Parliamentary Affairs

Tonge,, Jonathan;Wilks-Heeg,, Stuart;Thompson,, Louise

doi: 10.1093/pa/gsaa020pmid: N/A

The 2019 general election was another extraordinary chapter in a series of dramatic recent contests. Yet, given the consistent story from opinion pollsters (who had a good election) the extent of apparent wonderment at the broadcasters’ exit poll (enjoying another outstanding election) on election night was itself remarkable. A resounding Conservative majority was always a strong possibility. This was a mini-landslide forewarned—and in considerable detail. YouGov’s multi-level regression with post-stratification forecasts had catalogued the likely geography of Labour’s disaster.1 Nonetheless, the shock of the 2017 mislaying of the Conservatives’ majority by Theresa May loomed large over the election 30 months later, until that exit poll moment when the scale of the rout of Labour was revealed. Few had seriously doubted that the Conservatives under Boris Johnson would win the most seats; just not that many. In the weeks before the election, some commentators even speculated on the possibility of Jeremy Corbyn as Prime Minister, leading a minority rainbow government backed by virtually all the non-Conservative parliamentary forces (Finkelstein, 2019). The election night pre-exit poll forecasts of Boris Johnson and his team, while reflecting confidence in victory, all anticipated triumph by a smaller margin (Shipman, 2019). The scale of the Conservative victory, producing the UK’s first majority government for almost a decade, was quite something. For Labour, suffering a fall in vote share not far shy of 8%, the scale of the catastrophe is apparent when one remembers that 14 of the party’s seat losses occurred in constituencies held at every election since 1945—part of the collapse of the fabled ‘red wall’ of Labour strongholds, in which a majority had voted to leave the EU in 2016. The geographical parameters of this ‘red wall’ were imprecise. The term tended to be used most often in respect of seats in northern England which had been held by Labour for decades but was readily extended to similarly hitherto loyal Labour seats in the Midlands. Here, the scale of the move against Labour in long-time loyal seats such as Bassetlaw, which saw an 18% swing to the Conservatives, was remarkable. Yet, there was also a reasonable case for including solid Labour seats in north Wales as part of the ‘red wall’, as Jonathan Bradbury suggests in this volume. The Conservatives increased their vote share and seats, unprecedented for a party seeking a fourth consecutive term in office. In most of the constituencies captured it was a case of the Labour vote share plummeting, rather than the Conservatives’ vote share increasing by much. That, if anything, highlights the problems which confronted Labour more than it demonstrates hollowness in the Conservative victory. Following Johnson’s election as party leader in July 2019, the Conservatives had much sharper definition as the party of Brexit than under Theresa May. This is ironic given that May voted consistently in the Commons for Brexit, whereas Johnson did so only once. Johnson’s repeated assertions that May’s ‘Brexit with a Backstop’ (to keep Northern Ireland’s land border seamless) was the wrong type of EU departure were tactical. They looked hollow when he, upon elevation to the top job, quickly accepted what May had endorsed, dumping her awkward allies, the Democratic Unionist Party (DUP) in the process. The course was now set for Brexit to be delivered. If parliament continued to stall such a move, an election would be called, based upon Brexit delivery. Polling evidence suggested two key things. First, most voters had not changed their mind on Brexit (Curtice, 2020). Secondly, the Conservatives would win an election. Although Johnson’s election gambit initially met with hesitation from Labour, Corbyn’s party could hardly be seen to be ever running away from a contest. As such, they acquiesced. A more sensible electoral strategy for Labour might possibly—it’s difficult to argue more strongly—have been to let Johnson struggle on, in office but barely in power in respect of Brexit and with the goodwill of a honeymoon period likely to fade. Given Johnson’s Brexiteer credentials and the likelihood of him delivering the 2016 referendum verdict, the Conservatives unsurprisingly monopolised the Leave vote. The only other major player in that field, Nigel Farage’s Brexit Party, an easy winner of the 2019 European elections, stood aside in Conservative-held seats at the general election. Given the scale of the winning margin, it was clearly not a case of Farage acting as midwife to the delivery of Conservative election success. Nonetheless, there is evidence that Nigel Farage’s Brexit outriders offered a very modest boost. Where the Brexit Party stood, Labour’s vote fell by 8.6%, compared to 7.3% with the Brexiteers absent. The rise in the Conservatives’ vote averaged only 1.7% in constituencies with the Brexit Party present, compared to 2.5% elsewhere (Norris, 2019). Thus, Labour’s average percentage vote share was harmed nearly twice as much as was the Conservatives’ by the presence of Farage’s party. For the Prime Minister, his chief adviser, Dominic Cummings and election campaign manager, Isaac Levido, the outcome vindicated offering only a small core set of promises, dominated by the pithy appeal to ‘Get Brexit Done’. The slogan appeared to offer closure—of a sort—to a saga which had bedevilled UK politics since the June 2016 referendum. Having seen the fate of his predecessors, David Cameron and Theresa May, Johnson’s approach was simple; cut through the confusion; cast aside erstwhile allies; deliver on the instruction from voters; and leave the EU. The details of a trade deal seemingly unlikely to be advantageous to the UK could be sorted later. How a trading bloc of 66 million would achieve better terms by quitting one of over 500 million remained unexplained. Johnson calculated, correctly, that a combination of the satisfaction of the pro-Brexit demands of the committed; the end of the affair for the weary; the lack of desire for the second referendum offered by Labour and the split in the Remain vote, would allow the Conservatives to chart a course for victory. It came to pass. The Conservatives gained 56 seats that had voted to leave the EU in 2016. Labour lost 53 such constituencies. Johnson’s shrewd political judgement was, however, greatly abetted in its success, by the shortcomings of the leader of the Labour Party. Jeremy Corbyn entered the 2019 election with the worst ratings of any party leader in British political history. Having trailed Theresa May on the ‘who would make the best Prime Minister?’ question throughout her period in office, Corbyn found himself even further behind Johnson, despite the new Prime Minister’s ratings being only modest. Corbyn’s mandate from his party was manifestly large and had been confirmed twice but it had not eased the concerns of many within Labour’s parliamentary ranks that their leader could not win an election. Meanwhile, Johnson, although confronted by a difficult political in-tray on assuming the top job - one soon to become much worse - had a clear mandate from his parliamentarians and his party membership in his elevation. His internal control of his party appeared even more assured after the election. The Conservative parliamentary ranks were infused by new blood: 30% of its 2019 intake were Commons freshers. Johnson was entitled to take the view that many would not have arrived had it not been for him— although one might reasonably assert that Jeremy Corbyn also assisted their passage. The plan of the volume This volume analyses the 2019 general election by dissecting its context, results, parties, finances, geography and media. The analysis is divided into four sections: results and context, political parties, territorial dimensions and election themes. To begin the first section, David Denver outlines the election results and discusses how the opinion polls consistently pointed to a denouement involving a clear Conservative victory. That outcome is considered by Sir John Curtice in the broader context of whether a large overall majority represented a return to normality within our electoral system, after a decade of hung parliaments or narrow majorities. Does the outcome indicate that overall majorities are likely to be the norm once more or was this in some sense an exceptional result? That overall majority was a response to frustration with political stalemate and Louise Thompson contextualises the election in examining the parliamentary impasse that yielded such a decisive election result. Section two’s examination of the campaigns of the political parties begins with Sam Power, Tim Bale and Paul Webb exploring the Conservatives’ approach. They note how a tight campaign swerved potential dangers. Although Boris Johnson could be regarded as a potential asset, he was also high risk and it was decided to run a carefully controlled campaign, averting dangerous confrontations. The journalist, Andrew Neil, was left without a BBC interview with the Conservative leader. Beyond the endless insistence upon getting Brexit done, the Conservatives emphasised that the age of austerity and its attendant fiscal rules were in the rear-view mirror. Borrowing was permissible. Eunice Goes examines Labour’s campaign, which lacked the verve and mobilisation of 2017 as it charted a course towards a record-equalling fourth consecutive defeat. The lack of confidence within Labour’s ranks was displayed on Jeremy Corbyn’s campaign trail. Many of the seats visited by the Labour leader were already held by his party. The concern for the 76 Labour-held seats with a majority of less than 8,000 over the Conservatives was apparent. Goes analyses the impacts of Brexit, Corbyn and anti-Semitism, a negative triple whammy, upon Labour’s performance. David Cutts and Andrew Russell assess the performance of another party which, from a very different perspective from the Conservatives, at least offered clarity on Brexit. The Liberal Democrats’ 2019 decision to simply revoke Article 50 and cancel EU withdrawal drew accolades among the fellow-travellers at the party conference three months before the election. It also contributed to the Liberal Democrats becoming more exclusively a party backed by 2016 Remain voters—but not many of them and very few beyond. More broadly, the public had little knowledge of what else the Liberal Democrats stood for (see also Curtice 2019). This inability to harness the Remain vote placed the prospect of ‘Jo Swinson – our next Prime Minister’, as trumpeted on party leaflets, look on the implausible, even ludicrous, side of optimism. James Dennison examines the smaller parties, assessing the significance of the decision by Nigel Farage’s Brexit Party not to contest any of the 317 seats won by the Conservatives at the 2017 election. The Brexit Party offered a possible electoral repository for Labour Leave voters who could neither bring themselves to vote Conservative nor back a Labour Party led by Jeremy Corbyn. Dennison also examines the modest progress, in vote share at least, of the Greens, under their Remain Alliance with the Liberal Democrats and Plaid Cymru, noting that a sizeable minority of the Green vote had previously voted Labour. This section also considers the brief life of Change UK, the main changes wrought by its defectors from the Labour and Conservative parties being the regular ones to the new party’s name. Section three’s consideration of the election in the devolved nations begins with the analysis of Scotland by James Mitchell and Ailsa Henderson. As a pro-Remain party with a monopoly of a large pro-independence vote, the SNP was always likely to do well and so it transpired. The Conservatives’ attempt to harness the anti-independence majority was ranged against their minority status as a pro-Leave party, toned down where possible for Scottish consumption but unhelpful. Meanwhile, Labour struggled to define itself via either constitutional question. In Leave-voting Wales, Jonathan Bradbury analyses how and why Labour struggled to hold part of its ‘natural’ territory, conceding long-held seats in the north of the country and suffering losses in vote shares throughout. Northern Ireland’s concerns over Brexit led to election pacts which also took on a nationalist (anti-Brexit) versus unionist (pro-Brexit, albeit not on British Johnson’s terms) flavour. As Jonathan Tonge and Jocelyn Evans note, however, detachment from unionism and nationalism helped the Alliance Party, also anti-Brexit but opposed to Remain pacts, to make strides. The DUP and Sinn Féin were punished for the failure to restore the power-sharing Northern Ireland Assembly, a problem rectified within one month of the election. The fourth section covers various important aspects of the campaign, analysing some of its target recipients. In Britain Votes 2017, Justin Fisher noted how, despite being the party of government calling the snap election, the Conservative party machine was under-prepared. He outlines how this was not the case in 2019, with the Conservatives much better prepared for an early election, organisationally and financially. Katharine Dommett and Mehmet Emin Bakir examine how that preparation played out online, noting how both the Conservative and Labour parties invested heavily in Facebook advertising. Matthew Flinders’ analysis of the campaign notes how all political leaders were trying to overcome a lack of trust in them among the electorate. In that respect, Corbyn’s problems were acute but not isolated. Sarah Harrison considers these democratic frustrations among young voters and, notwithstanding a propensity to vote for alternatives to the prevailing Conservative government, does not find an exceptional level of disillusionment. Her findings highlight concerns that the problem is the democratic journey. The targeting of the campaign at women is assessed by Emily Harmer and Rosalynd Southern. Party manifestos concentrated their gendered concerns in familiar areas. The authors note, however, significant improvements in representation, with Labour and the Liberal Democrats now having more female MPs than male. Finally, Dominic Wring and Stephen Ward assess the role of the traditional media, noting that the press still matters as its older readership are likely to vote. They also observe how the backing of many newspapers for the Conservatives was the most gung-ho for some time. Our concluding analysis assesses the relative importance of the key aspects of the election; Johnson’s elevation to the Conservative leadership in the summer; the promise to ‘Get Brexit Done’, the failures of the Labour opposition under Corbyn and the apparent hubris of the Liberal Democrats. The sum of the parts was a victory which surprised in terms of its scale but not its victor. Footnotes 1 In 2017, YouGov pioneered the use of multi-level regression with post-stratification (MRP) to forecast a UK general election outcome, correctly predicting a hung parliament when virtually all conventional opinion polls pointed to a large Conservative majority. In contrast to opinion polls that estimate party shares of the vote from a representative sample of around 1,000 people, MRP forecasts use samples of tens of thousands of electors in combination with other data sources to estimate the number of seats each party will win. Under MRP, polling data are first modelled using numerous individual-level socio-demographic characteristics to estimate the probability of different groups of people voting for a party (multi-level regression). These probabilities are then applied to the socio-demographic make-ups of each individual constituency and weighted for other factors such as past voting patterns in that locality (post-stratification). In 2019,’ MRP’ entered the mainstream of the general election lexicon. The publication of YouGov’s MRP forecast, in late November, was eagerly awaited and its central forecast was a Conservative majority of 68 (a subsequent iteration closer to the election revised this down to 28). In a notable development, political parties also produced their own MRP forecasts to inform their campaign strategies – see Fisher, this volume. References Curtice J. ( 2019 ) ‘After “Stop Brexit”, the Public Draws a Blank on Lib Dem Policies’, The Times, 6 December. Curtice J. ( 2020 ) ‘ Brave New World: Understanding the 2019 General Election ’, Political Insight , 11 , 8 – 12 . Google Scholar Crossref Search ADS WorldCat Finkelstein D. ( 2019 ) ‘Corbyn Doesn’t Need to Win to Become PM’, The Times, 6 November. Norris P. ( 2019 ) ‘Divided we Fall: Was Nigel Farage the Kingmaker of the Johnson Victory?’. In Jackson D. , Thorsen E. , Lilleker D. , Weidhase N. (eds.) UK Election Analysis 2019: Media, Voters and the Campaign , Bournemouth , Centre for Comparative Politics and Media Research , pp. 38 – 39 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Shipman T. ( 2019 ) ‘Sedgefield Fell and They Erupted into Song – Things Can Only Get Better’, Sunday Times, 15 December. © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Denver*,, David

doi: 10.1093/pa/gsaa037pmid: N/A

Unlike the situation in 2017, the calling of the 2019 general election did not take the country by surprise. Boris Johnson had succeeded Theresa May as Prime Minister in July and formed a government avowedly committed to exiting the European Union (EU) as soon as possible. However, parliament had repeatedly frustrated all previous attempts to do so and the chances of MPs passing an act to implement a withdrawal agreement were slim. When parliament reconvened in September, therefore, the government twice introduced motions to allow an early general election to be held under the terms of the Fixed-term Parliaments Act. This required the support of two-thirds of MPs but on both occasions the numbers voting for an election fell well short of this. By mid-October, Johnson had successfully negotiated a new—if not dramatically different—Brexit deal with the EU and the relevant withdrawal agreement bill passed its second reading on the 22nd of the month. Any hopes that this meant that the end to the Brexit saga was in sight were quickly dispelled, however, as the House immediately rejected the proposed timetable for further consideration of the bill. Another attempt to force a general election was made on Monday 28 October but once again failed to obtain the votes necessary. Next day, however, the government sought to circumvent the Fixed-term Parliaments Act by bringing forward a short bill declaring that there would be an election on 12 December. This required only a simple majority to pass and the Liberal Democrats and Scottish National Party (SNP) indicated their support. Since this made a majority inevitable, Labour reversed its position of the day before and also voted for an election. Once again, the Fixed-term Parliaments Act had been shown to be a flawed piece of constitutional tinkering and the country embarked on its first December general election since 1923. 1. The inter-election cycle of party support After the 2017 election, the government’s agenda and the political scene generally were dominated by a single issue to an extent never before seen in peacetime—whether and how the UK should exit the EU in line with the result of the 2016 referendum (i.e. ‘Brexit’). Despite failing to make much progress on this process, the Conservatives enjoyed fairly steady support for a year before dipping slightly in the second half of 2018 (Figure 1.1). However, the so-called ‘Chequers meeting’ in July of that year, at which Mrs May informed her cabinet of her proposed post-Brexit relationship with the EU, proved a significant turning point for the party. Having previously been encouraged by her assertion that ‘no deal for Britain is better than a bad deal for Britain’, Brexiteers were now appalled by her proposals. Shortly after the meeting, David Davis and Boris Johnson (respectively Brexit Secretary, and Foreign Secretary) resigned along with two junior ministers, two vice-chairs of the Conservative party and four parliamentary private secretaries. Four months later, when the final proposed withdrawal agreement was revealed, more resignations from the government followed. The rallying cry of the European Research Group (ERG) of Eurosceptic Conservative backbenchers was now ‘Chuck Chequers’. Figure 1.1 Open in new tabDownload slide Trends in party support, July 2017–October 2019 Note: The data shown are the mean monthly voting intentions in published polls reported by Anthony Wells on his UK Polling Report website. Figure 1.1 Open in new tabDownload slide Trends in party support, July 2017–October 2019 Note: The data shown are the mean monthly voting intentions in published polls reported by Anthony Wells on his UK Polling Report website. In early December, the House of Commons was due to hold a ‘meaningful vote’ on the agreement but at the last moment this was postponed by the Prime Minister, claiming that she would seek changes from the EU. On the next day, Conservative malcontents (led by the ERG) gathered enough support to initiate a vote of confidence in Mrs May as Conservative leader. She managed to win the ensuing vote by 200 votes to 117 but was obviously seriously weakened. The withdrawal deal that the Prime Minister had negotiated with the EU finally came before the House of Commons in January 2019 and was defeated by a historic margin—432 to 202. There followed a chaotic period in parliament which included numerous further defeats for the government and the abandonment of the original date set for withdrawal from the EU (29 March). Unsurprisingly, support for the Conservatives declined sharply (Figure 1.1) and the proportion of poll respondents preferring Mrs May as Prime Minister nose-dived (Figure 1.2). Figure 1.2 Open in new tabDownload slide Best person for Prime Minister, July 2017–October 2019 Source: YouGov. Figure 1.2 Open in new tabDownload slide Best person for Prime Minister, July 2017–October 2019 Source: YouGov. Meanwhile, the Labour party had problems of its own. Following its unexpectedly good performance in 2017 support for the party drifted steadily downwards to the end of 2018. Doubts about the party leader, Jeremy Corbyn, as a potential Prime Minister re-emerged—at no point during this period did he lead Mrs May as the electorate’s preferred occupant of 10 Downing Street (Figure 1.2). More specifically, there was discontent over the leadership’s handling of Brexit and accusations that anti-Semitism had infected the party, with Jewish Labour MPs being subject to abuse by party members, and that these cases had not been adequately dealt with. As a consequence of these developments, in February 2019 seven Labour MPs resigned the Labour whip and formed what they called ‘The Independent Group’. They were shortly joined by another Labour MP and three Conservative Remainers. By April the ‘TIGS’, as they were swiftly christened, had become a registered party named ‘Change UK – The Independent Group’. More worryingly for the Conservatives, a spectre from the past also re-emerged at this time in the person of Nigel Farage and the Brexit Party he co-founded. He was joined by most of UKIP’s MEPs and indicated that the party would contest the forthcoming European Parliament elections which the government was forced to hold due to its failure to leave the EU by the appointed date. The impact on the electorate was immediate and striking. As Figure 1.1 shows, support for ‘others’, mainly the Brexit party which drew level with the Conservatives by April, increased rapidly. Voters who had previously been inclined towards UKIP now flocked to the new party as the former staggered from one internal crisis to another. The opinion poll results were reflected in local elections held across England in early May. Of the 248 councils involved, the Conservatives lost control of 44 and Labour of six. In the process, the former lost 1,330 seats and the latter 84. Although UKIP also lost ground, the gainers were the Liberal Democrats (+704 seats), Independents (+605) and Greens (+194). The BBC’s ‘projected national vote share’ was 28% for both the Conservatives and Labour, 19% for the Liberal Democrats and 25% for others. The Liberal Democrats and Greens tried to claim that the elections represented a success for anti-Brexit forces but, although this may have played a part, it seems more likely that many voters were simply going for ‘anyone but the big two’. After these bad results, Conservative calls for Mrs May’s resignation grew even louder and, if any further evidence that she could not carry on was needed, this was provided by the results of the European Parliament elections in late May. These shattered any notion that the 2017 general election had heralded a return to stable two-party politics in Britain as they were simply disastrous for both major parties. The Conservatives came fifth with only 9.1% of the votes in Great Britain—their worst performance in a national election since 1832; Labour managed just 14.1%—the party’s worst showing since 1910. In Scotland, Labour contrived to come fifth in votes and won no seats. The Liberal Democrats (20.3%) and the Greens (12.1%) did well but it was the newly-formed Brexit party which topped the poll with 31.6%. The latter led in every English region outside London, as well as in Wales, and even managed to win a seat in Scotland where it came second (14.8%), albeit a long way behind the SNP (on 37.8%). On the other hand, the other new party that had emerged—Change UK—sank virtually without trace and six of its 11 MPs resigned from the party shortly afterwards. The Prime Minister herself announced her imminent resignation on 24 May. In the Conservative leadership election which followed, the contestants were quickly whittled down to two—Boris Johnson and Jeremy Hunt—and in the run-off party members gave the former two-thirds of the 139,000 votes cast. Thus, in late July Johnson became the third Conservative Prime Minister in the space of three years. The change of personnel produced an immediate fillip in support for the Conservatives (Figure 1.1) although Labour also benefitted a little as the popularity of the Brexit party declined from its June peak and the Liberal Democrat advance stalled. Despite continuing difficulties with Remainers in his party, which resulted in a number having the whip withdrawn, Johnson’s Conservatives embarked on the 2019 campaign enjoying a clear lead over their rivals. He also quickly opened up a commanding lead over Jeremy Corbyn as the electorate’s preferred Prime Minister (Figure 1.2) although his support in this respect never exceeded 45% and his situation did not appear as rosy as Theresa May’s had been at the start of the 2017 campaign. The surprise announcement of an early election in 2017 had contributed to a sharp fall in the number of candidates nominated across Great Britain (the situation in Northern Ireland is discussed in a separate chapter) to 3,195. There was little change in 2019, however, with 3,218 coming forward. As usual, the Conservatives and Labour contested all but the (new) Speaker’s seat and the SNP all Scottish constituencies. However, the Liberal Democrats, Greens and Plaid Cymru agreed not to oppose one another in 60 seats—a tactic which had denied the Conservatives victory in the Brecon and Radnorshire by-election in August 2019—and there were, therefore, 611 Liberal Democrat candidates and 36 for Plaid. The number of Green Party candidates increased from 460 to 494 while, having eventually decided not to oppose sitting Conservatives, the Brexit party contested 275 seats. Among the assorted ‘others’ who, according to taste, enliven or add unnecessary clutter to the British electoral scene, by far the largest group comprised Independents (223, including six in Boris Johnson’s constituency) followed by UKIP with 31 (down from 614 in 2015 and 379 in 2017). A reminder of earlier political times—and confirmation that hope springs eternal in some breasts—is provided by the fact that more than 30 years after the two parties merged there were 19 candidates representing the Liberal Party and the same number carrying the flag for the Social Democratic Party (SDP). 2. Trends in party support during the ‘short’ campaign In some ways, the trends in party support during the 2019 campaign resembled those witnessed in 2017 (Figure 1.3). Figure 1.3 Open in new tabDownload slide Trends in voting intentions during the ‘short’ campaign Note: The data are three-day moving averages of polls reported by Mark Pack (https://www.markpack.org.uk/155623/voting-intention-opinion-poll-scorecard/). Figure 1.3 Open in new tabDownload slide Trends in voting intentions during the ‘short’ campaign Note: The data are three-day moving averages of polls reported by Mark Pack (https://www.markpack.org.uk/155623/voting-intention-opinion-poll-scorecard/). In both cases, Labour’s position improved steadily while the Liberal Democrats and others declined. In 2019, over the six weeks involved, the Liberal Democrats fell from the mid-to-high teens to around 12% of voting intentions while the Brexit party—the most significant of the others in the British context at the outset—slid from around 10% to about 3%, rendering it close to insignificance. The main difference between the two campaigns concerns the Conservatives. In 2017 their campaign was singularly inept. Their manifesto included controversial proposals relating to social care for the elderly and its launch produced an immediate change in the mood of the electorate. At the same time Theresa May, then Prime Minister, proved to be a poor performer on the national campaign stage and was widely ridiculed as ‘robotic’. On this occasion the Conservative manifesto (and, indeed, the whole campaign strategy) was ultra-cautious, while Johnson appeared affable and energetic, constantly reiterating that his aim was simply to ‘Get Brexit Done’ while avoiding potential ‘car crash’ moments such as a televised interview with the BBC’s Andrew Neil. On the other side, it was apparent that Mr Corbyn was not attracting the same enthusiastic audiences that he had in 2017. The consequence was that although Labour’s standing improved over the campaign so too did that of the Conservatives. On the basis of the data used to construct Figure 1.3 the latter’s lead over their main opponents fell below ten points on only 13 of the 43 days involved. Four of these cases occurred in the last week of the campaign. This, no doubt, encouraged optimists in Labour quarters hoping that the ‘miracle’ of 2017 could be repeated while, perhaps indicating some Conservative panic, a Daily Telegraph headline (11 December 2019) warned that the election was ‘on a knife edge’. Unfortunately for the headline writers, of the 12 polling firms which produced final predictions, the one used by the Telegraph (Savanta Com Res) was the least accurate, underestimating the Conservatives’ vote share by 5 points and overestimating that of Labour by 3. Nonetheless, the averages for all the final polls (Conservative 43.3%, Labour 33.5% and Liberal Democrats 11.6%) suggested that the Conservatives were on course for a clear victory. In Scotland, meanwhile, five campaign polls indicated that the SNP was set to improve upon its 2017 performance (averaging 42% of voting intentions) mainly at the expense of Labour. The latter (average 18%) consistently lagged in third place, well behind the Conservatives (28%). In Wales, in contrast, two commercial polls published once the campaign was in full swing reported that Plaid Cymru was making no headway and that, although Labour was losing support, it continued to be the most popular party. Despite the solid Conservative lead predicted by the Britain-wide polls, commentators were chary of extrapolating from voting intention figures to guessing what the election outcome might be in terms of seats in the House of Commons. In part this reflected the fact that in 2017 all the polls had seriously underestimated Labour’s performance but there were also worries about the potential impact of the Brexit Party and the possible extent and nature of tactical voting, which were difficult to be sure about in advance of the results. However, the much vaunted ‘Multiple Regression and Post-Stratification’ model developed by YouGov, which had proved spectacularly successful in 2017, finally predicted an overall Conservative majority of 28 seats, although the relevant report was hedged about with numerous cautionary warnings. In the event, the Conservatives far out-performed expectations in this respect. 3. The national result For the fourth successive election, the eagerly-awaited announcement of the results of the BBC/ITN/Sky exit poll, made as soon as voting was over, provoked emotions of elation, consternation and anguish among the various partisans although, on this occasion, there were few expressions of outright disbelief. The poll predicted a Conservative majority of 86 seats in the House of Commons with the party taking 368 constituencies to 191 for Labour, 55 for the SNP and 13 for the Liberal Democrats. As Table 1.1 shows, the final Labour total was slightly underestimated but the party nonetheless ended up with 60 seats fewer than in 2017. Table 1.1 Share of votes and number of seats won (Great Britain) and changes from 2017 . Share of . Change . Number . Change . . votes (%) . 2017–2019 . of seats . 2017–2019 . Conservative 44.7 +1.3 365 +48 Labour 32.9 −8.1 202 −60 Liberal Democrat 11.8 +4.2 11 −1 Green 2.8 +1.1 1 0 Brexit 2.1 — 0 − SNP/Plaid Cymru 4.5 +0.9 52 +13 Other 1.2 −1.5 1 0 . Share of . Change . Number . Change . . votes (%) . 2017–2019 . of seats . 2017–2019 . Conservative 44.7 +1.3 365 +48 Labour 32.9 −8.1 202 −60 Liberal Democrat 11.8 +4.2 11 −1 Green 2.8 +1.1 1 0 Brexit 2.1 — 0 − SNP/Plaid Cymru 4.5 +0.9 52 +13 Other 1.2 −1.5 1 0 Note: The Speaker, who was not opposed by the Conservatives, Labour or Liberal Democrats, is treated as an ‘Other’. Open in new tab Table 1.1 Share of votes and number of seats won (Great Britain) and changes from 2017 . Share of . Change . Number . Change . . votes (%) . 2017–2019 . of seats . 2017–2019 . Conservative 44.7 +1.3 365 +48 Labour 32.9 −8.1 202 −60 Liberal Democrat 11.8 +4.2 11 −1 Green 2.8 +1.1 1 0 Brexit 2.1 — 0 − SNP/Plaid Cymru 4.5 +0.9 52 +13 Other 1.2 −1.5 1 0 . Share of . Change . Number . Change . . votes (%) . 2017–2019 . of seats . 2017–2019 . Conservative 44.7 +1.3 365 +48 Labour 32.9 −8.1 202 −60 Liberal Democrat 11.8 +4.2 11 −1 Green 2.8 +1.1 1 0 Brexit 2.1 — 0 − SNP/Plaid Cymru 4.5 +0.9 52 +13 Other 1.2 −1.5 1 0 Note: The Speaker, who was not opposed by the Conservatives, Labour or Liberal Democrats, is treated as an ‘Other’. Open in new tab The SNP performance was less dramatic than the exit poll suggested but it racked up 48 of the 59 seats in Scotland (while Plaid Cymru retained its four seats in Wales). The actual Conservative majority over all others was 80 seats—easily enough to guarantee five years in government barring some unforeseen catastrophe. In terms of vote share, the Conservatives improved somewhat on their 2017 tally but Labour fell back sharply, giving an overall ‘swing’ of 4.7% from the latter to the former. This was the second largest swing between the two parties since 1997 (the largest, 5.1%, having been seen in 2010). Under their new leader, Jo Swinson, the Liberal Democrats achieved a healthy increase in vote share but the total (11.8%) was clearly a disappointment given their performances in the local and European elections only some seven months previously and the fact that in October they still averaged over 18% in the polls. It would appear that their standing declined as their key policy of simply ignoring the result of the Brexit referendum and remaining in the EU became more widely known, while their participation in the coalition government with the Conservatives from 2010 to 2015 still rankled with some Labour supporters. Despite winning more votes than in the previous election, the Liberal Democrats recorded a net loss of one seat and their losses included that of Ms Swinson herself, narrowly defeated in East Dunbartonshire. As expected, the Greens and the Brexit party were fringe players in the election although the only Green MP, Caroline Lucas, easily retained her Brighton Pavilion seat. The apparently weak performance of Nigel Farage’s party reflects in part, of course, the fact that it did not put forward candidates in Conservative-held seats. In the 275 constituencies it contested, the Brexit Party share of votes was a slightly more respectable 5.1% (the respective figure for the Greens being 3.5% in 494 seats). 3.1 Regional and urban–rural variations Although the shift from Labour to the Conservatives was evident in every region (and in Scotland and Wales), its extent varied substantially. Table 1.2 shows the regional swing figures. Table 1.2 Labour-Conservative swing in regions North East 8.3% to Con London 2.6% to Con North West 4.9% to Con South East 3.3% to Con Yorkshire/Humber 6.4% to Con South West 3.6% to Con East Midlands 6.4% to Con West Midlands 6.5% to Con Wales 5.2% to Con Eastern 5.4% to Con Scotland 2.5% to Con North East 8.3% to Con London 2.6% to Con North West 4.9% to Con South East 3.3% to Con Yorkshire/Humber 6.4% to Con South West 3.6% to Con East Midlands 6.4% to Con West Midlands 6.5% to Con Wales 5.2% to Con Eastern 5.4% to Con Scotland 2.5% to Con Note: The figure for Scotland is not very meaningful as the SNP came first or second in every constituency. Open in new tab Table 1.2 Labour-Conservative swing in regions North East 8.3% to Con London 2.6% to Con North West 4.9% to Con South East 3.3% to Con Yorkshire/Humber 6.4% to Con South West 3.6% to Con East Midlands 6.4% to Con West Midlands 6.5% to Con Wales 5.2% to Con Eastern 5.4% to Con Scotland 2.5% to Con North East 8.3% to Con London 2.6% to Con North West 4.9% to Con South East 3.3% to Con Yorkshire/Humber 6.4% to Con South West 3.6% to Con East Midlands 6.4% to Con West Midlands 6.5% to Con Wales 5.2% to Con Eastern 5.4% to Con Scotland 2.5% to Con Note: The figure for Scotland is not very meaningful as the SNP came first or second in every constituency. Open in new tab It was clearly most marked in the North East and above average in the rest of the North of England and the Midlands. In the South East and South West the Conservatives fared less well while the swing was smallest of all in London. Given that the capital also had much the largest pro-Labour swing in 2017, the 2019 outcome could add grist to the mill of those who argue that it houses a ‘metropolitan liberal elite’ who are out of touch with the rest of the country. Perhaps more plausibly, it may be the presence of large and increasing numbers of ethnic minority voters—who strongly favour Labour—which explains London’s divergence from national trends. The 2019 results in the regions are shown in more detail in Table 1.3. Table 1.3 Party shares of votes (%) and seats won in regions . Con . Lab . Lib Dem . Brexit . Green . SNP/PC . Other . North East 38.3 42.6 8.8 7.9 2.4 1.9 Seats 10 19 North West 37.5 46.5 7.9 3.9 2.5 1.8 Seats 32 41 1 1 Yorkshire/Humber 43.1 38.9 8.1 5.9 2.3 1.8 Seats 26 28 East Midlands 54.8 31.7 7.8 1.5 2.6 1.7 Seats 38 8 West Midlands 53.4 33.9 7.9 1.4 3.0 0.5 Seats 44 15 Eastern 57.2 24.4 13.4 0.4 3.0 1.7 Seats 52 5 1 London 32.0 48.1 14.9 1.4 3.1 0.5 Seats 21 49 3 South East 54.0 22.1 18.2 0.3 3.9 1.4 Seats 74 8 1 1 South West 52.8 23.4 18.2 0.4 3.8 1.6 Seats 48 6 1 Wales 36.1 40.9 6.0 5.4 1.0 9.9 0.6 Seats 14 22 4 Scotland 25.1 18.6 9.5 0.5 1.0 45.0 0.3 Seats 6 1 4 48 . Con . Lab . Lib Dem . Brexit . Green . SNP/PC . Other . North East 38.3 42.6 8.8 7.9 2.4 1.9 Seats 10 19 North West 37.5 46.5 7.9 3.9 2.5 1.8 Seats 32 41 1 1 Yorkshire/Humber 43.1 38.9 8.1 5.9 2.3 1.8 Seats 26 28 East Midlands 54.8 31.7 7.8 1.5 2.6 1.7 Seats 38 8 West Midlands 53.4 33.9 7.9 1.4 3.0 0.5 Seats 44 15 Eastern 57.2 24.4 13.4 0.4 3.0 1.7 Seats 52 5 1 London 32.0 48.1 14.9 1.4 3.1 0.5 Seats 21 49 3 South East 54.0 22.1 18.2 0.3 3.9 1.4 Seats 74 8 1 1 South West 52.8 23.4 18.2 0.4 3.8 1.6 Seats 48 6 1 Wales 36.1 40.9 6.0 5.4 1.0 9.9 0.6 Seats 14 22 4 Scotland 25.1 18.6 9.5 0.5 1.0 45.0 0.3 Seats 6 1 4 48 Note: The Speaker is counted as ‘Other’. Open in new tab Table 1.3 Party shares of votes (%) and seats won in regions . Con . Lab . Lib Dem . Brexit . Green . SNP/PC . Other . North East 38.3 42.6 8.8 7.9 2.4 1.9 Seats 10 19 North West 37.5 46.5 7.9 3.9 2.5 1.8 Seats 32 41 1 1 Yorkshire/Humber 43.1 38.9 8.1 5.9 2.3 1.8 Seats 26 28 East Midlands 54.8 31.7 7.8 1.5 2.6 1.7 Seats 38 8 West Midlands 53.4 33.9 7.9 1.4 3.0 0.5 Seats 44 15 Eastern 57.2 24.4 13.4 0.4 3.0 1.7 Seats 52 5 1 London 32.0 48.1 14.9 1.4 3.1 0.5 Seats 21 49 3 South East 54.0 22.1 18.2 0.3 3.9 1.4 Seats 74 8 1 1 South West 52.8 23.4 18.2 0.4 3.8 1.6 Seats 48 6 1 Wales 36.1 40.9 6.0 5.4 1.0 9.9 0.6 Seats 14 22 4 Scotland 25.1 18.6 9.5 0.5 1.0 45.0 0.3 Seats 6 1 4 48 . Con . Lab . Lib Dem . Brexit . Green . SNP/PC . Other . North East 38.3 42.6 8.8 7.9 2.4 1.9 Seats 10 19 North West 37.5 46.5 7.9 3.9 2.5 1.8 Seats 32 41 1 1 Yorkshire/Humber 43.1 38.9 8.1 5.9 2.3 1.8 Seats 26 28 East Midlands 54.8 31.7 7.8 1.5 2.6 1.7 Seats 38 8 West Midlands 53.4 33.9 7.9 1.4 3.0 0.5 Seats 44 15 Eastern 57.2 24.4 13.4 0.4 3.0 1.7 Seats 52 5 1 London 32.0 48.1 14.9 1.4 3.1 0.5 Seats 21 49 3 South East 54.0 22.1 18.2 0.3 3.9 1.4 Seats 74 8 1 1 South West 52.8 23.4 18.2 0.4 3.8 1.6 Seats 48 6 1 Wales 36.1 40.9 6.0 5.4 1.0 9.9 0.6 Seats 14 22 4 Scotland 25.1 18.6 9.5 0.5 1.0 45.0 0.3 Seats 6 1 4 48 Note: The Speaker is counted as ‘Other’. Open in new tab In Scotland, the partial recovery that Labour had achieved in 2017 fizzled out in this election as its vote share slumped by ten points, leaving Ian Murray in Edinburgh South as once again the party’s sole Scottish representative at Westminster. The Conservatives also fell back, losing seven seats, but nonetheless retained second place. The beneficiary of these developments was the SNP which lost no time in claiming that its overwhelming victory (in terms of seats, at least) was a mandate for a second independence referendum. In Wales, in contrast, nationalist support declined a little and, while Labour remained the largest party, the Conservatives were not far behind. Across England, the more pronounced swing to the Conservatives in the North and Midlands resulted in the party gaining in spectacular fashion some seats which had long been considered very safe for Labour. In Bolsover, for example, Dennis Skinner was ousted after almost 50 years as MP, while Blyth Valley went Conservative for the first time since the creation of the seat in 1950. Nonetheless, the north–south (outside London) divide in party support remains a significant feature of election outcomes. In England as a whole, the Conservatives had a lead of 13.3 points over Labour but in the three southern regions it was 31.5 points and in the Midlands 21.5. In the three northern regions, however, Labour led by 3.6 points. Turning to the smaller parties, the Liberal Democrats clearly had more appeal in the South than in the North and Midlands and the same is true of the Green Party. In the case of the Brexit Party the reverse is the case—the further south one travels in England, the poorer was the Brexit Party performance. In part this reflects the pattern of Brexit Party candidatures but in seats which it contested the party scored 7.5% in the North, 4.3% in the Midlands and 2.6% in the South. Another geographical divide that has characterised British elections is that between big cities and other heavily urban areas on the one hand and smaller towns and rural areas on the other. Labour always gets much better results in the former and the Conservatives in the latter. This difference remained fairly stark in 2019 as Table 1.4 shows. Table 1.4 The urban–rural divide 2019 (England) . Share of Votes . Change 2017–2019 . . Very . Mainly . Very . Mainly . . Urban . Urban . Mixed . Rural . Urban . Urban . Mixed . Rural . . % . % . % . % . . . . . Conservative 30.2 45.8 54.3 59.4 +0.4 +2.5 +1.5 +2.5 Labour 52.2 37.1 25.7 19.7 −7.2 −8.3 −8.5 −7.2 Lib. Democrat 10.3 10.6 13.1 16.4 +4.1 +3.7 +5.1 +5.7 Green 3.7 2.4 3.0 3.2 +1.5 +1.1 +1.1 +1.0 Other 3.6 4.2 3.9 1.3 +1.2 +1.1 +0.8 −2.0 (Constituencies) (138) (143) (159) (93) . Share of Votes . Change 2017–2019 . . Very . Mainly . Very . Mainly . . Urban . Urban . Mixed . Rural . Urban . Urban . Mixed . Rural . . % . % . % . % . . . . . Conservative 30.2 45.8 54.3 59.4 +0.4 +2.5 +1.5 +2.5 Labour 52.2 37.1 25.7 19.7 −7.2 −8.3 −8.5 −7.2 Lib. Democrat 10.3 10.6 13.1 16.4 +4.1 +3.7 +5.1 +5.7 Green 3.7 2.4 3.0 3.2 +1.5 +1.1 +1.1 +1.0 Other 3.6 4.2 3.9 1.3 +1.2 +1.1 +0.8 −2.0 (Constituencies) (138) (143) (159) (93) Open in new tab Table 1.4 The urban–rural divide 2019 (England) . Share of Votes . Change 2017–2019 . . Very . Mainly . Very . Mainly . . Urban . Urban . Mixed . Rural . Urban . Urban . Mixed . Rural . . % . % . % . % . . . . . Conservative 30.2 45.8 54.3 59.4 +0.4 +2.5 +1.5 +2.5 Labour 52.2 37.1 25.7 19.7 −7.2 −8.3 −8.5 −7.2 Lib. Democrat 10.3 10.6 13.1 16.4 +4.1 +3.7 +5.1 +5.7 Green 3.7 2.4 3.0 3.2 +1.5 +1.1 +1.1 +1.0 Other 3.6 4.2 3.9 1.3 +1.2 +1.1 +0.8 −2.0 (Constituencies) (138) (143) (159) (93) . Share of Votes . Change 2017–2019 . . Very . Mainly . Very . Mainly . . Urban . Urban . Mixed . Rural . Urban . Urban . Mixed . Rural . . % . % . % . % . . . . . Conservative 30.2 45.8 54.3 59.4 +0.4 +2.5 +1.5 +2.5 Labour 52.2 37.1 25.7 19.7 −7.2 −8.3 −8.5 −7.2 Lib. Democrat 10.3 10.6 13.1 16.4 +4.1 +3.7 +5.1 +5.7 Green 3.7 2.4 3.0 3.2 +1.5 +1.1 +1.1 +1.0 Other 3.6 4.2 3.9 1.3 +1.2 +1.1 +0.8 −2.0 (Constituencies) (138) (143) (159) (93) Open in new tab Labour won more than half of the votes (and the Conservatives made least progress) in the very urban constituencies (mainly in the cities) but was heavily outscored in those classified as ‘mixed’ or rural. The Liberal Democrats also had their best results in these sorts of areas. 4. Constituency variations in changes in party support Although most media reporting tends to give the impression that modern general elections are essentially national contests between competing party leaders, the fact remains that they involve 650 separate elections in the constituencies. Consequently, local personalities, issues, events and traditions, as well as demographic changes and constituency campaigning at the grass roots, all have a part to play in affecting results. In 2019, in particular, several incumbent Remain—supporting Conservative MPs, who had resigned from the party or had the whip withdrawn, attempted to retain their seats by standing again under a variety of labels. We would expect, then, to find much greater variations in both the direction and extent of changes in party support at constituency, as opposed to regional, level. In Britain as a whole, 29 seats swung to Labour while the size of the swing to the Conservatives varied (in England) from 0.1% (Sefton Central) to a massive 18.4% in Bassetlaw. Although they increased their support by just over four points overall, the Liberal Democrat vote share fell in 38 constituencies (and by more than 10 points in 5) but shot up by more than 20 points in 7. Finally, in Scotland, although the SNP improved its position in all but one constituency the upturn ranged from just over one percentage point to more than 14, while in Wales, in the 36 seats contested in both 2017 and 2019, Plaid fell back in 15, moved forward in 21 and was unchanged in one. Clearly, then, there were significant differences across constituencies in how the distribution of party support changed. A first step in analysing these differences is to consider how changes in support for the various parties were inter-related and Table 1.5 reports the relevant correlation coefficients (focussing on England and Wales in order to avoid complications caused by the strength of the SNP in Scotland). Table 1.5 Correlations between changes in vote shares (England and Wales) . Change . Change . Change . Change . . % Con . % Lab . % Lib Dem . % Brexit . Change % Lab −0.22 — — — Change % Lib Dem −0.50 −0.18 — — Change % Brexit −0.06* −0.55 −0.25 — Change % Green −0.02* 0.03* −0.25 –0.03* . Change . Change . Change . Change . . % Con . % Lab . % Lib Dem . % Brexit . Change % Lab −0.22 — — — Change % Lib Dem −0.50 −0.18 — — Change % Brexit −0.06* −0.55 −0.25 — Change % Green −0.02* 0.03* −0.25 –0.03* Note: The coefficients that are not statistically significant are asterisked. The N in all cases is 571. Here and in other tables relating to changes in votes the Speaker’s seats in 2017 and 2019 are excluded. Open in new tab Table 1.5 Correlations between changes in vote shares (England and Wales) . Change . Change . Change . Change . . % Con . % Lab . % Lib Dem . % Brexit . Change % Lab −0.22 — — — Change % Lib Dem −0.50 −0.18 — — Change % Brexit −0.06* −0.55 −0.25 — Change % Green −0.02* 0.03* −0.25 –0.03* . Change . Change . Change . Change . . % Con . % Lab . % Lib Dem . % Brexit . Change % Lab −0.22 — — — Change % Lib Dem −0.50 −0.18 — — Change % Brexit −0.06* −0.55 −0.25 — Change % Green −0.02* 0.03* −0.25 –0.03* Note: The coefficients that are not statistically significant are asterisked. The N in all cases is 571. Here and in other tables relating to changes in votes the Speaker’s seats in 2017 and 2019 are excluded. Open in new tab Negative coefficients indicate that where one of the parties concerned did better, the other had poorer results while positive figures mean that the two parties tended to rise and fall together. In a strictly two-party system where all constituencies had straight fights, we would find a perfect negative correlation between changes in their vote shares—an increase for one would be matched exactly by a decline for the other. The days of two-party dominance in Britain have long gone, however. This is amply illustrated by the coefficient measuring the relationship between changes in support for the Conservatives and Labour. Although this is in the expected direction (−0.22) and statistically significant, it is not large (and smaller than in 2017 when it stood at −0.27), meaning that changes in vote shares for the two leading parties were not actually very strongly related. This is due to the presence of large numbers of third-, fourth- and fifth-party candidates. Conservative vote changes were much more strongly related to Liberal Democrat performance (coefficient = −0.51). This is not surprising, perhaps, in the context of the 2019 election. The Conservatives campaigned for a swift exit from the EU while the Liberal Democrats advocated remaining a member. The relationship is consistent with the interpretation that a segment of voters with a strong preference for Remain were willing to switch their vote from the former to the latter. Also unsurprising is the fact that Liberal Democrat and Green changes are negatively related as these parties tend to appeal to the same sorts of voters. More puzzling is the fact that stronger Brexit performances were associated with smaller improvements by the Liberal Democrats. The strongest correlation of all, however, relates to the Brexit Party and Labour (−0.55). Clearly, the better the new party’s performance, the more Labour suffered. In contrast, there was no significant relationship between changes in support for the Conservatives and the ‘Brexiteers’. This gives some support to the argument of Nigel Farage during and after the election that his party’s presence helped the Conservatives and hurt Labour. Indeed, he suggested that the former won or came close in a number of seats which they would have had no chance of winning had not a Brexit candidate drawn votes from Labour. There is some substance to this as there were 20 seats won by the Conservatives in which their majority was smaller than the number of votes cast for the Brexit Party. Four of these were in Wales and most of the rest in the North of England and the Midlands. Bizarrely, however, they also include Kensington, and Carshalton and Wallington (both in London) although in these cases it was the tiny majorities rather than the strength of the Brexit candidates that was notable. The impact of Brexit candidacies on support for the other parties more generally is illustrated in Table 1.6, which shows the changes in vote shares for the latter in seats with and without a Brexit candidate in 2019. Table 1.6 The impact of Brexit candidates on changes in vote share 2017–2019 (England and Wales) Brexit candidate 2019 . No . Yes . Difference . Change % Con +1.9 +1.3 −0.6 Change % Lab −7.2 −8.7 −1.5 Change % Lib Dem +5.7 +2.6 −3.1 Change % Green +1.1 +1.4 +0.3 (Constituencies) (312) (259) Brexit candidate 2019 . No . Yes . Difference . Change % Con +1.9 +1.3 −0.6 Change % Lab −7.2 −8.7 −1.5 Change % Lib Dem +5.7 +2.6 −3.1 Change % Green +1.1 +1.4 +0.3 (Constituencies) (312) (259) Open in new tab Table 1.6 The impact of Brexit candidates on changes in vote share 2017–2019 (England and Wales) Brexit candidate 2019 . No . Yes . Difference . Change % Con +1.9 +1.3 −0.6 Change % Lab −7.2 −8.7 −1.5 Change % Lib Dem +5.7 +2.6 −3.1 Change % Green +1.1 +1.4 +0.3 (Constituencies) (312) (259) Brexit candidate 2019 . No . Yes . Difference . Change % Con +1.9 +1.3 −0.6 Change % Lab −7.2 −8.7 −1.5 Change % Lib Dem +5.7 +2.6 −3.1 Change % Green +1.1 +1.4 +0.3 (Constituencies) (312) (259) Open in new tab Across England and Wales, a Brexit presence made only a small difference to the Conservative performance, increased Labour’s decline by 1.5 points and reduced the Liberal Democrat advance by 3.1 points as compared to places where there was no Brexit candidate. In the latter case, given that the two parties espoused diametrically opposite policies it is not easy to explain why Jo Swinson’s party did significantly less well in seats where Brexit was in the field (as with the correlation noted above). This probably reflects the fact that the (Conservative-held) seats that Brexit did not contest are more inclined to the Liberal Democrats in any event although it may also be that, elsewhere, Brexit simply provided an alternative choice for voters unwilling to support either of the two main parties. The overall impact of the Brexit Party on Labour suggested in Table 1.6 may appear modest but it was regionally concentrated. In London the fall in Labour’s votes was almost identical in both categories of seat while in the rest of the south the party actually did better where there was a Brexit candidate than where none was standing. In the northern regions, on the other hand Labour fell by 3.3 points more in the ‘Brexit seats’ than in others while the relevant figure for Wales was 5.2 points. In Scotland, correlating the changes in the parties’ vote shares from 2017 produces a clear pattern. SNP increases were strongly related to Labour declines (coefficient = −0.70) and also, but less strongly, to changes in Conservative support (−0.21). The latter’s performance was more influenced by how well the Liberal Democrats did (−0.54) and this also affected Labour (−0.32). Illustrative of the complex and changed nature of Scottish electoral politics, there was no significant relationship between changes in support for the Conservatives and Labour across the constituencies. Changes in support for the major parties are explored further in Table 1.7 which shows, first, the direction and strength of the relationship between changes in vote shares for the major parties in English and Welsh constituencies and a standard set of social and demographic variables. The latter are taken from the 2011 Census. Although these data are now nine years old, few constituencies will have undergone dramatic changes in their make-up in the intervening period so they can still be usefully applied in analysis. Table 1.7 Bivariate correlations between changes in Conservative and Labour shares of vote 2017–2019 and constituency characteristics (England and Wales) . . . Lib . . . . Lib . . Con . Lab . Dem . . Con . Lab . Dem . Manual Workers (%) 0.66 −0.38 −0.51 % Prof./ Managerial −0.67 0.32 0.56 Owner occupiers (%) 0.14 −0.07* 0.09 % Private renters −0.26 0.29 −0.01* No qualifications (%) 0.62 −0.41 −0.43 % With degrees −0.68 0.35 0.46 Aged 65+ years (%) 0.22 −0.05* 0.02* % Aged 18-24 −0.09 0.17 −0.22 % Students −0.22 0.25 −0.12 Persons per hectare −0.23 0.15 0.02* Leave in EU ref. (%) 0.67 −0.40 −0.27 % Ethnic minority −0.20 0.18 0.02* . . . Lib . . . . Lib . . Con . Lab . Dem . . Con . Lab . Dem . Manual Workers (%) 0.66 −0.38 −0.51 % Prof./ Managerial −0.67 0.32 0.56 Owner occupiers (%) 0.14 −0.07* 0.09 % Private renters −0.26 0.29 −0.01* No qualifications (%) 0.62 −0.41 −0.43 % With degrees −0.68 0.35 0.46 Aged 65+ years (%) 0.22 −0.05* 0.02* % Aged 18-24 −0.09 0.17 −0.22 % Students −0.22 0.25 −0.12 Persons per hectare −0.23 0.15 0.02* Leave in EU ref. (%) 0.67 −0.40 −0.27 % Ethnic minority −0.20 0.18 0.02* Note: N = 571. All coefficients are statistically significant except those asterisked. Open in new tab Table 1.7 Bivariate correlations between changes in Conservative and Labour shares of vote 2017–2019 and constituency characteristics (England and Wales) . . . Lib . . . . Lib . . Con . Lab . Dem . . Con . Lab . Dem . Manual Workers (%) 0.66 −0.38 −0.51 % Prof./ Managerial −0.67 0.32 0.56 Owner occupiers (%) 0.14 −0.07* 0.09 % Private renters −0.26 0.29 −0.01* No qualifications (%) 0.62 −0.41 −0.43 % With degrees −0.68 0.35 0.46 Aged 65+ years (%) 0.22 −0.05* 0.02* % Aged 18-24 −0.09 0.17 −0.22 % Students −0.22 0.25 −0.12 Persons per hectare −0.23 0.15 0.02* Leave in EU ref. (%) 0.67 −0.40 −0.27 % Ethnic minority −0.20 0.18 0.02* . . . Lib . . . . Lib . . Con . Lab . Dem . . Con . Lab . Dem . Manual Workers (%) 0.66 −0.38 −0.51 % Prof./ Managerial −0.67 0.32 0.56 Owner occupiers (%) 0.14 −0.07* 0.09 % Private renters −0.26 0.29 −0.01* No qualifications (%) 0.62 −0.41 −0.43 % With degrees −0.68 0.35 0.46 Aged 65+ years (%) 0.22 −0.05* 0.02* % Aged 18-24 −0.09 0.17 −0.22 % Students −0.22 0.25 −0.12 Persons per hectare −0.23 0.15 0.02* Leave in EU ref. (%) 0.67 −0.40 −0.27 % Ethnic minority −0.20 0.18 0.02* Note: N = 571. All coefficients are statistically significant except those asterisked. Open in new tab The pattern of correlation coefficients shown for the Conservatives and Labour is very similar to the pattern found in 2017. In this election, as in the previous one, the inter-election change in support was better for the former and worse for the latter the more manual workers and people without educational qualifications there were in a constituency. In contrast, the more professional and managerial workers and people with a degree, the worse was the change in vote share for the Conservatives and the better (or less bad) the outcome for Labour. In short, working-class areas moved more from Labour to the Conservatives while middle-class areas moved less. One change from the comparable analysis of 2017 results concerns young people and students. In that election a promise to abolish university tuition fees had featured prominently in Labour’s campaign and the change in its vote share correlated positively with the percentage of those aged 18–24 years and the percentage of students in a constituency (0.34 in both cases) while Conservative change was negatively related (−0.20 and −0.35 respectively). In 2019, Labour’s policy on tuition fees remained the same but, as can be seen in Table 1.7, although the coefficients were in the same direction they were rather less strong than they had been previously. As far as the Liberal Democrats are concerned, their advance was clearly strongest in more middle-class constituencies and weakest in those that are more working-class. Repeating this analysis for Scotland reveals that there was no significant relationship between any of the social and demographic variables and changes in support for Labour and the SNP. Here, it was more a case of the nationalists gaining and Labour losing ground across the board. Again, as in 2017, the kinds of places where the Conservatives did best in comparison to the previous election are strikingly similar to those which had the largest Leave majorities in the 2016 EU referendum. Unsurprisingly, then, there is a strong positive correlation (0.67) between the estimated percentage Leave vote and the change in Conservative vote share and negative relationships for Labour (−0.40) and the Liberal Democrats (−0.27). The more Leave voters there were in a constituency, the better the Conservatives did and the worse the outcome for Labour and the Liberal Democrats, as compared with the results in 2017. This pattern was reversed, of course, the more Remain voters there were. 5. Patterns of party support in 2019 Having examined the extent and nature of electoral change between 2017 and 2019, the question remains as to what extent the pattern of absolute levels of party support across constituencies continues to display long-familiar features. Table 1.8 shows correlation coefficients measuring the associations between the shares of the vote obtained by the parties and a range of social and demographic variables. Table 1.8 Bivariate correlations between party shares of vote in 2019 and constituency characteristics (England and Wales) . . . Liberal . . Conservative . Labour . Democrat . Prof./Managerial (%) 0.09 −0.34 0.66 Manual Workers (%) −0.07* 0.28 −0.61 Owner occupiers (%) 0.70 −0.68 0.14 Social renters (%) −0.63 0.66 −0.29 Private renters (%) −0.52 0.46 0.07* Aged 18–24 years (%) −0.53 0.51 −0.13 Aged 65+ years (%) 0.63 −0.66 0.16 In agriculture (%) 0.39 −0.45 0.18 Persons per hectare −0.62 0.61 −0.05* With degrees (%) 0.16 −0.10 0.61 No qualifications (%) −0.04* 0.25 −0.61 Students (%) −0.56 0.50 0.01* With no car (%) −0.76 0.77 −0.26 Ethnic minority (%) −0.54 0.59 −0.11 . . . Liberal . . Conservative . Labour . Democrat . Prof./Managerial (%) 0.09 −0.34 0.66 Manual Workers (%) −0.07* 0.28 −0.61 Owner occupiers (%) 0.70 −0.68 0.14 Social renters (%) −0.63 0.66 −0.29 Private renters (%) −0.52 0.46 0.07* Aged 18–24 years (%) −0.53 0.51 −0.13 Aged 65+ years (%) 0.63 −0.66 0.16 In agriculture (%) 0.39 −0.45 0.18 Persons per hectare −0.62 0.61 −0.05* With degrees (%) 0.16 −0.10 0.61 No qualifications (%) −0.04* 0.25 −0.61 Students (%) −0.56 0.50 0.01* With no car (%) −0.76 0.77 −0.26 Ethnic minority (%) −0.54 0.59 −0.11 Note: N = 572 for Conservatives and Labour and 552 for the Liberal Democrats. All coefficients are statistically significant except those asterisked. Open in new tab Table 1.8 Bivariate correlations between party shares of vote in 2019 and constituency characteristics (England and Wales) . . . Liberal . . Conservative . Labour . Democrat . Prof./Managerial (%) 0.09 −0.34 0.66 Manual Workers (%) −0.07* 0.28 −0.61 Owner occupiers (%) 0.70 −0.68 0.14 Social renters (%) −0.63 0.66 −0.29 Private renters (%) −0.52 0.46 0.07* Aged 18–24 years (%) −0.53 0.51 −0.13 Aged 65+ years (%) 0.63 −0.66 0.16 In agriculture (%) 0.39 −0.45 0.18 Persons per hectare −0.62 0.61 −0.05* With degrees (%) 0.16 −0.10 0.61 No qualifications (%) −0.04* 0.25 −0.61 Students (%) −0.56 0.50 0.01* With no car (%) −0.76 0.77 −0.26 Ethnic minority (%) −0.54 0.59 −0.11 . . . Liberal . . Conservative . Labour . Democrat . Prof./Managerial (%) 0.09 −0.34 0.66 Manual Workers (%) −0.07* 0.28 −0.61 Owner occupiers (%) 0.70 −0.68 0.14 Social renters (%) −0.63 0.66 −0.29 Private renters (%) −0.52 0.46 0.07* Aged 18–24 years (%) −0.53 0.51 −0.13 Aged 65+ years (%) 0.63 −0.66 0.16 In agriculture (%) 0.39 −0.45 0.18 Persons per hectare −0.62 0.61 −0.05* With degrees (%) 0.16 −0.10 0.61 No qualifications (%) −0.04* 0.25 −0.61 Students (%) −0.56 0.50 0.01* With no car (%) −0.76 0.77 −0.26 Ethnic minority (%) −0.54 0.59 −0.11 Note: N = 572 for Conservatives and Labour and 552 for the Liberal Democrats. All coefficients are statistically significant except those asterisked. Open in new tab The figures relating to occupational class are striking. The level of Conservative support was not significantly related to variations in the proportion of manual workers in a constituency and very weakly related to the size of the solidly middle-class professional and managerial groups. The relevant coefficients for Labour are in the expected direction but do not indicate particularly strong relationships. In 2019, indeed, it is the Liberal Democrats whose local support was most strongly influenced by the class composition of constituencies: the more middle-class the constituency, the better they did. The weak class-party relationships in the case of the major parties did not come out of the blue, as it were, but represent a continuation and acceleration of a longer-term trend. This is illustrated in Figure 1.4 which charts the strength of the correlations between the occupational class variables and support for the Conservatives and Labour since the 2005 election. Figure 1.4 Open in new tabDownload slide Correlations between occupational class and constituency vote shares for major parties, 2005–2019 Figure 1.4 Open in new tabDownload slide Correlations between occupational class and constituency vote shares for major parties, 2005–2019 In Labour’s case, there has been a steady decline in the strength of these relationships which have become weaker at each successive election. On the Conservative side, the influence of the size of the middle class on their vote declined only a little to 2015 but then dropped sharply in the next two elections. It was in the 2015 election, however, that the weakening (negative) influence of the proportion of manual workers on variations in Conservative support became more apparent and this trend continued apace in 2017 and 2019. It should be emphasised that this sort of analysis of aggregate data does not, of itself, tell us anything about the extent of class voting among the electorate. It concerns variations among places (constituencies) not people. Nonetheless, opinion poll data relating to the 2019 election confirm the impression that class was not a major influence on party choice. An ‘on the day poll’ conducted on behalf of Lord Ashcroft reported that non-manual workers divided 44% Conservative to 32% Labour while for manual workers the respective figures were 46% and 34% (see https://lordashcroftpolls.com/2019/12/how-britain-voted-and-why-my-2019-general-election-post-vote-poll/). Returning to the data in Table 1.8, the type of housing prevalent in constituencies continued to influence election outcomes with those having more owner occupiers being more inclined to the Conservatives. Two elections ago it would have been surprising to find no Conservative benefit in seats where people with degrees were thicker on the ground but Brexit has changed that relationship. In 2019, however, larger proportions of graduates made for happier hunting grounds for the Liberal Democrats. University towns (and university areas in the cities) also proved more difficult for the Conservatives but provided a boost for Labour, as indicated by the coefficients for the percentage of students aged over 18 years in the constituencies. The remaining variables—age, % in agriculture, persons per hectare, those with no car and the proportion of ethnic minority residents—are all inter-related to a degree. The latter two, as well as younger people, are more numerous in cities while population density is a direct measure of urbanisation. As noted above, Labour remains especially strong in cities and it is unsurprising that the correlations involving these variables are all in the expected direction. The strength of the associations between levels of party support and the age structure of the relevant electorates is worth noting but the best single predictor of the share of votes obtained by the Conservatives and Labour is the percentage of household without a car (−0.76 for the former and +0.77 for the latter). Although it might be tempting to interpret this as indicating widespread dissatisfaction with the abysmal performance of some privatised railway companies in the years and months before the election, the likelihood is that it reflects levels of deprivation as well as the fact that car-owning is less necessary in city locations. In Scotland, the 2017 election appeared to confirm the view that the SNP had successfully taken over what were formerly the bases of Labour dominance there and, in general, this remained the case in 2019. Although the party advanced almost everywhere and the correlations are weaker, the size of the SNP vote in constituencies was positively and significantly associated with % social renters (0.45), % with no car (0.37) and % with no educational qualifications (0.23) while there were negative effects for % owner occupiers (−0.40), % professional and managerial (−0.39), % in agriculture (−0.34) and % aged 65 and over (−0.30). 6. Turnout Measuring turnout in Britain is not as straightforward as it may seem. It is defined as the percentage of those on the electoral register who cast a ballot. Unfortunately, the media (and even the House of Commons Library) generally base the calculation on the sum of votes cast for all parties and candidates that is the number of valid votes. However, this excludes those who voted but whose ballots were rejected for one reason or another—people who deliberately spoiled their paper, for example. In most cases, of course, the difference between valid votes and total ballots is small but in the Chorley constituency of the Speaker in 2019 there were 1,303 rejected ballots and including these clearly makes a difference to the turnout calculation. Across Britain, almost 113,000 ballots were rejected—about 42,000 more than in 2017. A more recent problem in calculating turnout relates to postal voting. Following the introduction of ‘on demand’ postal voting just before the 2001 general election, the proportion of votes cast in this way has risen steadily—from 2.4% of those included in the count in 1997 to 21.6% in 2017. Voting by post is more complicated than voting in person, however, and, as a consequence, a large number of postal ballots—in 2017, more than 140,000 (2.4% of all postal votes)—are disqualified on technical grounds. Whether these (attempted) votes should be included in the turnout calculation is a moot point. Some may be genuine attempts which involved a minor slip; others may be fraudulent. Measuring the ‘eligible electorate’ also raises issues. Electoral registers obviously exclude those who, for one reason or another, have never got round to (the relatively simple task of) adding their names. On the other hand, it is possible to be registered (but not, legally, to vote) in more than one place. Students, for example, may be registered at home and at their university. Since electoral registration is the responsibility of local authorities, however, there is no central register and hence no way of knowing the number of names entered more than once. Despite these niggling difficulties, the turnout figures given here refer to total votes cast (including those rejected at the count) as a percentage of the registered electorate. In advance of polling day, there were fears that an election held in the depths of winter would result in a significant drop in turnout but, although the weather was indeed wet and windy across much of the country, the Britain-wide figure of 67.7% was only slightly down on 2017. Table 1.9 shows that participation by the electorate was little changed in the South East and South West and increased in Scotland. Elsewhere, the largest decline (−2.7 points) was recorded in London (which had had the largest rise in 2017) with most of the others being close to a two point drop. In absolute terms, the southern regions outside London continue to have a somewhat higher turnout (70.4% taken together) than the three northern regions (65.1%). Table 1.9 Turnout in 2019 (%) . Turnout 2019 . Change 2017–2019 . North East 64.3 −1.9 North West 65.8 −2.2 Yorkshire/Humber 64.6 −1.9 East Midlands 67.4 −1.9 West Midlands 64.9 −2.0 Eastern 68.6 −1.4 London 67.6 −2.7 South East 70.5 −0.8 South West 72.3 +0.3 Wales 66.8 −1.9 Scotland 68.2 +1.7 Great Britain 67.7 −1.3 . Turnout 2019 . Change 2017–2019 . North East 64.3 −1.9 North West 65.8 −2.2 Yorkshire/Humber 64.6 −1.9 East Midlands 67.4 −1.9 West Midlands 64.9 −2.0 Eastern 68.6 −1.4 London 67.6 −2.7 South East 70.5 −0.8 South West 72.3 +0.3 Wales 66.8 −1.9 Scotland 68.2 +1.7 Great Britain 67.7 −1.3 Open in new tab Table 1.9 Turnout in 2019 (%) . Turnout 2019 . Change 2017–2019 . North East 64.3 −1.9 North West 65.8 −2.2 Yorkshire/Humber 64.6 −1.9 East Midlands 67.4 −1.9 West Midlands 64.9 −2.0 Eastern 68.6 −1.4 London 67.6 −2.7 South East 70.5 −0.8 South West 72.3 +0.3 Wales 66.8 −1.9 Scotland 68.2 +1.7 Great Britain 67.7 −1.3 . Turnout 2019 . Change 2017–2019 . North East 64.3 −1.9 North West 65.8 −2.2 Yorkshire/Humber 64.6 −1.9 East Midlands 67.4 −1.9 West Midlands 64.9 −2.0 Eastern 68.6 −1.4 London 67.6 −2.7 South East 70.5 −0.8 South West 72.3 +0.3 Wales 66.8 −1.9 Scotland 68.2 +1.7 Great Britain 67.7 −1.3 Open in new tab At constituency level, of course, there was much greater variation in turnout. At the bottom end, 15 constituencies failed to reach 55%, Hull East coming bottom of the pile with a turnout of 49.5%. At the other extreme, the figure was 75% or greater in 61 constituencies with the highest figure of all (80.4%) being recorded in East Dunbartonshire where the Liberal Democrat leader, Jo Swinson, lost her seat by 149 votes. Across England and Wales, the correlations between the change in constituency turnouts and changes in the parties’ vote shares are relatively weak, being −0.21 for the Conservatives, +0.20 for Labour and +0.23 for the Liberal Democrats. There was, then, a slight tendency for the latter two parties to have better results and the former to do less well where turnout declined least. In examining the sources of turnout variations across constituencies we encounter a familiar pattern. Table 1.10 shows correlations between the level of turnout in 2019 and variables indicating the socio-economic characteristics of constituencies. Table 1.10 Bivariate correlations between turnout in 2019 and constituency characteristics (Great Britain) Professional & Managerial (%) . 0.73 . % Manual Workers . −0.70 . Owner occupiers (%) 0.47 % Social renters −0.59 % Private renters −0.13 In agriculture (%) 0.31 Persons per hectare −0.27 With degrees (%) 0.63 % No qualifications −0.69 Aged 65+ years (%) 0.37 % Aged 18–24 −0.29 % Students −0.16 % Ethnic minority −0.28 % With no car −0.53 Constituency marginality 2017 0.13 Professional & Managerial (%) . 0.73 . % Manual Workers . −0.70 . Owner occupiers (%) 0.47 % Social renters −0.59 % Private renters −0.13 In agriculture (%) 0.31 Persons per hectare −0.27 With degrees (%) 0.63 % No qualifications −0.69 Aged 65+ years (%) 0.37 % Aged 18–24 −0.29 % Students −0.16 % Ethnic minority −0.28 % With no car −0.53 Constituency marginality 2017 0.13 Note: For all variables except marginality N = 631 (Chorley excluded) and for marginality N = 630 (Buckingham also excluded). All coefficients are statistically significant. Open in new tab Table 1.10 Bivariate correlations between turnout in 2019 and constituency characteristics (Great Britain) Professional & Managerial (%) . 0.73 . % Manual Workers . −0.70 . Owner occupiers (%) 0.47 % Social renters −0.59 % Private renters −0.13 In agriculture (%) 0.31 Persons per hectare −0.27 With degrees (%) 0.63 % No qualifications −0.69 Aged 65+ years (%) 0.37 % Aged 18–24 −0.29 % Students −0.16 % Ethnic minority −0.28 % With no car −0.53 Constituency marginality 2017 0.13 Professional & Managerial (%) . 0.73 . % Manual Workers . −0.70 . Owner occupiers (%) 0.47 % Social renters −0.59 % Private renters −0.13 In agriculture (%) 0.31 Persons per hectare −0.27 With degrees (%) 0.63 % No qualifications −0.69 Aged 65+ years (%) 0.37 % Aged 18–24 −0.29 % Students −0.16 % Ethnic minority −0.28 % With no car −0.53 Constituency marginality 2017 0.13 Note: For all variables except marginality N = 631 (Chorley excluded) and for marginality N = 630 (Buckingham also excluded). All coefficients are statistically significant. Open in new tab If the occupational class structure has ceased to be a reliable indicator of support for the major parties it nonetheless remains strongly associated with turnout levels. The more middle-class a constituency, the higher the turnout; the more working class, the lower the turnout. Other class-related variables—housing tenure, education and car ownership—show a similar pattern. Turnout is also lower the more urban a constituency and the more young people and ethnic minority citizens there are but the effect of the size of the student population is slight. It is worth mentioning again the point that this analysis does not tell us the extent to which people in the various groups listed turned out to vote. Rather, it tells us that the more professionals, owner occupiers, people with degrees, people employed in agriculture and older people there are in a constituency, the higher was its turnout. In the two previous elections, for reasons that no-one has been able to explain satisfactorily, the formerly strong relationship between marginality—the closeness of the contest in the preceding election—and turnout, which persisted over many years, disappeared. In 2019, however, there was something of a return to normality—the more marginal the seat the higher the turnout—but the relationship was very weak so that there remains a puzzle for electoral analysts given that the parties campaign harder than ever in more marginal seats while expending less effort in those that are hopeless prospects or very safe for them. 7. Explaining the outcome Although the size of the majority won by the Conservatives surprised many observers, the 2019 outcome could not have astonished them in the way that the 2017 contest had. After all, as has been seen, Labour lagged well behind in voting intentions throughout the campaign, the Conservatives avoided the gaffes that had derailed them in 2017 and the decline of the Brexit Party and the Liberal Democrats as the campaign progressed had been clearly charted. In Scotland, too, the progress made by the SNP was not unexpected (although Labour’s decline was probably greater than anticipated). The various factors which explain what happened in 2019 are explored elsewhere in this volume and will, no doubt, be thoroughly examined in future publications. As soon as the results of the exit poll were announced, however, and in the immediate aftermath of the election, attention focussed on Labour’s dismal performance and three topics dominated the discussion—Brexit, ‘Corbynism’ and Jeremy Corbyn himself. As far as Brexit is concerned, there was a widespread belief—certainly among Leave supporters—that Labour had helped to stymie the process by voting against withdrawal bills in the Commons, supporting extensions to the dates set for departing from the EU and shilly-shallying about whether a second referendum should be held. In addition, its own policy on the issue was unclear—at least as compared with those of its rivals. There seems little doubt that this hurt the party in the election. As has been seen, Labour lost votes to the Brexit Party and did worse than average in Leave-supporting areas. The Ashcroft poll referred to earlier found that among 2016 Leave voters 73% voted Conservative in 2019 while only 26% voted Labour. Among Remainers, the figures were 20% Conservative, 47% Labour and 21% Liberal Democrat. It is worth noting also—although it is largely irrelevant to Labour’s performance—that none of the Remain-supporting ex-Conservative MPs who sought re-election mentioned above came anywhere near success. Dominic Grieve, for example, who was perhaps the most prominent, lost by more than 15,000 votes in his constituency. If the 2017 election result could be characterised as ‘the revenge of the Remainers’, then it would appear that Leavers got some of their own back in 2019. By ‘Corbynism’ is meant the generally left-wing policies espoused by Labour under Corbyn’s leadership. Whether this contributed to Labour’s poor showing is rather more debatable. Certainly, some of the individual policies put forward in Labour’s manifesto—the proposal to renationalise the railways, for example, or to greatly increase spending on the National Health Service—commanded wide support. It may be, however, that voters were sceptical about the sheer scale of the spending commitments undertaken by the party and worried about where the money would come from to finance them. Reliable survey data will be required to throw more light on these issues. There is rather less room for doubt about the electoral impact of the Labour leader himself. As described above, he lagged well behind Theresa May as the electorate’s preferred Prime Minister for most of the inter-election period and the gap widened when Boris Johnson took over the top job. On another long-standing measure of leader popularity—Ipsos MORI’s satisfaction ratings—the news for Corbyn supporters was equally bleak. From February to June 2019 his average net score (% satisfied minus % dissatisfied with his performance) was −57 (compared with −37 for Mrs May) while in September and October, during Johnson’s brief ‘honeymoon’ period he plumbed new depths for an Opposition leader recording scores of −60 in both months. Corbyn’s ratings improved somewhat during the election campaign—in Ipsos MORI’s December figures he stood at −44 compared with −20 for Johnson—but there was no miraculous comeback such as had occurred in 2017. A plethora of polls confirmed Johnson’s lead as the preferred Prime Minister and just a week before the election Opinium asked respondents whether they could imagine Jeremy Corbyn holding the office. Only 29% said ‘Yes’ while 59% gave a negative response. Even among intending Labour voters 24% said that they could not imagine him as PM. These people had company in the Labour party. In what was surely an unprecedented development, 15 former Labour MPs placed advertisements in local newspapers on the eve of the election arguing that Corbyn was unfit to be Prime Minister. Whatever it was that made people warm to the Labour leader during the previous election campaign was patently absent on this occasion. Not long after the election, Brexit was finally achieved and Jeremy Corbyn announced that he would resign as Labour leader. Whether ‘Corbynism’ will continue to dominate Labour thinking remains to be seen. By the time the next general election comes round, however, it is certain that new issues, new personalities and simple ‘events’—including the still uncertain consequences of Brexit—will provide a new context and new questions on which a fickle and volatile electorate will be asked to pass judgement. Acknowledgements I am grateful to Mark Garnett and Rob Johns for comments on a draft of this chapter. Reference Daily Telegraph, ‘Wednesday Morning News Briefing: General Election on a Knife Edge’, accessed at https://www.telegraph.co.uk/news/2019/12/11/wednesday-morning-news-briefing-general-election-knife-edge/ on 31 March 2020 . © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Curtice,, John

