There is ongoing controversy as to whether political democracy inhibits or facilitates national economic growth. It is argued here that the answer to this question depends greatly on the regional political context within which democracy functions. In regions where social groups clamour for redistribution, as in Latin America, democracy may lead to populism and poor economic performance. Similarly, in regions where state elites are generally committed to promoting rapid industrialization, as in parts of Asia, democratic pressures may impede effective economic policy. However, in regions where patrimonialism is chronic, as in Sub-Saharan Africa, democracy may provide a useful mechanism for evicting grossly corrupt politicians and may therefore facilitate higher rates of economic growth. These regional arguments are tested statistically here and show that democratic governance constrains growth in Latin America and Asia yet facilitates growth in Africa. Sensitivity analyses indicate that these findings are fairly robust.

Capital requirements are an important regulatory device for the management of systemic risk in the banking industry. A puzzling fact is the existence of widespread over-compliance of banks regarding national and international regulatory capital requirements. Some light is shed on this pattern by studying bank capitalization decisions using 1,267 banks from twenty-nine OECD countries. The effect of market pressure (competition) on bank capitalization is found to be positive, whereas the direct effect of regulation is ambiguous. Three points can be made. First, the effectiveness of regulation must be evaluated taking into account the – often overlooked – indirect effects. Secondly, the impact of regulation can be amplified by competitive forces when there are synergies between regulatory and market pressures. Such synergies seem to have played a key role in the widespread adoption and effectiveness of international regulatory capital requirements (the Basle Accord). And thirdly, the prospects for international regulatory convergence in any area are influenced by the interconnectedness (synergies vs. rivalries) between regulatory and market processes.

This article analyses the role of parties in political systems with legislative initiative. The model explicitly takes into account the multi-district nature of elections so that voters recognize that they are not voting for a policy but for members of a policy-making body. In this setting, parties can attain their ideological goals without resorting to extreme discipline of their members in the legislature. With respect to the electorate, although voters are endowed with perfect information about their candidates' ideological stance, this model explains how the legislature will display correlation across ideology and party, a feature often attributed to imperfect information.

Scholars often use roll-call votes to study legislative behaviour. However, many legislatures only conclude a minority of decisions by roll call. Thus, if these votes are not a random sample of the universe of votes cast, scholars may be drawing misleading inferences. In fact, theories over why roll-call votes are requested would predict selection bias based on exactly the characteristics of legislative voting that scholars have most heavily studied. This article demonstrates the character and severity of this sampling problem empirically by examining European Parliament vote data for a whole year. Given that many legislatures decided only a fraction of their legislation by roll call, these findings have potentially important implications for the general study of legislative behaviour.

In this article, the reason why incumbent spending is less effective than challenger spending is explained. The argument is that incumbent spending efficiency depends on the marginality of seats: safe incumbent spending is less effective than marginal incumbent spending, since safe incumbents have to buy fewer extreme voters, whereas marginal incumbents can easily buy a larger number of swing voters. The analysis of the US Senate elections between 1974 and 2000 shows that safe incumbent spending is less effective than challenger spending, but marginal incumbent spending is not. The analysis also shows that the previous finding of less effective incumbent spending is largely due to the fact that the data for marginal and safe incumbents have been aggregated.

To what extent does public support for subnational officials fluctuate in response to local rather than national performance? Are the policy failures of subnational officials reliably punished by voters? Drawing upon both individual and aggregate level data, this article attempts to shed new light on these questions about the politics of decentralization by exploring electoral outcomes and public opinion at the subnational level in Argentina. Consistent with referendum voting models, this analysis suggests that the fate of candidates in both national and subnational elections is shaped by the performance of the incumbent presidential administration. Moreover, to the extent that subnational performance has an electoral impact, voters do not necessarily respond in ways that enhance electoral accountability. Voters not only blame and reward subnational officials for national performance, but also attribute responsibility for subnational performance to national authorities. The implications with respect to the impact of decentralized decision making on democratic accountability are decidedly mixed and anything but consistent with the argument that decentralization results in a closer match between citizen preferences and the allocation of public resources.

Immigration has become a highly salient political issue in many of the world's affluent democracies. Yet, the electoral dynamics of anti-immigrant sentiment remain barely understood. We distinguish two dimensions of concern about immigrants: material threat and cultural threat, and hold that the influence of both on the right-wing populist party vote is critically mediated by policy preferences to restrict immigration and to isolate Australia from foreign influence. The result is a path model of voting that allows material and cultural threat to influence policy preferences about how to deal with the ‘immigrant problem’, and allows both threat and policy preferences to affect voting for the far-right One Nation party in Australia. Our results confirm that popular concern about immigrants is multi-dimensional and that its two dimensions have different sources. We also demonstrate that anti-immigrant sentiment works indirectly through policy orientations to influence vote choice. Feelings about immigrants, in other words, have an electoral effect only when there is a good fit between the policy stances of voters and the policies promoted by the parties on offer.

