Intraparty Factions and Coalition Bargaining in Germany
Thomas Bräuninger and Marc Debus
Department of Politics and Management & Center for Junior Research Fellows, University of
Konstanz. Email: [email protected], [email protected]
This version: April 4, 2005
Paper presented at the ECPR Joint Session 2005, Granada,
Workshop 25 “Intraparty politics and its effect on government
formation and duration in parliamentary systems”
Abstract: Most formal theories of coalition formation are based on the assumption of parties
as unitary and forward-looking actors who bargain over portfolios and future governmental
policies in the government formation process (e.g. Laver/Shepsle 1996, Thies 2001). In this
paper we relax the unitary actor assumption and theorize about the conditions under which
intraparty factions play a role in coalition bargaining processes. Using a simple model we
argue that the outcomes of coalition formation processes (portfolio allocation, governmental
policy as stated in the coalition agreement) reflect the policy positions and intraparty strength
of party factions rather than the positions of parties as unitary actors. We evaluate the
argument in a case study of coalition formation in Germany in 1987/90 and 1998/2002. Party
manifestos, platforms of party working groups, and formal coalition agreements are used to
derive policy positions of all major parties (as unitary actors), their key factions, and future
governmental policies. Using the wordscore technique (Laver/Benoit/Garry 2003), we
estimate policy positions with respect to the three most important policy dimensions, the
socio-economic, the social, and the foreign affairs dimension. We find support for the
argument that the allocation of portfolios to members of party factions is used to enforce
coalition contracts when ministerial discretion is high.
Introduction
In most parliamentary democracies, single parties rarely obtain a majority of seats in
parliamentary elections so that attaining the control of the executive branch necessitates the,
mostly explicit, formation of a coalition of parties possessing a majority of seats in the
legislature. The study of the formation and termination of such coalitions is a major field of
interest in comparative politics. Almost all of these theories start with the assumption of
parties as unitary actors seeking for office or policy gains (Riker 1962; Leiserson 1968;
Axelrod 1970; De Swaan 1973; Dodd 1976; Peleg 1981; van Roozendaal 1992; Warwick
1994; Schofield 1995). Yet the conventional focus on the “partisan composition” of coalition
governments, whether as the explanandum (formation) or explanans (termination), has
somehow marginalized other issues like the questions as to what these collective actors intend
to do, what they expect to do, what they actually do, and how these government policies come
about (Strøm/Müller 1999). More recent studies have begun to study these issues in more
detail (Austin-Smith/Banks 1990; Laver/Shepsle 1996; Mitchell 1996; Timmermans 1998;
Müller/Strøm 2000a).
A major issue in these studies is the question of stability. In a wide range of settings,
majority rule does not produce stable outcomes (Arrow 1964; McKelvey 1986). Likewise,
government coalitions are susceptible to re-negotiations, their policy agreements are not selfenforcing per se, and “transferable utility” goods like office spoils provide room for making
and breaking governments. Given this, a number of stability-inducing mechanisms have been
identified, most notably, ministerial discretion (Austin-Smith/Banks 1990; Laver/Shepsle
1996), last offer authorities (Heller 2001), appointment of junior ministers (Thies 2001), or
parliamentary scrutiny (Martin/Vanberg 2004). In this paper we study a different mechanism:
the delegation of ministerial authority to members of intraparty factions. More specifically,
we relax the unitary actor assumption and theorize about the conditions under which
intraparty factions holding different preferences than the coalition core parties or their leaders
play a role in coalition bargaining processes.
We develop a simple model to show when and how the outcome of coalition formation
processes (portfolio allocation, the governmental policy program as stated in the coalition
policy agreement) reflect the policy positions and intraparty strength of party factions rather
than the positions of parties as unitary actors. We develop three simple arguments on how
intraparty factions can influence or, alternatively, can be used to secure coalition contracts
under the condition of ministerial discretion. We first contend that party factions hold
2
ministries if they make outcomes closer to the Pareto surface possible and parties are
interested in efficient outcomes. In a second setting, the (unweighted) bargaining outcome of
the governmental parties is assumed to be the negotiated coalition contract so that party
factions should hold ministries whenever they produce outcomes closer to the efficient (yet
unobserved) bargaining outcome. In a third setting, we test our theoretical considerations by
the use of a content analysis of coalition policy documents. We introduce the observed policy
program of the coalition and contend that portfolios are allocated to major intraparty groups in
such a way that the policy outcome under ministerial discretion is optimal with respect to the
coalition policy program.
The argument is evaluated in a case study of coalition formation in Germany in 1987,
1990, 1998, and 2002. We present data on the spatial positions of all political actors as well as
the coalition policy program in a three-dimensional policy space. Party manifestos, platforms
of party working groups, and formal coalition agreements are used to estimate the policy
positions of all major parties (as unitary actors), their key factions, and intended future
governmental policies. Using the wordscore technique (Laver/Benoit/Garry 2003), positions
are derived with respect to the three most important policy dimensions in Germany, the socioeconomic, the social, and the foreign affairs dimension. In what follows we first present our
argument on how intraparty politics might help us to explain the allocation of portfolios in
coalition governments. The third section introduces our case study of portfolio allocation in
Germany in the time period of 1987 to 2002. Section four evaluates the above argument on
the impact of intraparty politics on portfolio allocation in the light of these data. In the final
section, we draw conclusions from our analysis.
Delegating power to factions: Portfolio allocation with intraparty groups
In this section we develop our argument on how to explain the allocation of portfolios to
members of intraparty factions and on how intraparty politics can help to explain the
allocation of portfolios in coalition governments more generally. In a nutshell, our argument
is that in situations with more than one principal and a divergence of preferences, delegating
power to a third party agent with different interests may be Pareto-improving for the
principals. This holds beyond the notion that multiparty government creates a principal-agent
problem (Strøm/Müller 2001: 9f.).
To begin with, we note that coalition bargaining or government formation proceeds in
several steps and involves a number of decisions on the side of the actors involved. One is the
3
problem of who to ask to join a coalition, or, vice versa, whether or not to join once asked for.
A second problem is how the conflict of interest concerning future governmental policies
might be resolved within the coalition. Thirdly, a decision has to be taken how to delegate
authority for the execution and implementation of the agreed coalition policy program.
Obviously, all three questions are interrelated. Laver and Shepsle’s (1990; 1996) model of
portfolio allocation is the theoretically most sophisticated example of more recent research
that addresses all three questions in one setup.1
They presume that coalition cabinets are usually based on ministerial discretion. Cabinet
ministers have "considerable influence" over governmental decision-making in their area of
jurisdiction: "health policy is heavily conditioned by the partisan political agenda of the
minister of health, defense policy by the political views of the political party of the minister of
defense, and so on" (Laver/Shepsle 1998: 34). In short, the party that holds a portfolio sets its
policy (Laver/Shepsle 1996: 91). As a result, parties agree over portfolio allocations rather
than policies. Coalition programs may exist, but they do not reflect a compromise that is
negotiated by all coalition partners and covers all relevant policy areas and portfolios.
Ministerial discretion is apparently a key feature and critical assumption of the model. A
number of studies have challenged the assumption on both theoretical and empirical grounds.
One argument is that it seems unlikely that coalition partners are willing to accept Paretoinefficient outcomes, and the existence of formal coalition policy programs suggests that
Pareto-improvement is indeed feasible (Martin 2004; Martin/Vanberg 2004: 15f.). A second,
related argument refers to the allotment of junior ministerial positions to keep tabs on other
parties’ ministers and hereby implement policy packages that are Pareto-superior to those
implied in ministerial government (Thies 2001: 581). Thirdly, parties may make use of
parliamentary scrutiny of proposals from coalition partners’ ministers at the committee stage
which
should
render
the
enforcement
of
coalition
policy
agreements
possible
(Martin/Vanberg 2004). Finally, Bräuninger and Hallerberg (2005: 27) argue that countries
employ different cabinet decision-making rules, whereby ministerial discretion is one but only
one of the three types of rules considered in their analysis. In any case, the key problem is
how to put into effect a bargaining outcome when there is no third party equipped to monitor
or enforce the agreement.
A further possibility which to our knowledge has not been investigated in detail yet, is
the delegation of ministerial policy-making powers to members of party factions having
different interests but no bargaining power (see, however, Giannetti/Laver 2005). Suppose
1
For a similar work, see Austen-Smith and Banks (1990).
4
that as a result of the initial bargaining process, parties agree on a coalition contract that
entails some sort of policy program yet being more or less vague and incomplete. Given such
an agreement, candidates for a cabinet post having preferences that come close to the coalition
policy agreement, should be acceptable for both coalition partners. This holds irrespective of
whether or not the preferences of this individual party member deviate from those of its party
leaders. By contrast, in single-party governments, we would not expect to see members of
party factions holding cabinet posts. In this paper, we seek to explore this argument.2
We start with three assumptions. First, parties consist of a party core and, possibly,
several party factions. Party factions have no bargaining power in the process of coalition
formation. Secondly, ministers have full discretion over their portfolio once they are in office.
Thirdly, we assume that the bargaining process over the coalition policy program precedes the
stage of the division of cabinet posts; at least with respect to Germany, this is consistent with
what we know from actual coalition formation processes in the past (Ismayr 2000: 200). What
are the implications of these assumptions? We predict that cabinet posts are allocated to
members of the party core or party factions in such a way that the policy outcomes these
cabinet ministers produce come close to the policy program of the coalition government. We
therefore need to know what the coalition contract looks like.
Coalition policy programs are the result of a bargaining, i.e. a collective decisionmaking process of two or more actors having diverging interests or preferences. In such
situations numerous outcomes are in principal feasible and possible. Without further
knowledge on the actors’ bargaining power, reversion points or domain constraints, we have
to conjecture as to where the intended government policy as fixed in the coalition contract is
located. In what follows, we consider three scenarios from which we derive three alternative
hypotheses. To be more specific, let X be a multi-dimensional policy space and assume that
exactly one minister is responsible for one policy dimension or jurisdiction. Actors have
Euclidean preferences with an ideal point x ∈ X denoting the most preferred policy.
Ministerial candidates belong to one and only one party faction, or the party core. Party core
and factions are unitary actors but may diverge in what their most preferred policy in one or
more dimensions is. For a generic party A, let A0 denote the party core and A1, A2, A3 etc. the
party factions.