doi: 10.1093/pa/gsaa021pmid: N/A

1. Introduction Recently, Britain’s single-member plurality electoral system has not been living up to its traditional billing. Its advocates argue that because it is more or less guaranteed to give the winning party an overall majority of seats even though that party may win considerably less than 50% of the vote, the system helps facilitate alternating single-party majority government and ensures governments are directly accountable to voters (Norton, 1997; Bingham Powell, 2000; Renwick, 2011). Yet in both 2010 and 2017, the election produced a hung parliament that in one instance resulted in a coalition government and, in the other, a minority administration underpinned by a ‘confidence and supply’ arrangement with a small party. Meanwhile, although the 2015 contest did produce an overall majority, at 12 it was relatively small and indeed was deemed to be too small when the then Prime Minister, Theresa May, triggered the 2017 election in the hope that she would win a larger majority. But in the 2019 election, ‘normality’ was seemingly restored. The Conservatives were elected with an overall majority of 80. As a result, what had been a long-running impasse in the House of Commons about what should be done about Brexit was swiftly resolved, and less than two months after polling day, the UK left the EU. There could, it seemed, be no clearer demonstration of the claim that single-party majority administrations deliver ‘strong government’—in contrast to the weakness and immobilisme that are often thought to characterize coalitions and minority government and which appeared to be exemplified by the impasse over Brexit that had gripped the Commons between 2017 and 2019. In this article, we look more closely at how the electoral system operated in 2019. To what extent did it represent a return to normality—and the result indicate that overall majorities are likely to be the norm once more? Or was this in some sense an exceptional result that provides little indication as to what might happen in future? Why did the system work as it did in the 2019 contest—and how important was its use in determining the fate of Brexit? 2. Three comparisons In the election, the Conservatives secured a lead in votes of nearly 12 percentage points over Labour. It was the largest lead to be won by either party since 1997 when Labour was nearly 13 points ahead of the Conservatives. As Table 2.1 shows, since 1979, only one other contest—Mrs Thatcher’s landslide success in 1983—witnessed a significantly bigger lead for the winning party than the one that was in evidence in 2019. Against that backdrop, it would have been very surprising indeed if the 2019 contest had not resulted in anything other than a substantial majority for the winning party. Table 2.1 Winning party’s lead in votes and size of overall majority, 1979–2019 Election . % Lead in votes . Overall majority (seats) . 1979 7.0 43 1983 15.2 144 1987 11.8 102 1992 7.6 21 1997 12.8 179 2001 9.3 167 2005 2.9 66 2010 7.2 None 2015 6.6 12 2017 2.5 None 2019 11.7 80 Election . % Lead in votes . Overall majority (seats) . 1979 7.0 43 1983 15.2 144 1987 11.8 102 1992 7.6 21 1997 12.8 179 2001 9.3 167 2005 2.9 66 2010 7.2 None 2015 6.6 12 2017 2.5 None 2019 11.7 80 Source: Calculated from Rallings and Thrasher (2012), Cowley and Kavanagh (2016, 2018) and Baker et al. (2019). Open in new tab Table 2.1 Winning party’s lead in votes and size of overall majority, 1979–2019 Election . % Lead in votes . Overall majority (seats) . 1979 7.0 43 1983 15.2 144 1987 11.8 102 1992 7.6 21 1997 12.8 179 2001 9.3 167 2005 2.9 66 2010 7.2 None 2015 6.6 12 2017 2.5 None 2019 11.7 80 Election . % Lead in votes . Overall majority (seats) . 1979 7.0 43 1983 15.2 144 1987 11.8 102 1992 7.6 21 1997 12.8 179 2001 9.3 167 2005 2.9 66 2010 7.2 None 2015 6.6 12 2017 2.5 None 2019 11.7 80 Source: Calculated from Rallings and Thrasher (2012), Cowley and Kavanagh (2016, 2018) and Baker et al. (2019). Open in new tab Yet, while there have only been two elections during the last 40 years in which the winning party had a markedly bigger lead in votes than transpired in 2019, there were as many as four elections in which the winning party’s majority was bigger than the 80-seat one the Conservatives secured in 2019. Of particular note is the outcome of the 1987 election, in which the Conservative lead over Labour (11.7 points) was almost exactly the same as that in 2019, but the outcome was a majority of 102 rather than 80. Meanwhile, what was no more than a 9.3-point lead in 2001 resulted in a Labour majority of as many as 167 seats. It seems that the past record might have led us to expect a rather larger overall majority for the Conservatives than the one of 80 that they obtained. However, perhaps some inconsistency between the lead a party secures in votes and the outcome in seats should not surprise us. After all, under the single-member plurality electoral system, there is no formal mechanism that ensures that a party that wins any particular share of the vote across the country as a whole secures any particular proportion of the seats in the House of Commons. A general election simply consists of 650 separate constituency contests in which the (sole) winner in each case is the candidate with most votes. Nevertheless, it was observed in the immediate post-war period that in practice the electoral system appeared to produce results that were consistent with a ‘cube law’ (Butler, 1951; Gudgin and Taylor, 1979; Taagerpera and Shugart, 1989). This states that if the votes cast for the two largest parties are shared in the proportion A: B, the seats they win are shared in the proportion A3: B3—a ratio that exaggerates the lead of the largest party over its principal rival and—assuming few seats are won by any other parties—should ensure that in all but the narrowest of outcomes the party with most votes obtains a substantial overall majority. However, large though it might seem, the Conservatives’ majority in 2019 falls well short of the expectations of the cube law. In the election, the Conservatives won 57.6% of the votes cast for Conservative and Labour alone (otherwise known as the ‘two-party vote’). According to the law, that should have reaped a dividend of 71.4% of the seats—or 406, well above the Conservatives’ actual tally of 365.1 So, by this criterion too, it looks as though the Conservatives’ majority was rather less than we might have anticipated. The expectation that the electoral system should usually deliver the winner a substantial majority, even if that party secures much less than half of all votes cast, rests not only on its supposed tendency to operate in conformity to the cube law, but also on the expectation that few MPs from parties other than the Conservatives or Labour will be elected to the House of Commons. Even if some voters do vote for smaller parties, the winner take all nature of the system in every constituency is expected to make it difficult for those parties to win any seats—a feature that in turn is thought to discourage voters from voting for smaller parties in the first place (Duverger, 1954; Blais and Carty, 1991). Yet plenty of third-party MPs were elected in 2019—a grand total of not less than 82. In this respect, the outcome did not look anything like ‘normality’ at all. True, this tally was a little lower than in the three elections between 2005 and 2015, but it was still well above anything that was recorded before 1997. What has seemingly become a persistent ‘failure’ of the system at recent elections to exclude third parties from the Commons was once again in evidence. So, while the electoral system did deliver a substantial majority for the largest party in 2019, once we look underneath the surface there is reason to question whether the outcome did represent a return to ‘normality’. Whether judged by historical precedent or the expectations of the ‘cube law’, the Conservatives’ overall majority was rather less than might have been expected given the extent of their lead over Labour. Meanwhile, there were still many more third-party MPs elected than would once have seemed imaginable. There is evidently a need to examine further how the electoral system worked in 2019. 3. Marginal seats The early post-war literature that originally identified the ‘cube law’ did not indicate that it was an inevitable feature of how the system operated. Rather, it demonstrated that the way the single member plurality system translates votes into seats depends on the geography of party support—and in particular on how many seats are closely contested between the Conservatives and Labour (Kendall and Stuart, 1950). Formally, it indicated that the frequency distribution of the Conservative share of the two-party vote across all constituencies should be similar to that of a normal curve with a standard deviation (a measure of spread) of 13.7. In practice, what this means is that in a 600-seat legislature, 180 (or 30%) of the seats should be ones where—if the two largest parties enjoy the same share of the vote across the country as a whole—the share of the two-party vote secured by the winning party should be between 50% and 55%. In short, the cube law works so long as there are sufficient marginal seats. Table 2.2 assesses the geographical distribution of the two-party vote against these criteria at each election since 1955. In the first pair of columns, it shows the number of seats that satisfy our definition of a ‘marginal’ seat, and the proportion of the seats won by Conservative and Labour that this figure represents. In the second pair of columns, we give an indication of how the distribution of party support across all constituencies measures up against the more formal requirements that need to be satisfied for the law to operate. We show, first, the standard deviation of the distribution and secondly, the kurtosis, a measure of the extent to which there are more or fewer seats in the middle of the distribution than would be expected from a normal curve. Table 2.2 Distribution of the two-party vote, 1955–2019 Party . Marginals . Two-party vote . . No. . % . Standard deviation . Kurtosis . 1955 166 27.2 13.5 −0.25 1959 157 25.7 13.8 −0.29 1964 166 27.3 14.1 −0.45 1966 155 25.6 13.8 −0.46 1970 149 24.5 14.3 −0.27 1974 (Feb) 119 19.9 16.1 −0.68 1974 (Oct) 98 16.4 16.8 −0.82 1979 108 17.8 16.9 −0.87 1983 80 13.2 20.0 −1.05 1987 87 14.4 21.4 −1.03 1992 98 16.1 20.2 −1.03 1997 114 19.6 18.1 −0.85 2001 114 19.7 18.3 −0.82 2005 104 18.8 19.7 −0.96 2010 85 15.0 22.2 −1.08 2015 74 13.1 21.7 −1.19 2017 89 15.3 18.6 −0.89 2019 88 15.5 20.3 −0.69 Party . Marginals . Two-party vote . . No. . % . Standard deviation . Kurtosis . 1955 166 27.2 13.5 −0.25 1959 157 25.7 13.8 −0.29 1964 166 27.3 14.1 −0.45 1966 155 25.6 13.8 −0.46 1970 149 24.5 14.3 −0.27 1974 (Feb) 119 19.9 16.1 −0.68 1974 (Oct) 98 16.4 16.8 −0.82 1979 108 17.8 16.9 −0.87 1983 80 13.2 20.0 −1.05 1987 87 14.4 21.4 −1.03 1992 98 16.1 20.2 −1.03 1997 114 19.6 18.1 −0.85 2001 114 19.7 18.3 −0.82 2005 104 18.8 19.7 −0.96 2010 85 15.0 22.2 −1.08 2015 74 13.1 21.7 −1.19 2017 89 15.3 18.6 −0.89 2019 88 15.5 20.3 −0.69 Marginal Seat: Seat where Conservative share of two-party vote − (overall Conservative share of two-party vote - 50%) lies within the range 45–55%. Two-party vote: Votes cast for Conservative and Labour combined. Table based only on seats won by Conservative or Labour at that election and contested by both parties. Source:Curtice (2017a) and author’s calculations. Open in new tab Table 2.2 Distribution of the two-party vote, 1955–2019 Party . Marginals . Two-party vote . . No. . % . Standard deviation . Kurtosis . 1955 166 27.2 13.5 −0.25 1959 157 25.7 13.8 −0.29 1964 166 27.3 14.1 −0.45 1966 155 25.6 13.8 −0.46 1970 149 24.5 14.3 −0.27 1974 (Feb) 119 19.9 16.1 −0.68 1974 (Oct) 98 16.4 16.8 −0.82 1979 108 17.8 16.9 −0.87 1983 80 13.2 20.0 −1.05 1987 87 14.4 21.4 −1.03 1992 98 16.1 20.2 −1.03 1997 114 19.6 18.1 −0.85 2001 114 19.7 18.3 −0.82 2005 104 18.8 19.7 −0.96 2010 85 15.0 22.2 −1.08 2015 74 13.1 21.7 −1.19 2017 89 15.3 18.6 −0.89 2019 88 15.5 20.3 −0.69 Party . Marginals . Two-party vote . . No. . % . Standard deviation . Kurtosis . 1955 166 27.2 13.5 −0.25 1959 157 25.7 13.8 −0.29 1964 166 27.3 14.1 −0.45 1966 155 25.6 13.8 −0.46 1970 149 24.5 14.3 −0.27 1974 (Feb) 119 19.9 16.1 −0.68 1974 (Oct) 98 16.4 16.8 −0.82 1979 108 17.8 16.9 −0.87 1983 80 13.2 20.0 −1.05 1987 87 14.4 21.4 −1.03 1992 98 16.1 20.2 −1.03 1997 114 19.6 18.1 −0.85 2001 114 19.7 18.3 −0.82 2005 104 18.8 19.7 −0.96 2010 85 15.0 22.2 −1.08 2015 74 13.1 21.7 −1.19 2017 89 15.3 18.6 −0.89 2019 88 15.5 20.3 −0.69 Marginal Seat: Seat where Conservative share of two-party vote − (overall Conservative share of two-party vote - 50%) lies within the range 45–55%. Two-party vote: Votes cast for Conservative and Labour combined. Table based only on seats won by Conservative or Labour at that election and contested by both parties. Source:Curtice (2017a) and author’s calculations. Open in new tab The table shows that there has been a marked change in the geography of the two-party vote during the post-war period. In the 1950s and 1960s that geography more or less satisfied the requirements of the cube law (Curtice and Steed, 1982). The standard deviation was around 13.7. The kurtosis was a little less than that of a normal curve—and thus the proportion of seats that were marginal a little less than 30%—but it can be seen that overall the electoral system could be expected to produce an outcome that would more or less satisfy the expectations of the cube law. However, at the two elections in 1974, there was a marked drop in the number of marginal seats, and it fell yet further thereafter. Although thanks to her large lead in votes Mrs Thatcher still won a hefty three-figure majority in 1983, by this stage, the number of marginal seats had halved and the standard deviation of the two-party vote increased to 20. If the elections held in the 1980s had resulted in a narrower outcome in terms of votes, then they could well have resulted in hung parliaments. Meanwhile, although there was subsequently some variation in the number of marginal seats, the long-term decline that set in during the 1970s has never been reversed. The decline in the number of marginal seats has been attributed to the tendency from the 1950s onwards for northern and more urban Britain to become more strongly Labour, while southern and more rural parts of the country became more favourable to the Conservatives (Curtice and Steed, 1982, 1986). The result was a country whose political geography became much more polarised. However, the outcome of the 2019 election appears to have disrupted this pattern—the Conservatives captured from Labour for the first time ever a number of seats in Leave-voting areas in the North of England and the Midlands (see Power et al., and Goes, both this volume). Nevertheless, our table suggests this did little to change the overall geography of the two-party vote or the number of marginal seats, which, at 88, is much the same as it was in 2017. Labour’s loss of seats in Leave-voting constituencies that had once been bastions for the party was highly visible. However, in practice, the tendency for the Conservatives to perform more strongly in Leave-voting areas was far from confined to seats that were being defended by Labour. It was also clearly in evidence in Leave-voting areas that were being defended by the Conservatives. On average, there was a 7.5% swing from Labour to Conservative in Conservative-held seats where more than 60% voted Leave in 2016. Although this was somewhat lower than the equivalent figure among heavily Leave-inclined seats that Labour was defending (9.3%), it was still well above the GB-wide average of 4.7 points. Moreover, the proportion of Conservative-held seats (as of 2017) in which more than 60% voted Leave was, at 29%, just as big as (indeed, it was slightly higher than) the proportion of Labour seats that fell into that category (26%). In short, as well as turning some Labour seats blue, the tendency for Leave-voting Britain to swing more strongly to the Conservatives also resulted in many a Conservative-held seat becoming more strongly Conservative. As a result, the overall character of the distribution of the two-party vote was not radically changed. A look at what happened in the seats that on our definition were marginal in 2017 also helps explain why there was no marked change in the overall distribution of the two-party vote. On average, at 5.2%, the swing from Labour to Conservative in these seats was not very different from that across the country as a whole. Consequently, most of these remained marginal (indeed 62 of the 89 did so). At the same time, the swing from Conservative to Labour in those seats that can be defined as near-marginal for Labour in 20172 was, at 6.1%, sufficiently above the national swing to turn some of these seats into marginals, but only at a level that resulted in the replacement of those seats that ceased being so.3 So, the 2019 election did not result in any marked change in the overall distribution of the two-party vote. There continued to be many fewer seats that were marginal between the Conservatives and Labour than was required for the cube law to operate, and this helped ensure that the Conservatives’ overall majority was, indeed, rather less than we might have anticipated. But securing a large overall majority is also likely to be more difficult if more seats are won by parties other than the Conservatives and Labour, so it is to the continued high prevalence of third-party successes to which we turn next. 4. Third-party MPs Just as whether the electoral system conforms to the expectations of the cube law depends on the geography of party support, so also does the extent to which the system discriminates against smaller parties (Gudgin and Taylor, 1979). A small party whose vote is much the same from one constituency to another will, inevitably, tend to be a loser more or less everywhere. That is the fate that befell UKIP in 2015 when, even though the party won as much as 13% of the vote, it still only secured one seat. On the other hand, a small party whose vote is geographically concentrated may be relatively strong in—and thus capable of winning—some seats. One particularly striking example is Plaid Cymru who in both 2017 and 2019 won just 0.5% of the UK-wide vote but 0.6% (four) of the seats. The 1997 election witnessed a sharp increase in the number of third-party MPs (from 44 to 75), even though, at 26.1%, the share of the vote won by smaller parties was little different from that recorded at the three previous elections. It is an increase that has never been reversed. Most of it was accounted for by a change in the geography of the Liberal Democrat vote, which traditionally had been relatively evenly spread but now became somewhat more geographically concentrated, such that between 1997 and 2010 the party won between 46 and 62 seats (Russell and Fieldhouse, 2005; Curtice, 2009). However, a collapse in the party’s vote in 2015, after being in coalition with the Conservatives for five years, saw its representation fall to just eight seats. And although there was some recovery in its electoral support in 2019 (see Cutts and Russell, this volume), the party still won only 11 seats, one less than in 2017. However, the 2015 election also witnessed a dramatic increase in the number of SNP MPs. Scotland’s nationalist party only contests seats north of the border—but there it now stopped being a small party (Curtice, 2017b). Indeed, with as much as 50% of the Scottish vote, the single-member plurality electoral system worked in the party’s favour, crediting it with all but three of Scotland’s 59 constituencies. Although a dip in support in 2017 to 37% saw its number of MPs fall to 35, in 2019, the party’s vote rose once more and its representation increased to 48 (see Denver, this volume)—an increase of 13 seats that (bearing in mind the Liberal Democrats’ loss of one seat) wholly accounts for the overall increase of 12 third-party MPs compared with 2017. The 2019 election proved to be the third in a row in which the outcome in Scotland departed radically from that in the rest of Britain, with its representation dominated by a party other than the Conservatives or Labour. In that respect, Scotland is now much like Northern Ireland, where parliamentary representation has been dominated by small parties ever since the early 1970s. Britain’s electoral system cannot be expected to deliver normal service for as long as that proves to be the case. 5. Electoral bias So far, in assessing how the electoral system treats the Conservatives and Labour, we have examined the extent to which it exaggerates the lead of the largest party over the second party irrespective of which of those parties is in the lead. However, we cannot assume that the system treats the two parties in the same way. In particular, we might wonder what would have happened if it had been Labour rather than the Conservatives that had been 12 points ahead of their principal rivals in 2019? This question is addressed in Table 2.3. It shows what the outcome would be in seats for various possible outcomes in votes assuming that the geographical distribution of party support remained as it was in 2019. This is done by assuming that any change in the share of the vote won by Conservatives and Labour across the country as a whole is replicated in each and every constituency, while the level of support for all other parties (and the level of turnout) is unchanged. Thus, for example, in the second row of the table, we show what the outcome would be if the Conservative share of the vote fell by one point and the Labour share increased by one point in each and every seat, an outcome that would reduce the overall Conservative lead over Labour by two points—and reduce the Conservative tally of seats by eight. Table 2.3 Relationship between votes and seats following the 2019 general election Swing to Con from 2019 result . % Vote (GB) . Seats (UK) . Con . Lab . Con lead . Con . Lab . Others . Majority . 0.0 44.7 33.0 11.7 365 203 82 Con 80 −1.0 43.7 34.0 9.7 357 211 82 Con 64 −2.0 42.7 35.0 7.7 340 222 88 Con 30 −3.0 41.7 36.0 5.7 331 232 88 Con 12 −3.5 41.2 36.5 4.7 325 238 88 None −4.0 40.7 37.0 3.7 316 246 88 None −5.0 39.7 38.0 1.7 304 254 92 None −5.85 38.85 38.85 0.0 290 267 93 None −6.0 38.7 39.0 −0.3 288 269 93 None −7.0 37.7 40.0 −2.3 278 280 92 None −8.0 36.7 41.0 −4.3 269 288 93 None −9.0 35.7 42.0 −6.3 256 300 94 None −10.0 34.7 43.0 −8.3 244 311 95 None −11.0 33.7 44.0 −10.3 237 318 95 None −12.0 32.7 45.0 −12.3 229 326 95 Lab 2 −13.0 31.7 46.0 −14.3 224 331 95 Lab 12 −14.0 30.7 47.0 −16.3 216 342 92 Lab 34 −15.0 29.7 48.0 −18.3 201 358 91 Lab 66 −15.5 29.2 48.5 −19.3 194 366 90 Lab 82 Swing to Con from 2019 result . % Vote (GB) . Seats (UK) . Con . Lab . Con lead . Con . Lab . Others . Majority . 0.0 44.7 33.0 11.7 365 203 82 Con 80 −1.0 43.7 34.0 9.7 357 211 82 Con 64 −2.0 42.7 35.0 7.7 340 222 88 Con 30 −3.0 41.7 36.0 5.7 331 232 88 Con 12 −3.5 41.2 36.5 4.7 325 238 88 None −4.0 40.7 37.0 3.7 316 246 88 None −5.0 39.7 38.0 1.7 304 254 92 None −5.85 38.85 38.85 0.0 290 267 93 None −6.0 38.7 39.0 −0.3 288 269 93 None −7.0 37.7 40.0 −2.3 278 280 92 None −8.0 36.7 41.0 −4.3 269 288 93 None −9.0 35.7 42.0 −6.3 256 300 94 None −10.0 34.7 43.0 −8.3 244 311 95 None −11.0 33.7 44.0 −10.3 237 318 95 None −12.0 32.7 45.0 −12.3 229 326 95 Lab 2 −13.0 31.7 46.0 −14.3 224 331 95 Lab 12 −14.0 30.7 47.0 −16.3 216 342 92 Lab 34 −15.0 29.7 48.0 −18.3 201 358 91 Lab 66 −15.5 29.2 48.5 −19.3 194 366 90 Lab 82 Source: Author’s calculations. Open in new tab Table 2.3 Relationship between votes and seats following the 2019 general election Swing to Con from 2019 result . % Vote (GB) . Seats (UK) . Con . Lab . Con lead . Con . Lab . Others . Majority . 0.0 44.7 33.0 11.7 365 203 82 Con 80 −1.0 43.7 34.0 9.7 357 211 82 Con 64 −2.0 42.7 35.0 7.7 340 222 88 Con 30 −3.0 41.7 36.0 5.7 331 232 88 Con 12 −3.5 41.2 36.5 4.7 325 238 88 None −4.0 40.7 37.0 3.7 316 246 88 None −5.0 39.7 38.0 1.7 304 254 92 None −5.85 38.85 38.85 0.0 290 267 93 None −6.0 38.7 39.0 −0.3 288 269 93 None −7.0 37.7 40.0 −2.3 278 280 92 None −8.0 36.7 41.0 −4.3 269 288 93 None −9.0 35.7 42.0 −6.3 256 300 94 None −10.0 34.7 43.0 −8.3 244 311 95 None −11.0 33.7 44.0 −10.3 237 318 95 None −12.0 32.7 45.0 −12.3 229 326 95 Lab 2 −13.0 31.7 46.0 −14.3 224 331 95 Lab 12 −14.0 30.7 47.0 −16.3 216 342 92 Lab 34 −15.0 29.7 48.0 −18.3 201 358 91 Lab 66 −15.5 29.2 48.5 −19.3 194 366 90 Lab 82 Swing to Con from 2019 result . % Vote (GB) . Seats (UK) . Con . Lab . Con lead . Con . Lab . Others . Majority . 0.0 44.7 33.0 11.7 365 203 82 Con 80 −1.0 43.7 34.0 9.7 357 211 82 Con 64 −2.0 42.7 35.0 7.7 340 222 88 Con 30 −3.0 41.7 36.0 5.7 331 232 88 Con 12 −3.5 41.2 36.5 4.7 325 238 88 None −4.0 40.7 37.0 3.7 316 246 88 None −5.0 39.7 38.0 1.7 304 254 92 None −5.85 38.85 38.85 0.0 290 267 93 None −6.0 38.7 39.0 −0.3 288 269 93 None −7.0 37.7 40.0 −2.3 278 280 92 None −8.0 36.7 41.0 −4.3 269 288 93 None −9.0 35.7 42.0 −6.3 256 300 94 None −10.0 34.7 43.0 −8.3 244 311 95 None −11.0 33.7 44.0 −10.3 237 318 95 None −12.0 32.7 45.0 −12.3 229 326 95 Lab 2 −13.0 31.7 46.0 −14.3 224 331 95 Lab 12 −14.0 30.7 47.0 −16.3 216 342 92 Lab 34 −15.0 29.7 48.0 −18.3 201 358 91 Lab 66 −15.5 29.2 48.5 −19.3 194 366 90 Lab 82 Source: Author’s calculations. Open in new tab One point immediately stands out. Under these assumptions, Labour would need to be as much as 19 points ahead of the Conservatives in order to win an overall majority of 80. Other yardsticks also indicate that, for any given performance, the electoral system was inclined to reward the Conservatives more richly than Labour. If the two parties were to win the same share of the vote, the Conservatives, with 290 seats, would still be 23 seats ahead of Labour on 267 seats. If Labour had obtained the lead of 11.7 points that the Conservatives enjoyed, the party would have failed, albeit narrowly, to secure an overall majority at all. To win a minimal overall majority Labour would need a lead of just over 12 points, whereas the Conservatives only require to be 5 points ahead to win the 326 seats required for the barest possible overall majority. These figures further underline our argument that the outcome of the 2019 election did not represent a return to ‘normality’. Under our assumptions, even the relatively modest 80-seat majority that the Conservatives obtained would not have been achieved if Labour had enjoyed a near 12-point lead. Moreover, it should be noted that if the current electoral geography remains in place—and support for third parties remains as it was in 2019—a very wide range of results—anything from a Conservative lead of 5 points to a Labour one of 12—would result in a hung parliament. In short, the single-member plurality electoral system still cannot be relied upon to facilitate alternating single-party government by being more or less guaranteed to deliver an overall majority to one or other of the two largest parties. But why is the system inclined to reward the Conservatives more richly than Labour? Electoral bias of this kind can arise for two principal reasons (Gudgin and Taylor, 1979; Johnston, 1979). The first is that the seats won by one of the parties are on average smaller in size than those won by the other—either because fewer people are registered to vote or because the level of turnout is lower. The second is that the vote cast for one of the parties is more efficiently distributed than those won by the other. A party’s vote will be more efficiently distributed if more of its votes contribute to it winning seats narrowly—and less so if much of its support is either piled up in large majorities in safe seats or losing narrowly in lots of marginal seats. An indication of the extent to which the Conservatives and Labour are affected by these two patterns—and how this has varied over time—is given in Table 2.4 (Soper and Rydon, 1958). In the first column, we show the difference between the average (mean) share of the two-party vote won by the Conservatives across all constituencies and that party’s overall share of the votes cast in the country as a whole. If a party tends to perform more strongly in seats where fewer votes are cast, its average share of the vote across constituencies will be higher than its overall share of the vote—and thus a positive sign in the first column of the table is an indication of an advantage to the Conservatives while a negative sign points to one for Labour. In the second column, meanwhile, we show the difference between the share of the two-party vote won by the Conservatives in the median constituency and the party’s mean share across all constituencies. The share in the median constituency represents the tally that divides a party’s constituency performances into two equal groups, half being ones where the party does better than the median and half those where it does less well. If a party is to win more seats than its principal rivals, it will need to have won more than 50% of the two-party vote in the median constituency. Thus, if a party’s median share is more than its mean share, that is an indication that its vote is likely to be more efficiently distributed. Table 2.4 Measures of two-party bias, 1955–2019 Party . Conservative percentage of two-party vote . Mean −overall . Median −Mean . Median −overall . 1955 + 0.3 + 0.6 + 0.9 1959 + 0.4 + 0.8 + 1.2 1964 + 0.1 + 0.4 + 0.5 1966 −0.3 + 0.2 −0.1 1970 −0.9 + 0.8 −0.1 1970 (NT) −0.1 + 0.5 + 0.4 1974 (Feb) −0.1 − 0.5 −0.5 1974 (Oct) −0.3 + 1.4 + 1.1 1979 −0.7 − 0.5 −1.2 1979 (NT) −0.1 + 0.9 + 0.9 1983 −0.5 + 1.7 + 1.2 1987 −0.8 + 1.4 + 0.6 1992 −1.2 −0.0 −1.2 1992 (NT) −0.2 −0.7 −0.9 1997 −0.4 −1.6 −2.0 2001 −1.4 −1.5 −2.9 2001 (NT) −1.1 −1.4 −2.5 2005 −2.1 −1.1 −3.2 2005 (NT) −1.5 −1.0 −2.5 2010 −1.3 −0.8 −2.1 2015 −1.6 +2.1 +0.5 2017 −0.4 +0.6 +0.2 2019 −0.5 +2.0 +1.5 Party . Conservative percentage of two-party vote . Mean −overall . Median −Mean . Median −overall . 1955 + 0.3 + 0.6 + 0.9 1959 + 0.4 + 0.8 + 1.2 1964 + 0.1 + 0.4 + 0.5 1966 −0.3 + 0.2 −0.1 1970 −0.9 + 0.8 −0.1 1970 (NT) −0.1 + 0.5 + 0.4 1974 (Feb) −0.1 − 0.5 −0.5 1974 (Oct) −0.3 + 1.4 + 1.1 1979 −0.7 − 0.5 −1.2 1979 (NT) −0.1 + 0.9 + 0.9 1983 −0.5 + 1.7 + 1.2 1987 −0.8 + 1.4 + 0.6 1992 −1.2 −0.0 −1.2 1992 (NT) −0.2 −0.7 −0.9 1997 −0.4 −1.6 −2.0 2001 −1.4 −1.5 −2.9 2001 (NT) −1.1 −1.4 −2.5 2005 −2.1 −1.1 −3.2 2005 (NT) −1.5 −1.0 −2.5 2010 −1.3 −0.8 −2.1 2015 −1.6 +2.1 +0.5 2017 −0.4 +0.6 +0.2 2019 −0.5 +2.0 +1.5 NT, Notional results based on estimates of what the outcome would have been if that election had been fought on the new constituency boundaries that were introduced at the subsequent election. The 2001 redistribution (together with a reduction in the number of seats) was confined to Scotland, while the 2005 one only occurred in England and Wales. Two-party vote: Votes cast for Conservative and Labour combined. Figures based on all seats in Great Britain. Northern Ireland excluded. Source: Author’s calculations. Open in new tab Table 2.4 Measures of two-party bias, 1955–2019 Party . Conservative percentage of two-party vote . Mean −overall . Median −Mean . Median −overall . 1955 + 0.3 + 0.6 + 0.9 1959 + 0.4 + 0.8 + 1.2 1964 + 0.1 + 0.4 + 0.5 1966 −0.3 + 0.2 −0.1 1970 −0.9 + 0.8 −0.1 1970 (NT) −0.1 + 0.5 + 0.4 1974 (Feb) −0.1 − 0.5 −0.5 1974 (Oct) −0.3 + 1.4 + 1.1 1979 −0.7 − 0.5 −1.2 1979 (NT) −0.1 + 0.9 + 0.9 1983 −0.5 + 1.7 + 1.2 1987 −0.8 + 1.4 + 0.6 1992 −1.2 −0.0 −1.2 1992 (NT) −0.2 −0.7 −0.9 1997 −0.4 −1.6 −2.0 2001 −1.4 −1.5 −2.9 2001 (NT) −1.1 −1.4 −2.5 2005 −2.1 −1.1 −3.2 2005 (NT) −1.5 −1.0 −2.5 2010 −1.3 −0.8 −2.1 2015 −1.6 +2.1 +0.5 2017 −0.4 +0.6 +0.2 2019 −0.5 +2.0 +1.5 Party . Conservative percentage of two-party vote . Mean −overall . Median −Mean . Median −overall . 1955 + 0.3 + 0.6 + 0.9 1959 + 0.4 + 0.8 + 1.2 1964 + 0.1 + 0.4 + 0.5 1966 −0.3 + 0.2 −0.1 1970 −0.9 + 0.8 −0.1 1970 (NT) −0.1 + 0.5 + 0.4 1974 (Feb) −0.1 − 0.5 −0.5 1974 (Oct) −0.3 + 1.4 + 1.1 1979 −0.7 − 0.5 −1.2 1979 (NT) −0.1 + 0.9 + 0.9 1983 −0.5 + 1.7 + 1.2 1987 −0.8 + 1.4 + 0.6 1992 −1.2 −0.0 −1.2 1992 (NT) −0.2 −0.7 −0.9 1997 −0.4 −1.6 −2.0 2001 −1.4 −1.5 −2.9 2001 (NT) −1.1 −1.4 −2.5 2005 −2.1 −1.1 −3.2 2005 (NT) −1.5 −1.0 −2.5 2010 −1.3 −0.8 −2.1 2015 −1.6 +2.1 +0.5 2017 −0.4 +0.6 +0.2 2019 −0.5 +2.0 +1.5 NT, Notional results based on estimates of what the outcome would have been if that election had been fought on the new constituency boundaries that were introduced at the subsequent election. The 2001 redistribution (together with a reduction in the number of seats) was confined to Scotland, while the 2005 one only occurred in England and Wales. Two-party vote: Votes cast for Conservative and Labour combined. Figures based on all seats in Great Britain. Northern Ireland excluded. Source: Author’s calculations. Open in new tab One of the striking, but little remarked features of the 2019 election is that the contest was fought on the same constituency boundaries as every other election since and including 2010. This is despite the fact that the four Boundary Commissions (one for each part of the UK) presented recommendations for new, more equal constituencies to parliament in the autumn of 2018. These, though, were not put to MPs for their approval (Johnston et al., 2019)—most likely because of concern that, against the backdrop of severe internal party divisions over Brexit, some Conservative MPs whose political careers would be put at risk by proposals that also involved reducing the number of MPs from 650 to 600 would join the opposition in voting them down. Table 2.4 indicates why the Conservatives would be thought to have an interest in constituencies being made more equal in size. At every election since 1966, the mean Conservative share of the two-party vote has been less than the party’s overall share. Moreover, until 2017 at least, the gap has gradually widened over time, only to be reduced again after a boundary review. This is in part because the rules for drawing up constituency boundaries have hitherto resulted in smaller constituencies in Scotland and (more recently at least, especially in) Wales, where the Conservatives have performed less well, while for much of the post-war period, the pattern of population movement has seen people move away from (Labour-voting) inner cities towards (more Conservative-inclined) suburbs and rural areas (Champion, 2005; Rossiter et al., 2009). Meanwhile, turnout also tends to be lower in Labour-held seats. However, the 2017 election was the first at which the difference between the Conservatives’ mean share of the two-party vote and their overall share fell even though there had not been a boundary review. This was because (i) turnout increased more in Labour-held seats than in Conservative ones and (ii) there was a tendency for the Conservatives to perform more strongly in areas that voted Leave, a pattern that resulted in more of the party’s vote being garnered in smaller constituencies (Curtice, 2017a). The first of these patterns was not repeated in 2019. Rather, on average, turnout fell much more (by 2.6 points) in seats that Labour was defending than it did in those where the Conservatives were the incumbents (0.8 points). This pattern more or less wholly reversed the narrowing of the gap in turnout that was in evidence in 2015.4 On the other hand, the tendency in 2017 for the Conservatives to advance more strongly in areas that voted Leave—many of them places with smaller electorates—was repeated. The average electorate in the 54 seats the Conservatives gained from Labour was just 69,223. Not only is this lower than the GB-wide average of 73,245, but also than the figure of 72,242 in those seats that Labour defended successfully. In short, the Conservatives’ headline-grabbing gains occurred disproportionately in relatively small seats. Meanwhile, the swing to the Conservatives was also higher in seats with a lower turnout; in seats where less than 62% turned out to vote the average swing was 7.4 points, whereas in those where more than 72% turned out it was just 2.3 points. At the same time, the hitherto persistent tendency for the difference between the size of the registered electorate in Conservative and Labour-held seats to widen came to a halt. It has been evident for some time that the pattern of population movement out of the inner city into the suburbs and rural areas was slowing down (Champion, 2016). Meanwhile, the Electoral Commission estimated that in 2018 only 85% of those who should be on the electoral register (for parliamentary elections) in 2018 were actually registered, so there was plenty of room for improvement in the quality of the register—and especially so in more Labour-inclined urban areas (Electoral Commission, 2019). In any event, the level of registration increased just as much between 2017 and 2019 in seats that Labour was defending (by 1,134) as it did in those where the Conservatives were first (1,101). Although there was still a difference of 3,889 between the average electorate in seats Labour were defending (71,644) and that in those where the Conservatives were the incumbents (75,533), the fact the gap did not widen further stands in sharp contrast to previous experience.5 Fighting the 2019 election on constituency boundaries that were nearly 20 years out of date was not then as disadvantageous to the Conservatives as might have been anticipated. The difference in the average size of Conservative and Labour-held seats did not widen. Meanwhile, as in 2017, the Conservative vote tilted further towards seats with smaller electorates and lower turnouts. Only a larger drop in turnout in Labour-held seats worked in the opposite direction—and even here the resulting gap in turnout simply resulted in a return to the status quo ante. Still, while the Conservatives may not have been significantly further disadvantaged by differences in constituency sizes, they were hardly advantaged by them. To understand why the electoral system was inclined to treat the Conservatives more favourably than Labour, we need to look at the relative efficiency of the geographical distribution of the two parties’ votes. Indeed, Table 2.4 suggests that the Conservatives were advantaged on this score by as much as they have been at any election since 1955. Why this is the case is illustrated in Table 2.5, which provides fuller information on the distribution of the two-party vote in 2017 and 2019. As in Table 2.2, we assume that the two largest parties have an identical share of the vote nationally but that the geography of party support is as it actually was at the election in question. Thus, in 2019, for example, we assume that the Conservative share of the two-party vote is 7.6 points less than it actually was (and, in 2017, 1.5 points less), and that this reduction takes place everywhere. The table states the number of seats that on that basis each party would win with various shares of the two-party vote locally. Table 2.5 Distribution of the two-party vote, 2017 and 2019 Winning party share of two-party vote . Winning party: 2017 . Winning party: 2019 . Conservative . Labour . Conservative . Labour . 50–55 42 47 39 49 55–60 45 42 35 31 60–65 64 40 50 43 65–70 66 51 72 37 70–80 72 64 79 49 80–100 10 36 18 56 Winning party share of two-party vote . Winning party: 2017 . Winning party: 2019 . Conservative . Labour . Conservative . Labour . 50–55 42 47 39 49 55–60 45 42 35 31 60–65 64 40 50 43 65–70 66 51 72 37 70–80 72 64 79 49 80–100 10 36 18 56 Winning party share of two-party vote: Share of the two-party vote for the winning party assuming a 50:50 division of the overall two-party vote nationally, but an unchanged electoral geography. Two-party vote: Votes cast for Conservative and Labour combined. Table based only on seats won by Conservative or Labour at that election and contested by both parties. Source: Author’s calculations. Open in new tab Table 2.5 Distribution of the two-party vote, 2017 and 2019 Winning party share of two-party vote . Winning party: 2017 . Winning party: 2019 . Conservative . Labour . Conservative . Labour . 50–55 42 47 39 49 55–60 45 42 35 31 60–65 64 40 50 43 65–70 66 51 72 37 70–80 72 64 79 49 80–100 10 36 18 56 Winning party share of two-party vote . Winning party: 2017 . Winning party: 2019 . Conservative . Labour . Conservative . Labour . 50–55 42 47 39 49 55–60 45 42 35 31 60–65 64 40 50 43 65–70 66 51 72 37 70–80 72 64 79 49 80–100 10 36 18 56 Winning party share of two-party vote: Share of the two-party vote for the winning party assuming a 50:50 division of the overall two-party vote nationally, but an unchanged electoral geography. Two-party vote: Votes cast for Conservative and Labour combined. Table based only on seats won by Conservative or Labour at that election and contested by both parties. Source: Author’s calculations. Open in new tab The figures in the first row of the table, which encompass the seats we have already defined as marginal, are perhaps rather surprising. It is Labour, not the Conservatives, who tend to win more seats with small majorities; the party has 49 such seats as compared with the Conservatives’ 39. Moreover, the gap between the two parties in this respect widened between 2017 and 2019. However, if we turn our attention to the other end of the table, where we show the number of seats that each party wins with more than 80% of the two-party vote, we can see both that there are many more such seats in the Labour column (56) than the Conservative one (18), and that this gap is much wider than it was in 2017. In short, in 2019, Labour was ‘wasting’ many more votes than the Conservatives in piling up large majorities. This arose because, in seats where Labour had (on the basis of our calculation) more than 80% of the two-party vote in 2017, the Conservative share of the two-party vote increased on average by just 0.9 of a point. Meanwhile, in those seats where Labour’s share of the two-party vote had been between 70% and 80%, the average increase was, at 4.1 points, also relatively low. The headlines in the 2019 election focussed on Labour’s loss of some of its traditional strongholds to the Conservatives, but in practice, Labour was also disadvantaged because it was performing best of all in seats that were already relatively safe for the party.6 The resulting inefficiency in the distribution of its vote put it at a substantial disadvantage in its attempts to secure parliamentary representation—and unless reversed will leave the party at a disadvantage at future elections. In the meantime, it means that even though the Conservatives’ 80-seat majority can be regarded by some yardsticks as relatively low, it actually exaggerates the overall propensity of the system to deliver safe overall majorities irrespective of who is in the lead. The outcome was also a reflection of a system that was treating the Conservatives markedly more favourably than Labour. 6. An alternative system We have demonstrated that, despite appearances, the outcome of the 2019 election did not signal that the operation of the single-member plurality electoral system has returned to a ‘normality’ in which most of the time either the Conservatives or Labour will win an overall majority. Yet that does not mean that the way it operated in 2019 was anything other than decisive so far as the issue that precipitated the election was concerned—Britain’s future relationship with the EU. The result ensured that the UK left the EU at the end of January 2020. This is not necessarily what would have happened if a different, more proportional system had been in place. In Table 2.6 we show what the outcome would have been if everyone had voted the same way as they did under the single-member plurality system, but seats had been allocated using a regional party list system similar to that used in European elections in the UK. Even though we assume that a party would have to win 5% of the vote in a region before it was allocated any seats (in line with the rules for the allocation of list seats on the Greater London Assembly) and the use of a rule for allocating seats that is more generous to larger parties, the Conservatives would not have won an overall majority—and, crucially, neither would they have done so in combination with the Brexit Party. The two parties’ combined tally of 308 seats would have been insufficient to secure parliamentary approval for the revised Brexit deal that had been negotiated by the Prime Minister in October. Meanwhile, the parties in Great Britain that were willing to back a second EU referendum would have had 324 seats between them, which, together with the half-dozen seats that would have been won by parties in Northern Ireland that were opposed to Brexit (and were willing, unlike Sinn Féin, to take their seats), would have constituted a potential parliamentary majority for a very different course of action. Table 2.6 Projected outcome of the 2019 general election under regional proportional representation . Projected seats . % share of seats . % seats − % votes . Conservatives 302 46.5 +2.9 Labour 221 34.0 +1.8 Liberal Democrats 74 11.4 −0.2 SNP 26 4.0 +0.1 Brexit 6 0.9 −1.1 PC 3 0.5 0.0 Green 0 0.0 −2.7 Others (NI) 18 2.6 — . Projected seats . % share of seats . % seats − % votes . Conservatives 302 46.5 +2.9 Labour 221 34.0 +1.8 Liberal Democrats 74 11.4 −0.2 SNP 26 4.0 +0.1 Brexit 6 0.9 −1.1 PC 3 0.5 0.0 Green 0 0.0 −2.7 Others (NI) 18 2.6 — Seats allocated by government region. Total number of seats in each region proportional to current electorate using St Laguë divisor. Division of seats within each region determined by D’Hondt divisor but confined in each region to those parties that won at least 5% of the vote. Source: Author’s calculations. Open in new tab Table 2.6 Projected outcome of the 2019 general election under regional proportional representation . Projected seats . % share of seats . % seats − % votes . Conservatives 302 46.5 +2.9 Labour 221 34.0 +1.8 Liberal Democrats 74 11.4 −0.2 SNP 26 4.0 +0.1 Brexit 6 0.9 −1.1 PC 3 0.5 0.0 Green 0 0.0 −2.7 Others (NI) 18 2.6 — . Projected seats . % share of seats . % seats − % votes . Conservatives 302 46.5 +2.9 Labour 221 34.0 +1.8 Liberal Democrats 74 11.4 −0.2 SNP 26 4.0 +0.1 Brexit 6 0.9 −1.1 PC 3 0.5 0.0 Green 0 0.0 −2.7 Others (NI) 18 2.6 — Seats allocated by government region. Total number of seats in each region proportional to current electorate using St Laguë divisor. Division of seats within each region determined by D’Hondt divisor but confined in each region to those parties that won at least 5% of the vote. Source: Author’s calculations. Open in new tab Of course, in practice faced with a different system, some voters would have voted differently, and in particular, all of them would have had the opportunity to vote for one of the smaller parties such as the Brexit Party and the Greens—and not just those living in constituencies that those parties contested. This latter point, in particular, might well have resulted in a higher level of Brexit Party representation—the party’s average share of the vote in the seats it contested was above 5% in five regions, whereas it won over 5% of the overall vote (and thus has been allocated seats under our scenario) in just two. Even so, bearing in mind that some of the seats the Brexit Party might otherwise have won would have been secured at the expense of the Conservatives, it is still unlikely that there would have been a pro-Brexit majority in a House of Commons elected by a regional party-list system.7 We should, of course, be careful of drawing the conclusion that the UK would not have left the EU if it was using a more proportional electoral system to elect its MPs. After all, if a regional party-list system of the kind examined here had been in place in 2015, the Conservatives and the United Kingdom Independence Party (UKIP) would have had a majority between them (Curtice, 2015) and perhaps a coalition of the two parties would have delivered Brexit by 2019. Equally, there is no guarantee that any second referendum would have resulted in a majority vote for Remain. But it is certainly difficult to avoid the conclusion that, in 2019 at least, the use of the single-member plurality system played a crucial role in enabling the Prime Minister to implement Brexit.8 7. Conclusion After three elections in a row in which it failed to deliver a decisive overall majority for one party, the single-member plurality electoral system delivered a substantial majority that ended the impasse over Brexit that had dominated the business of the hung parliament that was elected in 2017. It appeared that ‘normal’ service had resumed so far as the electoral system is concerned. Yet in practice, the size of the Conservative majority reflected the large lead over Labour that the party enjoyed in votes and the fact that the system was treating it more favourably than its principal opponents. Even then, it can be argued that the majority was smaller than might have been anticipated. Meanwhile, if the positions of the two largest parties had been reversed Labour could well have failed to secure an overall majority, and there remains a high probability that future elections will result in more hung parliaments. There are two reasons why this is the case. First, despite the apparent dramatic changes in the geography of party support in 2019, there are still too few seats that are marginal between Labour and the Conservatives for the system to exaggerate the lead of the largest party over the second party to the extent it once did. Secondly, the system is still proving to be relatively poor at excluding third parties from the House of Commons, not least because of the collapse of the traditional two-party system in Scotland. These features may be hidden from view at present—but, unless they change, hung parliaments could well still prove to be a regular feature of British politics, a prospect that almost undoubtedly ensures that the debate about the merits of electoral reform is likely to continue. Footnotes 1 The result is the same if we undertake the calculation on the basis of only those votes cast in seats won by either the Conservatives or Labour, where, at 57.54% the Conservative share of the two-party vote was almost exactly the same as the 57.56% figure for the country as a whole. 2 These are seats where Labour would have won between 55% and 60% of the two-party vote in 2017 if, as a result of a uniform swing, the Conservatives and Labour had won the same share of the overall national vote. 3 Strictly speaking, the swing that matters here is that based on the change in the share of the two-party vote rather than that based (as quoted here) on the change in the overall share of the vote. However, the point still stands. The mean swing in the Conservative share of the two-party vote was 5.8%, little different from the overall national figure of 6.1%. The mean swing in Labour’s ‘near-marginals’ was 6.6%. 4 In 2019, turnout in seats that the Conservatives won in 2017 was 5.2 points higher than in seats Labour were defending; the equivalent gap in 2015 was 5.4 points. 5 Indeed, if we classify seats according to who won in 2019, we find that on this basis, the gap fell somewhat from 2,651 in 2017 to 2,470 in 2019. 6 We should note that another source of electoral bias that is not captured by the statistics in Table 4 is now working to the Conservatives’ disadvantage once more, after having disappeared in 2015 and 2017. The Conservatives waste more votes losing in seats that are won by third parties. On average the party won 23.9% of the vote in seats won by parties other than the Conservatives or Labour, while Labour won only 18.8%. In 2017, the two figures were virtually identical, while in 2015 it was actually Labour that ‘wasted’ more votes in this way. The explanation lies in the outcome in the 11 seats won by the Liberal Democrats, in which the Conservatives on average won 29.0% of the vote, whereas Labour secured only 7.5%. 7 This would also be the case if the Brexit Party did not stand at all, and all those who voted for the party backed the Conservatives instead. In those circumstances the Conservatives would have 313 seats, while 325 would have been won by British parties in favour of a second referendum and Northern Irish parties who oppose Brexit. With Sinn Féin winning but not taking up four seats, the latter figure would have constituted a majority irrespective of what the eight unionist MPs (who supported Brexit but not the Prime Minister’s deal) opted to do. 8 True, we should be careful about assuming that any alternative more proportional system would have resulted in a majority for another referendum. One instance where the position is less clear cut is the Single Transferable Vote system favoured by the Electoral Reform Society. According to a projection undertaken by the society itself, 315 Conservative and Brexit Party MPs would have been elected, just slightly less than the 317 representatives of British parties that were willing to back a second referendum (Garland et al., 2020). In this instance, the fate of Brexit would lie in the hands of the various parties in Northern Ireland. References Baker C. , Cracknell R. , Uberoi E. ( 2019 ) General Election 2019: Results and Analysis, London, House of Commons Library Briefing Paper CBP 8749, accessed at https://commonslibrary.parliament.uk/research-briefings/cbp-8749/ on 26 March 2020. Bingham Powell G. Jr. ( 2000 ) Elections as Instruments of Democracy: Majoritarian and Proportional Visions , New Haven, CT , Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Blais A. , Carty R. ( 1991 ) ‘ The Psychological Impact of Electoral Laws: Measuring Duverger’s Elusive Factor ’, British Journal of Political Science , 21 , 79 – 93 . Google Scholar Crossref Search ADS WorldCat Butler D. ( 1951 ) ‘An Examination of the Results’. In Nicholas H. (ed), The British General Election of 1950 , Basingstoke, England , Macmillan , pp. 306 – 333 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Champion A. ( 2005 ) ‘Population Movement within the UK’. In Chappell R. (ed) Focus on People and Migration , London , Palgrave Macmillan , pp. 91 – 114 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Champion A. ( 2016 ) ‘Internal Migration and the Spatial Distribution of the Population’. In Champion A. , Falkingham J. (eds.), Population Change in the United Kingdom , London , Rowman and Littlefield , pp. 125 – 141 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cowley P. , Kavanagh D. ( 2016 ) The British General Election of 2015 , London , Palgrave Macmillan . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Cowley P. , Kavanagh D. ( 2018 ) The British General Election of 2017 , London , Palgrave Macmillan . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Curtice J. , Steed M. ( 1982 ) ‘ Electoral Choice and the Production of Government: The Changing Operation of the Electoral System in the UK Since 1955 ’, British Journal of Political Science , 12 , 249 – 298 . Google Scholar Crossref Search ADS WorldCat Curtice J. , Steed M. ( 1986 ) ‘ Proportionality and Exaggeration in the British Electoral System ’, Electoral Studies , 5 , 209 – 228 . Google Scholar Crossref Search ADS WorldCat Curtice J. ( 2009 ) ‘Neither Representative nor Accountable: First past the Post in Britain’. In Grofman B. , Blais A. , Bowler S. (eds.), Duverger’s Law of Plurality Voting , New York, NY , Springer , pp. 27 – 46 . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Curtice J. ( 2015 ) ‘ A Return to Normality? How the Electoral System Operated ’, Parliamentary Affairs , 68 , 25 – 40 . Google Scholar Crossref Search ADS WorldCat Curtice J. ( 2017 a) ‘ How the Electoral System Failed to Deliver – Again ’, Parliamentary Affairs , 71 , 29 – 45 . Google Scholar Crossref Search ADS WorldCat Curtice J. ( 2017 b) ‘The Party and the Electorate’. In Hassan G. , Barrow S. (eds.), A Nation Changed? the SNP and Scotland Ten Years On , Edinburgh , Luath Press , pp. 32 – 39 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Duverger M. ( 1954 ) Political Parties: Their Organisation and Activity in the Modern State , London , Methuen . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Electoral Commission ( 2019 ) 2019 Report: Accuracy and Completeness of the 2018 Electoral Registers in Great Britain, accessed at https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/our-views-and-research/our-research/accuracy-and-completeness-electoral-registers/2019-report-accuracy-and-completeness-2018-electoral-registers-great-britain on 26 March 2020. Garland J. , Palese M. , Simpson I. ( 2020 ) Voters Left Voiceless: The 2019 General Election, London, Electoral Reform Society. accessed at https://www.electoral-reform.org.uk/latest-news-and-research/publications/the-2019-general-election-voters-left-voiceless/ on 26 March 2020. Gudgin G. , Taylor P. ( 1979 ), Seats, Votes and the Spatial Organisation of Elections , London , Pion . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Johnston R. ( 1979 ) Political, Electoral and Spatial Systems , Oxford , Clarendon Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Johnston R. , Pattie C. , Rossiter D. ( 2019 ) Boundaries in Limbo: Why the Government Cannot Decide How Many MPs there should be, accessed at https://blogs.lse.ac.uk/politicsandpolicy/boundaries-in-limbo/ on 26 March 2020. Kendall M. , Stuart A. ( 1950 ) ‘ The Law of Cubic Proportions in Election Results ’, The British Journal of Sociology , 1 , 183 – 197 . Google Scholar Crossref Search ADS WorldCat Norton P. ( 1997 ) ‘ The Case for First past the Post ’, Representation , 34 , 84 – 88 . Google Scholar Crossref Search ADS WorldCat Rallings C. , Thrasher M. ( 2012 ) British Electoral Facts 1832–2012 , London , Biteback . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Renwick A. ( 2011 ) A Citizens’ Guide to Electoral Reform , London , Biteback . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rossiter D. , Johnston R. , Pattie C. ( 2009 ), The Boundary Commissions: Redrawing UK’s Map of Parliamentary Constituencies , Manchester, England , Manchester University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Russell A. , Fieldhouse E. ( 2005 ) Neither Left nor Right? The Liberal Democrats and the Electorate , Manchester, England , Manchester University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Soper C. , Rydon J. ( 1958 ) ‘ Under-Representation and Electoral Prediction ’, Australian Journal of Politics & History , 4 , 94 – 106 . Google Scholar Crossref Search ADS WorldCat Taagerpera R. , Shugart M. ( 1989 ) Seats and Votes: The Effects and Determinants of Electoral Systems , New Haven, CT , Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Thompson,, Louise

