EUROPEAN UNION: Integration will remain elite project

Analysis

The 'contagion effect' of referenda was evident in the margin of 26 points by which the Dutch electorate -- voting three days after the French -- rejected the draft constitutional treaty (see FRANCE: De Villepin takes helm after strong 'no' vote - May 31, 2005 and see NETHERLANDS: Large 'no' vote may boost standing in EU - June 2, 2005). Even though the Dutch referendum was a consultative exercise, it hastened an early end of the ratification process (see EUROPEAN UNION: Budget row points to wider malaise - June 17, 2005). Instead of acting as powerful legitimising endorsements of further European integration, the French and Dutch 'no' votes stalled the ratification process and cast the wider EU project into crisis.

Public sentiment. It is unclear whether the results of referenda are reliable guides to public sentiment towards the EU. At issue is whether they represent the verdict of a genuinely plebiscitary exercise on the issue at hand, or whether they are ultimately statements of the level of public support for an incumbent government.

The question is important because referenda are widely used within the EU to settle matters relating to the future of integration. In some member states (for example France and Ireland), a referendum is required in situations where major constitutional changes are proposed. Elsewhere (notably the United Kingdom), they are used sporadically -- and perhaps strategically -- by divided governments or by governments seeking to defer responsibility to the people.

Analysis of referenda shows the sheer complexity of the domestic politics of European integration and how this varies from member state to member state. This is evident in two ways:

• Variable domestic discourses. Contrasts are evident between the French debate on the 'Anglo-Saxon' (or liberal economic) qualities of the treaty, and the putative UK debate, where the treaty is sometimes presented as a charter for interventionist threats to UK economic success.
• Referendum coalitions. Both the French and the Dutch campaigns united the extremes of right and left on behalf of a 'no' vote against the centrist establishment. This suggests, either that EU issues cannot be mapped onto a conventional left-right political spectrum or, more likely, that the referenda amounted to attacks on the political establishment in both countries.

Early opinion poll analysis suggests that in France, the 'no' vote was strongly motivated by economic considerations. Eurobarometer data published this week suggest that 31% of those voting against the constitutional treaty were driven in part at least by fears about unemployment. Scores were also high for general concerns about the economic situation and the supposed ultra-liberalism of the treaty. 'No' voters in the Netherlands nominated a diverse array of reasons for rejecting the treaty, the most prominent of which was a lack of information. In both cases, opposition to the national government figured prominently as a reason for the rejection of the constitutional treaty -- far above any fears about enlargement of the Union, or general feelings of euroscepticism.

EU referenda. Analysis of the broader context of EU referenda over time leads to several conclusions:

• Voting decisions in referenda on EU issues tend to be expressions of views on immigration, considerations of the state of the domestic economy and the evaluation of the incumbent government.
• EU referenda tend to say less about the issue at hand than they do about popular views of a government's performance or the balance of domestic party political forces.
• The 'information environment' during a referendum campaign can also have a significant impact on referenda, suggesting that outcomes are not preordained but are heavily influenced by the conduct of campaigns. While governments have a significant degree of control over the wording of the question to be asked, they are not able to set the debating parameters for the duration of the campaign.
• If a government is perceived to handle a referendum campaign badly, it is likely to be punished in the referendum vote.
• Referenda tend to yield more pro-EU integration results if there is a definite provision for their use in the country's constitution.

Prior to the French and Dutch examples, referenda usually produced results that endorsed the advancement of integration as sponsored by prevailing political elites. 'No' votes were anomalous, notably the two refusals of the Norwegian electorate to join the communities (see NORWAY: EU Rejection - November 29, 1994), or quickly rectified through a combination of bargained compromises and further referenda, as with the Danish vote on the Maastricht Treaty and the Irish referendum on the Nice Treaty (see EUROPEAN UNION: Denmark rejects the Euro - September 29, 2000, see EUROPEAN UNION: Discomfort after Irish 'No' - June 12, 2001 and see EUROPEAN UNION: Yes vote clears way to next accession - October 21, 2002).

Outlook. The recent referenda, and the ensuing summit decision to suspend ratification for the whole of the EU, have scuppered the constitutional treaty in its present form. Governments will develop justifiable anxieties about how a lost referendum campaign on an EU issue might have an impact on their long-term electoral fortunes. This will not only inject caution into the use of the referendum as a tool, but it may also affect what governments perceive to be politically possible in terms of European integration more generally.

This does not automatically imply that the EU is heading for a sclerotic period (see EUROPEAN UNION: French 'no' may give rise to new core - May 27, 2005). The result to date of attempts to ratify the constitutional treaty suggests that governments can receive political setbacks when they expose themselves to the will of their domestic populations. Yet the history of integration suggests that innovation and change will occur independently of grand revisions to the basic treaties.

Such innovation traditionally emerges from the corridors of EU institutions, leading to a continued democratic paradox in European integration. While the constitutional treaty was originally premised on the idea that the EU required a greater degree of openness and democratic transparency, its rejection in two national referenda is likely to lead to a period of increased opacity in decision-making. Political elites will now probably opt for intergovernmental bargaining that strays considerably from the treaty's aspirations of greater democratic openness, rather than appeal to direct democracy as a tool of European integration.