If voters use information about the economy to assess the competence of incumbents, a connection between economic conditions and incumbent success should only be discernible in settings where public policy might plausibly affect the economy, and where the assignment of government responsibility is relatively straightforward. Applying this logic to gubernatorial elections in the United States, we test the following hypothesis: the connection between economic conditions and incumbents' vote shares is mediated by the structure of the state economy. This hypothesis is premised on the idea that voters understand that raw macroeconomic aggregates – when driven by factors like weather, commodity prices and federal policy – are poor signals of incumbent performance. Using data from gubernatorial elections held between 1950 and 1998, we show that the connection between macroeconomic indicators and incumbent success is weak in states dominated by natural resources and farming but quite strong elsewhere. This finding helps explain why earlier studies found no connection between state-level economic conditions and gubernatorial elections.

The US federal bureaucracy implements the nation's laws while juggling its own preferences and the preferences of numerous stakeholders. This article begins to unpack the conditions under which the bureaucracy responds to its stakeholders during rule making, and it is argued that bureaucratic responsiveness is conditioned on the level of attention provided by Congress and the President. This argument is tested with a dataset of forty rules and 1,444 interest group comments. Interest group influence is found to be constrained by congressional – but not presidential – attention to rule making. The results also shed light on Wilson's interest group theory of politics and suggest that the fragmentation of group preferences is less important than the central message sent by groups.

Recent empirical studies on American elections suggest that campaigns provide voters with the necessary information to make reasoned voting decisions. Specifically, campaigns help voters learn about the electoral relevance of ‘fundamental variables’, such as the economy and party stances, that have been consistently shown to predict electoral outcomes. Do these findings generalize beyond the American case? This article uses cross-national survey data in order to subject this thesis to a more comprehensive test. The analysis provides further support for the hypothesis that campaigns ‘enlighten’ voters as the election draws near. Moreover, the article shows that some voters learn more from campaigns than others. Campaign effects are more pronounced among individuals with low political sophistication and those living in party list systems. Implications for future research are explored, suggesting a ripe research agenda using under-tapped cross-national data.

Scholars often observe that the foreign policies of states are not made in a vacuum but rather are determined or moulded to a significant degree by the external and internal actions of rivals. Domestic unrest is often considered a potential impetus for changing strategic behaviour. Leaders may be tempted to employ force externally to divert attention away from domestic unrest. The intended result is a ‘rally round the flag’ effect that culminates in higher approval/support for the executive as citizens forget about domestic problems and pay attention to a common adversary. One implication of this sort of ‘diversion’ is that potential scapegoats might employ strategic behaviour to avoid becoming a diversionary target. In other words, when they witness domestic unrest in a rival state, they worry that the rival may lash out at them and thus engage in ‘strategic avoidance’.

Conversely, strategic behaviour may lead to a greater chance that the potential ‘diverter’ will itself be targeted for hostile behaviour. Erstwhile scapegoats may view periods of social unrest such as elections, domestic political protests or unstable cabinet structures in the other country as convenient and favourable times to escalate hostility. Such situations are viewed as opportunities that are ripe for exploitation.

Alastair Smith's work has been extended to both the US case and a comparative cross-national study. Our purpose here is to extend this line of inquiry by looking at a region of the world locked in a long-term hostile relationship; namely, the Middle East. Our approach builds upon previous research that addresses the strategic interaction of enduring rivals.

The ‘varieties of capitalism’ literature identifies the labour market and labour relations system as an institutional feature by which liberal market economies (LMEs) and co-ordinated market economies (CMEs) of advanced countries may be distinguished. This literature argues for the resilience of CME institutional features to the allegedly convergence-inducing or homogenizing effects of globalization. Resilience to convergence has also revealed different patterns of institutional adaptation to globalizing pressures among CMEs. This article examines the adaptation of labour institutions to globalization using the Northeast Asian Newly Industrialized Countries of South Korea (hereafter Korea) and Taiwan as case studies. Examples are given of how Korea and Taiwan have pursued labour market liberalization, but within institutional conditions (policies and politics) distinct from those associated with transitions to the LME. The origins of this distinctive pattern of response are explained, pointing out the need for explanations based on globalization to be supplemented by ones that emphasize economic legacies and political logics.

