Applying insights from social-choice theory to illuminate the functioning of pluralitarian Westminster institutions, this article develops a coherent political answer to four puzzling questions about the economic liberalization that transformed New Zealand in 1984–93: why an anti-statist programme was initiated (and largely accomplished) by a labour party, why restructuring was more radical in New Zealand than in other democracies, why reformers were able to prevail through two elections and a change of government, and why they committed costly policy-sequencing errors. Understanding this remarkable case has implications for empirically grounded social-choice theory, the political theory of policy reform, and the evaluation of pluralitarian democracy – which New Zealanders themselves repudiated in 1993 by adopting proportional representation.

In the twenty years after 1945 both the United States and Britain created public funding regimes for social science, through the National Science Foundation (NSF) and the Social Science Research Council (SSRC) respectively. The historical and political contexts in which these institutions were founded differed, but the assumptions about social science concurred. This article uses archival sources to explain this comparative pattern. It is argued that the political context in both countries played a key role in the development of the two research agencies. In each country the need politically to stress the neutrality of social research – though for different reasons in each case – produced a bias towards positivist scientific methodology, untempered by ideology. This propensity created the trajectory upon which each country's public funding regime rests.

Constitutions devised by rank-and-file legislative politicians in the ‘Third Wave’ of democratization tend to exhibit an inverse relationship between party strength and executive strength. Party strength refers to the extent to which legislators campaign on the basis of their parties' reputations as providers of public policy as opposed to personal reputations as providers of more narrowly targeted services. Executive strength refers to the constitutional authority of the executive to influence policy independent of partisan support in the legislature. Politicians' interest in one configuration of power or another can be deduced logically from the situation in which they find themselves at the time they are designing a new democratic regime.

This article uses the British Election Panel Study to assess the impact of voters' and party positions vis-à-vis European integration on Conservative electoral support between 1992 and 1996. Over this period levels of public support for European integration declined markedly, so that by 1996 the Conservative party was even closer to aggregate public opinion, when compared with its main competitors, than it had been at the time of the 1992 election. However, an analysis of the proximity between individuals' positions on integration and the positions they then attributed to the parties indicates that Conservative divisions over Europe helped turn this potential electoral asset into a liability, leaving the party further from individual voters' own positions than were either of the other two main political contenders. Moreover, as issue proximity on integration predicts voting even when past vote and proximity on other issues are controlled for, it is likely that the European question will have resulted in electoral costs rather than the benefits it could have produced. One implication of these findings is that if the Conservatives hope to do well on this issue they will need to adopt a consistent Eurosceptic line, but such a strategy is unlikely to be easily maintained.

Recent studies of lynching have focused on structural theories that have been tested with demographic, economic and electoral data without much explanatory success. This article suggests that lynching was largely a reflection of a facilitating subculture of violence within which these atrocities were situationally determined by cultural factors not reported in census and economic tabulations, or election returns. Lynching declined in the twentieth century, in part, as a result of segregation and disfranchisement policies, but mainly because state executioners replaced lynch mobs in carrying out the will of the white majority.

Freedom of expression is celebrated as one of the glories of the American political system. But does all speech deserve immunity? In particular, should speech designed to vilify or degrade on the basis of race be protected? Opinions on racist speech are complicated because they must accommodate two fundamental democratic principles that operate at cross purposes: freedom of expression, which implies support for racist speech, and racial equality, which implies the opposite. Using data from the 1990 General Social Survey, we examine how Americans resolve this conflict. Our major finding is that the principle of free expression dominates the principle of racial equality. What contemporary legal scholars regard as a hard case entailing a collision of democratic principles, ordinary Americans seem to interpret as a straightforward application of just a single principle. This result mirrors and perhaps reflects a nearly century-long and mostly lop-sided debate favouring free speech among American elites.

Reif and Schmitt argued that elections to the European Parliament should be understood as second-order national elections, and advanced several predictions about the results of such elections. Those concerning the impact of government status, party size, party character and the national election cycle on electoral performance are examined here using data on four sets of European Parliament elections. In addition, the consequences of European Parliament elections for the next national election are explored. The analysis demonstrates the validity of most of Reif and Schmitt's original propositions, and further refines their analysis of the relationship between European and subsequent national elections. However, all propositions hold much more effectively in countries where alternation in government is the norm, suggesting that the distinction between first-order and second-order elections may not be so clear cut as Reif and Schmitt imagined.

