The German Party System(s) in 2005: A Return to Volkspartei Dominance

Abstract

The article assesses the socio-structural underpinnings and systemic dynamics of the contemporary German party system and identifies four phenomena: an increased level of fractionalisation that has made it more difficult for small parties to assume the ‘kingmaker’ or ‘pivot’ role; the continued strengthening of a two-bloc dynamic; the emergence and persistence of the new territorial cleavage in the united Germany; and, a skew in the party system to the left that has shifted the position of the median legislator. It is argued that all these changes have served to re-assert the dominance of the two Volksparteien and have been particularly advantageous to the SPD. The article concludes by arguing that the outcome of the 2005 federal election can thus be seen as very much in keeping with these trends.

The decision by Gerhard Schröder, on the evening of Sunday 22 May 2005, to call an early federal election short-circuited an electoral cycle that was due to end in September the following year. The immediate catalyst for Schröder's decision was the SPD's anticipated – but no less shocking – defeat in that Sunday's Land elections in their heartland state of North Rhine Westphalia. There was some debate as to the rationale behind Schröder's decision (and, indeed some questioned its legality, leading to an unsuccessful challenge in the Federal Constitutional Court). For some observers, Schröder's intention was to use a highly partisan campaign (Lagerwahlkampf) in order to re-establish discipline within the SPD and head off a rebellion by the left of the party's parliamentary Fraktion. Others argued that an early election was a means of preventing the emergence of a significant competitor on the SPD's left flank by denying the PDS and the newly formed ‘Electoral Initiative for Social Justice’, or WASG, the time to merge and form a genuinely ‘national’ party. And there were also those commentators who argued that Schröder's decision had nothing to do with such a rational calculus and that the chancellor either had simply had enough and was effectively throwing in the towel, or alternatively was staking all on one last throw of the electoral dice.

The ultimate outcome of Schröder's decision is dealt with elsewhere in this volume and it is beyond the scope of this article to assess the strategic thinking behind the calling of early federal elections. This article intends to map out the strategic environment on the eve of the federal elections of 2005. What this article will demonstrate, however, is that – despite appearances to the contrary in May 2005 – the underlying strategic environment within which the SPD was operating was quite benign. It is true that, at the start of the federal election campaign, the SPD appeared to be in disarray and Schröder's CDU challenger Angela Merkel enjoyed a commanding lead in the opinion polls. Nevertheless it will be argued that the longer term structural attributes of the German party system(s) worked in favour of both Volksparteien – and in particular to the advantage of the SPD.

The rest of the article is structured as follows. First, the article examines the underlying socio-structural drivers of the contemporary party system in an historical context. Second, the article charts the changing constitution of the German party system(s), with an emphasis on the impact of the changes of the last 25 years including Unification. Third, the article examines in some depth the strategic environment generated by party system change and assesses its potential impact on the fortunes of the political parties. Fourth, the article pursues the logic of the previous section and looks at the impact of the strategic environment on real world outcomes at both the federal and state levels. Finally, the article summarises the data and arguments made and what it tells us about the German party system(s) in September 2005.

THE SOCIAL BASE OF PARTY COMPETITION

Along with electoral systems and laws, political cleavages structure political competition and as a result shape party systems. Nevertheless, scholarly opinion is divided as to how these cleavages arise in the first place. Political sociologists have argued that political cleavages are underpinned by more embedded social cleavages,1 whilst other accounts argue that cleavages are less embedded, more contingent in nature, and thus more dependant on political agency.2 This article works from the assumption that political agency does matter and that political conflict is contingent on decisions taken by elites. However, it is also clear that these lines of conflict cannot be drawn down from the ether and that it is prudent to ground any account of the German party system(s) in its social context before focusing in more detail on contemporary developments and the current strategic environment in which political parties operate.

The notion of social cleavages is to be found in a quite diverse and contested literature.3 However, Lipset and Rokkan4 provide the seminal account of how cleavages shape party systems and party competition in advanced West European democracies. Lipset and Rokkan's comparative schema works from the premise that similarities between countries are grounded in a common experience of mass politics, universal suffrage, secret ballots, and open competition for votes by parties mobilising around cleavages. However, Lipset and Rokkan also work from the assumption that West European polities diverge in quite significant ways because of nation-specific variables such as representation criteria, election rules, and patterns of cleavages. Crucially for the purposes of this chapter, Lipset and Rokkan assume that these specific national characteristics persist and remain manifest in the warp and weft of modern party-political competition.

