ArticlePDF Available

Self-legitimation in the face of politicization: Why international organizations centralized public communication


Abstract and Figures

International organizations (I0) have centralized their public communication to a large extent over recent decades by undertaking a broader codification of communication tasks as well as a departmentalization of these tasks within units of IO bureaucracies. The paper provides the first systematic analysis of this important development in institutional design using a novel data set on the organization of public communication in 48 IOs between 1950 and 2015. It identifies self-legitimation as a key driver of centralization in the face of increased levels of politicization, that is, public awareness and activism directed at IOs. Empirically, the study suggests that the centralization of public communication significantly increases as transnational civil society organizes and gains access to IO decision-making. Further, politicization in terms of contentious activism and public scandals substantially accounts for varying levels of centralization across IOs.
This content is subject to copyright. Terms and conditions apply.
1 23
The Review of International
ISSN 1559-7431
Rev Int Organ
DOI 10.1007/s11558-017-9287-y
Self-legitimation in the face of
politicization: Why international
organizations centralized public
Matthias Ecker-Ehrhardt
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
Science+Business Media, LLC. This e-offprint
is for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at”.
Self-legitimation in the face of politicization:
Why international organizations centralized public
Matthias Ecker-Ehrhardt
#Springer Science+Business Media, LLC 2017
Abstract International organizations (I0) have centralized their public communication
to a large extent over recent decades by undertaking a broader codification of commu-
nication tasks as well as a departmentalization of these tasks within units of IO bureau-
cracies. The paper provides the first systematic analysis of this important development in
institutionaldesign using a novel data set on the organization of public communication in
48 IOs between 1950 and 2015. It identifies self-legitimation as a key driver of central-
ization in the face of increased levels of politicization, that is, public awareness and
activism directed at IOs. Empirically, the study suggests that the centralization of public
communication significantly increases as transnational civil society organizes and gains
access to IO decision-making. Further, politicization in terms of contentious activism and
public scandals substantially accounts for varying levels of centralization across IOs.
Keywords International organization .Institutional design .Communication .
Politicization .Legitimation .Transnational civil society .Non-state access .Political
protest .Political scandal
JEL Classification C23 .D73 .D83 .F53 .M31
1 Introduction
In the context of increased levels of public awareness and contestation, many IOs have
greatly developed their central capacities for public communication over recent
Rev Int Organ
DOI 10.1007/s11558-017-9287-y
Electronic supplementary material The online version of this article (doi:10.1007/s11558-017-9287-y)
contains supplementary material, which is available to authorized users.
*Matthias Ecker-Ehrhardt
Arbeitsstelle Transnationale Beziehungen, Außen- und Sicherheitspolitik, Freie Universität Berlin,
Ihnestr. 22, D-14195 Berlin, Germany
Author's personal copy
decades. For instance, NATOs Information Service (NATIS) expanded from a small
office in the early 1950s into full-fledged Bpublic diplomacy division^staffed with
skilled public relation practitioners who address a plurality of audiences including
journalists, NGOs, expert communities, and interested citizens. Similarly, the informa-
tion offices of the IMF or ASEAN have developed into much more ambitious com-
munication departments, which are charged with reaching out to a wider range of
societal actors that have been deemed relevant for the whole organization. In order to
master the increasing complexity of communication activities, these departments have
been tasked with organization-wide coordination as well as long-term planning, mon-
itoring, and evaluation.
This development is puzzling for a number of reasons: First, centralizing public
communication implies a significant loss of control for member states over how
internal negotiations are communicated back home to national constituencies and
consequently limits the extent to which governments are able to shape domestic
perceptions of their negotiating successes. Second, the decision to strengthen public
communication as an organizational task is a tough choice given the severe budget
constraints most IOs face over the course of their existence. It implies a substantive
prioritization of communication over the other tasksthat are classically assumed to be
at the core of most, if not all, IO functioninglike the facilitation of intergovernmental
negotiations or services to states by monitoring compliance, settling disputes, or
implementing policy programs on the ground (Abbott and Snidal 1998). Third, we
observe substantial variation in the organizational field: While there has been a
remarkable centralization of public communication competences in cases such as the
IMF, the International Criminal Court (ICC), or the International Coffee Organization
(ICO), we also see IOs that have been much less ambitious in this regard, including the
North-East Atlantic Fisheries Commission (NEAFC), the Shanghai Cooperation
Organization (SCO), or the Bank for International Settlements (BIS).
