What official new income statistics can tell us about living standards in 2021 and beyond


The financial year from April 2021 to March 2022 feels a long time ago. The furlough scheme and £20 a week benefit boost were in place for the first half of that year, Rishi Sunak was Chancellor, inflation averaged a ‘mere’ 4 per cent, and, even after some notable increases, the typical annualised energy bill only reached £1,277, around half its current (capped) level of £2,500. 

We now (at last) have detailed official household income and deprivation statistics for that year, with the release of the ‘Households Below Average Income’ results from the Department for Work and Pensions (DWP). Below we take an initial look at what these statistics can tell us and put them in the context of what may have happened since. 


Income levels 

In these survey-based statistics, the real household disposable income of the typical person (equivalised and after housing costs) rose by 1.6 per cent – taking incomes back to where they were in 2019-20 after a fall in 2020-21. This is shown in the chart below. As discussed further below, difficulties in conducting household surveys in the pandemic mean that these figures are more uncertain than usual. But across multiple data sources (DWP, ONS and RF modelling) there is consistency in the fact that the typical real household income in 2021-22 was broadly the same as in 2019-20 (i.e. pre-pandemic). 

The pandemic period was of course painful in many ways. But its primary consequences were not seen in our disposable incomes overall. That makes it very different from the current crisis: real incomes are projected to have fallen in 2022-23 and to fall again in 2023-24. On current economic forecasts we may not get back to the typical income levels of 2021-22 until 2026-27 or later. 



The new data shows how government support managed to protect, and in some cases increase, the incomes of poorer households during the pandemic – and this is reflected in absolute poverty numbers. 17 per cent of people were living in absolute poverty (after housing costs) in 2021-22, compared to 18 per cent in 2019-20 and – as the Chancellor would note – 21 per cent in 2009-10. These statistics are more uncertain than they would usually be, but both the absolute and relative poverty results for 2021-22 are slightly better than expected. This reading may not fit with arguably the most striking stat in the latest – the biggest rise in relative poverty (after housing costs) since the 1980s. But this does come after an even bigger fall in relative poverty in 2020-21. Taking a more robust two-year view, poverty rates are flat or down. 

However, just as with median income, things seem likely to have deteriorated in the more recent past: we project that there has likely been a rise in absolute child poverty in 2022-23.  



All of these statistics come with significant caveats. On top of the usual noise in survey-based results, Covid-19 created additional difficulties, with the survey response rate dropping from 50 per cent in 2019-20 to just 26 per cent in 2021-22 (although even the calculation of response rates is complicated by the pandemic). But there are also longer-term problems. Benefit receipt continues to be systematically under-estimated, and while the new release notes that “We are seeking to transform the [Family Resources Survey] by increasing the use of administrative data [which] should help correct for issues like benefit under-reporting and improve the quality of our income estimates” it may be a year or more until we see these corrected results. 

It is helpful to cross-check results from HBAI with separate ONS data and with the Resolution Foundation’s purely model-based nowcasts. But the existence of separate DWP and ONS data (as well as the variety of income measures on offer) also risks confusion where these disagree. ONS stats imply that the typical income of the poorest fifth (before housing costs) fell by 6 per cent between 2019-20 and 2021-22, but DWP stats in March show a rise of 2 per cent. The real answer may well be somewhere in between – but it would be more useful if these two sets of household income data were combined into a single source with less uncertainty. 


Food insecurity 

One welcome development in the provision of official data has been on food insecurity (since 2019-20) and food bank use (for the first time). Food insecurity was, thankfully, less prevalent in 2020-21 and 2021-22 than in 2019-20, affecting 7 per cent of households. But it remains shocking. Around one in eight (12 per cent) single parents experienced very low food security in 2021-22 – the most severe category, meaning that “at times during the last 30 days, eating patterns of one or more household members were disrupted and food intake reduced because the household lacked money and other resources for food” – and the same proportion had used a food bank within the previous 12 months. The overlap between these metrics is also a useful corroboration. In both cases, one in six households receiving Universal Credit were in the same boat.

What is particularly concerning is that these statistics are from 2021-22, before the £20 a week benefit boost ended; before real incomes fell and energy bills doubled; and before food price inflation of 18 per cent per year. Other survey evidence suggests that food insecurity dramatically increased in 2022-23. 

The new income statistics provide more evidence that policy did a good job of broadly protecting living standards during the pandemic, including among low-income households. The long-lasting national hit from higher energy (and food) prices is not something that households can be entirely shielded from in quite the same way. But it’s clear that levels of poverty and hunger can and should be changed through policy choices.