doi: 10.1093/pa/gsaa038pmid: N/A

It was clear from the outcome of the 2017 general election that the subsequent parliamentary term would be unusual. Theresa May had gone to the polls hoping for a clear mandate but instead found herself as a minority Prime Minister, relying on the support of the Democratic Unionist Party (DUP) to pass key pieces of legislation. May (2017) claimed that the election had brought certainty and clarity to Britain’s withdrawal from the EU but any semblance of this would quickly fade away. Instead, a set of unusual circumstances would take over and make the 2017–2019 parliament stand out as nothing less than extraordinary. Government defeats in the Commons of a scale never before recorded, the centrality of an activist Speaker and an unusual degree of cross-party cooperation from MPs on opposing sides of the chamber, facilitated a stalemate between parliament and government which stymied any progress towards Brexit. This contribution explores this state of brinkmanship between government and parliament from December 2018. It highlights the stubbornness of the government in its continued presentation of the same Brexit deal to parliament, assesses the significance of decisions made by the Speaker and examines the gradually emboldened cross-party group of MPs who would take advantage of procedural rules to put alternatives on the table. The combination of a government which failed to listen to parliament until it was too late and a parliament in which no single Brexit proposal could achieve a majority, left the UK’s withdrawal from the EU stuck on a merry-go-round of meaningful votes, indicative votes and general election votes from which the bypassing of the Fixed-term Parliaments Act (FTPA) and a general election was the only escape. 1. From a manageable to an unmanageable House of Commons With the Conservative Party falling eight seats short of a majority at the start of the 2017 parliament, it was unclear whether the government’s legislative programme was going to be manageable. A confidence and supply agreement negotiated with the DUP committed their Northern Irish MPs in supporting the government on key votes, including the necessary EU withdrawal legislation (Cabinet Office, 2017). Thus, in September 2017, MPs passed the second reading of the government’s European Union (Withdrawal) Bill in what the Prime Minister called a ‘historic decision’, by a fairly comfortable majority of 326 votes to 290 (May, 2017). There were no Conservative rebels and all the DUP’s MPs voted with the government, as agreed. Confident in its small parliamentary majority, the government took the bill into its committee stage in November. There was an expectation that a considerable amount of parliamentary time over the next 18 months would be spent scrutinising further withdrawal legislation and in making post-Brexit arrangements. After all, the government had promised that MPs would be ‘intimately’ involved at every stage (HC Debates, 13 November 2017, c.37). May faced her first defeat at the bill’s committee stage, when MPs narrowly passed an amendment by Dominic Grieve to give parliament a vote on the final Brexit deal (BBC, 2017). This ‘meaningful’ vote that Grieve and ten other Conservative rebels had forced on the government would go on to underpin the need for a 2019 general election. As just one of over 400 amendments tabled by MPs though, the government was not unduly fazed by the defeat, sacking one of the party’s Vice-Chairs, Stephen Hammond, for joining the rebellion (Sharman, 2017). Although they faced a tough time as the bill continued its scrutiny journey, particularly in the Lords, ministers successfully brought the EU Withdrawal Bill to its conclusion the following year and it received Royal Assent at the end of June 2018. By December 2018, however, the interaction between government and parliament became more hostile than it had been for several parliamentary sessions and the government’s parliamentary majority appeared much more unstable. May suffered three significant defeats in the Commons in December 2018, including on a motion holding the government in contempt of parliament for failing to release crucial Brexit documents (BBC, 2018). The BBC’s Laura Kuenssberg described it as an ‘excruciating’ day for May, with a record-breaking succession of defeats not seen since the 1970s (ibid). The Prime Minister also faced an additional internal party battle with her own backbenchers. On 12 December, Sir Graham Brady, the Chair of the Conservative Party’s 1922 Committee, announced that he had received 48 letters from Conservative MPs, triggering the 15% threshold for a vote of no-confidence in the party leader. Although a victory for May, one-third of Conservative MPs voted against her, an uncomfortably close result which would make delivering Brexit even more challenging. Over the next few months, this lack of confidence in the Prime Minister would underpin a succession of rebellions by Conservative MPs and see previously loyal backbenchers join opposition MPs in the ‘no’ lobby. 1.1 The balance between government and parliament begins to change To properly understand the changing relationship between government and parliament, we must take a closer look at the parliamentary business over the weeks surrounding the Conservative Party’s no-confidence vote in Theresa May. At the start of December, the government had put the negotiated withdrawal agreement before MPs for three days of debate. There was much unease about provisions for the Northern Ireland border, or ‘backstop’, across the House, including from the DUP. Recognising this concern, the Prime Minister delayed giving MPs the meaningful vote on the deal. In a rare admission from a Prime Minister, May admitted that if the vote had gone ahead as planned, the government would have lost ‘by a significant margin’ (HC Debates, 10 December 2018, c.23). The opposition parties were united in their frustration with this decision. The Scottish National Party (SNP) described it as a ‘stunning display of pathetic cowardice’ (Kirsty Blackman, HC Debates, 10 December 2018, c.28), while the DUP’s Westminster leader, Nigel Dodds, announced that the decision was ‘not credible’ (HC Debates, 10 December 2018, c.31). During three hours at the despatch box answering questions from MPs, May reaffirmed to the House that the choice before them would be the choice between ‘a deal, no deal and no Brexit’, with no majority among MPs for any alternative (HC Debates, 10 December 2018, c.64). She would not be pressed though on precisely when the vote would now happen. The views held by MPs on the type of Brexit which should be sought by the government and on the process by which it should be pursued and agreed by the House had long cut across party lines. But the degree of the divisions within and between parties on Brexit was becoming increasingly important. Father of the House Ken Clarke summarised the situation nicely when he observed: this House is divided not just into parties; it is divided into factions. It becomes clear that, at the moment, there is no predictable majority for any single course of action going forward (Ken Clarke, HC Debates, 10 December 2018, c.28). Some MPs began to believe that the Prime Minister had been ‘captured’ by the group of hard Brexiteer Conservative backbenchers who formed the party’s European Research Group (see HC Debates, 10 December 2018, c.40). This was also a period during which the Commons Speaker began to play a role in providing a platform for backbenchers to cause trouble for the government. One of the principal ways in which he would do this was through urgent questions (UQs), emergency questions that require a government minister to attend the House of Commons that day, making a statement and answering questions from backbenchers. This procedure was already a marked feature of Bercow’s speakership, but UQs relating to the Brexit process dominated UQs in the run up to the election (Priddy, 2020, p. 1), with 44 relating directly to Brexit.1 The UQ tabled by Labour’s Yvette Cooper, granted after the meaningful vote delay, committed the government to put the vote before the House of Commons before 21 January. In his answer to Cooper’s question, the Department of Exiting the EU minister, Robin Walker, further committed the government: if Parliament were to reject the deal, the Government would be required to make a statement on our proposed next steps and table a motion in neutral terms on that statement (HC Debates, 11 December 2018, c.151). This was a difficult period for executive–legislative relations, but the New Year brought even greater tension and a resurgent House of Commons. This was signalled most forcefully on 9 January when the government tabled its business motion for five further days of debate on the Prime Minister’s Brexit deal, to be followed by the long-awaited meaningful vote. The Speaker allowed an amendment by Conservative backbencher Dominic Grieve, requiring the government to table a new motion within three sitting days, setting out the way forward, should they lose the vote. When Conservative loyalists challenged the decision, Bercow maintained that it was standard practice and that the Speaker’s judgements should ‘not [be] questioned by Members of the House’ (HC Debates, 9 January 2019, c.366). He would later admit, however, that parliamentary officials had advised him not to allow a vote on Grieve’s amendment, justifying his decision as a means of ensuring the government did not ‘exert unfettered control’ over parliament (Bercow, 2020, p. 373). When the amendment passed by a narrow margin of just 11 votes (308 votes to 297), it marked the second time in just a few weeks that Grieve had upset the government’s planned parliamentary timetable. The Guardian’s description of Theresa May as an ‘increasingly boxed-in Prime Minister’ (Stewart and Walker, 2019) was very apt. When the vote was finally held, the government suffered a massive defeat (202 votes to 432), the heaviest defeat in ‘the parliamentary era’ (Stewart, 2019). It was indeed the biggest loss since official parliamentary records began. If the government were not humiliated enough by the size of the defeat, Jeremy Corbyn’s decision to table a motion of no-confidence in the Prime Minister intensified the hostile atmosphere on the green benches. During the debate, Conservative backbenchers did not defend so much on May’s leadership as a question whether a general election was the correct solution. The motion was defeated by 19 votes and May reiterated her desire to have individual meetings with the opposition party leaders and with ‘groups of MPs who represent the widest possible range of views’ (May, 2019a), in the hope of finding a way forward for her withdrawal deal. Jeremy Corbyn would later refuse to take part. Promising the House that she would approach discussions constructively, May warned that ‘we must focus on ideas that are genuinely negotiable and have sufficient support’ (HC Debates, 16 January 2019, c.1126). She also stuck to what would become an increasingly common line, that the only way to prevent a no-deal Brexit was for MPs to vote for the government’s deal. What was unusual about this tumultuous parliamentary period though was that, although it was clear that the position of the government was much weaker than normal, the House of Commons was equally weak. Although it was proving itself capable of frustrating the government and thwarting its attempts to forge ahead with May’s negotiated Brexit deal, it too found itself thwarted by an overwhelming lack of consensus on an appropriate road to Brexit. Corbyn appealed to the House after the first meaningful vote defeat, saying that ‘it is not enough for the House to vote against the deal before us, and against no deal; we also have to be for something’ (HC Debates, 15 January 2019, c1109, emphasis added). Such was the deadlock that Ken Clarke told MPs that they were facing a ‘constitutional crisis’ (HC Debates, 29 January 2019, c.693). At the end of January 2019, MPs debated the Brexit deal for a second time. Among the seven amendments selected by the Speaker for discussion were those tabled by Dominic Grieve and Yvette Cooper. Both amendments were designed to give the Commons the opportunity to avoid a no deal Brexit, by providing MPs with space in the parliamentary timetable the following week to debate and vote on a set of alternatives to May’s deal (Grieve’s amendment) and to let MPs decide whether to extend Article 50 in a binding vote (Cooper). For Grieve, it would provide ‘the House the space in which to find where the majority lies’ (HC Debates, 29 January 2019, c.674), but for the Prime Minister, it was an attempt to ‘usurp the proper role of the Executive’ (HC Debates, 29 January 2019, c.673). When MPs voted that evening, they demonstrated that a majority were against a no-deal Brexit (Conservative backbencher Caroline Spelman’s amendment) and also against the current iteration of the Northern Ireland backstop (Graham Brady’s amendment) but the other amendments failed. During a further general debate on Brexit on 14 February, MPs gain tried again to amend a government motion on Brexit. Although unsuccessful, Jeremy Corbyn warned the Prime Minister to stop ‘ignoring Parliament’ (HC Debates, 14 February 2019, c.1159). 2. Backbench MPs continue to push back against the government In a change of strategy to try to prevent a no-deal Brexit, Yvette Cooper presented a Private Member’s Bill in the Commons—the European Union (Withdrawal) (No. 4) Bill. This was a presentation bill and as such was the type of legislation which would normally be used by MPs simply to draw attention to an issue. The bill had cross-party support and was drafted with Conservative MP Oliver Letwin. It outlined a mechanism through which the Commons could instruct the Prime Minister to ask for an extension to the Article 50 period if the government lost the meaningful vote on the deal for a second time. It was a continuation of MPs’ parliamentary action to try to prevent a no-deal Brexit. As the end of February approached, it looked as though the government might be starting to heed some of the Commons’ concerns. Conservative MP Alberto Costa resigned from his position as Parliamentary Private Secretary to the Secretary of State for Scotland, David Mundell, in order to table an amendment to guarantee the implementation of citizens’ rights and the government agreed to support it. The Prime Minister also committed to some of the provisions set out in the Cooper-Letwin bill, which would not progress beyond its first reading stage. After negotiating further assurances in relation to the Northern Ireland backstop, May gave MPs the opportunity to vote on her Brexit deal for a second time on 12 March—just 17 days before the UK was expected to leave the EU. In what was becoming a remarkably familiar line, she told MPs that they faced ‘a very clear choice: vote for and support this deal … or risk no deal’ (HC Debates, 12 March 2019, c.209). After five hours of debate, they overwhelmingly rejected the deal once again, by 391 votes to 242. Although this was a smaller defeat for the government compared to the first attempt, it was still a reverse of a scale not typically seen in British parliamentary politics. A total of 75 Conservative MPs, plus the DUP, had walked through the no lobby, alongside the other opposition parties. Labour’s Emma Reynolds summed up the general feeling on the backbenches: This last-minute deal, which really has not changed much, is just the latest chapter in the Tory party’s Brexit divisions and melodrama. When the Prime Minister says that she is listening to Parliament, she is actually listening to hard-line Tory Brexiteers and her confidence and supply partners, the DUP. When she says that she is acting in the national interest, she is actually putting her party’s interests above the prosperity of our constituents. She encourages us to come together, but she has done little to reach out across the House to appeal to Labour Members and other Opposition MPs (HC Debates, 12 March 2019, c.277). May told MPs that she ‘profoundly regret[ted] the decision’ (HC Debates, 12 March 2019, c.295), but that she would stand by the concessions she had made about the next appropriate parliamentary steps. Fearing that the government would simply put the same deal before the House for the third time and worried that parliament was being ‘shabbily treated’ by the government (Bercow, 2020, p. 382), the Speaker once again intervened, citing the convention that the same question cannot be put before MPs twice in any session. 3. A state of gridlock By the end of March, MPs had formally voted against leaving the EU on the Prime Minister’s terms on two occasions by staggering majorities. They had also voted firmly against leaving the EU without any deal at all. As the clock ticked down to the supposed exit day of 29 March, the government talked of a short extension to the Article 50 period. It was clear what MPs did not want, but it remained unclear what path out of the EU could be acceptable to parliamentarians and uncertain if a short extension would suffice to attain sufficiency of consensus. It quickly became apparent that the government would not seek a compromise, tabling a motion making an extension of the Article 50 period conditional on MPs voting for the government’s deal. Labour MP Mary Creagh told the House that the government was holding them ‘prisoner’ (HC Debates, 13 March 2019, c.468). The following day MPs again tabled amendments pressing for alternatives including a second referendum, but all were defeated. They did, however, agree to the government’s motion to ask the EU for a delay to the Brexit period beyond 29 March. By this point not only the scale of the intra-party divisions, but also the scale of inter-party cooperation, was far higher than is the norm for British parliamentary politics. Whereas MPs usually work on a cross-party basis on shared policy demands, the poor handling of the withdrawal process by the government united MPs on all sides of the debate against the Prime Minister’s deal. Damien Green admitted candidly that most of the MPs involved in cross-party efforts to make progress had ‘not found [themselves] signing amendments together much in the past’ (HC Debates, 13 March 2019, c.422). The Conservative MP, Oliver Letwin, for example, supported the government’s deal in every meaningful vote, yet still worked with Yvette Cooper to give the Commons a voice in finding a better, more workable alternative to that same deal and ensuring that a no-deal Brexit was taken off the table. For some, this was an indication that the party system was no longer fit for purpose. Labour’s Jess Phillips told MPs that it had been ‘blown to pieces’ by the Brexit debate (HC Debates, 13 March 2019, c.435). 4. Finding a way through the impasse Since the beginning of the year, there had been a continual tension between the wishes of many MPs to explore alternative options for a way forward and the government’s continued insistence that the only way forward was to vote for the deal. The Chair of the Brexit select committee, Hilary Benn, had found his committee’s requests for a series of indicative votes on alternatives firmly declined by the Prime Minister. On 14 March, Benn tried once again to push for indicative votes, tabling an amendment to allow MPs to take control of the parliamentary timetable in order to do so. Although his amendment was selected by the Speaker, MPs narrowly rejected the move. It would take two more weeks before Benn would get his wish. In a debate on 25 March, Oliver Letwin moved an amendment calling for business on 27 March to be set aside for MPs to hold a series of indicative votes. Government Minister David Lidington told the House that it would ‘upset the balance between legislature and executive in a way that would set an unwelcome precedent’ (HC Debates, 25 March 2019, c.68) but backbenchers queried the claim that the government had a longstanding constitutional right to determine Commons business (see HC Debates, 25 March 2019, cc.80, 81). Letwin’s amendment passed comfortably (329 votes to 302) in what Vince Cable described as a ‘major constitutional innovation’ (HC Debates, 25 March 2019, c.144). 5. The Commons takes control The indicative votes were just that: indicative rather than definitive. A vote in favour would not bind the government to any course of action but would show where the majority Brexit view lay. It was left to the Speaker to decide which motions to select; eight were chosen and announced to the House on the afternoon of 27 March. A business motion moved by Oliver Letwin set aside not just that day but also the following Monday if no majority view was found. For MPs like Letwin, it was the culmination of an attempt to prove that sovereignty lay with parliament, rather than with the government (see HC Debates, 27 March 2019, c.343, 344). When the results of the votes were announced later that evening, not a single alternative had been able to secure a majority. The narrowest result, Ken Clarke’s customs union proposal, was defeated by eight votes. Several proposals had received more support from MPs than May’s deal had just a few weeks before, but they had not been able to achieve the necessary majority. Later that week the Prime Minister tried and failed for a third time to gain Commons approval for her Brexit deal. To get around the Speaker’s restrictions on putting the same question twice, the government asked MPs to vote only on the withdrawal agreement and not on the political declaration. Despite asking MPs to ‘look into [their] hearts and decide what is best for our constituents and our country’ (HC Debates, 29 March 2019, c.769), the third meaningful vote was defeated by 286 votes to 344. May summarised the extreme parliamentary stalemate in her address to the House immediately afterwards: I fear that we are reaching the limits of this process in this House. This House has rejected no deal; it has rejected no Brexit; on Wednesday it rejected all the variations of the deal on the table; and today it has rejected approving the withdrawal agreement alone and continuing a process on the future (HC Debates, 29 March 2019, c.775). Her words would be mirrored by activity in the Commons the following week when Oliver Letwin put a motion before the House to pursue a second round of indicative votes. This time the Speaker selected only four motions for MPs to consider: a customs union, a Common Market 2.0 motion (proposing membership of the European Free Trade Association and the European Economic Area plus a comprehensive customs arrangement with the EU), a confirmatory public vote and an SNP motion asking for an extension to the Brexit process or a clear vote between no-deal and a revocation of Article 50. All four were defeated, but Ken Clarke’s customs union motion lost out by a very narrow margin of just three votes. Also significant was the motion for a confirmatory public vote, defeated by just 12 votes (292 votes to 280). A sizeable number of MPs now clearly favoured giving the British public a further say. At this point, though, the UK was set to leave the EU in 11 days, yet MPs had rejected May’s Brexit deal and every alternative proposal put before them. MPs said it would be an ‘outrage’ if the government tried to hold a fourth vote (HC Debates, 1 April 2019, c.880). Exasperated by the failure of his own motion in the evening’s votes, Conservative Nick Boles announced his resignation from the Conservative Party, crossing the floor of the Commons to a round of applause from opposition MPs. The following day Yvette Cooper introduced another Private Member’s Bill, the European Union (Withdrawal) (No. 5) Bill, backed by MPs from all parties including Letwin, Grieve and Benn, requesting another Article 50 extension. The bill aimed to give clarity to the process of asking for an extension to exit day and to ensure that the Commons would continue to have a role to play in the process. Cooper’s bill completed its Commons stages in a single day, passing its third reading by just a single vote. Moving on to the House of Lords, it received Royal Assent just five days later. This forced the government to bring forward a motion extending the Article 50 deadline to 30 June, which was passed with a hefty majority. A new date of 31 October would later be agreed with the EU. 6. An increasingly fractured government and an isolated Prime Minister As the stalemate extended into May 2019, the government hoped to hold a fourth meaningful vote and to introduce the necessary Brexit legislation (the Withdrawal Agreement Bill). The already difficult parliamentary context was exacerbated though by the momentum forming for the Prime Minister’s departure. Already a group of 14 Conservative MPs had written to May to warn against including a customs union in any compromise deal with the opposition parties. It was no surprise, then, that at a meeting with the 1922 Committee of Conservative MPs, the Prime Minister agreed to formalise a timetable for her departure. Despite the overt lack of support now for May’s leadership from both the Conservative Party and from the Commons, the Prime Minister still outlined her plans for a ‘new deal’ to MPs (HC Debates, 22 May 2019, c.731). She outlined a ten-point plan to ‘end the political impasse’ which would be enshrined in the forthcoming Withdrawal Agreement legislation including a commitment to replace the Northern Ireland backstop by December 2020, a workers’ rights bill and a guarantee that MPs would be able to vote on any treaties governing the UK’s future relationship with the EU. Jeremy Corbyn told her that she had lost the ‘authority to deliver’ (HC Debates, 22 May 2019, c.734) and that, with no guarantee that her replacement would stand by any of these guarantees, it was a case of too little, too late. Just two days later, May formally announced her departure saying that she had done everything she could to convince MPs to back her deal (May, 2019b). The Conservative Party’s disastrous performance at the European Parliament elections just a few days later was a sign that the public had lost any faith they had in the government delivering the Brexit they had promised. Following a ballot of MPs and party members, Boris Johnson was elected as the next leader of the Conservative Party, taking over from Theresa May as Prime Minister at the end of July. 7. A new Prime Minister facing the same parliamentary problems Boris Johnson’s first speech from Downing Street saw him on a mission to prove ‘the doubters, the doomsters [and] the gloomsters’ that they were wrong about the government’s chances of honouring the 2016 referendum (Johnson, 2019a). Promising to leave by 31 October with ‘no ifs or buts’ (Johnson, 2019a) he had just 99 days to succeed in what his predecessor had failed to do over the last three years. In his first appearance before MPs in the Commons, Johnson promised to ‘turbo charge’ the Brexit process after the summer recess (HC Debates, 25 July 2019, c.1458). At the end of August, however, he announced that the Queen had approved a prorogation of parliament for most of September and early October. MPs were understandably concerned about such a move, which would inhibit parliamentary scrutiny at a critical time. As well as the lack of progress on parliament’s approval of the Brexit deal, other legislation was also building up, including the Trade Bill and the Agriculture Bill. The Commons Speaker told the press that it was ‘blindingly obvious’ that the aim of prorogation was to prevent scrutiny of Brexit (BBC, 2019). The Prime Minister’s determination to prorogue parliament was matched by a similarly determined Commons which did not want the government to go unscrutinised. When the House returned in early September, Hilary Benn introduced another piece of private member’s legislation—the European Union (Withdrawal) (No. 6) Bill—giving the Prime Minister until 19 October to pass his Brexit deal, to pass a no-deal motion, or to seek a further extension to Article 50. The bill, which even went so far as to set out word-for-word the text of the letter which the Prime Minister would be required to send to the European Council President, demonstrated that the Commons would not be relinquishing its new found power over the government any time soon. It set 31 January 2020 as the new deadline to leave the EU. Passing Benn’s bill required Oliver Letwin to secure an emergency debate in the Commons, in which MPs passed a motion to take control of parliamentary business once more. When the tellers announced the result (328 votes to 301), it was clear that the responsibility fell on the shoulders of 21 Conservative MPs who had joined their opposition colleagues in the ‘aye’ lobby. They included a now familiar set of faces (Grieve, Letwin and Clarke) but also a broader contingent of former cabinet ministers and previously loyal MPs (Anne Milton, Philip Hammond and Ed Vaizey). All would lose the party whip that night and sit as independent Conservative MPs.2 Their expulsion only served to exacerbate the government’s minority position in the Commons. When two ministers, Jo Johnson and Amber Rudd, resigned from the Cabinet Johnson’s position in the Commons fell even further. The previous parliamentary session seemed to be repeating itself. 8. The bumpy road to a general election In his formal response to the vote on Letwin’s motion, the Prime Minister told MPs they had left him with no choice but to push for a general election. Putting the blame squarely in parliament’s hands, he accused MPs of ‘wrecking’ the Brexit deal and bringing ‘more dither, more delay and more confusion’ (HC Debates, 3 September 2019, c.140). A government motion was subsequently tabled in line with the provisions of the FTPA. As Benn’s bill passed its second reading and committee stage the next day, Johnson moved the motion for an early election. When the result was announced, the Prime Minister found himself in the strange position of having won the vote (298 votes to 56) but losing the prize of a general election as he had not met the threshold of 434 votes specified in the legislation. The bill received Royal Assent on 9 September, the day on which parliament was due to be prorogued. Dominic Grieve successfully lobbied the Speaker for an emergency debate on the prorogation, forcing the government to come to the House and explain themselves. Raising his concerns with Bercow about the unprecedented parliamentary manoeuvre, he questioned the Prime Minister’s motives, citing the difference between his public statements and ‘what the evidence suggests is the reality’ (HC Debates, 9 September 2019, c.519). He was duly granted an emergency debate on the matter in which he requested copies of all formal and informal communication from government be presented to parliament through a humble address. The motion was agreed to by 311 votes to 302. Johnson then tried once again to get the House to agree to an early general election. Referring to Benn’s legislation as the ‘surrender bill’, he told MPs that an election was ‘the only way to break the deadlock in the House’ (HC Debates, 9 September 2019, c.616). For a second time, the government won the vote, largely thanks to Labour’s abstention (293 votes to 46) but failed to achieve the majority required under the FTPA. The SNP leader in the Commons congratulated Johnson for managing to lose ‘every vote he has brought to the House since he became Prime Minister’ (HC Debates, 9 September 2019, c.640). Bercow would later describe Johnson’s treatment of parliament as a ‘constitutional outrage’ (Proctor, 2019). When the Supreme Court unanimously ruled that the prorogation had prevented parliament from exercising ‘its constitutional functions’ and was thus unlawful, MPs returned to the Commons for what the Speaker said was the ‘most peculiar atmosphere’ he had ever known (Bercow, 2020, p. 1). The Prime Minister was met with cries of ‘resign!’ from the opposition benches (see HC Debates, 25 September 2019, c.774) and challenged them to table a no-confidence motion. The resulting hostile atmosphere in the House meant that, when the government tried to adjourn the Commons for Conservative party conference, MPs refused, defeating the motion by 289 votes to 306. 9. A new parliamentary session As a new parliamentary session began in mid-October 2019, following a lawful prorogation, MPs agreed to sit on a Saturday for the first time in 37 years to debate the government’s new Brexit deal and have a further attempt at a meaningful vote. The Speaker selected another cross-party amendment, moved by Oliver Letwin, to give the House the chance to scrutinise the text of the necessary withdrawal legislation before having any meaningful vote. As was now becoming a regular occurrence for Letwin’s amendments, the House gave its seal of approval (322 votes to 306) and, as such, the meaningful vote could not be put to the House. The government announced its intention to proceed with the vote the following week but the Speaker disallowed it on the grounds that it ‘would be repetitive and disorderly to do so’ (HC Debates, 21 October 2019, c.696). By then, the Prime Minister had anyway been obliged to follow the measures set out in the Benn legislation, sending a letter to the European Council requesting an extension to Article 50, which the Council later agreed to, giving a new Brexit date of 31 January 2020. A few days later, the EU (Withdrawal Agreement) Bill came before MPs. The government had allocated just three days for the bill to complete its scrutiny, something all parties warned was ‘simply not good enough’ for such important legislation which had only been printed the evening before (HC Debates, 21 October 2019, c.735; c.884). Although the bill passed its second reading with some ease, the programme motion for its committee stage was defeated (308 votes to 322). After the vote, Jeremy Corbyn described the Prime Minister as the ‘author of his own misfortune’ (HC Debates, 21 October 2019, c.926) and the government decided to pause the further progress of the bill. Frustrated once more by the Commons, the government put forward a third general election motion on the grounds that parliament had ‘run its course’ and was not ‘capable of delivering on the priorities of the people’ (HC Debates, 29 October 2019, c.54). Jeremy Corbyn pledged that Labour would not support an election motion until no-deal had been taken off the table and it was clear that people (particularly students) would not be disenfranchised by an election date right before Christmas, while the SNP wanted to force the Prime Minister to stop Brexit rather than risk giving him a mandate or majority to finish it. Only the Liberal Democrats expressed their support for an election. It was no surprise, then, that once again the required majority was not reached. Most Labour MPs abstained, along with several of the former Conservative MPs who had had the whip withdrawn. Johnson told MPs the result was ‘utterly bewildering’ and that the government would introduce a general election bill instead, to finally end the ‘dysfunctional parliament’ (HC Debates, 29 October 2019 c.79). As a standard piece of legislation, the Early Parliamentary General Election Bill required only a simple majority of MPs to vote in favour. Specifying 12 December 2019 as polling day under the FTPA, the two-clause bill passed its Commons stages in a single day, its third reading supported by 438 votes to 20. Johnson had finally managed to circumvent the Act which the party’s manifesto would later commit to removing altogether. 10. The 2019 general election and parliament As MPs returned to their constituencies to campaign in the 2019 general election, the Conservative government had set a series of political and parliamentary records. Over the preceding two years, over a quarter of the Cabinet had resigned, over 100 backbench MPs had voted to remove a Prime Minister from her position, the government had been defeated by the largest margin in parliamentary history and it had been censured by the Supreme Court for an unlawful prorogation. MPs were not relishing the thought of another general election so soon but it was the inevitable consequence of the state of gridlock and inertia which had brought the government’s legislative programme to a standstill. The campaign itself was mired in a sense of distrust or disbelief from the public at the behaviour of their elected representatives. For some, there was an irony in the perpetual voting on Brexit, given that the public had only had the chance to vote once in the referendum (e.g. HC Debates, 14 March 2019, c.609). MPs had become ever more conscious of the public’s view of the Commons as the 2017 parliament had progressed. They voiced their constituents’ concerns about the ‘childish antics’ of MPs (HC Debates, 29 January 2019, c.686) and the general sense of exasperation with parliament. Some felt that it was now their duty to let the public decide what the composition of the new parliament, and thus their preferred path out of the European Union, should be. Johnson’s campaign slogan of ‘Get Brexit Done’ tied neatly with this sense of frustration that constituents had been relaying to their MPs. As the results came in during the early hours of 13 December, it was clear that the 2019 parliament would be a very different beast to its predecessor. The large majority for the Conservatives put Prime Minister Boris Johnson in a much stronger parliamentary position, providing the parliamentary buffer needed to progress smoothly with the EU withdrawal process. Many of the Conservative rebels, such as Ken Clarke and Oliver Letwin, who had been such persistent thorns in the government’s side, had already announced their intention to stand down at the election. Others stood as independent or Liberal Democrat candidates. All of them failed to win their seats. They included Dominic Grieve, the architect of so many government defeats, who lost his Beaconsfield seat to Conservative candidate Joy Morrissey. Ten of the suspended Conservative MPs had the party whip restored in October 2019 but just four of these (Steve Brine, Greg Clark, Stephen Hammond and Caroline Nokes) contested the election. John Bercow, the Commons Speaker who had granted so many backbench amendments, also stood down from parliament. The Prime Minister therefore found himself relieved of the three major bugbears of his premiership so far; the lack of a parliamentary majority, the activist Speaker and most of the rebellious backbenchers, had now gone. 11. Conclusion The 2019 general election marked the end of an extremely volatile period of parliamentary politics. Pete Wishart’s description of the events of the 2017 parliament as a very ‘British parliamentary coup’, one ‘conducted with points of order and copies of Erskine May rather than through military means’, was very apt (HC Debates, 1 April 2019, c.804). Armed with the support of the Commons Speaker, MPs had provided a visible demonstration to the government that it did not have unfettered power. Although they were unable to put forward an alternative to the government’s Brexit deal that an absolute majority of MPs could get behind, they had shown the government that if it treated the parliament with contempt it would be quickly reined in. The power of the Commons was wholly unusual, but the resurgence was to be short-lived. With an ‘overwhelming mandate’ from the people at the general election, the Prime Minister’s promise of a ‘parliament that works for you’ (Johnson, 2019b) was confirmed by the passage of the troublesome Withdrawal Agreement Bill in January 2020 with a majority of 99. It seems unlikely that the relationship between parliament and government will be as fraught for the foreseeable future, but the memories of what the Commons can do to a struggling government will no doubt linger on. Footnotes 1 Figures from parliamentary search tool using the terms ‘Urgent questions’ AND brexit session : 17/19 2 Sam Gyimah joined the Liberal Democrats and he was joined by former colleague Antoinette Sandbach a few weeks later. References BBC ( 2017 , 13 December) ‘Brexit Bill: Government Loses Key Vote after Tory Rebellion’, accessed at https://www.bbc.co.uk/news/uk-politics-42346192 on 3 April 2020. BBC ( 2018 , 5 December) ‘Theresa May Suffers Three Brexit Defeats in the Commons’, accessed at https://www.bbc.co.uk/news/uk-politics-46446694 on 3 April 2020. BBC ( 2019 , 28 August) ‘Parliament Suspension: Queen Approves PM’s Plan’, accessed at https://www.bbc.co.uk/news/uk-politics-49493632 on 10 May 2020. Bercow J. ( 2020 ) Unspeakable: The Autobiography, London, Weidenfeld and Nicolson. Cabinet Office ( 2017 , 26 June) ‘Confidence and Supply Agreement between the Conservative and Unionist Party and the Democratic Unionist Party’, accessed at https://www.gov.uk/government/publications/conservative-and-dup-agreement-and-uk-. accessed at government-financial-support-for-northern-ireland/agreement-between-the-conservative-and-unionist-party-and-the-democratic-unionist-party-on-support-for-the-government-in-parliament#confidence-and-supply-agreement-in-the-uk-parliament on 24 April 2020. HC Debates (13 November 2017 ) c.37. HC Debates (10 December 2018 ) c.23. HC Debates (10 December 2018 ) c.28. HC Debates (10 December 2018 ) c.31. HC Debates (10 December 2018 ) c40. HC Debates (11 December 2018 ) c.151. HC Debates (9 January 2019 ) c.366. HC Debates (15 January 2019 ) c.1109. HC Debates (16 January 2019 ) c.1126. HC Debates (29 January 2019 ) c.674. HC Debates (29 January 2019 ) c.686. HC Debates (29 January 2019 ) c.693. HC Debates (14 February 2019 ) c.1159. HC Debates (12 March 2019 ) c.209. HC Debates (12 March 2019 ) c.277. HC Debates (12 March 2019 ) c.295. HC Debates (13 March 2019 ) c.468. HC Debates (14 March 2019 ) c.609. HC Debates (25 March 2019 ) c.68. HC Debates (25 March 2019 ) cc.80–81. HC Debates (25 March 2019 ) c.144. HC Debates (27 March 2019 ) cc.343–344. HC Debates (29 March 2019 ) c.769. HC Debates (29 March 2019 ) c.775. HC Debates (1 April 2019 ) c.804. HC Debates (1 April 2019 ) c.880. HC Debates (22 May 2019 ) c.731. HC Debates (22 May 2019 ) c.734. HC Debates (25 July 2019 ) c.1458. HC Debates (3 September 2019 ) c.140. HC Debates (9 September 2019 ) c.519. HC Debates ( 9 September 2019 ) c.616. HC Debates (9 September 2019 ) c.640. HC Debates (25 September 2019 ) c.774. HC Debates (21 October 2019 ) c.696. HC Debates (21 October 2019 ) c.735; c.884. HC Debates (21 October 2019 ) c.926. HC Debates (29 October 2019 ) c.54. HC Debates (29 October 2019 ) c.79. Johnson B. ( 2019 a, 24 July) ‘Boris Johnson’s First Speech as Prime Minister’, accessed at https://www.gov.uk/government/speeches/boris-johnsons-first-speech-as-prime-minister-24-july-2019 on 20 April 2020. Johnson B. ( 2019 b, 13 December) ‘PM Statement in Downing Street’, accessed at https://www.gov.uk/government/speeches/pm-statement-in-downing-street-13-december-2019 on 20 April 2020. May T. ( 2017 , 12 September) ‘PM Statement on EU Withdrawal Bill’, accessed at https://www.gov.uk/government/news/pm-statement-on-eu-withdrawal-bill-12-sept-2017 on 20 April 2020. May T. ( 2019 a, 16 January) ‘PM’s Statement at Downing Street’, accessed at https://www.gov.uk/government/speeches/pms-statement-at-downing-street-16-january-2019 on 14 April 2020. May T. ( 2019 b, 24 May) ‘PM’s Statement in Downing Street’, accessed at https://www.gov.uk/government/speeches/prime-ministers-statement-in-downing-street-24-may-2019 on 20 April 2020. Priddy S. ( 2020 , 7 April) Number of Urgent Questions in the Commons since 1997, London, House of Commons Library, CBP 08344, accessed at: file:///C:/Users/Team%20Knowhow/Downloads/CBP-8344.pdf on 11 April 2020. Proctor K. ( 2019 , 28 August) ‘Boris Johnson’s Move to Prorogue Parliament a ‘Constitutional Outrage’ Says Speaker’, The Guardian, accessed at https://www.theguardian.com/politics/2019/aug/28/boris-johnsons-move-to-prorogue-parliament-a-constitutional-outrage-says-speaker on 22 April 2020. Sharman J. ( 2017 , 13 December) ‘Stephen Hammond: Tory MP Sacked as Conservative vice-chairman after Brexit Rebellion’, The Independent, accessed at https://www.independent.co.uk/news/uk/politics/stephen-hammond-fired-sacked-brexit-rebellion-tory-mp-theresa-may-latest-a8108691.html on 26 April 2020. Stewart H. , Walker P. ( 2019 , 10 January) ‘Brexit: May Loses Grip on Deal after Fresh Commons Humiliation’, The Guardian, accessed at https://www.theguardian.com/politics/2019/jan/09/may-loses-grip-on-brexit-deal-after-fresh-commons-humiliation on 24 April 2020. Stewart H. ( 2019 , 16 January) ‘May Suffers Heaviest Parliamentary Defeat of a British PM in the Democratic Era’, The Guardian, accessed at https://www.theguardian.com/politics/2019/jan/15/theresa-may-loses-brexit-deal-vote-by-majority-of-230 on 13 April 2020. © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Power,, Sam;Bale,, Tim;Webb,, Paul

