Sixty years of democratic representation in Germany allow us to study the working of a specific type of electoral system, namely a mixed system combining proportional and majoritarian rules, in great detail. Mixed systems have figured as a reference point in many reform debates of the recent past. This is because they appear to combine advantageous traits of proportional and majoritarian rules, such as fairness, proximity between constituencies and representatives, and stable government majorities. Mixed systems have also attracted much scholarly attention of late, because they allow us to study the effects of electoral rules while holding many intervening variables constant. But they also attract interest because the proportional and majoritarian electoral tiers affect each other in ways that differ from what would have resulted under pure PR or plurality. All this makes mixed systems a fascinating object of study, and the German system is its oldest and prototypical exemplar.
|Product dimensions:||6.14(w) x 9.21(h) x 0.50(d)|
About the Author
Philip Manow is professor of comparative political economy at Bremen University. Previously he held positions at the Max-Planck Institute for the Study of Societies, Cologne, and at Konstanz and Heidelberg Universities. He was a visiting scholar at Harvard’s Center for European Studies, at the Centre Études Européenne, Sciences Po Paris, and was fellow at the Wissenschaftskolleg zu Berlin. His research interests cover democratic theory, the German political system, comparative political economy and European integration. He has published in Legislative Studies Quarterly, European Journal of Political Research, Politics & Society, Comparative Political Studies, West European Politics and the British Journal of Political Science, among others.
Read an Excerpt
Mixed Rules, Mixed Strategies
Candidates and Parties in Germany's Electoral System
By Philip Manow
ECPR PressCopyright © 2015 Philip Manow
All rights reserved.
The German electoral system
Mixed-member electoral systems (MM) have recently attracted increased academic interest. (Shugart and Wattenberg 2001b; Massicotte and Blais 1999; Moser and Scheiner 2004; Massicotte 2011; Moser and Scheiner 2012; Massicotte forthcoming). Regarding the number of studies treating these hybrid electoral systems, some have even spoken of an emerging 'cottage industry' (Nishikawa and Herron 2004: 753). That mixed systems attract this growing scholarly attention is explained by both a substantive and a methodological interest.
Regarding the substantive interest, MMs apparently combine the beneficial features of pure plurality and pure Proportional Representation (PR) systems (Shugart 2001b, 2001a; but see Bawn and Thies 2003). It, therefore, comes as no surprise that they have recently figured prominently in reform debates in many countries. Mixed-member electoral systems have become the 'most popular alternative to first past the post' systems (Blais 2008: 3) since they apparently meet the two prime goals of recent electoral reforms: proportionality and personalisation (Renwick 2010; Renwick 2011). They combine, as Moser and Scheiner note, the 'proportionality and small group representation commonly associated with PR and [...] the geographic representation of a particular locale and the large, catchall parties that are characteristic of SMD systems' (Moser and Scheiner 2012: 6). And – as at least the German case testifies – fair representation and personalisation does not need to come at the cost of lower government stability.
Today, around twenty-six countries apply mixed electoral rules for national elections. To this number one could add the state elections in the German and Austrian Länder (Massicotte 2011: 100–101), which increasingly emulate the two-vote proportional model of the federal level (Eder and Magin 2008). Overall, 'the use of majoritarian electoral systems has significantly declined while that of mixed systems has increased', as Bormann and Golder find in their recent overview of post-war electoral systems (Bormann and Golder 2013: 363). True, the prediction that the switch to mixed rules will be the most frequent electoral reform in the twenty-first century (cf. Shugart and Wattenberg 2001c) appears today as having been prematurely judged when observing the decreasing number of countries that have recently undertaken such a switch. Yet, all still agree that mixed systems 'remain an important family of electoral systems worth scrutinising' (Massicotte 2011: 101). And among those, the German case is arguably the best suited for such scrutiny.
