The impact of local identities on voting behaviour: A Scouse case study

Version 2.00

Abstract

The political salience of local identities has long been ignored in the British political science literature. This paper is a first step to address this, though a case study of the Scouse identity. The Scouse identity is primarily found in the English city of Liverpool and its environs. Using an original dataset of votes from across Merseyside, alongside the British Election Study, this paper explores the political salience and consequences of the Scouse identity and places it into a comparative context with sub-state national identities across Great Britain, namely Englishness, Scottishness, and Welshness. I find that holding a Scouse identity leads to higher levels of Labour support and lower levels of Conservative and Green support, that the political consequences of the Scouse identity vary across Merseyside, and that local identities can and do exist with sub-state national identities, challenging the myth of ‘Scouse not English’.

Introduction

National identities have long been seen as a key factor in shaping political behaviour within British politics, their importance turbocharged by the establishment of devolution to Scotland and Wales, the 2014 Scottish independence referendum and the 2016 referendum on leaving the European Union. There is now a rich and interesting sub-field exploring questions of English, Welsh and Scottish identity and their relationship with Britishness, as well as emerging scholarship on the national–territorial differentiation of Britishness, showing clearly that British identity means different things across different parts of the United Kingdom (Henderson et al. 2021).

Despite this welcome progress in understanding how national identities influence voting behaviour, political scientists still largely neglect one key element of an individual’s sense of self: their local identity. This study is an attempt to fill this lacuna, by using the case study of the Scouse identity, which is rooted in Liverpool, a city in the north west of England. Whilst there are many historical, social and cultural studies of the Scouse identity, the article represents the first time the Scouse identity was subject to a political science analysis.

This article makes three key contributions to the literature. The first contribution is that to shows how holding a local identity can have politically salient consequences when it comes to voting behaviour. The second is that to prove that local identities can matter just as much as, if not more than, national identities when it comes to shaping political behaviour. The third is to place the political consequences of the Scouse identity within a broader comparative context, alongside Englishness, Welshness and Scottishness.

Literature review — what role for local identities?

Scholars of voting behaviour have long understood the importance of identity on voters. The Columbia model of voting — first outlined in The People’s Choice (Lazarsfeld, Berelson, and Gaudet 1944) — argued that certain characteristics were heavily related to partisan behaviour. In this study, a series of familiar identity markers were shown as having the ability to accurately predict vote choice: class, race, religion, and place of residence (Hutchings and Jefferson 2018, 22). Although this model was criticised for being unable to cope with sudden swings in voting intention, identity — especially in the sense of class identity — has remained a salient part of many models of voting behaviour.

Identity is typically divided into social identity and personal identity (Tajfel and Turner 1979). Social identity refers to “a social category, a set of persons marked by a label and distinguished by rules deciding membership and (alleged) characteristic features or attributes”, whereas a personal identity “is some distinguishing characteristic (or characteristics) that a person takes a special pride in or views as socially consequential but more-or-less unchangeable” (Fearon 1999, 2). There is no necessary link between social and personal conceptualizations of identity, but in the case of a national (and local) identity, the two elements are related. It would be difficult to hold, say, a Scottish or Scouse personal identity without meeting at least some of the membership rules required to have it as a social identity.

Although interrelated, Fearon marks personal identity as the more important in understanding just how identity influences action, including voting. Fearon defines personal identity as

a set of attributes, beliefs, desires, or principles of action that a person thinks distinguish her in socially relevant ways and that (a) the person takes a special pride in; (b) the person takes no special pride in, but which so orient her behavior that she would be at a loss about how to act and what to do without them; or (c) the person feels she could not change even if she wanted to.

With regard to point (a), it is argued that “identity can typically be glossed as the aspects or attributes of a person that form the basis for his or her dignity or self-respect. Used in this sense, ‘identity’ has become a partial and indirect substitute for ‘dignity,’ ‘honor,’ and ‘pride’” (Fearon 1999, 11).

Identities shape behaviour because they help structure the range of possible choices available to an individual. Making decisions which go against one’s own identity is costly, as it acts as a challenge to the basis of one’s own dignity or sense of self. Returning to the Chicago model’s early attempts to understanding voting behaviour, the reason why identities like class, race, religion, and place of residence held such strong predictive power was (and is) because these identity categories carry a set of norms around behaviour, in the general form of “Members of category X are supposed to do (or ought to do) Y in situations A, B, C” (Fearon 1999, 27).

As such, an understanding of identity as the basis of an individual’s dignity or self-respect allows us to more clearly understand how it can affect an individual’s behaviour. Ultimately, decisions to act in a certain way are based on social norms — that is, informal understandings which are “(a) shared by other people and (b) partly sustained by their approval or disapproval” (Elster 1989, 99–100).

Typically, identities associated with political salience tend to be sticky, or slow to change. They are also, to some extent, self-perpetuating. Fearon argues that the pressure to act in accordance with your identity becomes internalised:

One might believe that following the norm is the right thing to do, whether because of early socialization or one’s independent judgement and experience. Relatedly, one might want to follow the norm because one would think badly of oneself otherwise — failing to follow the norm would undermine one’s pride, dignity, or self-respect. (Fearon 1999, 28)

Hence, one can see why, theoretically, individuals follow certain norms if they think that either the rules of their social identity group disallow it (e.g. ‘real Scousers don’t buy the Sun’) or that they would reduce feelings of their own self-worth. It is clear how and why this could find political expression.

Therefore, group identity is seen as important in shaping voting behaviour insofar as it shapes group consciousness, provides a readily-defined group of voters for parties to target, and a set of shared political and social symbols around which groups attach meaning and can be rallied to (Monroe, Hankin, and Vechten 2000; Achen and Bartels 2017).

The role of identities in British politics

The idea that identities might be important to understanding voting behaviour in the UK is nothing new. In post-war Britain class was seen as the main identity in determining voting behaviour with religion playing a role to a lesser extent (with the obvious exception of Northern Ireland) (Butler and Stokes 1974). This is perhaps best epitomized by the famous quote from Pulzer in 1967, when he declared that “Class is the basis of British party politics; all else is embellishment and detail” (Pulzer 1967, 98). But even then around 25 percent of British voters did not vote along class lines (Abramson 1972, 1291), and a steady, long-term process of class dealignment has weakened class-based voting to the point where, in 2001, “class effects on voting were negligible” (Clarke et al. 2004, 317). In 2019 Labour performed equally well among ‘middle-class’ ABC1 and ‘working-class’ C2DE voters (33 percent), whilst the Conservatives performed more strongly among working-class voters than middle-class voters (48 percent to 43 percent respectively) (YouGov 2019a).

Other identities feature strongly in predicting voting behaviour. Unsurprisingly, partisan identities strongly predict the party one will vote for, although partisan dealignment means that the usefulness of partisan identity has also declined (Crewe 1983; Dunleavy 2005). In 1964, 45 percent of the electorate had a ‘very strong’ partisan identification but, as shown in table 1, by 2001 this had dropped to just 13 percent of the electorate.

In the last decade a number of new identities have become increasingly important to understanding the voting behaviour of certain groups. The first is in Scotland, where the Unionist/independence identity cleavage has reshaped Scottish politics. Polling by YouGov a month after the 2014 Scottish independence referendum found 94 percent of Conservative voters, 82 percent of Labour voters and 79 percent of Liberal Democrat voters would vote no to independence, whereas 98 percent of SNP members would vote yes (YouGov 2014). By the time of the 2019 general election, these figures were 97 percent of Conservative voters, 70 percent Labour voters, and 94 percent Liberal Democrat would vote no to independence, while 90 percent of the SNP’s supporters backed independence (YouGov 2019b).¹

Like the Scottish independence referendum, the 2016 referendum on Britain’s membership of the European Union also either created new identities (or gave easy-to-understand labels to identities which were already emerging). Beyond simply ‘leavers’ and ‘remainers’, Goodhart conceptualised the UK as split into ‘somewheres’ and ‘anywheres’. For somewheres, their identity is rooted in a sense of place whereas for anywheres, it is much less so — it is instead rooted in a sense of socially-mobile cosmopolitanism (Goodhart 2017a). Goodhart himself noted how this analysis fed into Prime Minister Theresa May’s 2016 party conference speech (Goodhart 2017b), where she declared

too many people in positions of power behave as though they have more in common with international elites than with the people down the road, the people they employ, the people they pass in the street. But if you believe you’re a citizen of the world, you’re a citizen of nowhere. You don’t understand what the very word ‘citizenship’ means (May 2016).

For Sobolewska and Ford, the salient identity groups today are ‘identity conservatives’ and ‘identity liberals’, divided over their experiences of educational expansion, mass immigration and ethnic change (Sobolewska and Ford 2020). In this framework, identity liberals can be divided into two groups. ‘Conviction liberals’ are typically university educated, they “value individual freedoms very highly, have little attachment to traditional majority identities or values, and… are cosmopolitan, pro-migration and embrace diversity.” They see diversity as a social good in and of itself, but also see defending diversity and minorities as a salient element of their social and political identity. ‘Necessity liberals’ tend to be ethnic minority voters for whom “anti-racist and pro-diversity stances are not a matter of personal values but of necessity”. As a result, while these necessity liberals do “favour strong political action to defend diversity and prohibit expressions of prejudice and discrimination as this is a matter of basic self-interest for ethnic minority voters” this might not always extend to other non-dominant groups, such as LGBT voters (Sobolewska and Ford 2020).