doi: 10.1093/pa/gsaa022pmid: N/A

When all is said and done, what might be the most surprising thing about the Conservative Party’s victory in 2019 was how unsurprising it was. As the UK voted on 12 December, almost every poll had the Tories ahead by at least six percentage points (BBC, 2019). And yet much was made during the campaign itself of the result being supposedly hard to call (Pienaar, 2019). This was partly perhaps because commentators were chastened by unexpected results in 2015 and 2017. But it was also because, in early October, the British Election Study (BES) had shown that the electorate was the ‘most volatile in modern times’ (British Election Study, 2019), with even the godfather of psephology (indeed, the inventor of the term) Sir David Butler stating that he had ‘never felt more confused and uncertain’ about an electoral outcome (Payne, 2019). Yet the dawn which broke on 13 December brought Conservatives what for some of them must have been delayed gratification: this was the victory that 2017 was supposed to have been. The difference, as Johnson’s former colleague from City Hall Guto Harri argued, was in the execution: ‘he’s got the same deck of cards as Theresa May, but he’s played them differently’ (McTague, 2019). 1. ‘I’m not the artist; I’m merely the subject—it is for you to apply the rich chiaroscuro to your canvas’2 One of the more interesting analogies made during the general election came from an unusual source. Matthew McGregor—a veteran digital political strategist who had worked on Obama’s re-election campaign in 2012, but also extensively for the Labour Party and Blue State Digital—argued that to understand everything you needed to know about the Conservative Party strategy in 2019, you should go back not to 2017 but rather to 2008: I worked for the Labour Party in 2008, for Boris Johnson’s first mayoral run … and they had very little to say. I mean the entire campaign was, ‘I’m going to get rid of bendy buses’ … he makes himself a very, very big character outside of election time and then in election campaigns he makes himself as small a target as he possibly can and just tries to dance his way through without taking too many hits. So, they’ve built up all the momentum before the election campaign and now they’re trying to coast through quietly. (Commons People, 2019) It might well be the case, nonetheless, that Conservative Campaign Headquarters (CCHQ) had also learned the hard lessons from a 2017 campaign that put Theresa May front and centre—in terms of direct mail and leaflets, if not during TV debates. Part of this approach was down to Isaac Levido, the election strategist who had worked on the 2015 Cameron campaign and Scott Morrison’s surprise re-election in Australia earlier in 2019. Levido was said to favour a ‘safety-first’ approach and ‘insisted on strict message discipline … rarely veering from [Johnson’s] slogan to “Get Brexit Done”, and dodging the most difficult television interviews that risked tripping him up’ (Donaldson and Ross, 2019). The campaign, full of ‘clunking metaphors’ (Johnson, 2019) such as driving a JCB through a wall to symbolise breaking the parliamentary deadlock was, in fact, ‘classic Johnson, giving the appearance of chaotic joviality’ whilst in reality ‘saying nothing to distract from the campaign script’ (McTague, 2019). The Conservative Party manifesto itself is a neat encapsulation of this approach. It came in at just over 60 pages, nearly half that of Labour’s, and was so bereft of detail that PoliticsHome dubbed it the ‘meh-nifesto’ (Langford, 2019). It contained relatively modest spending proposals of £2.9 billion, especially when compared with the Liberal Democrats' £63 billion and Labour’s £83 billion. This made some in CCHQ anxious, with Labour being perceived as the party of change and the Tories as the party of more of the same (BBC, 2020). But it reflected a calculated gamble. Whilst the Tories might well have been seen as failing to properly address nearly ten years of austerity, they hoped that Labour’s more full-hearted approach to public spending would simply be seen as unrealistic (Partington, 2019). Ultimately, that gamble paid off. Moments like Culture Secretary Nicky Morgan coming unstuck on Good Morning Britain with regards to how many of the promised 50,000 new nurses were in actuality ‘new’, represented little more than one of a few fleeting ‘gotcha’ moments of the campaign, rather than more fundamental dissections of a manifesto which was otherwise seen as a more viable offer than Labour’s. Michael Gove, whilst disagreeing that the manifesto was ‘light’, admitted as much when suggesting that it was ‘written in such a way as to make sure that people had a clear sense of the direction we wanted to take the country and also to avoid some of the rows and distractions that had derailed Theresa’s campaign in 2017’ (BBC, 2020). It was, indeed, left to the public to apply the rich chiaroscuro to the (relatively) blank canvas. Whilst surely a hangover from the failures of 2017, this strategy might also have reflected an acceptance amongst party higher-ups, including Johnson himself, that he could be a rather ‘marmite’ figure. We know from work conducted by the campaign organisation Who Targets Me that all parties were relentlessly A/B testing all kinds of messaging during the early stages of the campaign on Facebook (Doward, 2019). This led to the Conservatives adopting varied marketing strategies, depending on the constituencies they were targeting. In Leave-voting constituencies, adverts solely featuring Johnson were favoured, whereas in Remain-voting constituencies, (negative) adverts solely featuring the opposition were favoured (Who Targets Me, 2019). In terms of campaign management, then, there was a pretty basic cost–benefit analysis at play. There was simply more to be lost than gained from Johnson being overly visible on the election trail. If that meant taking short-term reputational damage from, say, appearing only as an ice sculpture at a Channel 4 debate on climate change, or refusing a sit-down interview with BBC Rottweiler Andrew Neil, then so be it. As Michael Gove put it, when asked whether it was a mistake for Johnson not to be questioned by Neil (BBC, 2020): No. We won … with the best will in the world … the purpose of running an election campaign is to win so that you can govern the country well, not to agree to every broadcast bit. Indeed, those rare occasions when Johnson did encounter significant scrutiny suggested the Tories’ less-is-more strategy was a sensible one. The final days of the campaign witnessed farcical scenes featuring Johnson seeking refuge in a large fridge to hide from a particularly enthusiastic Good Morning Britain reporter, as well as a slightly more sinister pocketing of a reporter’s phone when he confronted the PM with a picture of a child forced to sleep on a hospital floor. Despite all this, when Johnson did put himself forward for scrutiny in head-to-head leaders’ debates he performed as well, if not marginally better than Jeremy Corbyn, at least in as much as these things can be measured. Both the BBC and ITV debates were effectively fought to a ‘score draw’. YouGov found 51% of voters backing Johnson as the winner in the ITV debates (Ibbetson, 2020) whereas the BBC debates (Smith, 2019)—held a full three weeks later—had Johnson on 52% to Corbyn’s 48%. If, on the issues there was little between the two, when the public were asked who would make the best prime minister, there was considerably more daylight. According to a post-election poll conducted by Lord Ashcroft, 49% of voters said that Johnson would make the best prime minister, with just 31% saying the same of Corbyn (Ashcroft, 2019a). Finally, a more light-hearted poll, released just days before the election might have been the clearest indication of the public’s true feelings, suggesting Johnson was simply seen as the best of a not particularly desirable bunch. YouGov (2019a) asked ‘If you had to, which of the main party leaders would you most like to spend Christmas Day with?’; here, excluding ‘don’t knows’, Johnson was the clear winner. He stood alone at 22%, with Corbyn on just 14%, Farage on 10% and Swinson and Sturgeon on 8%. With any election, you do need a little bit of luck, and many problems that the Conservatives faced—or might have faced—were quickly nullified by weaknesses in their opposition. For example, the first week of the campaign got off to a rocky start when Welsh Secretary Alun Cairns resigned over claims that he was aware of a former aide’s hand in ‘sabotaging’ a rape trial and Jacob Rees-Mogg apologised when appearing to imply that Grenfell Tower victims lacked common sense (O’Donoghue, 2019). However, at the same time, Corbyn had to contend with two (former) MPs doing the rounds of the weekend politics shows urging voters not to plump for the Labour Party. Similarly, the London Bridge terror attack might well have been a moment in which cuts to defencing and policing were brought front and centre, but in the eyes of many, Corbyn now lacked the moral authority to make that argument. The Conservatives might also have been tripped up by a rather unhelpfully timed visit from Donald Trump but he—aside from suggesting that the US would not ‘take’ the NHS ‘if you handed it to us on a silver platter’ and describing Johnson as ‘very capable’—largely stood aside from the fray (Langford, 2019). 2. The organisational strengths From Johnson’s victory in the leadership contest, we know, and have always known, that one thing Johnson can do is excite the Conservative base—and the organisational shortcomings that plagued the Conservative campaign in 2017 were lessened here. Whilst membership was still well below that of Labour (as of July 2019 the Conservatives had around 180,000), it still represented a modest increase on the 2017 number which stood at circa 150,000. Moreover, Labour membership, whilst still impressive, had reportedly fallen off considerably since ‘peak Jeremy’. PoliticsHome, in July 2019, had the ‘bleakest’ report that Labour membership had dropped to 485,000, from the 518,659 reported in the 2018 accounting returns to the Electoral Commission (Schofield, 2019), and nearly a full 80,000 down from the high of 564,443 reported in December 2017 (Audickas et al., 2019). The revival of Tory membership numbers did not, however, mean that the party’s grassroots activity during the election campaign matched, let alone surpassed, that of the Labour Party—nor, in proportionate terms, were its members generally as active as any other party's bar the UK Independence Party's (UKIP) or the Brexit Party’s, as Table 4.1 shows. Tory members reported having done an average of 1.4 campaign activities in 2019, which is significantly below Labour’s score of 2.4 activities. And there is little comfort to be gained from the fact that UKIP and the Brexit Party fared even worse in these terms, given that many of their members or supporters will have lived in seats where there was no candidate from their parties. Moreover, Conservative footsoldiers were generally less inclined to involve themselves in campaign activity than they were in 2017 or 2015 (Bale et al., 2019, p. 100). Party Members Project data reveal that some 57% of an enlarged Conservative membership admitted to having devoted no time at all to campaigning in 2019, compared to 44% in 2017 and just 29% in Cameron’s victorious campaign of 2015. Although the trend has been similarly negative across most of the parties, it has been more pronounced for the Tories. For instance, the respective figures for non-campaigners among Labour’s members are 27% in 2015, 33% in 2017 and 40% in 2019. Table 4.1 Which of the following things did party members do for their parties during the 2019 General Election campaign? Activity . Conservative . Labour . Liberal Democrat . UKIP/ Brexit . Green . SNP . Total . ‘Liked’ something by party/candidate on FB 39.4 56.4 45.5 35.0 48.0 65.7 48.4 (+0.02) (−7.5) (−17.7) (+0.2) (−15.6) (−5.0) (−9.0) Tweeted/re-tweeted party messages 22.8 39.4 32.1 18.2 33.8 37.1 30.7 (−1.5) (+0.5) (−8.0) (−0.6) (−6.0) (−6.6) (−5.0) Displayed election poster in window 13.6 40.2 31.0 11.6 22.4 46.1 27.7 (−8.0) (−16.0) (−17.2) (−22.0) (−17.3) (−12.9) (−16.7) Delivered leaflets 20.8 27.7 32.9 8.5 17.6 27.6 22.6 (−9.7) (−3.9) (−11.1) (−19.2) (−5.3) (−2.8) (−9.7) Attended public meeting or hustings 19.4 24.3 21.6 16.2 18.5 27.5 21.3 (−0.4) (−0.8) (−1.5) (−13.8) (−1.0) (−7.2) (−3.9) Drove voters to polling stations 3.3 4.1 2.1 1.0 0.8 5.6 2.9 (+1.0) (−0.5) (−0.6) (−3.8) (−1.2) (+0.1) (−0.7) Canvassed face to face or by phone 15.3 23.8 17.1 6.9 9.9 20.9 16.0 (−8.0) (−3.0) (−5.5) (−9.3) (−4.1) (+0.9) (−5.0) Helped run party committee room 3.8 5.2 6.4 1.4 2.8 5.7 4.2 (−3.2) (+1.2) (+0.9) (−3.4) (+0.4) (+1.3) (−0.6) Stood as candidate (councillor or MP) 0.9 2.2 2.7 1.3 2.3 0.3 1.6 (−1.3) (+1.4) (−0.9) (−3.5) (−1.0) (−0.7) (−0.7) Other 5.0 4.0 7.3 4.0 5.5 5.6 5.1 (−5.2) (−9.8) (−8.9) (−2.2) (−5.2) (−4.1) (−6.7) None 20.4 13.1 14.8 31.1 20.5 9.0 18.1 (−4.3) (+4.2) (+4.0) (+5.5) (+3.5) (+0.7) (+3.5) Campaign Activism Index—Mean 1.39 2.23 1.91 1.00 1.56 2.37 1.75 (−0.29) (−0.29) (−0.62) (−0.76) (−0.51) (−0.32) (−0.52) Number 1191 1377 1044 1112 1023 1038 6785 Activity . Conservative . Labour . Liberal Democrat . UKIP/ Brexit . Green . SNP . Total . ‘Liked’ something by party/candidate on FB 39.4 56.4 45.5 35.0 48.0 65.7 48.4 (+0.02) (−7.5) (−17.7) (+0.2) (−15.6) (−5.0) (−9.0) Tweeted/re-tweeted party messages 22.8 39.4 32.1 18.2 33.8 37.1 30.7 (−1.5) (+0.5) (−8.0) (−0.6) (−6.0) (−6.6) (−5.0) Displayed election poster in window 13.6 40.2 31.0 11.6 22.4 46.1 27.7 (−8.0) (−16.0) (−17.2) (−22.0) (−17.3) (−12.9) (−16.7) Delivered leaflets 20.8 27.7 32.9 8.5 17.6 27.6 22.6 (−9.7) (−3.9) (−11.1) (−19.2) (−5.3) (−2.8) (−9.7) Attended public meeting or hustings 19.4 24.3 21.6 16.2 18.5 27.5 21.3 (−0.4) (−0.8) (−1.5) (−13.8) (−1.0) (−7.2) (−3.9) Drove voters to polling stations 3.3 4.1 2.1 1.0 0.8 5.6 2.9 (+1.0) (−0.5) (−0.6) (−3.8) (−1.2) (+0.1) (−0.7) Canvassed face to face or by phone 15.3 23.8 17.1 6.9 9.9 20.9 16.0 (−8.0) (−3.0) (−5.5) (−9.3) (−4.1) (+0.9) (−5.0) Helped run party committee room 3.8 5.2 6.4 1.4 2.8 5.7 4.2 (−3.2) (+1.2) (+0.9) (−3.4) (+0.4) (+1.3) (−0.6) Stood as candidate (councillor or MP) 0.9 2.2 2.7 1.3 2.3 0.3 1.6 (−1.3) (+1.4) (−0.9) (−3.5) (−1.0) (−0.7) (−0.7) Other 5.0 4.0 7.3 4.0 5.5 5.6 5.1 (−5.2) (−9.8) (−8.9) (−2.2) (−5.2) (−4.1) (−6.7) None 20.4 13.1 14.8 31.1 20.5 9.0 18.1 (−4.3) (+4.2) (+4.0) (+5.5) (+3.5) (+0.7) (+3.5) Campaign Activism Index—Mean 1.39 2.23 1.91 1.00 1.56 2.37 1.75 (−0.29) (−0.29) (−0.62) (−0.76) (−0.51) (−0.32) (−0.52) Number 1191 1377 1044 1112 1023 1038 6785 Notes: All activities figures are percentages. Figures in brackets are changes since 2017. Campaign activism index is based on an additive scale that runs from 0 (no activity during the election campaign) to 9 (maximal activity during the campaign, excluding ‘other’). All relationships between party and type of campaign activity reported in this table are significant at p < 0.001. Source: Party Members Project Survey (2019). Open in new tab Table 4.1 Which of the following things did party members do for their parties during the 2019 General Election campaign? Activity . Conservative . Labour . Liberal Democrat . UKIP/ Brexit . Green . SNP . Total . ‘Liked’ something by party/candidate on FB 39.4 56.4 45.5 35.0 48.0 65.7 48.4 (+0.02) (−7.5) (−17.7) (+0.2) (−15.6) (−5.0) (−9.0) Tweeted/re-tweeted party messages 22.8 39.4 32.1 18.2 33.8 37.1 30.7 (−1.5) (+0.5) (−8.0) (−0.6) (−6.0) (−6.6) (−5.0) Displayed election poster in window 13.6 40.2 31.0 11.6 22.4 46.1 27.7 (−8.0) (−16.0) (−17.2) (−22.0) (−17.3) (−12.9) (−16.7) Delivered leaflets 20.8 27.7 32.9 8.5 17.6 27.6 22.6 (−9.7) (−3.9) (−11.1) (−19.2) (−5.3) (−2.8) (−9.7) Attended public meeting or hustings 19.4 24.3 21.6 16.2 18.5 27.5 21.3 (−0.4) (−0.8) (−1.5) (−13.8) (−1.0) (−7.2) (−3.9) Drove voters to polling stations 3.3 4.1 2.1 1.0 0.8 5.6 2.9 (+1.0) (−0.5) (−0.6) (−3.8) (−1.2) (+0.1) (−0.7) Canvassed face to face or by phone 15.3 23.8 17.1 6.9 9.9 20.9 16.0 (−8.0) (−3.0) (−5.5) (−9.3) (−4.1) (+0.9) (−5.0) Helped run party committee room 3.8 5.2 6.4 1.4 2.8 5.7 4.2 (−3.2) (+1.2) (+0.9) (−3.4) (+0.4) (+1.3) (−0.6) Stood as candidate (councillor or MP) 0.9 2.2 2.7 1.3 2.3 0.3 1.6 (−1.3) (+1.4) (−0.9) (−3.5) (−1.0) (−0.7) (−0.7) Other 5.0 4.0 7.3 4.0 5.5 5.6 5.1 (−5.2) (−9.8) (−8.9) (−2.2) (−5.2) (−4.1) (−6.7) None 20.4 13.1 14.8 31.1 20.5 9.0 18.1 (−4.3) (+4.2) (+4.0) (+5.5) (+3.5) (+0.7) (+3.5) Campaign Activism Index—Mean 1.39 2.23 1.91 1.00 1.56 2.37 1.75 (−0.29) (−0.29) (−0.62) (−0.76) (−0.51) (−0.32) (−0.52) Number 1191 1377 1044 1112 1023 1038 6785 Activity . Conservative . Labour . Liberal Democrat . UKIP/ Brexit . Green . SNP . Total . ‘Liked’ something by party/candidate on FB 39.4 56.4 45.5 35.0 48.0 65.7 48.4 (+0.02) (−7.5) (−17.7) (+0.2) (−15.6) (−5.0) (−9.0) Tweeted/re-tweeted party messages 22.8 39.4 32.1 18.2 33.8 37.1 30.7 (−1.5) (+0.5) (−8.0) (−0.6) (−6.0) (−6.6) (−5.0) Displayed election poster in window 13.6 40.2 31.0 11.6 22.4 46.1 27.7 (−8.0) (−16.0) (−17.2) (−22.0) (−17.3) (−12.9) (−16.7) Delivered leaflets 20.8 27.7 32.9 8.5 17.6 27.6 22.6 (−9.7) (−3.9) (−11.1) (−19.2) (−5.3) (−2.8) (−9.7) Attended public meeting or hustings 19.4 24.3 21.6 16.2 18.5 27.5 21.3 (−0.4) (−0.8) (−1.5) (−13.8) (−1.0) (−7.2) (−3.9) Drove voters to polling stations 3.3 4.1 2.1 1.0 0.8 5.6 2.9 (+1.0) (−0.5) (−0.6) (−3.8) (−1.2) (+0.1) (−0.7) Canvassed face to face or by phone 15.3 23.8 17.1 6.9 9.9 20.9 16.0 (−8.0) (−3.0) (−5.5) (−9.3) (−4.1) (+0.9) (−5.0) Helped run party committee room 3.8 5.2 6.4 1.4 2.8 5.7 4.2 (−3.2) (+1.2) (+0.9) (−3.4) (+0.4) (+1.3) (−0.6) Stood as candidate (councillor or MP) 0.9 2.2 2.7 1.3 2.3 0.3 1.6 (−1.3) (+1.4) (−0.9) (−3.5) (−1.0) (−0.7) (−0.7) Other 5.0 4.0 7.3 4.0 5.5 5.6 5.1 (−5.2) (−9.8) (−8.9) (−2.2) (−5.2) (−4.1) (−6.7) None 20.4 13.1 14.8 31.1 20.5 9.0 18.1 (−4.3) (+4.2) (+4.0) (+5.5) (+3.5) (+0.7) (+3.5) Campaign Activism Index—Mean 1.39 2.23 1.91 1.00 1.56 2.37 1.75 (−0.29) (−0.29) (−0.62) (−0.76) (−0.51) (−0.32) (−0.52) Number 1191 1377 1044 1112 1023 1038 6785 Notes: All activities figures are percentages. Figures in brackets are changes since 2017. Campaign activism index is based on an additive scale that runs from 0 (no activity during the election campaign) to 9 (maximal activity during the campaign, excluding ‘other’). All relationships between party and type of campaign activity reported in this table are significant at p < 0.001. Source: Party Members Project Survey (2019). Open in new tab Table 4.4 Logistic regression model of the Conservative Party vote, December 2019 Predictor . Logistic regression coefficient . Standard Error . Significance . Odds ratio . Competence: Johnson 0.159 0.021 0.000 1.172 (0=incompetent, 10=competent) Competence: Corbyn −0.304 0.019 0.000 0.738 (0=incompetent, 10=competent) Left-right scale 0.215 0.032 0.000 1.240 (0=left, 10=right) Libertarianism-authoritarianism scale 0.118 0.027 0.000 1.125 (0=liberal, 10=authoritarian) EU referendum vote −0.487 0.130 0.000 0.615 (Ref: = Leave) Expected general economic situation in country in next 12 months (ref: will get a lot better)   Will get a lot worse −1.379 0.466 0.003 0.252   Will get a little worse −1.190 0.457 0.009 0.304   Will stay the same −0.519 0.462 0.262 0.595   Will get a little better 0.265 0.493 0.592 1.303 Age (years) 0.008 0.003 0.016 1.008 Graduate 0.021 0.117 0.858 1.021 (Ref: = non-graduate) Gender 0.223 0.106 0.036 1.250 (Ref: = female) Social Grade ABC1 0.113 0.115 0.327 1.119 (Ref: = C2DE) Constant 0.562 0.547 0.304 1.754 Dependent variable: 1 = Voted Conservative, 0 = Voted for other party. Nagelkerke R2 = 0.535, Cox-Snell R2 = 0.369, n = 3719 Predictor . Logistic regression coefficient . Standard Error . Significance . Odds ratio . Competence: Johnson 0.159 0.021 0.000 1.172 (0=incompetent, 10=competent) Competence: Corbyn −0.304 0.019 0.000 0.738 (0=incompetent, 10=competent) Left-right scale 0.215 0.032 0.000 1.240 (0=left, 10=right) Libertarianism-authoritarianism scale 0.118 0.027 0.000 1.125 (0=liberal, 10=authoritarian) EU referendum vote −0.487 0.130 0.000 0.615 (Ref: = Leave) Expected general economic situation in country in next 12 months (ref: will get a lot better)   Will get a lot worse −1.379 0.466 0.003 0.252   Will get a little worse −1.190 0.457 0.009 0.304   Will stay the same −0.519 0.462 0.262 0.595   Will get a little better 0.265 0.493 0.592 1.303 Age (years) 0.008 0.003 0.016 1.008 Graduate 0.021 0.117 0.858 1.021 (Ref: = non-graduate) Gender 0.223 0.106 0.036 1.250 (Ref: = female) Social Grade ABC1 0.113 0.115 0.327 1.119 (Ref: = C2DE) Constant 0.562 0.547 0.304 1.754 Dependent variable: 1 = Voted Conservative, 0 = Voted for other party. Nagelkerke R2 = 0.535, Cox-Snell R2 = 0.369, n = 3719 Notes: The Left-Right and Libertarian-Authoritarian variables are simple additive scales constructed from well-known batteries of survey questions first devised in the 1990s by Heath et al. (1994). Source: BES Internet panel, Waves 17 and 19 (November and December 2019). Open in new tab Table 4.4 Logistic regression model of the Conservative Party vote, December 2019 Predictor . Logistic regression coefficient . Standard Error . Significance . Odds ratio . Competence: Johnson 0.159 0.021 0.000 1.172 (0=incompetent, 10=competent) Competence: Corbyn −0.304 0.019 0.000 0.738 (0=incompetent, 10=competent) Left-right scale 0.215 0.032 0.000 1.240 (0=left, 10=right) Libertarianism-authoritarianism scale 0.118 0.027 0.000 1.125 (0=liberal, 10=authoritarian) EU referendum vote −0.487 0.130 0.000 0.615 (Ref: = Leave) Expected general economic situation in country in next 12 months (ref: will get a lot better)   Will get a lot worse −1.379 0.466 0.003 0.252   Will get a little worse −1.190 0.457 0.009 0.304   Will stay the same −0.519 0.462 0.262 0.595   Will get a little better 0.265 0.493 0.592 1.303 Age (years) 0.008 0.003 0.016 1.008 Graduate 0.021 0.117 0.858 1.021 (Ref: = non-graduate) Gender 0.223 0.106 0.036 1.250 (Ref: = female) Social Grade ABC1 0.113 0.115 0.327 1.119 (Ref: = C2DE) Constant 0.562 0.547 0.304 1.754 Dependent variable: 1 = Voted Conservative, 0 = Voted for other party. Nagelkerke R2 = 0.535, Cox-Snell R2 = 0.369, n = 3719 Predictor . Logistic regression coefficient . Standard Error . Significance . Odds ratio . Competence: Johnson 0.159 0.021 0.000 1.172 (0=incompetent, 10=competent) Competence: Corbyn −0.304 0.019 0.000 0.738 (0=incompetent, 10=competent) Left-right scale 0.215 0.032 0.000 1.240 (0=left, 10=right) Libertarianism-authoritarianism scale 0.118 0.027 0.000 1.125 (0=liberal, 10=authoritarian) EU referendum vote −0.487 0.130 0.000 0.615 (Ref: = Leave) Expected general economic situation in country in next 12 months (ref: will get a lot better)   Will get a lot worse −1.379 0.466 0.003 0.252   Will get a little worse −1.190 0.457 0.009 0.304   Will stay the same −0.519 0.462 0.262 0.595   Will get a little better 0.265 0.493 0.592 1.303 Age (years) 0.008 0.003 0.016 1.008 Graduate 0.021 0.117 0.858 1.021 (Ref: = non-graduate) Gender 0.223 0.106 0.036 1.250 (Ref: = female) Social Grade ABC1 0.113 0.115 0.327 1.119 (Ref: = C2DE) Constant 0.562 0.547 0.304 1.754 Dependent variable: 1 = Voted Conservative, 0 = Voted for other party. Nagelkerke R2 = 0.535, Cox-Snell R2 = 0.369, n = 3719 Notes: The Left-Right and Libertarian-Authoritarian variables are simple additive scales constructed from well-known batteries of survey questions first devised in the 1990s by Heath et al. (1994). Source: BES Internet panel, Waves 17 and 19 (November and December 2019). Open in new tab On the other hand, there was one vitally important aspect of the campaign in which the Tories were able to completely dominate their opponents: money. The Conservative Party managed to raise far more in large donations than Labour, or indeed, anyone else. If we look at the final six weeks of the campaign, we can see from figures reported to the Electoral Commission as pre-poll donations that the Conservatives raised considerably more than their nearest rivals (Figure 4.1). Indeed, the Conservative Party’s near £19.5 million dwarfed the amount raised by Labour (£5.1 million), the Brexit Party (£4 million) and Liberal Democrats (£1.2 million) combined. In fact, the sum raised by the Conservatives is actually greater than the sum total of donations to all parties (including the Tories) during the pre-poll period in 2017 (£18.7 million). Figure 4.1 Open in new tabDownload slide Weekly donations at the 2019 general election reported to the Electoral Commission as pre-poll donations Source:Electoral Commission (2019). Figure 4.1 Open in new tabDownload slide Weekly donations at the 2019 general election reported to the Electoral Commission as pre-poll donations Source:Electoral Commission (2019). On top of this, when the accounts for the Q4 accounting period (2019) were released by the Electoral Commission, the enthusiasm that (large) donors had for the Conservative Party at this election was clear to see. The near £37.7 million (excluding public funds) that they managed to raise was, again, far higher than all other parties put together, which stood at just under £31.1 million (Electoral Commission, 2020). While it should come as no surprise that the Conservative Party is better at raising money through large donations than other political parties, the scale of money raised shows a level of support that had been lacking in 2017. The £37.7 million raised in the final quarter of 2019 also represents an increase in the total amount raised by the Conservative Party in 2017 as a whole through donations and fundraising (£34.9 million). The party, as an organisational force, was simply much more capable of exciting its donor base than last time around. 3. Was it ‘Brexit Wot Won It’? Winning the four-party politics of England and Wales Organisation aside, it was the Conservatives’ superior reading of the electoral landscape that represented their biggest strength in this election. As Curtice (2019) pointed out, Brexit reshaped ‘the basis of party support – again’. The election began with a gamble, initially on the part of the Liberal Democrats and then, when the writing was on the wall, Labour, that Remain and Leave affiliations would wreak havoc on traditional party loyalties. The problem was that the ‘havoc’ proved somewhat unidirectional. Labour banked on the fact that when push came to shove, voters were far more Remain than they were Conservative and far more Labour than they were Leave. On both counts, they were only half right. A snap post-election poll conducted by YouGov (Chorley, 2019), found that just 65% of Remainers who had backed the Conservative Party in 2017 backed them again. On the Remain side for Labour, 79% stuck with the party. However, if we look at the Leave numbers, we see that 92% of Leavers stuck with the Conservatives, while just 52% stuck with Labour, with 33% moving directly to a prime minister promising to ‘Get Brexit Done’. Moreover, the Remain vote was quite inefficiently split among Liberal Democrats, the Greens, the nationalist parties and the Labour Party. Of the aforementioned 2017 Conservative Remainers, 22% went to the Liberal Democrats, with the number that moved to Labour in single figures; a similar post-election poll conducted by Lord Ashcroft also showed 21% of Tory Remainers moving to the Liberal Democrats, with 8% moving to Labour (Ashcroft, 2019a). Ultimately, the Conservatives retained the support of enough Tory Remainers who simply could not countenance voting for a Corbyn-led Labour Party, whereas Labour lost a larger number of Leavers, whilst failing to compensate for those losses from defecting Remainers. The Conservatives’ ability to chart a path between these distinct, but overlapping, electoral coalitions suggests that the true turning point of the 2019 election occurred weeks before campaigning began when Boris Johnson managed to conclude a Withdrawal Agreement with the European Union that would win the backing of his parliamentary party, minus those he was prepared to expel. As such, voters could be sure that the election of a Conservative majority government would mean, at last, the UK’s departure from the European Union. Isaac Levido certainly argues as much: That deal was a real game-changer; it unified a number of the fringes of our voter coalition that we needed to bring together - namely, some of those that voted for us in 2017, but that were considering voting for the Lib Dems because they were concerned about no deal. Similarly, we had a lot of our 2017 voters, who were considering voting for the Brexit Party (BBC, 2020). Johnson’s deal was not all that new or different to a range of deals negotiated, and was rejected by Theresa May and her team prior to her downfall (Boffey et al., 2019). Indeed, the most significant change from that negotiated by May was that, despite Johnson’s protestations to the contrary, it effectively agreed to locate a customs border between Britain and Northern Ireland. But, notwithstanding the furious chagrin of the DUP, politics is as much about elite cues and perceptions of success, as about the substance of policy detail. And with the European Research Group (ERG) and the right-wing press (broadly) on board, Johnson could both trumpet his deal and turn it into a snappy slogan during the campaign itself. 4. A Brexit (Party) pact? The Conservatives also benefited from the announcement during the second week of the campaign that the Brexit Party would not contest those seats that May had managed to win or hold onto in 2017. The decision itself caused much hullaballoo about the extent to which a backroom deal had been made between the Tories (who categorically denied it) and the Brexit Party, with Nigel Farage even suggesting that allies of Johnson had offered the party eight peerages in return for standing down Brexit Party candidates (Hughes and Parker, 2019). Richard Tice, Brexit Party Chairman and widely seen as Farage’s deputy, suggested that back channels were indeed open prior to their decision (BBC, 2020): I wasn’t offered a peerage, but it did start with me because they started a sort of decapitation strategy with me at the top as chairman by suggesting that I take a safe seat … and indeed the seat of Rutland was mentioned … that was part of their ruthless behaviour. Deep down, there are levels of – bordering on corruption with a very small ‘c’ [sic]. However, analysing the exact effects of the Brexit Party’s self-denying ordinance is tricky. Though it certainly meant that the Leave vote did not split as much as the Remain vote, the precise electoral consequences are less clear. Although some are left in little doubt that the Brexit Party decision represents the single most important moment of the campaign—that ‘Farage won it for Johnson’ (Loucaides, 2019)—the numbers tell a slightly different story. YouGov’s Curtis (2019), for example, pointed to the fact that the Brexit Party was only standing down in Tory-held seats. The state of the polls prior to the announcement, it was argued, implied a 4% swing from the Labour Party to the Tories, which meant that in seats the Conservatives hoped to gain—and many of which they ended up winning—the Brexit Party would still be standing. In fact, of the 212 seats that the Conservative Party held from 2017 with Labour in second place, 86 had a Leave vote above 60% and in only five of those seats was the Conservative majority less than 10,000 (Connolly, 2019). It might well have been the case, then, that the Brexit Party announcement did very little to affect the overall outcome, short of a few seats here or there. Indeed, the effect the Brexit Party had in 2019 may well have been to prevent an even larger Tory majority. Of the 189 seats that the Labour Party managed to retain, there were 38 seats in which the combined total of the Brexit Party and the Conservatives was higher than Labour’s (Connolly, 2019). Of course, not every Brexit Party voter was a Conservative in disguise—as Ford and Goodwin (2014) were quick to point out when it came to UKIP support—but there were many Labour seats such as those of Ian Lavery, Yvette Cooper and former leader Ed Miliband that were won on vastly diminished majorities and were likely saved by the Brexit Party splitting the Leave vote. 5. Taking the ‘red wall’ Many of the abovementioned seats made up the so-called ‘red wall’ (Onward, 2019), the crumbling of which on 12 December made for the main drama of the broadcasters’ election night coverage. This was arguably an earthquake waiting to happen as culture and national identity cut new—and for the Conservatives profitable—paths in politics. That the election was about Brexit, there can be little doubt; but the underlying issues and trends for which Brexit acted as a filter had been bubbling under for some time. The Conservatives, in successfully pitching 2019 as a Brexit election (in a way that they failed to do in 2017) and then subsequently nullifying the Brexit Party challenge, managed to traverse this new landscape, although how long that landscape will remain as it stands now is an open question. It may well do. Beyond the traditional left–right divide that underlies party competition and voting behaviour in Britain, it has long since become commonplace to recognise the impact of a second line of conflict between ‘materialists’ and ‘postmaterialists’ (Inglehart, 1997), or between ‘Green-Alternative-Liberal’ and ‘Traditional-Authoritarian-Nationalist’ (GALTAN) voters (Hooghe et al., 2002). Ford and Jennings (2020) have suggested that these new faultlines have evolved due to four key sociodemographic developments: first, an educational expansion which has (1) driven the widening of a graduate class and (2) led to the decline of school leavers often employed in manual labour; second, an increase in mass migration which has led to an enlargement of ethnic diversity amongst electorates; third, the ageing of society, which has contributed to deepening generational divides in terms of both education and political values; and finally, a growing divide between prospering cities and declining towns and rural areas. These socio-demographic changes have helped boost the rise of the populist radical right—represented in the UK by the UKIP and then the Brexit Party. That radical right was able to pull together a coalition of voters from the left and (to a greater extent) the right disillusioned with ‘progressive’ politics (broadly defined), thereby exploiting the divide ‘between nationalists and cosmopolitans, liberals and conservatives, and cultural traditionalists and multiculturalists’ (Ford and Goodwin, 2017, p. 28; see also Norris and Inglehart, 2019). However, the Conservative Party moved quickly under Boris Johnson (and even to some extent under Theresa May) to steal their thunder—a move made easier, some argue, by Labour: according to one of its defeated MPs (Engel, 2019): Left behind? They just wanted to be left alone! They looked at their own values, their own sense of right and wrong, and they preferred it to the confusion and chaos of what the political elites in London seemed to be obsessed with: antisemitism, transgenderism and net-zero carbon admissions. It’s not that they disagreed with the parties’ positions on them, it’s that they have absolutely no bearing on their daily lives. The Conservative appeal amongst these voters is born out in polling. Over the past few elections, the Conservatives had been closing the gap on Labour amongst DE voters (essentially, lower working class and those not in work); in 2015, the Conservatives only won 29% of these voters to Labour’s 37%, whilst in 2017 the figures stood at 41% and 44%, respectively. By 2019, the Conservative vote amongst DE electors had risen to 47% with Labour’s falling to 34% (Chorley, 2019). Seats that the Conservatives gained from Labour in either 2017 or 2019 (Table 4.2) also had higher than average levels of deprivation, while house prices and wages were substantially below the national average and the seats contained very low numbers of graduates. Constituencies with these kinds of demographics 25 years ago, would have reasonably been described as the Labour ‘heartlands’.3 Table 4.2 Safe seats and seats that switched hands between the Conservatives and Labour at the 2017 and 2019 general elections . Seats consistently held by Labour . Labour gained from Con in GE2017 or GE2019 . Con gained from Labour in GE2017 or GE2019 . Seats consistently held by Conservatives . Deprivation Much higher than average levels of deprivation Slightly lower than average levels of deprivation Higher than average levels of deprivation Much lower than average levels of deprivation Housing Low numbers own outright or own with a mortgage High numbers of private renters. High levels of social housing. Average numbers own outright or own with mortgage High numbers of private renters. Low levels of social housing. Average numbers own outright or own with mortgage Low numbers of private renters. High levels of social housing. High numbers own outright or own with a mortgage. Low numbers of private renters. Very low levels of social housing. House prices Below average Average (though wide variation) Substantially below average Above average Wages Slightly below average Average Substantially below average Above average Age High working-age population Low numbers of pensioners High working-age population Quite low numbers of pensioners Quite low working-age population Quite high numbers of pensioners Low working-age population High numbers of pensioners Ethnicity Substantially above-average BAME population Above-average BAME population Below average BAME population Below average BAME population Education Average number of graduates Very high numbers of graduates Very low numbers of graduates Quite high numbers of graduates . Seats consistently held by Labour . Labour gained from Con in GE2017 or GE2019 . Con gained from Labour in GE2017 or GE2019 . Seats consistently held by Conservatives . Deprivation Much higher than average levels of deprivation Slightly lower than average levels of deprivation Higher than average levels of deprivation Much lower than average levels of deprivation Housing Low numbers own outright or own with a mortgage High numbers of private renters. High levels of social housing. Average numbers own outright or own with mortgage High numbers of private renters. Low levels of social housing. Average numbers own outright or own with mortgage Low numbers of private renters. High levels of social housing. High numbers own outright or own with a mortgage. Low numbers of private renters. Very low levels of social housing. House prices Below average Average (though wide variation) Substantially below average Above average Wages Slightly below average Average Substantially below average Above average Age High working-age population Low numbers of pensioners High working-age population Quite low numbers of pensioners Quite low working-age population Quite high numbers of pensioners Low working-age population High numbers of pensioners Ethnicity Substantially above-average BAME population Above-average BAME population Below average BAME population Below average BAME population Education Average number of graduates Very high numbers of graduates Very low numbers of graduates Quite high numbers of graduates Source:Cooper and Cooper (2020). Open in new tab Table 4.2 Safe seats and seats that switched hands between the Conservatives and Labour at the 2017 and 2019 general elections . Seats consistently held by Labour . Labour gained from Con in GE2017 or GE2019 . Con gained from Labour in GE2017 or GE2019 . Seats consistently held by Conservatives . Deprivation Much higher than average levels of deprivation Slightly lower than average levels of deprivation Higher than average levels of deprivation Much lower than average levels of deprivation Housing Low numbers own outright or own with a mortgage High numbers of private renters. High levels of social housing. Average numbers own outright or own with mortgage High numbers of private renters. Low levels of social housing. Average numbers own outright or own with mortgage Low numbers of private renters. High levels of social housing. High numbers own outright or own with a mortgage. Low numbers of private renters. Very low levels of social housing. House prices Below average Average (though wide variation) Substantially below average Above average Wages Slightly below average Average Substantially below average Above average Age High working-age population Low numbers of pensioners High working-age population Quite low numbers of pensioners Quite low working-age population Quite high numbers of pensioners Low working-age population High numbers of pensioners Ethnicity Substantially above-average BAME population Above-average BAME population Below average BAME population Below average BAME population Education Average number of graduates Very high numbers of graduates Very low numbers of graduates Quite high numbers of graduates . Seats consistently held by Labour . Labour gained from Con in GE2017 or GE2019 . Con gained from Labour in GE2017 or GE2019 . Seats consistently held by Conservatives . Deprivation Much higher than average levels of deprivation Slightly lower than average levels of deprivation Higher than average levels of deprivation Much lower than average levels of deprivation Housing Low numbers own outright or own with a mortgage High numbers of private renters. High levels of social housing. Average numbers own outright or own with mortgage High numbers of private renters. Low levels of social housing. Average numbers own outright or own with mortgage Low numbers of private renters. High levels of social housing. High numbers own outright or own with a mortgage. Low numbers of private renters. Very low levels of social housing. House prices Below average Average (though wide variation) Substantially below average Above average Wages Slightly below average Average Substantially below average Above average Age High working-age population Low numbers of pensioners High working-age population Quite low numbers of pensioners Quite low working-age population Quite high numbers of pensioners Low working-age population High numbers of pensioners Ethnicity Substantially above-average BAME population Above-average BAME population Below average BAME population Below average BAME population Education Average number of graduates Very high numbers of graduates Very low numbers of graduates Quite high numbers of graduates Source:Cooper and Cooper (2020). Open in new tab This is also representative of a shift in electoral geography (Ford and Jennings, 2020, p. 1713). Here we see young, ethnically diverse, graduate populations with socially liberal views agglomerating around cities and voting for leftist and/or socially liberal parties like Labour, the Greens and the Liberal Democrats, whereas the older, more ethnically homogeneous, non-graduate voters holding populist and socially conservative values opting for the Brexit Party and UKIP but, crucially, in 2019 the Conservatives (Ford and Jennings, 2020; see also Jennings and Stoker, 2016, 2019). This benefited the Conservatives both sociologically and institutionally. Whilst in the early 2000s, New Labour benefited from a First Past the Post electoral system that worked very well for them in terms of vote distribution, the boot is now firmly on the other foot. Conservative voters are now far more efficiently distributed across the UK, which minimises wasted ‘surplus’ votes in safe seats and ‘redundant votes’ in seats the Tories lose (Smith, 2020; see also Electoral Reform Society, 2020). The ‘bias’ in the electoral geography that favoured Labour in the Blair–Brown years had swung decisively towards the Tories. The Conservatives managed to claim large numbers of seats across the ‘red wall’, largely in the North and Midlands, due in large part to having a Brexit pitch that appealed—and a leader that seemed far less unappealing than Labour’s; of the 54 seats the Tories took from Labour in 2019, 50 of them had voted Leave in 2016 (Goodwin, 2019). After his brief moment of catch-up in the leader popularity stakes during the 2017 general election campaign, Corbyn had regularly ranked lower than his Conservative counterpart thereafter (YouGov, 2019b). British Election Study data gathered during the 2019 campaign showed that he was less well-liked than Boris Johnson in almost all demographic groups, excepting under 25-year-olds, Londoners and Scots. This means, among other things, that Corbyn was less well-regarded than his direct rival for the premiership among working-class voters and those northern and midland regions across which the ‘red wall’ stretched (Table 4.3). Given the growing body of evidence that confirms the significance of leader effects on voting behaviour, this was highly significant (Garcia, 2012; Curtice and Lisi, 2015). Table 4.3 Who was the best-liked major party leader, 2019? . Johnson . Corbyn . Swinson . Men 4.33 2.55 2.86 Women 4.01 2.79 2.91 ABC1 3.94 2.79 3.06 C2DE 4.44 2.55 2.58 Graduate 3.40 3.11 3.35 Non-graduates 4.78 2.34 2.53 18–25 2.51 4.24 3.43 26–35 3.39 3.14 3.37 36–45 3.88 2.57 3.09 46–55 4.42 2.24 2.58 56–65 5.00 1.77 2.29 Over 65 5.94 1.29 2.26 North East 4.21 2.83 2.75 North West 3.97 3.11 2.64 Yorks & Humber 4.09 2.81 2.73 East Midlands 4.71 2.37 2.80 West Midlands 4.75 2.45 2.75 East of England 4.83 2.22 2.83 London 3.34 3.52 3.35 South East 4.65 2.08 2.96 South West 4.68 2.34 2.92 Wales 4.08 2.88 2.66 Scotland 2.63 2.86 2.94 Total 4.17 (n = 30,227) 2.67 (n = 29,947) 2.88 (n = 26,818) . Johnson . Corbyn . Swinson . Men 4.33 2.55 2.86 Women 4.01 2.79 2.91 ABC1 3.94 2.79 3.06 C2DE 4.44 2.55 2.58 Graduate 3.40 3.11 3.35 Non-graduates 4.78 2.34 2.53 18–25 2.51 4.24 3.43 26–35 3.39 3.14 3.37 36–45 3.88 2.57 3.09 46–55 4.42 2.24 2.58 56–65 5.00 1.77 2.29 Over 65 5.94 1.29 2.26 North East 4.21 2.83 2.75 North West 3.97 3.11 2.64 Yorks & Humber 4.09 2.81 2.73 East Midlands 4.71 2.37 2.80 West Midlands 4.75 2.45 2.75 East of England 4.83 2.22 2.83 London 3.34 3.52 3.35 South East 4.65 2.08 2.96 South West 4.68 2.34 2.92 Wales 4.08 2.88 2.66 Scotland 2.63 2.86 2.94 Total 4.17 (n = 30,227) 2.67 (n = 29,947) 2.88 (n = 26,818) Notes: All figures are mean scores for a scale on which respondents were asked to rate leaders from 0 (strongly dislike) to 10 (strongly like). The italicized rows are the only categories in which Corbyn has a higher mean score than Johnson. Source: BES Internet panel, Wave 19 (December 2019). Open in new tab Table 4.3 Who was the best-liked major party leader, 2019? . Johnson . Corbyn . Swinson . Men 4.33 2.55 2.86 Women 4.01 2.79 2.91 ABC1 3.94 2.79 3.06 C2DE 4.44 2.55 2.58 Graduate 3.40 3.11 3.35 Non-graduates 4.78 2.34 2.53 18–25 2.51 4.24 3.43 26–35 3.39 3.14 3.37 36–45 3.88 2.57 3.09 46–55 4.42 2.24 2.58 56–65 5.00 1.77 2.29 Over 65 5.94 1.29 2.26 North East 4.21 2.83 2.75 North West 3.97 3.11 2.64 Yorks & Humber 4.09 2.81 2.73 East Midlands 4.71 2.37 2.80 West Midlands 4.75 2.45 2.75 East of England 4.83 2.22 2.83 London 3.34 3.52 3.35 South East 4.65 2.08 2.96 South West 4.68 2.34 2.92 Wales 4.08 2.88 2.66 Scotland 2.63 2.86 2.94 Total 4.17 (n = 30,227) 2.67 (n = 29,947) 2.88 (n = 26,818) . Johnson . Corbyn . Swinson . Men 4.33 2.55 2.86 Women 4.01 2.79 2.91 ABC1 3.94 2.79 3.06 C2DE 4.44 2.55 2.58 Graduate 3.40 3.11 3.35 Non-graduates 4.78 2.34 2.53 18–25 2.51 4.24 3.43 26–35 3.39 3.14 3.37 36–45 3.88 2.57 3.09 46–55 4.42 2.24 2.58 56–65 5.00 1.77 2.29 Over 65 5.94 1.29 2.26 North East 4.21 2.83 2.75 North West 3.97 3.11 2.64 Yorks & Humber 4.09 2.81 2.73 East Midlands 4.71 2.37 2.80 West Midlands 4.75 2.45 2.75 East of England 4.83 2.22 2.83 London 3.34 3.52 3.35 South East 4.65 2.08 2.96 South West 4.68 2.34 2.92 Wales 4.08 2.88 2.66 Scotland 2.63 2.86 2.94 Total 4.17 (n = 30,227) 2.67 (n = 29,947) 2.88 (n = 26,818) Notes: All figures are mean scores for a scale on which respondents were asked to rate leaders from 0 (strongly dislike) to 10 (strongly like). The italicized rows are the only categories in which Corbyn has a higher mean score than Johnson. Source: BES Internet panel, Wave 19 (December 2019). Open in new tab However, Tory success also owed much to longer-term institutional and sociological trends. These trends underpinned the Brexit vote, and may even point the way towards a future realignment in British politics. In Table 4.4, we bring together several of the various attitudinal and demographic factors reckoned to influence voting behaviour in a logistic regression model of voting for the Conservative Party in 2019. It confirms much of what we would expect. Using measures of perceived competence of the two direct rivals for the keys of Number 10 as indicators of leadership evaluations, we see that the higher that respondents rated Johnson, the greater the probability of them having voted for the Conservatives, whereas the opposite is true for Corbyn. We also see that more right-wing and socially authoritarian respondents were more likely to have voted Conservative. Compared to those who voted Leave in the 2016 EU referendum, those who voted Remain were significantly less likely to have voted Conservative in 2019. We can also take into account people’s evaluations of the country’s economic performance, given the widespread evidence that, however much ‘values voting’ has grown in importance recently, these kinds of ‘valence effects’ continue to impact voting behaviour in contemporary Britain (Clarke et al., 2004, 2009; Whiteley et al., 2013; Green and Jennings, 2017). Here, we find that, compared to those who expected the economy to get a lot better over the next year, those who expected it to get either a lot or a little worse were significantly less likely to have voted Tory in 2019. Finally, of the four major demographic controls entered into the model, only one of them had a statistically significant effect on voting for Conservative candidates. Whether people were male or female, graduate or non-graduate or manual or non-manual employees made no difference. What mattered was age: the older a voter was, the more likely they were to have opted for the Tories. 6. Must the Conservatives win? This is an election result that some see as tantamount to Labour’s last rites. But it is worth remembering that politics can change very quickly. In the wake of a crushing defeat in the 1959 general election, political scientists Mark Abrams and Richard Rose (1960) published their now classic reading of the runes Must Labour Lose?—only for Wilson to sweep to victory in 1964. In 1994, Anthony Heath and colleagues wondered if the 1992 election constituted Labour’s Last Chance? Thirteen years of New Labour dominance followed not long thereafter. And in 2005 Geoffrey Wheatcroft published The Strange Death of Tory England; a few months later, David Cameron became leader of the Conservatives and the rest, as they say, is history. Just as 2017 was the election in which everything went wrong for the Tories and 2019 was the election where everything went right, it might not do so next time around. They have a potentially fragile and fissiparous coalition of support, from traditional Tory shires to the previously solidly Labour North East, and of Leavers and Remainers. Table 4.2 shows the stark challenge that keeping these voters together will present: the column representing seats gained from Labour in 2017 and 2019 and the one showing consistently held Conservative seats present very real contrasts in terms of policy and demographics. In short, the Conservatives have a lot of plates to spin—and, likely, five years to keep them spinning. Boris Johnson acknowledged these challenges on the steps of Downing Street as dawn broke on 13 December when he observed that many of the votes that re-elected him had been loaned to his party and, as such, should not be taken for granted (Johnson, 2019). Although it might be easier for those who have at last broken their habit of voting Labour to vote Conservative next time too, it might just as well prove to be the case that, for these voters at least, the Tories end up being a one-time fling—‘a temporary and transactional swing to the Conservatives, rather than a conversion to conservatism’ (Ashcroft, 2019b). Moreover, many of these people voted for the Conservatives on the promise that they would ‘Get Brexit Done’ and for an ‘oven ready’ Brexit deal, which is actually, at best, parboiled. Moreover, the government, we should not forget, were facing a divided Labour opposition, with an unprecedently unpopular leader and a confusing pitch on the major issue of the day. Come the next election, the Tories might not be so lucky. That said, the Conservative Party still has more reasons to be cheerful than the Labour Party. The election showed that it had grasped many of the more fundamental challenges that the UK’s changing electoral landscape presents. Whether it can cope with the challenges presented by the government remains, however, to be seen. Acknowledgement This work was supported by the Economic and Social Research Council (grant number ES/M007537/1). Tim and Paul would also like to thank their colleague on the Project, Dr Monica Poletti. Footnotes 2 Remarks made to a journalist by Boris Johnson when asked why it had been such a dull campaign (reported by McTague, 2019). 3 We also see that those seats Labour gained from the Conservatives in either 2017 or 2019 are also indicative of this shifting electoral landscape in that they contain a high number of graduates, lower levels of deprivation and a high working-age population. 1 Lyrics by Alicia Keys (2007). References Abrams M. , Rose R. ( 1960 ) Must Labour Lose? Harmondsworth , Penguin . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ashcroft L. ( 2019 a) ‘Lord Ashcroft: How Britain Voted and Why. My 2019 Post-vote Poll’, accessed at https://www.conservativehome.com/platform/2019/12/lord-ashcroft-how-britain-voted-and-why-my-2019-post-vote-poll.html on 28 February 2020. Ashcroft L. ( 2019 b) ‘Lord Ashcroft: Would Victories for Johnson and Trump Herald the Triumph of Conservatism’, accessed at https://www.conservativehome.com/platform/2019/12/lord-ashcroft-will-victories-for-johnson-and-trump-herald-the-triumph-of-conservatism.html on 28 February 2020. Audickas L. , Dempsey N. , Loft P. ( 2019 ) ‘Membership of UK Political Parties’, accessed at https://researchbriefings.parliament.uk/ResearchBriefing/Summary/SN05125 on 28 February 2020. Bale T. , Webb P. , Poletti M. ( 2019 ) Footsoldiers: Political Party Membership in the 21st Century , London , Routledge . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC BBC ( 2020 , February 9) ‘The Inside Story of Election 19’ accessed at https://www.bbc.co.uk/sounds/play/m000f5rr on 28 February 2020. BBC ( 2019 , December 11) ‘General Election Poll Tracker: How Do the Parties Compare?’ accessed at https://www.bbc.co.uk/news/uk-politics-49798197 on 28 February 2020. Boffey D. , Henley J. , O’Carroll L. , Mason R. ( 2019 ) ‘Boris Johnson ‘on brink of Brexit deal’ after Border Concessions’, accessed at https://www.theguardian.com/politics/2019/oct/15/boris-johnson-close-to-brexit-deal-after-border-concessions on 6 March 2020. British Election Study ( 2019 ) ‘Press Release: Most Volatile British Electorate in Modern Times’, accessed at https://www.britishelectionstudy.com/bes-resources/press-release-most-volatile-british-electorate-in-modern-times/#.XlzzQRP7Rp8 on 28 February 2020. Chorley M. ( 2019 , December 17) ‘General Election Results: Working Class Switched to Tories’, accessed at https://www.thetimes.co.uk/edition/news/general-election-results-working-class-switched-to-tories-kfwptc6cr on 28 February 2020. Clarke H. , Sanders D. , Stewart M. , Whiteley P. ( 2004 ) Political Choice in Britain , Oxford , Oxford University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Clarke H. , Sanders D. , Stewart M. , Whiteley P. ( 2009 ) Performance Politics and the British Voter , Cambridge , Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Commons People ( 2019 ) ‘Weak Leadership. Inexperienced Strategists. Poor Policy – A Chastening Week For The Election Campaigns’, accessed at https://audioboom.com/posts/7435766-weak-leadership-inexperienced-strategists-poor-policy-a-chastening-week-for-the-election-camp on 28 February 2020. Connolly J. ( 2019 , December 19) ‘What Impact Did the Brexit Party Have on the General Election Result?’, accessed at https://blogs.spectator.co.uk/article/what-impact-did-the-brexit-party-have-on-the-general-election-result- on 28 February 2020. Cooper C. , Cooper L. ( 2020 ) ‘The Devastating Defeat: Why Labour Lost and How It Can Win Again’, accessed at https://www.europeforthemany.com/tdd-web.pdf on 28 February 2020. Curtice J. ( 2019 ) ‘Brexit Reshapes The Basis of Party Support – Again’, accessed at https://whatukthinks.org/eu/brexit-reshapes-the-basis-of-party-support-again/ on 28 February 2020. Curtice J. , Lisi M. ( 2015 ) ‘The Impact of Leaders in Parliamentary and Presidential Regimes’. In Costa Lobo M. and Curtice J. (eds.) Personality Politics: The Role of Leader Evaluations in Democratic Elections , Oxford , Oxford University Press , pp. 63 – 86 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Curtis C. ( 2019 ) ‘Farage’s Election Stand-down Will Make Little Difference’, accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/11/11/unless-polls-move-todays-announcement-farage-will- on 28 February 2020. Donaldson K. , Ross T. ( 2019 , December 13) ‘The Ruthless Vote Machine Behind Boris Johnson’s Big Win’, accessed at https://www.bloomberg.com/news/articles/2019-12-13/rampant-johnson-gets-roars-and-hugs-while-opponent-faces-despair on 28 February 2020. Doward J. ( 2019 ) ‘Voters ‘used as lab rats’ in political Facebook adverts, warn analysts’, accessed at https://www.britishelectionstudy.com/bes-resources/press-release-most-volatile-british-electorate-in-modern-times/#.XlzzQRP7Rp8 on 28 February 2020. Electoral Commission ( 2019 ) ‘Donations and Loans Received by Political Parties and Non-party Campaigners in the UK’, accessed at https://www.electoralcommission.org.uk/media-centre/donations-and-loans-received-political-parties-and-non-party-campaigners-uk-fourth-weekly-pre-poll on 13 May 2020. Electoral Commission ( 2020 ) ‘Record year and quarter for political party donations and loans in Great Britain’, accessed at https://www.electoralcommission.org.uk/media-centre/record-year-and-quarter-political-party-donations-and-loans-great-britain on 28 February 2020. Electoral Reform Society ( 2020 ) Voters Left Voiceless: The 2019 General Election , London , Electoral Reform Society . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Engel N. ( 2019 , December 16) ‘The Working Classes are Sick of Being Told What to Think’, accessed at https://www.thetimes.co.uk/edition/comment/the-working-classes-are-sick-of-being-told-what-to-think-9cz8l95xt on 28 February 2020. Ford R. , Goodwin M. ( 2014 ) Revolt on the Right: Understanding Support for the Radical Right in Britain , Abingdon , Routledge . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Ford R. , Goodwin M. ( 2017 ) ‘ Britain after Brexit ’, Journal of Democracy , 28 , 17 – 30 . Google Scholar Crossref Search ADS WorldCat Ford R. , Jennings W. ( 2020 ) ‘ The Changing Cleavage Politics of Western Europe ’, Annual Review of Political Science , 23 , 295 – 1720 . Google Scholar Crossref Search ADS WorldCat Garcia D. ( 2012 ) ‘ Party and Leader Effects in Parliamentary Elections: Towards a Reassessment ’, Politics , 32 , 175 – 185 . Google Scholar Crossref Search ADS WorldCat Goodwin M. ( 2019 , December 24), ‘Nine Lessons from the Election: Boris was Lucky – but He Also Played His Hand Right’, accessed at https://blogs.spectator.co.uk/article/nine-lessons-from-the-election-boris-was-lucky—but-he-also-played-his-hand-right on 28 February 2020. Green J. , Jennings W. ( 2017 ) The Politics of Competence: Parties, Public Opinion and Voters , Cambridge , Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Hooghe L. , Marks G. , Wilson C. J. ( 2002 ) ‘ Does Left/Right Structure Party Positions on European Integration? ’, Comparative Political Studies , 35 , 965 – 989 . Google Scholar Crossref Search ADS WorldCat Heath A. , Evans G. , Martin J. ( 1994 ) ‘ The Measurement of Core Beliefs and Values: The Development of Balances Socialist/Laissez Faire and Libertarian/Authoritarian Scales ’, British Journal of Political Science , 24 , 115 – 132 . Google Scholar Crossref Search ADS WorldCat Hughes L. , Parker G. ( 2019 ) ‘Farage Claims Tories Dangled Titles for Brexit Party Support’ accessed at https://www.ft.com/content/2c8bb0a8-06fc-11ea-a984-fbbacad9e7dd on 28 February 2020. Ibbetson C. ( 2020 ) ‘It’s a Draw, But Our Poll Shows Corbyn Came Across As More in Touch’ accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/11/21/corbyn-draws-boris on 28 February 2020. Inglehart R. ( 1997 ) Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies , Princeton , Princeton University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jennings W. , Stoker G. ( 2016 ) ‘ The Bifurcation of Politics: Two Englands ’, The Political Quarterly , 87 , 372 – 382 . Google Scholar Crossref Search ADS WorldCat Jennings W. , Stoker G. ( 2019 ) ‘ The Divergent Dynamics of Cities and Towns: Geographic Polarisation and Brexit ’, The Political Quarterly , 90 , 155 – 166 . Google Scholar Crossref Search ADS WorldCat Johnson B. ( 2019 ) ‘Boris Johnson: Perhaps My Campaign Was ‘Clunking’. But Sometimes, Clunking Is What You Need’, accessed at https://www.spectator.co.uk/article/boris-johnson-perhaps-my-campaign-was-clunking-but-sometimes-clunking-is-what-you-need on 28 February 2020. Keys A. ( 2007 ) Lesson Learned , New York , Universal Publishing Group . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Langford E. ( 2019 ) ‘AT-A-GLANCE: The Key Moments of the 2019 Election Campaign’, accessed at https://www.politicshome.com/news/uk/politics/news/108505/glance-key-moments-2019-election-campaign on 6 March 2020. Loucaides D. ( 2019 ) ‘The Brexit Party Folded, But Make No Mistake: Farage Won It for Johnson’, accessed at https://www.theguardian.com/commentisfree/2019/dec/13/brexit-party-nigel-farage-boris-johnson-labour-leavers on 28 February 2020. McTague T. ( 2019 ) ‘It’s Boris Johnson’s Britain Now’, accessed at https://www.theatlantic.com/international/archive/2019/12/boris-johnson-britain-uk-election/603466/ on 28 February 2020. Norris P. , Inglehart R. ( 2019 ) Cultural Backlash: Trump, Brexit and Authoritarian Populism , Cambridge , Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC O’Donoghue D. ( 2019 ) ‘Nine key moments from the 2019 general election campaign’, accessed at https://www.pressandjournal.co.uk/fp/news/politics/uk-politics/1908847/nine-key-moments-from-the-2019-general-election-campaign/ on 6 March 2020. Onward ( 2019 ) ‘Realignment: How Did the Tories, Win, How Did the Red Wall Fall, and What Does It Mean for the Future?’ accessed at https://www.ukonward.com/wp-content/uploads/2019/12/Realignment-Onward-deck-1.pdf on 14 March 2020. Partington R. ( 2019 ) ‘Conservative Party Manifesto: What It Says and What It Means’, accessed at https://www.theguardian.com/politics/2019/nov/24/conservative-manifesto-the-key-points-policies-boris-johnson on 6 March 2020. Payne S. ( 2019 ) ‘UK Set for ‘Most Unpredictable Election’ in Decades, Study Finds’, accessed at https://www.ft.com/content/3c889084-e9bf-11e9-85f4-d00e5018f061 on 28 February 2020. Pienaar J. ( 2019 ) ‘General Election 2019: A Campaign Unlike Any Other?’, accessed at https://www.bbc.co.uk/news/election-2019-50432050 on 6 March 2020. Schofield K. ( 2019 ) ‘EXCL Labour Membership Dips Below Half a Million as Tens of Thousands Leave Party’, accessed at https://www.politicshome.com/news/uk/political-parties/labour-party/news/105507/excl-labour-membership-dips-below-half-million on 28 February 2020. Smith M. ( 2019 ) ‘YouGov Snap Poll Finds Another Close Result as Viewers Split on Who Won BBC Leaders Debate’, accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/12/06/yougov-due-release-bbc-debate-snap-poll-930pm, on 28 February 2020. Smith T. ( 2020 ) ‘Why Did the Conservatives’ Large Lead in Vote Shares Produce Only an 80-Seat Majority?’, accessed at https://www.democraticaudit.com/2020/01/22/why-did-the-conservatives-large-lead-in-vote-shares-produce-only-an-80-seat-majority/ on 16 March 2020. Who Targets Me ( 2019 ) ‘Boris Johnson is Marmite (and the Tory Facebook ads know it)’, accessed at https://medium.com/@WhoTargetsMe/boris-johnson-is-marmite-and-the-tory-facebook-ads-know-it-b15de3d1f174 on 28 February 2020. YouGov ( 2019 a) ‘If You Had To, Which of the Main Party Leaders Would You Spend Christmas Day With?’, accessed at https://yougov.co.uk/topics/politics/survey-results/daily/2019/12/09/aa68a/1 on 28 February 2020. YouGov ( 2019 b) ‘Political Tracker Polls, Best Prime Minister’, accessed at https://d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/xqkhf8r3eb/YG%20Trackers%20-%20Best%20Prime%20Minister.pdf on 14 March 2020. Wheatcroft G. ( 2005 ) The Strange Death of Tory England , London , Allen Lane . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Whiteley P. , Clarke H. , Sanders D. , Stewart M. ( 2013 ) Affluence, Austerity and Electoral Change in Britain , Cambridge , Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Goes,, Eunice

