the disabled sample in terms of the types of impairments reported. Our assessment was that while some of the change in types of disability reported may have been genuine, a significant portion of it could have been due to sample bias. This was because we were unable to explain some of the changes in the grossed sample with reference to changes in the real world. For example, there were notable decreases in numbers reporting other types of impairment such as in hearing, memory, or vision, compared with previous years. This was likely a consequence of the change in mode rather than a real-world reduction in prevalence. It was not possible to further adapt the FRS grossing regime to adjust for this observable bias.

Changes in the prevalence of disability, or biases towards certain types of impairment, would not have been of concern had the incomes of disabled people who responded to the survey in FYE 2021 compared well to those who responded in FYE 2020. If the reductions in low-income rates matched non-disabled people that would also have provided some reassurance. However, those reporting memory, hearing and visual impairments in previous years were more likely to be in low-income households. Underrepresentation of this group could artificially reduce the overall numbers of disabled families in low income. In addition, the low-income rates for individuals with certain impairments were significantly lower than in previous years, at a scale that was difficult to validate. We were keen to assess whether this problem had persisted when preparing our FYE 2022 estimates.

Figure 4 provides a comparison of reported impairments for disabled individuals in the grossed FRS sample in each of the last 4 survey years. Please note that percentages do not sum to 100% because individuals are able to report multiple impairments.

Figure 4: Percentage of disabled people reporting each impairment type in the FRS, FYE 2019 to FYE 2022

There is still evidence of a mode effect in the FYE 2022 disability sample, with visual, hearing, and memory impairments again under-represented compared to before the coronavirus (COVID-19) pandemic. This largely affects the pensioner disability sample. However, we are reassured that growth in other impairment types largely corroborates external evidence on changes in the composition of the disabled population since the pandemic, for example see this publication on the employment of disabled people. Among the working-age population, the FRS has recorded more growth in reported mental health impairments since the pandemic, and in the prevalence of social or behavioural impairments amongst children.

Returning to the low income estimates, we assessed the extent to which the FYE 2022 changes in low-income rates for individuals in disabled households are being influenced by changes in impairment and whether new flows into disability might have different characteristics to the stock of disabled people prior to the pandemic.

Figure 5: Percentage of working-age disabled people reporting each impairment who are in low income, FYE 2019 to FYE 2022

Figure 5 gives a comparison of changes in the percentage of working-age disabled people in low income by impairment type since 2018/19 (note that the unpublished FYE 2021 estimates are not included in this chart due to the data quality issues summarised above). We were reassured that the low-income rates for each reported impairment type in FYE 2022 did not appear to be outliers, matching the FYE 2019 rates well. There is some evidence that the sample of those with hearing impairments is different to the years before the pandemic, reflecting a possible mode effect. Changes in low income rates measured since FYE 2020 do not seem to be disproportionately influenced by shifts in certain types of impairment.

Figure 6: Percentage of disabled pensioners reporting each impairment who are in low income, FYE 2019 to FYE 2022

Figure 6 presents the same chart for pensioners. While there is more variation in the low-income rates for pensioner impairment types, the FYE 2022 data is broadly in line with previous time periods. As with the working-age population, there is some evidence of a modest mode effect (most obviously for those who report visual impairments).

For researchers and users interpreting the disability estimates for FYE 2022, our advice is that we are content the estimates are reliable and comparable in the context of the broader time series. However, it should be noted that there is evidence that the change in mode is having a modest effect on the low-income estimates for less than 10% of the disabled population who report visual or hearing impairments.

Users should note that the last published estimates for FYE 2020 measured a statistically significant increase in the proportion of individuals in a disabled family who were in a low-income household, before housing costs, compared to FYE 2019. This year’s estimates are a statistically significant reduction compared to FYE 2020 and bring the estimates back into line with the longer-term trend.