With respect to methodological interest, mixed systems have become a favourite object of study since they seem to offer the opportunity to examine the effects of electoral rules within a quasi-experimental setting (Stratmann and Baur 2002; Moser and Scheiner 2004, 2012: 44–61). They allow us to hold constant a number of intervening variables when studying particular effects of electoral rules – like the business-cycle, the political culture, party system characteristics (e.g. cleavage structures), and the institutional characteristics of the political system. The promise is that a 'mixed-member electoral system allows us to hold the environment constant and isolate the effects of different electoral rules' (Moser and Scheiner 2012: 46). If we are able to control for many factors with well-established impacts on political outcomes, this could help a great deal in ascribing causal effects to electoral rules.
However, the growing literature on 'contamination' has argued that whereas the study of mixed member electoral systems might allow for the controlling of a number of intervening variables, the analysis of the 'pure' effects of electoral rules is hindered by their interaction (Ferrara, Herron, and Nishikawa 2005b; Herron and Nishikawa 2001; Cox and Schoppa 2002; Ferrara 2004; Nishikawa and Herron 2004; Ferrara and Herron 2005). These interesting contamination effects then, themselves, have motivated many recent studies, which analyse the consequences of mixing majoritarian elements and principles of proportional representation in one electoral system.
This might also explain the renewed interest in the German electoral system that is reflected in a series of recent publications. One of the prominent features of the German electoral system that links its plurality tier to its PR tier is the possibility for candidates to run in a district and on a party list simultaneously (dual candidacy). Another feature is the way mandates won in districts and those won via the party-list are charged against each other in order to secure overall vote-seat proportionality. One empirical indication that the plurality tier and the PR tier of the German electoral system do indeed contaminate each other is the observable non-Duvergerian tendency in its plurality tier: the effective number of district candidates is regularly higher than two. The consequences of these institutional features of Germany's mixed system for the strategies of candidates, members of parliament (MPs) and parties will be the focus of the following analyses.
Mixed electoral systems combine (some kind of) plurality with (some kind of) proportional representation (Massicotte and Blais 1999). Being in place since 1949, the German electoral system appears as the 'archetype' (Saalfeld 2005: 209) or 'mother' (Carey 2009: 32) of this type of electoral system, as it is its 'oldest and most-copied' exemplar (Cappocia 2002; Sieberer 2010: 488). The German electoral system combines regional closed-list PR with a nominal plurality vote in single-member districts (see also Nohlen 1978; Shugart and Wattenberg 2001c; Scarrow 2001; Klingemann and Wessels 2001b). Voters can cast two votes. With one vote they elect candidates in single seat districts with relative majority. With the other they vote for a closed regional party list. The list-votes decide the parties' overall seat shares for all – currently – 598 Bundestag seats, double the number of the – currently – 299 single member electoral districts. The candidate vote determines who will represent the district in parliament, and thereby, also the internal composition of each parliamentary party, especially the relative weights between district and list MPs. Since the districts won are subtracted from a party's overall seat share, calculated on the basis of all list votes, the higher the number of districts won, the lower the number of candidates that are drawn from the list. The seat-linkage between both components of the electoral system guarantees that proportionality dominates the translation of votes into seats. Accordingly scholars have labelled election systems, like the German one, Mixed-member Proportional systems, short MMP (Shugart and Wattenberg 2001c), or – in a more critical vein – as 'simply a more complicated way of getting the same basic PR outcome' (Bawn 1999: 490). This also explains why many previous studies of the German system have tended to ignore its plurality-tier – as the following chapters will show, at the cost of failing to fully understand its working logic and its interactive effects.
The combination of a PR and a plurality tier in the German electoral system provides all those who are interested in the effects of electoral rules with an interesting testing ground. In particular, those interested in the differences between list-PR and nominal vote in single-member districts, but also those who are interested in the mutual impact of electoral rules on each other ('contamination'), can gain many new insights. Analyses are also facilitated by the fact that data on various aspects of theoretical interest is available, and also, by the high degree of continuity of electoral rules. This enables us to study 60 years of an almost unaltered electoral system in its impact on parliamentary behaviour, campaign strategies, re-election probabilities or on descriptive representation, amongst others.