The political and economic couplings between wage bargaining institutions and macro-economic policy regimes are explored in this article. It is argued that in advanced industrialized democracies with well-organized unions and employers' associations, macro-economic performance (especially unemployment) is the outcome of an interaction between the centralization of the wage bargaining system and the monetary policy regime. Thus, a decentralized bargaining system in which the government in credibly committed to a non-accommodating monetary policy rule poses an institutional alternative to a centralized mode of wage regulation where the government enjoys macro-economic policy flexibility. Based on time-series data from ten highly organized market economies, I show that both of these institutional ‘equilibria’ produce superior employment performance, but also that the two systems are associated with very different distributional outcomes, and that they are supported by different coalitions of organized interests. In addition to predicting economic outcomes, the proposed model provides a theoretical framework for analysing institutional change in wage bargaining systems and in macro-economic policy regimes.

This article analyses Imperial Germany's legendary coalition of landed aristocracy and heavy industry around a policy of tariff protection. Using a simple model of voting behaviour, where party affiliation serves as a partial intervening variable between constituency interests and legislative votes on trade policy, I test hypotheses derived from three different interpretations of the ‘marriage of iron and rye’. Roll-call votes from four key divisions in the Reichstag are analysed in a number of forms, ranging from cross-tabulations to conditional logistic regression. Ronald Rogowski's ‘factor endowment’ model offers an important dynamic perspective that is lacking in the others, but his model must be reconciled with anomalies that arise in the short run. Rather than attempting to disentangle political party ideology from constituents' interests, more insight may be gained from understanding why the effects of the two causal factors were not fixed, and how they varied over time.

This article provides an empirical analysis of voting behaviour in the second ballot of the 1990 Conservative leadership contest that resulted in John Major becoming party leader and prime minister. Seven hypotheses of voting behaviour are generated from the extant literature relating voting to socio-economic variables (occupational and educational background), political variables (parliamentary experience, career status, age and electoral marginality) and ideological variables (drawn from survey data on MPs' positions on economic, European and moral issues). These hypotheses are tested using data on voting intentions gathered from published lists of MPs' declarations, interviews with each of the leadership campaign teams, and correspondence with MPs. Bivariate relationships are presented, followed by logistic regression analysis to isolate the unique impact that each variable had on voting. This shows that educational background, parliamentary experience and (especially) attitudes to Europe were the key factors determining voting. The importance of Europe in the contest is particularly instructive: the severe problems for Major's leadership which were caused by the issue can be attributed to, and understood in the context of, the 1990 contest in which he became leader.

Little is known about how ordinary Europeans feel about the central policy-making institutions of the European Union (EU). This has encouraged us to analyse mass attitudes towards the legitimacy of the European Court of Justice (ECJ). Relying on a cross-time (1992–93) panel analysis, as well as a cross-institutional analysis (the ECJ, the European Parliament and the high courts of the member states), we discover that (a) the ECJ does not possess a surplus of legitimacy, and it is doubtful whether the legitimacy shortfall is only a short-term function of the row over Maastricht; (b) attitudes toward the ECJ, although in the aggregate fairly stable, changed significantly over the one-year panel survey; (c) the European Parliament has little legitimacy it can share with the ECJ; and (d) although the national high courts do have greater legitimacy, there is little evidence that they are capable of transferring that legitimacy to the ECJ. We conclude with some speculation about whether the ECJ will be able to build greater legitimacy, and the consequences for the EU if the court fails to do so.

Strategic models of coalition bargaining formation have demonstrated the importance of institutional features for an understanding of cabinet formation in West European democracies. Yet little is know about the empirical regularities of government formation processes. In this article we analyse the duration of formation processes using a semi-parametric estimation procedure on a dataset of 304 government formations in thirteen multi-party democracies. The results are consistent with a bargaining approach under incomplete information.

This article identifies the different ways in which the Eurobarometer surveys measure party attachment in different sets of member states of the European Community or Union. The main variants are an absolute question, an ordinal question and a relative question. Using Eurobarometer data covering the period 1978–94, it is shown that these different questions produce predictably different estimates of the distribution of party attachment. The methodological interpretation is confirmed and a new equivalent measure of party attachment is constructed using the results of a specific test inserted for this purpose in Eurobarometer 41.1. The new measure is constructed in such a way as to preserve the comparability of Eurobarometer data on this topic over time. The new operationalization significantly improves the predictive power of the construct and it is recommended that it be adopted in future.