2
Note that we do not test the portfolio allocation model. We do not intend to explain the initial
formation of coalitions as the choice of a portfolio allocation in the “winset” of the predecessor
government. Instead we take the partisan composition of the government coalition as given. We
address the less ambitious question as to who becomes a minister for the portfolios in question given
that parties have already agreed on both forming a coalition and a coalition policy plan, and given that
they know that ministerial discretion is high afterwards.
5
In the first scenario we simply assume the coalition contract to be an efficient outcome
of the coalition bargaining of the party cores. As a result we expect the contract to be located
somewhere on the contract curve (for two parties) or, more general, the convex hull of the
ideal points of the coalition party cores. In such a situation and if ministerial discretion is
high, the allocation of cabinet posts to members of party factions can help to secure higher
benefits for both parties than a division of posts amongst party cores. Comparing portfolio
allocation of unitary parties with the one of parties with intraparty factions, we have the
following hypothesis.
Hypothesis 1: The actual non-unitary party cabinet is closer to the Pareto surface of the
coalition party cores than the respective unitary party cabinet.
Figure 1 clarifies the hypothesis in a two-dimensional setting with two parties A and B
consisting of a party core and two party factions. Factions A1 and A2 of party A have virtually
the same position on the foreign affairs dimension as the party core A0 whilst they differ with
respect to economic policies. By contrast, the factions of party B prefer different policies than
the party core on both dimensions. Suppose that we observe cabinet onon where an individual
from party faction A2 holds the economic ministry and a member of party faction B2 is the
minister of foreign affairs. Hypothesis 1 contends that the non-unitary party cabinet onon is
closer to the Pareto surface PS than the unitary party cabinet ouni that would be in place if
ministers did belong to party cores rather than party factions.
*** Figure 1 about here ***
For the second scenario we assume that parties do not only agree on Pareto-efficient policies
in the initial bargaining process but on a negotiated coalition contract that is the weighted or
unweighted mean position bu (or bw) of the parties in government (Figure 2). As a result
members of party factions should hold ministries whenever they produce outcomes closer to
the (efficient) bargaining outcome than the ones the unitary party cabinet would produce.
Hypothesis 2 and 3: If the coalition policy agreement is the unweighted (weighted)
mean position bu (bw) of the coalition parties then the actual non-unitary party cabinet is
closer to bu (bw) than the respective unitary party cabinet.
6
*** Figure 2 about here ***
Finally, for the third setting, we assign even more relevance to the coalition agreement as in
our second scenario. Suppose that we can observe the coalition contract (c) as it is often
published when the government gets into office. We contend that portfolios are allocated to
members of the party core or party factions in such a way that the policy outcome under
ministerial discretion is optimal with respect to the coalition policy program (Figure 3).
Comparing portfolio allocation of unitary parties with the one of parties with intraparty
factions, we have following hypothesis:
Hypothesis 4: If the coalition policy agreement is the observed coalition contract c then
the actual non-unitary party cabinet is closer to c than the unitary party cabinet.
*** Figure 3 about here ***
Intraparty groups and portfolio allocation in Germany since 1987
To test the above hypotheses, we analyze the programmatic positions of all political parties
represented in the German Federal Diet (Bundestag) and their most important intraparty
factions. Furthermore, we consider the coalition agreement between the government coalition
parties. The analysis is restricted to two time periods: the legislative periods from January
1987 to December 1990 and from October 1998 to September 2002.
Time frame of analysis
What are the reasons for selecting these time frames? First of all, the composition of
government is different. In the first period a coalition between the Christian Democratic
Union (CDU), their Bavarian counterpart Christian Social Union (CSU) and the liberal Free
Democratic Party (FDP) under chancellor Helmut Kohl (CDU) was in office, in 1998 a
coalition between the Socialdemocratic Party (SPD) and the Green Party (Bündnis90/Grüne;
in the following called Greens) under chancellor Gerhard Schröder (SPD) was formed
(Saalfeld 2000: 41).
Secondly and sticking to the unitary actor assumption for a moment, the first coalition
may be located at a moderate right-wing position on the general left-right ideological
dimension. Taking into account at least two of the key conflict dimensions, however, reveals a
7
huge gap in social policy between the progressive Free Democrats on the one hand and the
clearly conservative CDU/CSU on the other (Laver/Hunt 1992: 56). The moderate left-wing
so called “red-green” government does not possess such huge ideological differences on the
most important policy dimensions, according to recent expert surveys (Laver/Hunt 1992: 56,
197f.; Laver/Benoit 2005). It is therefore of interest whether in the latter case intraparty
groups of both coalition partners also have similar positions on key policy dimensions. For
the CDU/CSU-FDP cabinet, by contrast, the expectation might be that some intraparty
factions exists which try to close the programmatic gap between the coalition party cores. In
this case, these factions might help to stabilize a coalition between the Liberals and the
Christian Democrats. Such stabilizing intraparty groups, however, do not seem to be
necessary in a coalition between the ideological more coherent “red-green” government.
Third, the party constellation changed between the two observed time periods.
Between 1987 and 1990, the West-German party system was characterized by two party
blocks: SPD and Greens on one side, CDU, CSU and FDP on the other side (Jesse 2001: 73).
Only one state-level coalition, the SPD/FDP government in office in Hamburg from 1987 to
1991, was a “mixed” government (Bräuninger/König 1999: 219), i.e. it united parties that
belonged to different camps, namely government and opposition, at the federal level
(Kropp/Sturm 1998: 118; Jesse 2001: 79). The German unification in 1990 had major
implications for each party and the party system as a whole. Eastern Germany developed a
fundamentally different party system with three major parties – CDU, SPD and the leftsocialist, former communist PDS – and very weak small parties, notably Liberals and Greens
(Niedermayer 2001). Therefore – and because of a unique modification of the electoral
system for the federal election 1990 – the PDS was present in the Bundestag with a significant
share of seats until the elections in 2002. The mere existence of a left-socialist party reduced
the chances for attaining a red-green majority at the federal level. As a result, the blockthinking in coalition building vanished. Going back to the state-level, more than five states
were governed by mixed governments, mostly coalitions between the two large parties CDU
and SPD. Another implication of the demise of communism was the sharply decreasing
significance of Neo-Marxist theory inside the SPD and particularly the Greens. The leftfundamentalist wing of the latter party dropped away in 1991 (Poguntke/Schmitt-Beck 1994;
Falter/Klein 2003: 62), and the Social Democrats copied the “Third Way” of Tony Blair’s
“New Labour Party” for their 1998 electoral campaign (von Alemann 1999: 44;
Gallagher/Laver/Mair 2001: 182; Dürr 2002). The question here is whether the party as a
whole and intraparty groups of Greens and Social Democrats moved to the center or if there
8
are intraparty factions, which are still located on the left-wing of the political spectrum and
had an impact on portfolio allocation in the first red-green German government 1998. The
next section identifies the relevant intraparty factions in both time periods.
Intraparty Groups of German Parties
Perhaps the simplest criterion to identify important intraparty factions is their size, e.g.
measured as its share of members. Yet, this numerical way can not account for small but
influential intraparty groups. Examples for the latter, and also important to our question, are
factions consisting of representatives of big and small business. Their share of the population
in general and correspondingly their proportion of membership in German catch-all-parties
(Kirchheimer 1965) is low according to the social structure in modern Western democracies
(Lane/Errson 1999: 19f., 25f.). For this reason we have to use not only quantitative, but also
qualitative indicators to identify important factions in German political parties, for example
the importance of intraparty groups for collecting contributions and soft money.
Looking at the two major German parties3 the most important groups according to the
membership criterion are the lobbies of employees in CDU/CSU and SPD, respectively. In the
case of the Christian Democrats, the “Christlich-Demokratische Arbeitnehmerschaft” (CDA;
for the CSU: CSA) is the connection between CDU/CSU and labor groups. Ideologically, the
CDA draws heavily on the so-called “katholische Soziallehre” (“catholic social-doctrine”;
Pappi 1984: 13), and puts emphasis on welfare state issues and traditional values such as the
social function of the family. The organized Socialdemocratic counterpart is the
“Arbeitsgemeinschaft für Arbeitnehmerfragen” (AfA) which has strong ties to German labor
unions. Their policy priority is to expand or at least prevent the retrenchment of the German
welfare state. The intraparty opponent of the AfA is the group of Socialdemocratic selfemployed (“Arbeitsgemeinschaft Selbstständige in der SPD”, AGS). Since the roots of
Socialdemocratic parties are in the organized labor movement, the influence of this group as
compared to the impact of the AfA is considered to be low. Laterally reversed to that is the
situation in the CDU/CSU. Here the “Mittelstandsvereinigung” (representation of small and
middle-sized companies, MIT) is considered to be more influential than the CDA (Poguntke
1994: 200; Ismayr 2000: 108).
3
We here consider CDU and CSU as one party and hence as one unitary actor at the federal level.
Despite the fact that the two are formally independent parties, they form a joint parliamentary group in
the Federal Diet since 1949. Furthermore, in the empirical part of this paper we show that they have
nearly the same programmatic position in all relevant German policy dimensions.
9
One specific feature of German politics in the 1970s needs to be mentioned here. At
that time, policy-making was characterized by fierce ideological conflicts. Especially the
intraparty interest aggregation of the Socialdemocrats was aggravated by the antagonism
between more left-wing orientated, mostly younger politicians, for instance today’s
Chancellor Gerhard Schröder, and moderate, mostly senior Socialdemocrats like the then
acting Chancellor Helmut Schmidt. The supporters of both camps formed up in more informal
circles. These formally non-organized intraparty groups exist until today, and they have a
strong influence on the political direction of the party (Ismayr 2000: 110f.). The rightist group
is called the “Seeheimer Kreis” (“Seeheim Circle”). The leftist group of the party formed
under the name “Frankfurter Kreis” (“Frankfurt Circle”), or “Demokratische Linke 21”
(“Democratic Left 21”) since 2000. The more conservative “Seeheimer Kreis” attained key
influence during the late 1970s and is now the strongest intraparty supporter of the economic
reforms of the red-green government. The “Frankfurter Kreis” was very influential during the
long opposition time of the party from 1982 to 1998. We therefore consider these groups as
two important intraparty factions of the SPD and include them in our analysis.
