All three of the traditionally recognized new institutionalisms – rational choice, historical, and sociological – have increasingly sought to ‘endogenize’ change, which has often meant a turn to ideas and discourse. This article shows that the approaches of scholars coming out of each of these three institutionalist traditions who take ideas and discourse seriously can best be classified as part of a fourth ‘new institutionalism’ – discursive institutionalism (DI) – which is concerned with both the substantive content of ideas and the interactive processes of discourse in institutional context. It argues that this newest of the ‘new institutionalisms’ has the greatest potential for providing insights into the dynamics of institutional change by explaining the actual preferences, strategies, and normative orientations of actors. The article identifies the wide range of approaches that fit this analytic framework, illustrating the ways in which scholars of DI have gone beyond the limits of the traditional institutionalisms on questions of interests and uncertainty, critical junctures and incremental change, norms and culture. It defines institutions dynamically – in contrast to the older neo-institutionalisms’ more static external rule-following structures of incentives, path-dependencies, and cultural framing – as structures and constructs of meaning internal to agents whose ‘background ideational abilities’ enable them to create (and maintain) institutions while their ‘foreground discursive abilities’ enable them to communicate critically about them, to change (or maintain) them. But the article also points to areas for improvement in DI, including the theoretical analysis of processes of ideational change, the use of the older neo-institutionalisms for background information, and the incorporation of the power of interests and position into accounts of the power of ideas and discourse.

Why do certain European Union (EU) institutions look so much like institutions in its member states? The article asserts the reason for this similarity is that these structures have been ‘copied’ from the member states to the EU and provides a historical institutionalist explanation for this process of institutional transfer. The theory of isomorphism provides the analytical foundation for understanding institutional imitation in the EU, while the European ombudsman (EO) and the European Court of Auditors (ECAs) serve as the empirical case studies on which the theory is tested. The article finds that previously existing institutional arrangements matter not only by molding actor preferences in relation to an upcoming institutional transfer, but also by restricting their options and behavior. Specifically, the founding of the EO and the ECAs was the result of pressures exercised by the European Parliament, a supranational body, and the governments of the EU member states. In order to achieve the transfer, those actors had to navigate established institutional practices, primarily the ones associated with the negotiation of new treaties, which are necessary for the addition of new institutions to the EU institutional edifice.

Building on the literature on public finance, this article explores the consequences of political scale for government spending. The central argument is that the tendency for small political units to have big governments is not merely the result of economies of scale in the provision of public goods, but a reflection of the greater pressures for public spending faced by politicians in smaller and more homogeneous political units. The importance of such political pressures relative to other influences on spending is assessed on the basis of subnational data by comparing the relationship between size and spending under democracy and dictatorship. To the extent that government expansion is driven by citizen demands, the impact of size on spending may be expected to be more pronounced under democratic than authoritarian governance. Results from a time-series cross-sectional analysis of growth in government spending are consistent with this expectation. Government growth is shaped not only by the population size of political units but also by the interaction between regime and size. Analysis of spending patterns under democratic rule further indicates that size is an important determinant of spending even after controlling for variations in citizen preferences, political institutions, electoral competitiveness, and economic performance. The results have important theoretical implications for the study of fiscal policy and democratic governance around the world because they suggest that political scale conditions the linkages between citizens and the state, creating widely varying incentives for government growth across differently sized political units.

One of the most important activities of the European Union (EU) has been widening the Union, that is, enlargement. Tapping to what extent and why EU citizens support future enlargement rounds has become a popular endeavour among researchers and EU officials. During recent years, a number of studies have indicated that public support for EU integration is likely to also depend on how the national news media portray the EU. Based on an experimental survey design, we test the effects of two news frames on support for future Serbian EU candidacy. We find that exposure to news frames has considerable impact on general understanding of Serbian EU candidacy, issue interpretation, and policy support. This effect is moderated by political knowledge. Knowledgeable participants were able to express their thoughts on Serbian EU candidacy more elaborately, whereas low-knowledge individuals were overall more susceptible to framing effects. We discuss the implications of our findings for current debates such as Euroskepticism and decreasing public support for EU integration.

This article analyzes whether democratic deliberation enhances ‘civic virtues’ such as political knowledge, efficacy, trust, and preparedness for political and other collective action. The empirical analysis is based on an experiment held in November 2006 in Finland. The topic of this citizen deliberation experiment was nuclear power. Two treatments were enforced; half of the small groups made a decision through secret ballot, whereas the other half formulated a common statement. Even though both treatments were designed to be ‘deliberative’ with discussion rules and moderators, the common statement procedure was expected to lead to more thorough deliberation. Therefore, we anticipate more increase in civic virtues in the common statement treatment. Based on the empirical analyses, this expectation is partially confirmed. Most notably, the participants’ knowledge of energy-related issues increased more in the common statement treatment, but also the increase of political trust and readiness for collective action shows a similar pattern. Finally, deliberation did not increase the participants’ overall preparedness to act politically, but did not decrease it either.

This paper tests contested arguments within the institutionalist literature about the relation between institutional and attitudinal changes, using the reunified Germany as a case. Eastern Germany constitutes a case approaching a ‘natural experiment’ for the social sciences, being twice the receiver of externally imposed institutions. It, therefore, provides a unique opportunity to closely analyse institutional effects on attitudes, as in this particular case, the time order of institutional and attitudinal changes can actually be decided. Using data from the International Social Survey Program modules on ‘The Role of Government’ (1990, 1996, and 2006), attitudes towards government responsibilities are compared in Eastern and Western Germany, and to other countries. Results show a considerable convergence in attitudes between Eastern and Western Germany – attitudes in Western Germany are completely stable while attitudes in Eastern Germany become, overtime, more similar to those found in the West. Furthermore, comparisons of different birth cohorts show that while considerable attitude differences between Eastern and Western Germany are still found in 2006 among those who had their forming experiences before the fall of the wall, differences are virtually nil among those who were still children in 1989. In summary, the analysis provides strong support for the attitude-forming effects of institutions, and a clear vindication of institutional theories. It also points to generational replacement as a key mechanism in translating institutional change into attitudinal change.

This article supplements and further develops the almost exclusively American literature on the determinants of negative campaigning by analyzing the tone of the Danish parties’ election campaigns. It concludes that proximity to governmental power matters, as oppositional parties are more negative than incumbents. This is comparable to the American experiences. The prospect of electoral failure, however, does not affect the tone the same way as poor poll standings do in the US. Moreover, it is suggested that future studies of negativity might consider how different party organizations affect the campaign tone; at least this study finds indications that parties with large proportions of party identifiers are slightly more negative than other parties. Finally, it is found that parties campaign differently in different channels of communication; that is, they are generally more negative in channels that allow direct interaction among politicians. This finding poses the question whether some channels are better empirical sources for studies of negativity than others, which is addressed in the closing section of the article.