Universal Credit
Inequality & poverty

Five takeaways from new living standards data


The annual release of DWP’s Households Below Average Income (HBAI) figures is far less timely than other economic indicators and, as it is based on survey data, it is noisy. Yet, together with the ONS’s separate income survey data (expected later in Spring), it provides key insights into how living standards have changed for different groups. So here are five initial takeaways from this important new data, which covers the financial year 2022-23: the height of the cost of living crisis. 

1. Real incomes have taken a hit

With inflation of around 10 per cent in 2022-23, it is no surprise that this data shows a fall in real household incomes. Focusing on non-pensioner incomes after housing costs (a key measure of living standards), the typical real income fell by around 1 per cent, or £300 in annual terms. This is broadly in line with, although slightly lower than, the fall in aggregate mean Real Household Disposable Income, which fell by around 2 per cent in 2022-23 (the worst single year fall on record). Putting this together with the impact of Covid-19, the big picture here is that real incomes are likely to have fallen over this Parliament, the first time this has happened in the past century. 

2. Absolute poverty rose in 2022-23

That living standards hit also increased absolute poverty (abetted by the end of the £20 a week working-age benefit boost which had existed through 2020-21 and part of 2021-22). The share of people living in absolute poverty (after housing costs) rose from 17 per cent to 18 per cent in 2022-23: a rise of 600,000 people. This may not seem like such a big jump, but absolute poverty usually falls – given that real incomes usually rise – so the rise of 0.8 percentage points is in fact the largest since 1982 (and absolute poverty rates were far higher back then: in proportional terms the 2022-23 change is the worst on record). Within this, absolute child poverty rose by 2 percentage points to 25 per cent (the worst increase since 1981), with an extra 300,000 children falling into poverty. 

Stepping back, absolute poverty levels are still low by historic standards, with the overall rate in 2022-23 being unchanged from 2019-20. The Government may still use its talking point that the number of people in absolute poverty has fallen since 2009-10 (though now only by 1.1 million rather than 1.7 million). But we can, and should, be making faster progress. For comparison, over the previous 13-year period, from 1996-97 to 2009-10, absolute poverty levels fell by 7.8 million. The next Parliament will need to make big strides if it is to avoid embarrassment on the country’s international commitment to at least halve UK poverty (relative to 2015) by 2030. 

The new statistics also show that children in larger families (those with three or more children) now make up a narrow majority of the total number of children in absolute, or indeed relative, poverty even though they are only a third of children overall. The ongoing roll-out of the two-child limit is projected to further increase poverty rates for this group. 

3. Material deprivation and food insecurity have risen

Perhaps more striking still are the figures for deprivation and food insecurity. The IFS has set out how the number of people in material deprivation rose by 3 million between 2019-20 and 2022-23, including an enormous rise in the number of people unable to adequately warm their home, from 4 per cent in 2019-20 to 11 per cent in 2022-23. (Real income trends may not fully reflect such cost of living pressures given inflation was higher for low-to-middle income households than the overall inflation measure suggests.) 

Food insecurity also jumped up in 2022-23, rising from 7 to 10 per cent of households – or from 5 to 7 million people. Shockingly, 41 per cent of households on Universal Credit, and 32 per cent of all single parents have low or very low food security. More recent surveys from the Food Foundation suggest there may have been some improvement since the peak of the crisis, but there deserves to be a far more active policy agenda to ensure that no lower income households are going hungry.  

4. Inequality has, if anything, fallen since 2019-20

HBAI data is a crucial source for measuring UK inequality. The new stats suggest a slight uptick in 2022-23, but we should not read too much into this given survey noise (the ONS’s upcoming survey data could well show the opposite result, for example). We can say with more confidence though that there was – if anything – a fall in inequality between 2019-20 and 2022-23, with incomes at the very bottom of the distribution being supported by some permanent Universal Credit increases and 2022’s significant but temporary interventions (the Council Tax Rebate, Energy Bill Support Scheme and Cost of Living Payments). 

5. Welcome work to improve all of this data needs to progress

Finally, we also now have the first experimental output from the ‘Family Resources Survey (FRS) Transformation project’, which seeks to improve all such income statistics by making use of administrative data. 

The DWP’s report shows that around £40 billion a year of benefit income does not appear in the survey data (as we set out in 2018). DWP has been working hard on fixing this by linking survey responses to real benefit data for each household so that survey errors – such as mis-estimating benefit amounts or forgetting about particular benefits – can be corrected. Doing so would add in around £20 billion of benefit income to 2022-23’s survey data (e.g. boosting by around £4 billion the reported incomes of pensioners who have under-estimated their monthly State Pension income). 

But what about the other £20 billion? Well, the new release reveals for the first time that there is also “an underrepresentation of benefit recipients in the FRS sample”. Survey data is supplemented by a process of weighting, so that the data is representative of the whole population in terms of specific numbers, such as the population by age group, or by region. But it now appears that even with these methods, benefit recipients are being under-represented. For example, after correcting any incorrect survey responses about benefit receipt, the 2022-23 data includes 3.8 million Universal Credit recipients; but this is 1.1 million fewer than the known total. Adding in a 0.4 million undercount of Housing Benefit recipients, the survey is under-representing benefit claimants by 1.5 million at the very least (while over-representing non-claimants). The DWP sets out how this can be addressed by ensuring that survey weights match not only population figures but also benefit caseloads and employment numbers. 

The chart below shows the impacts of these two big changes – correcting survey responses using administrative data and revising population weights – which together would essentially resolve the problem of the missing £40 billion. 

This is, however, just an experimental release. DWP has not set out how these changes would affect income, poverty, inequality and deprivation statistics. And it is not straightforward to guess at the impacts because, although correcting under-estimated benefit receipt will boost incomes, correcting the under-representation of the benefit-receiving population essentially means adding more poor households to the data.  

What is clear is that there are important statistical revisions to come. This work is very welcome, and future developments will also make use of administrative earnings data, but it is unfortunate that we are entering election debates and post-election policymaking with statistics which we know are wrong.