The political economy literature has not considered in detail the dynamic interrelationship between one politico-economic system and another. To fill this lacuna, the author uses a VAR model to study the relationship between social democratic popularity and the economy not only within but also between the Scandinavian countries. In addition to the expected macroeconomic interdependence, there is empirical support for the traditional partisan theory. Economic voting is found only in Denmark. Social democratic popularity is not much affected by the economy, but it is found that the economy is influenced by social democratic popularity. An increase in the popularity of the social democrats eventually causes higher inflation. In particular, the popularity of the Swedish Social Democrats creates inflationary pressure in Sweden, and this pressure is exported to Denmark and Norway as well.

Social class has long been assumed to be the predominant social or structural determinant of voting behaviour. This article assesses the effect of class on voting behaviour at the 1992 general election by adopting the causal modelling perspective developed by Warren E. Miller and J. Merrill Shanks. It explores two mechanisms (party identification and left–right ideological positions) which may mediate the effect of class on voting behaviour. However, it demonstrates that wherever class is assumed to be located in the causal order, it does not dominate analysis of voting behaviour and left–right positions.

This article applies the portfolio allocation model of government formation to the formation of local government administrations in general, with a particular empirical application to the formation of administrations in British local authorities with hung councils. First, the implications of the portfolio allocation model for the formation of local administrations are discussed. Secondly, this model is operationalized in the context of British local politics, and specific empirical implications are explicitly derived from it. Thirdly, these implications are systematically evaluated against the recent experience of the formation of local administrations in Britain. The results suggest that policy is indeed important in the formation of British local administrations, but that more than a single left–right dimension of financial policy is needed to describe this process adequately.

Changes in activism among grassroots members of the Labour and Conservative parties in Britain are modelled using panel surveys in this article, with party members being surveyed before and after the 1992 general election. The evidence suggests that a decline in activism over time has occurred in both parties, but this decline is greater in the Labour party than in the Conservative party. This is attributed to a number of political, cultural and sociological changes in society over time, but in the short run the main factor is the outcome of the general election in 1992. These trends are modelled by means of a ‘general incentives’ theory of activism, which explains the decline in activism in terms of changing incentives for political action. Reductions in incentives for activism were, with one exception, greater for Labour party members than for Conservatives, which explains the differences between the two parties. This may produce a ‘spiral of demobilization’ in which electoral losses produce a decline in activism and campaigning at the local level, which in turn leads to further electoral losses.

The general elections in Chile and Brazil in 1989 marked ‘the first time that all the Ibero-American nations, excepting Cuba, enjoyed the benefits of elected constitutional governments at the same time’. This occurrence was not as dramatic or visible as the collapse of Communism and the transitions to democracy in Eastern and Central Europe, which began in the same year, but it did mark a historical watershed. After almost two centuries as independent states, the countries of Latin America now comprised a new democratic universe. Similarly to the Eastern European experience, the Latin American watershed presented new opportunities for comparative research into democratic governance. In particular, it created a new context for the study of the relationships between institutional design, party systems and governability – defined in the narrow sense of institutional efficacy, as expressed through government stability, legislative capacity and the avoidance of gridlock. This article sets out to review the recent research on these topics, in order to describe the predominant regime type in Latin America and differentiate its distinct variants, examining their impact on governability.

The Enlightenment model of social life is a seductive one. It depicts rational (or anyway reasoning) individuals choosing goals and plans and projects for themselves, with those autonomous individuals then coming together, of their own volition, in pursuit of shared interests and common goals. This founding fiction of the modern social world is well captured in the words of one Renaissance writer, who has God saying to Adam:

Taking it in turns is an endemic feature of political and social life. For instance, in a democratic system parties take turns in office according to the outcome of elections. Despite its empirical importance, taking it in turns has received virtually no attention from rational choice theorists. First, a general definition of taking it in turns is given. Secondly, the conditions for taking it in turns to be stable and efficient in Prisoners' Dilemma, Chicken, and Co-ordination supergames are discussed. The key result is that, where stable adherence to a taking it in turns pattern demands that players' continued adherence is conditional on the past adherence of others, players will stick with the pattern if they get their turn often enough. The precise conditions are shown to depend on the payoff structure and the discount rate. Potential problems are identified with initiating stable taking it in turns because of the existence of multiple equilibria and players' desire to get a stable taking it in turns pattern that gives them a high expected payoff. The paradigmatic version of this problem is who goes first when two players alternate between getting a high and a low payoff. The applicability of the results is illustrated by considering the questions of democratic stability and transitions to democracy.