Given these observations, two related research questions emerge: Beyond illustrative
examples, how much centralization of public communication do we observe across the
organizational field and over time? Moreover, what are the causal conditions that explain the
reallocation of organizational resources that is required to centralize public communication?
It is striking that public communication as an organizational feature of IOs has
hardly been the subject of any systematic investigation to date. Mainstream IR institu-
tionalism has more or less ignored public communication as a remarkable aspect of
Bdesign^(Koremenos et al. 2001), while existing research shows a number of lamen-
table deficiencies. For one, what we currently know comes exclusively from single case
studies that mostly examine changes in EU public communication over time
(Brüggemann 2008; Altides 2009; Meyer 2009; Nissen 2010; Biegoń2013; Altman
and Shore 2014); there are only a few additional studies on other regional or global IOs
(Alleyne 2003;Lehmann1999; Aghi and McKee 2000;Dimitrov2014; Servaes 2007;
Risso 2014; Gronau 2015). What is more, virtually all of these studies have limited
themselves to describing and normatively evaluating communication as a tool for
enhancing public transparency (Brüggemann 2008) or in the context of specific policy
programs (Lehmann 1999). A few remarkable exceptions have recently looked into
questions of IO public communication by conducting research on how IOs self-
legitimize in the public realm; however, again there is a strong focus on a few
prominent cases (Dingwerth et al. 2015; Gronau and Schmidtke 2016). It follows that
Ecker-Ehrhardt M.
Author's personal copy
we still lack systematic evidence as to how and why IOs develop central capacities for
organized public communication across cases.
Against this backdrop, the paper makes a number of important contributions to
existing research: First, it theorizes IO public communication by focusing on two
dimensions of structural differentiation: the internal explication of communication tasks
and the departmentalization of these tasks in communication units. Second, this paper
uses a novel dataset on public communication structures to add to our empirical
understanding of how IO public communication varies over time; this dataset makes
it possible to systematically analyze the centralization of public communication across
a stratified random sample of 48 IOs between 1950 and 2015.
Third, the paper further contributes to our understanding by theorizing and empir-
ically testing self-legitimation as a possible driver of centralization vis-à-vis alternative
explanations. Its central claim is that public communication largely aims at self-
legitimating IO policies and procedures in the face of politicization, that is, public
awareness and activism directed at IOs (Zürn et al. 2012;GronauandSchmidtke
2016). In the following, I elaborate on three observations that provide empirical
evidence for that claim. Each is based on different ways politicization can be expected
to manifest empirically and change the organizational demand for self-legitimation. For
one, a need for IO self-legitimation is assumed to significantly increase with the
transnationalization of civil societyin terms of the increased articulation of societal
demands directed at IOs as well as the organization of such demands in a transnational
sphere of non-state activism. By the same token, the need for self-legitimation is
assumed to increase with greater public contestation in the form of contentious activism
addressing IOs in the public realm as well as scandals triggered by allegations of gross
misconduct by IO leaders or staff. By implication, the fact that centralization goes along
with higher levels of institutional access granted to transnational civil society provides
correlational evidence for the asserted linkage between politicization, an organizational
drive for self-legitimation, and decisions to centralize public communication. In addi-
tion, the observation of substantial centralization in the context of public protest and
scandals further suggests that self-legitimation might play a key role in centralizing IO
public communication. Remarkably, self-legitimation provides a consistent explanation
for centralized IO public communication, while the empirical support for alternative
factorssovereignty costs, IO mandates for governing social change, and the demo-
cratic Bupload^of transparency normsis far more mixed, which further adds confi-
dence in the validity of main argument developed in this paper.