The Lipset-Rokkan model assumes that party competition in Europe is driven primarily by four types of cleavage conflict – between the centre and periphery, between church and state or between churches, between urban and rural interests, and between social classes. In Germany, as in many Western European countries, the dominant conflicts that have persisted over time are those of class and religion. The class cleavage is a ‘horizontal’ cleavage, and as such is easier to pacify than ‘vertical’ cleavages such as those associated with the confessional conflict.5 Thus in Germany – as elsewhere – post-war prosperity and the development of welfare capitalism has taken the sting out of class conflict to the extent that social class alone is no longer a reliable indicator of voter preferences.6 It has also been recognised that voters are affected by more than one cleavage and the typical example of such ‘cross-cutting’ cleavages is that of class and religion, buttressed by auxiliary structures such as churches and trade unions. In pre-war Germany, cross-cutting class and confessional cleavages undermined the stability of the party system,7 but the post-war Federal Republic is closer to the European norm in that the two crosscutting cleavages generate stability and prevent one dominant cleavage from splitting society along religious or class lines.8 Thus, the German party system in 2005 was underpinned by a reasonably strong confessional cleavage and a weaker class cleavage that is conditioned by either trade union membership or confessional affiliation.9

Other social variables have also had an impact upon the modern German party system, however. In the 1960s, Lipset and Rokkan argued that European party systems were ‘frozen’ at the introduction of universal male suffrage and thus reflected the pattern of cleavages that were manifest at that point.10 However, in Germany – as in most other European democracies – the cumulative effect of embourgoisement and social mobility eroded traditional loyalties and by the 1970s created a political opportunity structure11 for a ‘new politics’, de-coupled from the old cleavage structures. The way these changes impacted upon European polities varied from country to country, depending on local conditions.12 In Germany, the combined effects of electoral dealignment13 and post-materialist value orientations14 associated with the new politics were to introduce an element of unpredictability previously alien to the cosy political consensus of the post-war Federal Republic. And of course this element of unpredictability was further aggravated by Unification in 1990, which ‘injected new uncertainty into this already fluid political environment’.15 The cumulative impact of the social changes discussed above on (1) the constitution of the party system(s), (2) the strategic environment they generated, and (3) the patterns of party political competition and co-operation that arose from them, was profound and far reaching. It is to these themes that the article now turns.

THE CONSTITUTION OF THE GERMAN PARTY SYSTEM(S)

The social developments described in the pervious section are reflected in the development of the German party system(s) over the post-war period. Table 1 demonstrates how the German party system at the national level has developed over the period 1949–2002. The table demonstrates, first, a 30-year period of ongoing party system concentration, in which the party system is dominated by the two big Volksparteien, and, second, two systemic junctures which have served to break down this dominance and shift the centre of gravity within the party system towards the political left. The first of these junctures takes place in 1983, with the entry of the Greens, and the second takes place in 1990, following Unification, with the entry of the PDS.

Table 1 THE DEVELOPMENT OF THE GERMAN PARTY SYSTEM: FEDERAL ELECTIONS, 1949–2002

Much of the concentration of the German party system(s) over the period 1949–83 can be attributed to the social changes noted in the previous section. The decline in the salience of the confessional and, in particular, the class cleavage served to reduce polarisation within the party system by eroding the basis of electoral support for flanking parties of the right and left. However, political agency also played a part and over the period the two Volksparteien were reasonably successful in actively preventing the emergence of flanking parties. Of the two, the CDU/CSU was the most successful and pursued a conscious policy of integrating the political right. It did this by both absorbing smaller flanking parties and at the same time defending the rights of ethnic Germans expelled from the countries of Central and Eastern Europe. As a result, since the early 1960s, no party to the right of the CDU/CSU has managed to garner enough votes to scale the 5 per cent hurdle to Bundestag representation.

Up until the 1980s, the SPD had also enjoyed success in this regard. The party successfully survived the split that followed the Zwangsvereinigung and, by 1953, had successfully neutralised the KPD in West Germany.16 However, as Table 1 demonstrates, the German party system at the national level was reconstituted through two crucial historical junctures (1) in 1983, with the entry into the Bundestag of the environmentalist Greens; and (2) in 1990, when the PDS – the successors to the ruling party in the German Democratic Republic – survived the collapse of the GDR and established itself in the new all-German Bundestag.

By 2005, two strong tendencies could be identified within the German party system(s). First, the legacy of Unification means that German party politics was now subject to a strong territorial cleavage. In most of the old federal states there was what was effectively a two-bloc system, made up of four parties (the Greens, SPD, FDP, and CDU/CSU), arranged along a one-dimensional left-right continuum. By contrast, in most of the new federal states of the former German Democratic Republic there was a three-party system made up of the PDS, SPD, and CDU. Taken together, this means that the ‘national’ party system was made up of five parties: the PDS, Greens, SPD, FDP, and CDU/CSU. Second, this five-party system now had a wider ideological range, within which the ideological centre of gravity had shifted leftwards from where it had been for most of the post-war period.17 Taken in the round, the changes described in this section had a profound affect on the strategic environment in which political parties operated and the patterns of competition and co-operation that this generated. It is to the strategic environment in which political parties operate that the article now turns.

