Using House of Commons speeches on welfare from the late 1980s to 2015, Tom O’Grady finds that declining support for the benefits system was a top-down phenomenon. Shifts in political rhetoric – especially from Labour – did not occur after public opinion changed, but took place slightly before the public was changing its mind about benefits.
As Theresa May’s government continues the rollout of Universal Credit, changes in the UK’s welfare system are in the spotlight. The latest reforms represent the culmination of several decades of ‘welfare to work’ initiatives. These have taken Britain from a relatively generous welfare system to perhaps the least generous set of programs in the developed world. Benefits have been cut at the same time as the screws have been tightened on users of the system, subjecting them to ever more stringent conditions in return for their benefits.
Nonetheless, the reforms have proven popular with the public and were seen as a clear vote-winner by the Conservative Party. In ongoing work, I am trying to understand why. I argue that this shift in public opinion occurred mainly in response to changes in the way that politicians have framed and discussed both the welfare system and its users over the past thirty years. As the public heard increasingly harsh rhetoric, their opinions altered.
Figure 1 illustrates declining public confidence in the benefits system since the 1980s and early 1990s. It shows two measures from the British Social Attitudes Survey as an example. In the early 1990s, around 25% of people felt that ‘benefits are too high and discourage work’, whereas 50-60% did so by the early 2010s. Likewise, only around 25-30% of people disagreed that ‘most benefits recipients don’t deserve any help’, down from close to 50% in the early ’90s.
Figure 1: Changes in Public Opinion Toward Welfare and Welfare Recipients, 1987-2015
Public dissatisfaction reached its peak in the late 2000s and early 2010s, just as David Cameron came to power and began the Conservatives’ program of welfare reforms. Since then, only a slight softening in opinion is detectable. In fact, the biggest declines in confidence in the welfare system occurred not during the post-2010 era of radical reforms, but during the New Labour era of the mid-1990s to mid-2000s, when more tentative reforms began. Those earlier opinion shifts left the country in a position where further radical changes to the benefits system post-2010 were likely to prove popular. Why did the public lose faith in the benefits system? Several seemingly common-sense explanations can be put to bed.
First, perhaps the performance of the welfare system really did deteriorate over the 1990s and 2000s, and the collapse in public confidence in the system merely reflected this. But it would be hard to argue that welfare-to-work policies under the Major, Blair, and Brown governments failed to reduce unemployment. By the mid-2000s, unemployment was at an historic low, and had declined even amongst hard-to-reach groups like lone parents and young people. Nor is there any evidence of large rises in benefit fraud, even though the public’s perception of fraud did go up significantly. Actual spending on benefits generally fell, too, but the public continued to want ever lower spending on them.
Second, this wasn’t merely the product of the booming late-1990s and 2000s economy lowering the public’s demand for government help. If that were the case, we would expect to see the declines in support going into reverse when the economy took a turn for the worse again. But the declines in support for benefits largely persisted during and after the financial crisis of 2008.
Figure 2: Support for Government Providing Assistance to the Unemployment by Year and Income Quartile.
Source: International Social Survey Program Role of Government Module. 1= poorest quartile, 4=richest quartile. Question asks: “do you think it should be or should not be the government’s responsibility to provide a decent standard of living for the unemployed?” Possible responses range from 1 (Definitely should not be) to 4 (Definitely Should be).
Third, plunging support for the benefits system was not driven by specific subgroups who did well over the period, like higher-income people or those from skilled occupations. Instead, the shifts in opinion were broad-based. For instance, support fell in the poorest income quartile almost as much as amongst the wealthiest (Figure 2).
Instead, declining support for the benefits system was more of a top-down phenomenon. Using a new database of all speeches on welfare from the House of Commons from the late 1980s to 2015, my work documents a dramatic shift in politicians’ discourse. Thus occurred amongst all parties, but was especially notable from Labour politicians. I use a range of techniques to mine this database and understand how representations of the benefits system and its recipients changed over the period.
In the 1980s and early 1990s, benefits recipients were depicted mostly as deserving, and the system itself was talked about as a highly legitimate means of poverty alleviation. That began to change from the mid-1990s. Users of the system became stigmatised, and benefits were depicted as ineffectual or even wasteful. This is summarised in Figure 3, which uses a technique known as topic modelling to break down parliamentary speech about the welfare state into sets of topics. It shows that from the late 1990s to the mid-2000s – in a very sharp break from the 1980s – Labour devoted substantially less time to talking about the benefits system and its users in positive terms than it did to problems with the system and the need for reforms. Positive mentions of welfare plummeted, and were drowned out by more negative discourse.
Figure 3: Labour’s Discourse on Welfare Issues, 1987-2015
Source: Structural Topic Model run on a corpus of all speeches about welfare state issues in House of Commons from 1987-2015. Figure shows speeches by Labour MPs only. ‘Positive Depictions’ includes speeches referencing need for traditional welfare provision, depicting recipients as deserving. ‘Negative Depictions’ includes speeches referencing need for reform, fraud, problems with welfare system and depictions of recipients as undeserving or fraudulent. The remaining (un-assigned) percentage refers to speeches on other topics, such as public housing and pensions.
Explaining why Labour changed its tune on the welfare system would require a whole new article. But for now, what matters is that the shifts in rhetoric did not occur after public opinion changed. Politicians were not responding to public opinion; they were leading it. Shifts in discourse actually took place slightly before the public was changing its mind about benefits. In fact, I argue that the large reversal in public support for benefits can only be explained as a reaction to the discourse of politicians, filtered through the media. Modern theories of public opinion emphasise that people react strongly to how issues are framed by political elites. Importantly, this is a phenomenon that could affect all voters, rich and poor alike.
I have uncovered several key facts about the British experience that bolster this idea of a top-down change in opinion. For one, young people changed their views more than older people. Younger cohorts who were socialised in an environment of much harsher criticism of the benefits system began their lives more opposed to benefits than previous cohorts did at the same age. Those altered views have tended to stay that way over the past 20 years. Labour supporters, too, were more likely than those of other parties to shift toward opposition to benefits, which is consistent with a ‘follow-my-leader’ dynamic.
In addition, I have carried out an experiment that further demonstrates the power of political discourse. While no experiment carried out on contemporary subjects can prove a theory about long-term opinion changes, it can offer circumstantial evidence in favour. I found that when representative samples of the British public are exposed to the same rhetoric that was used by politicians over the period, their opinions shifted in exactly the same manner as the real over-time shifts in opinion that occurred over the 1990s and 2000s.
Taken together, this work shows why such a regressive program of reforms has proven highly popular. The public is simply unused to hearing welfare and welfare recipients talked about in positive terms. That may be starting to change, with increasing recognition of problems in the rollout of Universal Credit and the election of Jeremy Corbyn. But if past experience is anything to go by, it will take a long time for the public to turn around their opinions. By then, British welfare provision will be almost unrecognisable compared to a generation ago.
Tom O’Grady is Lecturer in Quantitative Political Science at UCL.
All articles posted on this blog give the views of the author(s), and not the position of LSE British Politics and Policy, nor of the London School of Economics and Political Science. Featured image credit: Pixabay, Public Domain.