There is a significant difference in opinion on Brexit between different age groups in the UK, with older citizens generally exhibiting more negative attitudes toward the EU than younger ones. But as Kieran Devine writes, while over 65s are typically treated as a single category in opinion polls, there are substantial generational differences within this group, with those who lived through the Second World War being far more likely to oppose Brexit.
The EU was set up in response to the horrors and destruction of the Second World War. In the wake of the Brexit referendum result, it was oft repeated that the older generations were more likely to have voted for Britain to leave the European Union. This presents something of a puzzle; why would older generations, likely to have experienced the impact of the war first-hand, seek to remove Britain from an institution that has helped maintain peace in Europe for more than seven decades? Might it be that the ‘over 65s’ category, containing individuals several decades apart in age, conceals distinct generational differences amongst this group?
To address these questions, I have conducted an Age-Period-Cohort (APC) analysis using Eurobarometer survey data. APC analysis is used to ascertain if distinct generational effects are present in public opinion. Generational effects refer to the influence of environmental factors during each generation’s formative period (approximately between ages 15-25) on their long term political opinions, such as the prevailing social attitudes of the time or the occurrence of important political events. The Second World War is undoubtedly just such an event that may have deeply influenced the opinions of the generation that came of age during wartime.
This APC analysis used a longitudinal dataset of Eurobarometer surveys covering the years 1970-2017. These biannual surveys, consisting of 1,000 face-to-face interviews with members of the British public, ask respondents a range of questions, including their opinions on European integration. When controlling for a range of factors that have been identified as influencing attitudes towards integration – education, occupation, left-right position, gender, urbanisation – generational effects are confirmed in the data.
Specifically, when defining a ‘war generation’ that experienced the majority of their formative period during the Second World War, as well as a number of other more recent generations, this war generation is revealed as displaying significantly more positive views towards European integration than the immediate post-war generations. In fact, the size of this generational effect between the war and post-war generations is approximately equivalent to the same change in attitude that would be expected from a two-year reduction in education levels, a factor well known to increase Euroscepticism.
Holding all values at their mean in the data, Figure 1 illustrates the curvilinear differences in generational attitudes towards the EU in the UK. These results reflect respondents’ answers to questions regarding how positively they view the image of the EU, and whether they rate the UK’s membership of the EU as a good or a bad thing. In these illustrations, higher values connote more negative attitudes towards European integration. They therefore show that, all else being equal, the war generation have more positive attitudes towards the EU than the immediately following generations. Indeed, only the most recent generation, the millennial generation, display more positive attitudes towards the EU than the war generation.
Figure 1: Image of the EU among different generations in the UK
Note: For both charts, a higher value indicates a more negative image of the EU.
One explanation for these results is that the war generation give a premium to the pacific benefits of European institutions. Having experienced first-hand the horrors of war, they place a high value on the founding principles of unity that the EU promotes. The most recent generations also view integration more positively, given that these individuals have grown up with the UK’s membership of the EU as the norm. The concept of not being a part of Europe – with its visible signifiers of flags, anthems and institutions – is likely to be discordant to those from the millennial generation. Conversely, the post-war and 60/70s generations in the UK have neither the memories of wartime nor the routinised experiences of EU membership during their formative years. They therefore display the most hostile attitudes towards integration.
Additional variables, not captured in the Eurobarometer Survey data, appear poor alternative explanations for these findings. Reduced urbanisation, increased Protestantism, changes in party affiliation, the spread of communication technologies, and media frames have all been shown in previous research to affect attitudes towards European integration. This previous research, however, would suggest that the historical trends in these factors should lead to a reduction in hostility towards integration from the war to post-war generations. They are not, therefore, likely to be explaining the observed effects.
Including individuals’ responses to other questions in the Eurobarometer survey can more clearly outline the explanatory logics underpinning these results. Mediation analysis was performed on the answers to the question “What does the EU mean to you personally?” in the analysed models. Crucially, one of the 14 multiple choice answers to this question was “Peace”. If the observed effects were operating through the proposed ‘peace hypothesis’, accounting for those answering peace to this question should reduce the observed generational effects.
The results of this analysis confirm that this is the case: the older generations are more likely to associate the EU with bringing peace, and when mediating for these attitudes, the generational effects in the initial models are reduced by around 20%. The war generation are more likely to associate the EU with peace, and thus have more positive attitudes towards integration.
However, this analysis also reveals additional elements that are driving the cohort effects between the war and the following generations. Indeed, the post-war generation are in fact more likely to associate the EU with bringing peace than their younger counterparts, and yet they display more negative overall attitudes towards integration.