THE STRATEGIC ENVIRONMENT

The development and constitution of the German party system(s) has led to the creation of a nation specific set of constraints and incentives that provide the environment within which party elites pursue strategic action. One of the key strategic considerations is the degree of stability within a given party system – a factor that is directly related to the level of fractionalisation within it.

The crudest measure of fractionalisation describes the total number of parties that exist within a party system. In theoretical terms, increased fractionalisation increases in turn the likelihood of ‘cycling’, by which any potentially winning actor or coalition can be trumped by another alternative.18 Increased fractionalisation often goes with higher levels of party system polarisation and, again, in theoretical terms, the more polarised the party system the less likely that a winning solution can be found.19 In West European democracies we have seen an increase in the degree of fractionalisation of party systems since 1980 and the average West European parliament now contains seven political parties. However, a more nuanced measure of fractionalisation is a count of the number of ‘effective’ parties within a party system (defined as parties that can be expected to have an impact upon the outcome of a legislative ‘game’).20 And using this measure the average West European party system only contains four ‘effective’ parties.21

In practical terms, then, the average West European party system remains relatively stable and this stability is buttressed to a greater or lesser extent by nation-specific system attributes such as voting rules (plurality versus proportional systems), barriers to representation (such as the five per cent hurdle at the federal level in Germany), laws regulating internal party democracy, and so on. These institutional variables impose what theorists call a ‘structure induced equilibrium’,22 the dynamics of which vary from country to country.

Writing at around the same time as Shepsle, Gordon Smith23 highlighted the importance of system attributes, both as means of restraining party system fragmentation and also as means of imposing the structure induced equilibria, noted above, upon them. For Smith, the key system attribute is ‘cohesion’, a quality which, if present in sufficient quantities, will enhance system stability. There are two types of cohesion, both of which impact upon parties strategic considerations. The first, ‘governing’ cohesion, determines the effectiveness of government in terms of its longevity, stability, and steering capacity. The second, ‘social’ cohesion, goes some way to determine the extent to which political parties are able effectively to integrate and aggregate competing societal interests and as a result satisfy voters' preferences and shut out potential competitors. For Smith, the UK-style plurality system – designed to produce majoritarian single-party government – is the best suited to deliver both strong, stable government and an integrated electorate. Nevertheless, Smith also held the 1970s German party system in high regard in this respect, dominated as it was by two ‘balanced clusters’ of the SPD and FDP in coalition (which lasted from 1969 to 1982), balanced by the CDU/CSU in opposition.24 It will be recalled from Table 1 that during the period when Smith was writing there were no flanking parties to either the left or right of the two Volksparteien. Between them the liberal FDP maintained a pivotal position in what Pappi famously described as a ‘triangular’ party system.25

The ‘Pappi model’ of a triangular party system presented a particular set of incentives and constraints to the three parties in the German party system of the period. The traditional left–right dimension of competition remained important but in a relatively consensual party system the parties developed a modus operandi in which all three were able to cooperate with each other along one of three issue dimensions. This presented the players with three different solutions. First, as during the Grand Coalition of 1966–69, the SPD and CDU/CSU were able to co-operate along the corporatist dimension of the triangle. Second, as during the period 1969–82, the SPD and FDP could co-operate along the ‘social liberal’ dimension. Finally, as had been the case in the period up until 1966 and again from 1982–98, the CDU/CSU and FDP could co-operate along the dimension of ‘bourgeois issues’ such as economic growth and prosperity. Crucially, however, it was the ability of any of the three parties to either exit an existing coalition or enter a new coalition that instilled discipline and gave the Pappi model its consensual dynamic. This was particularly important to the FDP, whose raison d'etre in the absence of significant and socially grounded electoral support was to act as the ‘kingmaker’ or pivot party within the system.

After 1983, however, the internal dynamics of the German party system began to change under the pressure of greater party system fragmentation and polarisation. Again, it will be recalled from Table 1 that 1983 and 1990 saw two critical junctures, in which the German party system fragmented, polarised, and skewed to the political left. The emergence of the Greens and then the PDS has turned the German party system into a genuinely multi-party system in which it is possible and even rational for smaller parties such as themselves to seek out niche positions on the dominant left–right continuum and mobilise around more peripheral segments of what is normally a bell-curve of electoral preferences. In addition, as will be come apparent below, the new party system increasingly failed to function according to the consensual triangular dynamic of the past and now more strongly resembled a relatively polarised ‘two-bloc’ system, primarily but not exclusively based on the left–right ideological dimension. The shape of the new party system is set out in the Figure below.