It is a fairly common criticism of deliberative democratic ideals that one cannot involve thousands, let alone millions, of people in a decision-making process and still retain its deliberative character, at least not in any sense as strict as those of the classic expositions of deliberative democratic principles. Various solutions to this scale problem have been proposed, but implicit in most deliberative designs is the media solution. People can watch the debate unfold on television, can listen to it on the radio, can follow it, or even contribute to a limited extent, online. This is supposed to ensure not only that deliberation is subject to the check of publicity, but also that it is dispersed throughout the public sphere.

This is precisely the idea behind James Fishkin's deliberative poll technique, designed to be televised so that deliberation among a few would be brought to ‘an audience of millions’, ‘bridging the gap’ between informed and uninformed opinion, and acting as a ‘catalyst’ to change the nature of discussion on a topic in a broader community. But what happens when we rely on the media to bear the weight of responsibility for generating links between insiders and outsiders in public deliberation? The media are not broad pipes which simply convey whatever is put into them but, like all institutions, shape that input in sometimes significant ways. If we are to understand the value of the media in addressing problems of deliberative scale, we need to understand those institutional effects.

In this Research Note I examine the effects the media had on the kinds of information transferred from deliberators to audience in a deliberative poll that took place in the United Kingdom in July 1998.

Those who have not witnessed the making of a government have reason to be happier than those who have. It is a thoroughly unpleasant and discreditable business in which merit is disregarded, loyal service is without value, influence is the most important factor and geography and religion are important supplementary considerations.

The Borden Ministry was composed under standard conditions, and was not, therefore, nearly as able, honest, or as industrious an administrative aggregation as could have been had from the material available … There were some broken hearts – in one instance, literally. In others, philosophy came to the rescue, but the pills were large and the swallowing was bitter.

Paul Bilkey

Prime ministers can typically rely on ideological agreement and norms of loyalty to deliver them a modicum of party cohesion. Beyond that, they have at their disposal a variety of institutional tools with which to enforce discipline. The powers to invoke the confidence convention and to dissolve parliament are the most well known and powerful of these tools, but these heavy-handed measures are ill-suited for securing unity on an on-going basis. The prime minister's monopoly over the distribution of preferment is a far more reliable means of ensuring members' loyalty. The rules of the game are simple: if the member of parliament (MP) wishes to climb the parliamentary career ladder, he or she must toe the party line. The prime minister's power over MPs' parliamentary careers is not without limit, of course. Some MPs must be brought into cabinet because they are too powerful and dangerous to leave on the backbench where they can openly challenge the prime minister.

Costs associated with voting affect an individual's willingness to turn out for an election as well as aggregate rates of voting across political jurisdictions. Barriers to participation also skew the social and economic composition of electorates. In this Research Note, we suggest that the costs of participation affect candidate behaviour as well – the strategic purposes of their appeals to voters and the media they choose to deliver messages. Why? By making the trip to the ballot box more or less difficult, electoral laws select voters with respectively less or more interest in and thus knowledge of politics. Given the systematic variations in how people with different levels of political knowledge learn during a campaign, we anticipate that election laws ultimately influence the communication tools that candidates use.

We propose that the costs of voting have a compositional effect on electorates: as voting becomes increasingly difficult, the average level of political knowledge and interest among voters should decrease. This is not due to a micro-level effect in which registration laws somehow make individual voters smarter or better informed. Institutionally imposed costs simply affect who can and will vote. For a brief example, if one state charged its citizens £50 to vote while another paid its citizens £50 at their polling site, we would expect quite sizeable differences in turnout and composition of the electorate across the two states. The main contribution of this note, however, is our recognition that campaigns adapt to these differences in systematic ways. Real world differences in the cost of voting are not as great as these financial disparities, but the idea is the same: costs borne by individuals will have selection effects that produce different types of electorates and prompt different campaign styles by candidates.

Every face-to-face Deliberative Poll (DP) to date has been the subject of a television broadcast. We consider these broadcasts a helpful adjunct to the design – a way of motivating both the random sample and the policy experts and policy makers to attend, of educating the broader public about the issues, and, perhaps, of nudging public opinion in the direction of the results. In ‘Rickety Bridges’, John Parkinson examines just one of these broadcasts, Channel 4's on the DP on the future of Britain's National Health Service (NHS) in 1998. Applying his coding of the contents to other DP broadcasts might or might not yield similar results, but we are happy to assume, for argument's sake, that it would. If DP broadcasts are generally doing what he describes the NHS DP broadcast as doing, they are doing pretty well, at least as far as the distribution of coverage is concerned. It is Parkinson's notion of what they should be doing that is mistaken. As a result, his critique is fundamentally misguided.