doi: 10.1093/pa/gsaa023pmid: N/A

When the Prime Minister Boris Johnson called an early election, few Labour candidates were expecting a victory or even a repeat of the better than expected results obtained in the 2017 general election. Equally few, however, were prepared for the party to suffer such a comprehensively devastating defeat. Yet Labour obtained its worst election result since 1935, attracting less than one-third of the votes cast and electing only 202 MPs. Crucially, Labour lost 61 seats (and won one), many in traditional heartlands in northern England and in the Midlands, but the party also lost a sizeable number of seats in Scotland and Wales. Unsurprisingly, as soon as the electoral results arrived, the blame game began. For many observers and Labour candidates, the leader of the party, Jeremy Corbyn, was the main cause of Labour’s defeat. After all, he agreed to hold an election at the worst possible time, led a disorganised campaign and was extremely unpopular with voters. ‘Corbynistas’, on the other hand, blamed Labour’s pro-European faction, which forced the party to adopt a new policy on Brexit which, in their view explains the loss of dozens of seats in the North of England and Midlands. Doubtless, these factors contributed to Labour’s devastating defeat; however, other long-term factors, such as a decade of turbulent politics, structural changes to politics and the emergence of new cleavages contributed to Labour’s fourth consecutive defeat since 2010. This chapter will explain the short-term and long-term factors that led to Labour’s defeat but it will start by contextualising Labour’s state of readiness for the election and analysing Labour’s 2019 electoral campaign. 1. From hubris to chaos Labour’s better-than-expected results at the 2017 general election offered some breathing space to its leader. The planned plots to oust Corbyn were put on hold and so were the ‘Save Jeremy’ campaign devised by his team (Shipman, 2018, p. 448). Moreover, as the new intake of MPs was supportive of his agenda, Corbyn was able to tighten his control over the party. Over time, the Left became the dominant voice of the party’s National Executive Committee and main decision-making structures of the party. When the party met in Brighton for its annual conference in September 2017 the mood was, to use the words of Corbyn’s former adviser Andrew Fisher, ‘hubristic’ (Evening Standard, 2020). Reflecting that mood, Corbyn told delegates that Labour had become ‘a government-in-waiting’ that was ‘setting the agenda and winning the arguments for a new common sense about the direction our country should take’ (Corbyn, 2019c). The Shadow Chancellor, John McDonnell, responded to Corbyn’s call by turbo-charging the party’s policy-making engine and by focussing on improving the party’s economic credibility. He led workshops on alternative forms of ownerships, listened to economists on productivity and launched several ‘tea-charm offensives’ to the City of London and to Britain’s boardrooms in the hope of reassuring them that business was ‘better off with Labour’. But McDonnell’s efforts to turn Labour into a credible government in waiting were undermined by two major crises which paralysed the party for the best part of the two years that preceded the 2019 election. The first one was Corbyn’s failure to tackle decisively anti-semitic behaviour within the party; the second was his inability to develop a coherent approach to Brexit, the most important question facing the country at the time. The crisis of anti-Semitism emerged soon after Corbyn’s election as Labour leader in 2015 but it had acquired dramatic proportions by the summer of 2018. The party tried slowly, albeit with little resolve, to address the issue by adopting new internal rules to discipline members found guilty of anti-semitic behaviour. But those efforts backfired when, in March 2018, Jeremy Corbyn expressed his opposition to the defacing of an anti-semitic mural. The backbencher Luciana Berger questioned the leader’s office about Corbyn’s apparent endorsement of the mural. Corbyn’s office backtracked immediately and apologised but this reaction was deemed inadequate by Berger and other MPs. Even McDonnell and Momentum condemned Corbyn’s office for Labour’s failure to deal with anti-semitic behaviour in a decisive manner (Kogan, 2019, pp. 349–351). The mural incident was not the final one in the party’s anti-Semitism crisis. In the summer of 2018, Labour’s internal debate about the adoption of the International Holocaust Remembrance Alliance’s (IHRA) definition of anti-semitism into the party’s code showed that Corbyn had not yet tackled the problem. The NEC’s proposed amendments to the definition were challenged by several frontbenchers who wanted Labour to adopt the full IHRA definition and by 68 British Rabbis, who published a letter in the Guardian (2018) condemning the party’s new code. Needless to say, the media coverage of the controversy did Labour no favours. This was a wound that continued to bleed. Another key moment in Labour’s anti-Semitism crisis emerged in February 2019, when Luciana Berger, together with seven other Labour backbenchers, defected from the party claiming institutional anti-Semitism. Labour’s anti-Semitism crisis escalated further when three months later, the Equality and Human Rights Commission launched an investigation into Labour’s anti-semitic behaviour. The extent of the party’s anti-semitic crisis was highlighted by the decision of the Jewish Labour Movement to refuse to campaign for the party at the 2019 election, quickly followed by the recommendation by the Chief Rabbi Ephraim Mirvis to his congregants to vote for whichever party that is more likely to defeat Labour. Following this, Labour’s reputation as an anti-racist party was in tatters. The second problem that undermined Labour’s popularity was Brexit. The party’s 2017 manifesto accepted the results of the referendum on membership of the European Union (EU) as well as the end of free movement, while making some positive noises about a softer Brexit and the benefits of immigration (Labour Party, 2017). This ambivalent position was electorally astute (Mellon et al., 2018, p. 735), but following the general election, Corbyn was under pressure to adopt a more pro-European stance. Confronted with a divided parliamentary party on the most important issue of the day, Corbyn’s approach to Brexit was marked by obfuscation in the period between the two general elections. The party’s Shadow Brexit Secretary, Keir Starmer, told Labour’s 2018 conference in Liverpool that ‘nobody is ruling out Remain as an option’ in a second referendum. Corbyn appeared less enthused by such a move. It took months of campaigning by pro-European groups and trade unions, and the disastrous results of the European Parliament elections in June 2019, for Labour to adopt a clearer policy on Brexit. Finally, at the party’s 2019 annual conference, Corbyn announced that a Labour government would first negotiate a new withdrawal agreement with Brussels and then it would subject that deal to a new referendum where voters would also be given the option to vote to remain in the EU. But to the surprise of many, the Labour leader announced that he would remain neutral on the question and would implement the will of the people. Labour’s new Brexit policy, which reflected the views of the majority of Labour members (Bale et al., 2020, p. 70; Fieldhouse, 2019), was greeted with relief by activists who had been campaigning for the last two years for a new referendum on EU membership. However, Labour’s new position was announced just a few weeks before the general election and meant that the party had little time to explain it to voters. As a result, many candidates feared the worst. For candidates who represented Leave seats, it was clear that their voters would react angrily at the party’s blatant disrespect for the results of the 2016 referendum on EU membership (Cruddas, 2019). Others feared that Corbyn’s neutrality on the issue made him look indecisive and opportunistic. The party’s handling of these crises undermined Labour’s popularity, even among ‘Corbynistas’. The infectious enthusiasm of the youngsters who chanted ‘Oh, Jeremy Corbyn!’ at music festivals in 2017 waned as a result of the party’s inability to discipline anti-semitic behaviour, ambivalence over Brexit and resistance to democratising its policy-making structures. Indeed, as the Leader of the Opposition’s Office tightened its grip over the party’s policy-making machine, disillusion grew amongst activists. That disillusion was well captured by the convener of the left-wing grassroot movement Compass, Neal Lawson, who remarked that the ‘Labour leader’s office today is at least as powerful, and unaccountable, as it was under Blair and Brown. Whatever else it is, this is not bottom-up democracy’ (2019, p. 180). In a similar vein, Jeremy Gilbert, an academic who supported Corbyn, admitted that the leader’s office was ‘acquiring a reputation for secrecy, authoritarianism and narrow-mindedness that may have well been an inevitable product of seeking success within the Westminster system, but which threatens to demoralise the membership base’ (2019, p. 87). This account was confirmed by Andrew Fisher’s letter of resignation from the party, in which he accused the leader’s office of a ‘lack of professionalism, competence and human decency’ (quoted by Line and Bloom, 2019). The scale of Labour’s unpopularity was extraordinary given the inability of the then Prime Minister Theresa May to get her withdrawal agreement supported by her party and approved in parliament. But instead of capitalising on May’s failings and unpopularity, Labour was consumed by its own internal crisis and was therefore unable to mount any effective opposition to the government. 2. Falling into the fox’s trap In the meantime, the Conservatives got their act together following Theresa May’s resignation and Boris Johnson’s election as leader of the party in the summer of 2019. Johnson faced the same institutional and political limitations (i.e. he did not control a parliamentary majority) as of May, but his style of governing was radically different. Instead of dithering and last-minute charm offensives to parliamentarians, Johnson relished the confrontation with the House of Commons and the Supreme Court over his decision to prorogue parliament in the hope of getting his EU withdrawal agreement approved with as little scrutiny as possible. Johnson’s use of prerogative power to prorogue parliament for five weeks was adjudged to have been exercised unlawfully by the Supreme Court on 24 September. Forced to retreat, the Prime Minister, in October, reluctantly asked the EU for another extension for article 50, delaying Britain’s departure, but his strategy of forcing an early election eventually paid-off. Labour, together with the other opposition parties was totally unprepared for this turn of events. In reality, they fell into Johnson’s wily fox trap by reluctantly agreeing to an early election to be held in early December. But the timing of the election could not have been worse for Labour. The party was trailing behind the Conservatives by 15 percentage points (YouGov, 2019a). Corbyn’s personal ratings were even more problematic. In one survey, he ranked as the most unpopular leader of the opposition since 1977 (Skinner et al., 2019). The same survey showed that 76% of public opinion was ‘dissatisfied’ with Corbyn as opposition leader and 77% believed he was doing a bad job at handling’s Britain’s withdrawal from the EU. At the start of the electoral campaign, only 21% of voters thought that Corbyn would be the best Prime Minister (YouGov, 2019a). 3. The advent calendar campaign With these polling figures, it is not surprising that Labour launched its electoral campaign almost a week before the Conservative Party. Party strategists hoped to reverse the trend in the opinion polls with an offensive campaign strategy and by diverting voters’ attention from Brexit with a populist-themed campaign adapted to the festive season. To use the words of one of the party’s strategists, the aim of the campaign was to channel the ‘Christmas cheer and spirit’ into the election, with ‘Vote Labour’ bobble hats, mulled wine in flasks on the doorstep, Christmas jumpers and all manner of creative ideas’ (Ryle, 2019). Labour’s electoral promises would be presented as if displayed in an Advent Calendar: each day the party would announce a brand-new policy idea which would target a different segment of the electorate. In this spirit, Corbyn launched the party’s Christmas-themed campaign, sprinkled with small doses of populist anger against the ruling elites. In the speech that launched the party’s campaign at Battersea’s Arts Centre, he announced a ‘once-in-a-generation chance to transform our country’ and promised to go after ‘the tax-dodgers, the landlords, the bad bosses, big polluters’ (Corbyn, 2019b). Labour’s strategy involved avoiding Brexit as a campaigning theme. Aware that the party’s new stance on Brexit was not easy to explain to voters, Corbyn used his second speech of the campaign to promise that a Labour government would get ‘Brexit sorted within six months’ (2019a). In a parallel move, Momentum launched a video where an enthusiastic young man explained Labour’s ‘simple’ approach to Brexit in 30 seconds but for the rest of the campaign, the party tried to avoid the subject. It turns out that these efforts were insufficient as most voters thought that Labour’s Brexit policy was unclear (Abraham, 2019). Having decided to treat Brexit as an aside, Labour’s offensive campaign strategy had four main components: Corbyn’s visits; the daily announcements from the party’s headquarters; the ground campaign and the digital campaign. To some extent, the tactics deployed at the 2017 election were fine-tuned; however, there was an important difference. Instead of adopting the defensive strategy used in 2017 (Middleton, 2019, p. 508), this time Labour conducted an offensive campaign. Labour strategists believed that Labour could have won more seats in 2017 if the party had adopted a more offensive strategy. This reasoning explains the decision to conduct most of Corbyn’s visits in marginal or Conservative-held seats. Over the six-week campaign (from 31 October to 11 December), he only visited 29 Labour-held seats, out of his visits to 80 seats.1 1 The number of visits was calculated by looking at Labour’s press releases, Corbyn’s Instagram, Facebook and Twitter feeds, as well as the national and local media coverage. Corbyn spent most of the campaign visiting marginal seats and Conservative seats like Harlow, Ashfield, Telford, Milton Keynes, Loughborough or seats held by the Scottish National Party (SNP) in Scotland (which he visited twice), but his visits to Labour-held seats in the North East of England and the Midlands were rarer. But Corbyn’s offensive visits strategy was also the result of the unwillingness of Labour candidates to be seen side-by-side with Corbyn (Labour MP, 2020). Corbyn enjoyed a few good moments on the campaign trail. For example, when Yorkshire and the Midlands were hit by flooding, Corbyn was able to present himself as an empathetic leader who visited the affected areas. This image contrasted to that of the Prime Minister, who stayed away from the media glare. But these positive moments were rare. In truth, the crowds at Corbyn events were smaller and less enthusiastic than in 2017, and his media performances were lacklustre and in some of them (e.g. his BBC interview with Andrew Neil), he came across as impatient and angry. In terms of digital campaign, Labour relied on the grassroot movement Momentum’s infrastructure which had been upgraded since the 2017 campaign. Momentum launched the new website MyCampaignMap.com which directed activists to their nearest marginal seats and was used to set up more than 21,000 canvassing events. Momentum also launched the ‘Labour Legends’ campaign, which recruited 1,400 individuals who volunteered a week of their time to campaign for the party (Clarke, 2019), and trained thousands of activists on how to have persuasive conversations with voters, using techniques borrowed from the campaign team of the American presidential candidate Bernie Sanders. However, Momentum’s contribution to the campaign was not the asset it had been in the previous election. Halfway through the electoral campaign, the site MyCampaignMap.com adopted Labour’s offensive strategy and started to divert activists from defensive seats to participate in ‘Unseat Campaigns’ in places like Wimbledon, Hendon, Uxbridge and Chingford and Woodford Green, which Labour had few chances of winning. In addition, the Bernie Sanders’ inspired training sessions turned out to be less effective. Many activists were turning up to campaign events not knowing how to handle hostile voters, and some did not know the name of the candidate they were campaigning for. There was also significant variation in the number of footsoldiers on the ground. In London and in the South East of England, Labour candidates could rely on the support of hundreds of Labour and Momentum activists, but it was far more difficult and expensive to send footsoldiers to campaign in seats in the North East and the Midlands. Indeed, in seats which had not been targeted by the party’s headquarters (Proctor, 2020), Labour candidates could rely only on a relatively small pool of party members to conduct the campaign. But Labour’s main campaign problems were not created by Momentum. The party’s campaign was also in the words of one observer ‘shambolically organised’ (Rodgers, 2020). Information was concentrated in Corbyn’s office, and consequently there was, according to a post-mortem, a ‘lack of communication and preparation over key policy lines, lack of clarity of how campaigns could secure the data of newly registered voters, the unreliability of digital tools’ and inadequate training of activists’ (Somerville, 2020). The daily policy announcements did not help as it contributed to a campaign without a coherent and focussed narrative (Rodgers, 2019). Apart from Fisher and McDonnell few people knew which flagship policies would be announced on a daily basis (Stewart, 2020). Without guidance from the party, candidates around the country were forced to improvise. As Chris Bryant put it, ‘you never knew if you were getting a ring or a satsuma’ out of the Advent Calendar (2020). Labour thought that these shortcomings could be overcome by the launch of the party manifesto. To that end, the party organised a big manifesto launch event in Birmingham on 21 November which involved the entire shadow cabinet. The manifesto, entitled It’s Time For Real Change, was more radical than Labour’s 2017 manifesto and it promised free broadband to all, a Green Industrial Revolution, the nationalisation of the railways, water and electricity, a four-day week, the scrapping of university tuition fees, the rise of the minimum wage, a windfall tax on oil companies, the promise of building 100,000 new council houses every year until 2024, higher investment in the NHS and other public services, pay rises for nurses, teachers and police officers, pension justice for women and crucially a second referendum on Brexit (Labour Party, 2019). The media reaction to this Christmas themed manifesto was either negative or tepid. Most newspapers perceived it to be too radical and too costly. For example, the Daily Mail called it the ‘Marxist Manifesto’ and ‘Corbyn’s 83% Tax Robbery’; the Daily Express called it ‘The £80 BN Raid on Your Wallets’, the Financial Times claimed that Labour’s manifesto stirred the ‘spectre of the 1970s for business’, while The Guardian neutrally claimed that Corbyn had unveiled ‘Labour’s most radical manifesto for decades’. The only supporting newspaper was the Daily Mirror whose front page claimed that Labour was ‘On Your Side’. 4. Alarm bells start ringing By the end of November, alarm bells started to ring at Labour headquarters. The ground and digital campaigns, together with the manifesto launch had a very modest effect upon the party’s lack of popularity. If Labour’s approval ratings were slowly rising in the polls, so were the Conservatives’. But the real bombshell of the campaign was YouGov’s multilevel with poststratification poll (see the Introduction to this volume) released in late November. The model, which confirmed Labour’s internal polling and the success of the Conservatives’ strategy of targeting seats in Labour’s ‘red wall’, projected a comfortable Conservative majority of 359 whilst Labour was expected to win only 211 seats (Wells, 2019). This devastating projection led to a change in the party’s strategy in the last two weeks of the campaign. The party shifted campaigning resources to defend constituencies which thus far had been deemed safe and started to focus on ‘bread and butter issues’ like council housing, more local buses and free prescriptions that were more likely to attract traditional working-class voters in the ‘red wall’. Because many of the seats under threat were strong Leave constituencies the party also started to place more pro-Leave voices at the centre of the campaign. For instance, Ian Lavery led a tour of Leave areas, whereas figures like Richard Burgon, Laura Pidcock, Angela Rayner, Jon Trickett were given more prominent roles in Labour’s media strategy. In contrast, figures associated with the Remain campaign, like Keir Starmer, Emily Thornberry, Diane Abbott, were told to keep off the airwaves. By the last week of the campaign, morale was extremely low among Labour activists and campaigners. Opinion polls suggested that the Conservatives retained a solid lead over Labour and the feedback from the candidates and canvassers confirmed that trend. Nonetheless, Corbyn pursued with his gruelling schedule of visits, having spent the last day of campaign visiting Scotland and Middlesbrough for the second time (Stewart, 2019). On election day, Labour was preparing for bad news, but the electoral results went beyond the party’s worst fears. Labour attracted only a 32.9% share of the vote across Great Britain (32.1% of the UK vote), with only 202 MPs elected, the lowest number of seats won since 1935. Labour retained only 72% of its 2017 voters. The loss of ultra-safe and traditional Labour seats like Bolsover, Sedgefield, Blyth Valley, Don Valley, Leigh, Workington and Bishop Auckland came to symbolise Labour’s existential crisis. However, the results showed that Labour’s crisis extended well beyond the famous ‘red wall’. The party lost seats everywhere and to every party and lost votes across all demographic groups (Cooper and Cooper, 2020, p. 12). Faced with such devastating results, Corbyn announced his resignation from the leadership of the party but failed to take responsibility for Labour’s defeat. In reality, his office blamed Keir Starmer for forcing the party to adopt a new Brexit stance which seemed to disrespect the result of the 2016 referendum. This interpretation of Labour’s crushing defeat was rejected by many. Indeed, three-quarters of Labour councillors said that Labour’s national leadership was the main reason for voters leaving the party (Butler, 2020). And for those candidates who lost their seats or who were re-elected on small majorities, the blame lay mostly with Corbyn, who was not only unpopular but seen as an ineffective leader (Merrick, 2020). This view was encapsulated by Alan Johnson’s outburst against Corbyn whom, he said, ‘couldn’t lead the working class out of a paper bag’ (ITV News, 2019) let alone to electoral victory. But the scale of Labour’s defeat cannot be attributed to a single factor or simply to contextual factors like Brexit or a poorly run electoral campaign, especially because this was the party’s fourth consecutive defeat since 2010. The following section will analyse the short-term and contextual factors as well as the long-term trends that explain Labour’s results. 5. Explaining the results The electoral results and survey and polling data show that there was an amalgam of reasons, both contextual and structural, which led so many voters to abandon Labour. Among the different factors that can explain the party’s defeat, Corbyn’s reputation, the party’s Brexit policy, a transactional manifesto which overpromised, a poorly run campaign together with long-term factors like structural political changes and new trends in voting behaviour, standout. These different factors did not work in isolation. In some instances—for instance, the Brexit and Corbyn factors—they reinforced each other. Doubtless, Brexit influenced voting behaviour. This was after all the issue that had dominated British politics since 2016 and the reason why an early election was called. Thus, it is not surprising that of the 60 seats that Labour lost, 52 represented Leave areas, though the party also lost Remain voters (Cooper and Cooper, 2020, p. 16). One analysis of the electoral results showed that only 14% of Leave voters voted for Labour in 2019 (down from 24% in 2017) and only 49% of Remain voters voted for the party (House of Commons, 2020). Lord Ashcroft’s polls, outlined in the opening results chapter, indicate a higher vote share for Labour among Leave voters, of 26%, compared with 47% of Remain voters. What is clear from both sets of figures is that Labour’s support among Leave voters was woefully low. Leavers did not see Brexit as ‘getting done’ by Labour. Analysis by the Resolution Foundation shows that ‘changes in Labour’s vote shares were more shaped by their Brexit vote’ than in the 2017 general election, but other factors, like voters’ perception of the competence of party leaders, were also contributing factors (Bell, 2019; see also Curtis 2019). Lord Ashcroft’s (2020) analysis of Labour’s defeat also claimed that two-thirds of Labour Party members, including three-quarters of those who voted for Corbyn as leader, said that Brexit dominated the election and had a bigger effect on the result than on how people felt about the parties, leaders and other policies (Ashcroft, 2020). However, it is not clear whether a pro-Leave position would have resulted in retaining seats in Leave areas. Lord Ashcroft’s survey indicated that 85% of Labour voters said they would have probably voted the same way had Brexit not been an issue (Ashcroft, 2020). In truth, candidates and activists who campaigned claimed that the Brexit factor could not be disentangled from voters’ attitudes towards the leader of the opposition. As Curtice explained, the ‘lack of leadership’ was a very big problem for Labour because, among other reasons, it led to indecision on Brexit and perceptions of competence and electability (2020). Perceptions of competence influence voting behaviour (Stewart and Clarke, 1992, p. 467), especially ‘when major events heighten the relevance of competence’ (Green and Jennings, 2018, p. 11), when parties prime questions of competence and performance and when high salience events ‘politicise the handling of an issue and make managerial competence’ more important (Green and Jennings, 2018, p. 221). After three years of parliamentary deadlock around Brexit, it can be argued that perceptions of competence seemed to be particularly relevant to voters in 2019. In truth, the Conservative Party primed competence as an important benchmark to evaluate party electability and centred its campaign on the promise of ‘getting Brexit done’. It is then not surprising that most analyses of Labour’s defeat identified Corbyn’s reputation as a leader as the main reason why so many voters abandoned the party. YouGov’s (2019a) report showed that Corbyn’s leadership was the main reason why Labour voters defected (Curtis, 2019). Lord Ashcroft’s (2020) report also showed that 53% of Labour defectors did not want Jeremy Corbyn to be Prime Minister. The same survey suggested that voters found Corbyn unpatriotic and lacking in leadership qualities. Corbyn’s unpopularity had many facets. As George Eaton (2019) explained, he was not trusted on national security; he was perceived as an unpatriotic, weak leader unable to manage a divided party and who did not look prime-ministerial. An illustration of the multi-faceted nature of popular hostility to Corbyn was offered by the former Labour whip, Graham Jones’ canvassing experiences: ‘door one wouldn’t vote for him because he was scruffy; the next person wouldn’t vote for him because of antisemitism in the party; the next because of connections to Hamas; and the next because he seemed unable to lead’ (Jones quoted Syal et al., 2019). Labour’s ambitious manifesto was also invoked as one the causes of its defeat, the party’s manifesto promises viewed as undeliverable, unrealistic and too expensive by many (Curtis, 2019). Similarly, Lord Ashcroft’s (2020) study identified Labour’s manifesto as a cause of Labour’s unpopularity. Labour’s badly organised campaign also played a role, especially because it contrasted with the Conservatives’ laser-focussed ‘Get Brexit Done’ campaign, though it must be said that electoral campaigns are rarely game changers. At best, campaigns have a mobilising effect which is reflected in the media coverage (Banducci and Karp, 2003, p. 463). But Labour’s 2019 results cannot be fully understood by considering only the short-term factors that influenced voting behaviour, especially because this was the party’s fourth consecutive electoral defeat. Indeed, it is important to consider that the election took place following ten years of high political turbulence. As the four general elections (two of which delivered hung parliaments), a referendum on Scottish independence and a national referendum on Britain’s relationship with the EU vividly illustrate, the British party system has suffered in the last decade several ‘electoral shocks’ which rendered voting behaviour more fluid, volatile and unpredictable (Fieldhouse et al., 2020, p. 30). In truth, the instability of the party system predates this decade of political turbulence (Goes, 2019, pp. 295–297). Hyper-globalisation, the rise in immigration flows and the adaptation by political parties to these changes led to class and partisan dealignments which in turn altered voting behaviour. In particular, the emergence in the last two decades of a globalisation cleavage led to significant changes in voter behaviour not only in Britain but also in most European party systems (Jennings and Stoker, 2016; Kriesi et al., 2006; Hobolt, 2016, p. 1260; Cutts et al., 2020, p. 17). As globalisation created ‘winners’ and ‘losers’ in society their antagonism manifested itself at economic, cultural and political levels. Thus, as shown by Kriesi et al., the ‘losers of globalisation’, who tend to be older, with fewer qualifications and live in rural, suburban or post-industrial areas, are more likely to vote for parties that defend tighter immigration controls, economic protectionism and authoritarian values, whereas the ‘winners’, who tend to be younger, with university qualifications and live in ethnically diverse urban areas, are more inclined to vote for parties that favour European integration, support immigration and cosmopolitan and libertarian values (2006, p. 924). It was precisely this globalisation cleavage which transformed the demographic and geographic profile of Labour support in the last two decades. Indeed, as Evans and Tilley show, the working classes started to desert the Labour Party after 2001 (2017, p. 153), but the 2015 general election was the first election since the war when the ‘working class voted for Labour at a lower rate than some middle-class groups’ (2017, p. 152). This new pattern of class voting was again in evidence at the 2019 general election. In terms of social class, Labour attracted only 33% of the vote of those categorised as C2DE, while the Conservatives attracted 48% of that social class. The old electoral laws of social class had been reversed. The Conservatives enjoyed significantly more support among the working class than did Labour. Meanwhile, Labour attracted 33% of the ABC1 social class, compared with 43% support for the Conservatives in that middle-class category (YouGov, 2019b). Labour’s support was evenly spread across the social classes, too thinly in both cases. New social class voting patterns are accompanied by age, educational and geographical dimensions, confirming Kriesi et al.’s globalisation cleavages. Age was the key indicator of voting behaviour at the 2019 election: 56% of 18- to 24-year-old voted Labour, while only 22% of the over-60s did likewise. Only 25% of voters without educational qualifications voted for Labour, while 43% of Labour voters had university degrees (McDonnell and Curtis, 2019). Age is correlated with education, as younger people tend to have higher educational qualifications. In terms of the geographical dimension, Jennings and Stoker showed that, at the previous two elections, support for Labour had tended to be concentrated particularly in urban areas connected to global growth, but declined in a cluster of coastal, post-industrial, suburban and rural areas (2017, p. 361; see also Jennings and Stoker, 2019 , p. 156). This trend was confirmed and consolidated in 2019. To understand how the globalisation cleavage played out at the 2019 election, we need to consider that the new pattern of class voting reflects economic and social values and also cultural identities normally expressed in public attitudes to immigration (Evans and Tilley, 2017, p. 186). Analysis by Evans and Chzhen confirmed the expression of this new class voting, by correlating the decline in Labour’s support in the 2005 and 2010 elections to popular disquiet with levels of immigration (2013, p.19). In 2019, that globalisation cleavage is associated with the Brexit factor. A closer look at Labour’s share of the vote in some of the ‘red wall’ seats over the last two decades shows us again the impact of globalisation in voting behaviour. As Table 5.1 shows Labour’s decline in support started in 2005, but became spectacularly visible following the 2008 global financial crisis and the 2016 referendum on EU membership and was confirmed at the 2017 and 2019 general elections. Table 5.1 Labour share of the vote in English ‘red wall’ seats, 2001–2019 (%) Constituency . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Ashfield 58.1 48.6 33.7 41.0 42.6 24.4 Barrow Furness 55.7 47.6 48.1 42.3 47.5 39.3 Bassetlaw 55.3 56.6 50.5 48.6 52.6 27.5 Bishop Auckland 58.8 50.0 39.0 41.4 48.1 35.9 Blyth Valley 59.7 55.0 44.5 46.3 55.9 40.9 Bolsover 68.6 65.2 50.0 51.2 51.9 35.9 Burnley 49.3 38.6 31.3 37.6 46.7 36.9 Darlington 56.3 52.4 39.4 42.9 50.6 40.5 Don Valley 54.6 52.7 37.9 46.2 53.0 35.2 Leigh 64.5 63.3 48.0 53.9 56.2 41.1 NW Durham 62.5 53.9 42.3 46.9 52.8 41.9 Redcar 60.3 51.4 42.3 43.9 55.5 37.4 Sedgefield 64.9 58.9 45.1 47.2 53.4 36.3 Stockton South 53.0 47.8 38.3 37.0 48.5 41.1 Workington 55.5 50.5 45.0 42.3 51.1 39.2 Constituency . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Ashfield 58.1 48.6 33.7 41.0 42.6 24.4 Barrow Furness 55.7 47.6 48.1 42.3 47.5 39.3 Bassetlaw 55.3 56.6 50.5 48.6 52.6 27.5 Bishop Auckland 58.8 50.0 39.0 41.4 48.1 35.9 Blyth Valley 59.7 55.0 44.5 46.3 55.9 40.9 Bolsover 68.6 65.2 50.0 51.2 51.9 35.9 Burnley 49.3 38.6 31.3 37.6 46.7 36.9 Darlington 56.3 52.4 39.4 42.9 50.6 40.5 Don Valley 54.6 52.7 37.9 46.2 53.0 35.2 Leigh 64.5 63.3 48.0 53.9 56.2 41.1 NW Durham 62.5 53.9 42.3 46.9 52.8 41.9 Redcar 60.3 51.4 42.3 43.9 55.5 37.4 Sedgefield 64.9 58.9 45.1 47.2 53.4 36.3 Stockton South 53.0 47.8 38.3 37.0 48.5 41.1 Workington 55.5 50.5 45.0 42.3 51.1 39.2 Source:House of Commons (2001, 2005, 2011, 2015, 2019, 2020). Open in new tab Table 5.1 Labour share of the vote in English ‘red wall’ seats, 2001–2019 (%) Constituency . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Ashfield 58.1 48.6 33.7 41.0 42.6 24.4 Barrow Furness 55.7 47.6 48.1 42.3 47.5 39.3 Bassetlaw 55.3 56.6 50.5 48.6 52.6 27.5 Bishop Auckland 58.8 50.0 39.0 41.4 48.1 35.9 Blyth Valley 59.7 55.0 44.5 46.3 55.9 40.9 Bolsover 68.6 65.2 50.0 51.2 51.9 35.9 Burnley 49.3 38.6 31.3 37.6 46.7 36.9 Darlington 56.3 52.4 39.4 42.9 50.6 40.5 Don Valley 54.6 52.7 37.9 46.2 53.0 35.2 Leigh 64.5 63.3 48.0 53.9 56.2 41.1 NW Durham 62.5 53.9 42.3 46.9 52.8 41.9 Redcar 60.3 51.4 42.3 43.9 55.5 37.4 Sedgefield 64.9 58.9 45.1 47.2 53.4 36.3 Stockton South 53.0 47.8 38.3 37.0 48.5 41.1 Workington 55.5 50.5 45.0 42.3 51.1 39.2 Constituency . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Ashfield 58.1 48.6 33.7 41.0 42.6 24.4 Barrow Furness 55.7 47.6 48.1 42.3 47.5 39.3 Bassetlaw 55.3 56.6 50.5 48.6 52.6 27.5 Bishop Auckland 58.8 50.0 39.0 41.4 48.1 35.9 Blyth Valley 59.7 55.0 44.5 46.3 55.9 40.9 Bolsover 68.6 65.2 50.0 51.2 51.9 35.9 Burnley 49.3 38.6 31.3 37.6 46.7 36.9 Darlington 56.3 52.4 39.4 42.9 50.6 40.5 Don Valley 54.6 52.7 37.9 46.2 53.0 35.2 Leigh 64.5 63.3 48.0 53.9 56.2 41.1 NW Durham 62.5 53.9 42.3 46.9 52.8 41.9 Redcar 60.3 51.4 42.3 43.9 55.5 37.4 Sedgefield 64.9 58.9 45.1 47.2 53.4 36.3 Stockton South 53.0 47.8 38.3 37.0 48.5 41.1 Workington 55.5 50.5 45.0 42.3 51.1 39.2 Source:House of Commons (2001, 2005, 2011, 2015, 2019, 2020). Open in new tab In the 15 seats displayed in Table 5.1, a pattern of electoral decline in Labour’s working-class heartlands can be identified between 2001 and 2019. In 2001, Labour won its comfortable electoral victory but by 2005, the large influx of immigrants from the new EU accession countries started to erode support for Labour in some of these seats. By 2010, two years after the global financial crisis, Labour suffered a devastating defeat which continued in 2015 but two years later there was surprising surge in support for Labour. In 2019, the decline in Labour support resumed, expressed in dramatic fashion. In seats like Ashfield, Redcar, Leigh, North West Durham and Stockton South, Labour registered notable vote share gains in 2017, but two years later Labour lost those seats to the Conservatives for the first time, in some cases in spectacular fashion. Labour’s woeful results in Leave-voting former heartlands indicate Brexit salience (see Curtice in this volume). In 2017, Labour promised to respect the results of the EU referendum and to reverse austerity measures and was rewarded with very big majorities in those seats, but in 2019, the promise of a second referendum on EU membership and lack of clarity over Labour’s position on Brexit proved unappealing, restoring the pattern of decline in the Labour vote in these areas. Thus, to a certain extent what was expressed in the last two general elections was not simply a Brexit effect but the long-term impact of the globalisation cleavage in the voting behaviour of traditional Labour voters (see also Flinders in this volume). 6. Conclusion: Labour on life support Labour’s decline in its traditional heartlands and concomitant transformation into a party of the ‘winners of globalisation’ led the backbencher Jon Cruddas to conclude that ‘paradoxically Corbyn completed Tony Blair’s project’ of transforming Labour into a party of the urban middle classes (2019). This is an interesting conclusion but there is no denying that these new trends in class, age and geographical vote pose an almost existential challenge to the Labour Party. As Lord Liddle argued, ‘Labour is still a viable party, but it might never again be a party of government’ (2020). Simply put, Labour cannot be a party of government by relying solely on the support of middle-class voters in English cities. Corbyn’s successor as Labour leader, Keir Starmer, elected in April 2020, was elevated to restore the party’s much battered credibility. As the candidate who attracted support from all Labour wings and won the leadership election with a resounding victory, the immediate expectations of Starmer were the unification of a bitterly divided party and the capacity to offer a more professional and forensic opposition to the Conservative government. Nonetheless, as the new Labour leader admitted in his acceptance speech, he faces a proverbial mountain to climb. To turn Labour into a party of government again, Starmer has to craft a political programme which keeps the urban, educated middle classes on board while bringing back into Labour’s fold the suburban, post-industrial communitarian working-class voters of the North of England, Midlands, Wales and also Scotland who have abandoned the party since 2005. This is a tall order. After all, these two sets of very different voters will struggle to find common ground on issues of law and order, immigration and national security but may agree on a social, democratic, economic approach that addresses the devastating socio-economic impact of a pandemic-led recession and of Britain’s withdrawal from the EU, the effects of which will be acute at the time of the next election. If that is the case, Starmer, may, but only just, find the solution to the electoral puzzle that has kept Labour out of power since 2010. Acknowledgements I would like to thank Lord Liddle, Lord Giddens, Jon Cruddas, Bridget Phillipson, Chris Bryant, Sienna Rodgers, Michael Chessum, Luke Cooper, Natalie Sedacca and other Labour MPs, councillors and party members who asked to remain anonymous, for sharing their time and insights. All mistakes in analysis are, of course, my own. References Abraham T. ( 2019 ) ‘Most Brits Uncertain on Labour’, YouGov, 5 November, accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/11/05/most-brits-uncertain-labours-brexit-policy on 10 January 2020. Ashcroft L. ( 2020 ) ‘Diagnosis of Defeat: Labour’s Turn to the Smell the Coffee’, Lord Ashcroft Polls, 12 February 2020, accessed at https://lordashcroftpolls.com/wp-content/uploads/2020/02/DIAGNOSIS-OF-DEFEAT-LORD-ASHCROFT-POLLS-1.pdf on 17 February 2020. Bale T. , Webb P. , Poletti M. ( 2020 ) Footsoldiers: Political Party Membership in the 21st Century , London , Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Banducci S. A. , Karp J. A. ( 2003 ) ‘ How Elections Change the Way Citizens View the Political System: Campaigns, Media Effects and Electoral Outcomes in Comparative Perspective ’, British Journal of Political Science , 33 , 443 – 467 . Google Scholar Crossref Search ADS WorldCat Bell T. ( 2019 , December 13) ‘Election Dissection: Top of the Charts’, Resolution Foundation, accessed at https://www.resolutionfoundation.org/comment/election-dissection/ on 17 March 2020. Bryant C. ( 2020 ) ‘Corbyn’s Legacy’, Mile End Institute Event, ‘What Next for Labour’, 20 January 2020. London. Butler P. ( 2020 , February 6) ‘Labour and Jeremy Corbyn under Fire from Councillors’ Leader’, The Guardian , accessed at https://www.theguardian.com/politics/2020/feb/06/labour-and-jeremy-corbyn-under-fire-from-councillors-leader on 20 February 2020. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Clarke L. ( 2019 , November 30) ‘Inside Momentum’s Data-Driven March to Put Corbyn in Number 10’, Wired, accessed at https://www.wired.co.uk/article/momentum-labour-party-my-campaign-map on 10 February 2020. Cooper L. , Cooper C. ( 2020 ) ‘The Devastating Defeat: Why Labour Lost and How It Can Win Again: Part 1, Europe for the Many’, 12, accessed at https://www.europeforthemany.com/the-devastating-defeat/ on 1 February 2020. Corbyn J. ( 2019 a, November 5) ‘Brexit Speech’, Harlow, accessed at https://labour.org.uk/press/jeremy-corbyns-brexit-speech-in-harlow/ on 31 March 2020. Corbyn J. ( 2019 b, October 31) ‘First Major Speech of the General Election Campaign’, Battersea Arts Centre, 31 October, accessed at https://labour.org.uk/press/jeremy-corbyns-first-major-speech-of-the-general-election-campaign/ on 1 February 2020. Corbyn J. ( 2019 c) ‘Speech to the 2017 Labour Party Annual Conference’, 27 September 2019, accessed at https://labour.org.uk/press/jeremy-corbyn-speech-to-labour-party-conference/ on 30 November 2019. Cruddas J. ( 2019 ) Private Interview, 8 October 2019. House of Commons, London. Curtice J. ( 2020 ) ‘Why Labour Lost’, Coffee House, The Spectator Blog, accessed at https://blogs.spectator.co.uk/2020/01/why-labour-lost/ on 18 January 2020. Curtis C. ( 2019 ) ‘In Their Own Words: Why Voters Abandoned Labour’, YouGov, 23 December, accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/12/23/their-own-words-why-voters-abandoned-labour on 10 January 2020. Cutts D. , Goodwin M. , Heath O. , Surridge P. ( 2020 ) ‘ Brexit, the 2019 General Election and the Realignment of British Politics ’, The Political Quarterly , 91 , 7 – 23 . Google Scholar Crossref Search ADS WorldCat Eaton G. ( 2019 , December 15) ‘How Labour Lost – And How It Can Recover from an Epic Defeat’, New Statesman, accessed at https://www.newstatesman.com/politics/uk/2019/12/why-labour-lost-and-how-it-can-recover-epic-defeat on 10 January 2020. Evans G. , Tilley J. ( 2017 ) The New Politics of Class: The Political Exclusion of the British Working Class , Oxford , Oxford University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Evans G. , Chzhen K. ( 2013 ) ‘ Explaining Voters’ Defection from Labour over the 2005-2010 Electoral Cycle: Leadership, Economics and the Rising Importance of Immigration ’, Political Studies , 61 , 138 – 122 . Google Scholar Crossref Search ADS WorldCat Evening Standard ( 2020 ) ‘Corbyn Adviser Says ‘Hubris Cost Us’’, 22 January 2020, accessed at https://www.standard.co.uk/news/londoners-diary/the-londoner-corbyn-adviser-says-hubris-cost-us-a4341276.html on 10 April 2020. Fieldhouse E. , Green J. , Evans G. , Mellon J. , Prosser C. , Schmitt H. , Van Der Eijk C. ( 2020 ) Electoral Shocks: The Volatile Voter in a Turbulent World , Oxford , Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fieldhouse E. ( 2019 ) ‘Labour’s Electoral Dilemma’, British Election Study, 19 October, accessed at https://www.britishelectionstudy.com/bes-findings/labours-electoral-dilemma/#.XlVexpP7Ts0 on 31 March 2020. Gilbert J. ( 2019 ) ‘Acid Corbynism for Beginners’. In Perryman M. (ed), Corbynism from Below , London , Lawrence and Wishart , pp. 80 – 103 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Goes E. ( 2019 ) ‘The British Party System(s): Fragmented, Unstable and Very Capricious’. In Lisi M. (ed), Party System Change, the European Crisis and the State of Democracy , Abingdon , Routledge , pp. 289 – 309 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Green J. , Jennings W. ( 2018 ) The Politics of Competence: Parties: Public Opinion and Voters , Cambridge , Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Guardian ( 2018 , July 18) ‘Letter to the Editor: Labour Party Must Listen to the Jewish Community on Defining Antisemitism’, accessed at https://www.theguardian.com/politics/2018/jul/16/labour-party-must-listen-to-the-jewish-community-on-defining-antisemitism on 10 April 2020. House of Commons ( 2020 ) General Elections 2019: Results and Analysis , London , House of Commons Library . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC House of Commons ( 2019 ) General Elections 2017: Results and Analysis, Briefing Paper CBP 7979, London, House of Commons Library. House of Commons ( 2015 ) General Election 2015, Briefing Paper, Number CBP7186, London, House of Commons Library. House of Commons ( 2011 ) General Election 2010, Final Edition, Research Paper 10/36, London, House of Commons Library. House of Commons ( 2005 ) General Election 2005, Final Edition, Research Paper 05/33, London, House of Commons Library. House of Commons ( 2001 ) General Election Results, 7 June 2001, Research Paper 01/54, London, House of Commons Library. Hobolt S. B. ( 2016 ) ‘ The Brexit Vote: A Divided Nation, a Divided Continent ’, Journal of European Public Policy , 23 , 1259 – 1277 . Google Scholar Crossref Search ADS WorldCat ITV News ( 2019 , December 13) ‘He Couldn’t Lead the Working Class Out of a Paper Bag’: Alan Johnson Launches Fierce Attack On Corbyn and Momentum’, accessed at https://www.itv.com/news/2019-12-13/alan-johnson-momentum-jeremy-corbyn-itv-news/ on 10 April 2020. Jennings W. , Stoker G. ( 2019 ) ‘ The Divergent Dynamics of Cities and Towns: Geographical Polarisation and Brexit ’, The Political Quarterly , 90 , 155 – 166 . Google Scholar Crossref Search ADS WorldCat Jennings W. , Stoker G. ( 2017 ) ‘ Tilting towards the Cosmopolitan Axis? Political Change in England and the 2017 General Election ’, The Political Quarterly , 88 , 359 – 339 . Google Scholar Crossref Search ADS WorldCat Kogan D. ( 2019 ) Power and Protest: The Battle for the Labour Party , London , Bloomsbury . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kriesi H. , Grande E. , Lachat R. , Dolezal M. , Bornschier S. , Frey T. ( 2006 ) ‘ Globalisation and the Transformation of the National Political Space: Six European Countries Compared ’, European Journal of Political Research , 45 , 921 – 956 . Google Scholar Crossref Search ADS WorldCat Labour MP. Private Interview, 10 March 2020. House of Commons, London. Labour Party ( 2019 ) It’s Time for Real Chance: The Labour Party Manifesto 2019 , London , Labour Party , accessed at https://labour.org.uk/wp-content/uploads/2019/11/Real-Change-Labour-Manifesto-2019.pdf on 28 February 2020. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Labour Party ( 2017 ) For the Many Not the Few: The Labour Party Manifesto 2017 , London , Labour Party . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lawson N. ( 2019 ) ‘Labouring under Illusions’. In Perryman M. (ed) Corbynism from Below , London , Lawrence and Wishart . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Liddle L. ( 2020 ) Private Interview, 8 January 2020. Line H. , Bloom D. ( 2019 ) ‘ Top Jeremy Corbyn Aid Resigns Attacking 'Blizzard of Lies’ in the Party ’, Mirror, 22 September 2019, accessed at https://www.mirror.co.uk/news/politics/top-jeremy-corbyn-aide-resigns-20141185, on 6 January 2020. OpenURL Placeholder Text WorldCat Mellon J. , Evans G. , Fieldhouse E. , Green J. , Prosser C. ( 2018 ) ‘ Brexit or Corbyn? Campaign and Inter-Election Vote Switching in the 2017 UK General Election ’, Parliamentary Affairs , 71 , 719 – 737 . Google Scholar Crossref Search ADS WorldCat McDonnell A. , Curtis C. ( 2019 ) ‘How Britain Voted in the 2019 General Election’, YouGov, 17 December, accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/12/17/how-britain-voted-2019-general-election on 10 January 2020. Merrick R. ( 2020 ) Corbyn’s ‘Cabal’ of Top Aides Sabotaged Election Campaign, Says Labour Leader’s Close Friend’, The Independent, 8 February, accessed at https://www.independent.co.uk/news/uk/politics/corbyn-general-election-labour-alan-simpson-seamus-milne-karie-murphy-a9324066.html on 12 March 2020. Middleton A. ( 2019 ) ‘ For the Many Not the Few: Strategising the Campaign Trail at the 2017 General Election ’, Parliamentary Affairs , 72 , 501 – 521 . Google Scholar Crossref Search ADS WorldCat Proctor K. ( 2020 ) ‘Labour’s Canvassing Strategy Had ‘Major Deficiencies’: Leaked Report Says’, The Guardian, 7 February, accessed at https://www.theguardian.com/politics/2020/feb/07/labours-canvassing-strategy-had-major-deficiencies-leaked-report-says on 12 March 2020. Rodgers S. ( 2020 ) ‘Corbyn’s Legacy’, Mile End Institute event, ‘What Next for Labour?’ 20 January 2020. Rodgers S. ( 2019 ) ‘How Labour Lost the Red Wall’, The House Magazine, 16 December, accessed at https://www.politicshome.com/news/uk/political-parties/labour-party/house/house-magazine/108584/sienna-rodgers-how-labour-lost on 10 January 2020. Ryle J. ( 2019 ) ‘Labour Must Make This a Christmas Election, Not a Brexit Election’, LabourList, 29 October, accessed at https://labourlist.org/2019/10/labour-must-make-this-a-christmas-election-not-a-brexit-election/ on 11 January 2020. Shipman T. ( 2018 ) Fallout: A Year of Political Mayhem , London: William Collins. Skinner G. , Pedley K. , Gottfried G. , Garrett C. ( 2019 , September 20) ‘Jeremy Corbyn Has Lowest Leadership Satisfaction Rating For Any Opposition Leader Since 1977’, Ipsos MORI, accessed at https://www.ipsos.com/ipsos-mori/en-uk/jeremy-corbyn-has-lowest-leadership-satisfaction-rating-any-opposition-leader-1977 on 10 March 2020. Somerville E. ( 2020 , February 8) ‘Labour Campaign Weaknesses Exposed as ‘Major Deficiencies In Election’, I Newspaper, accessed at https://inews.co.uk/news/labour-election-defeat-jeremy-corbyn-canvassers-1393517 on 27 February 2020. Stewart H. ( 2020 , February 21) ‘Crushed By Brexit: How Labour Lost the Election’, The Guardian, accessed at https://www.theguardian.com/politics/2020/jan/21/crushed-by-brexit-how-labour-lost-the-election on 27 February 2020. Stewart H. ( 2019 , December 13) ‘Clashing Egos and policy incontinence’: Inside Labour’s Campaign’, The Guardian, accessed at https://www.theguardian.com/politics/2019/dec/13/inside-labours-campaign-behind on 6 January 2020. Stewart M. C. , Clarke H. D. ( 1992 ) ‘ The (Un) Importance of Party Leaders: Leader Images and Party Choice in the 1987 British Election ’, Journal of Politics , 54 , 447 – 470 . Google Scholar Crossref Search ADS WorldCat Syal R. , Mason R. , Stewart H. , Perraudin F. ( 2019 , December 17) ‘Corbyn, Antisemitism and Brexit: Labour MPs on Why They Lost’, The Guardian, accessed at https://www.theguardian.com/politics/2019/dec/17/corbyn-antisemitism-and-brexit-labour-mps-on-why-they-lost on 10 January 2020. YouGov ( 2019 a) ‘Westminster Voting Intention Tracker 2017–2019’, accessed at https://yougov.co.uk/topics/politics/articles-reports/2020/02/05/political-trackers-31-jan-2-feb-2020-update on 21 February 2020. YouGov ( 2019 b, December 17) ‘How Britain voted in the 2019 General Election’, accessed at https://yougov.co.uk/topics/politics/articles-reports/2019/12/17/how-britain-voted-2019-general-election on 31 March 2020. Wells A. ( 2019 , November 27) ‘YouGov MRP: Conservatives 359, Labour 211, SNP 43, LD 13, Plaid 4, Green 1’, accessed athttps://yougov.co.uk/topics/politics/articles-reports/2019/11/27/yougov-mrp-conservatives-359-labour-211-snp-43-ld- on 27 January 2020. © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Cutts,, David;Russell,, Andrew

