The 2019 UK General Election had seismic consequences for British politics. After three years of political turmoil following the 2016 referendum on Britain’s membership of the European Union (EU), the 2019 election marked a victory for the Leave side of the Brexit debate, putting to rest questions of a second referendum and any chance of Parliament blocking the Withdrawal Bill. The United Kingdom left the EU on January 31, 2020. Although there were clear consequences for Britain’s EU membership, there is debate about whether 2019 was a “Brexit election” (Prosser Reference Prosser2020)—even a critical election (Green Reference Green, Allen and Bartle2021)—or the continuation of long-term realignments in British politics (Cutts et al. Reference Cutts, Goodwin, Heath and Surridge2020; Jennings and Stoker Reference Jennings and Stoker2017). By most accounts, Brexit dominated the 2019 election as a political issue, but whether this represents a key moment in a process of realignment of voters in Britain remains to be seen.
From the perspective of Electoral Shocks (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020), the question of whether the 2019 election was a Brexit election or “a continuation of longer-term trends in dealignment” (Cutts et al. Reference Cutts, Goodwin, Heath and Surridge2020, 8) poses a false dichotomy. According to the theory of electoral shocks, electoral change results from the interplay of key short-term events and long-term trends in the electoral context. Two such trends make British electoral politics particularly vulnerable to shocks: (1) the declining attachment of British voters to political parties, and (2) an increase in electoral support for smaller parties. Together, these trends have rendered the British electorate more volatile—individual-level volatility followed an upward trend from the 1960s until 2015 (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020)—and thus more susceptible to the disruption caused by major political events. Importantly, election outcomes—reflecting parties’ gains and losses from voter volatility—depend on the response of political actors to electoral shocks. With respect to Brexit, this suggests that we should not ask whether the 2019 result was the result of Brexit or long-term shifts in alignment but instead—in the context of an unstable and dealigned electorate—how did Brexit affect the outcome of the election and are the resulting patterns of electoral behavior suggestive of realignment?
This article addresses these questions and, although it is premature to announce a stable realignment of British politics, we assess the extent to which recent electoral change is consistent with realignment. Indicators of fundamental shifts that would signal realignment include a surge in volatility followed by a restabilization, substantial shifts in the social and geographical bases of support, and the emergence of new political identities based on the new electoral cleavage (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020; Green, Palmquist, and Schickler Reference Green, Palmquist and Schickler2002; Key Reference Key1955; Mayhew Reference Mayhew2000).
A BREXIT ELECTION?
Prime Minister Boris Johnson called the 2019 General Election with a pledge to “get Brexit done.” Having been forced to request an extension to the Brexit negotiations by Parliament, Johnson eventually gained parliamentary assent for an early General Election in December, all but ensuring that the election would be primarily about Brexit. The two major parties took very different positions on Brexit: the Conservatives were committed to leaving the EU on January 31, 2020, whereas Labour promised a second referendum, although trying to maintain an ambiguous position on how it would campaign in such a referendum. Other options existed for Leave voters (i.e., the Brexit Party) and Remain voters (i.e., the Liberal Democrats, the Scottish National Party [SNP], Plaid Cymru, and the Greens) and there were formal and less formal electoral pacts along Brexit lines. On the Leave side, the Brexit Party decided not to stand candidates against incumbent Conservative Members of Parliament, minimizing the chance of splitting the Leave vote in Conservative-held seats. The Liberal Democrats, Greens, and Plaid Cymru formed the “Unite to Remain” alliance; however, neither of these strategic arrangements had a significant effect on the outcome of the election (Mellon Reference Mellon2022). The campaign focused heavily on the question of the United Kingdom’s withdrawal for the EU despite attempts by Labour to move the campaign toward other issues on which it was less conflicted (Cutts et al. Reference Cutts, Goodwin, Heath and Surridge2020; Prosser Reference Prosser2020). The British Election Study (BES) Internet Panel (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Bailey, Schmitt and van der Eijk2021, Reference Fieldhouse, Evans, Green, Mellon, Prosser and Bailey2023) indicated that Brexit was the most important issue to voters throughout the campaign. In the post-election wave of the panel, 52% of respondents gave a Brexit-related answer to the question of what the most important issue was facing the country.Footnote 1
The election delivered a decisive victory for the Conservatives with 43.6% of the vote and 365 seats, a clear majority of 80 seats. Labour won only 32.1% of the vote and 202 seats, a decrease of 60 seats since the 2017 election. The Brexit Party won only 2% of the vote and no seats, and the Liberal Democrats increased their vote share to 11.6% (up from 7.4%), winning 11 seats (one less than in 2017).