Looking at the smaller German parties during in the time period in question, only the
PDS has a well-structured party organization with formally existent intraparty groups
(Poguntke 2001: 259). Given that the Left-Socialists were not involved in any coalition
building process at the national level, we are not going to analyze the programmatic positions
of their intraparty factions. More important to our analysis are intraparty factions of Liberals
and the Green Party. Both parties have a low and more informal organizational structure.
Especially the organization of the FDP looks like one of a “party of notabilities” or an “Elite
party” (Poguntke 1994: 201f.; Mair 1997: 97f., 110). German Liberals traditionally are
separated into a left-libertarian and a national wing, with the latter focusing more on
economic
issues
(Kirchner/Broughton
1988;
Lösche/Walter
1996).
The
informal
conglomeration of the first mentioned group is called the “Freiburg Circle” (FDP-FR). The
hardcore economic liberals convene in the working group “Liberal Business Owners” (FDPMIT). Because of their informal structure, we only could obtain programmatic documents of
these groups for the time period from 1998 to 2002. Considering the Green Party, a well
structured and detailed party organization comparable to those of other German parties does
not exist due to the “grassroots democracy”-approach of the German Greens (Poguntke 2001:
259). However, for the observation period 1998-2002 we use programmatic documents of the
Green intraparty “Committee for economic and welfare affairs” (GR-ECO) because senior
politicians in this time period belong to the reform-orientated, economically more liberal party
10
wing. In the next paragraph, we identify relevant policy dimensions in German party
competition and the most salient government offices within the coalition building process.
Key Policy Dimensions and Government Ministries
According to seminal cleavage theory (Lipset/Rokkan 1967; see also Lijphart 1999: 79ff.), the
German society and party system is structured by two key conflicts. The first and most
important one is a result of the industrialization in the late 19th century and is about how much
influence the government should have on economic issues. This socio-economic conflict,
identified as the key policy dimension in almost all Western Democracies by Anthony Downs
(1957) and Kenneth A. Arrow (1964), has established a party conflict in Germany between
Social Democrats and Liberals being opposed at the ends of the dimension, and the
CDU/CSU as the median player. But how could SPD and FDP build and maintain the longlasting coalition from 1969 to 1982 given the wide ideological range between both? The
simple reason is that on the second key policy dimension reflecting the conflict between
progressive and more conservative perspectives about the order of society, Social Democrats
and Liberals nearly have the same, progressive position, whereas the more clerical CDU/CSU
is at the other end of the dimension (Pappi 1984: 12f.). Following this argument, German
parties should try to capture those Government Ministries, where central questions of these
conflicts were decided.
Another strategy to identify the most salient policy dimensions is more empirical and
relates to the conflict structure in the present. The results of the Laver und Hunt (1992: 105)
expert survey on the question which country-specific government portfolios are the most
important ones, leads the authors to the conclusion that “if policy is what drives the
government formation process, then the salience of key cabinet portfolios will be closely
related to the salience of the dimensions over which they have jurisdiction”. Following this
argument, the foreign and the economic domain are the two most important policy areas in
Germany (Laver/Hunt 1992: 106; Laver/Shepsle 1996: 130f.; Saalfeld 2000: 41f.;
Druckman/Warwick 2005: 39). This, however, partly conflicts with the above findings from
the historical-sociological approach.
We therefore take the results of both approaches into account. We consider all three
dimensions that structure the German party competition: the economic, the social and the
foreign policy dimension. In a second step, we seek to find pairs of dimensions and portfolios,
i.e. we search for those portfolios where each conflict dimension is strongest implemented.
The identification of the portfolios for the foreign and the social dimension is unproblematic.
11
We refer to the Foreign Ministry in the first and – once more referring to the Laver und Hunt
(1992: 106) expert survey – to the Ministry of the Interior in the latter case.4 More
problematic is the allocation of one portfolio to the socio-economic dimension. The German
Cabinet structure allows three possibilities: first, the finance portfolio as used by Laver and
Shepsle (1996) for clarification of their model, secondly the economic portfolio, and thirdly
the Ministry for Labor and Social Affairs. Despite the fact that the Minister of Finance
possesses veto power in the budget negotiations (Saalfeld 2000: 66) we refer to the Ministry
for Labor and Social Affairs as the key socio-economic portfolio as it is here where the
conflict between the different points of view regarding welfare state expansion and market
regulation clash. In contrast to this, the economic portfolio lost importance and agendasetting-power over time (see e.g. the results of a new portfolio ranking conducted by
Druckman and Warwick 2005: 39).
We next have a look at the partisan composition of the portfolios under investigation.
Considering the research on the “monitoring” function of junior ministers in coalition
governments (Thies 2001) we also look at the distribution and intraparty group membership
of these so-called “Parlamentarische Staatsekretäre”. The intention for the creation of this
positions in 1967 was not only to unburden the respective minister but also to establish a
circle of politicians as potential successors for open cabinet posts (Schindler 1999: 1109).
Following the monitoring-hypothesis and Thies’ (2001: 592) findings on the importance of
intraparty factions during single-party governments in Japan, the same could be the case in
Germany. If a ministry is completely controlled by one party and the office-holder is a
member of a particular, for example left-wing intraparty group, then some right-wing factions
in the same party could try to balance policy in this area by having one of their members
nominated as a junior minister. Table 1 shows the distribution of senior and junior ministers
in the four key departments for the four government coalitions considered.
***Table 1 about here***
As Table 1 shows, for most departments and governments, senior and junior ministers are
from one and only one coalition party. The exceptions are the ministry of justice in the two
governments of Chancellor Helmut Kohl and the foreign ministry in the two governments of
Chancellor Gerhard Schröder. In each of the last mentioned cases, the respective smaller
4
For the social dimension, we are also taking into account the ministry of justice and the
programmatic positions of officeholders there.
12
coalition party holds the position of the senior minister, whilst the junior minister post is
assigned to a Social Democrat. Thus, only in two of the cases there is evidence for the
“monitoring-hypothesis”. However, keeping in mind the traditional programmatic direction of
each coalition party, Thies’ (2001) argument makes some sense. In the 1987 und 1990 cabinet
formation, the FDP received the ministry of justice. Because of their more progressive
attitudes in social issues, at least one junior minister was a member of the CDU/CSU that had
a conservative position on these policy issues. The same pattern becomes apparent in the
coalition negotiations between the Social Democrats and the Greens in 1998 and 2002. In the
end the Greens received the foreign ministry but – maybe due to their pacifistic and antimilitaristic origins and statements during the 1998 election campaign (Rüdig 2002: 82) – the
SPD held one junior minister in each of the two governments (see Saalfeld 2000: 71).
Looking at differences between intraparty factions in the composition in these four
ministries and cabinets, there is some evidence for the expectation that intraparty factions of
German political parties indeed serve as a corrective, especially in the two cabinets of
Chancellor Schröder. The Minister for Labor and Social Affairs in 1998-2002, Walter Riester,
was an active member of the AfA faction, which is labor union friendly and welfare stateorientated. By contrast, one of his junior ministers, Gerd Andres, belongs to the economically
reform-oriented “Seeheim Circle”. In the Interior and Justice departments, the respective
senior minister was not an (active) member of any intraparty faction. However, most of the
junior ministers belong to the “Seeheim Circle”. This may hint to the implementation of a
more conservative policy in these areas.
In the following section, we apply a new method for estimating (intra)party policy
positions (Laver/Benoit/Garry 2003). On this basis, we explore if first the qualitative analysis
of programmatic behavior of intraparty factions above matches to what we find in the data.
Secondly, we use the same method to analyze the (published) coalition policy agreement and
hereby estimate the position of a coalition government. According to Timmermans (1998:
423; see also Müller/Strøm 2000b: 140f., 147ff.), such contracts between coalition partners do
not only have a symbolic and conflict prevention function, their task also is to build a frame
of the political agenda for the following years of the election period. Therefore, coalition
agreements approximate the intended future policy of the respective government; taking into
consideration and balancing the positions of party cores as well as intraparty factions that
usually have representatives participating in the coalition negotiations (Timmermans 1998:
419).
13
This leads us to another point. As Maor (1992; 1995) points out, parties with a
decentralized organizational structure are able to handle intraparty conflicts in a better way as
strongly centralized parties. The risk of losing members at the ground or the party elite-level
is higher for centralized political parties than for parties with a more heterogeneous internal
structure (Maor 1995: 66). With respect to our question, Maor’s findings suggest that there
exists something like a “strength of weakness” in particular for catch-all-parties enclosing a
wide range of social groups. During coalition negotiations, such parties can always remind
their potential partner that one or more intraparty groups of them could have problems with
the proposed compromise in a specific policy-area. Under certain circumstances, e.g. the
potential coalition partner being not in a pivotal position or being highly office-oriented, this
yields a strong bargaining potential for the catch-all-party.
Estimating the Policy Positions of German Parties, Intraparty Groups and Coalition
Agreements
Though a number of methodologies and studies on concerning the measurement of party
positions exist, most of these do not provide information as to what the policy-area specific
positions of political parties are. Notably exceptions are recent expert surveys (Laver/Hunt
1992; Laver/Benoit 2005) and the data provided by the “Comparative Manifesto Project”
(CMP) that is based on manual content analysis of election manifestos (e.g., Budge et al.
2001). The problem with the first is that expert surveys are usually conducted sporadically;
data therefore refer to specific points in time and it is seldom possible to track potential
changes in the policy positions of political parties between successive elections. The CMPdata take possible changes in policy priorities of political parties into account. However, being
based on salience theory (Robertson 1976; see also Budge 2001: 82) the whole approach has
somewhat come under attack (Laver 2001: 70f.).5
5
Salience theory is based on the assumption that parties do not formulate positions for or against one
issue in their election manifestos. Moreover, it is assumed that first, parties avoid negative
formulations, and second, the frequency of mentioning one policy issue tells us something about the
position of a party in the respective policy area. However, how often an issue is articulated in a party
document reveals something about the importance of the respective policy area, but nothing about the
party’s position there. A second problem is how to identify party positions concurrently for different
policy areas and separately for each election. The standard Left-Right-dimension included into the
dataset, for instance, is based on categories whose assignment as “left” or “right” is questionable. The
categories of “Freedom and Human Rights” and “Democracy” are examples in this respect.
Furthermore, the structure of the CMP-dataset does not allow for principal component analysis
separately for each direction because of the then arising small N, large V-problem (for an alternative
methodology based on the transposed data matrix, see Pappi/Shikano 2004).