As mentioned, the renewed interest in the German electoral system has not been only academic. Its apparent advantages – fair representation, i.e. a high degree of proportionality, a moderate effective number of parties (or a reasonable degree of party fragmentation), elections that regularly lead to stable coalitions, responsiveness of the party system to the representation of new societal interests due to moderate entry-thresholds (see the entry of the Green party in 1980 and of the PDS/ die Linke [excommunists] after 1990; cf. Capoccia 2002), a combination of local and national interest representation, an element of 'personalisation' – have also qualified it to be a point of reference in electoral reform debates all over the world, from Japan, New Zealand, Great Britain, Italy, Canada, Scotland and Wales to Venezuela and Angola (cf. Shugart 2001b, 2001a; Massicotte forthcoming, chapter 2: 15–16; Moser and Scheiner 2012: 5). In Germany itself, the electoral system 'has been generally accepted since the early 1970s and carries a significant degree of legitimacy' (Saalfeld 2005: 226). It is generally considered to be 'tried and tested' (Nohlen 2014: 394).
This all, also, explains the increased need for an encompassing, up-to-date treatment of Germany's mixed-member electoral system in English. Yet, it does not exist. One important contribution to fill this lacuna will be Louis Massicotte's systematic and comprehensive study of mixed electoral systems (Massicotte forthcoming), in which German federal and state elections are covered in minute detail. Apart from this important and timely study, the last monographic study on Germany's electoral system in English has been published more than twenty years ago (Jesse 1990). Yet, within these twenty years – not least due to German unification – electoral rules and electoral behaviour have changed in multiple (small) ways, and subsequently, the functioning logic of Germany's mixed system appears to have changed (quite profoundly) as well. Amongst others, we witness the following changes since unification:
a higher number of effective parties (due to the entrance of the ex-communists),
smaller districts and states with fewer districts in the East,
lower turnout, higher voter volatility and less party-identification in East and West, and subsequently
an increase in the disproportionality of the plurality tier, which led to a steeply increasing number of surplus seats (see, in particular, Chapters Two and Three).
These changes brought some features of Germany's mixed electoral system to the fore that had previously been attributed to the electoral rules as such, but were rather 'contextual' as now becomes apparent. But the effects of an electoral system – as Bingham Powell reminds us – depend on the interaction of '(t) he decisions of parties as they offer candidates in the election; the decision of voters as they vote; [and] the rules that aggregate the citizen partisan choices to determine the winning representatives' (Powell 2000: 23). With changes in voters' and parties' strategies, we have recently observed changing effects of the electoral rules, too. These changes stress the necessity to re-assess an electoral system with such an exemplary status and to make another effort at uncovering its inner working logic. This is the aspiration of this book.
But promoting a better understanding of the German system is not the only, and not even the major, objective of the following chapters, it is rather the means to another end. The prime objective of this book is to address a number of current research controversies and debates in electoral and legislative studies – using the German case as an exemplary case that lends itself very well to their discussion. Therefore, the following chapters clearly reach beyond the German case as such. In particular, as I want to show in the following chapters, studying mixed systems promises to contribute substantially to at least five debates in political science: (1) the contamination literature, (2) the mandate-divide debate, (3) the literature on descriptive representation, (4) the debate on the cartel party and on decreasing democratic accountability and (5) pre-electoral coalition formation. I will briefly sketch these literatures and show how they relate to my study.