Benefit receipt

The FRS asks respondents about receipt of the full range of state benefits, and amounts received, including Universal Credit (UC). There is a known undercount for several of these benefits, which means that when grossed to population totals, the FRS still estimates that there are fewer receiving the benefits than is reported by administrative data. This difference can be due to 2 factors: underreporting by the respondent, and sample bias which means the FRS is not capturing those individuals and they remain underrepresented in the sample. Further information on the level of undercount can be found in FRS Methodology Table M6a on the FRS homepage.

In FYE 2021, we found that the HBAI estimates captured the growth in the Universal Credit caseload seen over the whole of FYE 2021, but there was variation in the degree of undercount across the year. We believed this was likely related to changes in fieldwork introduced throughout the year to boost FRS response rates, with certain approaches being more likely to elicit responses from Universal Credit claimants. This had the most effect on our low-income rates for children.

This year, the level of undercount of Universal Credit in the grossed FRS population is stable across the year. There has been an improvement in the undercount compared to previous time periods, with the FYE 2022 FRS population ‘missing’ around 28% of Universal Credit claimants compared to 35% in FYE 2021 and 32% in FYE 2020. Data matching of the FRS with Universal Credit administrative data confirmed that this reflected better reporting of the benefit by respondents rather than marking a change in the representativeness of the sample (for example, change in mode had led to more Universal Credit claimants responding to the survey). This shift made intuitive sense given that the population are now more familiar with Universal Credit, particularly post-pandemic.

Ethnicity

In FYE 2021, several non-white ethnic groups (notably Asian – Indian, Pakistani, and Chinese) were underrepresented in the grossed FRS population of individuals. This reflected the fact that a smaller proportion of non-white households were sampled, which cannot be adjusted for by our grossing. The HBAI estimates of low-income rates by ethnic group were subject to much smaller sample sizes and showed higher variation than in previous years, even after smoothing the estimates based on a rolling 3-year average.

Figure 7 shows that the FYE 2022 FRS grossed ethnicity population is much closer to levels measured before the pandemic (although the under-representation in the Asian – Bangladeshi population should be noted). In addition, measured low-income rates are comparable with previous time periods. In the FYE 2022 publication all estimates we provide will exclude the FYE 2021 data from the calculation of the rolling 3-year average, and we would advise users of the data to adopt the same approach. More information on this is provided in the ‘Impact on Headline Statistics’ section below.

Figure 7: FRS population by ethnic group, FYE 2019 to FYE 2022

Employment and Self Employment

During FYE 2022, the Labour Force Survey reported an increase in the official employment rate compared to FYE 2021 which included a further reduction in the number of self-employed workers compared to the pre-pandemic period.

The FRS asks questions on employment status, so any movement in employment levels seen during FYE 2022 should be reflected in HBAI estimates. We have compared levels and rates of employment and self-employment reported in the FRS with those reported by the Labour Force Survey (LFS) for FYE 2022 and found a similar level of comparability as seen in the years prior to the pandemic.

Users may wish to note that the change in self-employed numbers between FYE 2021 and FYE 2022 in the LFS and in the FRS is different. Where the LFS saw a continued reduction in the number of self-employed workers in FYE 2022, the FRS measured an increase of 300,000 since FYE 2021. This reflects significant under-representation of the self-employed population in the FYE 2021 FRS, leading to the estimated reduction in self-employment being much greater than in the LFS. There are differences in the measurement of self-employment between the 2 data sources which may have contributed to the divergence. However, the change in self-employment since the start of the pandemic is comparable across both sources. The Labour Force Survey (LFS) is considered the definitive source where numbers participating in the labour market are concerned.

Household Composition

Last year’s technical report highlighted that the FRS measured several changes in household composition and formation in the population during FYE 2021. We judged that some of these changes may have been a genuine response to the coronavirus (COVID-19) pandemic. For example, there was an increase in multi-generational households (as adult children moved in with their parents or parents moved in with their children) in FYE 2021, compared with previous years.