The Figure demonstrates a party system divided into two blocs along the left–right ideological dimension, but also cross-cut by a libertarian–authoritarian dimension made salient by the emergence of post-materialism, noted earlier. In addition the figure identifies three issue dimensions along which political parties might forge common policy positions. Of these three issue dimensions, the corporatist issue dimension remains from the old Pappi model. In 2005, this thus provided the potential for the formation of a Grand Coalition along the lines of 1966–69. However, the sublimation of the FDP into the bourgeois bloc over the previous 20 years effectively neutralised the old ‘social liberal’ issue dimension for the time being. This was confirmed by the FDP's ruling out of the possibility of a coalition with the SPD before the federal election.26 In its place, however, there was a possibility of co-operation between the CDU and FDP along the bourgeois issue dimension or a continuation of co-operation along the ‘new politics’ dimension if the incumbent Red-Green coalition was re-elected.

Figure 1 THE STRATEGIC ENVIRONMENT IN THE 2005 GERMAN PARTY SYSTEM

NOTES: PARTY POSITIONS ARE APPROXIMATE AND DERIVED FROM JUDGEMENTAL DATA RATHER THAN CONTENT ANALYSIS OF MANIFESTOS.

In 2005, therefore, the strategic environment described in the Figure presented both the CDU/CSU and SPD with a strategic dilemma. On the one hand, as ‘Catch-all’ parties they had no choice but to compete for the median voter by underplaying their ideological side and appealing to the centre of the bell-curve of voter preferences along the dominant left–right ideological dimension. On the other, the mathematics of coalition formation meant that unless they were to enter a Grand Coalition they more than likely had to rely on flanking parties to the left or right in order to form a government, which meant they also had to appeal in that direction. This dilemma has been particularly problematic for the SPD, which has had to cope with the emergence of two quite different competitors on its left flank. The emergence of the Greens, following the first critical juncture of 1983, presented the SPD with a particular set of problems as to how best to compete with a new party whose support was (1) mainly drawn from younger and/or well-educated voters, and which in ideological terms was (2) to the left along the left–right ideological dimension and also (3) more libertarian in outlook. This strategic problem was aggravated by the arrival of the PDS following the second juncture of 1990. Given the PDS's position as the successor party to the ruling party of the German Democratic Republic, its emergence meant that the SPD now also had to compete with a party that (1) drew most of its support from a different social milieu to that of the Greens and was (2) significantly to its left along the left–right ideological dimension but (3) by 2005 had come to occupy roughly the same position along the libertarian–authoritarian dimension.

To sum up therefore, the two Volksparteien have tried to prevent the emergence of parties on their political flanks. The CDU/CSU has been successful in this regard, whilst the SPD has had to compete with at first one and then two flanking parties – albeit competing along different ideological dimensions. For smaller parties, the presence of the two big Volksparteien within the party system makes it rational for them to try and achieve a position where they hold the balance of power between the two Volksparteien, as the FDP managed to achieve up until 1983. However, in the 2005 German party system such a pivotal position appeared beyond the reach of the FDP, Greens or PDS.

One of the reasons why none of the smaller parties have managed to re-create the pivotal position enjoyed by the FDP before 1983 is that the increased level of fractionalisation within the party system means that the electoral arithmetic simply does not add up in the way that it did in the past. But once again we return to the issue of political agency – and in particular the relatively successful strategy by which the SPD has coped with the threat of the Greens and PDS. Much has been made in the past of the SPD's strategic dilemma.27 However, it has become apparent in recent years that the SPD has managed successfully to adapt to both these new electoral competitors, mainly because the bases of support for the two parties are territorially distinct. In 2005, the PDS continued to garner significant support in the new federal states but had conspicuously failed to make an impact in the old federal states. By contrast, the Greens' initial respectable level of support had dissipated in most of the new federal states and it remained by-and-large a party of the affluent west. Because of this territorial distinctiveness the SPD has rarely been forced to compete with both parties simultaneously but rather has been able to develop different strategies of competition and cooperation – depending on which part of the Federal Republic it is campaigning in – and thus at the same time retain a reasonably credible policy mix. And in 2005, the SPD's strategy appeared to have worked very well. In the new federal states, the PDS had lost support to the SPD in the 2002 Bundestag elections and failed to scale the 5 per cent hurdle. As a result the PDS' share of seats fell from 36 in the 1998–2002 Bundestag to just two directly elected seats after the 2002 elections. And while the SPD had not managed to defeat the Greens in electoral terms, through a strategy of sustained co-operation it had managed to pull the party into its political orbit.28 As a result the SPD and Greens had governed in a formal federal-level coalition since 1998, following over a decade of such coalitions at the Land level.29 At the same time, the SPD had been able to cooperate with both the FDP and CDU at the Land level and hold out the possibility of similar arrangements at the federal level should the Red-Green coalition break down. These patterns of competition and co-operation are discussed in greater depth in the next section.