I begin by conceptualizing the centralization of public communication and mapping
its variation over time. In the second part, I theorize the politicization-legitimation
linkage. The third part of the paper puts these explanations to the test empirically by
using negative binomial regression analysis. The fourth part concludes by laying out a
number of implications for further research.
In the following discussion, I refer to Bpublic communication^as organizational structures
that enable IOs to regularly communicate with nongovernmental audiences, including
media organizations, experts, lobby groups, movements, and laypersons. This definition is
Self-legitimation in the face of politicization: Why international...
Author's personal copy
meant to encompass a diverse set of concepts applied elsewhere in International Relations,
like Bpublic diplomacy^(L'Etang 2009; Altman and Shore 2014), Bdevelopment
communication^(Servaes 2007)orBpublic information^(Coldevin 2001;Brüggemann
2008). What public communication shares with these concepts is a similar interest in the
organization of Bboundary spanning^(Aldrich and Herker 1977)activitiessuchasthe
production and channeling of text, audio, or video material to media organizations, the
organization of public symposia, workshops, radio programs, websites, or social media
activities. However, in contrast to these alternative concepts, the notion of Bpublic
communication^is deliberately agnostic regarding the specific goals of communication
and the normative standards communication may or may not live up to.
Further, I focus on communication activities that send messages to nonstate publics
in a Btop-down^manner, that is, to those at the receiving end of international authority
(Zürn et al. 2012). In this way, the study of public communication complements an
important strand of research that has recently addressed institutionalized ties with
nonstate actors under the heading of Bcivil society participation^or Btransnational
access^. Such concepts typically refer to institutional mechanisms whereby nonstate
actors Bmay take part in the policy process of an IO^(Tallberg et al. 2013:25).While
some scholars have stressed symmetric communication as a desirable, but rarely found,
aspect of institutional access (Steffek et al. 2008), this research arguably focuses on
Bbottom-up^processes of political participation. I suggest understanding the study of
Bpublic communication^and Btransnational access^as necessary complements of a
comprehensive research agenda that aims at better understanding the interlinkages
between IOs and nonstate environments.
Finally, my primary concern in this paper is the centralization of public communi-
cation, that is, the development of central administrative capacities tasked with regu-
larly communicating with non-governmental audiences. According to Abbott and
Snidal, the centralization of institutional tasks is one of the main characteristics that
distinguish IOs from other forms of international institutions that lack Baconcreteand
stable organizational structure and an administrative apparatus managing collective
activities^(Abbott and Snidal 1998: 9). Understanding whether and why important
tasks are performed by specific administrative units is at the core of IO studies
(Koremenos et al. 2001: 771). In this tradition, I understand the centralization of IO
public communication as having two dimensions, namely (a) the codification of
communication tasks assigned to the IO central administration (including those spec-
ifying target audiences as well as management tasks) and (b) the departmentalization of
assigned tasks into administrative units.
To start with, the (a) codification of public communication as a core objective of IO
administration indicates the delegation of a general competence for communication for
Concepts such as Bdevelopment communication^(Servaes 2007)andBpublic diplomacy^(Altman and
Shore 2014) reflect framing strategies of communication practitioners, which seek to legitimize communica-
tion as a governance tool. Similarly, the concept of Bpublic information^is far too narrow as it has been
predominantly used by political actors as well as academics to signal a strong focus on communication as a
mechanism to enhance IO transparency and accountability (UNGA 1946; Brüggemann 2008). In the same
vein, my definition comes close to the professional understanding of public relations as Bthe management of
communication between an organisation and its publics^(Grunig and Hunt 1984:68). However, as a
professional label Bpublic relations^is strongly associated with professional standards such as
Bexcellence^(Grunig 1992), which are problematic for analyses that are meant to include organized IO
communication regardless of professionalism.
Ecker-Ehrhardt M.