Finally, ‘identity conservatives’ operate on the opposite side of the divide. They tend to be white voters who leave school with few or no qualifications. They recognise that their share of the population is falling, and hence so is the interest politicians show them. The rise in ethnic diversity — lauded by the conviction liberals — is seen as a threat to identity conservatives’ place in the world (Sobolewska and Ford 2020).

In addition to how academics define these identity groups, voters themselves also have a strong sense of their own Brexit identity. In December 2021, over five years after the referendum campaign, 41 percent of the public had a ‘very strong’ remain or leave identity, compared to just 18 percent who had a very strong party identity. 33 percent had a ‘fairly strong’ remain or leave identity, compared to 38 percent for a ‘fairly strong’ partisan identity. For ‘not very strong’ the figures were 9 percent for Brexit identity and 25 percent for a partisan identity, while for ‘none’ they were 17 percent and 18 percent respectively (Ipsos 2021).²

Place is central to identity, and hence voting behaviour. Jennings and Stoker argue that we are seeing a division within England, between cosmopolitan areas which are global in outlook and pluralistic in their identity, and ‘provincial backwaters’, where Euroscepticism and anti-immigration sentiment is higher, where people are attitudinally illiberal and are more likely to identify as English rather than British (Jennings and Stoker 2016; Jennings and Stoker 2017). This is a trend which began before, but was turbocharged by, the EU referendum.

Immediately after the referendum ‘left behind’ became buzzphrase du jour. Left-behind areas are “the cities, towns and regions located beyond the large and successful agglomerations that had become the sites of economic growth in the 21st century” (Leyshon 2021), and a higher-than average (and, for the commentariat, higher-than-expected) support for leave in these areas was seen as ‘the revenge of the forgotten’, both against the establishment and, due to claims these areas were now increasingly Conservative-leaning, a revenge of the forgotten against Labour too. Indeed, Labour was seen to have abandoned these left-behind voters as far back as 1994 with the creation of New Labour. The left-behind thesis has been criticised from numerous angles, including for ignoring ethnic minority leave voters (Martin, Sobolewska, and Begum 2019), the conflation of economic and cultural phenomena (Calvert Jump and Michell 2021), the statistical evidence that the most economically deprived areas, and voters, still tended to vote Labour (when they did actually vote) (Furlong 2019), and that deprivation was only weakly correlated with voting to Leave (Calvert Jump and Michell 2021). Despite this, the idea of the ‘left-behind voter’ shaped the ideas of policy wonks and politicians alike, including prime ministers Theresa May (through tackling the ‘burning injustices’ within society) and Boris Johnson (by ‘levelling up’ the country).

Following the 2019 general election, however, ‘left behind’ has been replaced by the ‘red wall’ as the key realignment in British (although really English and, to a lesser degree, Welsh) politics — at least for journalists and commentators. The red wall refers to a number of seats across North and Midlands where, based on constituency demographics, the Conservative Party historically underperformed and the Labour Party has been dominant (Kanagasooriam and Simon 2021). Many of these seats fell to the Conservatives in 2019, after the party recorded massive swings towards them in 2017. For some, the ‘red wall’ and ‘left behind’ areas overlap (Cooper and Cooper 2020), whereas for others ‘left behind’ represents something closer to a state of mind, or way of seeing the world, whereas the red wall is more a geographic descriptor.

The political salience of more traditional national identities has also received increased attention. The UK is a multi-national state, and so how those national identities operate (and inter-operate) is of vital importance to understanding political trends. As McCrone notes, “National identity relates to political behaviour and attitudes, but not in a straightforward way… We should think of national identity as a frame for understanding, not as a determinant of how people vote.” (McCrone 2020) Like with other identities, national identities shape possibilities, rather than determining outcomes.

Scottish and Welsh national identities have been brought into focus with the introduction of devolution, whilst Englishness has received increasing attention due to broader issues surrounding the West Lothian question, a perceived lack of ‘English’ institutions to mirror the Scottish and Welsh parliament, and the relative strength of the leave vote in England. Following the Brexit vote, European identity unsurprisingly became politically salient and contested. There was even a campaign to ‘declare yourself European’ on the 2021 census (Stay European 2021).³

As noted above, while the Scottish independence referendum reconfigures politics along a sharper Unionist/independence identity cleavage, this turbocharged processes which were already in motion. Whilst the Unionist/independence identity may be sharp, historically Scottishness as an identity is generally blunter. In the Scottish independence referendum ‘Scottishness’, or the Scottish national identity, was available to both sides of the independence debate (Keating and McEwen 2020). While there has been a “tightening up” of the relationship between national identity and political choice since the 2014 referendum, the two concepts do still remain relatively autonomous (McCrone 2020). Analysis of the EU referendum found that Scottish and British identifiers both voted Remain in equal proportions, and that whilst all major parties had majority support for remain, the two parties with the largest leave minorities were the Conservatives (45 percent) and the SNP (37 percent) (McCrone 2020).

Unlike in Scotland, where the EU referendum produced a clear answer, and one which stood in stark contrast to England, the case of Wales — and thus Welsh national identity — is more complex. Wales voted to leave the European Union by 52.5 percent to 47.5 percent. Among voters who identified as Welsh only, 29 percent voted to leave, whereas among those who saw themselves as Welsh British the figure was 58 percent (Wyn Jones and Larner 2021). Also unlike in Scotland, where Conservative unionism was able to establish itself and where Labour’s dominance only really began in 1964, Wales has always been electorally lopsided, being first dominated by the Liberal Party and then the Labour Party (Wyn Jones and Scully 2006). Indeed, Labour has won every general election in Wales since 1922.

Another difference between Wales and Scotland is that calls for Welsh independence have always been muted, although it has been somewhat boosted by Brexit, the rhetoric of Welsh Labour leader Mark Drakeford (Larner 2019), and perceived differences in performance between Wales and England in handling covid-19. Support for Welsh independence has thus spread its appeal beyond those who identity as ‘Welsh only’ (Wyn Jones 2019), although it is important not to overstate this: support for Welsh independence is still heavily concentrated within the ‘Welsh Only’ group, with no other mainstream identity group reporting support of greater than 15 percent (Wyn Jones and Larner 2021). The lead for no to independence has shrunk from 53 percent (70 percent against to 13 percent in favour) the week before the Scottish referendum to a low of 7 percent (49/42) in April 2021. The figure is now sitting at around 32 percent (53/21), but as Wyn Jones and Larner note, “the period since 2016 has in fact seen an overall increase in support for devolution across most identity groups – including some of those most supportive of Brexit” (Wyn Jones and Larner 2021).

One final difference between Scotland and Wales is that Welsh Labour seems much more confident in embracing Welsh identity than their Scottish counterparts, which is partly why Plaid Cymru has not been able to replicate the success of the SNP. In 2000 the Welsh Labour leader Rhodri Morgan labelled his party Welsh Labour rather than New Labour (Bory 2020) while his successor Carwyn Jones successfully utilised “a specific ‘Welsh Labour’ rhetoric” (Moon 2017).

The political salience of Englishness has historically been largely muted, subsumed into a wider sense of Britishness. However an IPPR report in 2012 found evidence of the emergence of an ‘English political community’, which is marked by “notable concerns within England about the seeming privileges of Scotland, in particular, in a devolved UK, a growing questioning of the capacity of the current UK-level political institutions to pursue and defend English interests, and one underpinned by a deepening sense of English identity” (Wyn Jones et al. 2012, 2). For these authors, this English political community had emerged as a ‘backlash’ against the establishment of the devolved institutions in Scotland and Wales.

Kenny, however, argues that English national identity was growing before the establishment of devolution, spurred on prior to devolution by a “broader return across Europe to forms of belonging associated with historic forms of national and regional identity” and a “growing sense of turbulence and uncertainty associated with changes in the global economy and geo-politics” which fed into a deepening loss of confidence in the UK state. Faced with a sense of straining state capacity, devolution was one answer (albeit not for England), and increased European integration was another (Kenny 2014, 27). As a result, Englishness became more closely aligned with Euroscepticism (Wellings 2012).

So, whilst the English political community was not created by grievances over devolution, just like how the Scottish independence referendum turbocharged the politically salience of a pre-existing identity, the establishment of devolution turbocharged the political salience of Englishness. This is explained by the concept of devoanxiety, the “sense that devolution has left the Celtic periphery of the state in general, and Scotland in particular, unfairly privileged at the expense of England” (Henderson and Wyn Jones 2021, 103–123). However, until the rise of UKIP and the EU referendum, the willingness of the Conservative Party to also, reluctantly, accept devolution meant that these voters were politically sidelined (Wellings 2012).