PATTERNS OF COMPETITION AND CO-OPERATION

As noted earlier in this article, in 2005 the German party system displayed a strong territorial cleavage and a wider ideological range, within which the ideological centre of gravity has skewed leftwards. These changes had two main effects on the relative strengths of the parties. First, as discussed earlier, the emergence of a less fluid ‘two-bloc’ system made it harder for any of the smaller parties to assume the pivotal position enjoyed by the FDP before 1983. It appears that this development has enhanced the relative dominance of the Volksparteien, although it will require one or two more Bundestag elections to be sure that this is the case.30 Second, the skew towards the left has led to the SPD – rather than the FDP – increasingly being the party with the median legislator along the left–right continuum within the legislature. In theoretical terms, this makes the SPD the Mparty, without which no ideologically connected majority coalition can form and which should make it decisive in any coalition game.31 The shift of the median legislator from FDP to SPD is demonstrated in Table 2.

Table 2 BUNDESTAG ELECTIONS 1983–2002: MPARTIES AND COALITION OUTCOMES

	Date of Bundestag Election
	Seats in Bundestag
	6 Mar. 83	25 Jan. 87	2 Dec. 90	16 Oct. 94	27 Sep. 98	22 Sep. 02
CDU/CSU	244	223	319	294	245	248
FDP	34	46	79	47	44	47
SPD	193	186	239	252	298	251
Greens	27	42	8	49	47	55
PDS	–	–	17	30	35	02
Total seats	498	497	662	672	669	603
Minimum Winner	250	249	332	337	335	302
Mparty	FDP	FDP	FDP	FDP	SPD	SPD
MpartyK?	No	No	No	No	Yes	Yes
Coalition	CDU/CSU- FDP	CDU/CSU- FDP	CDU/CSU- FDP	CDU/CSU- FDP	SPD-Greens	SPD-Greens
Degree of change	None	None	None	None	Total	None
Source: Forschungsgruppe Wahlen, 1983, 1987, 1990, 1994, 1998, 2002.

In addition to tracking the shift of Mparty from the FDP to the FDP, Table 2 demonstrates two other points. The first, relatively self-evident, point is that – in keeping with the theoretical predictions noted above – throughout the run of data the Mparty always becomes a member of the winning coalition. The second, more significant, observation is that not only did the SPD become Mparty after the 1998 elections; the outcomes of the 1998 and 2002 coalition negotiations also made it MpartyK. The term MpartyK refers to a given political party that ‘owns’ not only the median legislator within the legislature but also the median legislator within the winning coalition. In real world coalition outcomes, the position of the median legislator and the median legislatorK are often different and it would be expected that ceteris paribus subsequent coalition policy positions would fall within the interstices of the preference curves of the median legislator and median legislatorK.32 The important thing to remember, however, is that any political party which is both Mparty and MpartyK is doubly decisive within the coalition game, in that (1) it should be in all majority ideologically connected coalitions and (2) subsequent policy outcomes should be close to its ideal position. Because of its size and the ideological range of the coalitions in which it took part, the position of MpartyK was something the FDP was never able to achieve.

Thus, the SPD as Mparty and MpartyK would be expected to enjoy a degree of political leverage within the winning coalition greater than that expected from its relative size. And on the basis of what we know of the 1998–2005 Red-Green coalition, this was the case.33 Despite effectively saving the ruling coalition in the 2002 federal election through its better than expected electoral performance, the Greens never enjoyed the kind of influence enjoyed by the FDP in previous coalitions.34 Indeed, with the exception of the Foreign Ministry (which under normal circumstances is always allocated to the junior coalition party in Germany) the Greens singularly failed to secure any blue-chip portfolios in either 1998 or 2002.35 This compares badly with the range of portfolios secured by the FDP in previous coalitions.

Thus we can observe a new two-block party system at the federal level. But what was the picture at the Land level? The pattern of party political competition in Land parliaments as of 1 September 2005 is set out in Table 3, below.

