Analysing multiparty competition in plurality rule elections

Abstract

Abstract:The study of general election outcomes can be helped by finding better approaches for visualizing large quantities of information and asking questions about its patterning. We review the Nagayama or 'all possibilities' triangle display, and show that it can only be legitimately used to show an overall 'field' of results that is logically feasible, called the effective space of competition, which varies with the number of observable parties. We apply this reductionist view to analyse outcomes in three leading plurality rule systems (the USA, India and Great Britain), focusing on evidence of the Duvergerian psychological effect acting on voters during campaign periods. The ECS view illuminates some key differences across countries, and variations with rising numbers of parties competing. We next consider a more holistic approach, the 'crown' diagram, which links electoral district outcomes more closely to the most important politico-ideological dimension in each country. Both views suggest some tentative evolutionary hypotheses for the variegated development of plurality rule systems over time. Britain is a highly nationalized party system, but one that has moved substantially away from Duvergerian predictions of two-party focusing, and towards multi-party politics. The USA seems to be a case of 'stunted development'. And India shows a partial Duvergerian conformity, yet combined with a substantial vertical scatter of non-Duvergerian results. Applications to over-time and regional analysis within countries are also sketched. In the comparative analysis of elections and party systems we have yet to develop logically acceptable ways to chart the district-level outcomes of multi-party elections, and to assess the clustering or patterning of outcomes in systematic ways. In this paper we show that the Nagayama or 'all possibilities' triangle has major defects, but can be reformulated and re-applied in two new ways. The first captures the 'layer cake' character of general election outcomes in a reductionist fashion, showing how the number of observable parties competing for votes at district level influencesoutcomes. An alternative variant (the crown diagram) gives a more holistic picture of outcomes, shifting attention to the performance of the top two parties or blocs linked to the predominant political-ideological dimension in a political system. We link this second innovation to a tentative logic of evolutionary development applicable to plurality rule election systems. Graphic representations of multi-party competitionIn his discussion of 'paradigms' in science, Thomas Kuhn (1996) emphasized that the term is meant to cover not just very basic or fundamental ideas that may lie at the heart of salient encompassing theory changes, but also to whole congeries (or 'swarms') of complementing ideas, methods, and practices, many of which concern instrumentation, measurement, schemas, and the analysis, representation and visualization of data (Buchanan, 2000, p. 233-4). There are still widespread problems of defect...

Neither the T Index nor the D2 Score Measure “Two-Partyness”: A Comment on Gaines and Taagepera

“…I have implemented some minor consistency and ENPv calculation corrections to the original data. I also thank Rekha Diwakar of the University of Sussex for the India data here, drawn from Dunleavy and Diwakar (2013). Naturally all interpretations of any of this data are mine alone.…”

“…In fact, we know empirically that this is perverse -that in the most conventionally recognized two-party system, the United States, very high levels of P 1 2 P 2 leads routinely occur, and only a tiny fraction of results empirically occur at P 1 2 P 2 leads below 5%, as Table 4 demonstrates. Now there are good reasons for believing that in party system evolution terms (as in so many other respects) the United States may be an exceptional case, a strangely "frozen", super-Duvergerian party system still showing remarkably few of the multi-partism trends characteristic of most large and modern liberal democracies (see Dunleavy & Diwakar, 2013). Yet Table 4 shows that Ghana and Bangladesh have more districts than the United States where P 1 2 P 2 leads are below 5%.…”

Strategic Voting in Slovak Regional Elections of 2013 and 2017: Evidence for Duverger’s Law in Elections of Presidents of Regional Self-Government?

“…Some empirical studies (Gaines 1999;Diwakar 2007;Chytilek 2007;Grofman et al 2009;Dunleavy -Diwakar 2011) found various violations of the Duvergerian assumptions at the level of individual countries (e.g., India, Canada, UK, USA). However, a recent analysis working with large data sets of elections in SMDs confirmed that district magnitude had (at the district level) the effect that Duverger had expected, although the effect of electoral institutions could be contingent on and (at the district level) inhibited by country-specific conditions such as social cleavages that generate demand for additional parties.…”

Green governance? Local politics and ethical businesses in Great Britain

“…Equally, local politics and voting patterns are very different from national ones. So, it is possible to identify a far greater contribution of these “radical third party” elements at local Council level than through Parliamentary constituencies (Dunleavy & Diwakar, ).…”