doi: 10.1093/pa/gsaa024pmid: N/A

The year 2019 was supposed to be the year of the Liberal Democrats’ rehabilitation. They had apparently emerged from the long, dark shadows cast by their coalition with the Conservatives, and their electoral trouncing in 2015 and poor showing in 2017. They had enjoyed a period of stability, and remarkable progress, under the leadership of Vince Cable who inherited a party that had been worn down to a husk. As Labour shifted to the left and the Conservatives chose full-on Euroscepticism and inner turmoil, the Liberal Democrats started to grow. A rearguard action in local government, European Election success and victory in a Westminster by-election saw off a centre party threat seeking to claim their territory. Brexit—and the need to stop it—was the party’s passport back to the top table of British politics. When Cable stood down as leader, he was succeeded by a candidate who seemed to fit the bill for what the party most desired. Jo Swinson was young, female and had a reputation in the Westminster circle for competence and empathy. Under Swinson, the Liberal Democrats grasped the opportunity with both hands to reinvent themselves as the only GB wide, unambiguously anti-Brexit party. With pre-election polls placing them on the coattails of Labour, the Liberal Democrats then went for the jugular. Swinson began the campaign in spectacularly upbeat fashion, reasserting that a majority Liberal Democrat government would revoke Article 50 and announcing that she was running to be the new Prime Minister. A few weeks later, the party’s campaign had turned to ashes, the number of Westminster Liberal Democrats actually fell and Swinson herself, for the second time in four years, lost her own seat to the Scottish National Party (SNP). Everything that might have gone wrong went spectacularly awry. So what was to blame for the Liberal Democrats’ failure to sustain pre-election expectations? Why in the so-called ‘Brexit election’ did the party whose raison d’être was to stop Britain leaving the European Union (EU) fail to make significant electoral headway? Why did the Liberal Democrats get their political and electoral strategy so badly wrong? We examine what happened and why and assess whether this represented another missed opportunity, or whether the failings of the campaign mask the underlying story of slow, relative and partial, recovery in Liberal Democrat fortunes. 1. Building momentum After Tim Farron’s resignation, Sir Vince Cable was elected unopposed as leader, with the unenviable task of reconnecting with an increasingly polarised and divided electorate. The Cable era can be divided into two parts. Initially, Cable was keen to play the ‘statesmanlike’ card by making a virtue out of his longevity in public life. On Brexit, for instance, he immediately painted himself as the ‘political adult’ in the room and while ruling out formal pacts with the others sought to call out ‘sensible grown-ups’ from both parties who shared his and the party’s position to keep Britain in the single market and customs union. He recognised that the party needed more media exposure to rally the troops and sought to provide clearer political messaging and policy positions in order to keep the Liberal Democrat badge in the forefront of public debate and people’s minds. Yet, despite all the positive rhetoric, optimism and denial that he was merely a short-term caretaker, the reality was something different. The first 18 months of Cable’s leadership was pedestrian, lacklustre and did little to raise the party’s profile. The Liberal Democrats poll rating barely moved much above 8% while the 2018 local election results were, in terms of its nationwide share of the vote, one of the worst local election performances by the Liberal Democrats since their formation. Unsurprisingly, a year on from becoming leader, Cable confirmed that he would stand down after Brexit was resolved or stopped and after he had transformed the inner workings of the party. Yet, writing off Cable proved to be premature. During the first half of 2019, both the Conservatives and Labour hit internal strife and showed signs of fracturing as they faced increasing public derision over Brexit. When the inevitable happened, the new Independent Group of 11 MPs was vocal in expressing their desire not to join the Liberal Democrats, claiming the party brand had been damaged by coalition with the Conservatives. Cable responded in a measured manner, calling for electoral alliances around shared interests such as a second Brexit referendum but noted the structural electoral hurdles for two centrist parties competing for the same voters. His decision in mid-March to step down after the local elections, to make way for a new generation, simultaneously enabled the party to talk about renewal and a fresh start but strategically was designed to offset the novelty of its new centrist opponent. And with Change UK not standing in the local elections and Brexit dominating political discourse, the door seemed ajar for the Liberal Democrats to make a political statement. Over the next month, the Liberal Democrats not only gained significant electoral impetus but also they mortally damaged their Centrist adversaries. First, through local election success—704 net gains and 12 councils—the party won 16% of seats up for election, with 1,351 councillors returned and now controlled 23 councils, its highest number since 2010. With visibility increasing, a seemingly decisive shift in the national polls occurred. The Liberal Democrats immediately sought to build on this through the launch of their European election manifesto and campaign slogan, ‘Bollocks to Brexit’, which generated even more media exposure. Like the local elections, it was the Liberal Democrats that cannibalised the pro-Remain vote in the European elections, winning 16 seats, its largest number since 1979. With 3.4 million votes and 20% of the national vote, the party topped the poll in London and was the largest party in 44 local areas, 29 of which were in the capital and the South East. Cable passed the baton onto Swinson with the Liberal Democrats seemingly back in the electoral fight. 2. Swinson: shifting position Emboldened by the legitimacy given to it from pro-Remain supporters in local and European elections, a record membership of 120,000 and the subsequent considerable uptick in national polls, the Liberal Democrats became increasingly visible and garnered far more national media attention. Swinson hit the ground running with success in the Brecon and Radnorshire by-election while defections from Change UK and, in time, from three Conservative MPs who crossed the floor of the House also provided publicity and a legitimacy boost. From the early days of Swinson’s leadership, there were two distinct shifts from the Cable regime. On becoming leader, Swinson unequivocally set out her stall to ‘Stop Brexit’ and remove any ambiguity about the Liberal Democrats’ pro-Remain credentials. Part of the reasoning was to exploit the uncertainty around Labour’s ‘renegotiation strategy’ position and their shift towards a ‘second referendum’ position with strings attached. As a consequence, the Liberal Democrats signalled to Remain voters that they would revoke Article 50 if they won outright power and would revert to supporting a second referendum with a Remain option if they fell short. The Liberal Democrats also adopted an unambiguous equidistance stance to lure recruits from both Labour and the Conservatives by ruling out supporting a Johnson- or Corbyn-led administration. Compared to Cable, Swinson also ‘ratcheted up’ the anti-Corbyn tone and rhetoric. She consistently dismissed a Corbyn-led caretaker government to avoid a no-deal Brexit and instead put forward other alternative caretaker PMs. Eyeing disenchanted Labour voters, Swinson went beyond Corbyn’s reluctance to back a second referendum. She also signposted the Labour leadership’s handling of anti-Semitism and its damaging economic policies, while portraying Corbyn as a threat to national security, using examples of his response to the Salisbury poisoning incident and support for authoritarian regimes. Both the Revoke policy and the rigid equidistance stance represented political gambles. Each risked alienating potential Conservative and Labour tactical switchers. They also jeopardised message clarity and handed their rivals potential attack lines warning that supporting the Liberal Democrats could let in their opponent by mistake. With both leaders and parties not overwhelmingly popular in their own right, the Liberal Democrats began to project Swinson as a competent, decisive leader who owned policy positions like Revoke and was not afraid to make tough choices. When the Labour leadership bowed to pressure from Swinson and the SNP for an early general election, it was clear that the party would go ‘all in’ on this one-club electoral strategy. 3. The political and electoral strategy After their 2017 general election performance, the Liberal Democrats still needed to climb an electoral mountain even to get back to the point prior to entering coalition with the Conservatives. There were 17 seats where the party needed a swing of 10% or less, 12 of which were won by the Conservatives in 2017. On a swing of 20% or less, this figure increased to 52 seats with 40 held by the Conservatives, 7 by Labour and 5 by nationalist parties in Scotland and Wales. Of these 35 most marginal, only Dorset West and St Albans had not been held by the Liberal Democrats between 2001 and 2010. Despite evidence of growing individual switching, for any Liberal Democrat revival to occur, it was highly likely it would be in the places where local credibility remained a factor. But there were other problems to overcome. As Figure 6.1 shows, only 16 of the 52 seats where the Liberal Democrats required a swing of 20% or less had a Remain vote above 55%. Nine of these 16 seats were held by Labour or the SNP. Yet, 22 of these seats had a majority for Leave, with 12 recording a Leave vote of 55% or more. Ten of these 12 were held by the Conservatives. If the Liberal Democrats were to make an electoral breakthrough, they needed the support of both a highly efficient tactical ‘Remain Alliance’ and disgruntled moderate Leave voters. This was a monumental challenge. Figure 6.1 Open in new tabDownload slide Liberal Democrat 2019 possible constituency targets (swing 20% or less) by 2016 % Remain vote Figure 6.1 Open in new tabDownload slide Liberal Democrat 2019 possible constituency targets (swing 20% or less) by 2016 % Remain vote To overcome the millstone of credibility, part of the electoral strategy involved trying to persuade switching MPs to remain in their incumbent seats and fight on the Liberal Democrat badge. Two of its new MPs (Sandbach and Wollaston) fought where they were originally elected. Other high-profile defectors were parachuted into Remain seats where party polling suggested they had a viable chance of winning or causing an upset. Three of these MPs (Berger, Gyimah and Ummuna) were strategically placed into strong London Remain seats, while Smith and Lee moved to Remain Altrincham and Sale West and Wokingham, respectively. The party sought to offset credibility concerns by re-selecting either former MPs or those who stood in key target seats last time out. It hoped that the personal votes of former MPs would simultaneously nullify any incumbent advantage and act as a focal point for tactical pro-Remain switching in these seats. The Liberal Democrats also relied on their trusted local base to help spring a surprise. Recent local gains against the Conservatives in North Devon and Winchester provided some hope that this would translate to the parliamentary level. And the party agreed a Unite to Remain electoral pact with the Greens and Plaid Cymru in 60 seats where only one of these party’s candidates would stand to maximize the chances of getting MPs who opposed Brexit elected. With a free-run in 43 of these seats, the party hoped it would boost its electoral chances in key targets. A key plank of the party’s electoral strategy was to minimize the potential squeezing from its rivals. Evidence from the British Election Study (Fieldhouse et al., 2019) suggested that the Liberal Democrats drew 49% of its vote from Labour and 20% from Conservative 2017 voters in the European elections. Holding onto these voters who switched was vital. Again, credibility was critical, given that in many potential seats the Liberal Democrats were third or a distant second. To offset this, the party deliberately avoided mentioning the 2017 constituency result and used more favourable local polling, evidence from Remain driven tactical voting sites and European election data. This led to some embarrassing enquiries about the creative use of data and dubious bar charts in some Liberal Democrat literature. The dual credibility goal was to ensure that the Liberal Democrats were regarded as the main Remain party and that they not Labour (in most cases) were best placed to stop the Conservatives winning a majority. The biggest political and electoral gamble was the party’s Revoke policy. On the face of it, the strategy did have some merit. Senior party strategists were concerned about the fragility of the Liberal Democrat vote—a post-European election poll found that only 31% of Liberal Democrat voters would definitely continue to support the party at the general election whereas 29% may change their mind and vote for someone else. Without a distinctive platform, the fear was that they risked being swallowed up by one of its main rivals. With Labour now offering another referendum, the Revoke position meant the Liberal Democrats could continue to present itself as the true party of Remain. Given its weak partisan base and desire to build support in Labour seats, shoring up its vote through this Remain narrative seemed a credible option. Leaving aside the judgment on whether the party’s new policy was sufficiently liberal or democratic, it was a clear risk nevertheless. Only 54% of those who voted Liberal Democrat in June 2019 either approved or strongly approved of cancelling Brexit.1 While the policy appealed to voters in much of London and its wealthy suburban belt, outside of four or five London seats, it still needed to convert Labour and Green voters en masse to translate votes into seats. There was also a worry that a policy of revoking Article 50 could seriously backfire in Conservative battlegrounds. At the individual level, it was inconceivable that such a policy would appeal to Leavers while the Conservative Remain vote could also be turned off by such as hard-line policy. At the aggregate level, many of its previous strongholds in the South West of England were strong Leave areas. To win back these seats and others elsewhere, the Liberal Democrats were now reliant on the Remain vote mobilizing behind them and the Brexit party splitting the Leave vote. In Scotland, the Liberal Democrats had to ‘out-Remain’ the Remain SNP in a country where growing support for the SNP’s IndyRef2 platform was likely to split the Remain vote and lead to the party battling other parties for the Remain plus Union voter (see the Mitchell and Henderson contribution to this volume). Away from Brexit, it would still need to appease core Labour voters in any tactical coalition, given the Liberal Democrats austerity record. For all the talk of individual volatility, it still looked electorally complicated for the Liberal Democrats when the anchor of credibility was not present (Cutts and Russell, 2018). 4. The electoral outcome The 2019 general election saw the Conservatives gain 43.6% of the UK vote and win an 80-seat majority with 48 net gains. Labour saw their vote plummet by almost 8 percentage points to 32.1% and 60 seat losses. While the combined Conservative and Labour vote share did not reach the heights of 2017, more than 75% of voters still supported the two main parties. With the SNP also gaining ground in Scotland, internal party fears that the Liberal Democrat vote could be squeezed during the election campaign became reality. Nonetheless, the party did poll nearly 3.7 million votes, 1.31 million more than in 2017 despite the 2-point decline in turnout and increased its vote share by 4.2 percentage points to 11.5% (Table 6.1). The improvement was steady but not the spectacular increase hoped for prior to the campaign. To put this into perspective, the Liberal Democrat vote was higher than in 2015 and 2017 but half of what it achieved nine years ago pre-coalition and lower than at any election between 1974 and 2010. Two years previously, the party lost support but achieved a net gain of four seats due to effective targeting. In 2019, support increased in 574 of the 611 constituencies where the Liberal Democrats stood candidates. Yet the party suffered a net loss of one seat, gaining three seats and losing four, and the number of Liberal Democrat MPs in Westminster fell to11. Table 6.1 Summary of Liberal Democrat general election performance, 1992–2019 LD . 1992 . 1997 . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Votes (000) 5999 5243 4814 5985 6836 2416 2372 3696 UK vote (%) 17.8 16.8 18.3 22.0 23.0 7.9 7.4 11.5 Seats won 20 46 52 62 57 8 12 11 Seats won (%) 3.2 7.0 7.9 9.6 8.8 1.2 1.8 1.7 Votes: seatsa 1.12 2.74 2.84 2.82 2.48 1.01 1.62 0.96 Lost deposits 11/632 13/639 1/639 1/626 0/631 341/631 375/629 136/611 LD . 1992 . 1997 . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Votes (000) 5999 5243 4814 5985 6836 2416 2372 3696 UK vote (%) 17.8 16.8 18.3 22.0 23.0 7.9 7.4 11.5 Seats won 20 46 52 62 57 8 12 11 Seats won (%) 3.2 7.0 7.9 9.6 8.8 1.2 1.8 1.7 Votes: seatsa 1.12 2.74 2.84 2.82 2.48 1.01 1.62 0.96 Lost deposits 11/632 13/639 1/639 1/626 0/631 341/631 375/629 136/611 Note: These are UK-wide vote share percentages so differ slightly from the GB-only figures reported by David Denver in ‘Results’. a Votes: Seats ratio derived from dividing LD seats won by LD share of the vote. In 1992, the Liberal Democrats stood in 632 constituencies; and in 2017, they stood in 629. In 2019, the party only stood in 611 seats following their pact with Plaid Cymru and the Greens. Open in new tab Table 6.1 Summary of Liberal Democrat general election performance, 1992–2019 LD . 1992 . 1997 . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Votes (000) 5999 5243 4814 5985 6836 2416 2372 3696 UK vote (%) 17.8 16.8 18.3 22.0 23.0 7.9 7.4 11.5 Seats won 20 46 52 62 57 8 12 11 Seats won (%) 3.2 7.0 7.9 9.6 8.8 1.2 1.8 1.7 Votes: seatsa 1.12 2.74 2.84 2.82 2.48 1.01 1.62 0.96 Lost deposits 11/632 13/639 1/639 1/626 0/631 341/631 375/629 136/611 LD . 1992 . 1997 . 2001 . 2005 . 2010 . 2015 . 2017 . 2019 . Votes (000) 5999 5243 4814 5985 6836 2416 2372 3696 UK vote (%) 17.8 16.8 18.3 22.0 23.0 7.9 7.4 11.5 Seats won 20 46 52 62 57 8 12 11 Seats won (%) 3.2 7.0 7.9 9.6 8.8 1.2 1.8 1.7 Votes: seatsa 1.12 2.74 2.84 2.82 2.48 1.01 1.62 0.96 Lost deposits 11/632 13/639 1/639 1/626 0/631 341/631 375/629 136/611 Note: These are UK-wide vote share percentages so differ slightly from the GB-only figures reported by David Denver in ‘Results’. a Votes: Seats ratio derived from dividing LD seats won by LD share of the vote. In 1992, the Liberal Democrats stood in 632 constituencies; and in 2017, they stood in 629. In 2019, the party only stood in 611 seats following their pact with Plaid Cymru and the Greens. Open in new tab Table 6.2 Liberal Democrat performance, 2019 general election: national and regional breakdown National and Regional . 2019 % LD Vote . 2017 % LD Vote . Change 2017–19 . Seats 2019 . Seats 2017 . Change 2017–19 . Country  UK 11.5 7.4 +4.2 11/632 12/632 −1  England 12.4 7.8 +4.6 7/533 8/533 −1  Scotland 9.5 6.8 +2.8 4/59 4/59 0  Wales 6.0 4.5 +1.5 0/40 0/40 0 Region  East Midlands 7.8 4.3 +3.5 0/46 0/46 0  Eastern 13.4 7.9 +5.5 1/58 1/58 0  London 14.9 8.8 +6.1 3/73 3/73 0  North East 6.8 4.6 +2.3 0/29 0/29 0  North West 7.9 5.4 +2.5 1/75 1/75 0  South East 18.2 10.5 +7.7 1/84 2/84 −1  South West 18.2 14.9 +3.2 1/55 1/55 0  West Midlands 7.9 4.4 +3.5 0/59 0/59 0  Yorkshire and the Humber 8.1 5.0 +3.1 0/54 0/54 0 National and Regional . 2019 % LD Vote . 2017 % LD Vote . Change 2017–19 . Seats 2019 . Seats 2017 . Change 2017–19 . Country  UK 11.5 7.4 +4.2 11/632 12/632 −1  England 12.4 7.8 +4.6 7/533 8/533 −1  Scotland 9.5 6.8 +2.8 4/59 4/59 0  Wales 6.0 4.5 +1.5 0/40 0/40 0 Region  East Midlands 7.8 4.3 +3.5 0/46 0/46 0  Eastern 13.4 7.9 +5.5 1/58 1/58 0  London 14.9 8.8 +6.1 3/73 3/73 0  North East 6.8 4.6 +2.3 0/29 0/29 0  North West 7.9 5.4 +2.5 1/75 1/75 0  South East 18.2 10.5 +7.7 1/84 2/84 −1  South West 18.2 14.9 +3.2 1/55 1/55 0  West Midlands 7.9 4.4 +3.5 0/59 0/59 0  Yorkshire and the Humber 8.1 5.0 +3.1 0/54 0/54 0 Open in new tab Table 6.2 Liberal Democrat performance, 2019 general election: national and regional breakdown National and Regional . 2019 % LD Vote . 2017 % LD Vote . Change 2017–19 . Seats 2019 . Seats 2017 . Change 2017–19 . Country  UK 11.5 7.4 +4.2 11/632 12/632 −1  England 12.4 7.8 +4.6 7/533 8/533 −1  Scotland 9.5 6.8 +2.8 4/59 4/59 0  Wales 6.0 4.5 +1.5 0/40 0/40 0 Region  East Midlands 7.8 4.3 +3.5 0/46 0/46 0  Eastern 13.4 7.9 +5.5 1/58 1/58 0  London 14.9 8.8 +6.1 3/73 3/73 0  North East 6.8 4.6 +2.3 0/29 0/29 0  North West 7.9 5.4 +2.5 1/75 1/75 0  South East 18.2 10.5 +7.7 1/84 2/84 −1  South West 18.2 14.9 +3.2 1/55 1/55 0  West Midlands 7.9 4.4 +3.5 0/59 0/59 0  Yorkshire and the Humber 8.1 5.0 +3.1 0/54 0/54 0 National and Regional . 2019 % LD Vote . 2017 % LD Vote . Change 2017–19 . Seats 2019 . Seats 2017 . Change 2017–19 . Country  UK 11.5 7.4 +4.2 11/632 12/632 −1  England 12.4 7.8 +4.6 7/533 8/533 −1  Scotland 9.5 6.8 +2.8 4/59 4/59 0  Wales 6.0 4.5 +1.5 0/40 0/40 0 Region  East Midlands 7.8 4.3 +3.5 0/46 0/46 0  Eastern 13.4 7.9 +5.5 1/58 1/58 0  London 14.9 8.8 +6.1 3/73 3/73 0  North East 6.8 4.6 +2.3 0/29 0/29 0  North West 7.9 5.4 +2.5 1/75 1/75 0  South East 18.2 10.5 +7.7 1/84 2/84 −1  South West 18.2 14.9 +3.2 1/55 1/55 0  West Midlands 7.9 4.4 +3.5 0/59 0/59 0  Yorkshire and the Humber 8.1 5.0 +3.1 0/54 0/54 0 Open in new tab For the second time in four years, Swinson experienced defeat in her East Dunbartonshire seat. In 2015, her vote held up relatively well when other Liberal Democrat incumbents suffered dramatic drops in support. This time, with her as party leader, the effect was hugely symbolic and catastrophic. The SNP was extremely effective at squeezing the Labour vote and building an anti-Swinson alliance to secure victory. The loss of Stephen Lloyd in Eastbourne was less surprising and less high profile, while defeat for Tom Brake in Carshalton and Wallington appeared to be a tactical blunder. Internal Liberal Democrat analysis suggested that Brake would hold the seat and central office then deployed valuable resources to nearby Wimbledon where the party was challenging the Conservatives. Warning signs emerged on polling day as efforts were made to save Brake but to no avail. Elsewhere, with Norman Lamb not standing in Norfolk North, the Liberal Democrat vote plunged by more than 18 percentage points as the Conservatives regained the seat. The Liberal Democrats did gain North East Fife from the SNP helped by pro-Union Conservative tactical switching and narrowly held onto Caithness, Sutherland and Easter Ross and Edinburgh West. Across England, the two gains in ultra-Remain Richmond Park and St Albans were secured with comfortable majorities with both Liberal Democrat candidates polling more than 50% of the vote. In 6 of the 11 seats won, the party gained more than half of the votes cast. Elsewhere the party fell agonisingly short in four seats: Cheltenham, Sheffield Hallam, Wimbledon and Winchester. Wimbledon was one of seven seats (excluding Buckingham) where the party increased their vote by more than 20 percentage points but failed to win the seat. Despite some strong performances, all the Labour and Conservatives defectors who stood as Liberal Democrats in 2019 fell short. Perhaps the worst kept secret of the Liberal Democrat campaign was the big effort in Esher and Walton to unseat the incumbent, prominent Conservative Brexiteer Dominic Raab but despite increasing their vote by almost 28 points the party narrowly failed. Two years ago, the Liberal Democrats polled more than 30% of the vote in 28 constituencies, lost 375 deposits and only came second in 38 seats. In 2019, the party did advance, winning more than 30% of the vote in 51 seats and losing deposits in only 138 constituencies. They are now second in 91 seats, 80 of which where the Conservatives are the incumbent. Of those second places, the Liberal Democrats are now 10% behind the winning party in 15 seats compared to 9 two years ago. Ten seats are ultra-marginal (5% or less behind the incumbent) with eight of these held by the Conservatives. Optimists might claim the Liberal Democrats have the platform to make serious electoral inroads into the Conservatives next time. At the national and regional levels, the Liberal Democrat vote remains uneven and the post 2010 north-south divide in party support is widening (Table 6.2). Despite holding four seats in Scotland, its vote continued to be squeezed in the face of SNP resurgence and pro-Union supporters opting for the Conservatives unless the Liberal Democrats were better placed. Liberal Democrat support rose by less than 3 points in Scotland but it fared far worse in Wales. Even accounting for the Unite to Remain pact, the party lost Brecon and Radnorshire, 16 deposits in the 32 seats they stood and only managed to increase their vote by 1.5 percentage points across the country. The Liberal Democrats gained support across all regions in England, with the largest growth in London and the South East. Of the 29 seats where the party is now fewer than 10 percentage points behind the incumbent, 14 are in London and the South East (and only eight of the party’s 138 lost deposits nationally were suffered in these areas). The Liberal Democrats continue to poll relatively strongly in its area of traditional strength, notably the South West with 11 of the 55 seats in the region recording Liberal Democrat vote shares of 30% or more. However, as in 2017, party support continues to bottom-out with increases below the national average. Aside from regaining Bath in 2017, they failed to recover lost ground elsewhere and the task seems to be getting more difficult. Only 5 of the 29 seats where the Liberal Democrats are 10% behind the incumbent now are in the South West with the Conservatives retaining huge majorities in previous Liberal Democrat strongholds. Liberal Democrat support continues to hold up in eastern England and remains above the national average but elsewhere the picture looks bleak. Across the North of England and the Midlands, party support is anything from 3.4% to as much as 4.7% below the national vote share. In 2017, the difference between the five northern and midland regions and the four southern regions of England was roughly 6.5 points. By 2019, this gap had grown to 9 percentage points. To reiterate the growing divide, around 87% of lost party deposits in England were in the North and Midlands. Forty per cent of all UK wide lost party deposits were in the North West and Yorkshire and the Humber. The Liberal Democrats failed to record any constituency vote shares above 30% across the whole of the Midlands. Only 4 of the 29 seats where the party was second and less than 10% behind the winner in 2019 are in the North of England. There remains a clear geographic divide in Liberal Democrat representation. For the second successive election, Westmoreland and Lonsdale is the only northern English Liberal Democrat seat. 5. The constituency battleground Table 6.3 examines the Liberal Democrats’ 2019 performance by seat type. In 2015 and 2017, while incumbency mattered, those seeking re-election were not guaranteed immunity from any national surge because of their personal standing in the seat (Cutts and Russell, 2017). The loss of Swinson and Brake reinforces this. Nonetheless, where a Liberal Democrat incumbent stood again, support rose by a modest 2 percentage points, lower than the national increase in party support and roughly half that achieved by incumbent candidates in 2017. For Liberal Democrat candidates in a number of incumbent seats, support had begun to reach its ceiling. Moreover, the party was far less defensive than two years ago targeting many more seats including some where the Liberal Democrats were languishing in third place. In non-held seats, support increased by more than 4 points. Evidence suggests that the floor of the Liberal Democrat vote (the 375 seats where the party lost its deposit in 2017) rose by 3.1 percentage points. Yet closer inspection suggests that, while being in second place mattered, the role of political agency continues to be more important. Table 6.3 Liberal Democrat performance by incumbency and seat type, 2019 general election Seats . 2019 % LD Vote . 2017 % LD Vote . Change ±17–19 . Incumbency  LD 2017 incumbent seats (12) 44.6 44.2 +0.4  LD incumbent candidates (10) 44.9 42.9 +2.0  LD non-held seats (599) 10.9 6.6 +4.3 Seat type (LDs second place)  LD all second place (38) 33.1 29.1 +4.0  Con-LD seats (29) 35.4 29.7 +5.7  Lab-LD seats (7) 24.5 26.1 −1.6  SNP/PC-LD seats (2) 30.2 30.9 −0.7 Historical Legacy  LD legacy February 1974 seats (13)a 19.9 21.9 −2.0  LD heartland 1992 seats (18) 26.2 27.0 −0.8  LD breakthrough 1997 seats (28)a 31.8 28.5 +3.3  LD pre-coalition 2010 (56)a 27.6 27.6 0.0 Seats . 2019 % LD Vote . 2017 % LD Vote . Change ±17–19 . Incumbency  LD 2017 incumbent seats (12) 44.6 44.2 +0.4  LD incumbent candidates (10) 44.9 42.9 +2.0  LD non-held seats (599) 10.9 6.6 +4.3 Seat type (LDs second place)  LD all second place (38) 33.1 29.1 +4.0  Con-LD seats (29) 35.4 29.7 +5.7  Lab-LD seats (7) 24.5 26.1 −1.6  SNP/PC-LD seats (2) 30.2 30.9 −0.7 Historical Legacy  LD legacy February 1974 seats (13)a 19.9 21.9 −2.0  LD heartland 1992 seats (18) 26.2 27.0 −0.8  LD breakthrough 1997 seats (28)a 31.8 28.5 +3.3  LD pre-coalition 2010 (56)a 27.6 27.6 0.0 Notes: Percentages derived from summing LD votes cast/Total Valid Votes Cast × 100. The 2017 constituencies exclude Buckingham and Brighton Pavilion as the LDs did not stand a candidate in either constituency. In 2019, the Liberal Democrats stood in 611 seats. a Legacy seats are one less because the party did not stand in Isle of Wight; in 2010 they are one less because the party did not stand in Bristol West. Open in new tab Table 6.3 Liberal Democrat performance by incumbency and seat type, 2019 general election Seats . 2019 % LD Vote . 2017 % LD Vote . Change ±17–19 . Incumbency  LD 2017 incumbent seats (12) 44.6 44.2 +0.4  LD incumbent candidates (10) 44.9 42.9 +2.0  LD non-held seats (599) 10.9 6.6 +4.3 Seat type (LDs second place)  LD all second place (38) 33.1 29.1 +4.0  Con-LD seats (29) 35.4 29.7 +5.7  Lab-LD seats (7) 24.5 26.1 −1.6  SNP/PC-LD seats (2) 30.2 30.9 −0.7 Historical Legacy  LD legacy February 1974 seats (13)a 19.9 21.9 −2.0  LD heartland 1992 seats (18) 26.2 27.0 −0.8  LD breakthrough 1997 seats (28)a 31.8 28.5 +3.3  LD pre-coalition 2010 (56)a 27.6 27.6 0.0 Seats . 2019 % LD Vote . 2017 % LD Vote . Change ±17–19 . Incumbency  LD 2017 incumbent seats (12) 44.6 44.2 +0.4  LD incumbent candidates (10) 44.9 42.9 +2.0  LD non-held seats (599) 10.9 6.6 +4.3 Seat type (LDs second place)  LD all second place (38) 33.1 29.1 +4.0  Con-LD seats (29) 35.4 29.7 +5.7  Lab-LD seats (7) 24.5 26.1 −1.6  SNP/PC-LD seats (2) 30.2 30.9 −0.7 Historical Legacy  LD legacy February 1974 seats (13)a 19.9 21.9 −2.0  LD heartland 1992 seats (18) 26.2 27.0 −0.8  LD breakthrough 1997 seats (28)a 31.8 28.5 +3.3  LD pre-coalition 2010 (56)a 27.6 27.6 0.0 Notes: Percentages derived from summing LD votes cast/Total Valid Votes Cast × 100. The 2017 constituencies exclude Buckingham and Brighton Pavilion as the LDs did not stand a candidate in either constituency. In 2019, the Liberal Democrats stood in 611 seats. a Legacy seats are one less because the party did not stand in Isle of Wight; in 2010 they are one less because the party did not stand in Bristol West. Open in new tab Heading into the 2019 general election, the Liberal Democrats were second in 38 seats, 29 of which were Conservative–Liberal Democrat battlegrounds. Previously, the tactical unwind of the centre-left vote in these battleground seats was a major factor in the party’s collapse. Likewise in 2017, while success against the Conservatives transpired where the Liberal Democrats were able to curb the Labour vote, some tactical unwind occurred, with Labour even relegating the Liberal Democrats to third place in five seats (Cutts and Russell, 2017). In 2019, tactical unwind not only stopped but had reversed, although in modest and uneven fashion. In the 29 Conservative–Liberal Democrat battlegrounds, the party made above-average headway, increasing support by 5.7 points. There is, however, a lot of unevenness, with the party losing ground in five seats and simultaneously recording double-digit increases in another six. In the two seats, the Liberal Democrats won, both the Conservatives and Labour lost support. Across the 29 seats, the Conservatives marginally increased their support but Labour saw their vote drop by 6.6 points, with their vote declining in all but one of these constituencies. The Liberal Democrats clearly began to regain some of the centre-left tactical switchers that had left it after 2010 but its failure to make any deep inroads into the Conservative vote placed a ceiling on its ability to win seats. In the seven seats, they were directly challenging Labour the party simply failed to breakthrough and lost 1.7 percentage points of their vote. Again, Liberal Democrat vote change was uneven with four of the seven seats recording a decline in support. Two recorded little change while the party’s vote rose by nearly 10 points in Hornsey and Wood Green. Labour also experienced a decline in support but only by 2.2 percentage points. Clearly, the failure to secure the marginal seats of Cambridge and Sheffield Hallam and to make further inroads against Labour generally was a considerable setback. The tactic of re-selecting a previous Liberal Democrat MP to stand in the seat they once represented had limited success. Of those, only Sarah Olney in Richmond Park saw a substantial increase in support and actually won back the seat. In two potentially winnable seats—Ceredigion and St Ives—previous MPs lost ground. The Liberal Democrat retreat in their traditional heartlands continued although there are signs that the rate of decline is beginning to slow as support reaches its floor. Of the 14 Liberal constituencies held in February 1974, Orkney and Shetland remained the only seat held. The Liberal Democrats largely stood still in all 1,992 seats despite regaining North East Fife. Many of these seats were the foundation for party growth during 1990s and 2000s but were wiped out during the coalition years. In 2019, the drop in Liberal Democrat voting looks to have bottomed-out but there are few signs of any long-term recovery. On the contrary, for the second successive election, the Liberal Democrats improved their performance in the ‘breakthrough’ seats which the party won in 1997 at the peak of the anti-Conservative tactical alliance. They recorded on average more than 30% of the vote in these predominantly Conservative-leaning constituencies and increased their support overall by more than 3 points. Across the 56 seats that the party won in 2010 and stood a candidate in 2019, there was little evidence of any resurgence. This is mainly because any growth where the party was fighting the Conservatives was largely offset by more challenging conditions in seats where Labour and the SNP were the primary competitors. Despite the churn, 8 of the 11 seats won in 2019 were held by the party in 2010; 5 were first gained in 1997; 4 of the 11 seats also elected Liberal Democrat representatives 27 years ago. The story, therefore, remains similar to two years ago: while the party has failed to recover in its traditional heartland areas, the Liberal Democrats historical legacy still remains vital to both its support and Westminster representation. 6. Brexit and the Liberal Democrat vote After the 2019 election, only 1 of the 11 seats represented by Liberal Democrat MPs voted Leave in the 2016 EU Referendum. In 6 of the other 10 seats it now holds, support for Remain in the referendum exceeded 60%. Three of the four losses were Leave seats. In 2017, there was a positive relationship (correlation of 0.15* significant at 99% level) between Liberal Democrat constituency vote change and Remain vote, albeit the line was relatively flat. Two years later, the relationship is slightly stronger (0.24* significant at 99% level) although there is some indication that the Liberal Democrat vote increased more in softer Remain seats as well as in those places that strongly voted Remain (see Figure 6.2). Figure 6.2 Open in new tabDownload slide Liberal Democrat vote change 2015–2017 and 2017–2019, by % Remain vote Figure 6.2 Open in new tabDownload slide Liberal Democrat vote change 2015–2017 and 2017–2019, by % Remain vote Overall, party support was on average more than 6 percentage points higher in Remain areas than in Leave seats and around 4 percentage points higher than the party’s national vote share. When broken down into ‘soft’ (50–59.9%) and ‘hard’ (with a 60%+ vote) categorisations, it is evident that the party’s vote was 5.5 points lower in ‘hard Brexit’ than ‘soft Brexit’ areas but both saw increases in Liberal Democrat support (see Table 6.4). Many of these included seats where Labour was fighting off a Conservative surge. Even on this limited evidence (and acknowledging ecological fallacy issues), it is probable that while Remain voters did not abandon Labour for the Liberal Democrats in droves, the party’s ability to lift the floor of its vote harmed Labour. The party increased their vote share, on average, by 5.6% across Remain constituencies but the growth was nearly 1 percentage point stronger in ‘soft’ than ‘hard Remain’ seats. In a sizeable minority of these ‘hard Brexit’ seats, the SNP was rampant which restricted growth. Many others were safe Labour seats of which more than 30 were in London. The party simply lacked the longstanding credibility in these seats as a viable voting option so any possible surge in support was always going to have a ceiling. Table 6.4 Liberal Democrat performance in Remain and Leave seats, 2019 general election Seats . 2019 % LD Vote . Change ±17–19 . Remain/Leave  All Leave seats (381) 9.2 +3.3  All Remain seats (230) 15.4 +5.6 Leave  ‘Soft Brexit’ seats 50.1–59.9% (231) 11.0 +3.8  ‘Hard Brexit’ seats 60%+ (150) 6.5 +2.6 Remain  ‘Soft Remain’ seats 50.1–59.9% (145) 15.6 +5.9  ‘Hard Remain’ seats 60%+ (85) 15.0 +5.0 Seats . 2019 % LD Vote . Change ±17–19 . Remain/Leave  All Leave seats (381) 9.2 +3.3  All Remain seats (230) 15.4 +5.6 Leave  ‘Soft Brexit’ seats 50.1–59.9% (231) 11.0 +3.8  ‘Hard Brexit’ seats 60%+ (150) 6.5 +2.6 Remain  ‘Soft Remain’ seats 50.1–59.9% (145) 15.6 +5.9  ‘Hard Remain’ seats 60%+ (85) 15.0 +5.0 Note: 611 seats where the Liberal Democrats stood candidates. Open in new tab Table 6.4 Liberal Democrat performance in Remain and Leave seats, 2019 general election Seats . 2019 % LD Vote . Change ±17–19 . Remain/Leave  All Leave seats (381) 9.2 +3.3  All Remain seats (230) 15.4 +5.6 Leave  ‘Soft Brexit’ seats 50.1–59.9% (231) 11.0 +3.8  ‘Hard Brexit’ seats 60%+ (150) 6.5 +2.6 Remain  ‘Soft Remain’ seats 50.1–59.9% (145) 15.6 +5.9  ‘Hard Remain’ seats 60%+ (85) 15.0 +5.0 Seats . 2019 % LD Vote . Change ±17–19 . Remain/Leave  All Leave seats (381) 9.2 +3.3  All Remain seats (230) 15.4 +5.6 Leave  ‘Soft Brexit’ seats 50.1–59.9% (231) 11.0 +3.8  ‘Hard Brexit’ seats 60%+ (150) 6.5 +2.6 Remain  ‘Soft Remain’ seats 50.1–59.9% (145) 15.6 +5.9  ‘Hard Remain’ seats 60%+ (85) 15.0 +5.0 Note: 611 seats where the Liberal Democrats stood candidates. Open in new tab Figure 6.3 provides further insight into how Brexit shaped Liberal Democrat support in the 2019 election against their main rivals. Here, we mark England and Wales constituencies as to whether they backed Remain (hollow triangle, where support for Leave <45%); were comparatively evenly balanced (hollow square, where support for Leave ≥45% and ≤55%) or whether they backed Leave (black circle, where support for Leave >55%). For the Liberal Democrats, there is a significant negative effect for Conservative and Labour vote change on Liberal Democrat vote change, suggesting that Liberal Democrat support held up better in seats where their rivals did not make as much ground. In 2019, we can see how the Liberal Democrats made modest gains in Leave and largely working-class seats from an extremely low base predominantly where the Conservatives gained moderate support and Labour collapsed. In Remain and largely middle-class seats, there are places where the Liberal Democrats saw their vote surge while Labour fell away but there are also a number of strong Remain areas where the Labour vote held up. Liberal Democrat gains and Conservative losses are predominantly in the evenly balanced Brexit areas and Remain seats notwithstanding a small cluster of the latter where the party made little progress. While this did not translate into Liberal Democrat seats in 2019, it does illustrate Britain’s changing electoral map where previous Conservative middle class, largely Remain areas might be potentially vulnerable to the Liberal Democrats in future elections. Figure 6.3 Open in new tabDownload slide How Brexit shaped Liberal Democrat gains and losses, 2019 general election Figure 6.3 Open in new tabDownload slide How Brexit shaped Liberal Democrat gains and losses, 2019 general election 7. Why did it go wrong for the Liberal Democrats? 7.1 Jo Swinson’s disastrous campaign Leaders are critical nowadays in British elections but they are especially vital for third parties. Not only do they enable them to reach out to the electorate but they provide a vehicle to enhance electoral credibility and create goodwill towards the party brand. Traditionally, the Liberal Democrats always did better when they were given more exposure with the popularity of the leader a key determinant in the party’s success. With Johnson marmite for many and Corbyn unpalatable for more, the Liberal Democrats sensed an opportunity. Notwithstanding internal party enthusiasm for Swinson, many saw her relative obscurity both an advantage and a vehicle through which the Liberal Democrats could capture votes from disenchanted voters. As a consequence, the party’s campaign, branding and literature were built around her. From the orange battlebus emblazoned with her photo and the phrase ‘Jo Swinson’s Liberal Democrats’ to personalising high-profile policy announcements on free childcare for those aged two to four, the ‘presidential style’ campaign was all about selling Swinson the person to the public at large. Far from being an electoral asset, Swinson quickly became a liability. At the party’s formal election campaign launch, Swinson insisted she was running to become Prime Minister. The comment haunted her throughout the campaign with many considering it arrogant and unrealistic. As the party’s polling began to slide, Swinson resolutely kept talking up her Prime Ministerial chances, which served to make her sound fanciful and out of touch. This was largely of the party’s own making but there was also a bind the Liberal Democrats now consistently find themselves in, since their time in coalition. The party ruled out working with a Johnson-led Conservative administration and a Labour one led by Corbyn but did not explicitly dismiss other alternatives. While this permitted some deflection on questions of potential coalition building, it meant that the party needed to persuade voters to support it in their own right and selling Swinson as a potential, credible Prime Minister was an important plank in this strategy. Crucially, Swinson’s approval ratings barely got above 25% throughout her period as leader. In the early days, many polled could not make a judgement but as the public got to know and see Swinson more, the more unimpressed they became. Whereas Swinson’s net approval rating was below Corbyn, ultimately, only 19% of the public approved of Swinson, a figure below even 21% for the Labour leader. Swinson simply failed to cut through. Swinson’s inability to free herself from the shackles of the Conservative–Liberal Democrat 2010–2015 coalition also damaged the party’s appeal. Swinson’s own track record in the austerity government dogged her throughout the campaign. Not only she was the subject of a concerted online campaign from Labour supporters and activists, she faced anger from voters in person. During her BBC Question Time appearance, Swinson came under repeated attack from younger audience members for supporting austerity and was forced to say ‘sorry that we did not win more of those fights in coalition’ (Independent, 2019a). Swinson was also confronted by a student in Glasgow who blamed her and the Liberal Democrats for enabling austerity cuts and then later in the campaign by a protestor in Streatham over the effect of the coalition government's policies on young people. There was also a sense that the apology for austerity cuts became more sincere as the campaign progressed. Early on, Swinson not only spoke about being upfront about the Liberal Democrats failings in coalition but also defended their role in government pointing to policy successes such as same-sex marriage and taking lower incomes out of paying tax. A few weeks later the tone changed. The week before polling day, Swinson apologised for voting for austerity cuts, supporting policies such as the ‘bedroom tax’ and admitted that austerity went too far (Independent, 2019b). Austerity remained a millstone around the party’s neck. Among key voters that the party needed onside, austerity still mattered and Swinson represented the era when the Liberal Democrats were the poster boys of economic cuts. If the Liberal Democrats wanted to draw a line under this period, Swinson was emblematic of why this could not happen. Swinson’s exclusion from the first televised head-to-head election debate was a damaging blow. It might have provided a critical space for the Liberal Democrat leader’s credibility, just as it did for Nick Clegg in 2010. A three-way leaders’ debate could have given the party and Swinson equal footing with her main rivals on a national stage. The vigour with which the party attacked and contested the decision suggested that they thought it might be a pivotal moment. Given the ‘presidential nature’ of the Liberal Democrat campaign, it represented a prize opportunity for Swinson to present herself as the moderate, progressive leader reaching out to discontented voters. Nevertheless, in different TV debate formats, it was not as if Swinson ‘stole the show’, so it is not guaranteed that Swinson and the Liberal Democrats would have gained a huge advantage from taking part. 7.2 Party strategy Aside from adopting a backfiring ‘Presidential style’ campaign, the biggest strategic error was the policy to revoke Article 50. Throughout the campaign, it came under sustained attack from both the main parties. On the one hand, it was counterproductive because it could only be implemented if the party won a majority which was highly unlikely. It also muddied the Liberal Democrats’ Brexit message which was previously simple and clear. It both hurt and contradicted the moderate, pragmatic appeal of the party’s image to voters. Opponents questioned whether winning a majority at Westminster under first-past-the-post constituted an unequivocal mandate. For the Liberal Democrats, a party that had long supported proportional representation recognizing such a mandate seemed inconsistent and in the long run an act of self-harm. Polling two weeks before election day suggested that 28% of the public supported the Liberal Democrats’ position of revoking Article 50 and stopping Brexit completely (YouGov, 2019). Yet, only 50% of Remain voters backed it while 35% opposed. More than four-fifths of Leave voters unsurprisingly did not support the policy. The policy was clearly polarising and caused a great deal of public resentment. Moreover, there was barely any support among Conservative voters, with only 3% in agreement and 90% opposed. The policy of revoking Article 50 arguably placed a ceiling on party support and hampered efforts to persuade large numbers of Conservative moderates to lend them their vote. To make matters worse, in the final weeks of the campaign, the Liberal Democrats began to row back and returned to what Layla Moran called ‘Plan A’ of a second referendum. This shifted position reaffirmed that it ‘bombed’ with those voters the party wanted to win over. Like previous efforts to be equidistant, such a stance simply failed to reap electoral rewards. While criticisms of Boris Johnson and his Brexit policy was wholly unsurprising, the acerbic tone levelled at Corbyn and Labour was strategically a little more difficult to fathom. Early on it appeared that the party saw an electoral opening. By winning over Labour Remain voters and moderates from other parties, the Liberal Democrats could do irreparable damage to Labour. Yet while Corbyn was clearly unpopular, it became abundantly evident as the campaign progressed that he was the only leading politician—and Labour the only party—that could achieve what the Liberal Democrats wanted—to stop Brexit. Despite this, the attacks continued diluting the party’s supposedly unequivocal anti-Brexit stance. This had two damaging knock-on effects. Firstly, many Labour Remainers simply saw through Liberal Democrat attempts to position themselves in key seats as the only viable Remain option. As Lord Ashcroft’s (2019) post-election polling shows, 84% of 2017 Labour Remainers stayed with Labour. Secondly, for Remain Conservatives, already nervous about a Corbyn-led Labour government, constant confirmation from the Liberal Democrats about how bad it would be scared them off from switching. When the opposition is unpopular, it makes it difficult for a third party to win seats from the incumbent. Further fuelling that unpopularity and supporting an extreme policy which could lead to resentment from voters in those key seats it is trying to attract, made little strategic sense and unsurprisingly failed. On the face of it, cooperation with Labour together with a carefully thought out political strategy to get a ‘people’s vote’ with Remain on the ballot through a Labour-led administration might have changed electoral dynamics. Simply put, equidistance gave the party a high-risk shot at a ‘home run’ return to the Kennedy days of Westminster representation, but ultimately it was somewhat easily and predictably ‘struck out’ by far cannier rivals. 7.3 The campaign The party went into the election with a highly ambitious 80-seat target strategy as internal strategists believed that the electoral momentum was with them. It staked that volatility would triumph over local credibility with scores of voters abandoning place-based political loyalties as they switched to them. Yet, despite the odd surge and the leapfrogging of Labour in some seats, longstanding credibility once again proved far more vital. As the campaign went on, the target list was amended week by week based on internal polling and local canvass returns with layers of key seats withdrawn central support. Come polling day, only a select few target seats remained, illustrating how hopelessly optimistic the original strategy was. Some of the damage was done early on. Nigel Farage’s decision to remove the Brexit party from standing in Conservative-held seats was out of Liberal Democrat control but might have been anticipated as Farage had come under severe pressure from donors, high-profile supporters and the Brexit-leaning media in advance of the campaign. For the Liberal Democrats, already struggling to unify the Remain vote in marginal battlegrounds, a united Leave vote was far too powerful to overcome. Other problems were of the Liberal Democrats’ own making. In an attempt to offset an absence of local credibility, the Liberal Democrats used European election constituency results, internal polling and external multilevel regression with post-stratification (MRP) polling from tactical voting websites and local constituency polls to tell voters it was them, not Labour, most likely to oust Conservative incumbents. The party’s use of data and badly drawn bar charts angered opposition activists and voters alike and quickly attracted media attention with Swinson having to face down accusations of misleading voters. It became an unwelcome distraction. And if Labour suspicions about the Liberal Democrats pro-Remain credentials at all costs were already somewhat waning, candidate disputes in Canterbury and High Peak added fuel to the anti-Labour conspiracy. While the decision to stand a candidate against pro-Remain Labour MP Rosie Duffield in marginal Canterbury did not stop her winning, her compatriot in High Peak was not so lucky, losing by 600 votes with the new Liberal Democrat candidate decisively winning more than 5% of the vote. Even though there was no formal Remain pact with Labour, it seemed to contradict the Liberal Democrats' widely aired stated ambition of stopping Brexit and opened it up to accusations that its real goal was to damage Labour. Beyond revoking Article 50, the Liberal Democrats’ manifesto remixed and extended some 2017 policies such as 1p in the pound on income tax to raise £7 billion for NHS and free childcare for two to four year olds. Like 2017, there was a renewed attempt to win back centre-left voters who had deserted the party since 2010, with proposals to generate 80% of electricity from renewables, freeze train fares, recruit 20,000 new teachers, lift the minimum wage for those on zero-hour contracts and build 300,000 new homes a year. On the economy though, the Liberal Democrats were more fiscally conservative than their rivals, with stringent borrowing limits. Any additional revenue was due to come from the ‘Brexit bonus’ from remaining in the EU. Although the Liberal Democrats’ efforts to showcase their broader policy portfolio were somewhat overshadowed by ‘Brexit’, longstanding problems on political identity and policy appeal remained. Like 2017, aside from Europe where 11% of respondents named the Liberal Democrats as the best party, the Liberal Democrats were unable to make much headway in challenging their rivals’ ownership of salient issues (see Table 6.5). There were some improvements on Education but apart from Europe all other issues remained in single digits. Simply put, there were few signs that the policies presented in the manifesto translated to the public and little evidence that the party had managed to address longstanding concerns by showcasing a distinctive identity. Table 6.5 Best party on issues: YouGov tracking poll, 25–26 November 2019 (%) Issue . Conservatives . Labour . Lib Dems . Others . None . DK . Health 26 35 6 3 8 21 Immigration 30 18 8 14 7 24 Law and order 36 19 6 5 8 26 Education 26 30 9 3 8 24 Taxation 34 24 7 3 8 24 Unemployment 28 27 5 4 9 27 Economy 37 19 7 4 8 26 Housing 22 32 6 3 9 28 Britain’s EU exit 31 13 11 13 11 20 Defence and security 39 15 5 5 7 30 Issue . Conservatives . Labour . Lib Dems . Others . None . DK . Health 26 35 6 3 8 21 Immigration 30 18 8 14 7 24 Law and order 36 19 6 5 8 26 Education 26 30 9 3 8 24 Taxation 34 24 7 3 8 24 Unemployment 28 27 5 4 9 27 Economy 37 19 7 4 8 26 Housing 22 32 6 3 9 28 Britain’s EU exit 31 13 11 13 11 20 Defence and security 39 15 5 5 7 30 Open in new tab Table 6.5 Best party on issues: YouGov tracking poll, 25–26 November 2019 (%) Issue . Conservatives . Labour . Lib Dems . Others . None . DK . Health 26 35 6 3 8 21 Immigration 30 18 8 14 7 24 Law and order 36 19 6 5 8 26 Education 26 30 9 3 8 24 Taxation 34 24 7 3 8 24 Unemployment 28 27 5 4 9 27 Economy 37 19 7 4 8 26 Housing 22 32 6 3 9 28 Britain’s EU exit 31 13 11 13 11 20 Defence and security 39 15 5 5 7 30 Issue . Conservatives . Labour . Lib Dems . Others . None . DK . Health 26 35 6 3 8 21 Immigration 30 18 8 14 7 24 Law and order 36 19 6 5 8 26 Education 26 30 9 3 8 24 Taxation 34 24 7 3 8 24 Unemployment 28 27 5 4 9 27 Economy 37 19 7 4 8 26 Housing 22 32 6 3 9 28 Britain’s EU exit 31 13 11 13 11 20 Defence and security 39 15 5 5 7 30 Open in new tab 8. Conclusion After being routed in 2015, the Liberal Democrats were on life support, battling for electoral survival. Two years ago, the party had stabilised but was still in a critical condition. In 2019, the Liberal Democrats seemed to be on the precipice of a full recovery but remain in intensive care. Most of the damage was self-inflicted. The campaign proved to be a disaster and much of the political strategy was ill-thought out. Embarking on a ‘Presidential style’ campaign with a leader whose popularity plummeted on close public inspection damaged the party’s ability to reach out to disenchanted voters (and suggests no-one in Liberal Democrat HQ had paid any attention to Theresa May’s disastrous campaign of 2017). Its seemingly kamikaze Revoke policy muddied the Liberal Democrat Brexit message, polarised the public and undermined the party’s moderate-pragmatic brand. The unrelenting personal attacks on Corbyn backfired, as Conservative and Labour voters hardened their positions for different reasons. Tactical mistakes, targeting blunders and data misrepresentation combined to make this a campaign to forget. Swinson’s own defeat in a seat the party was adamant throughout they would defend exemplified the campaign’s shortcomings. However, these failings actually mask some small but important advances. Leaving aside the drop in representation, the party saw an upsurge in support and is in a far healthier position than two years ago. Amidst Britain’s changing electoral geography, it is now in prime position against the Conservatives in a cluster of seats and has an opportunity to develop its local brand in many others. There are also strong signs of growth among longstanding demographics, particularly graduates living in suburban, cosmopolitan areas who in recent elections looked elsewhere. Despite the disappointment, there are reasons to be optimistic. Yet longstanding issues persist. The party’s social and partisan base remains relatively weak and it is still largely reliant on votes that are lent rather than owned. Aside from Europe, the party continues to lack a political identity, focussing instead on quick fix, eye-catching proposals rather than building a long-term policy platform that appeals to those who share wider liberal values. Excluding its high-profile Remain credentials, it remains the case that, outside the most avid of political observers, most voters would be unable to say what the Liberal Democrats stand for, or recollect any of their policies. The association with the Coalition and its austerity policies also continues to damage the party’s brand with key sections of the electorate. After three consecutive election setbacks, now ought to be the time to change direction and adopt a different approach. The Liberal Democrats need to rethink their political identity, probably embrace a more centre-left agenda and possibly abandon equidistance. Given the changing electoral map, it makes political sense to take a more strident anti-Conservative line and forge new alliances with those on centre left of British politics. With Corbyn gone and a new leader with reputational competence and ‘soft Left’ credentials at the helm, the task of persuading moderates in Conservative facing seats to switch should now be easier. But the party needs to be careful as they too could lose ground to Labour in the places where they have made progress if credibility is weak. And if Labour surges, winning back Labour-held seats may prove impossible. This will be doubly difficult if the party does not choose to break with the past. The logic here suggests the next Liberal Democrat leader ought to have been elected since the Coalition years so that they and the party can move forward without the baggage of austerity. While the Liberal Democrats are not awash with potential candidates, recent electoral churn means that there are a number of credible options. Rebuilding the connection with Labour and Green supporters could be the difference between making significant electoral gains or remaining a spectator on the sidelines. The Liberal Democrats still face a monumental task. With Brexit seemingly done and dusted, political and economic fallout from the COVID-19 pandemic is likely to dominate the political discourse. With traditional left-right debates about the role, size and funding of the state likely to usurp recent cultural drivers of voting, the Liberal Democrats will face a renewed challenge of connecting with the public, providing policy distinctiveness and retaining political relevance. The 2019 election demonstrated that even when the Liberal Democrats recovered their relevance they still found it difficult to be popular. How the party defines and positions itself in the next few years could dictate its viability and long-term future. The Liberal Democrats might be on the road to recovery but the path ahead remains hazardous. Footnotes 1 To the BES question of whether parliament should cancel Brexit, 54% of Liberal Democrat voters in June 2019 either approved or strongly approved. For Labour switchers, it was 57% but far less for Conservative switchers (46%). References Ashcroft L. ( 2019 , December 13) ‘How Britain Voted and Why’, accessed at https://lordashcroftpolls.com/2019/12/how-britain-voted-and-why-my-2019-general-election-post-vote-poll/ on 20 April 2020. Cutts D. , Russell A. ( 2017 ) ‘The Liberal Democrats: Green Shoots of Recovery or Still on Life Support?’, In Tonge J. , Bandeira C. , Wilks-Heeg S. (eds.) Britain Votes 2017 , Oxford , Oxford University Press , pp. 72 – 90 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cutts D. , Russell A. ( 2018 ) ‘The ‘Exit from Brexit’ illusion: why the Liberal Democrats cannot capture the 48%, LSE Blog (British Politics and Policy)’, accessed at https://blogs.lse.ac.uk/brexit/2018/11/09/the-exit-from-brexit-illusion-why-the-liberal-democrats-cannot-capture-the-48/ on 19 April 2020. Fieldhouse E. , Green J. , Evans G. , Mellon J. , Prosser C. ( 2019 ) ‘British Election Study Internet Panel Wave 16’, accessed at https://www.britishelectionstudy.com/data-object/wave-16-of-the-2014-2019-british-election-study-internet-panel/ on 2 April 2020. Independent ( 2019 a, November 22) ‘Question Time Debate: Jo Swinson admits “getting it wrong on austerity”’, accessed at https://www.independent.co.uk/news/uk/politics/election-debate-jo-swinson-become-prime-minister-regret-austerity-a9214526.html on 20 April 2020. Independent ( 2019 b, December 4) ‘Jo Swinson knows she will lose’, accessed at https://www.independent.co.uk/news/uk/politics/election-jo-swinson-interview-andrew-neil-liberal-democrats-coalition-welfare-cuts-a9233351.html on 19 April 2020. YouGov ( 2019 ) ‘Sunday Times Survey Results’, accessed at https://3859gp38qzh51h504x6gvv0o-wpengine.netdna-ssl.com/files/2019/12/YouGov-Sunday-Time-opinion-poll-late-November-2019.pdf on 20 April 2020. © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Dennison,, James