Author's personal copy
the whole IO. However, the range of more specific tasks to be performed to achieve this
general objective qualifies the centralization of public communication in important
First of all, we observe a significant variation across individual IOs and over time in
how the communication with specific target audiences is codified in terms of specific
tasks. While classic IO public communication tends to be broad and generally focused
on addressing Bthe public^via reports and press releases (seeking coverage in mass
media outlets), the more focused targeting of specific stakeholdersincluding layper-
sons, journalists, experts, lobby groups, representatives of affected communities etc.
has increased in importance over recent decades (Dingwerth et al. 2015).
Moreover, the notion of public communication has been refined in several IOs to
explicitly include several management tasks.Reflectingamainstreamingofcommunica-
tion competences across the organization, such management tasks first of all include the
coordination of different parts of the organization that interact with nonstate audiences.
Further tasks related to the management of communication activities include the strategic
planning of communication activities in the form of a long-term programming of objec-
tives and responsibilities and the systematic research into audiences in order to evaluate
the effectiveness of communication activities. Illustrating the increasing relevance of
management tasks, the executive orders on public communication issued by the
Organization of American States (OAS) have expanded over recent decades to cover
strategic planning (Bthe development and of advertising strategies on key issues as well
as communication plans and strategies for the whole organization^), coordination
(Bassistance to all areas of the Organization indesigning,organizing,andstructuring
external website content and for the use of virtual social networks^)andsystematic
research on resonance (Bmonitoring of principal media outlets of the American hemi-
sphere to evaluate OASmedia impact,^OAS 2012:67).
Turning to the second dimension, however, the centralization of public communi-
cation becomes manifest and organizationally effective only in (b) the departmentali-
zation of codified communication tasks. Here, we find significant differences in how
assigned tasks are allocated to units of varying organizational weight. There is not only
variation in budgets and staff sizes, but also in form and location across the organiza-
tional chart. More specifically, once the head of the communication unit gains access to
the top tier of organizational management, this implies a significant empowering of
public communication (as well as Bexcellence^in terms of public relations theory, cf.
Grunig 1992). To illustrate, the long-term development of NATOsInformationService
(NATIS) from its peripheral status as a subunit of the Political Affairs branch in the
1950s to the current Public Diplomacy Division (PDD) located at the main department
level and headed by an Assistant Secretary-General may be an uncontroversial case of
enhancing centralized public communication by including communication managers
into NATOs top-level stratum of civilian leadership (Risso 2014).
Former orders to compare with include OAS 1977;OAS1982;OAS1992;OAS1997.
Note that the Bcentralization^of public communication as discussed here does not imply or even suggest that
Bdecentralized^capacities to communicate on the level of local offices or projects are unimportant or even
reduced. On the contrary, where significant organizational resources are spent on public communication at the
project level, such spending typically calls for the enhancement of structures at headquarter level in order to
allow for proper coordination, planning, and programming of communication efforts Bon the ground^
(Coldevin 2001; Servaes 2007).
Self-legitimation in the face of politicization: Why international...
Author's personal copy
My measure of the centralization of public communication combines both dimen-
sions in a multiplicative index, which weights codification (the range of observed
communication tasks assigned to IO central administration) by departmentalization
(the degree to which these tasks are matched by organizational capacities).
Centralization ¼Codification xDepartmentalization
Turning to operationalization, the degree of codification was measured by a count of
tasks as identified in IO documents. Six task were coded, three concerning the
communication with specific categories of publics (general public, expert publics, other
nonstate publics) plus three management tasks (coordination, research, strategic
planning; see Table A.2 in the online appendix for the operational definition and
illustrative examples). The coding was based on a comprehensive collection and
reading of relevant documents, including all annual reports, budget proposals, hand-
books, communication plans, and published case study research by other scholars if
available. Where necessary, I contacted IO secretariats directly. Due to severe restric-
tions on information regarding the organizational resources explicitly allocated for
public communication (budget, staff), I used the location of the main communication
unitif existingin the administrative hierarchy as a proxy for departmentalization.