(1) Electoral Contamination: Recent literature has directed attention towards the possible contamination effects between the tiers of a mixed electoral system (Herron and Nishikawa 2001; Ferrara, Herron, and Nishikawa 2005b; Cox and Schoppa 2002). One central indicator that is said to indicate contamination is the number of parties competing in a district for the plurality vote. In a mixed member electoral system, this number does not follow Duverger's Law, that is, it does not converge into two (parties or candidates), but remains systematically higher. This effect, however, is debated (cf. Moser and Scheiner 2012; Chapter Two), and seems to depend, at least partly, on voters' information about likely election outcomes, i.e. on the stability of the party system (Herrmann 2012). The main contamination argument holds that more than two parties usually contest a district because smaller parties run constituency candidates in actually hopeless races with the expectation that candidates will positively influence small parties' list vote share (Hainmüller, Kern, and Bechtel 2006; Hainmüller and Kern 2008). However, as of yet, the literature has largely failed to demonstrate that running for a seat in a district indeed has a positive impact on the PR-votes gained in that district – at least for those parties that should be the most affected by the majoritarian rules in the district: small ones. This has given rise to the counter argument that causality is in fact reversed: it is not candidates entering district races that boost the PR-vote of their party, but where a party can count on stronger voter support it will also nominate district candidates. Below, I will present empirical evidence that a positive impact from the plurality towards the proportional tier and from the candidacy to the improved vote share (but not vice-versa) does in fact exist. This will also allow us to identify a kind of 'second order' contamination effect, that, as of yet, has not been noted in the literature. The higher the number of parties contesting a district, the stronger the disproportionality with which nominal votes are translated into district seats in the plurality tier. This increase in disproportionality, in turn, has feedback effects on the functioning of the PR tier, which will be studied in more detail in Chapters Two–Four, and on candidates' strategies, which will be studied in more detail in Chapters Six–Seven.
(2) Mandate-Divide: With respect to the impact that electoral rules have on MPs' strategies and behaviour, the literature offers diametrically opposed views. Some claim that mixed electoral rules will induce mixed representation roles, while others assert that legislative behaviour depends on the way the members of parliament were elected. They thus identify two distinct legislator types in the German parliament and claim that list- and district-MPs show distinct parliamentary behaviour, that they employ different campaign strategies, and that they are differently willing to toe the party line (see Stratman and Baur 2002; Bawn 1999; Lancaster and Patterson 1990; Bawn and Thies 2003; Zittel and Gschwend 2007; Sieberer 2010). This is flatly denied by others: 'Contrary to widespread opinion, it is of absolutely no importance whether a mandate is obtained through the constituency or the Landesliste' (Burkett and Padgett 1987: 130; cf. Jesse 1988: 120). The controversy points to a larger literature evaluating the incentives that electoral rules provide for the representation of constituency- vs.-party- or lobbying-group-interests (Bawn and Thies 2003; Andeweg and Thomassen 2005) or for MPs' legislative activities (Herron 2002). Because it renders both legislator types, Germany's electoral system seems particularly suited for looking closer into the arguments on either side, to assess empirically their relative explanatory power and to provide us with a more precise picture of the nexus between electoral rules and representative roles (see also Bailer et al. 2013; see Chapter Nine).
Excerpted from Mixed Rules, Mixed Strategies by Philip Manow. Copyright © 2015 Philip Manow. Excerpted by permission of ECPR Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
List of Figures and Tables vii
List of Abbreviations xv
Chapter One – Introduction 1
PART I – PARTIES IN GERMANY’S MIXED-ELECTORAL SYSTEM 19
Chapter Two – The Contaminated Cube Rule 21
Chapter Three – Of the Cannibalistic Consequences 43
Chapter Four – Contesting Districts: Why small parties enter (hopeless) district races 61
Chapter Five – Coordinating: Locally or nationally? Or what if German parties suddenly behaved in a Duvergerian fashion? 87
PART II – CANDIDATES IN GERMANY’S MIXED
ELECTORAL SYSTEM 105
Chapter Six – Candidatures: Turning candidates into MPs 107
Chapter Seven – Combining Candidatures: Hedging against electoral uncertainty 127
PART III – MPS IN GERMANY’S MIXED ELECTORAL SYSTEM 143
Chapter Eight – Careers: Electoral rules and legislative turnover 145
Chapter Nine – Characteristics: Electoral pathways to the
Bundestag and MPs’ parliamentary strategies 159
Chapter Ten – Conclusion 181
Appendix to Chapter Two 189