14
Even more important, there is, to our knowledge, no dataset that would provide us
with information on the positions of both “parties” as unitary actors and their (key) intraparty
factions. We therefore use the language blind and non-manual wordscore-method
(Laver/Benoit/Garry 2003). The basic idea of the wordscore procedure is to compare the
frequency distribution of words in a text, the position of which is known, to the word
distribution of a text of the same character whose position is unknown. Laver, Benoit and
Garry (2003: 314f.) refer to these two sorts of documents as “reference texts” and “virgin
texts”, respectively. In a nutshell, the position of a virgin text changes if the frequency of
some “signal” words goes up or down. Laver, Benoit and Garry (2003: 326f.) as well as
Gianetti and Laver (2005) use the wordscore technique not only to estimate the position of
political parties but also the positions of Irish MPs and cabinet members in Italy. In these
studies, “virgin texts” are speeches of the respective politician, while “reference texts” are
election manifestos calibrated with expert survey data. In this paper, we use programmatic
documents of the respective intraparty groups as virgin texts. The main reason is that the
character of both text sorts and hence the set of words used are more similar to each other
than the sets of words in election manifestos and speeches of politicians.
Finally, we are also going to estimate the policy-specific position of German coalition
agreements with election manifestos as “reference texts”. A regular way to think about the
(policy-area specific) “position of a coalition” of actors is to use the arithmetic mean of the
position of each participant, sometimes weighted by their share of seats in the parliament (e.g.
Cusack 2001). We here refer to the published coalition policy agreement as a more
independent source of information. In coalition contracts parties formulate their program for
the next years in government considering the actual circumstances in the respective country,
whilst parties in order to mobilize their voters have to formulate more radical or opportunistic
policy positions in election manifestos. There is another aspect which makes the coalition
agreement in Germany useful for this study. In German coalition bargaining processes, the
respective parties usually first talk about policies and at last about the allocation of offices to
parties and individuals (Ismayr 2000: 200; Saalfeld 2000: 47). Following this, one would
expect that cabinet posts – senior and junior ministries – are allocated in a way so that the
result of the policy-area specific coalition negotiations is best implemented by the distribution
of offices not only to parties but also intraparty groups.
15
Analysis: Intraparty groups and the coalition policy program
Descriptive Results
Before going on to evaluate the three hypotheses on the allocation of portfolios to intraparty
factions, we have a look at the descriptive results. Where are German political parties, their
intraparty factions and coalition agreements located on key policy dimensions? Figures 4 and
5 show these positions on the economic, foreign and social dimension during the time periods
from 1987 to 1990 and 1998 to 2002 (for the detailed data, see appendix A and B).6
***Figures 4 and 5 about here***
The figures reveal that the programmatic positions of German political parties are – with few
exceptions – relatively stable over time. The position of the Greens on economic issues in
2002 clearly has changed from their more left position four years before. In the same time
period, the SPD formulated a clearly more progressive position in their 2002 election
manifesto than in their 1998 program. It seems that the “New Labour” and “Third Way”imitation of German Social Democrats in 1998 was more concentrated on social then on
economic issues, according to their more left wing position in the latter policy in their election
manifesto 1998. All other parties are placed where they should be according to the qualitative
literature: the liberal FDP is the economically right-wing party during both time periods,
whilst they have a progressive position on social issues. The CDU/CSU, by contrast,
formulated a moderate right-wing position in economic issues. In questions on social policy,
the CDU/CSU is the right-wing party. In the 1990 election the Bavarian CSU drafted a
separate manifesto in addition to a joint one with the CDU. As Figure 4 shows, CSU and
CDU/CSU had almost the same programmatic position on any policy dimension. So it indeed
makes sense to analyze both parties as an unitary actor. The left-socialistic PDS is the leftwing party on economic7 and, together with the Greens, on social issues. Furthermore the
PDS is the only major outlier on the foreign policy dimension. In 1998 and 2002 the PDS
stood against a German participation in European security and peacekeeping missions
6
We assign expert survey results of 2003 (Laver/Benoit 2005) to the German election manifestos for
2002. In contrast to Laver, Benoit and Garry (2003: 325f.), we use manifestos of the 1987 election as
“reference texts”, because the Laver and Hunt expert survey was conducted in the first half of 1989
(1992: 36) and can not account for the circumstances around the German unification process, which
had an enormous impact on the 1990 election and hence for the then formulated election manifestos.
7
With the exception of the 1990 election where no reference text and hence no reference score for the
party is available.
16
(Laver/Benoit 2005), whilst in 1990, the former Communists as well as the Greens favored
closer relations to the Soviet Union (Laver/Hunt 1992: 197). In the latter mentioned policy
area, all other parties are located close to each other. But there are interesting variances in the
positioning of the parties, particularly in the last time period observed here. In 2002, the
Greens favored a German participation in military peace keeping mission just as the SPD did.
In previous time periods when they were in opposition, the pacifistic-orientated Green Party
formulated a more ambivalent position on this question in their election manifesto.
Considering the position of intraparty groups, the range the parties cover on each
dimension widens: as can be seen from Figures 4 and 5, there is a large differentiation in the
economic and social policy area, whilst intraparty groups do not differ that much from the
positions of their party cores on foreign policy issues. The reason for this is simply that nearly
all intraparty factions are primarily orientated on economic and, partly, on social issues. In the
11th election period from 1987 to 1990, where we could obtain intraparty documents of CDU
and SPD only, the “Frankfurt Circle” (FK) receives the most left-wing position on each
dimension whereas the position of the AGS, the social democratic self-employed, marks the
right wing of the intraparty organizations in the SPD. In contrast to the expectations above,
there is no huge difference between the position of the “Seeheim Circle” and the SPDemployee organization “AfA”. In case of CDU/CSU, the position of their employeeorganization CDA reaches deeply into areas on the policy dimensions covered by SPDorganizations. In contrast to the Social Democrats, no intraparty group but the party core of
the CDU/CSU determines the right border of their intraparty political spectrum. Only in 1987
the small business association MIT, marked the right end on the economic policy dimension.
In the other cases, the position of the MIT was nearby the CDU/CSU position, so one could
argue, in accordance with descriptive studies (Ismayr 2000: 108), that this group has the most
influence in the CDU/CSU as a whole.
The most striking findings become evident when looking at the estimated positions of
the CDU/CSU-FDP coalition in 1987 and 1990. In the economic policy area, despite the
right-wing position of both coalition parties, the coalition agreement is located in the center of
the dimension. Once the position of intraparty groups are taken into account, the coalition
agreement is located inside the range of the programmatic positions of the coalition parties
and their factions. The foreign policy dimension shows the same picture. Only in social
policy, where the FDP as the smaller coalition partner has a progressive position, the coalition
agreement is located near one of the coalition party cores. But once more, the CDA is also the
left-wing policy player on this dimension. In sum, when interpreting the programmatic
17
position of the coalition agreements of 1987 and 1990, it seems that intraparty groups have to
be a part of any explanation.
Figure 5 shows positions for 1998 and 2002. Broadly speaking, there are no major
changes in the positions of SPD- and CDU/CSU-intraparty factions between the 11th and the
14th election period. In economic policy, the “Democratic Left 21” (DL21), the successor
group of the “Frankfurt Circle”, maintains its left-wing position but now alongside with the
PDS. Surprisingly, there is no change in the distance between the “Seeheim circle” and the
“AfA”. Both social democratic intraparty groups have almost the same programmatic position
on all three policy dimensions. There are some changes in the programmatic behavior of
intraparty factions of CDU/CSU. Whereas CDA and their Bavarian counterpart CSA slightly
drifted to the right in economic questions, the new position of the CDU/CSU businessassociation is due to the position of the Liberals. On social issues, the ideal point of
CDU/CSU is located to the right of all Christian Democratic intraparty groups. Only the
position of the CSA comes close to the location of the CDU/CSU-election manifestos in 1998
and 2002. For the latter period, we are also able to estimate the positions of intraparty factions
of FDP and Green Party. As Figure 5 shows, the more left-libertarian “Freiburg Circle” (FDPFR) has almost the same position as the Greens or the left-wing of the Social Democrats in
social and foreign policy, whereas the “Freiburg-Circle” has a much more moderate
orientation in economic issues. The programmatic position of the liberal self-employed (FDPMIT) form up the right border in the economic dimension, whereas this faction has a more
conservative position in social policy. In questions on foreign affairs, FDP-MIT together with
the business-circle in the CDU/CSU is the group that is most favourable about German
participation on military peacekeeping missions. The single intraparty faction of the Greens
(GR-ECO) indeed holds a more moderate position on all three dimensions than the Green
party as a unitary actor.
Hence, the moderate position of the Green party and of most intraparty factions in the
SPD should enable the red-green coalition government to formulate reform-orientated,
moderate positions in the coalition agreement, especially in economic terms. But taking the
policy-area specific position of the coalition contract into account, the economic policy
agreement between SPD and Greens is leaning more towards the position of the left than to
the more moderate groups. Despite the fact that in 2002, the positions of SPD and Greens
moved to the center in economic affairs, the policy agreement almost remained at its 1998
position. In social and foreign policy, however, the coalition position moved in the same
direction both coalition parties have moved. There are two possible explanations for this.
18
First, Social Democrats could fear about losing traditional voters when implementing a radical
economic reform concept. Secondly, they try to bring members of those intraparty factions
into office, which implement the policy formulated in the coalition agreement. Whilst we are
not able to test these competing explanations, we are going to evaluate the latter in the
remainder of this paper.
Evaluation of Hypotheses
Our main concern in this paper is the role of party factions in the allocation of portfolios. We
assume that individual senior politicians belonging to party factions rather than the party core
have preferences that deviate from that of the party core. Then, if ministers do have
considerable discretion once they are in office – that is to say, they are not only difficult to
control by the parliamentary coalition or cabinet members of the coalition partner but also by
their own party core – the appointment of party faction members results in (forecasts of)
coalition policy outcomes that are different from those the appointment of core party members
would produce. Hence the strategic allocation of portfolios to party faction members can
secure policy outcomes or even allow for party coalitions that would not have been possible
otherwise.