doi: 10.1093/pa/gsaa026pmid: N/A

This contribution considers how niche parties react when they lose their niche, using the cases of three parties in the turbulent period prior to the 2019 UK general election: the Brexit Party, the Green Party and Change UK. I overview the background of these parties before showing that each lost its respective policy niches to larger, more established parties. I show that each responded with some combination of directly competing with the mainstream party; electorally cooperating with them or other parties; or diversifying into something distinct from their mainstream analogue. I explain how each party’s approach partially explains their 2019 general election result, as well as European Parliament elections result, using British Election Study data. I suggest that this ‘compete, cooperate or diversify’ approach provides a theoretical framework for understanding how niche parties are likely to react to losing their niche elsewhere. 1. Niche parties in the UK prior to the 2019 UK general election Even though they ultimately received relatively few votes, the pivotal role played by ‘niche parties’ during the 2017–2019 parliament does much to explain the outcome of the election and the subsequent direction of British politics. Such parties have received various academic definitions, with perhaps the most succinct being that they are ‘parties that compete primarily on a small number of non-economic issues’ (Wagner, 2012, p. 845). There have been a number of attempts in the academic literature to make sense of niche parties. Adams et al. (2006) show that niche parties, first, do not respond to shifts in public opinion by changing their policy positions, whereas mainstream parties do, and, secondly, are electorally punished when moderating their positions, unlike mainstream parties (see also, Ezrow, 2008). Meguid (2005) argued that the electoral fortunes of niche parties are subject to strategic manipulation of the salience of issues by mainstream parties, which may adopt ‘accommodative’ or ‘adversarial’ strategies in response (see also, Abou-Chadi, 2016). However, to date, there have been no analyses of how niche parties react to the actions of mainstream parties. Although in the notably specific context of a first-past-the-post electoral system, the tumultuous period in British politics prior to the 2019 elections in which a number of ‘niche parties’ rose and fell in prominence (see Figure 7.1) provides three cases to investigate their reactions. Figure 7.1 Open in new tabDownload slide General election polling, February 2019–December 2019 Source: Wikipedia aggregation of polls. Figure 7.1 Open in new tabDownload slide General election polling, February 2019–December 2019 Source: Wikipedia aggregation of polls. 2. How the Brexit Party won Britain’s most valuable niche On 15 January 2019, Theresa May’s Conservative government lost the first ‘meaningful vote’ on its Withdrawal Agreement with the European Union (EU) by a margin of 230 MPs, the largest defeat of a government in modern parliamentary history. The crushing defeat, the first of three on the UK–EU deal, came 14 months after the High Court of Justice ruled that the government would have to put any withdrawal agreement to a vote in parliament, upending the government’s Brexit strategy, casting the referendum result into doubt and leading to painful parliamentary wrangling between and within the mainstream parties. Five days later, Nigel Farage, the former leader of the UK Independence Party (UKIP) warned of attempts by Conservative and Labour MPs to rule out no deal, form a committee to take over the Brexit process entirely, or even end it altogether (Farage, 2019). He went on to state: A lot of new parties have been formed in the course of the last year or two. A registration has been put into the Electoral Commission for ‘The Brexit Party’ […] it’s got my full support in every way […] I want to make it really clear to all those people in Westminster on both parties that if they think they can betray Brexit and get away with it scot-free, they can’t. Elsewhere, Farage argued that there was widespread electoral demand for a party with a clear position on Brexit, compared to the ‘political elite [that] want to stop Brexit in its tracks [while] the Prime Minister doesn’t have the strength or inclination to see this through’ (Wooding, 2019). Survey data largely supported his claim; in early 2019, nearly half of the UK electorate believed that ‘no party was best able to handle’ the issue that they considered most important—usually Brexit—up from 20% two years earlier (see Figure 7.2). Meanwhile, the percentage listing the Conservatives had fallen from around a third to 13% and the figure for UKIP had slowly risen to 8%. Figure 7.2 Open in new tabDownload slide Perceptions of the party best able to handle the single most important issue facing the country, June 2016–December 2019 Q: ‘Which party is best able to handle the issue that you think is the single most important one affecting the country at the present time?’ Source:Fieldhouse et al. (2020). Figure 7.2 Open in new tabDownload slide Perceptions of the party best able to handle the single most important issue facing the country, June 2016–December 2019 Q: ‘Which party is best able to handle the issue that you think is the single most important one affecting the country at the present time?’ Source:Fieldhouse et al. (2020). Farage by this point had entirely separated himself from UKIP. In the aftermath of the 2016 referendum, he had given the Conservatives his tacit support to pursue Brexit unencumbered by electoral competition to its right. Eventually, in 2018, he left UKIP as it unsuccessfully attempted to both compete directly on Brexit with the Conservatives, who were still fairly trusted on the issue, and diversify into a more explicitly Islamophobic populist radical right party that publicly courted controversial far-right figures (Dennison, 2018). On 5 February 2019, the Brexit Party was officially registered with the Electoral Commission. Farage, appearing on various news outlets, stated that he would stand as a candidate for the European Parliament in May should Brexit be delayed. As the government failed to pass its second and third ‘meaningful votes’, some Conservative MPs openly suggested they might vote for the Brexit Party in the now inevitable European elections amid chaos inside the Conservative party. Farage, by then officially party leader after Islamophobic social media activity by the founder had been uncovered, held the official party launch two days after a six-month extension was agreed between the EU and UK. Referring to his former party, Farage concluded that ‘middle England, decent people’ would not vote for a party ‘completely obsessed’ by Islam with a ‘fairly loutish fringe’ and associated ‘with violence, criminal records and thuggery’ (Halliday and Walker, 2019). He stated, ‘In terms of policy, there’s no difference [from UKIP], but in terms of personnel there is a vast difference’. These personnel, introduced as candidates for the upcoming elections, represented a broad range of professions, while others with political backgrounds included the sister of high-profile Tory Brexiteer Jacob Rees-Mogg and ‘radical libertarian’ campaigner Claire Fox, underscoring the party’s single-issue credentials. Seven Members of the European Parliament (MEPs) defecting from UKIP followed. The party took steps to keep the focus purely on Brexit, including, aside from its easily interpretable name, withholding any manifesto until after the European Parliament elections and, even more unusually, having only three members so as to avoid the division that defined the internal politics of UKIP. Farage himself maintained control over the selection of the executive committee and candidates. The party itself was controlled by a parent company, The Brexit Party Limited, which had just two directors, Farage and the party’s chairman Richard Tice (an election candidate), along with five undisclosed shareholders. Crucially, as it would later turn out, the company’s memorandum of association gave the leader freedom to work with other democratic parties ‘for particular purposes consistent with its aims’, presumably working with a future ‘hard-Brexit’ Conservative Party (McTague, 2019). Published following a freedom of information request, the party’s constitution described its broader policy objectives in vaguely Thatcherite terms. Over the following weeks, other high-profile candidates were gradually announced, maintaining media coverage of the new party. The party was soon able to claim over 100,000 ‘registered supporters’ and topped every opinion poll for the European elections from late April onward. Estimates for the Conservatives declined from nearly 30% at the end of February to just over 10% by the election, while those for UKIP declined from 17% to around 3%. Meanwhile, Farage toured Leave voting areas, holding rallies focussed squarely on the undemocratic ‘betrayal’ of the Leave vote by a Remain parliament and no longer bothering with criticism of the EU itself. Promising that a vote for the Brexit Party was the best way to express desire for a ‘proper’, ‘clean’ or ‘World Trade Organisation (WTO) terms’ Brexit, Farage was met with crowds chanting his name. At the 2019 European Parliament elections, the Brexit Party won 30.5% of votes, 10.9 percentage points more than the second-placed Liberal Democrats. This share netted 29 seats, four in 10 of the total British delegation. The party’s predecessors, UKIP, managed 3.2% of the vote. The Brexit Party won the most seats in every constituency in England and Wales, except London and picked up an additional seat in Remain-leaning Scotland. Post-election surveys made clear that the Brexit Party had unequivocally won the most prized niche in British politics in the space of just two months. The Brexit Party won 70.3% of Leave voters and 82% of those who both voted Leave in 2016 and saw ‘Europe’ as the most important issue affecting Britain. In terms of so-called EU referendum identities, now more common than party identities, the Brexit Party won 75% of Leave identifiers. The Conservatives won <9% of each of these groups (Fieldhouse et al., 2020). The socio-demographic and attitudinal leanings of Brexit Party voters were reasonably consistent with those of UKIP voters in previous European elections. The average age, 59, was six years older than that of the average voter, 56% of the party’s voters were male, the party’s votes disproportionally came from the south, their income was only slightly lower than that of the average voter and the mean self-placement on a 0–10 left–right spectrum was 6.33 (Fieldhouse et al., 2020). The Brexit Party’s voters placed Theresa May at 4.1 and the Conservatives at 4.7 on a 0–10 scale of likeability, while 70% stated that the government was handling the Brexit process badly. More measured, however, were attitudes to the deal itself. Less than half believed that the government’s deal with the EU did ‘not at all’ honour the result of the EU referendum, suggesting ire was directed more at the inability to ‘leave’ rather than the nature of the deal itself. Perhaps most significantly of all, 64% of Brexit Party voters stated in June that they planned to vote for the party again at the next general election. 3. From losing Britain’s biggest niche to unilateral electoral cooperation In the month following the European Parliament election, the Brexit Party regularly topped polling for the next general election, with one poll estimating that the party would win 306 MPs, 20 short of an outright majority while the party came within just 683 votes of upsetting Labour in a by-election on 7 June, with 29% of the vote. Owing to the Brexit Party’s drubbing of the Conservatives, who received just 8% of the vote, at the European Parliament elections, Theresa May announced that she would resign as Prime Minister (see Denver, this volume). Attention thereafter turned to the Conservative leadership election, with commentators speculating about which candidate might be best able to win votes back from Farage’s party. Polling showed that at least a third of Conservative Party members would vote for the Brexit Party if there were a general election before the UK had left the EU. By aping many of the noises being made by the Brexit Party, Boris Johnson, promising a ‘do-or-die’ Brexit, even if proroguing parliament were required, quickly became the choice of hard-line Brexiteers within the Conservative Party and, thus, the favourite to win the contest. Farage’s initial strategic reaction to events was competitive. The Brexit Party leader rubbished the Brexit credentials of the Tories and the idea of an electoral pact with them, asking ‘Just to keep the Tories inside No 10 and us in the EU? … I don’t trust any of them to deliver a genuine Brexit and unless that situation changes, we are gearing up as an organisation to fight every seat in the country’ (Adu, 2019). Having praised the campaign of Rory Stewart, the main candidate supporting May’s deal, he reserved his ‘strongest criticism’ for Boris Johnson, labelling him ‘untrustworthy’ after voting ‘aye’ at the third meaningful vote and being ‘very confused’ about the EU’s single market during the 2016 referendum campaign. Installed as Prime Minister in late July, Boris Johnson took a hard-line stance by proroguing parliament and removing the whip from 21 Conservative MPs who had voted in favour of the Benn Bill, which legislated against a no-deal Brexit. One former Conservative cabinet minister announced her intention to resign as an MP, stating: ‘My concerns about the Conservative Party becoming the Brexit Party have come to pass’. Johnson announced his intention to seek an early election, something that MPs then rejected thrice before forcing the government to request an extension until 31 January 2020. Johnson did so, while washing his hands of responsibility for ‘parliament’s letter, not my letter’. During this period, the intensity and complexity of parliamentary wrangling and the novelty of Johnson’s premiership overshadowed the Brexit Party. One of the main exceptions came as Johnson’s government put a minimally revised deal—lambasted by Farage as ‘appalling’—to parliamentary vote, passing second reading on 22 October. Though the Brexit Party at this stage maintained a highly competitive stance towards the Conservatives, Johnson had done what May had been unable to do, in receiving the support of a majority of MPs for a deal, leaving it ‘frozen’ at second reading, just seven days before the House of Commons finally agreed to a general election and with the Conservatives again seen as the best party to deal with Brexit (see Figure 7.2). Cowed by the possible loss of their niche, the Brexit party turned from electoral competition to cooperation as the best way to maintain relevance. The calls for cooperation were initially aggressive and maintained attempts to undermine their rival’s credibility on the issue. On 2 November, Farage issued an ‘ultimatum’ to the Prime Minister that the Brexit Party would stand candidates across the country unless he scrapped the deal with the EU. Johnson flatly rejected electoral cooperation and underlined his sole ability in parliament to ‘Get Brexit Done’. The next day Farage stated that he personally would not run in the election, allowing him to campaign across the country, while attacking Johnson’s deal as ‘not Brexit’, otherwise ‘we wouldn’t need to fight against him in this election’ (Sandle, 2019). Simultaneously, he contradicted this dogmatism by suggesting that his own offer of a ‘Leave alliance’, reportedly made as early as September, had been rejected. Meanwhile, the Conservatives polling lead steadily increased as pro-Brexit voters again faced the inescapable logic of first-past-the-post. On 11 November, the last day for candidates to register, Farage declared that his party would not field candidates in the 317 seats in which there was an incumbent Conservative MP. He justified the decision by citing recent promises by Johnson that the one-year post-Brexit transition period would not be extended and that there would be greater regulatory divergence with the EU. Even most Brexit Party candidates, whose campaigns were abruptly ended, expressed support for the decision, given the obvious electoral arithmetic; indeed, much of the pressure to not field a full slate of candidates came from within the Brexit Party. However, with a one-sided ‘Leave alliance’ of sorts in place, Farage refocussed on attacking Labour for ‘betraying’ its Leave-supporting voters. Underscoring the party’s decline into second-tier players that fulfilled a niche-within-a-niche, Farage initially responded ambiguously when asked if he would vote for the Conservatives. At the general election, the party contested 275 seats, winning 644,257 votes and 2% of the national vote share (5.1% in seats in which it stood candidates). The party gained around 30% in two Barnsley seats, though Labour maintained healthy majorities in each case. In terms of socio-demographic and attitudinal profile, Brexit Party voters were similar to their European Parliament voters. Indeed 83% of Brexit Party voters at the general election had voted for them seven months earlier at the European Parliament election, with another 10% opting for UKIP. In the aftermath of the election, there was considerable speculation about whether Farage’s decision had cost Labour or the Conservatives more seats. Proponents of the former theory, in some cases seemingly keen to explain the Labour Party’s loss as a Brexit one-off rather than a rejection of the direction of the party itself, claimed that ‘Farage won it for Johnson’, arguing that the Brexit Party won the votes of ‘Labour leavers’ who would never vote Conservative. One aggregate-level analysis estimated that Farage’s decision to stand down ultimately doubled Johnson’s majority (Norris, 2019). However, later-released individual-level data puts into doubt these arguments (Fieldhouse et al., 2020). Whereas ‘only’ 45% of all voters placed Corbyn at nought on a zero to ten scale of likeability, 73% of Brexit Party voters did so and only 4% reported liking him (giving him six or higher). In contrast, 69% of Brexit Party voters reported liking Boris Johnson and only 6% placed him at zero. That said, 29% had voted for Labour in 2017, <37% who had voted Conservative but more than the 22% who had voted for UKIP, at that point struggling. Overall, the Brexit Party’s voters typically had little history of voting Labour and, even if they did, would have been unlikely to do so again with Jeremy Corbyn as Labour leader, instead of feeling considerable sympathy with Boris Johnson. Finally, with Brexit complete and electoral competition and cooperation now moot, Farage spent the aftermath of the election pointing to his plans for the ultimate diversification of the Brexit Party, renaming it the ‘Reform Party’, a name already registered prior to the election, with the first priority of reform to be the House of Lords. Whether constitutional reform has the niche-issue vote potential as Brexit had very much remains to be seen. 4. The Greens: balancing competition, cooperation and diversification Unlike the other niche parties at the 2019 general election, the Green Party of England and Wales (GPEW) had already lost their most significant niche of anti-austerity politics in the previous parliament following the election of Jeremy Corbyn as leader of the Labour Party (Dennison, 2015, 2016, 2018). As a result, the party had experienced balancing competition, cooperation and diversification. At the party’s 2017 autumn conference, the GPEW explicitly debated how best to perform this balance going forward. Caroline Lucas, the party’s co-leader, defended the strategy of cooperation that had seen the party stand aside in 22 marginal seats in which the Labour or Liberal Democrat candidate supported electoral reform, arguing that the decision had avoided a Conservative majority. Delegates argued it had caused a negative impact on the overall vote share, which indeed had more than halved, with a subsequent loss in funding. In reality, these losses were more to do with the loss of their niche and, partially as a result, greater tactical voting in 2017. Some went so far to describe the fangled ‘progressive alliance’ as ‘treachery’ or ‘surrender’, particularly as the other parties failed to return the favour. Delegates passed a motion calling for ‘new tests’ on future pacts and for the party leadership to stop referring to Green participation in any ‘progressive alliance’. Continuing the strategic balancing act of the last parliament, Lucas refocussed on diversification, stating that by ‘focusing on the environment and Brexit’ the Greens would win voters back and arguing that ‘Corbyn’s position on Brexit betrays young people’. Outside of the party, in early 2018, pro-Corbyn Guardian columnist, Owen Jones, called for the collapse of the Green Party into the Labour Party stating ‘In the political era before 2015 the Greens functioned as a left-wing alternative to Labour; but the ascent of Corbynism has rather dented that purpose. Truthfully, the Green party’s prospects are poor’ (Jones, 2018). In May, despite Jones’ predictions, the GPEW saw their councillors increased by 8 to 39 on a vote share of 6.5% at the 2018 local elections. Caroline Lucas, in her second leadership stint, announced that she would not be standing at the party’s upcoming biennial leadership contest, after two years dedicated in large part to internal restructuring and performing a three-way strategic balancing act of ‘competition, cooperation and diversification’. She stated ‘We have not been eclipsed by the rise of Jeremy Corbyn, but instead have used these unique circumstances to push for even more radical change’ before highlighting the Greens’ positions on post-material issues like shorter working weeks, ending ‘pointless jobs’, environmentalism, peace and drug legalisation, as well as electoral reform. Incumbent co-leader Jonathan Bartley and two-time London mayoral candidate and London Assembly member, Siân Berry, by this point both also councillors, were declared new co-leaders that September, with over three-quarters of the party vote. Bartley used his victory speech to criticise ‘treacherous’ Corbyn who had ‘fudged’ the issues of Brexit, airport expansion, nuclear power and electoral reform, as well as overlooking wealth redistribution. Finally, Bartley criticised Labour’s commitment to credit-fuelled growth and subsidies for both fossil fuels’ and the arms trade—perhaps attempts by the former advisor to John Major to pique the interest of more fiscally conservative voters, in the vein of the German Greens. As the unforeseen 2019 European Parliament election approached, the Greens again began to find their relevance increase. The party gained a ‘sensational’ 184 seats at the English local elections, the biggest percentage gain of any party and leading to membership surge of over 10,000 members. The party’s European campaign completely diversified from the socio-economic focus of the preceding decade, now put simply as ‘Yes to Europe, No to Climate Chaos’. Party co-leader, Berry, stated ‘The Greens are the most united, strongest pro-EU party on the ballot’. Survey evidence suggested that the British public, or at least the portion that felt able to express a view on the subject, agreed, consistently placing the party as the most united in British politics (see Figure 7.3). Vince Cable, leader of the Liberal Democrats, proposed running joint tickets with the Greens and Change UK on the issue of a second referendum; however, the party turned down the offer, citing ideological differences and the Greens’ focus on climate change. Behind the scenes, the Greens did not want to risk wasting the more proportional nature of the electoral system and had the wind in their sails following an uptick in the salience of their original environmental niche, amid widespread flooding and increasing global awareness of climate issues (see Figure 7.4). Figure 7.3 Open in new tabDownload slide Perceptions of the extent of unity and division in political parties, June 2017–December 2019 Q: ‘How united or divided are each of these parties?’ (i) very divided; (ii) fairly divided; (iii) neither divided nor united; (iv) fairly united; (v) very united. Source:Fieldhouse et al. (2020). Figure 7.3 Open in new tabDownload slide Perceptions of the extent of unity and division in political parties, June 2017–December 2019 Q: ‘How united or divided are each of these parties?’ (i) very divided; (ii) fairly divided; (iii) neither divided nor united; (iv) fairly united; (v) very united. Source:Fieldhouse et al. (2020). Figure 7.4 Open in new tabDownload slide Most important issues facing the country, June 2017–January 2020 Q. ‘Which of the following do you think are the most important issues facing the country at this time? Please tick up to three’ Source: YouGov Top Issue Tracker. Figure 7.4 Open in new tabDownload slide Most important issues facing the country, June 2017–January 2020 Q. ‘Which of the following do you think are the most important issues facing the country at this time? Please tick up to three’ Source: YouGov Top Issue Tracker. At the 2019 European Parliament elections, the GPEW won an impressive 12% of the vote (an increase of five points) and seven seats (up from three), with over 1.8 million voters choosing them. Compared to the mean voter at the election, the average Green party voter was five years younger and five percentage points more likely to be female, while on average placing themselves at 3.5 on a 0–10 left–right spectrum. Lucas explained the result with the party’s twin campaign on the climate and the EU, though Green parties across the EU had performed well. The party’s general election polling was by now regularly into double figures for the first time since early 2015. The party’s post-Corbyn diversification had until then been primarily in terms of salience, with more attention being given to Europe, the environment and post-material policies, rather than positioning. Most of these policies were already in place while the party was still operating as an anti-austerity niche party. However, by late September, the Greens also held a unique position in Europe after the Liberal Democrats changed their policy from favouring a second referendum to immediate revocation of Article 50 to remain in the EU. Lucas (2019), wasted no time in labelling the move as ‘arrogant, self-indulgent, cynical and very dangerous’. By mid-November, however, the Greens had managed to build a ‘Remain alliance’ with the Liberal Democrats and Plaid Cymru across 60 seats, which would see the Greens given a free run in eight seats and stand aside in 43. The alliance, called ‘United to Remain’ and led by now-Liberal Democrat Heidi Allen, was criticised by Labour shadow chancellor John McDonnell who stated ‘To those Greens and others who are getting into bed with the Lib Dems: I remember what the Lib Dems did when they were in government. I say, go and ask any disabled person in this country’ (Proctor, 2019). During the election campaign, and with the environment now considered the third most important issue affecting Britain by the public, the Greens focussed a full 45.9% of its social media activity on environmental issues (Deacon et al., 2019). Berry encapsulated the party’s strategic refocus by telling activists that making the election about climate was even more important than Brexit. This was only further buoyed by their inclusion in a leaders’ debate specifically on the issue of climate change in late November. However, by this point, the other parties were again attempting to win over environmental voters, with Labour describing its centrepiece economic plan as a ‘green industrial revolution’ and the Liberal Democrats pledging tighter deadlines for net-zero carbon emissions. At the election, the Greens won 2.7% of the national vote, up from 1.9% two years earlier, again returning one MP, Caroline Lucas. The party saved its deposit in 31 constituencies, up from five, though was unable to produce a genuine electoral threat for a second seat. Green voters held a similar profile to those at the European Parliament elections, albeit less skewed either socio-demographically or attitudinally, placing themselves at a fairly centrist 4.0 on a 0–10 left–right spectrum, with 15% considering themselves right of centre. In 2017, 39% of 2019 Green voters had voted for Labour, 17% had voted Conservative and 25% had voted Green. Notably, Green Party voters expressed strong dislike for Jeremy Corbyn, with 82% placing him at less than five on a 0–10 likeability scale (Fieldhouse et al., 2020). 5. Change UK: a very parliamentary niche party For a group receiving 18% in a nationwide poll five days after being formed, it seems fair to say that the 2019 general election, nine months later, did not turn out as planned for The Independent Group (TIG; later known as Change UK—TIG and, finally, TIG for Change). Initially, a coalition of seven MPs simultaneously resigning from a Labour Party they described as ‘hijacked by the machine politics of the hard left’, ‘institutionally antisemitic’ with a culture of ‘bullying and bigotry’ and ‘facilitating Brexit’, the group was immediately compared to the Social Democratic Party of the 1980s. Although the launch was overshadowed by one MP making a racial faux-pas, over the next two days another Labour MP and three Tory MPs, whose party was at that point still considering a no-deal Brexit, joined TIG. In total, four of the party’s MPs had recently been deselected by their constituency Labour Party. In the following weeks, the group remained squarely in the media limelight and attracted large donations and revolting councillors, as pro-EU forces sensed that they may now somehow be able to overcome parliamentary Brexit gridlock in their favour and anti-Corbyn Labour supporters hoped the group might offer the centre-left a fresh start. Immediately, commentators speculated over how TIG and the Liberal Democrats would interact with each other, given their ideological similarity, the precedent of the Social Democratic Party (SDP)-Liberal alliance and merger and that both had equal parliamentary representation. Three of the group’s most high-profile parliamentarians, Chuka Umunna, Chris Leslie and Anna Soubry, adopted a competitive stance. Umunna and Leslie immediately ruled out a merger and highlighted their group’s ambition to ‘build a new alternative’, while Soubry urged Liberal Democrats to join their group. Meanwhile, the Labour Party quickly attempted to stymie further defections by announcing that it would back moves for a second referendum. The group officially registered as a party in March. This was again overshadowed by its proposed ‘Change UK – The Independent Group’ name being challenged by petition website change.org. The new party, with ex-Conservative Heidi Allen chosen as its interim leader, declared it would put forward a large number of candidates if there were forthcoming European elections. Liberal Democrat Jo Swinson called for the two parties to find ways of ‘working together’, a proposal rejected by TIG on the grounds that the party’s brand was still damaged by its participation in the 2010–2015 coalition. A reportedly leaked internal party strategy document explicitly outlined how they planned to win over Liberal Democrat activists, members and resources. With the European Parliament elections now seemingly inevitable, TIG again rejected offers of electoral cooperation from the Liberal Democrats, this time to stand joint pro-Remain candidates. Instead, on 23 April, TIG announced a full list of 73 candidates, including former Labour, Conservative and Liberal Democrat MPs, and a broad range of professional backgrounds, including Boris Johnson’s sister and a former BBC correspondent, to some extent echoing the Brexit Party’s earlier launch. Two of the candidates were quickly forced to step down after ‘inappropriate’ past tweets were discovered. At a question-and-answer session, the party claimed that, for now, only a party commitment to a second referendum mattered and the new policies would be announced soon. Soon after, the party’s ‘TIG’ logo and its use of a hashtag were both rejected by the Electoral Commission, meaning that its candidates would have to run with blank spaces next to their names in the upcoming elections. More broadly, the party’s unclear branding and name were negatively compared to those of the immediately understandable Brexit Party. Even Rachel Johnson, the party’s lead candidate for the South West, publicly criticised the name and party leadership, likening the outfit to a sinking ship. Meanwhile, the lead candidate for Scotland defected to the Liberal Democrats and encouraged others to do the same after homophobic remarks by another candidate, the former Deputy Prime Minister of Poland, were uncovered. More problematically, as shown in Figure 7.5, less than half of voters were able to correctly name the party’s official Brexit position, the lowest of any party. Figure 7.5 Open in new tabDownload slide Attitudes to Brexit by party supported, June 2019 Q: ‘Some people want to leave the European Union immediately without a deal. Other people would like to cancel Brexit and remain in the EU. Where would you place yourself and the political parties on this scale?’ Source:Fieldhouse et al. (2020). Figure 7.5 Open in new tabDownload slide Attitudes to Brexit by party supported, June 2019 Q: ‘Some people want to leave the European Union immediately without a deal. Other people would like to cancel Brexit and remain in the EU. Where would you place yourself and the political parties on this scale?’ Source:Fieldhouse et al. (2020). 6. Losing their niche and splintering into cooperation, competition and diversification At the European Parliament elections, Change UK achieved a disappointing 3.3% of the vote, seventh place and no seats. In contrast, the Liberal Democrats comfortably came in second place with 19.6% of the vote in an election obviously characterised by the painful consequences of the 2016 referendum. The party’s voters were slightly younger than average, gender-balanced, with above-average incomes and disproportionately likely to hail from London or the South East of England, while placing themselves at a very centrist 4.8 on at 0–10 left–right spectrum on average (Fieldhouse et al., 2020). At the 2017 general election, half of 2019 Change UK voters had chosen Labour, 36% the Conservatives and 10% the Liberal Democrats. At the December 2019 general election, 40% would go on to vote for Labour, 18% for the Conservatives and 33% for the Liberal Democrats. With the battle to win the pro-Remain niche over, Change UK’s MPs split into three ways over the correct course of action for the party. Before the result, Allen and Umunna were already openly hinting at electoral cooperation with the Liberal Democrats at the next general election and stating their remorse over the lack of a pact. Another group preferring to maintain competition with the Liberal Democrats was led by Anna Soubry who soon became party leader, attacking Allen’s remarks as ‘bizarre’. Meanwhile, a third preferred to diversify by returning to its initial parliamentary niche of being a loose independent group. One commentator wryly pointed out that ‘bluntly, the decision is likely to be made for’ all three splinters (Bush, 2019). Eventually, five TIG MPs joined the Liberal Democrats, including Umunna and Allen. Three of these had initially joined a non-party parliamentary group called ‘The Independents’, set up by a handful of former TIG MPs and operating for the rest of the Parliament as a ‘cooperative of independents’. At the 2019 general election, the remaining three MPs of Change UK, now known as TIG for Change, failed to retain any of their seats, with their best result being Anna Soubry’s 8.5%. Soon after, Soubry announced the party’s deregistration. 7. Conclusion The three niche parties in the UK prior to the 2019 general election reacted to losing their niche with some combination of ‘compete, cooperate or diversify’. The rapid success of the Brexit Party directly led to Boris Johnson’s election as leader of the Conservatives, robbing them of their Brexiteer niche. This led them to first attempt to undermine Johnson’s Brexiteer credentials, with little success, before demanding cooperation on their own terms and then resigning themselves to unilateral electoral cooperation. After the election, party leader Farage hinted at diversification into the ‘Reform Party’. The GPEW, which had lost its anti-austerity, far left niche to Labour in 2016, spent 2017–2019 continuing the strategic balancing act that had maintained place in second tier of the party system. Diversification to focussing on issues like Europe and (back to) the environment proved successful as the salience of both issues rose, while they formed a more favourable ‘remain alliance’ than the ‘progressive alliance’ of two years earlier. Finally, Change UK (under various other names) took a highly competitive stance towards its obvious rivals, the Liberal Democrats, before its pro-Remain niche had been properly secured. With the obvious loss of this niche in the aftermath of the European Parliament elections, its 11 MPs divided into those cooperating via joining their erstwhile rivals, continuing to compete as a party or diversifying into a ‘non-party’ called ‘The Independents’. The three different approaches partially resulted from the different leadership structures of the parties, with Farage able to dictate transition from one strategy to another, the pluralist Greens forced to balance various forces and the ‘very parliamentary’ niche party of TIG having more leaders than members, each clustering around one strategy after the party lost its niche. The ‘compete, cooperate or diversify’ theoretical framework seems to also work well for the fates of recent British niche parties, such as UKIP in its heyday, or the Greens in the early 1990s, although it may not work so well in more proportional electoral systems in which niche parties do not risk ‘dividing the vote’, making cooperation less imperative. The model also highlights the difficulties faced by niche parties when they lose their niche, with only the Greens, after three years of a difficult balancing act, able to claim some level of success at the 2019 general election. Acknowledgment I would like to acknowledge the generous support of the Swedish Research Council (Vetenskapsr䣥t; grant no. 2019/00504). References Abou-Chadi T. ( 2016 ) ‘ Niche Party Success and Mainstream Party Policy Shifts – How Green and Radical Right Parties Differ in Their Impact ’, British Journal of Political Science , 46 , 417 – 436 . Google Scholar Crossref Search ADS WorldCat Adams J. , Clark M. , Ezrow L. , Glasgow G. ( 2006 ) ‘ Are Niche Parties Fundamentally Different from Mainstream Parties? the Causes and the Electoral Consequences of Western European Parties’ Policy Shifts, 1976 ’, American Journal of Political Science , 50 , 513 – 529 . Google Scholar Crossref Search ADS WorldCat Adu A. ( 2019 ) ‘Brexit Pact’, The Sun, accessed at https://www.thesun.co.uk/news/9309594/tories-pact-nigel-farage-general-election-johnson/ on 27 April 2020. Bush S. ( 2019 ) ‘Terrible Election Results and Nearing Zero in the Polls: What is the Future of Change UK?’ New Statesman, accessed at https://www.newstatesman.com/politics/uk/2019/06/terrible-election-results-and-nearing-zero-polls-what-future-change-uk on 27 April 2020. Deacon D. J. , Goode D. , Smith D. , Wring D. , Downey J. , Vaccari C. ( 2019 ) ‘General Election 2019: Report 5: 7 November–11 December’, Centre for Research in Communication and Culture, Loughborough, Loughborough University. Dennison J. ( 2015 ) ‘ The Other Insurgency? The Greens and the Election ’, Parliamentary Affairs , 68 , 188 – 205 . Google Scholar Crossref Search ADS WorldCat Dennison J. ( 2016 ) The Greens in British Politics: Protest, anti-Austerity and the Divided Left . Basingstoke , Palgrave Macmillan . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dennison J. ( 2018 ) ‘ The Rug Pulled from under Them. UKIP and the Greens ’, Parliamentary Affairs , 71 , 91 – 108 . Google Scholar Crossref Search ADS WorldCat Ezrow L. ( 2008 ) ‘ Research Note: On the Inverse Relationship between Votes and Proximity for Niche Parties ’, European Journal of Political Research , 47 , 206 – 220 . Google Scholar Crossref Search ADS WorldCat Farage N. ( 2019 , 20 January) The Nigel Farage Show, LBC. Accessed at https://www.youtube.com/watch?v=HZxJ5J9Jt1Q on 4 June 2020. Fieldhouse E. , Green J. , Evans G. , Mellon J. , Prosser C. ( 2020 ) ‘British Election Study Internet Panel Waves 1–19’, accessed at https://www.britishelectionstudy.com/data-objects/panel-study-data/ on 25 April 2020. Halliday J. , Walker P. ( 2019 ) ‘Annunziata Rees-Mogg to Stand as MEP for Farage’s Brexit Party’. The Guardian, accessed at https://www.theguardian.com/politics/2019/apr/12/annunziata-rees-mogg-to-stand-as-mep-for-farages-brexit-party on 27 April 2020. Jones O. ( 2018 , 22 February) ‘The Greens’ Best Hope is to Sign Up with Labour’, The Guardian, accessed at https://www.theguardian.com/commentisfree/2018/feb/22/greens-labour-jeremy-corbyn on 27 April 2020. Lucas C. ( 2019 , 18 September) ‘Lib Dem ‘Revoke and Remain’ Stance is Arrogant’, The Guardian, accessed at https://www.theguardian.com/politics/2019/sep/18/caroline-lucas-lib-dems-revoke-and-remain-stance-brexit-is-arrogant on 27 April 2020. McTague T. ( 2019 ) ‘Nigel Farage’s Startup Politics’, Politico, accessed at https://www.politico.eu/article/nigel-farage-brexit-party-start-up-politics-eu-election/ on 27 April 2020. Meguid B. ( 2005 ) ‘ Competition between Unequals: The Role of Mainstream Party Strategy in Niche Party Success ’, American Political Science Review , 99 , 347 – 359 . Google Scholar Crossref Search ADS WorldCat Norris P. ( 2019 ) ‘Was Farage the Midwife Delivering Johnson’s Victory? The Brexit Party and the Size of the Conservative Majority’, LSE British Political and Policy, accessed at https://blogs.lse.ac.uk/politicsandpolicy/ge2019-brexit-party-impact/ on 27 April 2020. Proctor K. ( 2019 , 7 November) ‘Lib Dems, Greens and Plaid Cymru reveal Remain Election Pact’. The Guardian, accessed at https://www.theguardian.com/politics/2019/nov/07/lib-dems-greens-and-plaid-cymru-reveal-remain-election-pact on 27 April 2020. Sandle P. ( 2019 ) ‘Brexit Party Leader Nigel Farage Will Not Run in UK Election’, Reuters, accessed at https://www.reuters.com/article/uk-britain-election-farage/brexit-party-leader-nigel-farage-will-not-run-in-uk-election-idUSKBN1XD05Y on 27 April 2020. Wagner M. ( 2012 ) ‘ Defining and Measuring Niche Parties ’, Party Politics , 18 , 845 – 864 . Google Scholar Crossref Search ADS WorldCat Wooding D. ( 2019 ) ‘Newkip’. The Sun, accessed at https://www.thesun.co.uk/news/brexit/8236819/nigel-farage-new-brexit-party/ on 27 April 2020. © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Mitchell,, James;Henderson,, Ailsa

doi: 10.1093/pa/gsaa027pmid: N/A

Recent trips to the polls in Scotland seem to evoke a series of binaries: Yes versus No, Remain versus Leave, Indyref versus Brexit, surges and losses. The 2019 UK election is no exception for it allows us to look at simple themes, such as whether the Scottish National Party (SNP) would regain those seats it lost in 2017 and offer more nuanced analysis about the relative importance of different constitutional issues on voter choice. In what follows, we outline the wider political context of the 2019 election, the relevant themes within the campaign and then explore the results, at each point identifying the various challenges facing parties and how they navigated an electorate that now finds itself divided into four tribes: Yes/Remain, Yes/Leave, No/Remain and No/Leave. 1. Background and summary results In the early years of devolution, it was assumed that elections to the Commons would be first order and elections to the Scottish Parliament would be second order.1 Allied with a more proportional electoral system, the SNP was expected to perform better in elections to Holyrood but struggle to challenge Labour’s dominance representing Scotland in the Commons. Holyrood elections created opportunities for the SNP unavailable in elections to the Commons. The ‘blackmail’ potential would be augmented by a ‘governing potential’, the possibility of becoming the governing party in Edinburgh (Mitchell et al., 2011, pp. 10–33). The SNP averaged under 20% of the vote in the first three Commons elections after devolution but its constituency vote in Holyrood averaged close to 30%, seemingly confirming the second-order thesis. These expectations have been challenged and patterns disrupted in recent elections. The SNP landslide in 2015 reduced its Labour, Conservative and Liberal Democrat opponents to one seat each. The independence referendum that took place eight months before had caused this dramatic change. But this was only the first shock to the Scottish party system. The Brexit referendum in 2016 had a similar, though less significant, impact. The 2017 general election witnessed the SNP lose a quarter of its vote from two years before, falling from 50% to 36.9% and losing 21 seats. The Conservatives pushed Labour into third place in Scotland for the first time since 1935.2 The 2019 European elections offered continued pain for Labour. Their least-worst swing was −6 in Orkney and worst was −24 in North Lanarkshire. They were in fourth place in two-thirds of councils and sixth in a further 12. At the time, this was attributed to ambiguous or inconsistent policies on both Brexit and independence. These two issues—Brexit and Scottish independence—have recently dominated Scottish politics, creating remarkable turbulence in electoral behaviour over the last five years. However, there has been little shift in public opinion on independence or Brexit. Indeed, there is evidence that opinion on independence has become more entrenched. But while these issues continue to be the major drivers of Scottish electoral behaviour, the importance of each has varied in impact over the last three general elections. The turbulent nature of Scottish electoral behaviour is evident in the last three UK general elections (Table 8.1). We have to go back to Labour’s 1997 landslide to see significant shifts in share of the vote when Labour increased its Scottish share of the vote by 6.6 points from 1992 and the Conservatives’ share tumbled by 8.2 points. The succession of dramatic changes over three elections suggests an unsettled electorate. The days when becoming a Scottish MP was an assured job for life for some have gone at least for the moment. Table 8.1 Vote shares and seats in Scotland, 2015, 2017 and 2019 UK general elections . 2015 . 2017 . 2019 . % vote (change) . Seats (change) . % vote (change) . Seats (change) . % vote (change) . Seats (change) . Conservative 14.9 (−1.8) 1 (−) 28.6 (+13.7) 13 (+12) 25.1 (−3.5) 6 (−7) Labour 24.3 (−17.7) 1 (−40) 27.1 (+2.8) 7 (+6) 18.6 (−8.5) 1 (−6) Liberal Democrats 7.5 (−11.3) 1 (−10) 6.8 (−0.7) 4 (+3) 9.5 (+2.8) 4 (−) SNP 50.0 (+30.1) 56 (+50) 36.9 (−13.1) 35 (−21) 45.0 (+8.1) 48 (+13) . 2015 . 2017 . 2019 . % vote (change) . Seats (change) . % vote (change) . Seats (change) . % vote (change) . Seats (change) . Conservative 14.9 (−1.8) 1 (−) 28.6 (+13.7) 13 (+12) 25.1 (−3.5) 6 (−7) Labour 24.3 (−17.7) 1 (−40) 27.1 (+2.8) 7 (+6) 18.6 (−8.5) 1 (−6) Liberal Democrats 7.5 (−11.3) 1 (−10) 6.8 (−0.7) 4 (+3) 9.5 (+2.8) 4 (−) SNP 50.0 (+30.1) 56 (+50) 36.9 (−13.1) 35 (−21) 45.0 (+8.1) 48 (+13) Note: Figures in brackets refer to (change from previous election). Open in new tab Table 8.1 Vote shares and seats in Scotland, 2015, 2017 and 2019 UK general elections . 2015 . 2017 . 2019 . % vote (change) . Seats (change) . % vote (change) . Seats (change) . % vote (change) . Seats (change) . Conservative 14.9 (−1.8) 1 (−) 28.6 (+13.7) 13 (+12) 25.1 (−3.5) 6 (−7) Labour 24.3 (−17.7) 1 (−40) 27.1 (+2.8) 7 (+6) 18.6 (−8.5) 1 (−6) Liberal Democrats 7.5 (−11.3) 1 (−10) 6.8 (−0.7) 4 (+3) 9.5 (+2.8) 4 (−) SNP 50.0 (+30.1) 56 (+50) 36.9 (−13.1) 35 (−21) 45.0 (+8.1) 48 (+13) . 2015 . 2017 . 2019 . % vote (change) . Seats (change) . % vote (change) . Seats (change) . % vote (change) . Seats (change) . Conservative 14.9 (−1.8) 1 (−) 28.6 (+13.7) 13 (+12) 25.1 (−3.5) 6 (−7) Labour 24.3 (−17.7) 1 (−40) 27.1 (+2.8) 7 (+6) 18.6 (−8.5) 1 (−6) Liberal Democrats 7.5 (−11.3) 1 (−10) 6.8 (−0.7) 4 (+3) 9.5 (+2.8) 4 (−) SNP 50.0 (+30.1) 56 (+50) 36.9 (−13.1) 35 (−21) 45.0 (+8.1) 48 (+13) Note: Figures in brackets refer to (change from previous election). Open in new tab There has been only one election to the Scottish Parliament over the same period. In May 2016, the SNP retained its position as Scotland’s largest party but lost the overall majority of seats it had won five years before. The SNP formed a minority government with 63 seats in the 129 seat Parliament. The Conservative Party gained most, winning 31 seats (up 16 from 2011) and seven ahead of Labour (down 13). With six seats, the Scottish Greens, beneficiaries of the more proportional electoral system, pushed the Liberal Democrats with five seats into the fifth position. It is far from clear that Holyrood elections are second-order contests though turnout remains lower and smaller parties perform better in Holyrood elections, but the latter could be explained by the electoral system. It appears that Scottish concerns are driving Scottish elections at Holyrood and at Westminster as never before. The disadvantages faced by the SNP in elections to the Commons have declined significantly. The Scottish and European questions have each become simple binary choices: for or against independence and for or against European Union (EU) membership. This polarisation has worked to the advantage of the SNP and Conservatives as each has a clearly defined position on these questions. Ironically, the main thrust of Conservative campaigning in Scotland has been to keep the Scottish question alive, presenting the party as the most effective opposition to the SNP and seeking to marginalise Labour and the Liberal Democrats in the process. In early August 2019, for example, the outgoing Scottish Conservative leader, Ruth Davidson, accused Labour of ‘rank betrayal’ by opening the door to a second independence referendum, after Shadow Chancellor John McDonnell said Labour would not block a referendum if the Scottish Parliament and Scottish people decide that (The Herald, 2019). In the past, the Conservatives had been damaged electorally as the Unionist Party par excellence in Scotland (Mitchell, 1990) but its hardline position on the union was now working to its advantage. If anything, the Liberal Democrats have also hardened their opposition to independence in recent years. The party continued to pay lip service to federalism but made no effort to develop or prioritise the policy and its rhetoric has become vehemently anti-SNP and anti-independence. The old Liberal insistence that it was neither unionist nor nationalist but federalist has all but disappeared. Brexit presented major challenges for the Scottish Tories. While the Scottish Tory leadership had opposed Brexit during the 2016 referendum, it fell into line after the referendum result but got caught up in the internal party battles over the nature of the Brexit deal. Ruth Davidson insisted that any deal must be the same throughout the UK and argued against any special provisions for Northern Ireland, fearing this would set a precedent for a different arrangement for Scotland that the Scottish government proposed (Express, 2017). Davidson undermined Theresa May’s tactic of threatening to withdraw from negotiations with the EU by saying it was ‘certainly possible’ that the Scottish Tories might oppose policy decided by the UK Cabinet. The key issue that arose was whether the group of Scottish Tory MPs elected in 2017 would back May or Davidson in such a scenario. Newspaper reports suggested that the Scottish Tories might break away from the party south of the border, a platform on which her main opponent had stood in the Scottish Tory leadership contest in 2011 and which Davidson had then strongly opposed (The Telegraph, 2017; Convery 2014). In the aftermath of the 2017 election, it had been widely reported that the new group of Scottish Tory MPs would act as a bloc and be accountable to Davidson (The Guardian, 2017; The Courier, 2017). This view was encouraged by the Scottish leader’s claim that the group of Scottish Tory MPs would argue Scotland’s case ‘forcefully’ with the Prime Minister and the assertion that the Scottish party had ‘complete autonomy over policy, candidates, campaigning, finance’, while the party’s MPs would take the Tory whip in the Commons (BBC News, 2017). But it soon became clear that the MPs saw themselves primarily as accountable to the leadership in London rather than Edinburgh. This became even more obvious after Boris Johnson became the leader and effectively took back control of the party in Scotland. Most Scottish Tory MPs subsequently adopted a hardline position on Brexit. The SNP also had challenges. It had all but swept the board when independence had cast its shadow over the 2015 election by attracting a high turnout amongst the 45% who had voted ‘Yes’ in the 2014 referendum. Its three main opponents fought amongst themselves (and against lower turnout) for the 55% who had voted ‘No’. But the Brexit referendum in 2016 pushed that issue up the agenda, adding an unhelpful dimension for the SNP. In 2017, this meant that supporters of independence who also supported Brexit would be cross-pressured. There was less incentive to vote amongst supporters of independence, while politics operated in the shadow of the Brexit referendum. The SNP had misread the public mood and came out of the 2017 election bruised, losing 21 seats. As the results in Table 8.2 show, over 40% of those who had voted Yes in 2014 and Leave in 2016, voted for a party other than the SNP, with more than half turning to Labour, although SNP won almost 80% of those who had voted SNP in 2015 and Remain in 2016 (Henderson and Mitchell, 2018) and a similar proportion of Yes/Remain voters. As the political agenda shifted to Brexit, the SNP was in danger of arguing for independence when many of its erstwhile voters opposed the SNP policy on EU membership or thought independence was not the key issue in 2017. Table 8.2 Voting in the 2017 UK general election in Scotland, by constitutional preferences (%) . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 3 15 32 69 Labour 17 24 42 19 Liberal Democrats 2 3 13 5 SNP 77 58 13 5 Other 1 0 0 2 N 439 286 571 354 . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 3 15 32 69 Labour 17 24 42 19 Liberal Democrats 2 3 13 5 SNP 77 58 13 5 Other 1 0 0 2 N 439 286 571 354 Notes: The 2019 Scottish Election Survey asked respondents in wave 1 to recall how they voted in the previous UK general election (2017), in the 2014 independence referendum and the 2016 Brexit referendum. Respondents who did not vote in all three contests are therefore excluded from this table. Source: Scottish Election Study 2019. Open in new tab Table 8.2 Voting in the 2017 UK general election in Scotland, by constitutional preferences (%) . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 3 15 32 69 Labour 17 24 42 19 Liberal Democrats 2 3 13 5 SNP 77 58 13 5 Other 1 0 0 2 N 439 286 571 354 . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 3 15 32 69 Labour 17 24 42 19 Liberal Democrats 2 3 13 5 SNP 77 58 13 5 Other 1 0 0 2 N 439 286 571 354 Notes: The 2019 Scottish Election Survey asked respondents in wave 1 to recall how they voted in the previous UK general election (2017), in the 2014 independence referendum and the 2016 Brexit referendum. Respondents who did not vote in all three contests are therefore excluded from this table. Source: Scottish Election Study 2019. Open in new tab Labour has suffered the classic third-party squeeze on both independence and Brexit, not helped by a perceived lack of clarity on its Brexit position. While Labour’s policy remains steadfastly opposed to Scottish independence, many of its former voters supported the idea. Labour’s best prospects lay in changing the focus of attention away from these issues, framing the election on other matters. In 2019, the NHS was mentioned as a very important issue for 67% of its voters, with independence only in fifth place as a matter of importance. The problem for Labour was that its own voters were atypical of the electorate as a whole in listing matters of importance in the election (Ipsos MORI, 2019). Half of Labour’s lost deposits across Britain were in Scotland and the Liberal Democrats lost 13 deposits in Scotland. While the Liberal Democrats strived to present themselves as the European party par excellence by promising to reverse the Brexit process without a referendum, they struggled to convince voters that they were credible as the main pro-Remain party in Scotland compared with the SNP. The Scottish Greens are now the fourth party in Holyrood but continue to struggle to break through at Westminster. The party contested 22 candidates across Scotland in 2019, up from only three contested two years before but lost all deposits (the Scottish Greens’ best result was in Edinburgh East where they won 4.3%). The Brexit Party won a seat at the European Parliament elections in May 2019 with 14.8% of the vote, coming second behind the SNP (37.8%). In the general election, the Brexit Party contested 15 seats and UKIP only seven. Louis Stedman-Bryce, Brexit Party MEP for Scotland, resigned from the party just before the UK election because he felt that the party had failed to represent Leave voters by standing down in Tory-held seats and because the party’s candidate in Glenrothes was homophobic. The Brexit Party and UKIP lost all deposits and the best result for a Brexit Party candidate was for the Glenrothes candidate who had been dropped during the campaign by the party. 2. Results on the ground As ever, the aggregate results hid many variations across constituencies in Scotland. The SNP gained 14 seats but also lost a seat. The SNP had come from fourth to take North East Fife in 2015. Stephen Gethins, the SNP’s high-profile Westminster spokesperson on Brexit, held the seat two years later by two votes to make it the most marginal constituency in the UK. The seat had previously been held by Menzies Campbell, former Liberal Democrat leader, from 1987 until his retirement in 2015 and had long been a battleground between Liberal Democrats and Tories. The equivalent seat in Holyrood is held by Willie Rennie, Scottish Liberal Democrat leader. The Liberal Democrat strategy had been to appeal to the sizeable Conservative support in the constituency to vote tactically. This resulted in the Conservatives suffering their largest decrease in Tory support in the UK at the election. The Scottish Greens changed their minds and decided not to contest the seat in favour of the SNP though it was far from clear that Green-inclined voters in places like St Andrews would necessarily follow Green Party activists in supporting the SNP. In nearby Kirkcaldy and Cowdenbeath, held by Gordon Brown until he retired, Neale Hanvey was suspended from the SNP after social media posts were discovered in which he compared Israeli treatment of the Palestinian people to the holocaust and included an image of George Soros as a puppet master (BBC News, 2019). The SNP had taken the seat in 2015 but Labour had regained it in 2017 with a tiny majority of 259 over the SNP. The SNP leadership instructed local activists to campaign instead in North East Fife for Stephen Gethins. Hanvey and local activists ignored the SNP leadership and he fought the seat as an independent, although his name appeared on the ballot paper as SNP candidate as his suspension came after the close of nominations. He won the seat, defeating Labour’s Shadow Scottish Secretary with a 1243 majority. While the SNP lost a seat in the east of Scotland to the Liberal Democrats, it gained Jo Swinson’s East Dunbartonshire seat in the west. Swinson, UK Liberal Democrat leader, had lost the seat in 2015 to John Nicolson of the SNP but regained it in 2017. Nicolson chose not to stand in East Dunbartonshire but instead became SNP candidate in Ochil and South Perthshire (though he confused the two seats speaking at hustings in Ochil a fortnight before polling). He nonetheless won the seat from the Conservatives, while Amy Callaghan won his old seat for the SNP in a tight contest with the Liberal Democrat leader. Ian Murray managed to hold onto Edinburgh South for Labour with a significantly reduced but still substantial majority over the SNP and returned to being Labour’s only Scottish MP, a position he held after 2015. The constituency includes predominantly middle-class areas of Edinburgh such as Morningside and Colinton (its median income is 14% higher than the Scottish average) leading to exaggerated comments that Murray had the largest Tory majority in Scotland with reference to tactical voting to keep the SNP from taking the seat. Murray had a high profile as a critic of Jeremy Corbyn and strongly anti-Brexit (which is helpful in a constituency that voted 78% Remain). He is also avowedly anti-independence, but the constituency is less hostile to independence than it is to Brexit, and the way in which these two issues are navigated will likely determine his future success in the constituency. 3. Polls and expectations Polls consistently placed the SNP ahead of the other parties throughout the short parliament elected in 2017 (Figure 8.1). The SNP lead over its main challenger was under 10 points in only six polls early in the period between the 2017 and 2019 elections and was 20 points or more on five occasions. SNP support dipped in late 2017 and through the following year, with odd exceptions, it appeared to have passed its peak. But from early in 2019, its support increased and rarely fell much below 40%. Its 45% share of the vote in the election was higher than any poll had shown it throughout the short parliament. Labour and Conservatives competed to be the main challengers to the SNP in share of the vote, though the electoral system, which historically had worked to Labour’s great advantage, has been seriously undermining the party since 2015. Scottish Conservative support dipped over the winter of 2017–2018 and there was no evidence that the party was able to build on its 2017 election performance. The Tories appeared to have peaked in 2017. Even before Ruth Davidson’s resignation as leader in August 2019, the Scottish Tories looked likely to struggle to hold onto the seats they had gained, although they picked up some support in the polls towards the end of 2019 under the new Acting Leader, Jackson Carlaw. The party went into the election looking likely to lose seats but as the campaign progressed, the Conservatives became increasingly confident that they would hold and indeed gain seats, including Lanark and Hamilton East where the Tory candidate dispensed with a dog whistle and trumpeted her belief that the area was a ‘solid unionist royalist Rangers supporting heartland’ (The Times, 2019). But the constituency result, a 5,187 SNP majority, suggested that this old-style Protestant unionist vote was not so easily mobilised. On winning the Edinburgh Central Scottish parliamentary constituency in 2016, Ruth Davidson had acknowledged that she was ‘under no illusion that everybody who voted for me in that seat is a true-blue, dyed-in-the-wool Tory, and neither are they in places up and down Scotland’ but had voted for her and her party ‘to do a very specific job, and that is to hold the SNP to account’ (BBC News, 2016). The Tories had essentially rebalanced, if not quite renounced, their message as the Scottish Conservative and Unionist Party in favour of its unionist message. The independence referendum and continued importance of the issue provided the Tories with a fillip after decades of decline and an opportunity to build support for the party’s wider agenda. But the party played it safely under Davidson’s leadership, emphasising its unionism rather than seek to convert unionists who had voted for the party into Conservatives. Labour too had its challenges. The party made a recovery in 2017 winning six additional seats at the expense of the SNP. Labour came within 1.5 points of the Conservatives in 2017 but the party’s six-seat gains amounted to half of the number of extra seats captured by the Tories. Labour’s best showing in the polls, in the 18 months between the 2017 and 2019 elections, had come early in the parliament when it led the Tories for second place from the autumn of 2017 until spring 2018. But Labour failed to maintain that lead and slipped back into third place. The result of the European elections in May 2019 saw Labour support collapse in Scotland. The party came in fifth with 9.3%, behind the SNP (37.8%); Brexit Party (14.8%); Liberal Democrats (13.9%); and the Conservatives (11.6%). 4. Campaign foci Polls showed that Brexit was the dominant issue in the election in Scotland, as across the UK. Two weeks before polling day, 56% of Scottish respondents cited Brexit as the issue which would be ‘very important’ in helping decide how to vote. This was followed by health on 44% and independence/devolution coming in third on 34%. Brexit was the dominant issue even amongst those intending to vote SNP with 55% citing the issue and with independence and health on 46% and 45%, respectively. Brexit was even more emphatically the number one issue for Conservative-inclined voters, despite the Scottish Tories making opposition to independence/an independence referendum central to their campaign (Ipsos MORI, 2019). The SNP manifesto maintained that ‘An SNP election victory will be a clear instruction by the people of Scotland that a new referendum on independence should be held next year, on a precise date to be determined by the Scottish Parliament’ (Scottish National Party, 2019, p. 10) and insisted that the SNP already had a mandate for a second independence referendum given the material change in conditions brought about by Brexit. In the fortnight before the official launch of the campaign, SNP leader Nicola Sturgeon repeatedly used the #indyref2020 hashtag and on 27 October tweeted that ‘#indyref2020’ would be at the ‘heart of the @SNP campaign’. But the SNP messages were quite different during the campaign. Brexit dominated news releases issued by SNP headquarters and tweets from the SNP leader during the election and #indyref2020 was dropped. There was only one further explicit tweet from Sturgeon referring to #indyref after the official start of the campaign and that was in response to Labour’s changing position on an independence referendum. Labour had initially said it would oppose another independence referendum in its first term in office, then shifted to opposing one in its first two years in office, provoking the SNP leader to tweet that, ‘By the end of the week, at this rate, Corbyn will be demanding #indyref2020.’ Sturgeon’s references to a second independence referendum were at best ambiguous during the campaign. This typically involved urging voters to support the SNP ‘to put Scotland’s future in Scotland’s hands’ rather than arguing explicitly for a referendum and avoided suggesting that a referendum would be held in 2020. Even references to ‘independence’ were rare.3 There was little surprise in this shift in focus, given polling evidence that most voters, including many potential SNP voters, were not keen on a referendum in the immediate future. Scottish voters tended to divide evenly on whether there should be a further referendum in the next five years in Panelbase polls towards the end of 2019 (What Scotland Thinks, 2019). In an Ipsos MORI poll taken two weeks before the 2019 election, 43% strongly opposed and 7% opposed an independence referendum in the next five year and only 33% strongly supported and 9% supported one, while Scotland split 50:50 on independence. Sturgeon had failed to convince enough of her own supporters, far less anyone else, of the case for another referendum (Ipsos MORI, 2019). It was not surprising that the SNP quietly dropped direct references to a referendum during the election. A range of other matters was mentioned much more often in Sturgeon’s tweets but Brexit was the dominant issue and was linked to a range of policy matters. Escaping Brexit, escaping Boris Johnson and escaping the Tories featured heavily in a campaign that reversed past SNP campaign emphasis on positive campaigning (Pattie et al., 2011). But while the SNP might struggle to claim it had a mandate for an independence referendum in 2020 based on its campaign, it could at least take solace in the outcome of the Conservative campaign. The Tories emphasised opposition to an independence referendum, arguing persistently and unambiguously that if people voted SNP, they would be giving the SNP a mandate for a second referendum. The Conservative campaign was largely a rerun of recent Tory campaigns in Scotland. Campaign literature throughout Scotland was consistent to the point of uniformity leaving very limited space for other issues. The message was conveyed with military discipline and repetitively. Uniform campaign literature was delivered across Scotland with little scope for individual candidates to offer an alternative from the dominant message of opposition to an independence referendum. This included a facsimile of a hand-written letter to all voters from Ruth Davidson, even though she had already announced her resignation as leader. While the Scottish Tories could not avoid Brexit, given it was the dominant theme of the party across Britain, it posed challenges for them given that the leadership of the Scottish Tories and most of their Members of the Scottish Parliament (MSPs) had opposed Brexit in the 2016 referendum. During that referendum, Ruth Davidson, credited with having revived the fortunes of the party north of the border, had been a leading figure in the Remain campaign during the 2016 referendum and had not hidden her opposition to and indeed contempt for Boris Johnson. During the UK Tory leadership contest after Theresa May stood down, Davidson initially backed Sajid Javid, then Michael Gove after Javid was knocked out of the contest before supporting Jeremy Hunt, along with 18 of the 31 Scottish Tory MSPs, in a vain bid to block Boris Johnson. Davidson resigned as a leader at the end of August but her successor was not in post at the time of the election, leaving Jackson Carlaw, her deputy, as Acting Leader.4 Carlaw faced the election in 2019 not confirmed as party leader and with a low profile, in part due to the massive effort that the Scottish Conservatives and media had put into building the party around Ruth Davidson. Richard Leonard suffered similarly from a low profile. He became Scottish Labour leader in November 2017, in a contest in which he was portrayed as Corbyn’s candidate against Anas Sarwar, the ‘New Labour’ candidate. This contest had followed the sudden and unexpected resignation of Kezia Dugdale, who had been elected Scottish Labour leader in August 2015. Leonard had only been elected to Holyrood in 2016 as a List member for Central Scotland. He had little support amongst Labour MSPs and struggled to gain traction amongst the electorate. Willie Rennie, Scottish Liberal Democrat leader, had an advantage over other opposition leaders in Scotland and his own UK party leader in having been a leader since 2011. Surveys showed that Sturgeon had by far the most positive satisfaction ratings in Scotland, as Table 8.3 indicates. Table 8.3 Satisfaction with the Prime Minister/First Minister/party leaders (%) . Satisfied . Do not know . Dissatisfied . Boris Johnson 22 4 74 Jeremy Corbyn 22 9 69 Jo Swinson 27 25 48 Nicola Sturgeon 48 4 48 Jackson Carlaw 18 39 43 Richard Leonard 17 33 50 Willie Rennie 31 31 38 . Satisfied . Do not know . Dissatisfied . Boris Johnson 22 4 74 Jeremy Corbyn 22 9 69 Jo Swinson 27 25 48 Nicola Sturgeon 48 4 48 Jackson Carlaw 18 39 43 Richard Leonard 17 33 50 Willie Rennie 31 31 38 Source: Ipsos MORI for STV, 27 November 2019. Open in new tab Table 8.3 Satisfaction with the Prime Minister/First Minister/party leaders (%) . Satisfied . Do not know . Dissatisfied . Boris Johnson 22 4 74 Jeremy Corbyn 22 9 69 Jo Swinson 27 25 48 Nicola Sturgeon 48 4 48 Jackson Carlaw 18 39 43 Richard Leonard 17 33 50 Willie Rennie 31 31 38 . Satisfied . Do not know . Dissatisfied . Boris Johnson 22 4 74 Jeremy Corbyn 22 9 69 Jo Swinson 27 25 48 Nicola Sturgeon 48 4 48 Jackson Carlaw 18 39 43 Richard Leonard 17 33 50 Willie Rennie 31 31 38 Source: Ipsos MORI for STV, 27 November 2019. Open in new tab 5. The impact of the Scottish and Brexit referendums As noted above, the Scottish and Brexit questions overshadowed the election. How this played out for each party depended on the relative importance of each issue combined with the party’s position on these and how they each dealt with the challenges they faced. Taking the four main parties and two key issues, we can see how each party performed across the two questions (Table 8.4). Table 8.4 Voting in the 2019 UK general election in Scotland by preferences on Scottish and European questions (%) . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 1 27 23 67 Labour 9 16 29 14 Liberal Democrats 2 3 20 7 SNP 88 50 25 9 Other 1 3 1 3 N 550 250 636 362 . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 1 27 23 67 Labour 9 16 29 14 Liberal Democrats 2 3 20 7 SNP 88 50 25 9 Other 1 3 1 3 N 550 250 636 362 Source: Scottish Election Study 2019. Open in new tab Table 8.4 Voting in the 2019 UK general election in Scotland by preferences on Scottish and European questions (%) . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 1 27 23 67 Labour 9 16 29 14 Liberal Democrats 2 3 20 7 SNP 88 50 25 9 Other 1 3 1 3 N 550 250 636 362 . Yes Remain . Yes Leave . No Remain . No Leave . Conservative 1 27 23 67 Labour 9 16 29 14 Liberal Democrats 2 3 20 7 SNP 88 50 25 9 Other 1 3 1 3 N 550 250 636 362 Source: Scottish Election Study 2019. Open in new tab The SNP won 88% support (up from 77% in 2017) amongst those who supported its policies of independence and remaining in the EU but won only half the vote of those who supported independence but supported leaving the EU, down from 58% in 2017 (see Tables 8.2 and 8.4 for 2017 and 2019 figures). Over a quarter of voters who wanted independence and Brexit voted for the Conservative Party, up from 15% in 2017, suggesting that such voters saw Brexit as more important than independence. The Conservatives won 67% amongst those who accepted its position of opposing independence and EU membership, only slightly down in 2017. It is no surprise to find that the Conservatives again hardly polled at all amongst supporters of independence and the EU though the SNP managed to win almost 1 in 10 of those who opposed both independence and EU membership, slightly up in 2017. The Liberal Democrats might have expected to do well amongst opponents of independence who supported remaining in the EU given the clear position that party had adopted on these issues. In fact, the Liberal Democrats won only one in five such voters, fewer than any of the other three main parties amongst this group, though an improvement on 2017 when they won 13%. Labour’s best result was amongst those who opposed independence and supported membership of the EU, but the party only won 29% of this group, the largest collection of voters, significantly down from the 42% in this group won in 2017. The SNP shift in focus onto Brexit during the campaign played down its support for a second independence referendum. This appealed to the majority on the issue that had the highest salience for the public. Once the election was over, however, the SNP returned to its demand for a referendum. In a speech delivered the day after the election, Nicola Sturgeon claimed that the SNP victory ‘renews, reinforces and strengthens the mandate we have from previous elections to offer the people of Scotland a choice over their future’. Given the lack of priority during the campaign attached to a second independence referendum, the SNP proposed a novel form of mandate theory to claim there should be a second referendum, by noting that the Conservatives had ‘said that a vote for them was to deny people in Scotland the right to decide our own future’ (SNP News release, 13 December 2019). As the election receded into the past, the SNP increasingly claimed a mandate for a referendum in 2020 on the basis of the election result. What Table 8.4 does not capture, however, is the differing ways that the two constitutional issues affected political parties. This is more obvious when we look at three parties in particular: the SNP, Conservatives and Labour. In their early analysis of 2019 Scottish Election Study results, Johns et al. (2020) noted that Conservative voters were far more likely to say that Brexit was the most important issue when determining how they would vote, indeed they were six times more likely to mention Brexit rather than independence. For the SNP, however, independence was seen as most important by over one-third but Brexit was cited by just over a quarter. This suggests that the SNP were taking a significant body of independence supporters for granted and focussing during the campaign on reaching out to others on Brexit. The two constitutional issues together helped to earn support for the SNP. These are, it should be noted, self-reported assessments, separate from models of voting behaviour as such. Labour, however, is in a slightly different position. Labour voters were more likely to say that Brexit rather than independence determined their view (over 40% versus over 15%). When the Scottish Election Study team modelled support for Labour defectors, however (Henderson et al., 2020), it is clear that Labour’s failure to capture the shift in support for independence—within the electorate as a whole but also amongst its previous voters—was, all things being equal, an important source of declining support. Critically, it was not how individuals had voted in 2014 but how they felt about independence in 2019 that cost Labour. Those who were least Left-wing and those who disliked Corbyn were also more likely to defect to other parties. This makes clear that it was not Labour’s ambivalence on Brexit but its hostility to independence, that cost it most votes in 2019 which, coming on top of losses incurred since 2015, has left the party in a parlous position. 6. Conclusion As Scotland moved on to the Holyrood election in 2021, the same issues that dominated the political agenda in December 2019 looked likely to determine the outcome of the election. Arguments over another independence referendum and about Brexit and its form looked set to continue to polarise opinion. Other matters seemed likely to be crowded out or subsumed within these constitutional questions. The SNP and Conservatives share an interest in keeping the Scottish and European questions uppermost in the electorate’s mind but this could also present dangers for each party. The SNP leadership could face pressure from within the party and wider movement if polls continue to show that support for independence remains static or if it is unable to deliver a second referendum. The Conservatives face dangers if Brexit turns out, or more accurately is perceived to be damaging, especially in those parts of Scotland in which it has made gains in recent years or if its emphasis on independence contributes to the SNP’s claim that it has a mandate for a second referendum. Labour has changed its leader in London but with a Corbyn-supporting Scottish leader and a new Deputy Scottish leader on the Blairite wing of the party, it may continue to expend energy on internal battles. As with the Liberal Democrats, Labour has yet to articulate a position on the Scottish and European questions that might allow it to cut out a distinct and convincing alternative to the positions of the SNP and Conservatives. The impact of the COVID-19 crisis is difficult to assess at the time of writing but the recent past suggests that this global pandemic will be presented to the electorate through these constitutional prisms. While this much seems likely, recent history suggests that the turbulent nature of Scottish politics makes prediction hazardous. Tribes and turbulence can have unpredictable dynamics. Funding Scottish Election Study (ESRC award ES/N018060/1). Figure 8.1 Open in new tabDownload slide Opinion polls in Scotland, June 2017–December 2019 Source: Electoral Calculus. Figure 8.1 Open in new tabDownload slide Opinion polls in Scotland, June 2017–December 2019 Source: Electoral Calculus. Footnotes 1 The extent to which Holyrood elections are second order was challenged from the first elections to the Scottish Parliament. Paterson et al. (2001, p. 44) concluded that the ‘first Scottish election was neither clearly first order nor was it undoubtedly second order’. 2 It is sometimes mistakenly assumed that the Conservatives were the largest party in 1955 but this is an ahistorical reading of results that lumps together support for the Scottish Unionist Party (forerunner of today’s Scottish Conservative and Unionist Party) and other Unionists/Liberal Unionists. Labour was Scotland’s largest party since 1935 until 2015. 3 When independence appeared, it was in retweets from Sturgeon. At the start of the campaign, Sturgeon responded to a tweet by a party colleague who had paid tribute to a party activist who had died who had ‘worked incredibly hard for @theSNP and the independence cause’. 4 Carlaw would go on to win the election as Scottish Tory leader in February 2020. References BBC News ( 2016 , 6 May) ‘Holyrood 2016: Scottish Tory Leader Ruth Davidson Wins Edinburgh Central’, accessed at https://www.bbc.co.uk/news/election-2016-scotland-36219380 on 19 April 2020. BBC News ( 2017 , 10 June) ‘Davidson Says Scots Tory MPs Will Argue Scotland’s Case “forcefully”’, accessed at https://www.bbc.co.uk/news/uk-scotland-scotland-politics-40232374 on 19 April 2020. BBC News ( 2019 , 29 November) ‘SNP Members Urged Not to Campaign for Axed Candidate’, accessed at https://www.bbc.co.uk/news/election-2019-50711777 on 19 April 2020. Convery A. ( 2014 ) ‘ The 2011 Scottish Conservative Party Leadership Election: Dilemmas for Statewide Parties in Regional Contexts ’, Parliamentary Affairs , 67 , 306 – 327 . Google Scholar Crossref Search ADS WorldCat The Courier ( 2017 , 12 June) ‘Exclusive: Scottish Conservative MPs to Be Led by David Mundell and Report to Ruth Davidson’, accessed at https://www.thecourier.co.uk/fp/news/politics/scottish-politics/446686/exclusive-scottish-conservative-mps-led-david-mundell-report-ruth-davidson/ on 19 April 2020. Express ( 2017 , 5 December) ‘Davidson Wades into Brexit Row as May Battles Cabinet’, accessed at https://www.express.co.uk/news/uk/888226/brexit-deal-northern-ireland-scotland-single-market-ruth-davidson-theresa-may on 19 April 2020. The Guardian ( 2017 , 11 June) ‘Scottish Tories expected to act as a bloc to protect Scotland’s interests’, accessed at https://www.theguardian.com/politics/2017/jun/11/scottish-tories-expected-vote-bloc-protect-scotlands-interests on 19 April 2020. Henderson A. , Mitchell J. ( 2018 ) ‘Referendums as Critical Junctures? Scottish Voting in British Elections’. In Tonge J. , Leston-Bandeira C. , Wilks-Heeg S. (eds.) Britain Votes 2017 , Oxford , Oxford University Press , pp. 109 – 24 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Henderson A. , Johns R. , Larner J. , Carman C. ( 2020 ) ‘ Scottish Labour as a Case Study in Party Failure: Evidence from the 2019 UK General Election in Scotland ’, Scottish Affairs , 29 , 127 – 140 . Google Scholar Crossref Search ADS WorldCat The Herald ( 2019 , 6 August) ‘Labour Would Not Block a Second Scottish Independence Referendum’, accessed at https://www.heraldscotland.com/news/17819772.john-mcdonnell-labour-not-block-independence-referendum/ on 19 April 2020. Ipsos MORI ( 2019 , 27 November) ‘Scottish Political Monitor for Scottish Television’, accessed at https://www.ipsos.com/ipsos-mori/en-uk/scotland-voting-intention-snp-pole-position-run-election on 20 April 2020. Johns R. , Henderson A. , Larner J. , Carman C. ( 2020 ) ‘ Brexit or Independence: Scotland’s General Election ’, Political Insight , 11 , 28 – 31 . Google Scholar Crossref Search ADS WorldCat Mitchell J. ( 1990 ) Conservatives and the Union , Edinburgh , Edinburgh University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mitchell J. , Bennie L. , Johns R. ( 2011 ) The Scottish National Party: Transition to Power , Oxford , Oxford University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Paterson L. , Brown A. , Curtice J. , Hinds K. , McCrone D. , Park A. , Sprogston K. , Surridge P. ( 2001 ) New Scotland, New Politics ?, Edinburgh , Polygon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pattie C. , Denver D. , Johns R. , Mitchell J. ( 2011 ) ‘ Raising the Tone? The Impact of “Positive” and “Negative” Campaigning on Voting in the 2007 Scottish Parliament Election ’, Electoral Studies , 30 , 333 – 343 . Google Scholar Crossref Search ADS WorldCat Scottish National Party ( 2019 ) Stronger for Scotland. SNP manifesto for the 2019 election, Edinburgh, SNP. The Telegraph ( 2017 , 10 June) ‘Ruth Davidson Planning Scottish Tory Breakaway as She Challenges Theresa May’s Brexit Plan’, accessed at https://www.telegraph.co.uk/news/2017/06/09/ruth-davidson-planning-scottish-tory-breakaway-challenges-theresa/ on 19 April 2020. The Times ( 2020 , 9 December) ‘Brexit and Scottish Independence Have the Marginal Seat of Lanark & Hamilton East Feeling Blue’, accessed at https://www.thetimes.co.uk/article/brexit-and-scottish-independence-have-the-marginal-seat-of-lanark-amp-hamilton-east-feeling-blue-09k8qjnpx on 19 April 2020. What Scotland Thinks ( 2019 ) accessed at https://whatscotlandthinks.org/questions/when-do-you-think-anotherscottish-independencereferendum-should-be-held-asked-a/ on 20 April 2020. © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Bradbury,, Jonathan