Englishness also has an image problem among cultural elites. As Kenny notes, “For many commentators, Englishness is irretrievably tainted by its regressive, conservative, and ethnically charged character” (Kenny 2014). Whereas Scottishness, Welshness, and Britishness are seen as cuddly, civic conceptions of nationalism, for some reason Englishness is treated differently. For Kenny, this is rooted in Thatcherite conservatism, which “ensured that progressives became increasingly wary of Anglicized expressions of nationhood, and opted to identify instead with the nationalisms emerging in Scotland and Wales, and, in the 1990s, with the civic liberalism associated with Britishness” (Kenny 2014).

Unlike with Scottish and Welsh identity, however, Englishness does not seem to be linked to demands for an English parliament. Henderson succinctly outlines the tension at the heart of English nationalism: “England is too big to accommodate by itself, the English want England to be accommodated by itself, but not necessarily with an English Parliament” (Henderson 2021).

Henderson and Wyn Jones note how “English nationalism combines both concern about England’s place within the United Kingdom as well as fierce commitments to a particular vision of Britain’s past, present, and future” (Henderson and Wyn Jones 2021, 4). They also note how Britishness means different things in different parts of Britain, and that “many of the attitudes that attach to Englishness in England attach to Britishness in Scotland and Wales” (Henderson and Wyn Jones 2021, 6). For some, the fact Britishness “has been construed differently in different parts of the state… is the genius of traditional unionism” (Keating and McEwen 2020). For others, this is hugely problematic: is such a Janus-faced identity robust enough to heal the divides of the Scottish independence or EU membership referendum?

Regardless, it is clear that national identities matter, and that they are now receiving appropriate academic interest. There remains, however, one key area of identity formation which is neglected in the academic literature on British psephology: the role of local identities on voting behaviour. There is no need to be overly prescriptive about the exact level of ‘localness’ — politically salient identities can operate on the parish level, the city level, the city region level, or the regional level.

There have been a handful of recent high-profile attempts to mobilise sub-national identity groups in the political sphere, with limited success, including the Yorkshire Party, the Lincolnshire Independents, and the Northern Independence Party, but there has been next to no attempt to explore which local identities may have political salience⁴. One study has sought to explore the role of local (as well as national) identities on support for metro mayors, finding that while a relationship did exist between holding a Scouse identity and supporting the Liverpool City Region metro mayor in a bivariate model with controls, this relationship disappeared in the full model. In this case, geographic place mattered more than identity (Jeffery 2022).

This study will be the first to attempt to measure the political salience of a local identity, using the Scouse identity as a case study. The Scouse identity is a good case study to use because it is a clear example of local identity which is often portrayed as politically salient (see Jeffery 2017; Wilks-Heeg 2019; Jeffery forthcoming). Indeed, Merseyside’s politics is so different to the rest of Northern England in its anti-Conservatism that in their 2021 study of the red wall, Kanagasooriam and Simon decided to include a dummy variable for one area only: Merseyside (Kanagasooriam and Simon 2021, Table 1). And, as Liverpool’s reputation has recovered from the nadir of the 1980s, and people move out of the city to its environs, Merseyside is becoming increasingly Scouse, although this is an uneven process (Wilks-Heeg 2019).

Finally, the Scouse identity is an interesting choice due to its perceived tension with the English national identity, an idea which is widely reported as a fact among the city’s cultural elites. As politically salient identities typically require an outgroup against which to define themselves (see Tajfel and Turner (1979), pp. 43-44 for theory, and Jeffery (forthcoming) for how this operated in Liverpool), this suggests that the Scouse identity may indeed be a fruitful avenue for exploration.

These points are brought together in an article in the Liverpool Echo⁵ immediately after the 2019 general election. The headline was “‘Scouse not English’ goes viral as Merseyside remains defiant after the election”, as “Using the viral hashtag #scousenotenglish, scores of Twitter users spoke of how proud they are to be from Liverpool following the Labour victory in the city” (Hadfield 2019).

Research Questions

In order to explore the political salience of the Scouse identity, this article will explore four key questions:

1. How does Scouse identity relate to party choice?
2. How does Scouse identity relate to EU referendum choice?
3. How does Scouse identity, and the political salience thereof, vary across Merseyside?
4. How does Scouse identity interact with national identities?

Data and methodology

This study makes use of two datasets: the British Election Study and a poll of voters from across Merseyside, commissioned by the author and conducted by Panelbase. The poll has a representative sample of 616 voters from across Merseyside and the questions used were taken from the British Election Study to allow for direct comparison between the two datasets. The one exception to this was vote choice, where respondents in the Scouse survey were asked to rank parties. A respondent’s vote choice was estimated by taking their highest-ranked party.

For respondents in the British Election Study, wave 15 (March 2019) to wave 21 (May 2021) was used, and for all variables used the respondent’s most recent response for a given variable. This was done to maximise the response rate across all the variables used. There were some minor issues of question wording for views on the European Union, where waves 15 to 19 asked respondents whether they would stay or remain, whilst waves 20 and 21 asked if they would rejoin the EU or stay out.

The control variables for this study are gender, age (as an ordered factor, from ‘<25’, ‘25-34’, ‘35-44’, ‘45-54’, ‘55-64’, and ‘65+’), socio-economic group (ABC1 and C2DE), income (as an ordered factor, ‘Under £8,500’, ‘£8,500 — £13,499’, ‘£13,500 — £23,999’, ‘£24,000 — £33,999’, ‘£34,000 — £49,999’, ‘£50,000 — £70,000’, and ‘Over £70,000’).

Other variables used include the British Election Study’s left-right, authoritarian-libertarian, and populism scales, as well as an equality scale based on respondents’ views on whether attempts at equality for women, gays, and ethnic minorities has gone too far, and an immigration scale based on whether immigration is positive along cultural and economic dimensions.

In order to measure identity, respondents were asked to rank how strongly English or British they feel on a scale of 1 (not at all) to 7 (very strongly). In the Merseyside survey respondents were asked how Scouse they on the same scale, as well as whether they would consider themselves a Scouser (as a binary yes/no question). Scottish and Welsh respondents identity from the British Election Study to use as a point of comparison.

In order to answer the research questions, a series of logistic regression model are presented, with party choice as a binary variable for Labour, Conservatives, Green, Liberal Democrats, Reform UK and UKIP, as well as a binary variable for whether a respondent would rejoin the European Union.

Results and analysis

The results of the first set of regression analysis are shown in table 2. In answer to the question of how Scouse identity relates to party choice, it is quite clear that Scouse identifiers are more likely to support the Labour Party and less likely to support the Conservative Party, and interestingly, the Green Party. Scouse identifiers are no more likely to want to rejoin the EU than non-Scouse identifiers across Merseyside.

Table 2: Regression models for vote choice

	Labour	Con	Lib Dem	Green	Reform	UKIP	Rejoin
Male	-0.26	-0.52	0.58	0.96	-0.40	0.37	0.25
	(0.22)	(0.30)	(0.43)	(0.52)	(0.65)	(0.57)	(0.27)
Age	-0.23 **	0.22 *	0.15	0.06	0.56 *	-0.34	-0.30 ***
	(0.07)	(0.10)	(0.14)	(0.17)	(0.22)	(0.21)	(0.09)
ABC1	-0.26	0.57 *	0.18	-0.20	0.82	-0.19	0.20
	(0.16)	(0.22)	(0.35)	(0.36)	(0.48)	(0.41)	(0.20)
Income	0.05	0.08	0.11	-0.42 **	-0.14	-0.12	0.06
	(0.07)	(0.09)	(0.14)	(0.16)	(0.19)	(0.16)	(0.08)
LR scale	0.43 **	-0.47 *	-0.49	0.16	-0.48	-0.19	0.04
	(0.17)	(0.21)	(0.32)	(0.36)	(0.38)	(0.41)	(0.22)
AL scale	-0.19	0.70 **	0.43	-0.87 *	0.29	0.65	-0.50 *
	(0.17)	(0.24)	(0.34)	(0.39)	(0.51)	(0.46)	(0.22)
Populism	0.47 *	-0.84 ***	-0.41	-0.28	1.22 *	0.02	-0.14
	(0.19)	(0.25)	(0.38)	(0.41)	(0.51)	(0.46)	(0.24)
Equality	-0.32 *	0.26	-0.15	-0.09	0.28	0.44	-0.74 ***
	(0.14)	(0.18)	(0.29)	(0.34)	(0.32)	(0.33)	(0.17)
Immigration	0.12	-0.10	0.29	0.05	-0.30	-0.26	0.25 ***
	(0.07)	(0.09)	(0.15)	(0.17)	(0.19)	(0.17)	(0.08)
Britishness	0.09	-0.03	0.21	-0.03	-0.23	-0.09	-0.02
	(0.07)	(0.10)	(0.17)	(0.15)	(0.18)	(0.18)	(0.10)
Englishness	-0.01	0.03	-0.09	-0.11	0.33	0.07	-0.22 *
	(0.08)	(0.10)	(0.15)	(0.16)	(0.24)	(0.20)	(0.10)
Scouser	0.83 ***	-1.01 ***	0.81	-2.63 ***	-0.83	0.58	0.35
	(0.21)	(0.27)	(0.45)	(0.67)	(0.59)	(0.60)	(0.26)
N	540	540	540	540	540	540	454
AIC	617.70	400.04	225.46	165.41	132.94	154.36	432.75
BIC	673.49	455.83	281.25	221.20	188.73	210.15	486.28
Pseudo R2	0.26	0.31	0.11	0.35	0.35	0.17	0.43
*** p < 0.001; ** p < 0.01; * p < 0.05.