Table 3 PARTY SYSTEMS IN THE GERMAN LÄNDER PARLIAMENTS AS OF 1 SEPTEMBER 2005

Taken in the round, in Table 3 there were a total of eight different parties in the various Land legislatures, ranging from the PDS on the left of the party system to the DVU on the right.36 But, of these eight parties only the CDU and SPD were present in all 16 Land parliaments; thus confirming the relative dominance of the two Volksparteien. 37 Table 3 also demonstrates the clear territorial cleavage noted earlier. Thus we find that there were both ‘type 1’ and ‘type 2’ party systems at the Land level.38 In the old federal states we find type 1 systems, made up of the Greens, SPD, FDP, and CDU or CSU. In 2005 such systems existed in Baden-Württemberg, Hesse, Lower Saxony, North Rhine Westphalia, Rhineland Palatinate, Saarland, and Schleswig-Holstein. There were also variants of the type 1 system in Bremen and Bavaria (where the FDP was excluded), and Hamburg (which also included the PRO, the ex-Schill party). Type 2 systems, although made up of three parties (the PDS, SPD, and CDU) lacked the triangular dynamic of the old Pappi model found in West Germany in the 1960s and 1970s, and in 2005 were found in Mecklenburg-West Pomerania and Thuringia. There were also two variants, to be found in Brandenburg (which also included the far-right DVU) and Saxony-Anhalt (which included the FDP). Very interestingly, we find two instances of the development of what might be called a ‘Type 3’ system which encompasses all of the parties found in the ‘national’ system. Thus, in Berlin there was a hybrid system – made up of a type 1 sub-system in the old West Berlin and a type 2 sub-system in that part of the city that used to be part of the German Democratic Republic – and in Saxony there was a party system that included the PDS, Greens, SPD, FDP, CDU and far-right NPD. The Berlin and Saxony party systems also encapsulated the other two phenomena discussed earlier, namely higher levels of party system fractionalisation and a greater ideological range within the system. Saxony in particular displayed a significant level of fractionalisation and had the largest ideological range of any of the Land party systems.

CONCLUSIONS

So what do the arguments and data discussed in this article demonstrate? In many ways, the picture that emerges is confusing and potentially contradictory. On the one hand, we find higher levels of party system fractionalisation than was the case in the 1950s, 60s, and 70s. At first glance, such party system deconcentration should work against the interests of the two big Volksparteien. Indeed, in the 1980s and early 1990s, the SPD did struggle to adjust to the presence of first one and then two competitors on the left of the party system. For its part, however, the CDU/CSU has managed to prevent the emergence of a sustained challenge on its right flank and thus avoided the strategic difficulties encountered by the SPD.

This article has argued, however, that a closer analysis of party system change in Germany reveals that the strategic environment has actually become more benign for the Volksparteien and in particular the SPD. There are four main reasons for this. First, the degree of fractionalisation that has taken place has served to make it more difficult for a single small party to assume the ‘kingmaker’ or ‘pivot’ role enjoyed by the FDP in the old triangular system. Second, the loss of the kingmaker has both enhanced – and been enhanced by – the emergence of a two-bloc dynamic, in which one of the two Volksparteien almost inevitably dominates each bloc. Third, the emergence and persistence of the new territorial cleavage in the united Germany means that the SPD has managed to forge a modus operandi that allows for competition and/or co-operation with the Greens and PDS in territorially discrete arenas. This has served to offset the problem generated by the need to compete and co-operate with the two parties along distinct ideological and issue dimensions and has helped the SPD maintain a reasonably credible policy mix. Finally, the skew in the party system to the left has shifted the position of the median legislator to the SPD's advantage, to the extent that in the 1998 and 2002 federal elections it was not only Mparty but also MpartyK. As discussed, in theoretical terms this makes the SPD decisive in any coalition game that requires an ideologically connected majority outcome.

So where did that leave the political parties at the start of the final weeks of campaigning in September 2005? We now know the outcome of the federal election and this is dealt with elsewhere in the volume. What we can say here is that the arguments made in this article indicate that the formation of a Grand Coalition at the end of 2005 was not inevitable but – I would argue – was more likely than was perhaps recognised at the time.39 And despite the electoral strength of the FDP, the party system dynamics described in this article were to play their part in consigning them once again to the opposition benches. Thus, although voters did punish the outgoing Red-Green coalition, it was the Greens that bore the brunt of public anger. Moreover, the expected challenge from the PDS/Party of the Left was not sufficient to make them serious players in the post-election coalition negotiations.40 For their part, the SPD were to remain in power and – as junior partner to the CDU/CSU – continue to play a potentially decisive role in the government of the Federal Republic. In short, the outcome of the 2005 federal election was very much in keeping with what this article argues has taken place – namely a return to Volkspartei dominance.