doi: 10.1093/pa/gsaa025pmid: N/A

It is often forgotten that Boris Johnson cut his general election teeth, aged 32, standing as the Conservative candidate for Clwyd South in north Wales in the 1997 general election. He managed a respectable second place in losing to Labour amidst the New Labour landslide and it was to be another four years before he finally entered parliament. But as Johnson hit the campaign trail yet again in November 2019 it was as an incumbent Prime Minister seeking the parliamentary majority that had eluded his predecessor, Theresa May, in 2017. Just as May had noted, Johnson was aware that Wales offered a possible source of the gains that were needed, with several Labour marginal seats across both north and south Wales. While Labour still held 28 of the 40 seats in Wales and the Conservatives just eight, the Conservative vote share had steadily risen with each general election since 1997, to reach just over a third in 2017. Labour had proven themselves time and again a resilient force in Wales but 2019 looked as if it could be the time the dam broke. When the results appeared on the morning of 13 December, Labour was still the largest party in Wales, but in the election as a whole Johnson emerged with an 80-seat majority and part of the story of that success could be found in Wales. The dam had not fully broken but the Conservatives had achieved their highest vote share in Wales since 1900 and their highest number of MPs in Wales since 1983. Simon Baynes, the Conservative candidate in Clwyd South, had finally succeeded where his boss had failed 22 years before and Conservative victories across north Wales contributed to the knocking down of key parts of Labour’s ‘red wall’ that helped to achieve the Conservative majority. How did the Conservative erosion of Labour’s fortress happen, and how did political actors view its likely causes and consequences? In answering these questions, we must address the ways in which the UK election campaign played out in Wales. To do this, the chapter will first review the election campaign and party strategies in Wales and then examine party performance, before turning to the post-election debate. The focus here is on revealing what the parties in Wales did, analysing perceptions of the causes of the result and its implications. 1. The campaign It is important to note of course that Boris Johnson was fighting the long election campaign from the moment he became Prime Minister in July 2019, heading a minority Conservative government. His first target had been the Brexit Party, which in Wales, as in England, had emerged victorious in the European elections in May. Johnson set about establishing that the Brexit Party was no longer needed and that the Conservatives under him would deliver on the defining issue of the day: negotiation of withdrawal from the EU and then the UK’s departure from it. Following this, he set about winning the battle in the polls against Labour by first, promising to deliver Brexit, and secondly, offering a targeted set of spending priorities on the NHS, police and education. In contrast, Jeremy Corbyn as Labour leader fashioned a complex position over Brexit to try and address the conundrum that most Labour MPs wanted to remain in the EU while most Labour seats had voted to leave in the 2016 referendum. The policy involved renegotiation, followed by a second referendum, Johnson’s hopes in Wales were bolstered by the fact that, to the horror of the Labour establishment, there had been a 52.5% Leave vote in Wales in the 2016 referendum, just as there had been in the UK as a whole. The Labour Party, successful in holding back Theresa May’s efforts in 2017 in Wales under Welsh First Minister, Carwyn Jones, was now under the leadership of Mark Drakeford. Plaid Cymru offered a potential heightened threat, having come second in the European elections in Wales, ahead of Labour, on a Remain platform and were now under a charismatic new leader, Adam Price. But the polls had been steadily encouraging and the expectation was that Brexit would be the key issue, a matter on which the Conservatives hoped to capture all the Leave support against a set of parties either not offering clarity or still supporting Remain. When the election campaign began the hopes of the Conservatives appeared to be borne out. The first Welsh opinion poll in early November indicated a Labour lead over the Conservatives of just 1% (Table 9.1). There was still a reasonably strong showing for the Brexit Party, third in the poll on 15%. The projection from the poll, on the basis of a uniform swing since the 2017 election, was that the Conservatives would gain nine seats from Labour, including five seats in the north (Alyn and Deeside, Clwyd South, Delyn, Vale of Clwyd and Wrexham) and four in the south (Newport West, Cardiff North, Bridgend and Gower) (Scully, 2019). This would have taken the Conservatives to 17 out of Wales’ 40 seats. It was also predicted that Labour would lose Ynys Mon to Plaid Cymru, meaning that they would lose ten seats overall, and be reduced to a total of 18 seats. Plaid Cymru were predicted to stay on four seats as the poll also predicted that the Liberal Democrats would take Ceredigion from them. As revealing were Welsh views on who would make the best Prime Minister, with 41% choosing Boris Johnson and just 26% Jeremy Corbyn. Voters’ feelings about who they blamed for Brexit having not yet happened did not point to any particular electoral implications, as 46% blamed MPs on all sides, 20% Boris Johnson and the Conservative government and 14% Jeremy Corbyn and the Labour party. Table 9.1 Welsh Political Barometer opinion polls: shares of the vote during the election campaign (%) . Cons . Lab . Lib Dem . Plaid Cymru . Brexit . Others . 31 Oct–4 Nov. (N=1,136) 28 29 12 12 15 4 22–25 Nov. (N=1,116) 32 38 9 11 8 2 6–9 Dec. (N=1,020) 37 40 6 10 5 2 . Cons . Lab . Lib Dem . Plaid Cymru . Brexit . Others . 31 Oct–4 Nov. (N=1,136) 28 29 12 12 15 4 22–25 Nov. (N=1,116) 32 38 9 11 8 2 6–9 Dec. (N=1,020) 37 40 6 10 5 2 Source: ITV Cymru Wales and Cardiff University, carried out by YouGov. Open in new tab Table 9.1 Welsh Political Barometer opinion polls: shares of the vote during the election campaign (%) . Cons . Lab . Lib Dem . Plaid Cymru . Brexit . Others . 31 Oct–4 Nov. (N=1,136) 28 29 12 12 15 4 22–25 Nov. (N=1,116) 32 38 9 11 8 2 6–9 Dec. (N=1,020) 37 40 6 10 5 2 . Cons . Lab . Lib Dem . Plaid Cymru . Brexit . Others . 31 Oct–4 Nov. (N=1,136) 28 29 12 12 15 4 22–25 Nov. (N=1,116) 32 38 9 11 8 2 6–9 Dec. (N=1,020) 37 40 6 10 5 2 Source: ITV Cymru Wales and Cardiff University, carried out by YouGov. Open in new tab The Welsh Conservative campaign had early problems. Alun Cairns resigned as Secretary of State on 6 November, following accusations that he had supported the nomination of a party staff member, Ross England, as a Conservative candidate for the National Assembly despite knowing about his role in the collapse of a rape court case. Cairns was not replaced and as a result there was no Welsh figurehead to the campaign. At the same time, they had a string of candidacy difficulties. Cairns continued to stand as a candidate in the Vale of Glamorgan and attracted considerable media attention. Francesca O’Brien, the candidate in Gower, received calls to stand down over reports that in 2014 she had written on her Facebook page, after watching the TV programme Benefits Street: ‘my blood is boiling, these people need putting down’ (Western Mail, 5 November 2019: 8). The party in Wales stood by her nonetheless. The Conservatives were also seeking to get two seats back that they had won in 2017 but subsequently lost. In Aberconwy, Guto Bebb gave up the party whip over Brexit but was standing down at the election. The Conservatives had also lost Brecon and Radnorshire as recently as August 2019 to the Liberal Democrats, after the 2017 Conservative victor, Chris Davies, was convicted in April 2019 of submitting false expense claims. Davies re-emerged as the party’s candidate in Ynys Mon, only to pull out just prior to the election, leaving the party to scramble around for a new candidate. The Welsh Conservative Party’s manifesto, ‘Get Brexit Done, Unleash Wales’ Potential’, was launched in the target seat of Wrexham on 24 November (Welsh Conservative Party 2019) and the two themes, of getting Brexit done and unleashing a tide of investment in Wales in the aftermath, were repeated throughout the campaign. This was backed up by a strong focus on developing the Welsh economy and passing on the targeted funding increases on the police, health and education to the Welsh budget to improve public services. In so doing they also promoted some specific policies that suggested the direct interest of a UK Conservative government in improving Wales’ fortunes over the heads of the Labour-led government in the National Assembly for Wales. This drew on the already agreed UK government-backed city or growth deals for Cardiff, Swansea, mid Wales and north Wales. The election saw the promotion specifically of the Marches Growth deal to improve cross-border road infrastructure. Closely related to this, Johnson, along with other key party visitors to Wales, including the then chancellor Sajid Javid, repeatedly focused on the transport needs of south and north Wales. In the south they stressed that even though the Welsh government had abandoned the M4 relief road plan to explore alternatives, the UK government had allocated the funding necessary from the UK level and wanted to make it happen. Speaking on a campaign trip to Caerphilly, Johnson told journalists that ‘we’re going to rely on our superior powers of persuasion. I think people looking at the congestion now…the Brynglas tunnels, it’s been famously said, are the nostrils of the Welsh dragon which are currently stuffed-up. We intend to apply the Vicks inhaler. I think the overwhelming case is to get it done’ (Western Mail, 12 December 2019). The Conservatives also promised to fund a West Wales Parkway railway station outside Swansea and along the north Wales coast they pledged to upgrade the A55 through the National Road Fund. The final component of the narrative was damning criticism of Labour’s left ideology in Wales and their record both on the economy and in public service performance, as revealed, as one example, in relatively poor education results. In contrast, as Johnson said in his Wrexham address, ‘We will put Wales first, we will put you first… the Welsh dragon will roar louder than ever before’ (Western Mail, 25 November 2019). On the face of it, Labour’s Welsh campaign was a fairly coherent alternative. It was led by Mark Drakeford and provided a close fit with policies being championed by the British leadership. The manifesto, Standing up for Wales, also launched in Wrexham, committed Labour to a green industrial revolution with investment in renewable energy, promised a boost to small and medium enterprises through the Development Bank Wales, and publicly owned railways (Labour Party 2019). Corbyn’s spending promises offered an extra £3.4 billion for the Welsh government budget, to be spent on health, education and other public services, allowing the provision of free social care and the scrapping of tuition fees. There were commitments to end the pay cap for the public sector, get rid of universal credit, introduce free broadband for every home and business and legislate for a real living wage of at least £10 an hour. Drakeford was proud that the Welsh government had been the first in the UK to declare a climate emergency and this priority ran through everything the party would do in government in Wales. In choosing the manifesto title, ‘Standing up for Wales’, Drakeford consciously repeated the slogan used by his predecessor, Carwyn Jones, at the 2017 election. Jones had used this to suggest that in voting for Labour candidates, voters should see their vote as primarily for MPs who would represent Wales at Westminster and less as MPs who would support Jeremy Corbyn as Prime Minister. On this occasion, Drakeford did in part stand out from the Corbyn leadership by identifying a specifically Welsh position over Brexit. Drakeford had originally supported Corbyn’s Brexit stance but he was alone in his cabinet in doing so and changed his mind. In contrast to Corbyn’s neutrality, if and when a second referendum was held by a Labour government, Drakeford (2019, p. 21) committed the Welsh party to clearly campaigning for Remain, insisting ‘Welsh Labour will campaign wholeheartedly and unapologetically to Remain, because we believe Wales is better off as part of Europe’. As Shipton (2019a) observed, this was significant because ‘belated as this was, it enabled strongly Remain Welsh Labour candidates to appear credible when they said they had the backing of their party in a way the European Parliament candidates hadn’t…Welsh Labour simply declared itself a Remain party’. Other than over Brexit though, Drakeford did not depart from the British party line. He had been the first Welsh Labour cabinet member to support Corbyn’s candidacy in 2015 and strongly supported Corbyn now. He declined to redevelop Carwyn Jones’ 2017 rhetorical narrative of appealing to voters to support Wales in voting for Labour rather than Jeremy Corbyn for prime minister. The Corbyn agenda largely stood to succeed or fail on its own merits in Wales, just as it did anywhere else. Both the Conservatives and Labour were parties standing for office that had to incorporate their positions on Brexit within broader programmes for government. In contrast, the other parties defined themselves more explicitly in relation to how they would influence the handling of Brexit by whichever government took office. On the one hand, the Brexit Party provided the option to vote for keeping the political class to their promise on leaving the EU. As the election proceeded, the party leader, Nigel Farage, stood candidates down in many seats where the Conservatives were expected to win, to leave a focus on Labour held Leave seats. Their argument was that the Brexit Party was in the best position to challenge Labour in such seats. This had big implications for Wales as 21 of the 29 seats that voted Leave in 2016 had a sitting Labour MP. Welsh Labour’s hardening of their pro-Remain position in a second referendum underlined the Brexit Party’s resolve to campaign in these seats, threatening both Labour’s position, as well as, potentially at least, Conservative prospects for taking any of these where they were target seats. Throughout the campaign the Conservatives attacked the Brexit Party for threatening to split the Leave vote and allow Labour MPs to sneak back in. On the other side, on 7 November it was announced that the pro-EU electoral pact between the Liberal Democrats and Greens, seen in England across 49 seats, was being extended to Wales to include Plaid Cymru. The pact covered 11 seats in all, meaning that in each case two of the parties would stand aside for the other to stand as the clear pro-Remain candidate for a second referendum. Plaid Cymru were the biggest winners from this, gaining a free run in seven seats, the Liberal Democrats in three and the Greens one. This replicated the electoral pact between the three parties in the Brecon and Radnorshire by-election held earlier in 2019, in which the Liberal Democrat candidate, Welsh party leader Jane Dodds, had been given a free run and won. The declared intention was to give a Remain voting candidate the biggest chance of victory in each case; but the anti-Brexit coalition was always incomplete without Labour joining. This was never a realistic option given Labour’s history in Wales and Labour actively sought to blunt its appeal. A Welsh Labour spokesperson suggested that at its core the ‘pact appears to be nothing more than Plaid and the Lib Dems propping each other up in seats they’re worried about losing. Plaid are worried about Arfon. The Lib Dems don’t think they’ll hold Brecon and Radnorshire’ (BBC, 2019). The pact did not cover Ceredigion, where Plaid Cymru held the seat and the Liberal Democrats were the main competitors. The pact was not supported by everyone, with the Liberal Democrat candidate in Pontypridd resigning and running as an independent. Among the electoral pact partners, Plaid Cymru achieved the most visibility during the campaign, led ably by both the Assembly party leader, Adam Price, and the UK parliamentary leader, Liz Saville Roberts. As well as seeking to defend their four seats, two of which were very marginal, they had the modest target of trying to win one more, Ynys Mon, where they launched their campaign on 4 November. As well as campaigning against Brexit, they consistently highlighted the Corbyn leadership’s greater interest in higher spending in Scotland than in Wales, as well as the deficiencies in Labour’s Welsh government record. Their manifesto, Wales, It’s Us (Plaid Cymru, 2019) promoted policies focused on an economic strategy based on investing in renewable energy, infrastructure and digital technology; a free national social care service; universal free child care and a new £35 a week payment for every child in low-income families; a programme of affordable homes investment and rent relief for people who pay more than 30% of their income on rent; devolution of the police and welsh justice system; and the creation of a new £50 million crime prevention fund to recruit 1,600 new police officers. As usual, Plaid were competing on the same left of centre class and nation ground as Labour but they were realistic that there were still very few constituencies where they had an established platform on which to compete. Adam Price’s eye was on preparing the ground for a bigger shot at the 2021 Assembly elections. The campaign overall lacked drama and colour. Perhaps this was a function of it being a winter election and politicians being tired at the end of a long year. Leaders’ debates also failed to take off. These were curious affairs, as the Conservatives called on David Davies, Conservative chair of the Welsh Affairs select committee, to represent them in lieu of the departed Cairns. Labour, by custom put up their shadow secretary of state, as opposed to the Labour First Minister. But on this occasion the shadow secretary of state, Christina Rees, was not deployed, nor was the possible alternative of Mark Drakeford. Instead, the party preferred to use Nick Thomas Symonds, MP for Torfaen. Pitched against Adam Price and Jane Dodds, the still very unknown leader of the Liberal Democrats, it all looked a bit random; more like occasional editions of the BBC’s Question Time than the major set piece election events such debates originally were intended to be. As the campaign went on Labour under Corbyn was at least rallying on the campaign trail and in Wales, as usual, the party was putting in a large organisational effort. A second opinion poll, released on 25 November, showed Labour up nine points from the first poll and the Conservatives up four points (see Table 9.1). Labour’s relative improvement appeared to be limiting the likely damage as at this point just four seats were predicted to change hands, two in north Wales (Vale of Clwyd and Wrexham) and two in south Wales (Cardiff North and Gower). Corbyn was clearly more popular than at the start of the campaign. The poll indicated that 38% of Welsh electors saw Johnson as the best Prime Minister and 35% Corbyn. By now, it was apparent that there was a similar trend happening in Wales as in England in that the two main parties were squeezing the smaller ones, notably the Brexit Party. However, the third and final opinion poll, released in early December, suggested that in comparison with the 2017 election there had been no Corbyn surge, and there was no particular mitigating Welsh Labour appeal. The third poll predicted a further two point rise for Labour to 40% but a five point rise for the Conservatives to further squeeze the other parties. Labour was again predicted heavy losses, this time eight seats: five in north Wales (Wrexham, Vale of Clwyd, Delyn, Clwyd South and Alyn and Deeside) and three in south Wales (Bridgend, Gower and Cardiff North). By this point, open criticism of the Labour campaign specifically in Wales was beginning to break out into the press. Welsh Labour made a strong case for how hard they had fought the election, in that they had targeted resources on defensive Labour/Tory marginal and target seats, with a focus generally on both north Wales and south Wales seats. They claimed that there had been over 2,000 campaign events, over 150 key campaign visits and strenuous efforts at doorstep and telephone canvassing. Amidst criticism of Drakeford personally, it was claimed that he had visited 30 constituencies. But critics pointed out how little Corbyn had visited Wales during the campaign; that Drakeford had been pretty invisible; and that the ‘Standing up for Wales’ campaign had lacked substance and had not been well communicated. Organisational problems were raised: that the campaign launch had been in Cardiff South and Penarth, convenient to the Assembly, but a safe seat and with very few people present; leaflets had not been printed in time to deliver before postal votes were sent out; there had been a shortage of leaflets in key seats; and the example was raised of busing campaigners into campaign in Arfon, which was not a key seat. On the eve of the poll a senior Welsh Labour politician claimed: ‘Lessons have to be learnt. Members and staff have worked really hard but there is a lack of strategic leadership within Transport House’. Communications staff in Transport House, Welsh Labour’s headquarters in Cardiff, also briefed the media that Corbyn had made it harder for them in Wales. (Western Mail, 11 December 2019). The omens on election eve were that Brexit, a campaign that did not take off and organisational problems might well make the difference this time. 2. Party performance Labour were the largest party in Wales but they did indeed lose six seats (see Table 9.2). In the south they limited the damage to Bridgend but in the north they lost a clutch of seats: Ynys Mon, Vale of Clwyd, Clwyd South, Delyn and Wrexham. The northern seats were seen as part of the hitherto apparently unbreachable Labour ‘red wall’ and the loss of Wrexham was particularly totemic, having been held by the party since the 1930s. Apart from Alyn and Deeside, in north east Wales, held with a majority of just 213 votes, all of Labour’s seats were now across south Wales. In the seats held, Labour’s strongest performances came in the four Cardiff seats, with Cardiff Central the safest Labour seat in Wales, with a majority of 17,179. However, even in all the seats Labour retained, there was a swing to the Conservatives. In the south east this included large swings in Torfaen (8.3%), Newport East (8.1%) and Islwyn (7.9%); in the valleys there was a 7.7% swing in the Rhondda, 7.9% in Merthyr Tydfil and Rhymney, and 7.6% in Ogmore; and in the industrial south west swings in Aberavon of 8.6%, Llanelli 8.8% and Neath 8.8%. Overall, Labour’s share of the vote in Wales of 40.9% was eight points down on 2017. This was still better than their results in Wales in 2010 and 2015 and they were still the largest party in Wales. Nevertheless, their seat tally of 22 of the 40 seats was the party’s worst since 1983. It was a grim night for the party. Table 9.2 UK General Election Result in Wales, 2019 . Seats . Change from 2017 . Vote share (%) . Change in % vote share since 2017 . Conservative 14 +6 36.1 +2.5 Labour 22 −6 40.9 −8.0 Lib Democrat 0 − 6.0 +1.5 Plaid Cymru 4 − 9.9 −0.5 Brexit 0 − 5.4 +5.4 Others 0 − 1.7 −0.8 . Seats . Change from 2017 . Vote share (%) . Change in % vote share since 2017 . Conservative 14 +6 36.1 +2.5 Labour 22 −6 40.9 −8.0 Lib Democrat 0 − 6.0 +1.5 Plaid Cymru 4 − 9.9 −0.5 Brexit 0 − 5.4 +5.4 Others 0 − 1.7 −0.8 Open in new tab Table 9.2 UK General Election Result in Wales, 2019 . Seats . Change from 2017 . Vote share (%) . Change in % vote share since 2017 . Conservative 14 +6 36.1 +2.5 Labour 22 −6 40.9 −8.0 Lib Democrat 0 − 6.0 +1.5 Plaid Cymru 4 − 9.9 −0.5 Brexit 0 − 5.4 +5.4 Others 0 − 1.7 −0.8 . Seats . Change from 2017 . Vote share (%) . Change in % vote share since 2017 . Conservative 14 +6 36.1 +2.5 Labour 22 −6 40.9 −8.0 Lib Democrat 0 − 6.0 +1.5 Plaid Cymru 4 − 9.9 −0.5 Brexit 0 − 5.4 +5.4 Others 0 − 1.7 −0.8 Open in new tab Meanwhile, Boris Johnson could rightly see the election in Wales as a cause for celebration. The party regained the two seats that they had lost between the 2017 and 2019 elections, Aberconwy and Brecon and Radnorshire. Beyond that they won the six seats from Labour to take 14 in all; Bridgend in the south and the five north Wales seats, including Ynys Mon which had been a Plaid Cymru target. Local constituency difficulties largely melted away as Alun Cairns won the Vale of Glamorgan with an increased majority, though the problems in Gower were a factor in them not winning the seat back. This meant that the Conservatives held all but two of the seats across north Wales, and down the eastern half of Wales you could walk continuously from Wrexham in the north to Monmouth in the south and never leave Conservative held seats. In addition, they held two of the Pembrokeshire and Carmarthen seats in the far south west. The Conservatives lay in second place in 21 of Labour’s 22 seats, having reduced Labour’s majorities in all. They also cut Plaid Cymru’s majority in Carmarthen East to less than 2,000 votes. The Conservatives came out of the election looking genuinely like a competitive party nationwide. Overall, the Conservative vote share of 36.1% represented a further 2.5 percentage point increase from 2017 and their best share of the vote since 1979. Labour may have still been the largest party in Wales but the Conservative performance impressed. The results reflected a squeeze of support for other parties. This left Plaid Cymru deeply disappointed at once again not making a breakthrough. They took comfort from increasing their majorities in Arfon at the expense of Labour and Ceredigion at the expense of the Liberal Democrats, as well as holding on to their two other seats. But beyond winning their four seats they came second nowhere and third in 15 seats spread across north and south Wales. Plaid came third even in their top target seat of Ynys Mon. The Liberal Democrats virtually disappeared as a serious party, coming second only in two seats, competitive only in Brecon and Radnor and third in only nine seats (including Ceredigion, where they were passed by the Conservatives). Perhaps the most significant of the results for the smaller parties was that for the Brexit Party. Their vote was squeezed down to 5.4% by election day, with some votes going to the Conservatives and some to Labour as the campaign wore on. Even so their performance was 3.4 percentage points higher than UKIP in 2017 and included a second place in Blaenau Gwent and nine third place finishes, all in Labour held seats in south Wales. It is hard to say conclusively, but the Brexit Party’s performance may have made the difference in four seats where Labour held on from the Conservative challenger by fewer votes than were won by the Brexit party candidate coming in third place. These were one in the north (Alyn and Deeside) and three in south Wales (Newport East, Newport West and Torfaen). But for Nigel Farage’s insistence that the Brexit Party had a role to play in the election, the Labour and Conservative parties may have ended up even on 18 seats each. This reflects more broadly that the election was a night of fine margins. These four Labour holds were all with majorities of less than 2,000 votes but the six Conservative gains from Labour were also all with majorities of 2,131 votes or fewer. So often in tight elections the small margins had broken Labour’s way. The Brexit Party may have saved Labour even more blushes, but overall the fine margins for once favoured the Conservatives. In the first election poll, 60% of voters had said that Brexit was the most important issue to them and consistently during the campaign it remained the top issue, followed by health. In reflecting on the results further it would appear that they seem to suggest that the party’s positions on the issue of Brexit helped the Conservatives the most (see Table 9.3). In the 2016 European Union referendum, 29 of Wales’ constituencies had voted Leave. Seven of the eight seats that the Conservatives retained were Leave voting seats (the exception was Monmouth) and all six of the seats that the Conservatives won from Labour were Leave voting seats. Labour retained 15 Leave seats and Plaid Cymru one, though in all cases with substantially reduced majorities. Eleven constituencies had voted Remain in 2016. None of these changed hands in the 2019 election. It is likely that in these cases Labour candidates were helped by Welsh Labour’s unequivocal pro-Remain position as it allowed candidates to reflect this. Two marginal seats that stayed Labour were Gower and Cardiff North. Anna McMorrin in Cardiff North campaigned explicitly as the pro-Remain candidate and it was reported that ‘the Liberal Democrat candidate has nobly suggested to his supporters that they might consider voting for Ms McMorrin in the spirt of opposing Brexit’ (Shipton, 2019b). Generally, however, the numbers suggested that Labour probably lost more than it gained in Wales from not clearly supporting the completion of Brexit. Voters in all parts of Wales moved from Labour to the Conservatives and party positions on Brexit appear to have played a big part. Table 9.3 Party performances in Wales, according to Leave or Remain constituency voting in the 2016 EU Referendum . Leave voting constituencies . Remain voting constituencies . Labour Hold 15 7 Cons Hold 5 1 Cons Hold (lost between 2017 and 2019) 2 0 Cons Gain from Labour 6 0 PC Hold 1 3 Total 29 11 . Leave voting constituencies . Remain voting constituencies . Labour Hold 15 7 Cons Hold 5 1 Cons Hold (lost between 2017 and 2019) 2 0 Cons Gain from Labour 6 0 PC Hold 1 3 Total 29 11 Source: Adapted from House of Commons Library: Brexit: Estimates of Votes by Constituency. Open in new tab Table 9.3 Party performances in Wales, according to Leave or Remain constituency voting in the 2016 EU Referendum . Leave voting constituencies . Remain voting constituencies . Labour Hold 15 7 Cons Hold 5 1 Cons Hold (lost between 2017 and 2019) 2 0 Cons Gain from Labour 6 0 PC Hold 1 3 Total 29 11 . Leave voting constituencies . Remain voting constituencies . Labour Hold 15 7 Cons Hold 5 1 Cons Hold (lost between 2017 and 2019) 2 0 Cons Gain from Labour 6 0 PC Hold 1 3 Total 29 11 Source: Adapted from House of Commons Library: Brexit: Estimates of Votes by Constituency. Open in new tab The election post-mortem started almost immediately in the Labour Party, reflecting much pent-up frustration at Corbyn’s leadership and the reception candidates had found in canvassing for their leader. Several Welsh Labour MPs publicly criticised him. Chris Bryant, MP for the Rhondda, said ‘During the last general election in 2017, people on the doorstep would call him a terrorist sympathiser. This time round it was “He is a terrorist”. A lot of that may be unfair, but the Labour Party has a lot of soul searching to do, and so does Jeremy Corbyn. For a lot of Labour voters, the Labour Party has swayed a long way from the party they have known and loved. Whether that is over Brexit or our defence policy or our patriotism’ (Western Mail, 14 December 2019, pp. 4–5). Anna McMorrin asserted: ‘It’s on the watch of this leadership of our party that we have lost decent hardworking MPs who fought every day for their constituents and the millions of people who need a Labour government and I hope that he will take responsibility so we can move on and rebuild the party’ (Western Mail, 14 December 2019, p. 4). The most entertaining attack came from former Labour MP for Pontypridd, Kim Howells, who was deeply critical of Corbyn’s post-election decision that he would stay on until a new leader had been elected: ‘It is just insane, but it is an indication of that smug, sanctimonious, deluded, middle class, London group of MPs who run the Labour party and have ruined it’ (Western Mail, 17 December 2019). He attacked Corbyn for his lack of patriotism, always finding fault in Britain and siding with Britain’s enemies. Alun Davies, AM for Blaenau Gwent, wrote in his blog: ‘We need to recognise that this defeat was not only a defeat for the man, but a defeat for what he stood for as well. His failure was personal, political, policy and principle. Corbyn must go but for Labour to recover Corbynism must go as well’ (Western Mail, 17 December 2019). In response, the left of the party sought to defend the election policy agenda generally and blame the specific issue of Brexit that made the election very hard for Labour. Mark Drakeford described the results as ‘deeply disappointing’ but argued that the losses in Wales were by relatively small margins of 2,000 or less, with Labour still the largest party. The seats that the Conservatives had won he considered to be ones that had only been lent to the Conservatives and they wanted them back. He acknowledged ‘we have a particular issue in the north of Wales as well, we know we have to address, and we’ll be thinking hard and we’ve been thinking that already’ (Western Mail, 20 December 2019). The Conservatives celebrated while the other parties felt utterly marginalised by the two-party battle. Only Plaid Cymru still had a foothold in the party system and looked forward immediately to the 2021 Assembly elections. Observers of representative democracy in Wales could take some generally positive conclusions from the election. The turnout was a bit lower than in 2017 but it was a winter election. The number of women MPs increased by three to 14 (35%) overall, just slightly above the proportion for the House of Commons as a whole. Sarah Atherton (Wrexham), Virginia Crosbie (Ynys Mon) and Fay Jones (Brecon and Radnorshire) were the Conservatives’ first female MPs ever elected in Wales. There were still no BME MPs though. The seats to vote shares were more proportional than in previous elections, with Labour winning 55% of the seats on 40.9% of the vote, the Conservatives 35% of the seats on 36.1% of the vote, and Plaid Cymru 10% of seats on 9.9% of the vote. Labour clearly still received a bonus in how the simple plurality system worked. The election indicated overall that Labour resilience generally was still there but, as Table 9.4 shows, the historic trend suggests that Labour has now had a relatively bad result in Wales in three of the last four elections. Its decline from total dominance was emphasised in 2019 by its lowest seat haul since 1983, because for once Labour relatively speaking lost out more at the margins. Overall, the trend appears to be towards a more competitive two-party system between Labour and the Conservatives with Plaid Cymru as the third party, strong in its core areas but otherwise still struggling to break a duopoly. Table 9.4 UK General Election results in Wales, 1979–2019 . Con . Lab . Lib Dem (formerly Lib and Lib-SDP) . Plaid Cymru . Others (incl. UKIP and Green) . 1979 Vote share (%) 32.2 47.0 10.6 8.1 2.2 Seats 11 21 1 2 1 1983 Vote share (%) 31.0 37.5 23.2 7.8 0.4 Seats 14 20 2 2 0 1987 Vote share (%) 29.5 45.1 17.9 7.3 0.2 Seats 8 24 3 3 0 1992 Vote share (%) 28.6 49.5 12.4 8.8 0.7 Seats 6 27 1 4 0 1997 Vote share (%) 19.6 54.7 12.4 9.9 3.4 Seats 0 34 2 4 0 2001 Vote Share 21.0 48.6 13.8 14.3 2.3 Seats 0 34 2 4 0 2005 Vote share (%) 21.4 42.7 18.4 12.6 4.9 Seats 3 29 4 3 1 2010 Vote share (%) 26.1 36.2 20.1 11.3 6.2 Seats 8 26 3 3 0 2015 Vote share (%) 27.2 36.9 6.5 12.1 17.3 Seats 11 25 1 3 0 2017 Vote share (%) 33.6 48.9 4.5 10.4 2.5 Seats 8 28 0 4 0 2019 Vote share (%) 36.1 40.9 6.0 9.9 7.1 Seats 14 22 0 4 0 . Con . Lab . Lib Dem (formerly Lib and Lib-SDP) . Plaid Cymru . Others (incl. UKIP and Green) . 1979 Vote share (%) 32.2 47.0 10.6 8.1 2.2 Seats 11 21 1 2 1 1983 Vote share (%) 31.0 37.5 23.2 7.8 0.4 Seats 14 20 2 2 0 1987 Vote share (%) 29.5 45.1 17.9 7.3 0.2 Seats 8 24 3 3 0 1992 Vote share (%) 28.6 49.5 12.4 8.8 0.7 Seats 6 27 1 4 0 1997 Vote share (%) 19.6 54.7 12.4 9.9 3.4 Seats 0 34 2 4 0 2001 Vote Share 21.0 48.6 13.8 14.3 2.3 Seats 0 34 2 4 0 2005 Vote share (%) 21.4 42.7 18.4 12.6 4.9 Seats 3 29 4 3 1 2010 Vote share (%) 26.1 36.2 20.1 11.3 6.2 Seats 8 26 3 3 0 2015 Vote share (%) 27.2 36.9 6.5 12.1 17.3 Seats 11 25 1 3 0 2017 Vote share (%) 33.6 48.9 4.5 10.4 2.5 Seats 8 28 0 4 0 2019 Vote share (%) 36.1 40.9 6.0 9.9 7.1 Seats 14 22 0 4 0 Turnout: 1979: 79.4%; 1983:76.1%; 1987: 78.9%; 1992:79.7%; 1997: 73.6%; 2001: 60.6%; 2005: 62.4%; 2010: 64.9%; 2015: 65.6%; 2017: 68.6%; 2019: 66.6%. Open in new tab Table 9.4 UK General Election results in Wales, 1979–2019 . Con . Lab . Lib Dem (formerly Lib and Lib-SDP) . Plaid Cymru . Others (incl. UKIP and Green) . 1979 Vote share (%) 32.2 47.0 10.6 8.1 2.2 Seats 11 21 1 2 1 1983 Vote share (%) 31.0 37.5 23.2 7.8 0.4 Seats 14 20 2 2 0 1987 Vote share (%) 29.5 45.1 17.9 7.3 0.2 Seats 8 24 3 3 0 1992 Vote share (%) 28.6 49.5 12.4 8.8 0.7 Seats 6 27 1 4 0 1997 Vote share (%) 19.6 54.7 12.4 9.9 3.4 Seats 0 34 2 4 0 2001 Vote Share 21.0 48.6 13.8 14.3 2.3 Seats 0 34 2 4 0 2005 Vote share (%) 21.4 42.7 18.4 12.6 4.9 Seats 3 29 4 3 1 2010 Vote share (%) 26.1 36.2 20.1 11.3 6.2 Seats 8 26 3 3 0 2015 Vote share (%) 27.2 36.9 6.5 12.1 17.3 Seats 11 25 1 3 0 2017 Vote share (%) 33.6 48.9 4.5 10.4 2.5 Seats 8 28 0 4 0 2019 Vote share (%) 36.1 40.9 6.0 9.9 7.1 Seats 14 22 0 4 0 . Con . Lab . Lib Dem (formerly Lib and Lib-SDP) . Plaid Cymru . Others (incl. UKIP and Green) . 1979 Vote share (%) 32.2 47.0 10.6 8.1 2.2 Seats 11 21 1 2 1 1983 Vote share (%) 31.0 37.5 23.2 7.8 0.4 Seats 14 20 2 2 0 1987 Vote share (%) 29.5 45.1 17.9 7.3 0.2 Seats 8 24 3 3 0 1992 Vote share (%) 28.6 49.5 12.4 8.8 0.7 Seats 6 27 1 4 0 1997 Vote share (%) 19.6 54.7 12.4 9.9 3.4 Seats 0 34 2 4 0 2001 Vote Share 21.0 48.6 13.8 14.3 2.3 Seats 0 34 2 4 0 2005 Vote share (%) 21.4 42.7 18.4 12.6 4.9 Seats 3 29 4 3 1 2010 Vote share (%) 26.1 36.2 20.1 11.3 6.2 Seats 8 26 3 3 0 2015 Vote share (%) 27.2 36.9 6.5 12.1 17.3 Seats 11 25 1 3 0 2017 Vote share (%) 33.6 48.9 4.5 10.4 2.5 Seats 8 28 0 4 0 2019 Vote share (%) 36.1 40.9 6.0 9.9 7.1 Seats 14 22 0 4 0 Turnout: 1979: 79.4%; 1983:76.1%; 1987: 78.9%; 1992:79.7%; 1997: 73.6%; 2001: 60.6%; 2005: 62.4%; 2010: 64.9%; 2015: 65.6%; 2017: 68.6%; 2019: 66.6%. Open in new tab 3. Conclusion The fulfilment of Boris Johnson’s desire for a decisive majority in order to ‘Get Brexit Done’ was aided by the results in Wales. Whether this outcome included specifically Welsh factors is a matter for debate. There was a Welsh campaign, characterised by party strategies, manifestos and leaders’ debates, conducted amid an awareness that the electorate invariably returned Labour as the largest party. It would be hard to say that the distinct Welsh Conservative campaign offered a great deal to the Conservatives’ UK-wide cause; but it is likely that the reach of Johnson’s British-wide message and campaign, tailored to Wales, was important in positively making a difference to Conservative fortunes in Wales. This was the factor that made the difference from 2017. At the same time, the Welsh Labour campaign had its problems. It was not as effective as the Labour campaign in 2017 and it is likely that this in fact was the most significant specifically Welsh dimension to the election that made a difference. Having said that clearly Corbyn was not as effective as in 2017 as well. The implications of the result fed into British-wide analyses of political change. The trend towards support for the (small ‘u’ unionist) centre-ground of Welsh politics seen in 2017 was confirmed; it is simply that the Conservatives won more of it than in the previous contest. The implications of this were hard to predict. Labour was still the largest party with a proven record of resilience. They could improve their performance after a slightly under-par election organisationally, and with a different British leader and Brexit behind them the party could yet increase their vote share again. Nevertheless, the Conservatives clearly had a more solid platform for the future. This certainly had implications for future UK elections, and possibly also for Conservative fortunes in the National Assembly elections in 2021. In this context, it is perhaps worth reflecting on the 2019 election in Wales specifically as a success for Johnson. In launching the party’s manifesto on Bangor-on-Dee’s racecourse in late November, he had commented that ‘I started this project 22 years ago’. In a way that Theresa May had signally failed to understand from walking holiday visits, Johnson had got both what the Welsh electorate disliked about the Conservative Party and what it might support ingrained in his DNA from the oldest form of political training; the party apprenticeship of taking a good political beating in an unwinnable seat. As a result, when it mattered, he had a brand of populist Toryism that cut through to appeal to both conventional Conservative voters and tap into anti-Welsh establishment feeling. This was done concisely and cleverly, in a language which was engaging and insistent without forcing people to feel like they were being disloyal to their country. This more than anything else may be the defining legacy of the 2019 election in Wales. References BBC. ( 2019 ) ‘General election: Plaid, Lib Dems and Greens in pro-EU pact in 11 Welsh seats’, 7 November 2019, accessed at https://www.bbc.co.uk/news/uk-wales-50325815 on 7 January 2020. Drakeford M. ( 2019 ) ‘Richer Lives, Greener Future for all and a Chance to Stay in the EU’, Western Mail, 11 December, p. 21 . Cymru Plaid. , ( 2019 ) Wales, It’s Us , Cardiff , Plaid Cymru . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Scully R. ( 2019 , 4 November) ‘The First Welsh Poll of the General Election’, accessed at https://blogs.cardiff.ac.uk/electionsinwales/2019/11/04/the-first-welsh-poll-of-the-general-election-2/ on 2 April 2020. Shipton M. ( 2019 a), ‘Don’t Assume the Odds-on Favourite will Cruise to Victory’, Western Mail, 30 October, pp. 22 – 23 . Shipton M. ( 2019 b) ‘Brexit will be Determining Factor in this Historic Election’, Western Mail, 12 December, p. 9 . Welsh Conservative Party ( 2019 ) Get Brexit Done, Unleash Wales’ Potential , Cardiff , Welsh Conservatives . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Welsh Labour. ( 2019 ) Standing up for Wales . 2019 Election Manifesto, Cardiff , Welsh Labour . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Western Mail ( 2019 , 5 November) ‘Top Tories Standing by Election Hopeful in Facebook “put them Down” Row’, p. 8 . Western Mail ( 2019 , 26 November) ‘Johnson Pledges Key Roads Boost as he Launches Welsh Manifesto’, pp. 6 – 7 . Western Mail ( 2019 , 11 December) ‘Labour Activists Accuse Party of Poor Campaign’, p. 8 . Western Mail ( 2019 , 12 December) ‘Boris Johnson Visits Caerphilly in Last Push for Election Votes’ pp. 6 – 7 . Western Mail ( 2019 , 14 December) ‘Corbyn to Resign as Leader in Early 2020 after Polls Nightmare’, pp. 4 – 5 . Western Mail ( 2019 , 14 December) ‘Leader under Fire as Labour Endures a Battering in Wales Despite Keeping Control of Cardiff’, p. 4 . Western Mail ( 2019 , 17 December) ‘Former Labour Minister’s Scathing Attack on Corbyn’, p. 6 . Western Mail ( 2019 , 20 December) ‘Interview with Mark Drakeford’, pp. 8 – 9 . © The Author(s) 2020. Published by Oxford University Press on behalf of the Hansard Society; all rights reserved. For permissions, please e-mail: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)