The Politics of European Referendum Outcomes and Turnout: Two Models

Abstract

This survey briefly examines the EU accession process for the countries included in this collection and considers the distinctive features of this set of referendums, particularly the comparatively high levels of Yes vote and low levels of turnout. It argues that, although they represent a distinctive sub-type of European referendum, they can be used as a basis to draw at least tentative comparative and theoretical conclusions. Consequently, it then posits causal models that both provide an analytical framework for this collection and, more broadly, attempt to explain the results and turnout in these and, potentially, other (European) referendums.

Notes

We are grateful to all the contributors to this volume for their extremely valuable input in helping us to develop the thinking that we set out in this introduction. These ideas were developed at two seminars for the contributors to this special issue held in June 2003 and October–November 2003 at the Sussex European Institute.

On the EU accession process, see, for example, Cameron (2004).

Slovenia was the only one of the post-communist states due to join NATO in 2004 (Bulgaria, Estonia, Latvia, Lithuania, Romania and Slovakia were the others) that held a referendum on this issue. There were calls for a NATO accession referendum to be held in Slovakia, with opinion polls indicating that the outcome would be very close, but this was opposed by most of the political elite. Grassroots supporters of this option failed to gather the requisite number of signatures. In the event, fears that Slovenians might use the NATO referendum as an opportunity to protest against the (at that time planned) US-led invasion of Iraq, proved unfounded and Slovenians voted to join NATO by a margin of two to one.

For more on the Lithuanian referendum see Mazylis and Unikaite (2003).

As Fowler explains in her article, Hungary had a less onerous threshold that 25% of the total electorate, as well as a simple majority, had to vote Yes.

See also van der Eijk et al. (1996) for the same thesis applied to European Parliament elections.

Much of this is based on analysis of the various Danish European referendums. See, for example, Siune et al. (1994) and Svensson (2002).

For a good review of the literature that is broadly sympathetic to the first approach, see Franklin and Marsh (2001).

Developed on the basis of two workshops involving the participants involved in this project in June 2003 and October–November 2003.

We acknowledge that this can, in turn, stem from either the credibility of the individual concerned or the institution that they occupy or form part.

In the case of voting in the two Irish referendums on the Nice Treaty there is evidence that the opposite was the case, with the low level of turnout correlating positively to a No vote and vice versa. See Gilland (2002a; 2002b).

We are particularly grateful to Brigid Fowler for pointing this out to us.

Although it is also offered further support by the outcome of another recent Scandanavian referendum on European issues, the Swedish EMU referendum. See Aylott (2003).

In the case of Malta, although this country does not have a weak, organised civil society as a legacy of communism, one might also expect a low level of civic engagement given the way that the two main parties have traditionally completely dominated and controlled the organised public sphere.

According to the 1953 Danish Constitution, sovereignty can only be delegated to an international authority if a 5/6 majority is achieved in the Folketing or if a simple majority is achieved and the proposal is backed by a referendum. We are grateful to Clive Archer for pointing this case out to us.

There are, of course, many other possible variables that may account for varying levels of turnout that we have excluded because we believe that they only have a marginal effect or because no generalisable propositions can be made. For example, we believe that the weather, which is often cited as being crucial, particularly by non-academic commentators, falls into this category.