Notes

1. See S. Lipset and S. Rokkan, ‘Cleavage Structures, Party Systems, and Voter Alignments: An Introduction’, in S. Lipset and S. Rokkan (eds.) Party Systems and Voter Alignments: Cross-National Perspectives (New York: Free Press, 1967); R. Rose and D. Urwin, ‘Social Cohesion, Political Parties and Strains in Regimes’, Comparative Political Studies 2 (1969), pp.7–67; R. Rose and D. Urwin, ‘Persistence and Change in Western Party Systems since 1945’, Political Studies 18 (1970), pp.287–319; E. Allardt and S. Rokkan, Mass Politics (New York: Free Press, 1970); R. Rose, Electoral Behaviour: A Comparative Handbook (New York: Free Press, 1974).

2. G. Loewenberg, ‘The Making of the German Party System: Political and Socio-Economic Factors’, Polity 1 (1968), pp.86–113; P. Converse and H. Valen, ‘Dimensions of Cleavage and Perceived Party Distance in Norwegian Voting’, Scandinavian Political Studies 6 (1971), pp.107–52; S. Berglund and U. Lindstrom, ‘The Scandinavian Party System(s) in Transition(?): A Macro-level Analysis’, European Journal of Political Research 7 (1979), pp.187–204.

3. See P. Dunleavy and C. Husbands, British Democracy at the Crossroads (London: Allen & Unwin, 1985); A. Zuckerman, ‘Political Cleavage: A Conceptual and Theoretical Analysis’, British Journal of Political Science 5 (1975), pp.231–48; P. Pesonen, ‘Dimensions of Political Cleavage in Multi-party Systems’, European Journal of Political Research 1 (1973), pp.109–32; D. Rae and M. Taylor, The Analysis of Political Cleavages (New Haven: Yale University Press, 1970); for instance, H. Daalder, ‘Parties, Elites, and Political Developments in Western Europe’, in J. LaPolambara and M. Weiner (eds.), Political Parties and Political Development (Princeton: Princeton University Press, 1966).

4. Lipset and Rokkan, ‘Party Systems and Voter Alignments’.

5. T. Landman, Issues and Methods in Comparative Politics: An Introduction (London/New York: Routledge, 2000), p.129.

6. Forschungsgruppe Wahlen, election data 1976–87.

7. W. Claggett et al., ‘Political Leadership and the Development of Political Cleavages: Imperial Germany, 1871–1912’, American Journal of Political Science 26/4 (1982), pp.643–63.

8. C. Lees, Party Politics in Germany: A Comparative Politics Approach (Basingstoke: Palgrave Macmillan, 2005).

9. Forschungsgruppe Wahlen, 2002; M. Dogan, ‘Class, Religion, Party. Triple Decline of Electoral Cleavages in Western Europe’, in L. Karvonen and S. Kuhnle (eds.) Party Systems and Voter Alignments Revisited (London: Routledge, 2001).

10. Lipset and Rokkan, ‘Party Systems and Voter Alignments’.

11. H. Kitschelt, The Logics of Party Formation (Ithaca, NY: Cornell University Press, 1989); H. Kitschelt, ‘Political Opportunity Structures and Political Protest: Anti-nuclear Movements in Four Democracies’, British Journal of Political Science 16/1 (1986); S.G. Tarrow: Power in Movement: Social Movements, Collective Action and Politics (Cambridge: Cambridge University Press, 1994).

12. R. Dalton et al., Electoral Change in Advanced Industrial Democracies (Princeton; Guildford: Princeton University Press, 1984).

13. S. Padgett (ed.), Parties And Party Systems in the New Germany (Aldershot: Dartmouth, 1993); P. Conradt et al. (eds.), Germany's New Politics: Parties and Issues in the 1990s (Oxford: Berghan, 1990); P. Converse and H. Valen, ‘Dimensions of Cleavage and Perceived Party Distance in Norwegian Voting’, Scandinavian Political Studies 6 (1971), pp.107–52; I. Crewe, ‘Introduction: Electoral Change in Western Democracies: A Framework for Analysis’, in I. Crewe and D. Denver (eds.), Electoral Change in Western Democracies: Patterns and Sources of Electoral Volatility (London: Croom Helm, 1985).

14. R. Ingelhart, Culture Shift in Advanced Industrial Society (Princeton: Princeton University Press, 1990); R. Ingelhart, The Silent Revolution: Changing Values and Political Styles among Western Publics (Princeton: Princeton University Press, 1977); S. Flanagan, ‘Changing Values in Advanced Industrial Society’, Comparative Political Studies 14 (1982), pp.403–44; see also R. Dalton, Citizen Politics: Public Opinion and Political parties in Advanced Industrial Democracies, 2^nd edn. (Chatham, N.J.: Chatham House, 1996); R. Dalton, and M. Kuechler (eds.), Challenging the Political Order: New Social and Political Movements in Western Democracies (Oxford: Oxford University Press, 1990).