BREXIT—AN ELECTORAL SHOCK
To understand the Brexit referendum vote and its consequences for British politics, we must place it in the context of partisan dealignment and changing-issue salience and understand how these interacted with changes in party positions. The period of New Labour government around the turn of the century witnessed a lessening of perceived differences between the major parties on traditional economic issues (Green and Hobolt Reference Green and Hobolt2008). The early 2000s also marked an increase in the salience of immigration and a decline in the perceived ability of the major parties to control it, particularly among voters with more socially conservative values and a more ethnocentric outlook (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020; Sobolewska and Ford Reference Sobolewska and Ford2020). Together, this created a widening gap between the more liberal and highly educated cosmopolitan voters and those “left behind” by globalization (Jennings and Stoker Reference Jennings and Stoker2017), paving the way for the rise of the UK Independence Party (UKIP) (Evans and Mellon Reference Evans and Mellon2019; Ford and Goodwin Reference Ford and Goodwin2014). As the issues of immigration and EU membership became more closely entwined following the expansion of the EU in 2004, first Labour and then the Conservative–Liberal Democrat coalition governments were unable to control immigration from the EU (Evans and Mellon Reference Evans and Mellon2019). With the Conservatives facing competition on the right from UKIP, the Conservative government promised a referendum on EU membership, resulting in the vote to leave in 2016. However, it was the strategic positioning of the parties after the referendum that accelerated the electoral sorting of the major parties by Brexit preferences, with the Conservatives claiming to be the party of Brexit (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020). This is illustrated in figure 1, which shows the proportion of each of the Conservative Party’s and the Labour Party’s support composed of Leavers and Remainers, respectively. Before the referendum, there was an approximately equal proportion of Leavers and Remainers among Conservative supporters and a modest majority of Remainers among Labour voters, but this shifted decisively after the referendum and continued thereafter.
Percentage of Conservative and Labour Party Supporters Who Would Vote for Leave and Remain, Respectively
Notes: Each series shows current EU Vote intention before referendum and hypothetical second referendum vote afterwards. “Don’t knows” and “would not vote” are included in the denominator. Source: British Election Study Internet Panel, waves 1–24.
VOLATILITY AND BREXIT
The 2019 election experienced a relatively low level of aggregate volatility,Footnote 2 returning to a level witnessed in most historical elections in the United Kingdom (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020, 11). More important for our understanding of electoral change, however, figure 2 shows that there also was a substantial decrease in individual-level volatility compared to the two previous General Elections.
Individual-Level Voter Volatility in 12 Pairs of Elections, 1964–2019
Source: British Election Studies, 1964–2019.
This decrease in volatility could indicate a return to relative stability based on new alignments around Brexit. However, changes to volatility also reflect long-term trends in the electoral context, including the rise and fall of smaller parties and the weakening of partisan attachments (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020). On this basis, we might have expected a decrease in volatility due to the strong performance of both the Conservatives and Labour in 2017 and a relatively stable level of partisan identification. In total, this would have predicted a 1.7-percentage-point decrease in vote switching in 2019.
Individual volatility had peaked in 2015 with the collapse of the Liberal Democrat vote following coalition government, the rise of the SNP in Scotland and the UKIP in England, and the overall increase in vote share of smaller parties across the country. This volatility was an important steppingstone in the subsequent sorting along the “cultural dimension” (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020; Surridge Reference Surridge2018). In particular, many of those who voted UKIP in 2015 subsequently switched to the Conservatives in 2017 or 2019 (many via the Brexit Party), which contributed to the sorting pattern shown in figure 1 (Evans, de Geus, and Green Reference Evans, de Geus and Green2021). Volatility was lower in 2017 than 2015 but nevertheless was historically very high, with much of the switching driven by Brexit preferences (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020). By 2019, because the sorting process had almost reached saturation point, some reduction in volatility was almost inevitable unless the Brexit divide were to unwind. Most Leavers already had switched to the Conservatives and most Remainers to one of the pro-EU parties. Although volatility was at a lower level than in the previous two elections, switching in 2019 nevertheless was structured by Brexit. There is evidence of this in figure 3, which shows how Leave and Remain voters changed their votes between the 2017 and 2019 General Elections.