It is also possible to go beyond statistical significance and analyse the relative importance of the Scouse identity variable in the regression models compared to the other variables. This is presented as figure 1, with the relative importance rank of the Scouse variable shown as the black line. The full table can be found in table 4, in the appendix.

Figure 1: Chart showing the relative importance of the Scouse identity variable

Figure 1 clearly shows that the Scouse identity variable is the most important in determining support for the Labour, Conservative and Green parties, and despite not being statistically significant in the regression model, the second-most important variable in the Liberal Democrat model (after views on equality). Interestingly, for Reform UK, UKIP, and rejoining the EU — all variables with a strong European aspect to them — the Scouse identity is less important, ranking fifth for the two populist radical right parties and sixth for determining support for rejoining the European Union.

There is also evidence that the Scouse identity is associated with certain politically-related norms which those who do hold the identity would be reluctant to break, and which are generally not held by those within Merseyside who are not Scousers. Figure 2 shows responses to three statements: ‘Real Scousers should vote Labour’, ‘Real Scousers should not vote Conservative’, and ‘Real Scousers should not buy the Sun newspaper’.

Figure 2: Response to survey question by Scouse identity

Interestingly, Scousers are slightly more anti-Tory than they are pro-Labour: 64 percent of Scouse identifiers either agree or agree strongly that ‘real’ Scousers don’t vote Tory, compared to around 57 percent for those who agree or strongly agree that ‘real’ Scousers should vote Labour. For non-Scousers the values are 33 percent and 26 percent respectively.

To put this into comparison, the strongest norm associated with the Scouse identity is whether real Scousers should buy the Sun newspaper, for obvious historic reasons (Jeffery forthcoming; Foos and Bischof 2022). 88 percent of Scouse identifiers agree or strongly agree that ‘real’ Scousers should not buy the newspaper, compared to 56 percent of non-Scouse identifiers. Taken together, this is clear evidence that the Scouse identity contains politically-salient norms, and also that the Scouse identity is politically consequential more broadly.

Geographic variation of Scouse identity

It is also worth exploring whether Scouse identity plays out differently across Merseyside. Figure 3 shows the geographic distribution of the Scouse identity across the five council areas comprising Merseyside.

Figure 3: Map showing strength of Scouse identity.

Unsurprisingly, Liverpool has the highest level of Scouse identity, followed by Knowsley to its east. Sefton to the north has the third highest level of Scouseness at 53 percent, but this hides geographic variation — the south of the borough is more strongly Scouse compared to the top half, which includes the town of Southport where there is a movement for it to leave Merseyside and rejoin Lancashire.

Wirral is the fourth most Scouse, again with some geographic variation: the east of the peninsular is more Scouse than the west, and this also reflects the Labour-Conservative divide. Sefton is the least Scouse ward, and this reflects a broadly low level of Scouseness across the council area as a whole.

In order to test whether the impact of the Scouse identity plays out differently in different parts of Merseyside, an interaction effect was added to the previous models controlling for whether a respondent lives in Liverpool or not. This is shown in table 3.

Table 3: Regression models for vote choice, with interaction effects

	Labour	Con	Lib Dem	Green	Reform	UKIP	Rejoin
Male	-0.28	-0.52	0.58	1.01	-0.40	0.47	0.25
	(0.22)	(0.30)	(0.43)	(0.52)	(0.66)	(0.58)	(0.27)
Age	-0.23 **	0.19	0.16	0.08	0.61 **	-0.38	-0.28 **
	(0.08)	(0.10)	(0.15)	(0.18)	(0.22)	(0.22)	(0.09)
ABC1	-0.27	0.62 **	0.18	-0.25	0.74	-0.12	0.18
	(0.17)	(0.23)	(0.35)	(0.37)	(0.49)	(0.42)	(0.20)
Income	0.05	0.07	0.11	-0.42 **	-0.15	-0.13	0.07
	(0.07)	(0.09)	(0.14)	(0.16)	(0.19)	(0.16)	(0.08)
LR scale	0.41 *	-0.50 *	-0.48	0.24	-0.42	-0.22	0.07
	(0.17)	(0.21)	(0.33)	(0.38)	(0.40)	(0.40)	(0.23)
AL scale	-0.19	0.69 **	0.45	-0.90 *	0.33	0.61	-0.50 *
	(0.17)	(0.25)	(0.34)	(0.40)	(0.53)	(0.46)	(0.22)
Populism	0.48 *	-0.80 **	-0.43	-0.32	1.21 *	0.09	-0.16
	(0.19)	(0.25)	(0.38)	(0.42)	(0.52)	(0.46)	(0.24)
Equality	-0.32 *	0.26	-0.15	-0.04	0.28	0.44	-0.74 ***
	(0.14)	(0.18)	(0.30)	(0.36)	(0.33)	(0.33)	(0.17)
Immigration	0.13	-0.09	0.29	0.03	-0.32	-0.25	0.25 ***
	(0.07)	(0.09)	(0.15)	(0.18)	(0.20)	(0.17)	(0.08)
Britishness	0.08	-0.04	0.21	-0.02	-0.25	-0.07	-0.01
	(0.08)	(0.10)	(0.17)	(0.15)	(0.18)	(0.17)	(0.10)
Englishness	-0.02	0.01	-0.08	-0.08	0.42	0.05	-0.21 *
	(0.08)	(0.10)	(0.15)	(0.16)	(0.25)	(0.20)	(0.10)
Scouser	0.13	-0.93 *	1.26	-0.63	-0.51	1.00	0.64
	(0.34)	(0.45)	(0.69)	(0.85)	(1.15)	(0.79)	(0.42)
Liverpool	-0.39	-0.58	0.53	0.77	1.42	-0.54	0.51
	(0.33)	(0.41)	(0.74)	(0.63)	(0.77)	(1.20)	(0.44)
Liverpool * Scouser	1.11 *	0.22	-0.83	-3.34 *	-1.34	-0.26	-0.68
	(0.45)	(0.61)	(0.91)	(1.38)	(1.43)	(1.35)	(0.58)
N	540	540	540	540	540	540	454
AIC	615.31	401.52	228.65	163.24	133.62	156.93	435.13
BIC	679.68	465.90	293.03	227.62	197.99	221.30	496.90
Pseudo R2	0.28	0.32	0.12	0.38	0.38	0.18	0.44
*** p < 0.001; ** p < 0.01; * p < 0.05.

Firstly, there are statistically significant variables related to identity or location for Labour, the Conservatives and the Greens, similar to the previous models. In the Conservative model the Scouser variable is negative and statistically significant, while the Liverpool variable and the interaction effect are not statistically significant. This suggests that regardless of where one is in Merseyside, holding a Scouse identity reduces the liklihood of voting Conservative - ie. the effect is uniform across Merseyside.

For Labour and the Green models neither the Scouser nor the Liverpool dummy variables are statistically significant, but the interaction effect is. For Labour, this means that in Liverpool the Scouse identity means you are more likely to vote Labour than those who do not hold a Scouse identity, but this is not the case outside of Liverpool, where there is no difference based on identity. For the Green Party model, respondents in Liverpool with a Scouse identity are much less likely to vote for the Green Party than those without a Scouse identity, but this relationship does not hold outside Liverpool.

As such, there is a geographic element to the importance of Scouse identity: for the Conservatives, Scouse identity had a negative impact on liklihood to support the party, whereas the effects for Labour and the Greens are restricted only to Liverpool. This could be because, when a Green candidate stands, they can take vote share away from both Labour and the Conservatives. However, in Liverpool, voting Green generally means you are voting against a sitting Labour councillor, or in a seat where Labour are second place. As such, a Green vote is effectively an anti-Labour vote and so it is not surprising the relationship for the Scouse identity acts in opposite directions. This also suggests that the Scouse identity is more strongly an anti-Conservative identity than pro-Labour.

However, there is sufficient evidence here to support the claim that Scouse identity does play a role in shaping political preferences, especially with regard to support for Labour, Conservative and Green parties. Labour and the Conservatives are simple enough to understand: as Jeffery (2017) notes, the Scouse identity became imbued with anti-Conservative flavour in the 1980s, a development which faced very few sources of opposition on the local level.

Scouse identity and national identities

The final element of this analysis explores the extent to which Scouse identity interacts with national identities — in this case Englishness and Britishness. This takes work by Henderson et al. (2021) as a guide, and utilises their relative territorial identity (RTI) measure to analyse complex multilevel local and national identities. Before that, however, it is worth discussing the relationship between local and national identities.

