Talking the politics of welfare

Liam Stanley - Associate Fellow, SPERI, & Lecturer in Politics, University of Sheffield

Given this large gap, it is perhaps unsurprising to see that public attitudes towards unemployment benefit have significant hardened in recent times.  The British Social Attitudes survey routinely asks people whether they believe that unemployment benefit is either too low and causes hardship or too high and discourages work.  As the graph below shows, views on this have shifted from a majority answering ‘too low, cause hardship’ and towards the answer ‘too high, discourage work’.

Unsurprisingly then perhaps, specifically speaking of ‘welfare’ can have negative connotations.  A classic study by Tom W. Smith demonstrated how survey participants provided systematically different answers when asked a variety of questions about whether spending should be increased on ‘welfare’ compared to ‘assistance for the poor’ and ‘caring for the poor’.  Smith found that asking about ‘welfare’ produced consistently ‘much more negative evaluations’ than ‘the poor’.

This is particularly relevant for another area in which both the previous Coalition and now Conservative governments have started using the word welfare: namely, in ‘annual tax summaries’, the individualised documents that HM Revenue and Customs (HMRC) has started sending to every taxpayer detailing how the state has spent their revenues over the last year.  The seemingly innocuous documents stirred controversy when they were launched, with the TUC declaring them ‘propaganda’. At the centre of the controversy was the way in which ‘welfare’ – which ended up as the largest category – was calculated.

Normally, budget information follows the official Public Expenditure Statistical Analysis (PESA) – which in turn is structured around the United Nations’ Classification of the Function of Government (COFOG).  Welfare is not mentioned in either of these systems.  They instead both write of the more neutral ‘social protection’.  For the purposes of the tax summary, the PESA category of ‘social protection’ was split up into ‘pensions’ and ‘welfare’.  While the ‘pensions’ category was relatively straightforward, the resulting ‘welfare’ category included a number of areas – such as military pensions or long-term social care – that are not normally associated with welfare.

The government justified this change on the basis that the ‘difficult statistical language’ of PESA can be problematic to understand.  The Institute for Fiscal Studies (IFS), however, politely disagreed, and argued that the welfare category included ‘spending that would normally be classed as “welfare”’.  The IFS then recommended an alternative tax summary (pictured below) that remains consistent with PESA, but disaggregates ‘social protection’ into four smaller categories.

In a recent study I conducted with my colleague Todd Hartman, we wondered whether it would make a difference whether people received either the HMRC tax summary with ‘welfare’ at the top or the IFS alternative that split social protection up. Our recent SPERI brief outlined how those who received the HMRC version are: less likely to agree with how the Government spent their money; less likely to indicate that current Government spending is a good use of taxpayer money; and likely to guess that the Government spends more on welfare.

So, it seems that speaking of ‘welfare’ over, say, ‘social protection’ or ‘social security’ does matter, and maybe matters very much.  Two questions arise from this. First, when and why did the government – and society – start talking about welfare instead of national insurance and benefits?  In regards to the ‘when’ aspect, Google N-gram – which allows you to examine how often a word or phrase is used within a particular corpus of books (in this case ‘British English’) – provides some clues. Mentions of national insurance for many years outweighed welfare benefits, but this trend reversed during the 1980s.  When else would we expect?

The ‘why’ aspect is unavoidably a political issue that merits further discussion.  But, for the moment, whether ‘welfare’ or ‘national insurance’ is the right or wrong label, the difference clearly highlights that there is something at stake in the word welfare that goes beyond just semantics.  So, next time you are coining your next piece of social protection legislation, it might be worth taking a moment to reflect on the choice of words and the impact they may have.