Commentators have often sought to explain Brexit as a form of resistance to globalisation, but Jan Eichhorn argues we should treat such conclusions with caution. His analysis highlights that attempts to link Brexit to particular social classes or age groups underplay much of the division within these categories, and that identification with Englishness offers one of the best clues to understanding voting behaviour during the referendum.
Many newspapers have linked the Brexit decision to people characterised as the “losers of globalisation” who supposedly saw the referendum as their opportunity to rebel. This narrative is present in different political orientations and media outlets, ranging from the Financial Times to the Guardian or the Washington Post. The issue is also linked to class, with The Telegraph, for example, proclaiming that “Middle class liberals were the only social group to emphatically back Remain”. These stories leave us with a feeling that the so-called “Brexiteer” is a clearly defined person, who felt they needed to rebel against the status quo because of a (perceived) personal position of disadvantage.
Figure 1: Brexit vote by socio-occupational class (BES 2016) – only voters included
And indeed, we do see differences in support for Brexit based on social position. As figure 1 illustrates (showing data from the British Election Study collected shortly after the referendum), people in lower socio-occupational classes had a greater tendency to support leaving the EU (61 and 64 per cent respectively in the two lowest groups) compared to those in the highest socio-occupational groups (41 to 50 per cent respectively). However, the transition by social class is rather gradual – and not always linear. Higher intermediary middle class categories are split roughly evenly over Brexit, for example. Average differences between the lowest and most other socio-occupational class groups are only about 10 to 14 percentage points (with only one exception: higher professional occupations).
While one’s personal economic position is related to Brexit attitudes, we are far from seeing a country divided into distinct groups of social classes, where the advantaged wholeheartedly embrace the openness of the EU and the disadvantaged reject it outright. The figures actually paint a very different picture. Nearly half of those in higher middle class positions also support the UK exiting the EU. While percentages are lower than in some other groups, it suggests that divides run deeply through all groups of the population. It appears to be a matter of degree, not type. Indeed, other researchers have warned about over-simplification and the identification of a singular type of “Brexiteer”, based on social class alone.
Figure 2: Brexit vote by age (BES 2016) – only voters included
The relationship between age and Brexit attitudes, for example, is much stronger than that of social class (see figure 2) and, importantly, education has been shown to be strongly related to a person’s views about whether the UK should leave or remain. However, even such comparisons do not allow for perfect typologies. Did “the old” as a group cause Brexit? Those aged 66 or above were most likely to vote for it (around 60 per cent), but even amongst them four in ten opposed leaving the EU. And while most young people wanted to remain, around 30 per cent still opted for Brexit. So caution is most apt. When discussing demographic differences, we should talk about relative differences and tendencies, but avoid painting simplistic pictures that identify singular, homogenous groups for blame, when those groups do not actually exist in such a distinctive form.
When trying to understand the motivations of people, we also need to go beyond the material, physical characteristics we usually consider when differentiating groups. Otherwise, it would be hard to explain why, for example, regions in the South of England that are rather wealthy also voted for Brexit, contrary to the often claimed narrative about Brexit being a problem of an apparently poorer North. Questions of identity, for example, matter as well, and substantially so, as figure 3 reveals.
Figure 3: Identification with Englishness and Britishness by Brexit attitudes (BES 2016) – only voters in England included
People in England who feel strongly attached to their English national identity are much more likely to support Brexit than those who do not. Of those who chose the highest value for English identity on a 7-point scale, over 70 per cent voted to leave the UK. Conversely, over 80 per cent amongst those who only emphasise their Englishness slightly (2 on a 7-point scale) voted to remain. National identity mattered strongly in this referendum, but is rarely talked about to the same extent as questions of class or even age, although the divide is much more dramatic and cuts across different socio-economic groups in the population.
But even here, we have to remain careful not to jump to over-simplistic conclusions. Identities are complex and it appears that Britishness, while slightly related to views on Brexit, was much less important in differentiating between groups than English identity was, confirming other research that a particular aspect of the Brexit decision had to do with perceptions of Englishness specifically – which is rarely addressed.