Unlike studies which focus on Britishness and sub-state national identities, where there is only relationship of interest in each model (i.e. how a respondent sees themselves in terms of Scottishness and Britishness, Welsh and Britishness, English and Britishness), including Scouse identity adds an extra layer of interaction. This is shown in figure 4.

Figure 4: Theoretical model of identities in Merseyside

In order to explore the relationship between Scouseness and national identities, Henderson et al’s relative territorial identity (RTI) measure, which aims “to capture the priority given to sub-state or state-level identity” (Henderson et al. 2021), will be used. Firstly, respondents are asked to place themselves on a scale of 1 to 7 based on how strongly they identify with a given identity, with 7 being the strongest and 1 the weakest. The respondent’s Britishness score is then subtracted from their sub-state identity score to produce a 13-point measure of relative identity, which is then rescaled to between -1 and 1. A measure of -1 means a respondent is completely British not English/Scottish/Welsh/Scouse, whereas +1 means the respondent is completely English/Scottish/Welsh/Scouse and not British. 0 means that a respondent is equally British and English/Scottish/Welsh/Scouse. The RTI measure for Englishness versus Scouseness is also included, in line with figure 4.

Figure 5: Distribution of the relative territorial identity (RTI) variable

Figure 5 shows the distribution of the RTI measure across each identity of interest. Starting with the RTI for English-Scouse, there is very little evidence of the ‘Scouse not English’ trope often deployed across Merseyside. In fact, among the Merseyside dataset, more people feel exclusively English than feel exclusively Scouse. The same is true for the British-Scouse RTI, and interestingly, if you separate out respondents in Liverpool, a plurality of respondents are equally Scouse and English/British, rather than exclusively one or the other. The vast majority of respondents feel some attachment to Englishness/Britishness as well as their Scouse identity. Further, a lower share of respondents feel exclusively Scouse compared to those who feel exclusively Scottish, or even exclusively Welsh. As such, it is immediately clear that the Scouse identity does co-exist with the English and British national identities.

The final step in this analysis is to explore the impact of RTI on voting behaviour, again using a series of logistic regression models with party choice as a dummy variable. The full tables for each model are presented in the appendix, but figure 6 shows the predicted probably of voting for a given party at each level of RTI. Relationships which are statistically significant are denoted by a solid line.

Figure 6: Relationship between RTI and support for rejoin/political parties

In terms of rejoining the European Union, the more you identify as Scouse relative to English the greater the likelihood you have of wanting to rejoin the EU. Like Henderson et al (2021) there is a positive relationship between British-Scottish RTI and pro-EU sentiment and a negative relationship between British-English RTI and pro-EU sentiment. However, unlike in Henderson et al’s study, this analysis finds no statistically significant relationship between British-Welsh RTI and view on the European Union. There is no statistically significant relationship for Scouse identity relative to Britishness.

When it comes to party-political behaviour there is evidence that the more Scouse you are, relative to both English and British, the more likely you are to vote for the Labour Party. The only other statistically significant relationship is for the British-Scottish RTI, which is negative. This is unsurprising, given that Scottishness is a clear correlate with supporting both Scottish independence and the Scottish Nationalist Party.

When it comes to supporting the Conservative Party, the more one identifies as Scouse, Welsh or Scottish, relative to British, the less likely one is to vote Conservative, whereas the reverse is true for English. There is also a negative relationship between English-Scouse RTI.

The only other relationships of note for the two Scouse RTIs are for the Green Party, whereby the more Scouse you are relative to English/British, the less likely you are to support the Green Party. Interestingly, the reverse is true for British-English RTI, where the more English one feels the greater the likelihood of supporting the Green Party.

Although there is no statistically significant relationship for either of the Scouse RTIs, it is interesting to see that the more Welsh/Scottish/English one feels, relative to British, the less likely one is support the Liberal Democrats. There is a negative relationship between British-Welsh RTI and UKIP support, and a positive relationship between British-English RTI and UKIP support. There is no territorial dimension to support for Reform UK.

One final point to note here is the role of populist values. Whereas in all the models every statistically significant variable acts in the expected direction, populism does not. In the Scouse models for likelihood to vote Labour and to vote Conservative populism operates in the opposite direction than expected. For Labour voting the more populist one is the more likely one is to vote Labour, whereas in Wales and England there is no relationship and for Scotland there is a negative relationship between Labour support and increased populist values. On the other hand, in England and Scotland higher populism scores are related to an increased likelihood of voting Conservative, but for the Scouse models they are negatively related.

One reason why this might be the case is that the population of Merseyside has a higher level of baseline populist values compared to the result of the United Kingdom, and those who identify as Scouse are more likely to have a higher populist score, and are more likely to vote Labour (and less likely to vote Conservative). Thus, this is a correlation rather than causation. There is also the idea that the Scouse identity was politicised through conflict with central government, and as such anti-elite, populist elements were a core part of the politicisation of Scouseness, and which have survived to this day. This ties into a broader emerging analysis regarding Scouse identity being a specifically left-wing populist identity (Jeffery 2021).

Conclusions

This study had two broad aims. The first was to show that local identities can matter just as much as national identities and the second was to begin to understand the political consequences of the Scouse identity. In both cases, this study has made significant contributions to our understanding.

Firstly, it is evident that local identities matter. As figure 1 and table 4 show, Scouse identity is one of the key variables in predicting Labour, Conservative, and Green support. Secondly, the political consequences of holding a Scouse identity are not uniform across Merseyside: while a Scouse identity is related to a lower likelihood of voting Conservative across the county, in the case of the Labour and Green party the Scouse identity is only politically salient in Liverpool.

The political salience of local identities also varies across parties in other ways. Liberal Democrat support in Liverpool is not influenced by Scouseness, and neither are views on Reform UK nor UKIP. When it comes to support for the European Union, in line with other studies, English identity is found to be associated with Euroscepticism while Scottish identity is associated with pro-European sentiment. The Scouse identity is associated with a more pro-European position than the English identity, but that no difference exists between Britishness and Scouseness.

This study has also highlighted the relationship between identities, place and values, namely how voters in Merseyside are more populist than England as a whole. In regression analyses not presented here, both Scouseness and living in Merseyside are related to higher levels of populism than in England as a whole, and if broken down by local authority the relationship holds across all boroughs except Sefton - the area with the lowest level of Scouse identifiers. This suggests that the political environment in Merseyside is more populist than elsewhere, fuelled by the inherent anti-elite/anti-government sentiment present in Liverpool at the time when the Scouse identity was being politicised. This also goes some way to explaining why a Labour vote is correlated with an increased level of populism and a Conservative vote is associated with lower levels of populism, the opposite of what is seen across the rest of England.

This analysis raises further questions in the study of British politics. Firstly, which local identities are politically salient? Secondly, why are some local identities politically salient whilst others are not? And thirdly, how do these local identities interact with other identities? Hopefully, this study of the Scouse identity is only the beginning for the study of the political importance of local identities.

Bibliography

Academic Sources

Abramson, Paul R. 1972. “Intergenerational Social Mobility and Partisan Choice.” American Political Science Review 66 (4): 1291–1294. doi:10.2307/1957180.

Achen, Christopher, and Larry Bartels. 2017. Democracy for Realists: Why Elections Do Not Produce Responsive Government. ebook. Princeton and Oxford: Princeton University Press. https://doi.org/10.1515/9781400888740.

Bory, Stéphanie. 2020. “Welsh Labour: From ‘Fantastic Results’ in June 2017 to the ‘Crumbling of the Red Wall’ in December 2019.” Revue Française de Civilisation Britannique XXV (3). doi:10.4000/rfcb.5948.

Butler, D., and D. Stokes. 1974. Political Change in Britain. Second. London and Basingstoke: Macmillan Press LTD.

Calvert Jump, Robert, and Jo Michell. 2021. “Did ‘Left Behind’ Voters Cause Brexit?” UK in a Changing Europe. https://ukandeu.ac.uk/did-left-behind-voters-cause-brexit/.

Clarke, Harold D., David Sanders, Marianne C. Stewart, and Paul Whiteley. 2004. Political Choice in Britain. Oxford University Press. doi:10.1093/019924488X.001.0001.

Cooper, Luke, and Christabel Cooper. 2020. “‘Get Brexit Done’: The New Political Divides of England and Wales at the 2019 Election.” The Political Quarterly 91 (4): 751–761. doi:10.1111/1467-923X.12918.

Crewe, Ivor. 1983. “The Electorate: Partisan Dealignment Ten Years On.” West European Politics 6 (4): 183–215. doi:10.1080/01402388308424445.

Dunleavy, Patrick. 2005. “Facing Up to Multi-Party Politics: How Partisan Dealignment and PR Voting Have Fundamentally Changed Britain’s Party Systems.” Parliamentary Affairs 58 (3): 503–532. doi:10.1093/pa/gsi049.

Elster, Jon. 1989. “Social Norms and Economic Theory.” Journal of Economic Perspectives 3 (4): 99–117. doi:10.1257/jep.3.4.99.