15. Dalton et al., ‘Two German Electorates?’, in G. Smith, and W. Patterson (eds.), Developments in German Politics (Basingstoke: Macmillan, 1992), p.55.

16. Although already neutralised as an electoral force the KPD was banned by the Federal Constitutional Court in 1956.

17. Lees, Party Politics in Germany.

18. D. Black ‘On the rationale of group decision making’, Journal of Political Economy 56 (1948), pp.23–34; K.J. Arrow, Social Choice and Individual Values (New York: John Wiley & Sons, 1951).

19. R.G. Niemi, ‘Majority decision making with partial unidimensionality’, American Political Science Review 63 (1969), pp.488–97; G. Tullock, and C.D. Campbell, ‘Computer Simulation of a Small Voting System’, Economics Journal 80 (1970), pp.97–104; K. Koga and H. Nagatani, ‘Voter Antagonism and the Paradox of Voting’, Econometrica 42 (1974), pp.1045–67.

20. See M. Laakso and R. Taagepera, ‘“Effective” Number of Parties: A Measure with Application to West Europe’, Comparative Political Studies 12/1 (April 1979), pp.3–27; also C. Lees and A. Taylor, ‘Explaining the 2005 Coalition Formation Process in Germany – A Comparison of Power Index and Median Legislator Coalition Models’, Politics 26/3 (2006), pp.151–60.

21. J.-E. Lane and S. Ersson, Politics and Society in Western Europe, 4^th edn (London: Sage, 1999), pp.142–3.

22. K. Shepsle, ‘Institutional Arrangements and Equilibrium in Multidimensional Voting Models’, American Journal of Political Science 23 (1979), pp.27–60.

23. G. Smith, ‘Western European Party Systems: On the Trail of a Typology’, West European Politics 2/1 (1979).

24. Ibid., p.135.

25. F. Pappi, ‘The West German Party System’, West European Politics 7 (1984), pp.7–26.

26. Since the 2005 federal elections there has been a renewal of interest in the Social-Liberal model. The main driver for this has been the election of Kurt Beck as Chairman of the SPD in May 2006. As Minister President of the state of Rhineland-Palatinate, Beck has in the past ruled in coalition with the FDP. Given that Beck will almost certainly be the SPD's chancellor-candidate at the next federal elections, the prospect of a federal-level coalition involving the SPD and FDP is far more likely than it has been in the recent past.

27. See for instance C. Lees, The Red-Green Coalition in Germany: Politics, Personalities, and Power (Manchester: Manchester University Press, 2000).

28. Lees, Party Politics in Germany.

29. Lees, The Red-Green Coalition in Germany.

30. See G. Smith, ‘The ‘New Model’ Party System’, in S. Padgett, W. Paterson and G. Smith (eds.), Developments in German Politics 3 (Basingstoke: Palgrave, 2003), p.98.

31. Robert Axelrod developed the idea of ideological connectedness as a response to the predictive weaknesses of ‘pure’ office-seeking models of coalition formation. Axelrod's work was subsequently refined by, amongst many others, Abram de Swaan, who is associated with initial work on the median legislator/Mparty model. See R. Axelrod, Conflict of Interest (Chicago: Markham, 1970); A. de Swaan, Coalition Theories and Cabinet Formation (Amsterdam/Oxford, 1973: Elsevier). Another method of calculating the decisiveness of a political party in a given coalition game is through the use of power indices such as the standardised Banzhaf index. Such indices are contested in the literature (see Maclean et al., 2005, for instance).

32. A. de Swaan, Coalition Theories And Cabinet Formation (Amsterdam/Oxford: Elsevier, 1973).

33. See Lees, The Red-Green Coalition in Germany, Party Politics in Germany.

34. Ibid.

35. It might be argued that Green portfolios that oversee such key policy areas and large budgets as health or agriculture are not second-order ministries. However, it is generally agreed that first-order ministries are those of finance, foreign affairs, environment, defence, transport, and the interior. Please refer to Müller and Strøm, 2003, for a useful generic ranking of specific ministries in a comparative context.

36. This includes the Sűdschleswigsche Wählerverband (SSW), which exists to defend the interests of the territorially distinct Danish ethnic minority in Schleswig-Holstein. For the purposes of this article, however, it cannot be considered part of the ‘national’ party system.

37. Smith 2003.

38. Lees, Party Politics in Germany.

39. For a more developed argument in this vein, using power index analysis, see Lees and Taylor, ‘Explaining the 2005 Coalition Formation Process’, pp.151–60.

40. Ibid.