Socio-demographic factors relate to a person’s likelihood of supporting Brexit, but without considering a broader spectrum of attitudinal concerns, such as identity, we will not be able to develop a genuine understanding of Brexit. The considerations presented here should act as a cautionary reminder that we should be vigilant in avoiding the mistake of taking on extensive, typologising narratives about political decisions. We should challenge explanations that misrepresent attitudinal tendencies in some groups of the population as the identification of particular groups that should be blamed as a whole for decisions that we might not be comfortable with and that challenge ideas of open societies.
Instead of singling out distinctive groups based on a few demographic characteristics, we should instead try to understand the interplay of the various tangible and intangible factors that may affect the views of people across different groups of the population. Simplistic typologies may create plausible stories, but instead of providing starting points for solutions, they are likely to further the construction of artificial dividing lines that we may bring into existence ourselves. We all, as readers, researchers, and members of societies should there take this on as a task in our daily lives: to challenge those explanations that might just be too comforting and easy to be accurate.
________
Note: this article was originally published on d|part and LSE Brexit. Data reference: Fieldhouse, E., J. Green., G. Evans., H. Schmitt, C. van der Eijk, J. Mellon and C. Prosser (2015) British Election Study Internet Panel Wave 9. DOI: 10.15127/1.293723.
About the Author
Jan Eichhorn is the research director of d|part and oversees the work on the Situation Room project. He also teaches Social Policy at the University of Edinburgh. Twitter: @eichhorn_jan
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.
Imagine a fair ground. Everyone passes a stall where they are super exhorted to buy a hoop and throw it on items which promise them a prize. They didn’t want to play but joined in weakly and now demand the promised prize – which as any remainer knows is an empty or poisoned chalice which affects everyone. The stall keeper keeps demanding the rest of us pay them – with our freedom of movement, lives, jobs, multi national families. It ism’t going to happen that we do that voluntarily. This is actually parliament and politicians’ finest hour standing up for what they know is right – to remain.
Reading Hon’s informed comment above….none of which I would dispute….it brings home how much we…Leavers and Remainers…..are overthinking the Referendum. The simple numbers are that out of a population of 65 million, 17 million voted to leave. Who they were, rich, poor, educated, racist, black, white, old, young, we actually have a pretty clear idea of the demographics…..obvious from day one analyses on the tv. What’s important to me in a Democracy, is how can only 26% of the population drag us down such a dangerous and contentious road? The title of Harold Clarke’s study: ‘Why Britain voted for Brexit……’ will mislead anyone who reads it. Britain DID NOT VOTE FOR Brexit. 26% of Brits voted for Brexit. They were influenced by an extraordinarily effective programme of disinformation and propaganda. One which the Third Reich would have been proud of. Is the propensity to be influenced by such a programme a function of education or knowledge? I’m too humble to make this sort of judgement. But I’d rather go to a trained brain surgeon than one who got his instructions from the side of a bus.
@Roger Cubberley, I share much of your sentiments, though I would add some nuances and caveats. Electoral results always have to be interpreted through the lens of the electoral system in order to make useful generalisations about what the country wants, given heterogeneity of views within the electorate. From the referendum setup – which both sides tacitly or explicitly acknowledged – Brexit would happen if a simple majority of voters voted Leave. There is much to criticise about the setup of a referendum which could have irrevocable consequences for a generation, instead of the usual 4-5 years of an electoral cycle. Politicians should have built in safeguards such as a supermajority to effect major constitutional change – which clearly is the case for leaving the EU as large swathes of legal statutes are dependent on EU law, or mechanism for the electorate to re-evaluate in light of new information at a later date. Nonetheless, it is hard to make a political and moral case for changing the rules after the result. Moreover, Parliament had the chance in March 2017 to vote against triggering Article50, therefore exercising Parliament’s constitutional prerogative superceding the advisory nature of the referendum. Yet owing to constituency and internal party dynamics, the outcome was a supermajority of MPs (a large majority of whose constituencies voted Leave) voting for departure from the EU. In terms of number and percentage of votes for altering the status quo, the Brexit referendum is among the highest compared to other national elections in the UK (some care needed to compare a 2-sided vote with multi-sided vote in general elections e.g. is a vote for the third largest party a vote for or against the largest party?), though one should also remember the number and percentage of votes for Remain was among the highest gathered by the losing side (again, care needed with defining the losing side in multiparty votes) – hence their views on what happens next must be taken into account. The figure of 26% of UK population voting for Brexit can be compared against the 23% and 24% of the population voting for Conservative and for Labour in their respective landslide general elections in 1983 and 1997 (the legitimacy of those electoral outcomes is seldom challenged), rather than read in isolation as what appears to be a small minority. (again, some care needed here – for example, in the months after the 1997 Labour landslide, opinion polls showed satisfaction with Tony Blair as PM reached as high as 90%; in contrast, opinion on Brexit remains as divided as ever.)