Whilst a test of the second question is beyond the scope of this study, we can address
the first one: Is the non-unitary party cabinet (a set of ministers belonging to party factions)
closer to the “intended” coalition policy outcome than the unitary party cabinet (the same set
of ministers belonging to parties as unitary actors)? This necessitates specifying what the
“intended” coalition policy outcome is or was.8 In the theoretical part of this paper we
considered three straightforward options from which we derived three hypotheses. We first
assumed that coalition parties prefer outcomes close the Pareto surface to outcomes that are
more distant to the Pareto surface. Second, we considered the (weighted) center of the Pareto
surface as a likely outcome of the coalition policy bargaining and therefore as the “intended”
coalition policy. Thirdly, we use the observed coalition policy program as a proxy for what
the coalition intends to do in the succeeding legislative term. The expectation with all three
scenarios is that the appointment of party faction members allows to get closer to the intended
outcomes.
8
Note that the “intended” policy outcome in the Laver/Shepsle (1996) model is endogenous: when
making governments parties select portfolio allocations or policy outcomes from a finite set a feasible
ones (those in the so-called winset of the status quo). In our model, government formation is not to be
explained, and in the absence of a status quo policy any given coalition can select any policy as the
“intended” coalition policy program. The question then becomes how to make the implementation of
(most of) this program possible.
19
***Table 2 about here***
Table 2 shows the results. The columns list Euclidean distances between cabinets and
alternative intended policy outcomes in the three-dimensional policy space of economic,
social and foreign policy. For senior ministers, the position of the cabinet is straightforward. It
is the combined position of the minister of labor on the economic dimension, the positions of
the ministers for interior and foreign affairs on the social and the foreign policy dimension,
respectively. Accordingly, the positions of the junior ministers are used for the “junior
minister cabinet”; if there is more than one junior minister in a department the average of their
positions is used. Two things are apparent from the table. First, there is absolutely no
evidence for the “Pareto improvement” hypothesis suggesting that either actors have no
intention to get closer to the Pareto surface, or party factions do not play a role in the strategic
calculations of minister appointments.9 The same applies to the “unweighted bargaining”
scenario. Second, there is considerable evidence for the “weighted bargaining” hypothesis and
even more so for the “coalition contract” hypothesis. Under both assumptions, the party
faction cabinets are closer to the intended outcome than the unitary actor cabinets. This holds
for five, respectively six of the eight senior and junior minister cabinets in the time period
under consideration.
Conclusion
The results of the empirical analysis are mixed and somewhat uncomforting. First, looking at
parties as unitary actors, the German case study does provide support for the Laver/Shepsle
(1996) standard portfolio allocation approach. We find no evidence that junior ministers are
placed on the side of senior ministers in order to serve as watchdogs (Thies 2001) who either
control the execution of the coalition contract or balance the partisan policy-making of the
senior minister. By contrast, most junior ministers are from the same party or even party
faction as the senior minister is.
Second, we find that considering intraparty politics in studying portfolio allocation and
coalition bargaining is worthwhile. The empirical results suggest that the appointment of
individual ministers who belong to party factions is used to secure the enforcement of the
coalition contract or, at least, make such an enforcement more likely or make deviations from
9
Or, the assumption of ministerial discretion is unjustified. This, however, is a last resort objection to
our analysis as the model is based on the ministerial discretion assumption.
20
the contract less severe. This holds for both senior and junior ministers. By contrast, there is
no evidence that the allocation of portfolios to members of party factions is used to get
outcomes that are simply more efficient from the (unitary) coalition parties’ point of view.
The expected outcomes under portfolio allocation with intraparty factions are, as compared to
the standard allocation, neither closer to the contract curve of the unitary parties, nor are they
closer to a fictitious and efficient bargaining outcome on the contract curve.
These results, however, are unsatisfying and disturbing. With respect to our German
case study, we find that intraparty groups of major German parties are largely organized
along the socio-economic conflict which is the most important cleavage in the German party
system. This is reflected in both the (nominal) labeling and self-reference of those groups, and
the ideal positions of these factions in the multi-dimensional policy space. Accordingly, we
expect intraparty politics to be most relevant for the appointment of the (junior) Minister of
Economy. While this is in line with our expectation and a noteworthy insight in the Germany
party system, it is also a misfortune for the empirical analysis as it reduced the number of
cases on which we could “test” our hypothesis (which is already low in the case study design
we provide).
At the theoretical level, we end with more questions raised than answered. Assuming
ministerial discretion we find that intraparty factions might be relevant when it comes to the
explanation of portfolio allocations. But in our model, ministers (are supposed to) execute a
negotiated coalition contract, while the content or spatial location of that contract is far from
being obvious. In many cases it is neither on (or close to) the contract curve of the coalition
parties as unitary actors. It is, however, almost always in the (uni-dimensional) range of the
positions of governmental actors when we consider both party cores and their factions as key
actors in the coalition bargaining process. As we have no theory at hand that could plausibly
explain how these contracts come about (and also because of the small number of cases in our
approach) in such situations, we had to measure and enter the contract into the analysis as an
exogenous variable. This is a misery that calls for further studies into the conditions of
coalition bargaining with intraparty factions (see, however, Luebbert 1986; Bäck 2004).
Finally, with respect to the often emerging and – after the last state elections – current
discussion about building of a so called “great coalition” between Christian and Social
Democrats in Germany, our findings suggest that such a coalition would not be able to
implement major reforms in the socioeconomic policy area. Despite the fact that such a
coalition would have a broad majority in the lower and the upper chamber and therefore could
implement their policy views a lot easier than under the existence of a large opposition party
21
as a veto player (König 2001, Tsebelis 2002: 145), the ideological positions of intraparty
factions in both German catch-all parties would not easily allow for major economic and
welfare policy reforms. The reason is that both large parties in Germany have strong
economically left-orientated intraparty wings which, as the analysis above suggests, can not
be ignored in the formulation of the coalition agreement. This supports the “strength of
weakness” conjecture which follows from Maor’s (1995) findings. Finally, one might
speculate whether not only the “cooperative federalism” of the German political system but
also the programmatic positions and strength of intraparty groups in German catch-all parties
is one of the reasons which hinder economic reforms in Germany.
References
Arrow, Kenneth A. (1964): Social Choice and Individual Values. 2nd edition. New York: Wiley.
Austen-Smith, David; Banks, Jeffrey (1990): Stable Governments and the Allocation of Policy Portfolios. In:
American Political Science Review 84: 3, 891-906.
Axelrod, Robert (1970): Conflict of Interest. A theory of divergent goals with applications to politics. Chicago:
Markham.
Cusack,
Thomas
(2001):
Partisanship
in
the
Setting
and
Coordination
of
Fiscal and Monetary Policies. In: European Journal of Political Research 40: 1,
93- 115.
Bäck, Hanna (2004): Intraparty Politics and Coalition Formation Evidence from Swedish Local Government.
Paper prepared for delivery at the Nordic Local Government Studies Conference in Oslo, 26–28
November.
Bräuninger, Thomas; Hallerberg, Mark (2005): Cabinet Decision-Making and Policy Outcomes in
Parliamentary Democracies. Manuscript. Atlanta: Emory University.
Bräuninger, Thomas; König, Thomas (1999): The checks and balances of party federalism: German federal
government in a divided legislature. In: European Journal of Political Research 36: 207-234.
Budge, Ian; Klingemann, Hans-Dieter; Volkens, Andrea; Bara, Judith; Tanenbaum, Eric (eds; 2001): Mapping
Policy Preferences. Estimates for Parties, Electors and Governments 1945-1998. Oxford: Oxford
University Press.
Budge, Ian (2001): Theory and Measurement of Party Policy Positions. In: Budge, Ian; Klingemann, HansDieter; Volkens, Andrea; Bara, Judith; Tanenbaum, Eric (eds): Mapping Policy Preferences. Estimates
for Parties, Electors and Governments 1945-1998. Oxford: Oxford University Press, 75-90.
De Swaan, Abram (1973): Coalition Theories and Cabinet Formation. Amsterdam, New York, San Francisco:
Elsevier.
Dodd, Lawrence C. (1976): Coalitions in Parliamentary Government. Princeton: Princeton University Press.
Downs, Anthony (1957): An Economic Theory of Voting. New York: Harper.
Druckman, James N.; Warwick, Paul V. (2005): The missing piece: Measuring portfolio salience in Western
European parliamentary democracies. In: European Journal of Political Research 44: 1, 17-42.
Dürr, Tobias (2002): Die Linke nach dem Sog der Mitte. Zu den Programmdebatten von SPD, Grünen und PDS
in der Ära Schröder. In: Aus Politik und Zeitgeschichte 52: 21, 5-12.
Falter, Jürgen; Klein, Michael (2003): Der lange Weg der Grünen. München: Beck.
Gallagher, Michael; Laver, Michael; Mair, Peter (2001): Representative Government in Modern Europe.
London: McGraw-Hill.
Gianetti, Daniela; Laver, Michael (2005): Policy positions and jobs in the government. In: European Journal of
Political Science 44: 1, 91-120.
Heller, William B. (2001): Making Policy Stick: Why the Government Gets What It Wants in Multiparty
Parliaments. In: American Journal of Political Science 45: 4, 780-798.
Ismayr, Wolfgang (2000): Der deutsche Bundestag. Opladen: Leske und Budrich.
Jesse, Eckhard (2001): Die Parteien im westlichen Deutschland von 1945 bis zur deutschen Einheit 1990. In:
Gabriel, Oscar W.; Niedermayer, Oskar; Stöss, Richard (eds): Parteiendemokratie in Deutschland.
Bonn: Bundeszentrale für politische Bildung, 59-83.
22
Kirchheimer, Otto (1965): Der Wandel des westeuropäischen Parteiensystems. In: Politische
Vierteljahresschrift 6: 1, 20-41.
Kirchner, Emil J.; Broughton, David (1988): The FDP in the Federal Republic of Germany: the requirements of
survival and success, in: Kirchner, Emil J. (eds): Liberal Parties in Western Europe. Cambridge, New
York: Cambridge University Press, 62-92.
Kürschners Volkshandbuch (1987): Deutscher Bundestag: 11. Wahlperiode. Darmstadt.
Kürschners Volkshandbuch (1991): Deutscher Bundestag: 12. Wahlperiode. Darmstadt.
Kürschners Volkshandbuch (1999): Deutscher Bundestag: 14. Wahlperiode. Darmstadt.
Kürschners Volkshandbuch (2003): Deutscher Bundestag: 15. Wahlperiode. Darmstadt.
König, Thomas (2001): Bicameralism and Party Politics in Germany: an Empirical Social Choice Analysis. In:
Political Studies 49: 3, 411-437.