Flow of Vote 2017–2019 by Brexit Vote
Crucially, Labour lost almost a third (32%) of its Leave voters to the Conservatives, costing numerous seats in Leave-voting constituencies. Labour losses to the Brexit Party (5%) were modest and outweighed by desertions from the Conservatives to the Brexit Party. However, although 2019 accelerated the sorting process, Labour’s defeat was not attributable solely to its loss of Leave voters. Labour also failed to win a sufficient proportion of the votes of Remainers, which reflected the unpopularity of leader Jeremy Corbyn and the lack of a clear commitment to oppose Brexit compared to the Liberal Democrats, Greens, SNP, and Plaid Cymru (Labour Together 2020). Although Labour Leave voters were more likely to defect than Labour Remain voters, because the latter group was significantly larger, in absolute numbers, Labour lost as many Remainers as Leavers. That is, half of the 28% of 2017 Labour voters who defected to another party in 2019 were Remain voters.
Figure 3 also shows how the Conservatives consolidated their position as the party of Brexit. Between the referendum and the 2017 General Election, there was a sharp change in perceptions about where the Conservatives stood on Europe. In particular, Leave voters increasingly regarded the Conservatives as an anti-EU party (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020, 170), whereas Labour’s perceived position as a pro-EU party remained stable. These perceptions of party image continued to diverge going into the 2019 General Election. During the campaign, the mean placement of the Conservatives on a 10-point EU scale (where 10 is the most anti-EU) was 7.6, compared to 7.0 during the 2017 campaign. Labour moved in the opposite direction, from 3.8 during the 2017 campaign to 3.3 in 2019. With the collapse of UKIP and the failure of the Brexit Party to make significant inroads into either Labour or Conservative support, particularly given the electoral system, the Conservatives were the only party likely to deliver Brexit.
This ongoing sorting process meant that by 2019, the electorate was more divided by Brexit than in 2017. Examining the preelection Brexit preferences of 2019 major party voters, we find that more than 80% of those who voted Conservative were Leavers, and a similar proportion of Labour voters were Remainers.
CONSOLIDATION OF NEW CLEAVAGE PATTERNS?
A changing geographic basis of support can be indicative of realignment (Archer and Taylor Reference Archer and Taylor1981), and one of the most remarked-on features of the 2019 General Election was the collapse of Labour’s “red wall” (Kanagasooriam and Simon Reference Kanagasooriam and Simon2021). The loss of previously safe seats in Labour’s working-class industrial heartlands—from North Wales to the North East of England, including places held by Labour for decades such as Stoke-on-Trent, Bolsover, and Sedgefield—was a vivid symbolic representation of a Brexit realignment. The encroachment of the Conservatives in Northern England, while failing to gain ground in the South, marked a significant step in the reversal of the otherwise long-standing North–South divide (Johnston, Pattie, and Allsopp Reference Johnston, Pattie and Allsopp1988). Following the Brexit vote, at the 2017 General Election, Labour had made greater headway in constituencies that voted to remain in the EU. In 2019, it lost support disproportionately in seats that most strongly voted Leave and that were formerly Labour seats.Footnote 3 Similarly, the Conservatives strengthened their vote share in Leave areas while losing support in predominantly Remain areas. In another study, we demonstrated how this changing geography is produced by the Brexit sorting process (Fieldhouse and Bailey Reference Fieldhouse and Bailey2023).
The decrease in Labour’s red-wall seats represented not only a shift in electoral geography; it also reflected a continuation of the changing class basis of British electoral politics and the weakening of Labour’s working-class base (figure 4). The softening of this base has been apparent since New Labour, when—under Tony Blair—Labour reached out to a broader, more middle-class support base. This appeal was based on economic centralism (i.e., “the Third Way”) and social liberalism (Evans and Tilley Reference Evans and Tilley2017). However, after the immigration increase that followed the accession of EU states in 2004, immigration became increasingly salient. Many disenchanted, working-class, socially conservative, and anti-EU voters began to find new electoral homes, initially with the Conservative Party. When the Conservatives failed to deliver the lower immigration that they had promised, many of these voters switched to UKIP in 2015. The realignment of anti-EU working-class voters was accelerated by the Brexit vote, driving Leave voters to the Conservatives in 2017, the Brexit Party in the 2019 European Parliament Elections, and—in even greater numbers—to the Conservatives in the 2019 General Election (Evans, de Geus, and Green Reference Evans, de Geus and Green2021; Evans and Tilley Reference Evans and Tilley2017).
