English national identity has been the topic of breathless discussion and debate for the past few years. In the aftermath of the 2014 Scottish Independence Referendum, David Cameron announced that “the millions of voices of England must also be heard.” In 2015, the Conservatives raised the spectre of Scottish nationalism in a bid to win English voters, and subsequently instituted “English votes for English Laws“. In 2016, commentators from across the spectrum described the vote to Leave the EU as an act of English nationalism (never mind that Wales also voted to Leave).
It came as a big surprise to me, then, that English nationalism is actually in fairly rapid decline, and has been so since at least 2012. According to the Annual Population Survey (APS), 58% of people in England identified as “English” in 2011. 10 years later, this number had fallen by 14% to only 44%. Meanwhile, the number identifying as “British” surged from 41% to 59%.
Indeed, British identity is increasing at the expense of English/Scottish/Welsh identities in all three countries, albeit much slower in Scotland and Wales. The change in England is so rapid that it cannot be explained solely by composition (that is, English-identifying older people being replaced by British-identifying younger people). Many people now have different national identities to 10 years ago, which can be confirmed using British Election Study (BES) individual-level data.
This raises some interesting questions about the political effects of national identity. As English identity declines, should we expect euroscepticism to decline, too (as the supposed causal link would imply)? Or, will English identity become even more exclusive/political as it shrinks to a small minority? I took a look at the BES data to see how the effects of national identity have changed over time.
The plot below summarises twelve models, each explaining voting intention for one party/issue in England, Scotland and Wales. Each model is a panel logistic regression, with the dependent variable being support for the outcome in question (e.g. 1 for intending to vote to Leave the EU, 0 for intending to vote to Remain). Each model includes controls for age, education, gender, household income and, where sample sizes allow, ethnicity, as well as ‘national identity’ scales interacted with the survey date. Each national identity scale is self-reported from 1 (Not at all) to 7 (Very Strongly) British, English, Welsh and Scottish. British and English identity are measured across all three countries, while Welsh and Scottish identity are only asked in those countries.
Most of the headline numbers here are as we’d expect. The largest average effects are the effect of British identity on voting Conservative in Scotland (positive), British identity on voting SNP (negative), Scottish identity on voting SNP (positive), British identity on voting Plaid Cymru (negative) and British identity on voting Conservative in Wales (positive). The effect of national identity tends to be smaller in England, but there is a clear positive effect of English identity on voting for Eurosceptic parties and voting to Leave the EU.
These effects tend to be relatively stable over time but some have experienced significant change. In Scotland, for example, Britishness has become a much more significant predictor of support for Leaving the EU. This is possibly due to vocal support for the EU among Scottish nationalists and the increasing association of Brexit with the Westminster government.
In England, the effect of Englishness on likelihood to vote for UKIP or the Brexit Party has shrunk. This is likely due to the decline of UKIP and rise of the Conservative Party in its place, which can be seen in the growing effect of Englishness on support for the Conservatives. That said, there seems to be little sign that Englishness is becoming a more significant motivating factor over this time period (2016-2021). The effect of Englishness on support for Leaving the EU and for Labour remained practically consistent.
On the other hand, there are some indications that Britishness is becoming more associated with supporting Leaving the EU and the Conservatives in England. This runs counter to the argument put forward by some that it is the relative strength of Englishness/Scottishness/Welshness versus Britishness which has most effect on political attitudes. It could suggest that as English-identifiers move towards identifying as British, they are retaining their eurosceptic Conservative lean. This raises questions about the true ‘causal’ effect (or direction of causality) of national identity.
How changing dynamics of national identity will affect party politics is hard to say. It seems likely that as English identity declines, English politicians will move more emphasis towards British nationalism, but as yet it is too early to say whether British identity is being effectively “activated” by any political tradition.