Kropp, Sabine; Sturm, Roland (1998): Koalitionen und Koalitionsvereinbarungen. Theorie, Analyse und
Dokumentation. Opladen: Leske und Budrich.
Lane, Jan-Erik; Errson, Svante (1999): Politics and Society in Western Europe. 4th edition. London, Thousand
Oaks, New Delhi: Sage.
Laver, Michael (2001): Position and salience in the policies of political actors, in: id. (ed): Estimating the Policy
Positions of Political Actors. London, New York: Routledge, 66-76.
Laver, Michael; Shepsle, Kenneth (1990): Coalitions and cabinet government. In: American Political Science
Review 84: 3, 873-890.
Laver, Michael; Hunt, W. Ben (1992): Policy and Party Competition. New York, London: Routledge.
Laver, Michael; Shepsle, Kenneth (1996): Making and Breaking Governments. Cambridge: Cambridge
University Press.
Laver, Michael; Shepsle, Kenneth (1998): Events, Equilibria, and Government Survival, in: American Journal
of Political Science 42: 1, 28-54.
Laver, Michael; Benoit, Kenneth; Garry, John (2003): Extracting Policy Positions from Political Texts Using
Words as Data. American Political Science Review 97: 2, 311-331.
Laver, Michael; Benoit, Kenneth (2005): Party Policy in Modern Democracies. London (forthcoming).
Leiserson, Michael (1968): Factions and Coalitions on One- Party Japan. In: American Political Science Review
62: 3, 770-787.
Lijphart, Arend (1999): Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries.
New Haven, London: Yale University Press.
Lipset, Seymour M.; Rokkan, Stein (1967): Cleavage Structures, Party Systems and Voter Alignments: An
Introduction. In: Lipset, Seymour M.; Rokkan, Stein (eds): Party Systems and Voter Alignments:
Cross-National Perspectives. New York, London: Collier-Macmillan, 1-64.
Lösche, Peter; Walter, Franz (1996): Die FDP. Richtungsstreit und Zukunftszweifel. Darmstadt:
Wissenschaftliche Buchgesellschaft.
Luebbert, Gregory M. (1986): Comparative Democracy. Policymaking and Governing Coalitions in Europe and
Israel. New York: Columbia University Press.
Mair, Peter (1997): Party System Change: Approaches and Interpretations. Oxford: Clarendon Press.
Maor, Moshe (1992): Intra-Party Conflict and Coalitional Behaviour in Denmark and Norway: The Case of
“Highly Institutionalised“ Parties. In: Scandinavian Political Studies 15: 2, 99-116.
Maor, Moshe (1995): Intra-Party Determinants of Coalition Bargaining. In: Journal of Theoretical Politics 7: 1,
65-91.
Martin, Lanny W. (2004): The Government Agenda in Parliamentary Democracies. In: American Journal of
Political Science 48: 3, 445-461.
Martin, Lanny W.; Vanberg, Georg (2004): Policing the Bargain: Coalition Government and Parliamentary
Scrutiny. In: American Journal of Political Science 48: 1, 13-27.
McKelvey, Richard D. (1986): Covering, Dominance, and Institution-Free Properties of Social Choice. In:
American Political Science Review 30: 2, 283-314.
Mitchell, Paul (1996): The Life and Times of Coalition Governments. Coalition Maintenance by Event
Management. Florence: Doctoral dissertation European University Institute.
Müller, Wolfgang C.; Strøm, Kaare (2000a): Coalition Governments in Western Europe. Oxford: Oxford
University Press.
Müller, Wolfgang C.; Strøm, Kaare (2000b): Die Schlüssel zum Zusammensein: Koalitionsabkommen in
parlamentarischen Demokratien. In: van Deth, Jan; König, Thomas (eds): Europäische
Politikwissenschaft: Ein Blick in die Werkstatt. Frankfurt, New York: Campus, 136-170.
Niedermayer, Oskar (2001): Nach der Vereinigung: der Trend zum fluiden Fünfparteiensystem. In: Gabriel,
Oscar W.; Niedermayer, Oskar; Stöss, Richard (eds): Parteiendemokratie in Deutschland. Bonn:
Bundeszentrale für politische Bildung, 107-127.
Pappi, Franz Urban (1984): The West German Party System. In: Bartolini, Stefano; Mair, Peter (eds): Party
Politics in Contemporary Western Europe. London: Cass, 7-26.
23
Pappi, Franz Urban; Shikano, Susumu (2004): Ideologische Signale in den Wahlprogrammen der deutschen
Bundestagsparteien 1980 bis 2002. Arbeitspapiere – Mannheimer Zentrum für europäische
Sozialforschung Nr. 76, Mannheim.
Peleg, Bazalel (1981): Coalition formation in simple games with Dominant Players. In: International Journal of
Game Theory 10: 1, 11-33.
Poguntke, Thomas (1994): Parties in a Legalistic Culture: The Case of Germany. In: Katz, Richard S.; Mair,
Peter (eds): How Parties Organize. Change and Adaption in Party Organizations in Western
Democracies. London: Sage, 185-215.
Poguntke, Thomas (2001): Parteorganisationen in der Bundesrepublik Deutschland: Einheit in der Vielfalt? In:
Gabriel, Oscar W.; Niedermayer, Oskar; Stöss, Richard (eds): Parteiendemokratie in Deutschland.
Bonn: Bundeszentrale für politische Bildung, 253-273.
Poguntke, Thomas; Schmitt-Beck, Rüdiger (1994): Still the Same with a New Name? Bündnis90/Die Grünen
after the fusion. In: German Politics 3: 1, 91-113.
Riker, William (1962): The Theory of Political Coalitions. New Haven: Yale University Press.
Robertson, David (1976): A Theory of Party Competition. London, New York, Sydney, Toronto: Wiley &
Sons.
Rüdig, Wolfgang (2002): Germany. In: Müller-Rommel, Ferdinand; Poguntke, Thomas (eds): Green Parties in
National Governments. London, Portland: Cass, 78-111.
Saalfeld, Thomas (2000): Germany: Stable Parties, Chancellor Democracy, and the Art of Informal Settlement.
In: Müller, Wolfgang C.; Strøm, Kaare (eds): Coalition Governments in Western Europe. Oxford:
Oxford University Press, 32-85.
Schindler, Peter (1999): Datenhandbuch zur Geschichte des Deutschen Bundestages 1949 bis 1999. Band 1.
Baden-Baden: Nomos.
Schofield, Norman (1995): Coalition Politics. A Formal Model and Empirical Analysis. In: Journal of
Theoretical Politics 7: 2, 245-281.
Strøm, Kaare; Müller, Wolfgang C. (eds; 1999): Political Parties and Hard Choices. In: id.: Policy, Office, or
Votes? How Political Parties in Western Europe Make Hard Decisions. Cambridge: Cambridge
University Press, 1-35.
Strøm, Kaare; Müller, Wolfgang C. (2001): Coalition Agreements and Governance. Paper prepared for the
Annual Meeting of the American Political Science Association, Panel “Bargaining, Transaction Costs,
and Parliamentary Coalitions”. San Francisco.
Thies, Michael F. (2001): Keeping Tabs on Partners: The Logic of Delegation in Coalition Governments. In:
American Journal of Political Science 45: 3, 580-598.
Timmermans, Arco (1998): Conflicts, Agreements, and Coalition Governance. In: Acta Politica 33: 4, 409-432.
Tsebelis, George (2002): Veto Players. How Political Institutions Work. New York, Princeton: Princeton
University Press.
von Alemann, Ulrich (1999): Der Wahlsieg der SPD von 1998: Politische Achsenverschiebung oder glücklicher
Ausreißer? In: Niedermayer, Oskar (ed): Die Parteien nach der Bundestagswahl 1998. Opladen: Leske
und Budrich, 37-62.
van Roozendahl, Peter (1992): Cabinets in Multi-Party Democracies. Amsterdam: Thesis Publishers.
von Winter, Thomas (1990): Die Sozialausschüsse der CDU. Sammelbecken für christdemokratische
Arbeitnehmerinteressen oder linker Flügel der Partei? In: Leviathan 18: 3, 390-416.
Warwick, Paul (1994): Government Survival in Parliamentary Democracies. New York: Cambridge University
Press.
Sources
CDU/CSU
Christlich-Demokratische Arbeitnehmerschaft (CDA) (1987): Hamburger Erklärung. Hamburg.
Christlich-Demokratische Arbeitnehmerschaft (CDA) (2001): Grundsatzprogramm: Erst der Mensch, dann der
Markt. Bonn.
Christlich-Demokratische Union Deutschlands (CDU) (1986): Weiter so, Deutschland - für eine gute Zukunft.
Das Wahlprogramm von CDU und CSU für die Bundestagswahl 1987. Bonn.
Christlich-Demokratische Union Deutschlands (CDU) (1990): Ja zu Deutschland – ja zur Zukunft:
Wahlprogramm der Christlich Demokratischen Union Deutschlands zur gesamtdeutschen
Bundestagswahl am 2. Dezember 1990. Bonn.
24
Christlich-Demokratische Union Deutschlands (CDU) (1998): Zukunftsprogramm der CDU Deutschlands.
Bonn.
Christlich-Demokratische Union Deutschlands (CDU) (2002): Leistung und Sicherheit. Regierungsprogramm
2002-2006. Berlin.
Christlich-Soziale Arbeitnehmerschaft in der CSU (CSA) (2000): Dem Menschen verpflichtet – christlich und
sozial.
München.
Download:
http://www.csu.de/csa/csa_Themen_Leitsaetze_2000?verband=ag_csa&reiter=reiter4 (02.11.2004).
Christlich-Soziale Union (CSU) (1990): Heimat Bayern. Zukunft Deutschland. Mit uns. CSU. Programm der
Christlich-Sozialen Union zur Bundestagswahl am 2. Dezember 1990. München.
Mittelstandvereinigung der CDU/CSU (MIT) (1987): Strukturreform des Gesundheitswesens – Strategien zur
Begrenzung der Sozialkostenlawine. Bonn: Mittelstands-Verlagsgesellschaft, 15-24.
Mittelstandvereinigung der CDU/CSU (MIT) (1989): Europäisch denken und europäisch handeln. Bonn:
Mittelstands-Verlagsgesellschaft, 21-27.
Mittelstand- und Wirtschaftsvereinigung der CDU/CSU (MIT) (2003): Zukunft für Deutschland. Freiheit und
Verantwortung.