Analysis of FYE 2022 FRS data shows that the percentage of adults living with cohabiting adults, the percentage of adults living with unrelated (not cohabiting) people, and the number of households containing multiple generations is similar to FYE 2021. Although the household composition of some more specific groups (for example, childless single individuals) has shown a larger shift back towards the pre-pandemic picture, overall household composition has largely remained as it was, rather than reverting to a pre-pandemic situation.

Family Type

There remains evidence that the FYE 2022 FRS raw sample includes higher proportions of some family types than periods prior to the pandemic, for example, more pensioner couples and fewer working-age couples with children. There are also fewer households sampled with 3 or more benefit units living in the same household. As this is largely corrected for by the FRS grossing regime, we are confident it is having minimal impact on the headline estimates. However, it does mean certain under-represented households sampled are weighted more highly than in previous years. For example, compared to pre-pandemic, the grossed FRS population has a higher proportion of benefit units with 3 or more children, and fewer with only 1 child. There are also fewer benefit units where the age of the youngest child is between 0 and 4 years.

The FYE 2022 estimates have also recorded a 300,000 jump in the population of children in lone parent families since FYE 2020. Estimates of the number of lone parent families in the population are provided from the Labour Force Survey (LFS) and directly used to weight the HBAI sample. Further data points are needed to assess the degree to which this reflects a genuine increase, is due to LFS sample variation, or is a consequence of recent changes to weighting introduced within the LFS in response to sample under-representation during the pandemic. Our assessment is that this jump has had a negligible impact on low-income rates for children.

Pensioners

The achieved FRS sample for the last 2 survey years has contained a higher proportion of pensioners, particularly pensioner couples, compared to the pre-pandemic period.

When preparing the FYE 2021 estimates, we found some evidence that the pensioners sampled were more affluent compared to previous time periods – more likely to be owner occupiers, more likely to receive an occupational or personal pension, more likely to have income from investments, and on average to have access to higher levels of savings. Figure 8 compares receipt of occupational or personal pensions and value of savings in the pensioner population over the last 4 survey years. This shows that the FYE 2022 population is more closely aligned with the pre-pandemic period on these metrics, compared to FYE 2021.

Figure 8: Level of savings and occupational or personal pensions receipt in FYE 2019 to FYE 2022

The FYE 2022 pensioner low-income estimates have shown larger increases across all measures compared to FYE 2021. While some of this increase is real, reflecting the context of below-inflation uprating of the State Pension in FYE 2022, the degree of change is also likely to have been somewhat affected by changes in pensioner sample composition between the 2 survey years. This broader context should be borne in mind when interpreting the observed changes in pensioner low income rates.

Impact on headline statistics

Low income rates by UK region and ethnicity

The HBAI publication estimates the number and proportion of people in low income households for each UK region and ethnic grouping using a 3-year rolling average. This is an established method used to smooth out observed variation in single-year estimates for these groups, which are subject to smaller sample sizes.

From FYE 2022, all 3-year rolling averages calculated and published for any period including FYE 2021 will use 2 data points only. This follows our decision to not publish breakdowns of the headline FYE 2021 estimates. This approach will also be adopted for several other publications where use is made of the HBAI regional estimates to inform their statistics. These are Children in Low Income Families (CILIF) local area statistics, and statistics on incomes in the devolved administrations, published by Scotland, Wales and Northern Ireland.

Change in material deprivation estimates

In the HBAI publication, we publish estimates of the percentage of benefit units with children who are materially deprived, including in combination with a series of low income thresholds, and an estimate of the percentage of pensioner households who are materially deprived. For the first time this year, we have also published indicators of combined low income and material deprivation for working-age adults.

We measure material deprivation by asking respondents a series of questions about their household circumstances and access to a range of goods and services, which, if lacked, contribute towards a deprivation score for that household. Those households who score above a set threshold are materially deprived. Please see the HBAI Quality and Methodology Information Report for detailed information on the approach.