Fearon, James D. 1999. “What Is Identity (as We Now Use the Word).” Unpublished Manuscript, Stanford University, Stanford, Calif. http://www.academia.edu/download/39722/758yoziw9eo36uov7hv.pdf.

Foos, Florian, and Daniel Bischof. 2022. “Tabloid Media Campaigns and Public Opinion: Quasi-Experimental Evidence on Euroscepticism in England.” American Political Science Review 116 (1): 19–37. doi:10.1017/S000305542100085X.

Furlong, Jamie. 2019. “The Changing Electoral Geography of England and Wales: Varieties of ‘Left-Behindedness’.” Political Geography 75 (November): 102061. doi:10.1016/j.polgeo.2019.102061.

Goodhart, David. 2017a. The Road to Somewhere: The Populist Revolt and the Future of Politics. London: Hurst & Company.

Goodhart, David. 2017b. “Don’t Blame Me for May’s Election Woes.” Catholic Herald. https://catholicherald.co.uk/dont-blame-me-for-mays-election-woes/.

Hadfield, Charlotte. 2019. “’Scouse Not English’ Goes Viral as Merseyside Stands Defiant.” Liverpool Echo. https://www.liverpoolecho.co.uk/news/liverpool-news/scouse-not-english-how-merseyside-17420602.

Henderson, Ailsa. 2021. “England and the Union.” Discover Society. https://discoversociety.org/2021/06/08/england-and-the-union/.

Henderson, Ailsa, Ed Gareth Poole, Richard Wyn Jones, Daniel Wincott, Jac Larner, and Charlie Jeffery. 2021. “Analysing Vote-Choice in a Multinational State: National Identity and Territorial Differentiation in the 2016 Brexit Vote.” Regional Studies 55 (9): 1502–1516. doi:10.1080/00343404.2020.1813883.

Henderson, Ailsa, and Richard Wyn Jones. 2021. Englishness: The Political Force Transforming Britain. First. Oxford University Press. doi:10.1093/oso/9780198870784.001.0001.

Hutchings, Vincent, and Hakeem Jefferson. 2018. “The Sociological and Social-Psychological Approaches.” In The Routledge Handbook of Elections, Voting Behavior and Public Opinion, edited by Justin Fisher, Edward Fieldhouse, Mark Franklin, Rachel Gibson, Marta Cantijoch, and Christopher Wlezien, 21–29. London ; New York, NY: Routledge.

Ipsos. 2021. “Both Remainers and Leavers retain a strong Brexit identity 5 years on.” Ipsos. https://www.ipsos.com/en-uk/both-remainers-and-leavers-retain-strong-brexit-identity-5-years.

Jeffery, David. forthcoming. The Strange Death of Tory Liverpool. Liverpool: Liverpool University Press.

Jeffery, David. 2017. “The Strange Death of Tory Liverpool: Conservative Electoral Decline in Liverpool, 1945.” British Politics 12 (3): 386–407. doi:10.1057/s41293-016-0032-6.

Jeffery, David. 2021. “The Myth of ’Scouse Not English’ - Scouse Identity: History, Mystery, and Political Relevance Today.” https://www.youtube.com/watch?v=0H3Mw-mM-QY.

Jeffery, David. 2022. “Understanding Drivers of Support for English City-Region Devolution: A Case Study of the Liverpool City Region.” British Politics, January. doi:10.1057/s41293-021-00199-z.

Jennings, Will, and Gerry Stoker. 2016. “The Bifurcation of Politics: Two Englands.” The Political Quarterly 87 (3): 372–382. doi:10.1111/1467-923X.12228.

Jennings, Will, and Gerry Stoker. 2017. “Tilting Towards the Cosmopolitan Axis? Political Change in England and the 2017 General Election.” The Political Quarterly 88 (3): 359–369. doi:10.1111/1467-923X.12403.

Kanagasooriam, James, and Elizabeth Simon. 2021. “Red Wall: The Definitive Description.” Political Insight 12 (3): 8–11. doi:10.1177/20419058211045127.

Keating, Michael, and Nicola McEwen. 2020. “The Independence Referendum of 2014: Issues and Outcome.” In The Oxford Handbook of Scottish Politics, 649–674. Oxford University Press. doi:10.1093/oxfordhb/9780198825098.013.34.

Kenny, Michael. 2014. The Politics of English Nationhood. Oxford University Press. doi:10.1093/acprof:oso/9780199608614.001.0001.

Larner, Jac. 2019. “How Brexit Reinvigorated the Welsh Independence Movement.” LSE Brexit. https://blogs.lse.ac.uk/brexit/2019/09/27/how-brexit-reinvigorated-the-welsh-independence-movement/.

Lazarsfeld, P. F., B. Berelson, and H. Gaudet. 1944. The People’s Choice. New York, N.Y: Columbia University Press.

Leyshon, Andrew. 2021. “Economic Geography I: Uneven Development, ‘Left Behind Places’ and ‘Levelling up’ in a Time of Crisis.” Progress in Human Geography 45 (6): 1678–1691. doi:10.1177/03091325211011684.

Martin, Nicole, Maria Sobolewska, and Neema Begum. 2019. “Left Out of the Left Behind: Ethnic Minority Support for Brexit.” SSRN Electronic Journal. doi:10.2139/ssrn.3320418.

May, Theresa. 2016. “Theresa May 2016 Speech at Conservative Party Conference.” https://www.ukpol.co.uk/theresa-may-2016-speech-at-conservative-party-conference/.

McCrone, David. 2020. “Nationality and National Identity.” In The Oxford Handbook of Scottish Politics, 21–41. Oxford University Press. doi:10.1093/oxfordhb/9780198825098.013.2.

Monroe, Kristen Renwick, James Hankin, and Renée Bukovchik Van Vechten. 2000. “The Psychological Foundations of Identity Politics.” Annual Review of Political Science 3 (1): 419–447. doi:10.1146/annurev.polisci.3.1.419.

Moon, David S. 2017. “Red Dragon FM: Carwyn Jones’s ‘Welsh Labour Rhetoric’.” In Voices of the UK Left, edited by Judi Atkins and John Gaffney, 123–146. Cham: Springer International Publishing. doi:10.1007/978-3-319-51902-9_7.

Pulzer, Peter. 1967. Political Representation and Elections: Parties and Voting in Great Britain. New York: Praeger.

Simpson, Mollie. 2022. “Bored of Publishing Clickbait, the Liverpool Echo Moves to Crush Its Competition.” The Post. https://www.livpost.co.uk/p/bored-of-publishing-clickbait-the.

Sobolewska, Maria, and Robert Ford. 2020. Brexitland: Identity, Diversity and the Reshaping of British Politics. First. Cambridge University Press. doi:10.1017/9781108562485.

Stay European. 2021. “Census 2021: Declare Yourself European .” Stay European. https://www.stayeuropean.org/census/.

Tajfel, H., and J. C. Turner. 1979. “An Integrative Theory of Intergroup Conflict.” In The Social Psychology of Intergroup Relations, edited by W. G. Austin and S. Worchel. Monterey, CA: Brooks/Cole.

Wellings, Ben. 2012. English Nationalism and Euroscepticism. Oxford: Peter Lang.

Wilks-Heeg, Stuart. 2019. “Safe Labour Suburbia? The Changing Politics of the Merseyside Suburbs.” The Political Quarterly 90 (1): 53–63. doi:10.1111/1467-923X.12528.

Wyn Jones, Richard. 2019. “Divided Wales.” Cardiff University Politics & Governance Blog. https://web.archive.org/web/20210413161615/https://blogs.cardiff.ac.uk/brexit/2019/09/23/divided-wales/.

Wyn Jones, Richard, and Jac Larner. 2021. “What about Wales? Brexit and the Future of the UK.” Discover Society: New Series 1 (2). doi:10.51428/dsoc.2021.02.0004.

Wyn Jones, Richard, Guy Lodge, Ailsa Henderson, and Daniel Wincott. 2012. The Dog That Finally Barked: England as an Emerging Political Community. London: Institute for Public Policy Research.

Wyn Jones, Richard, and Roger Scully. 2006. “Devolution and Electoral Politics in Scotland and Wales.” Publius: The Journal of Federalism 36 (1): 115–134. doi:10.1093/publius/pji034.

YouGov. 2014. “YouGov / The Times Survey Results.” YouGov. http://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/fyv6pya3md/Times-results-141030-scotland-FULL.pdf.

YouGov. 2019a. “How Britain Voted in the 2019 General Election | YouGov.” YouGov. https://yougov.co.uk/topics/politics/articles-reports/2019/12/17/how-britain-voted-2019-general-election.

YouGov. 2019b. “YouGov / The Times Survey Results.” YouGov. https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/9f359xvq6w/TheTimes_191206_Scotland_Results_IndependenceW.pdf.

Software

Francois, R. (2020). bibtex: Bibtex Parser. R package version 0.4.2.3. <URL: https://github.com/romainfrancois/bibtex>.

Gohel, D. (2022). flextable: Functions for Tabular Reporting. R package version 0.7.0. <URL: https://CRAN.R-project.org/package=flextable>.