Class and Vote, 2015–2019
Figure 4 shows how Labour continued to have an advantage in 2015 among routine and semi-routine workers, despite its poor overall performance. By 2017—the first General Election after the Brexit vote—this lead had evaporated. In 2017, for the first time on record, the Conservatives won a larger proportion of both of these classes than Labour, despite Labour polling more than 40% of the vote overall. The loss of working-class votes to the Conservatives was compensated for in 2017 by a relatively strong performance among middle-class voters, especially professionals. In 2019, this historic reversal became even starker. Labour maintained its improvement among middle-class voters but lost additional ground to the Conservatives among working-class voters. A key factor in these shifts was sorting by Brexit preferences: Labour held its ground among the more pro-Remain middle classes (46% of whom voted Leave) and lost ground among the pro-Leave working classes (62% of whom voted Leave).Footnote 4
These changing patterns are not simply about class but also are attributable to the increasing importance of education as a predictor of vote choice. Education is correlated with ethnocentrism and liberal–authoritarian values (Evans, Heath, and Lalljee Reference Evans, Heath and Lalljee1996; Kinder and Kam Reference Kinder and Kam2010) and increasingly with Euroscepticism (Hakhverdian et al. Reference Hakhverdian, van Elsas, van der Brug and Kuhn2013). Reflecting this pattern, education was a strong predictor of the Brexit vote in 2016 and subsequently became an increasingly powerful predictor of party choice (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020). This becomes apparent when examining the choices of voters without any qualifications shown in figure 5. In 2015, 49% supported the Conservatives among those who voted for either of the major parties. This percentage increased to 61% in 2017 and to 74% in 2019.
Education and Vote, 2015–2019
Another increasingly important electoral cleavage that has been heightened by Brexit is age. Although older voters have long been regarded as more conservative, the increasing importance of the cultural-values dimension (especially immigration) and Brexit has intensified the effect of age on vote choice. This is illustrated in figure 6, which shows the relationship between age and the Conservative Party share of the two-party vote in 2015, 2017, and 2019. Age already was a strong predictor of Conservative Party voting in 2015, but this relationship strengthened in 2017 and remained so in 2019.
Age and Vote 2015–2019
Overall, recent elections have experienced a shift in electoral cleavages with Labour losing a substantial proportion of its support among its traditional heartlands, especially more socially conservative, older, working-class voters—while gaining support among younger, highly educated, and socially liberal voters. The Conservatives increased their support in those groups that deserted Labour. Although these changes are part of a long-term shift in political alignments reflecting economic and social change (Jennings and Stoker Reference Jennings and Stoker2017), the catalyst for the dramatic changes since 2016 was Brexit. It is the combination of long-term secular trends with the impact of electoral shocks that produces large-scale electoral change (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020). We cannot yet know whether these changes are transient or represent a more permanent realignment, but one clue lies in the shifting nature of political identities.
BREXIT IDENTITIES
Underlying changes in party identification normally are expected to accompany realignment alongside changes in voting behavior (Green, Palmquist, and Schickler Reference Green, Palmquist and Schickler2002; Miller Reference Miller1991). However, another symptom of realignment in an age of party dealignment may be the rise of competing social and political identities, filling the vacuum created by the decline of traditional party identification (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020; Hobolt, Leeper, and Tilley Reference Hobolt, Leeper and Tilley2021). The consolidation of new political identities connected to vote choice may be responsible for the decrease in volatility from the peak of 2015. In 2016, following the EU referendum, we introduced questions in the BES designed to measure identification with the Leave and the Remain sides of the Brexit debate, mirroring measures of affective partisanship based on social-identity theory (Bankert, Huddy, and Rosema Reference Bankert, Huddy and Rosema2017; Huddy, Mason, and Aarøe Reference Huddy, Mason and Aarøe2015). In 2017 and 2019, both Remain and Leave identities were stronger, on average, than traditional party identification measured on the same scale (Fieldhouse et al. Reference Fieldhouse, Green, Evans, Mellon, Prosser, Schmitt and van der Eijk2020). Figure 7 shows how these Leave and Remain identities grew in parallel in the period following the referendum in 2016, peaking in the period between the referendum and the 2017 General Election. Brexit identities receded only slightly after 2019 despite the passage of time since the referendum. Of course, over time, explicit identification with sides taken in the referendum may fade, morph into a more general liberal–conservative political identity, or—if parties compete around the cultural dimension—reinforce traditional party identities. For now, however, the continued strength of these EU referendum identities provides further evidence of a realignment along the Brexit divide.