Almost 2 years on, I still struggle to understand the cognitive and emotive motivations of the Leave voters – whether they really understand the consequences (not many people have good understanding of the workings of the EU, and even less on the Customs Union and the Single Market), the huge downside risks, the elusive nature of the purported upsides, whether they misdirected their legitimate dissatisfactions with their socioeconomic circumstances at the wrong cause. (The only undisputed upside to departure from EU and Single Market is full control over immigration policy, though I take the view a restrictive immigration policy will do more harm than good.) Those Leave voters swayed by the promise on the side of the bus, surely have enough time by now to figure out the implausibility of the promise. Yet the reality is throughout the past year, opinion on the desirability of Brexit remains finely balanced (https://en.wikipedia.org/wiki/Brexit#Public_comment_pre-_and_post-Article_50_notification). Nobody can reliably predict the outcome of a second referendum, if it ever takes place.
I have read about the often cited characteristics of Leave voters e.g. level of education, age, income. However, one needs to be careful that some of the purported explanatory variables are positively correlated e.g. the older generation has lower % of people with degrees due to the expansion of higher education over the past decades, education is correlated with income, hence some of the variables may lack explanatory power when analysed under multivariate statistical models.
Great article. Ties in with something Yanis Vourafakis said about devolution for Scotland and Wales being a case of a small step in the right direction being worse than no step at all. Devolution for other countries in UK may have left England feeling that it’s interests are not being given priority. Since England as an entity makes up a huge chunk of UK -devolution for the whole of England wouldn’t make sense. The only place England can really get devolution from is the EU. Parallel devolution for the regions of England would have allowed the areas which feel aggrieved a way of airing their issues and getting more control. As it stands a major driving force for Brexit appears to be British exceptionalism which us really English exceptionalism. Unfortunately the whole of the UK is being drawn into this pointless and self defeating expression of English identity.
Before i married my German born wife I told her how lucky she was. After all I explained “I am an Englishman which means that I have won first prize in the lottery of life”. She did not take me seriously and nonetheless we married and lived happily ever after.
NOT ANY MORE YOU HAVEN.T
Like most complex political developments, Brexit has a multiple of causes. It is worth trying to understand the social dynamics of each factor on its own, as well as quantifying the contribution of each factor to the overall picture which would require multivariate statistical analysis. See multivariate study by Harold Clarke & et al (2017) “Why Britain Voted for Brexit: An Individual-Level Analysis of the 2016 Referendum Vote” which notes “National identities were influential as well, but operated further back in the set of forces affecting attitudes towards the EU” https://academic.oup.com/pa/article/70/3/439/3109029
Eichhorn should engage with the article Clarke (2017) and the very detailed study in the book “Brexit why Britain voted to leave the European Union” by the same author, before deciding if there is anything new to add.
Some studies examine causes at individual level, other studies identify causes at a constituency level.
Eichhorn singles out identification of Englishness as the most important cause. However, this factor is irrelevant to explaining voting pattern in Wales, which also has a majority voting Brexit, and in Scotland where a large minority also voted this way. Understanding voting pattern in these two regions are important – given the tightness of the national vote, a marginal smaller number of people in these regions voting for Brexit would have swayed the national result.