Conducting additional mediation analysis on alternative answers to the question of “EU meaning” reveals the importance of issues surrounding immigration and sovereignty to the post-war and 60s/70s generations. Almost 22% of the cohort effects are described by the post-war generation as being far more likely to associate the EU with a loss of control of the nations’ borders and a threat to national identity. Similarly, 45% is mediated by concerns surrounding a lack of the effective democratic functioning of the EU, something that has been linked in previous research to notions of an ‘anti-democratic’ EU eroding British sovereignty. It is thus not simply the scarring effects of wartime explaining the war generation’s positivity towards integration, but rather that the two following generations are also particularly hostile to European institutions.
Explanations for these results can be found in British history; the post-war and 60s/70s generations were the first to confront the fall of empire during their formative years, as well as the first mass immigration from the Commonwealth. This fuelled insecurities over British identity, coming to the fore in such instances as Enoch Powell’s “Rivers of Blood Speech” and the Immigration Acts of the 1960s. These results therefore support the notion that it is during times when identities are threatened that they become mobilised as points of political salience, and that these heightened political environments can shape individuals’ opinions long into the future.
What do these results suggest for future UK public opinion on European integration? Firstly, extolling the virtues of the pacific benefits of the EU are likely to fall on ever more deaf ears – the generation with which this argument will have its greatest impact are severely dwindling in number. Secondly, the degree to which individuals feel their British identity is threatened may govern their level of positivity towards European institutions. Thirdly, the prevailing political environment shapes the long-term opinions of those in their formative years. Given the current ubiquity of the Brexit debate, today’s arguments and events surrounding integration will almost certainly have a significant impact on the most recent generation, namely those born after the millennium. In exactly what way these debates will shape public opinion, however, remains to be seen.
Note: the above was first published on LSE EUROPP.
Kieran Devine is co-founder of the civic education social enterprise Connective Realities and holds an MSc in Comparative Politics from the London School of Economics.
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).
Pity you didn’t put the X-Axis of those charts the other way around. Convention is that the origin starts at 0,0
So people would be inclined to read those charts backward.
Should start with low age leading to high age, not the other way around, unless you are deliberately trying to confuse people.
If the “war generation” has more positive views towards European integration than the immediate post-war generations, then it’s probably over-optimistic to project that simple demographic trends (i.e. over-65 Leave supporters dying off and younger Remain supporters reaching voting age) will soon bring about a majority for Remain in the country. It is likely that the mortality rate is higher among the “war generation” than it is among younger over-65s. For a few more years, we might see more Remainers dying off than Leavers.
My father fought in WW2 and my mother was in London during the war. Both were very informed, politically interested socialists. They knew that the EEC was set up as a bulwark against socialism (which is illegal under EEC/EU competition law) and that it was intended to usur our parliamentary sovereignty.
The ‘Yes’ campaign misleadingly sought to convince voters that this was just about trade and that without it jobs would be at risk. Within a few years, unemployment had risen to over three million. People were conned by a campaign that was supported by every national newspaper (bar the Morning Star!) and that vastly outspent the ‘no’ campaign.
My parents were sufficiently well informed not to be fooled but sadly most were not and voted for spurious economic reasons. Having said that there certainly were idealists like your parents and other members of my family, who believed that regardless of concerns about sovereignty or socialism, the EEC would help to reinforce the peace.
This is the most noble reason for wanting to remain in the EU, but the EU’s post Cold War behaviour suggests it is less of a force for peace than might be imagined.
The situation in Northern Ireland has some parallels with the Second World War situation.
I personally come from a unionist background, so might be more expected to vote in favour of Leave, a la DUP.
However, I also experienced the sectarian horrors of the Troubles, and have no desire to see any return to a border, hard or soft, in Ireland.
I can see the peace-creating aspects of the EU, and this tips my judgement in favour of Remaining in the EU. This helps explain the 55% vote in favour of Remain in Northern Ireland, as many other soft unionists must have reasoned in a similar fashion.
What is the sample size for the War Generation? It seems like it might be too small to withstand all these control variables without losing most of its degrees of freedom.
What birth years did you use to define the ‘War Generation’?
My father fought in World War II, and both grandfathers in World War I. Although my family escaped comparatively lightly compared to many, it did suffer casualties in both wars. At the time of the first EC referendum in 1975 I was a teenager too young to vote, but I remember my father saying then that joining would mean I’d never have to fight in another war between western European nations. This suggests that voters in 1975 were well aware that the EEC was going to be something bigger than just the Common Market, contrary to claims by some Brexiteers – indeed, would the Labour government even have bothered to hold a referendum if it had only been about a trade deal? And would anyone have voted ‘No’ (which 37% did) unless they thought it involved something more?
I’m glad that my father didn’t live to see the referendum to take the UK out of the EU, or the nasty, divided, angry and intolerant country that it has turned us into.