Henry, L. and H. Wickham (2020). purrr: Functional Programming Tools. R package version 0.3.4. <URL: https://CRAN.R-project.org/package=purrr>.

Hugh-Jones, D. (2021). huxtable: Easily Create and Style Tables for LaTeX, HTML and Other Formats. R package version 5.4.0. <URL: https://hughjonesd.github.io/huxtable/>.

Müller, K. and H. Wickham (2021). tibble: Simple Data Frames. R package version 3.1.6. <URL: https://CRAN.R-project.org/package=tibble>.

Marwick, B. (2022). wordcountaddin: Word counts and readability statistics in R markdown documents. R package version 0.3.0.9000.

McLean, M. W. (2014). Straightforward Bibliography Management in R Using the RefManager Package. arXiv: 1403.2036 [cs.DL]. <URL: https://arxiv.org/abs/1403.2036>.

—–— (2017). “RefManageR: Import and Manage BibTeX and BibLaTeX References in R”. In: The Journal of Open Source Software. DOI: 10.21105/joss.00338.

—–— (2020). RefManageR: Straightforward BibTeX and BibLaTeX Bibliography Management. R package version 1.3.0. <URL: https://github.com/ropensci/RefManageR/>.

Pedersen, T. L. (2021). ggforce: Accelerating ggplot2. R package version 0.3.3. <URL: https://CRAN.R-project.org/package=ggforce>.

R Core Team (2022). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Vienna, Austria. <URL: https://www.R-project.org/>.

Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. ISBN: 978-3-319-24277-4. <URL: https://ggplot2.tidyverse.org>.

—–— (2019). stringr: Simple, Consistent Wrappers for Common String Operations. R package version 1.4.0. <URL: https://CRAN.R-project.org/package=stringr>.

—–— (2021a). forcats: Tools for Working with Categorical Variables (Factors). R package version 0.5.1. <URL: https://CRAN.R-project.org/package=forcats>.

—–— (2021b). tidyverse: Easily Install and Load the Tidyverse. R package version 1.3.1. <URL: https://CRAN.R-project.org/package=tidyverse>.

Wickham, H., M. Averick, J. Bryan, et al. (2019). “Welcome to the tidyverse”. In: Journal of Open Source Software 4.43, p. 1686. DOI: 10.21105/joss.01686.

Wickham, H., W. Chang, L. Henry, et al. (2021). ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics. R package version 3.3.5. <URL: https://CRAN.R-project.org/package=ggplot2>.

Wickham, H., R. François, L. Henry, et al. (2022). dplyr: A Grammar of Data Manipulation. R package version 1.0.8. <URL: https://CRAN.R-project.org/package=dplyr>.

Wickham, H. and M. Girlich (2022). tidyr: Tidy Messy Data. R package version 1.2.0. <URL: https://CRAN.R-project.org/package=tidyr>.

Wickham, H., J. Hester, and J. Bryan (2022). readr: Read Rectangular Text Data. R package version 2.1.2. <URL: https://CRAN.R-project.org/package=readr>.

Xie, Y. (2016). bookdown: Authoring Books and Technical Documents with R Markdown. ISBN 978-1138700109. Boca Raton, Florida: Chapman and Hall/CRC. <URL: https://bookdown.org/yihui/bookdown>.

—–— (2022). bookdown: Authoring Books and Technical Documents with R Markdown. R package version 0.26. <URL: https://CRAN.R-project.org/package=bookdown>.

Zhu, H. (2021). kableExtra: Construct Complex Table with kable and Pipe Syntax. R package version 1.3.4. <URL: https://CRAN.R-project.org/package=kableExtra>.

Appendix

Table 5: Regression models for rejoining the EU by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	0.38	0.38	0.10	-0.07	-0.05
	(0.26)	(0.26)	(0.19)	(0.15)	(0.05)
Age	-0.33 ***	-0.33 ***	-0.18 *	-0.25 ***	-0.24 ***
	(0.09)	(0.09)	(0.07)	(0.06)	(0.02)
ABC1	0.22	0.22	0.40 **	0.03	0.17 ***
	(0.19)	(0.19)	(0.15)	(0.12)	(0.04)
Income	0.07	0.07	0.09	0.08	0.07 ***
	(0.08)	(0.08)	(0.07)	(0.05)	(0.02)
LR scale	0.14	0.16	0.76 ***	0.77 ***	0.81 ***
	(0.22)	(0.22)	(0.13)	(0.11)	(0.03)
AL scale	-0.56 *	-0.57 **	-0.29 *	-0.32 **	-0.48 ***
	(0.22)	(0.22)	(0.13)	(0.10)	(0.03)
Populism	-0.21	-0.21	-0.89 ***	-0.71 ***	-0.63 ***
	(0.23)	(0.23)	(0.15)	(0.13)	(0.04)
Equality	-0.65 ***	-0.66 ***	-0.49 ***	-0.42 ***	-0.40 ***
	(0.17)	(0.17)	(0.13)	(0.11)	(0.03)
Immigration	0.26 ***	0.26 ***	0.50 ***	0.51 ***	0.54 ***
	(0.07)	(0.07)	(0.07)	(0.06)	(0.02)
English-Scouse	0.59 *
	(0.27)
British-Scouse		0.48
		(0.27)
British-Welsh			0.06
			(0.19)
British-Scottish				0.89 ***
				(0.16)
British-English					-0.47 ***
					(0.07)
N	469	469	882	1431	14088
AIC	447.99	449.83	763.46	1195.40	12099.15
BIC	493.65	495.48	816.06	1253.33	12182.24
Pseudo R2	0.42	0.41	0.56	0.54	0.55
*** p < 0.001; ** p < 0.01; * p < 0.05.

Table 6: Regression models for voting Labour by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	-0.36	-0.37	-0.15	-0.13	-0.23 ***
	(0.21)	(0.21)	(0.16)	(0.15)	(0.04)
Age	-0.23 ***	-0.24 ***	-0.18 **	-0.25 ***	-0.30 ***
	(0.07)	(0.07)	(0.06)	(0.05)	(0.01)
ABC1	-0.28	-0.27	-0.11	-0.21	-0.12 ***
	(0.16)	(0.16)	(0.13)	(0.11)	(0.03)
Income	0.03	0.03	0.04	0.03	-0.05 ***
	(0.06)	(0.06)	(0.06)	(0.05)	(0.01)
LR scale	0.39 *	0.41 *	0.89 ***	0.66 ***	0.98 ***
	(0.16)	(0.16)	(0.12)	(0.11)	(0.03)
AL scale	-0.03	-0.05	-0.30 **	0.13	-0.24 ***
	(0.16)	(0.16)	(0.12)	(0.10)	(0.03)
Populism	0.50 **	0.48 **	-0.01	-0.25 *	-0.05
	(0.18)	(0.18)	(0.12)	(0.11)	(0.03)
Equality	-0.33 *	-0.35 **	-0.32 **	-0.13	-0.30 ***
	(0.13)	(0.13)	(0.11)	(0.10)	(0.03)
Immigration	0.10	0.11	0.13 *	0.05	0.19 ***
	(0.07)	(0.07)	(0.06)	(0.06)	(0.02)
English-Scouse	0.77 ***
	(0.22)
British-Scouse		0.67 **
		(0.22)
British-Welsh			0.20
			(0.16)
British-Scottish				-0.85 ***
				(0.14)
British-English					-0.03
					(0.07)
N	562	562	907	1456	14577
AIC	646.82	650.10	971.44	1305.48	14526.74
BIC	694.47	697.74	1024.36	1363.60	14610.20
Pseudo R2	0.25	0.24	0.32	0.11	0.33
*** p < 0.001; ** p < 0.01; * p < 0.05.

Table 7: Regression models for voting Conservative by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	-0.44	-0.38	0.03	0.21	0.00
	(0.29)	(0.29)	(0.18)	(0.16)	(0.04)
Age	0.18	0.19 *	0.21 **	0.27 ***	0.27 ***
	(0.09)	(0.09)	(0.07)	(0.06)	(0.02)
ABC1	0.52 *	0.49 *	-0.11	0.13	0.03
	(0.21)	(0.21)	(0.13)	(0.13)	(0.03)
Income	0.11	0.12	-0.00	0.02	0.05 ***
	(0.09)	(0.09)	(0.06)	(0.05)	(0.01)
LR scale	-0.40 *	-0.41 *	-1.06 ***	-1.22 ***	-1.26 ***
	(0.20)	(0.21)	(0.12)	(0.12)	(0.03)
AL scale	0.59 *	0.58 *	0.60 ***	0.47 ***	0.75 ***
	(0.24)	(0.24)	(0.13)	(0.12)	(0.03)
Populism	-0.90 ***	-0.86 ***	-0.07	0.37 **	0.11 **
	(0.25)	(0.25)	(0.14)	(0.13)	(0.04)
Equality	0.28	0.32	0.49 ***	0.26 *	0.34 ***
	(0.18)	(0.18)	(0.12)	(0.12)	(0.03)
Immigration	-0.08	-0.09	-0.13 *	-0.40 ***	-0.29 ***
	(0.09)	(0.09)	(0.06)	(0.06)	(0.02)
English-Scouse	-1.33 ***
	(0.30)
British-Scouse		-1.25 ***
		(0.30)
British-Welsh			-0.54 **
			(0.18)
British-Scottish				-1.57 ***
				(0.17)
British-English					0.42 ***
					(0.07)
N	562	562	907	1456	14577
AIC	409.76	412.03	862.63	1092.16	13403.61
BIC	457.40	459.68	915.54	1150.28	13487.06
Pseudo R2	0.32	0.32	0.44	0.53	0.48
*** p < 0.001; ** p < 0.01; * p < 0.05.