In FYE 2021, several of the questions asked as part of the measure were affected by government restrictions introduced in response to the coronavirus (COVID-19) pandemic. This meant that it was not possible for those sampled to access several social opportunities or services during periods of lockdown, regardless of deprivation or financial constraint. These included opportunities such as school trips, socialising with friends or family, attending organised activities or pursuing hobbies, going on holiday, and getting a haircut. Some of those in the sample may have responded to these questions with their ordinary circumstances in mind. Others may have responded according to their actual (lockdown affected) circumstances.

For FYE 2022, although the impact on survey responses was less marked than last year, restrictions remained in place throughout the first quarter of the survey year and the effect the coronavirus (COVID-19) pandemic had on social interactions continued to unwind during the remainder of the survey year as restrictions were removed and society began returning to normal. Therefore, for both FYE 2021 and FYE 2022, all estimates of material deprivation, including those combined with low income measures, are not strictly comparable with the pre-pandemic period. Changes in recorded material deprivation for FYE 2022 may not fully reflect the real change in household circumstances compared to FYE 2021 or the pre-pandemic period. Further details are provided below.

Child and adult material deprivation

In the methodology, any person who responds with “does not apply” is treated as not being deprived of the item in question. In Figure 9, below, we have provided information on the proportion of ‘does not apply’ responses to all coronavirus (COVID-19) pandemic affected questions asked of the child compared to previous years.

Figure 9: Proportion of ‘does not want/need/does not apply’ responses to coronavirus (COVID-19) pandemic affected questions in the child material deprivation measure, FYE 2019 to FYE 2022

Compared to FYE 2021, fewer respondents used the ‘not applicable’ category when answering the questions, but levels have not returned to those in the 2 survey years prior to the pandemic. Closer examination of the data revealed that, particularly for the school trip and organised activity questions, responses to the question changed across the year as coronavirus (COVID-19) pandemic restrictions eased and society began to return to normal. By the end of March 2022, only 7% of respondents selected this category with regards to the question on school trips, compared with around 18% across the first quarter of the survey year.

It is difficult to quantify the extent to which the changes we have recorded in material deprivation since FYE 2020 reflect lockdown constraints and how much are genuine changes, or whether different responses to the questions would have resulted in higher levels of material deprivation. Some households will have already accumulated enough points from other questions to be considered materially deprived. Others will have remained below the threshold even if they would ordinarily have answered ‘yes’ to a small number of the questions. The likelihood of a household accumulating enough points to cross the material deprivation threshold depends not just on their responses to individual questions, but groups of questions, and the importance each question makes to the overall deprivation score is also weighted according to the prevalence of access to that item in the population.

Pensioner material deprivation

Unlike the child deprivation measure, it was not possible for pensioner households to initially respond ‘does not apply’ to the questions. Instead, if the pandemic limited their ability to experience an opportunity or access a service, they would answer “no – I cannot access this” and provide more detail on the reason why from a predefined list including health and financial reasons. In Figure 10, below, we have provided the proportion of ‘no’ responses to all coronavirus (COVID-19) pandemic affected questions asked of the pensioner compared to previous years.

Figure 10: Proportion of ‘no’ responses to coronavirus (COVID-19) pandemic affected questions in the pensioner material deprivation measure, FYE 2019 to FYE 2022

Compared to FYE 2021, fewer respondents used the ‘no’ category when answering the questions, but levels have not returned to those in the 2 survey years prior to the pandemic. Once again, it is difficult to quantify what the responses to the affected questions would have been in ‘normal’ circumstances, and how that would have affected the recorded levels of pensioner material deprivation in the FYE 2022 publication. It can be noted, however, that for most of the questions we judged to be unaffected by the pandemic - e.g., ability to replace the cooker, pay bills, or access a telephone - higher levels of deprivation or more ‘no’ or ‘can’t afford’ responses were recorded compared to FYE 2021, but still below those recorded pre-pandemic.