Kölner
Leitsätze.
Download:
www.mittelstanddeutschland.de/DOWNLOAD/2003/kleitsaetze_endfassung.pdf (04.11.2004)
Mittelstand- und Wirtschaftsvereinigung der CDU/CSU (MIT) (2004): Fortschreibung der Kölner Leitsätze.
Download: http://www.mittelstand-deutschland.de/05-politik.htm (04.11.2004)
SPD
Arbeitgemeinschaft Selbstständige in der SPD (AGS) (1986): Auf die kleinen und mittleren Betriebe kommt es
an. Hamburg.
Bundeskonferenz der Arbeitsgemeinschaft für Arbeitnehmerfragen in der SPD (AfA) (2004): Solidarisch und
gerecht! Den Wandel sozial gestalten. Erfurt. Download: http://afa.spd.de/servlet/PB/menu/1500191/
(01.11.2004)
Bundesvorstand der Arbeitsgemeinschaft für Arbeitnehmerfragen in der SPD (AfA) (1986): Leitlinien
sozialdemokratischer Wirtschaftspolitik. Bonn.
Bundesvorstand der Arbeitsgemeinschaft Selbstständige in der SPD (AGS) (1990): Arbeitsgemeinschaft setzt
sich durch. Bonn.
Arbeitgemeinschaft Selbstständige in der SPD (AGS) (2002a): Zukunft braucht Selbstständigkeit. Berlin.
Download: http://ags.spd.de/servlet/PB/menu/1265382/index.html (01.11.2004)
Arbeitgemeinschaft Selbstständige in der SPD (AGS) (2002b): Reformvorschlag Rentensystem. Berlin.
Download: http://ags.spd.de/servlet/PB/menu/1265382/index.html (01.11.2004)
Arbeitgemeinschaft Selbstständige in der SPD (AGS) (2002c): Arbeitmarktreform. Berlin. Download:
http://ags.spd.de/servlet/PB/menu/1265382/index.html (01.11.2004)
Arbeitgemeinschaft Selbstständige in der SPD (AGS) (2004): Mittelstand im Mittelpunkt. Memorandum für eine
aktive sozialdemokratische Mittelstandspolitik. Berlin.
Demokratische Linke 21 (DL21) (2001a): Selbstverständnis Forum Demokratische Linke 21. Download:
http://www.forum-dl21.de/Ueber_das_Forum_DL21.7.0.html (03.11.2004)
Demokratische Linke 21 (DL21) (2001b): Zur Zukunft der Parteilinken. Download: http://www.forumdl21.de/Ueber_das_Forum_DL21.7.0.html (03.11.2004)
Frankfurter Kreis (FK) (1986a): Die Zukunft gestalten: Arbeit für alle, Verbesserung der Lebensqualität,
Ökologische Gesamtreform. Thesen der sozialdemokratischen Linken. Bonn.
Frankfurter Kreis (FK) (1986b): Für eine neue Friedens- und Sicherheitspolitik der SPD. OerErkenschwick/Bonn.
Seeheimer Kreis (Seeheim) (1987): Eine Seeheimer Beitrag zur Sozialdemokratischen Programmdiskussion.
Bonn.
Seeheimer Kreis (Seeheim) (2003a): Vom bürokratischen Sozialstaat zum sozialen Bürgerstaat. Download:
http://www.seeheimer-kreis.de/hintergrund/krise_als_chance (02.11.2004)
Seeheimer Kreis (Seeheim) (2003b): In der Krise liegt die Chance – Die Sozialdemokratie im Umbruch.
Download: http://www.seeheimer-kreis.de/hintergrund/krise_als_chance (02.11.2004)
Seeheimer Kreis (Seeheim) (2003c): „Mut zur Veränderung“ – Agenda 2010: Reformen für Deutschland.
Download: http://www.seeheimer-kreis.de/hintergrund/krise_als_chance (02.11.2004)
Sozialdemokratische Partei Deutschlands (SPD) (1987): Zukunft für alle - arbeiten für soziale Gerechtigkeit und
Frieden, Regierungsprogramm der Sozialdemokratischen Partei Deutschlands. Bonn.
Sozialdemokratische Partei Deutschlands (SPD) (1990): Der neue Weg : ökologisch, sozial, wirtschaftlich,
stark. Regierungsprogramm 1990 – 1994. Bonn.
Sozialdemokratische Partei Deutschlands (SPD) (1998): Arbeit, Innovation und Gerechtigkeit: SPDWahlprogramm für die Bundestagswahl 1998. Bonn.
25
Sozialdemokratische Partei Deutschlands (SPD) (2002): Erneuerung und Zusammenhalt. Regierungsprogramm
2002-2006. Berlin.
FDP
Bundesvereinigung Liberaler Mittelstand (FDP-MIT) (2003): Ein Programm für den Mittelstand. Berlin.
Download: http://www.liberaler-mittelstand.de/pdf/Flyer-Programm.pdf (01.11.2004)
Freiburger Kreis (FDP-FR) (1999a): Würzburger Erklärung des Freiburger Kreises. Download von:
http://www.freiburgerkreis.de/ (01.11.2004).
Freiburger Kreis (FDP-FR) (1999b): Für eine Erneuerung liberaler Politik. Download von:
http://www.freiburgerkreis.de/ (02.11.2004).
Freiburger Kreis (FDP-FR) (2000): Nürnberger Erklärung des Freiburger Kreises. Positionsbestimmung der
F.D.P. Download von: http://www.freiburgerkreis.de/ (02.11.2004)
Freie Demokratische Partei (FDP) (1986): Zukunft und Leistung. Die Wahlplattform der F.D.P. zur
Bundestagswahl 1987. Bonn.
Freie Demokratische Partei (FDP) (1990): Das liberale Deutschland. Programm der F.D.P. zu den
Bundestagswahlen am 2. Dezember 1990. Sankt Augustin.
Freie Demokratische Partei (FDP) (1998): Es ist ihre Wahl: das Wahlprogramm der F.D.P. zur Bundestagswahl
1998. Bonn.
Freie Demokratische Partei (FDP) (2002): Bürgerprogramm 2002. Programm der FDP zur Bundestagswahl
2002. Mannheim.
Greens
Bündnis 90/Die Grünen (GR) (1998): Grün ist der Wechsel. Programm zur Bundestagswahl 1998. Bonn.
Bündnis 90/Die Grünen (GR) (2002): Grün wirkt! Unser Wahlprogramm 2002-2006. Berlin.
Bündnis 90/Die Grünen – Sozialpolitische Kommission (GR-ECO) (2003): Sozial ist nicht egal! Für die Reform
der
Sozialversicherungssysteme.
Berlin.
Download:
http://www.gruenepartei.de/cms/files/dokbin/32/32675.reader_sozialpolitik_komplett.pdf (01.11.2004)
Die Grünen (GR) (1987): Bundestagswahl 1987: Farbe bekennen. Bonn.
Die Grünen (GR) (1990): Das Programm zur ersten gesamtdeutschen Wahl 1990. Bonn.
PDS
Linke Liste/Partei des Demokratischen Sozialismus (PDS) (1990): Wahlprogramm der Linken Liste/PDS zur
Bundestagswahl 1990. Berlin.
Partei des Demokratischen Sozialismus (PDS) (1998): Programm der PDS zur Bundestagswahl 1998. Berlin.
Partei des Demokratischen Sozialismus (PDS) (2002): Die linke Kraft. Wahlprogramm der PDS zur
Bundestagswahl 2002. Berlin.
Coalition Agreements
Christlich-Demokratische Union, Christlich-Soziale Union, Freie Demokratische Partei (1987):
Koalitionsvereinbarungen für die 11. Legislaturperiode des Deutschen Bundestages vom 10. März
1987. Bonn.
Christlich-Demokratische Union, Christlich-Soziale Union, Freie Demokratische Partei (1991):
Koalitionsvereinbarung für die 12. Legislaturperiode des Deutschen Bundestages. Bonn.
Sozialdemokratische Partei Deutschlands, Bündnis 90/Die Grünen (1998): Aufbruch und Erneuerung –
Deutschlands Weg ins 21. Jahrhundert. Koalitionsvereinbarung zwischen der Sozialdemokratischen
Partei Deutschlands und Bündnis 90/Die Grünen 20. Oktober 1998. Bonn.
Sozialdemokratische Partei Deutschlands, Bündnis 90/Die Grünen (2002): Koalitionsvertrag 2002 – 2006:
Erneuerung – Gerechtigkeit – Nachhaltigkeit. Für ein wirtschaftlich starkes, soziales und ökologisches
Deutschland. Für eine lebendige Demokratie. Berlin.
26
Figure 1:
Predicted distance of unitary and non-unitary party cabinets to the Pareto
Foreign policy
surface
A1
A0
A2
PS
B2
onon
ouni
B1
B0
Economic policy
Figure 2:
Predicted distance of unitary and non-unitary party cabinets to a fictitious
Foreign policy
bargaining outcome
A1
A0
A2
bu
onon
bw
ouni
B1
B2
B0
Economic policy
27
Figure 3:
Predicted distance of unitary and non-unitary party cabinets to an observed
Foreign policy
coalition contract
A1
A0
A2
c
B2
onon
ouni
B1
B0
Economic policy
28
Table 1:
Senior and Junior Ministers
Term/Government Department
Minister
Junior Minister
1987-1990
(CDU/CSU-FDP)
Labor
Foreign
Interior
Vogt, CDU-CDA; Höpfinger, CSU-CSA
Adam-Schwaetzer, FDP; Schäfer, FDP
Waffenschmidt, CDU; Spranger, CSU
1990-1994
(CDU/CSU-FDP)
Labor
Foreign
Interior
Labor
Blüm, CDU-CDA
Genscher, FDP
Zimmermann,
CSU
Blüm, CDU-CDA
Genscher, FDP
Schäuble, CDU
Riester, SPD-AfA
Foreign
Interior
Fischer, Grüne
Schily, SPD
1998-2002
(SPD/Greens)
Günther, CDU; Seehofer, CSU
Adam-Schwaetzer, FDP; Schäfer, FDP
Waffenschmidt, CDU; Lintner, CSU
Andres, SPD-Seeheim; Mascher, SPD
Verheugen, SPD; Volmer, Grüne
Körper, SPD-Seeheim; SonntagWohlgast, SPD
2002Labor/Economy Clement, SPD
Andres, SPD-Seeheim; Schlauch, Grüne;
(SPD/Greens)
Staffelt, SPD-Seeheim
Foreign
Fischer, Grüne
Bury, SPD; Müller, Grüne
Interior
Schily, SPD
Körper, SPD-Seeheim; Vogt, SPD
Source: Short biographies in the official Bundestag handbook (Kürschners Volkshandbuch 1987,
1991, 1999, 2003); official websites of intraparty groups (http://www.parlamentarische-linke.de/,
http://www.seeheimer-kreis.de/organisation/leitungskreis, http://www.kas.de/archiv/acdp/
832_webseite.html). A senior or junior minister is coded as a party faction member if he or she
explicitly mentions the faction in a short biography or is mentioned as a active member of the
respective intraparty group.