Change in Leave and Remain Identity Strength Over Time
CONCLUSIONS
This article describes how the Brexit shock continued to affect electoral choices in 2019. We argue that the shifting electoral cleavages around social class, geography, education, and age; the rise of alternative political identities; and the resulting reversal in electoral volatility in the 2019 General Election are all indicative of an ongoing process of electoral realignment. The process of realignment has its roots long before the referendum of 2016; however, to properly understand electoral change and realignment, we must recognize this interplay of long-term secular trends and electoral shocks. The politicization of Brexit in 2019, with the rallying cry of the Conservatives to “get Brexit done” and the increasing divide of the two major parties on the issue of Brexit, meant that party choice for the two largest parties was structured around Brexit in 2019 even more than in 2017. This explains why traditional Labour “heartland” constituencies fell to the Conservatives, resulting in a comfortable parliamentary majority for Prime Minister Boris Johnson to take Britain out of the EU.
The Brexit shock continued to affect electoral choices in 2019. We argue that the shifting electoral cleavages around social class, geography, education, and age; the rise of alternative political identities; and the resulting reversal in electoral volatility in the 2019 General Election are all indicative of an ongoing process of electoral realignment.
Although the patterns of support we describe are consistent with the early stages of a realignment initiated by the EU referendum in 2016, it is still too early to determine whether these changes can be considered either a long-term critical realignment (Nardulli Reference Nardulli1995) or a temporary but substantial interruption to “politics as usual.” BES data show that support for Brexit has declined since Britain left the EU on January 31, 2020. In June 2020 (i.e., wave 20 of the BES), half of all respondents who stated that they would vote in another referendum said that they would vote to re-join the EU; the other half stated that they would vote to stay out. However, by December 2022 (i.e., wave 24), the proportion stating that they would re-join had increased to 56%. Since Britain left the EU in 2020, the sorting of Conservative and Labour voters by their Brexit preference remains very high (see figure 1). Approximately three quarters of Conservative voters continue to be drawn from those who want to stay out of the EU, with a similar proportion of Labour voters drawn from those who want to re-join.
Whether the Brexit alignment persists will depend on the electoral strategies of the major parties. As we discuss, the sorting process—which dramatically accelerated after the referendum in 2016—means that the characteristics of supporters of each major party have changed. Conservatives are now older, less educated, and more socially conservative; Labour voters are younger, more educated, and more socially liberal. As long as the parties continue to appeal to their new support bases, the shadow of Brexit will continue to dominate electoral alignments. If this becomes the case, we might expect Brexit identities to evolve into more general—but equally important—social conservative–liberal identities and/or be associated with the respective party identities. Alternatively, left–right economic issues may return to the fore as the after-effects of Brexit, COVID-19, and the cost-of-living crisis expose the political differences between the “haves and the have-nots.” The politics of competence (Green and Jennings, Reference Green and Jennings2017) may become of greater importance to the levels of party support, even as the realignment continues. Whichever scenario emerges will depend on not only voters and the strategic response of political parties but also the unforeseeable electoral shocks that inevitably are around the corner.
ACKNOWLEDGMENT
The authors gratefully acknowledge the support of the Economic and Social Research Council in funding this research (Grant Nos. ES/S015671 and ES/K005294).
DATA AVAILABILITY STATEMENT
All data are available at www.britishelectionstudy.com/data/#.ZCv-dnbMIuU. Research documentation and data that support the findings of this study are openly available on the PS: Political Science & Politics Harvard Dataverse DOI:10.7910/DVN/K7I3JT.
CONFLICTS OF INTEREST
The authors declare that there are no ethical issues or conflicts of interest in this research.
Open access
