Monday, 13 April 2015

Short summary of PhD, April 2015.

How the British public changes its support for membership of the European Union.

This thesis examines over time trends in the public’s support for EU membership. Specifically, it investigates the effects of two major political events on public attitudes towards the EU. The first is the financial crisis of 2007 and its aftermath. The second is the highest level of immigration to Britain since records began, following the expansion of the EU in 2004 to include seven former 'eastern bloc' countries.

The 20th Century project of European integration, in both its initial design and subsequent management, has been the preserve of political elites. Unsurprisingly therefore, most academic research has been conducted on elite-level behaviour and attitudes. Whilst the volume of work on the mass public's relationship with integration has increased, at the same time voters have been recoiling from the European project across Europe, regularly voting against it at country-based referendums. 

This has often been a source of frustration to European elites, and EU supporting academics too, who have often shown little respect for public opinion on the EU.  In perhaps the best known and most widely read history of Britain's relationship with integration, top of most undergraduate reading lists, the late Hugo Young (1998) wrote that public attitudes were always 'changeable, ignorant and half-hearted', rarely achieving 'what pollsters call salience' (1998: 287). As a matter of study, 'if it revealed a consistent pattern', it was 'that the people tended to go wherever they were led by the political class' (508). In short, existing theories dispute whether voters have the capacity to respond to new information on the EU in a coherent manner at all, and if this is the case, locating meaningful patterns in public opinion is likely to be a frustrating task. When patterns or structure in public opinion has been located, it has generated conflicting expectations in public behaviour (Boomgarden et al, 2011).  All in all, the mechanisms by which the public changes its support for EU membership remain unclear.  The unique contribution of this thesis is to set out a single consistent rationale for why and when voters choose to learn from new information when revising EU attitudes, and when they do not, based upon a Bayesian process of adaption.

The ideas of English statistician, philosopher and Presbyterian minister Thomas Bayes (1701-1761) have previously found application in more common areas of voting behaviour study, particularly how voters identify to political parties (party ID). In this thesis, Bayes' logic is now applied for the first time to understand public attitudes towards the EU.  The central Bayesian assumption is that for EU support, like party ID, voters make the same probabilistic calculation of the importance of new experiences set against their own existing prior beliefs.  They choose whether to update these prior beliefs as a ‘running-tally’ of new information and experiences, but only when they consider the new information sufficiently valid to update their old preferences (Fiorina, 1981). This approach, grounded in assumptions of voter rationality rather than psychology, provides a powerful framework for understanding the notable changes in preference for the EU - both for and against - that have occurred in Britain during the last ten years, cutting-through existing theories that lack any 'a priori' logic for why voters, by themselves, can adapt their preferences.

By contrast to the economic approach of my thesis, existing theories of public opinion towards the EU have tended to follow a more psychological tradition in voter behaviour studies. At other times, theories have diverged from the mainstream body of voting behaviour literature all together, portraying voter behaviour on EU matters as an exceptional concern compared to behaviour towards parties and at domestic election time. Students of 'European Studies' are taught of the complexity and remoteness of supra-national government, new cultural values of 'Europeanisation', its unique institutional structures and the way it engenders for the voter questions of national identity, cutting across the main left-right field of political contestation that shapes domestic politics. This 'sui generis' quality to European political affairs, has become a favourite justification for the independence of the field of European Studies from political science. It is one of its own, with its unique characteristics.

All too often however as a result, the uniqueness of EU opinion theory has become a self-fulfilling prophecy in research, and useful core assumptions used to ground voter behaviour theories have become lost.  In particular, work on attitudes towards the EU has become complicated with a thick under-brush of post hoc justifications for two unrelated and potentially conflicting attitude dimensions structuring the voter decision: economics and identity. Crucially, there is no supporting theory as to how the voter, himself, connects the two to make rounded and evolving evaluations. Instead, we are left with 'non-rational' models of behaviour that claim change is the result of forces external of the voter, somehow mobilising at different times the unrelated (“soft-economic" and “hard-identity”) dimensions in his attitudes (Hooghe and Marks, 2009). Emotional forces of nationalism especially, as well as domestic party competition involving populist parties, are used to explain voter behaviour (Taggart, 1998). Following this traditional academic narrative on voter support for the EU, politicians of the highest level have perpetuated the gloomy assessment of the voters' capability to make sense of Europe. Most recently for example, Tony Blair claimed during the 2015 General Election campaign that British voters cannot be trusted to make ‘sensible choice’ on the EU (Telegraph, 07/04/15).

This thesis rejects this pessimistic view of British voters and the Europe issue. Using evidence from 2007-2009 IntUne surveys, it shows that the attitudes of these voters towards EU membership are one-dimensional, and bi-dimensionality in the structure of attitudes is an artefact of the measurements tools not the empirical data itself.  The voter can balance “soft” economic appraisals with “hard” identity based ones, and the later may embrace nationalist and populist feelings. But the preferences that are formed best reflect a singular understanding of his interests. The voter is not as irrational as the bi-dimensional model insists.

The PhD also reviews the history of the debate between those for and against Britain’s membership of the EU from 1970, and how the issue has been framed within the traditional party political system and more recently by insurgent movements: the Referendum Party and Ukip. Evidence from a series of 144 British Election Study monthly surveys, conducted between 2004 and 2015, shows the non-rational approach is poorer at explaining over time change within the historical development of the issue, than the new Bayesian model.  Public opinion has evolved in a predictable way to changing information generated by the specific political context.  Most recently, record immigration to Britain from the EU, the financial crisis both at home and in the Eurozone, and subsequent economic recovery have become important determinants of EU support.  This runs contrary to the unsatisfactory and counterfactual conclusion of the main existing work published on the financial crisis and EU support, that the Eurozone's difficulties did not 'substantially bring back in economic factors as an important source of Euroscepticism' (Serrachio et al, 2013:51).

The results of the research reported in this thesis shed favourable light on the capacity of voters to respond to new information, a conclusion that belies the prevailing view of British voters as uninterested and uniformed on the EU membership question. They are also increasingly relevant to the decision as to whether an EU referendum should be called. Further, the conclusions also provide economic and political scenarios under which public opinion may change in the future, developing new insights on the likely outcome of such a referendum.  Such a referendum would shape not just the future of Britain's relationship with Europe, but also the nature of domestic politics in Britain.