Table 2:
Distance of unitary and faction party cabinets to coalition policy agreement
Senior Minister
H1
(Pareto surface)
H2
(Unweighted bargaining)
H3
(Weighted bargaining)
H4
(Coalition contract)
Junior Minister
Year
Unitary Faction Hypothesis
Unitary Faction Hypothesis
1987
1990
1998
2002
1987
1990
1998
2002
1987
1990
1998
2002
1987
1990
1998
2002
2.98
2.87
2.89
1.08
4.26
4.22
8.26
2.65
9.60
9.85
10.63
7.05
7.85
9.74
6.45
1.45
2.98
2.87
1.45
0.73
4.26
4.22
8.13
2.47
9.60
9.85
9.88
7.16
7.85
9.74
6.07
1.47
6.38
8.81
4.44
1.08
7.77
9.68
9.16
2.65
7.50
7.73
12.21
7.05
6.95
8.31
6.43
1.45
not in favor
not in favor
not in favor
(not in favor)
not in favor
not in favor
not in favor
(not in favor)
in favor
in favor
not in favor
(not in favor)
in favor
in favor
in favor
(not in favor)
6.38
5.11
2.28
1.68
7.77
6.22
4.74
2.09
7.50
7.83
8.14
7.90
6.95
8.06
1.90
2.67
not in favor
not in favor
not in favor
not in favor
not in favor
not in favor
in favor
in favor
in favor
in favor
in favor
not in favor
in favor
in favor
in favor
not in favor
29
Figure 4:
Position of German Political Parties, Intraparty Factions and Coalition
CSU
CDU/CSU
FDP
Governments, 1987-1990
Coal
AGS
MIT
25
FDP
CDU/CSU
SPD
GR
1987
15
AFA
5
Coal
-5
Seeheim
FK
1990
PDS
GR
SPD
Economic Policy
5
15
1987
CDU/CSU
CSU
MIT
AGS
15
25
CDU/CSU
SPD
FDP
5
GR
-5
Co al Seeheim
CDA
AFA
FK
1990
Coal
FDP
SPD
PDS
GR
Social Policy
1987
25
Coal
FDP
-5
Coal
SPD
1990
CDU/CSU
CSU
Foreign Policy: Pro friendly relations with USSR vs. anti
30
Figure 5:
Position of German Political Parties, Intraparty Factions and Coalition
Governments, 1998-2002
FDP
CDU/CSU
FDP-M IT
MIT
FDP
Coal
15
SPD
PDS
1998
5
GR
-5
AFA
FDP-FR
Seeheim
C DU/CSU CDA
CSA
GR-EC O
AGS
DL21
2002
GR
PDS
Coal
SPD
Economic Policy
25
2002
5
1998
15
CDU/CSU
FDP-MIT
25
CDU/CSU
15
SPD
FDP
5
Coal
PDS
GR
1998
FDP-FR
-5
GR-ECO
DL21
CDA
Seeheim
AFA
MIT
AGS
CSA
Coal
SPD
PDS
FDP
GR
Social Policy
2002
25
PDS
-5
PDS
FDP-MIT Coal
CDU/CSU
SPD
MIT
FDP
GR-ECO
AGS CSA CDU/CSU
SPD
Coal
CDA GR
Seeheim
AFA
GR
DL21
FDP-FR
Foreign Policy: In favour of envolvement in European security and peacekeeping
missions vs. opposes any involvement
31
Appendix A: Positions of German political parties, intraparty factions and coalition agreements, 1987-90
Party/
Economic Policy
Intraparty
Foreign Policy
Share of
Social Policy
Total scored
Total words
scored
faction/
Reference or
95% lower
95% upper
Reference or
95% lower
95% upper
Reference or
95% lower
95% upper
Coalition
transformed
Confidence
Confidence
transformed
Confidence
Confidence
transformed
Confidence
Confidence
Agreement
score
Interval
Interval
scores
Interval
Interval
score
Interval
Interval
CDU/CSU 1987
13.53
-
-
9.83
-
-
14.42
-
-
18,994
18,994
100
SPD 1987
6.53
-
-
4.61
-
-
6.68
-
-
15,325
15,325
100
FDP 1987
15.68
-
-
6.61
-
-
6.84
-
-
7,870
7,870
100
Grüne 1987
5.21
-
-
4
-
-
2.9
-
-
29,356
29,356
100
CDU/CSU 1990
16.28
14.99
17.56
10.71
10.01
11.41
16.06
14.8
17.32
5,467
6,725
81.3
CSU 1990
15.95
14.97
16.93
10.84
10.3
11.38
16.3
15.33
17.27
10,341
11,897
86.9
SPD 1990
6.81
5.63
7.98
5.87
5.26
6.48
8.92
7.84
10
6,882
9,007
76.4
FDP 1990
17.21
16.59
17.84
7.63
7.34
7.91
8.25
7.72
8.77
27,846
37,399
74.5
Grüne 1990
1.28
-0.2
2.76
2.77
2.01
3.53
1.51
0.11
2.9
4,428
5,794
76.4
PDS 1990
7.93
6.86
8.99
3.59
3.09
4.09
1.44
0.52
2.35
10,229
13,830
74
Coalition 1987
9.56
8.1
11.02
6.42
5.68
7.16
7.65
6.34
8.97
5,660
7,395
76.5
words
words
(in %)
Coalition 1990
10.59
9.6
11.59
6.49
5.99
6.99
8.24
7.36
9.12
11,621
14,623
79.5
MIT
15.16
13.36
16.95
8.5
7.61
9.38
10.96
9.37
12.56
3,463
4,191
82.6
CDA
8.02
6.35
9.69
4.77
3.95
5.59
5.44
3.98
6.9
4,109
4,961
82.8
AfA
9.87
8.08
11.66
5.7
4.83
6.57
6.99
5.44
8.53
3,178
3,657
86.9
AGS
14.03
12.53
15.54
7.3
6.58
8.02
9.84
8.57
11.11
4,567
5,430
84.1
Seeheim Circle
8.57
6.96
10.17
5.88
5.08
6.68
7.82
6.39
9.25
3,871
4,525
85.5
Frankfurt Circle
1.16
-0.1
2.43
1.92
1.31
2.52
0.01
-1.07
1.1
7,336
8,991
81.6
32
Appendix B: Positions of German political parties, intraparty factions and coalition agreements, 1998-2002
Party/
Economic Policy
Intraparty
Foreign Policy
Share of
Social Policy
Total scored
Total words
scored
faction/
Reference or
95% lower
95% upper
Reference or
95% lower
95% upper
Reference or
95% lower
95% upper
Coalition
transformed
Confidence
Confidence
transformed
Confidence
Confidence
transformed
Confidence
Confidence
Agreement
score
Interval
Interval
score
Interval
Interval
score
Interval
Interval
CDU/CSU 2002
14.4
-
-
6.6
-
-
15.9
-
-
20,536
20,536
100
SPD 2002
9.3
-
-
5.5
-
-
7.3
-
-
21,273
21,273
100
FDP 2002
18.7
-
-
6.7
-
-
5.3
-
-
32,607
32,607
100
Grüne 2002
11
-
-
6.6
-
-
2.4
-
-
24,829
24,829
100
PDS 2002
3
-
-
14.2
-
-
4.9
-
-
14,031
14,031
100
CDU/CSU 1998
11.68
10.91
12.45
5.38
4.86
5.89
17.71
16.66
18.76
16,245
18,055
90.0
SPD 1998
7.51
6.7
8.31
6.76
6.2
7.33
14.04
12.98
15.1
13,838
15,003
92.2
FDP 1998
18.48
17.72
19.25
6.21
5.75
6.67
8.63
7.77
9.48
21,293
23,971
88.8
Grüne 1998
6.16
5.54
6.78
9.7
9.26
10.14
-2.17
-2.97
-1.37
30,561
35,126
87.0
PDS 1998
-2.58
-3.4
-1.75
19.74
19.13
20.35
-0.03
-0.85
0.79
23,294
26,261
88.7
Coalition 1998
9.23
8.42
10.03
7.33
6.78
7.88
8.29
7.24
9.33
14,794
16,606
89.1
Coalition 2002
8.7
8.11
9.3
5.47
5.07
5.87
6.62
5.86
7.38
24,427
27,221
89.7
MIT
18.85
17.73
19.97
4.89
4.19
5.59
6.79
5.41
8.16
10,895
12,834
84.9
CDA
12.26
11.29
13.23
7.3
6.65
7.94
5.48
4.24
6.72
10,261
13,226
77.6
CSA
13.23
11.71
14.76
6.85
5.83
7.87
14.17
12.16
16.18
4,467
5,010
89.2
AfA
10.85
9.7
12
8.54
7.77
9.3
6.06
4.82
7.31
9,407
10,694
88.0
words
words
(in %)
AGS
14.28
13.56
15
6.02
5.55
6.5
6.99
6.11
7.87
23,262
27,653
84.1
Seeheim Circle
11.46
9.96
12.96
8.3
7.31
9.3
5.27
3.62
6.92
5,668
6,397
88.6
DL 21
6.26
4.58
7.94
10.27
9.07
11.47
5.21
3.34
7.08
4,260
4,952
86.0
FDP-MIT
23.62
21.49
25.74
4.8
3.58
6.03
13.29
10.79
15.8
2,855
3,273
87.2
FDP-FR
11.3
8.81
13.79
10.48
8.87
12.1
2.25
-0.41
4.91
2,166
2,626
82.5
GR-ECO
13.73
12.35
15.1
5.78
4.85
6.72
4.5
2.84
6.16
5,984
6,737
88.8
33