Although the role of issues in elections has been widely studied, different approaches to the definition, classification, and measurement of issues render cross-national comparisons difficult. Further, little attention has been paid to the stability or instability of issues across time, and the possible implications of issue change for voting behaviour and election outcomes. In this analysis, a common classification scheme is employed for measuring the salience of issues in the elections of four countries, and the linkage of issues to political parties in the thinking of individual voters is explored. Panel data are employed to test the stability of attitudes towards political issues across a series of three elections in each country. Although there are wide differences in the specific array of issues in each country, issues can be broadly typed into five general categories of public concerns. Even given the common emphasis on economic issues throughout the 1970s, considerable instability exists across time in all categories of issues. This instability makes it more difficult to sustain a consistent issue/party link over more than one election, and thereby has implications for the accountability of parties in issue/policy terms.

The literature suggests two major theoretical explanations for public resistance to taxation. The first emphasizes a cognitive, instrumental, utility-maximizing response to objective elements of taxation and government spending. The second treats contemporary tax revolts as the result of more enduring ideological predispositions, disaffection from politics in general and the result of social mobilization by anti-government protagonists. In this article we confront the self-interest model and the symbolic politics model with empirical evidence at both the aggregate level and at the micro-level. In the macro-analysis we examine the relationship between the level of taxation and the acceptance of tax cheating in twelve Western nations; in the micro-analysis we perform a more detailed study of Norway, employing survey data. The over-all results clearly show the strongest support for the symbolic politics model, while the self-interest explanation is only weakly supported by the data. An important finding is that political attitudes and values have their strongest and most consistent impact among the politically uninvolved. This result further substantiates the symbolic politics model.

The focus is on the problem of dimensionality of left–right evaluations in mass publics. Starting from findings of earlier research showing that the left–right dichotomy can be interpreted in terms of a partisan and a value/issue component, it is argued here that the latter component has to be further differentiated into a more general value preference and a pro-/anti-establishment component. A subsidiary hypothesis maintains that the establishment component shows substantial life-cycle effects. This differentiation thus can help to clarify conceptually the aging–conservatism thesis. The analysis reveals an age-specific split on the meaning of Left and Right. While older cohorts interpret these terms predominantly via the traditional partisan frame of reference, in younger cohorts the value- and anti-establishment components overwhelmingly explain left–right evaluations. These effects are strongest in the group of university-educated younger-age cohorts. This is a reversal of the traditional alignments of the better educated; in Germany they have had a position closer to the Right. Education thus has become the basis of a new left–right conflict. In addition, in all the younger cohorts, the value components are increasingly detached from the traditional partisan frame. The ‘floating’ New Left has distanced itself from the established Left–Right camps to focus upon the new German party, the Greens.

This is an attempt to formulate a conditional model of coalition formation. Predictions from this model are tested against data on coalition formation in Dutch local government in 1982. Starting from a simple model of the behaviour of party politicians, it is assumed that the coalition behaviour of parties depends on the competitiveness of the (local) political system. The choice of a coalition strategy (minimizing size, minimizing policy range or proportional representation) is contingent upon two dimensions of the competitiveness of local politics. These are the amount of electoral volatility and the extent to which local election results follow national trends. Our conditional theory predicts different coalition behaviour in polities differing with regard to these two factors. Though the hypothesis was falsified, a more detailed analysis showed that (local) political competitiveness did make a difference in the choice of at least some of the coalition strategies. These differences, moreover, were in line with what was to be expected from the theoretical model.

The life expectancy of all political systems depends, among other factors, on the ability to resist military pressure and other means of coercion, that is, to mount and sustain a sufficient ‘security-related effort’(SRE). Yet over the last decades, SRE in the West has increasingly fallen short of what is required, especially vis-à-vis the main systemic competitor, the Soviet bloc. This trend can be explained by the interplay of two kinds of long-term change in Western societies. The first is the gradual penetrating and remoulding of their structures by capitalism, replacing both threat and integrative relations by exchange as a social organizer and thus delegitimizing the military (and other means of organized coercion), and at the same time rendering war much more destructive and costly by harnessing modern science and technology to the design and production of weaponry. The second process is the diffusion of representative, subsequently democratized, systems of government: decision-making by representative assemblies generates, via the mechanism of log-rolling, allocative decisions biased towards providing private goods (e.g. by tax reductions), merit goods, and transfer payments, and consequently disfavouring public goods such as deterrence of potential aggressors.

In the context of a general policy-based theory of how parties behave in democratic governments, a specific hypothesis on party factional behaviour is proposed and tested for the post-war period in 20 democracies. The hypothesis is that reshuffles (replacements and transfers of individuals between ministries) are more frequent where Prime Ministers have more freedom of action and decline where they have less freedom. As Prime Ministerial power is greatest in single-party governments, these are expected to produce more reshuffles – an expectation strongly confirmed by the evidence.