Hierarchical, Decentralized, or Something Else? Opposition Networks in the German Bundestag

• Author: Sebastian Jäckle,Thomas Metz
• Published date: 01 May 2016
• DOI: http://doi.org/10.1111/lsq.12122
• Date: 01 May 2016

THOMAS METZ

SEBASTIAN J€

ACKLE

University of Freiburg

Hierarchical, Decentralized,

or Something Else?

Opposition Networks

in the German Bundestag

Members of the German parliament may force government to publicly answer

questions by issuing minor interpellations (kleine Anfragen). We use 3,608 interpella-

tions from the session 2009–13 that have been signed by authoring and supporting

members to construct the social network of support relations among members within the

three opposition parties. We find that parties differ markedly in terms of internal struc-

ture. While social democrats organize hierarchically, Greens cooperate horizontally. The

network for socialist Linke in contrast shows signs of homophily and social segregation.

Our approach yields a novel perspective on intraparty politics in parliamentary systems

which are notoriously difficult to analyze.

Introduction

In politics, You are who You know. While institutions and strategy

make up an essential part of collective organizations, personal connec-

tions are at least equally important since whatever the rules of the game

are, it is impossible to understand outcomes without a sense of who inter-

acts with whom to produce them. In this perspective, politics is a

network. Within democracies, parties are the essential tie connecting citi-

zens with government (and opposition). Analyses of “making and

breaking” (Laver and Shepsle 1996) coalition governments have only

until recently started to include intraparty politics in their models, relax-

ing the unitary actor assumption (see Giannetti and Benoit 2009).

1

Yet,

measuring intraparty differences remains a problem, particularly in

parliamentary systems such as Germany where strong party discipline

makes it diff‌icult to extract factions based on, for example, roll-call votes.

The accessibility of empirical intraparty data also hampers the analysis

LEGISLATIVE STUDIES QUARTERLY, 41, 2, May 2016 501

DOI: 10.1111/lsq.12122

V

C2016 Washington University in St. Louis

of candidate-selection models (Hazan and Rahat 2006, 117) and, more

generally speaking, a more profound study of intraparty democracy (see

Cross and Katz 2013).

In this study, we introduce social network analysis as an alternative

possibility to partially open the black box of the party, focusing on

organizational aspects that structure intraparty politics. To do so, we

investigate the social network among members of the opposition in the

German parliament during the legislative term 2009–13. We infer social

connections by tracking which members supported other members’ par-

liamentary requests for information, so called Minor Interpellations

(kleine Anfragen), or MI for short. These interpellations are an important

means of the opposition to raise public attention, criticize government,

and try to muster public support for the next election. Since they have to

be signed by either a parliamentary group or at least 5% of the members

of the Bundestag, MIs require parliamentarians to work together. Map-

ping out this cooperation as an interpersonal network of intraparty

support offers a unique window into who cooperates within the parlia-

mentary groups, who is responsible for conf‌lict with government, and

who plays a central role in expressing the party’s agenda. We assume

the structuring principles for the MI support network not only to

determine the work on these interpellations, but the work within the

opposition more generally.

The question of how parliamentarians in the opposition work

together is important from at least three vantage points: First, in a parlia-

mentary system like Germany, the temporal alternation between

government and opposition is one of the main elements in the separation

of powers. Therefore, it is important to understand the opposition’s inter-

nal organization, both in terms of internal working and structures of the

different parties and in terms of the cooperation between the parties on

an aggregate level. Second, coming from an elite perspective which

mainly focuses on those in power, the idea of an alternation between

government and opposition raises the interesting question of who is cen-

tral in this powerless elite, framing a government in waiting. And, third,

and most importantly: not much is known about intraparty politics in

general. While intraparty differences are relevant for many subjects rang-

ing from coalition formation to public policy and candidate selection,

their empirical measurement is still in its infancy. The problem is espe-

cially salient for parliamentary systems where strong party discipline

foils the application of methods developed for presidential systems

(such as roll-call votes). One way to approach this dilemma are auto-

mated text-analysis techniques such as Wordf‌ish (Slapin and Proksch

2008) or Wordscores (Laver, Benoit, and Garry 2003). Applying the

502 Thomas Metz and Sebastian J€

ackle

latter method to speeches of Irish legislators and German MPs, Laver

and Benoit (2002) and Bernauer and Br€

auninger (2009) analyze the

ideological intraparty heterogeneity in Ireland and Germany, respec-

tively. Hanna B€

ack presents another possibility: she conducted a survey

asking representatives in Swedish municipalities about factionalization

and intraparty democracy (B€

ack 2008, 79–80).

Our approach is different, as we do not focus on the ideological

differences within a party (unfolded, for example, in speeches of MPs)

but on the way social contacts of parliamentarians organize opposition

work in the Bundestag and structure the internal functioning of the par-

liamentary group. Such a perspective on intraparty organization also

highlights important implications in terms of democratic theory: while a

party that is only loosely organized internally is probably a more friendly

territory for diverging viewpoints, a stronger organization (especially

when coupled with a hierarchical structure) would indicate the party’s

possibility to pose a clearer alternative to voters, although at the price of

reduced diversity. Taking a more output-oriented perspective, the clarity

of intraparty structures could not only transport a party’s external appear-

ance as a corporate entity in general but also indicate in a wider sense its

potential to formulate and attain policy goals.

Our analysis is to some extent an inductive and descriptive

endeavor. We want to show which parliamentarians work together, what

the complete network of cosignees looks like, and whether specif‌ic pat-

terns and structures become apparent within it. To do so, we will f‌irst

justify our approach, outline existing work in the literature, and state

four preliminary hypotheses. We will then describe our data and our

method. After that, we will concentrate on aggregate characteristics of

the cooperation patterns. In a second step, we investigate the global

opposition network derived from the MIs and then focus on the structure

of the networks for each of the three opposition parties separately. We

discuss our f‌indings in the concluding section.

Theory and Related Work

Why Look at Minor Interpellations and What is That, Anyway?

Like in other democracies, oversight and control of the executive

are important tasks of the German parliament. Among the numerous

instruments the Bundestag has at its disposal is that members and parlia-

mentary groups may force the executive to provide information and

answer questions. Basically, there are four different ways of inquiring

2

(see Siefken 2010, 21).

503Hierarchical, Decentralized, or Something Else?