On 21 May 2018, Professor Jane Green delivered the fifth annual Sir Roger Jowell Memorial Lecture, in honour of the famed statistician (and NatCen's co-founder). In this blog she shares some of her insights from the lecture, on the subject of electoral shocks and the increasing volatility of voters in British elections:
Sir Roger Jowell made a tremendous contribution to our understanding of British elections. His work with the British Election Study taught us about over-time declines in party attachments (partisan dealignment), the importance of political context and how this is related to the issues, attitudes and dimensions that matter to elections and electoral choice.
Sir Roger’s focus on context inspired his cross-national work with the European Social Survey and the International Social Survey Programme. His focus on over-time context to understand British politics called for continuity in survey items, just as he pioneered with the British Social Attitudes surveys - recognising that we cannot understand the unique period we are in without seeing it within the bigger picture. This is a priority that we – the current BES team - have embraced and invested in, focusing on continuity in measurement, restoring survey questions from earlier periods, and understanding our current electoral climate in the context of over-time change.
The task we face is no less than understanding the changing nature of the British party system; not the votes for one particular party, not favouring one type of statistical model or one kind of explanatory variable, but understanding the system-wide changes we have witnessed in such a short time in British politics. From the steepest change in level of voting for ‘others’ between the general elections in 2010 and 2015, to the largest Conservative-Labour combined share in 2017 since 1970, just two years later; the last two general elections have been the most volatile at the aggregate level since 1931. We also find that 2015 and 2017 were among the most volatile elections at the individual level. Switching party choice between elections has risen steadily, the flip-side of the decline in partisan attachments.
Existing explanations help us understand various specific aspects of the changing political landscape, but none are well suited to the task of understanding volatility at the level of the party system.
What are electoral shocks?
In the book we are completing (Electoral Shocks: Understanding the Volatile Voter in a Turbulent World, by Edward Fieldhouse, Jane Green, Geoffrey Evans, Jonathan Mellon, Christopher Prosser, Hermann Schmitt and Cees van der Eijk, forthcoming with Oxford University Press), we argue for the importance of a series of electoral shocks and the environment of partisan dealignment as the reason for why these shocks are now having especially destabilising effects.
Shocks are not just minor events that come and go, influencing a measure such as prime ministerial approval. They are shocks to the party system that may have important effects on the long-term reputations of parties and the issues and dimensions that matter to politics. They could be absorbed into the current party system and existing party competition, and so they are not defined by their consequences.
Their effects depend on a) the level of partisan attachment, which determines the propensity of people to switch loyalties in response to shocks; and b) the politicisation of shocks in politics, the media, and in party competition, which determine the success – or otherwise – in the ways in which parties gain or lose votes in response to them.
Electoral shocks are noticeable over prolonged time periods, salient, and hard to ignore. They have the potential to be noticed and recognised by people who do not commonly have much interest in politics, and by people who might otherwise select information that fits with their beliefs and preconceptions. This is an argument I made with Will Jennings in The Politics of Competence.
They are political, and so are therefore relevant to the potential shaping of the party system. They offer politicians new political opportunities, ways of competing, bringing uncertainty into strategies, electoral responses, and election outcomes.
They are also atypical, representing a change to the status quo. Each might have been predicted in some quarters in different ways, but they are unusual, significant and abrupt political changes.
The impacts of electoral shocks
In Electoral Shocks, we study the effects of five shocks and demonstrate their impacts on the 2015 and 2017 elections,
- The rapid rise immigration from Eastern Europe, leading to the rising salience of non-left-right issues, a stronger relationship between attitudes to Europe and immigration, and support for UKIP in 2015.
- The Global Financial Crisis as a long-term shock to Labour’s reputation and resentments that drove UKIP support in 2015.
- The 2010-2015 coalition, which damaged the Liberal Democrats' reputation, helping to explain the outcomes in 2015 and 2017.
- The 2014 Scottish independence referendum which re-aligned voters with parties on the independence dimension.
- The 2016 EU referendum. Perhaps the largest of all shocks, which caused a realignment of voters to parties on the liberal-authoritarian dimension, alongside attitudes to Europe and immigration, and the importance of different cleavages based on age, education and income.
For each of these shocks we analyse the complex causal mechanisms behind their effects, offering new insights into theories of electoral behaviour and explanations of the 2015 and 2017 elections.
The British party system is weaker, more unstable and more vulnerable to the destabilizing effects of shocks due to partisan dealignment. Newer parties have also been weaker and more vulnerable to shocks because of the absence of more strongly rooted partisan support. We show that the effects of each of our shocks are conditional on different levels of party identification in the electorate, such that shocks that matter now would have had less dramatic effects in periods in which British voters demonstrated stronger loyalties to political parties.
What happens next?
Crucially, we don’t yet see any resurgence in the levels of partisan identification in Britain, and we don’t yet have strong empirical reasons to expect this to happen in the near future. Our ability to understand the likely instability or stability of British politics going forward depends on two questions that our book places centre stage:
- Will a new Brexit-based attachment make the party system more resistant to shocks?
- Will the outcome of Brexit further destabilise the party system and lead to new and unexpected electoral choices and outcomes?
We might not be able to predict the future, but our work helps us understand the factors that will be critical to what lies ahead.
Sir Roger Jowell covered four general elections in the seventeen years he was a BES Co-Director. I have covered four major elections in the four years I have held the same role. We live in extraordinary political times. The next four years promise to be as fascinating as the last four, and it is a huge honour and privilege to follow in Sir Roger’s footsteps to be a part of it.