THE AIMS OF DP BROADCASTS

Parkinson's critique rests on an inappropriate standard. His central claim is that Channel 4's broadcast of the NHS DP did not replicate the participants' experience. Of course it did not. No broadcast could ever give viewers the same experience they would have if they were actually part of the DP's on-site, weekend-long deliberations. The broadcast in that case would have to be weekend-long, and there would actually have to be multiple broadcasts – as many as there are participants – since every participant's experience is different.

In her provocative article, ‘Agency, Truth and Meaning: Judging the Hutton Report’, Diana Coole makes two important claims: first, that ‘political inquiry is impoverished to the extent that theorists … ignore the more mundane reports and statements that help constitute everyday political life’; and secondly, that a theoretical analysis of the Hutton Report shows it to have been informed by a particular conception of truth (Lord Hutton's own conception) that largely guaranteed its conclusions and served to make Dr Kelly and the BBC ‘victims of a particular sense of truth’.

I share Coole's belief that political theorists should contribute more to the analysis and understanding of ‘mundane’ texts, but I have grave reservations about the kind of analysis she favours, and about its implications in this particular case. It seems to me that her focus on truth is de-politicizing, while her conclusion that Dr David Kelly was a ‘victim’ rests upon a simplistic understanding of moral agency and responsibility. If we wish to bring the resources of theory to bear on everyday political texts, we do better, I suggest, to focus on moral and political philosophy than on epistemology.

TRUTH

Coole claims that the conclusions of the Hutton Report were informed (even determined) by a ‘positivist-juridical (empiricist-legalistic)’ conception of truth, one which displayed ‘antipathy towards the idea that truth is a matter of interpretation and hence dependent on subjective judgements’. As this conception worked its way through the inquiry it ‘guaranteed’ partisan conclusions that were at odds with the self-understanding of other protagonists in the affair, most notably that of Dr David Kelly.

I am grateful to James Fishkin and Robert Luskin for taking the time to consider and respond to my remarks on the dangers of relying on the news media to form a bridge between participants and non-participants in deliberative events. Clearly we are in agreement that there is a difference between deliberations as experienced by participants and a television audience. However, Fishkin and Luskin have misrepresented my aims and claims about that difference, and thus their response is less telling than it might have been, directed as it is at a man of straw.

At the outset, let me emphasize that the purpose ofmyresearch note was not to subject deliberative polls to a thorough critique: one case study would indeed have been an inadequate foundation for such work. Rather, the case study was a device used to highlight a feature common to many attempts to put deliberative democratic principles into practice, not just the deliberative poll (DP). The underlying issue is, to state it again, one of scale: millions of people cannot strictly deliberate together. Walzer proposes a maximum number of twenty individuals, although the literature on small groups suggests that the actual limit may be between five and seven, both which seem disturbingly small for purportedly democratic processes. Even if the deliberative limit were several hundred, however, the scale issue still arises and causes problems for the democratic legitimacy of such events: how can their decisions be binding on others when those others have not had the opportunity to have their views transformed by the encounter with the better argument?

My purpose in the article was twofold: to commend to political theorists a method of critical analysis that can usefully be applied to texts that circulate within public life, and to apply it myself in reading the Hutton Report. Like many people, I was incredulous when the report appeared and suspicious of its conclusions. But I was also curious as to how it managed to appear simultaneously both objective and biased. Consequently, I wanted to investigate how political power was operating at a deeper discursive level. I welcome this opportunity to respond to Susan Mendus's detailed and thoughtful response, both to defend my methodology and to clarify those points where I think she misinterprets its implications. Because her reply is so clearly structured, I have replicated that structure in my responses.

TRUTH

I distinguished in the article between two competing views of truth, each of which may be situated in presuppositions regarding agency, language and meaning. In the first part of her reply, Mendus contests my observations regarding the empiricist and legalistic view of truth that I associated with the Hutton Report. She makes three observations here.

First, she is sceptical that Hutton subscribed to this positivist-juridical conception at all and as evidence, she cites an early section of the report where he explains his intention to reproduce parts of the inquiry's transcript of evidence so that the public can understand why he reached his conclusions. ‘These are not’, Mendus concludes, ‘the words of a man who denies the need for interpretation.’