Table 8: Regression models for voting Green by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	0.96 *	1.02 *	-1.02	-0.52	-0.00
	(0.45)	(0.45)	(0.87)	(0.61)	(0.10)
Age	0.11	0.12	-0.30	-0.41 *	-0.13 ***
	(0.15)	(0.15)	(0.28)	(0.20)	(0.03)
ABC1	-0.02	-0.01	-0.32	1.09	0.03
	(0.33)	(0.33)	(0.59)	(0.78)	(0.08)
Income	-0.37 *	-0.38 **	0.22	-0.05	-0.10 **
	(0.14)	(0.14)	(0.24)	(0.18)	(0.03)
LR scale	0.10	0.07	0.17	-0.27	0.05
	(0.32)	(0.32)	(0.48)	(0.44)	(0.07)
AL scale	-1.16 ***	-1.15 ***	0.14	-0.84	-0.26 ***
	(0.34)	(0.34)	(0.52)	(0.45)	(0.07)
Populism	-0.22	-0.18	-0.49	-0.10	0.14
	(0.36)	(0.37)	(0.55)	(0.42)	(0.08)
Equality	-0.03	-0.02	-0.95	-0.92 *	-0.17 *
	(0.30)	(0.30)	(0.50)	(0.43)	(0.07)
Immigration	0.12	0.10	-0.30	-0.41	0.17 ***
	(0.16)	(0.16)	(0.28)	(0.27)	(0.04)
English-Scouse	-1.07 *
	(0.49)
British-Scouse		-1.03 *
		(0.49)
British-Welsh			-0.34
			(0.74)
British-Scottish				-0.34
				(0.57)
British-English					0.64 ***
					(0.18)
N	562	562	907	1456	14577
AIC	199.93	200.35	102.77	142.67	3716.59
BIC	247.57	248.00	155.68	200.79	3800.05
Pseudo R2	0.21	0.21	0.12	0.20	0.05
*** p < 0.001; ** p < 0.01; * p < 0.05.

Table 9: Regression models for voting Liberal Democrat by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	0.44	0.45	0.25	-0.10	0.12 *
	(0.41)	(0.41)	(0.26)	(0.19)	(0.05)
Age	0.18	0.16	0.10	-0.03	0.09 ***
	(0.14)	(0.14)	(0.10)	(0.07)	(0.02)
ABC1	0.14	0.15	0.28	0.09	0.15 **
	(0.33)	(0.33)	(0.22)	(0.16)	(0.05)
Income	0.08	0.08	-0.02	0.00	0.09 ***
	(0.13)	(0.13)	(0.09)	(0.06)	(0.02)
LR scale	-0.46	-0.42	0.15	-0.28 *	-0.02
	(0.31)	(0.31)	(0.17)	(0.12)	(0.04)
AL scale	0.54	0.50	0.19	0.02	-0.05
	(0.32)	(0.32)	(0.18)	(0.13)	(0.04)
Populism	-0.50	-0.50	-0.29	-0.47 **	-0.42 ***
	(0.35)	(0.35)	(0.19)	(0.15)	(0.04)
Equality	-0.15	-0.15	0.07	0.29 *	0.10 *
	(0.28)	(0.28)	(0.18)	(0.14)	(0.04)
Immigration	0.28	0.27	0.42 ***	0.29 ***	0.31 ***
	(0.15)	(0.15)	(0.10)	(0.08)	(0.02)
English-Scouse	0.44
	(0.42)
British-Scouse		0.23
		(0.42)
British-Welsh			-0.66 *
			(0.26)
British-Scottish				-0.77 ***
				(0.19)
British-English					-0.23 **
					(0.08)
N	562	562	907	1456	14577
AIC	238.81	239.60	479.08	858.92	9997.94
BIC	286.46	287.24	531.99	917.04	10081.40
Pseudo R2	0.09	0.09	0.10	0.10	0.11
*** p < 0.001; ** p < 0.01; * p < 0.05.

Table 10: Regression models for voting UKIP by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	0.38	0.36	-0.28	-13.89	0.29
	(0.55)	(0.55)	(0.75)	(22837.71)	(0.22)
Age	-0.37	-0.36	0.10	-5.04	-0.28 ***
	(0.21)	(0.21)	(0.30)	(5778.15)	(0.08)
ABC1	-0.20	-0.19	-0.44	-12.83	-0.12
	(0.40)	(0.40)	(0.59)	(14693.01)	(0.16)
Income	-0.11	-0.11	-0.08	3.01	-0.12
	(0.16)	(0.16)	(0.25)	(7247.97)	(0.07)
LR scale	-0.12	-0.14	-0.24	12.06	0.13
	(0.40)	(0.39)	(0.43)	(21283.86)	(0.14)
AL scale	0.58	0.59	-0.49	3.18	0.08
	(0.44)	(0.45)	(0.61)	(7028.02)	(0.16)
Populism	0.04	0.04	1.22	3.68	0.68 ***
	(0.46)	(0.46)	(0.66)	(13461.99)	(0.18)
Equality	0.48	0.47	-0.00	-5.83	0.24
	(0.33)	(0.33)	(0.43)	(21586.09)	(0.14)
Immigration	-0.29	-0.28	-0.71 *	-4.78	-0.35 ***
	(0.17)	(0.17)	(0.32)	(8005.52)	(0.08)
English-Scouse	-0.01
	(0.56)
British-Scouse		0.14
		(0.59)
British-Welsh			-2.01 *
			(0.82)
British-Scottish				-31.26
				(13055.33)
British-English					0.59 *
					(0.30)
N	562	562	907	1456	14577
AIC	152.68	152.62	96.97	22.00	1039.74
BIC	200.33	200.27	149.89	80.12	1123.19
Pseudo R2	0.16	0.16	0.33	1.00	0.12
*** p < 0.001; ** p < 0.01; * p < 0.05.

Table 11: Regression models for voting Reform UK by RTI

	Scouse (English)	Scouse	Welsh	Scottish	English
Male	-0.19	-0.17	-0.27	0.19	0.06
	(0.60)	(0.60)	(0.33)	(0.68)	(0.12)
Age	0.53 **	0.56 **	-0.00	-0.20	0.05
	(0.20)	(0.20)	(0.13)	(0.23)	(0.05)
ABC1	0.88	0.87	-0.13	0.75	-0.07
	(0.47)	(0.47)	(0.25)	(0.60)	(0.09)
Income	-0.16	-0.18	-0.04	-0.18	-0.14 ***
	(0.18)	(0.18)	(0.11)	(0.22)	(0.04)
LR scale	-0.65	-0.73 *	0.09	-0.20	0.12
	(0.36)	(0.36)	(0.21)	(0.43)	(0.08)
AL scale	0.32	0.38	0.20	0.57	0.20 *
	(0.49)	(0.50)	(0.26)	(0.50)	(0.09)
Populism	1.22 *	1.20 *	0.28	0.80	0.45 ***
	(0.49)	(0.49)	(0.26)	(0.52)	(0.10)
Equality	0.29	0.25	0.16	0.70	0.19 *
	(0.31)	(0.32)	(0.20)	(0.45)	(0.08)
Immigration	-0.34	-0.37	-0.42 ***	0.25	-0.32 ***
	(0.19)	(0.19)	(0.12)	(0.22)	(0.04)
English-Scouse	-0.42
	(0.56)
British-Scouse		0.13
		(0.57)
British-Welsh			-0.37
			(0.32)
British-Scottish				-0.26
				(0.64)
British-English					0.26
					(0.16)
N	562	562	907	1456	14577
AIC	134.19	134.68	334.62	128.54	2845.00
BIC	181.83	182.33	387.53	186.66	2928.46
Pseudo R2	0.33	0.32	0.17	0.11	0.13
*** p < 0.001; ** p < 0.01; * p < 0.05.

---

1. These figures exclude those who respond ‘don’t know’ or ‘would not vote’.↩︎

2. These figures exclude those who respond ‘don’t know’ or ‘would not vote’.↩︎

3. At the time of writing, 52,958 people had pledged to write in ‘European’ in the census, markedly lower than the over 390,000 people in England and Wales (0.7 percent of the population), who reported their religion as Jedi in the 2011 census. The political salience of the Jedi is also a sadly understudied area.↩︎

4. For an attempt across the North of England, see Jeffery (forthcoming, chap. 8)↩︎

5. The Liverpool Echo is nominally the local newspaper but under Reach PLC it is increasingly an advertising company with some journalists attached (Simpson 2022)↩︎