Accordingly, departmentalization is measured on a three-point scale, where the first
point indicates there is no central administrative unit predominantly responsible for
public communication (score = 0), the second point indicates a communication unit that
is a subordinate to a main department (score = 1), and the third point indicates a
communication unit with direct access to the head of administration (either as a staff
unit directly connected to the head of administration or a main department in the line;
score = 2). The resulting index theoretically ranges from 0 to 12.
Empirically, the analysis focuses on 48 IOs that are (1) formal entities with (2) a
permanent and separate secretariat, administration, or headquarters, and (3) a set of
principals consisting of at least three states whose delegates are members of central
decision-making bodies in the respective IO.
The selected IOs come from a stratified-
random sample drawn by Tallberg and colleagues, including general as well as issue-
specific IOs and global as well as regional ones (Tallberg et al. 2013). From their
sample of 50 IOs, only 48 turned out to be sufficiently accessible to code the core
features of organized public communication for most years of the IOs existence (see
Appendix Table A.1 for a comprehensive list of the IOs covered).
As can be seen in Fig. 1, we predominantly observe an increase in centralization
over time. In fact, the levels of centralization observed in 2015 were higher for 31 IOs
than the corresponding levels at the beginning of the period studied. However, signif-
icant variation across IOs remains in terms of the timing and intensity of centralization.
We find cases with singular leaps in centralization, as in the case of the Northwest
Criterion 1 and 2 distinguish IOs from temporary institutions and conferences created on an ad hoc basis and
without the necessary capacity to act autonomously of their principals; through criterion 3 all those organi-
zations are set aside that are predominantly non-governmental and bilateral institutions as well as mere
emanationsof existing IOs. Pevehouse et al. 2004; Wallace and Singer 1970.
The two IOs that were not included due to a lack of reliable information about public communication
structures are the Andean Community and the Fund for the Development of Indigenous Peoples in Latin
America and the Caribbean.
Ecker-Ehrhardt M.
Author's personal copy
Atlantic Fisheries Organization (NAFO) but also much more gradual changes over
longer time spans, as in the case of NATO or the Council of Europe (CE). Finally,
several cases show constant levels of centralizationvarying from complete absence of
centralized communication capacities, as in the case of the North American Free Trade
Fig. 1 Centralization of public communication over time
Self-legitimation in the face of politicization: Why international...
Author's personal copy
Agreement (NAFTA), to constantly high levels of centralization, as in the case of the
United Nations (UN). Such range of case-specific variation deserves a theoretically
sound explanation, to which I turn in the next section.
3 Theory: Self-legitimation in the face of politicization
How can the observed variation be accounted for? Existing evidence suggests under-
standing public communication as anorganizationalfunctionthataims to self-
legitimate IO policies and procedures in times of politicization, that is, when public
awareness and contestation of IOs is high (Zürn et al. 2012; Dingwerth et al. 2015;
Gronau and Schmidtke 2016).
Research has provided ample evidence for the empirical significance of politiciza-
tion. Public opinion studies now reveal a remarkable attitudinal awareness of major
global and regional IOs in terms of structured beliefs and evaluations accessible to
citizens across continents (Norris 2000; Boomgaarden et al. 2011; Johnson 2011;
Dellmuth and Tallberg 2015; Dellmuth 2016). Scholars of social movements and
NGO advocacy have extensively studied how IOs have become a new focus of
transnational activism (Keck and Sikkink 1998; O'Brian et al. 2000; Tarrow 2001;
della Porta 2007). Similarly, research on parliamentary debates (de Wilde 2011), party
manifestos (Ecker-Ehrhardt 2014) and mass media communication (Bennett et al.
2004; Nullmeier et al. 2010) suggests that a couple of prominent IOs have become
major reference points of political discourse.
Why should we care? To start with, politicization is desirable for normative reasons
and therefore worthy of a systematic analysis of its causes and consequences. While the
shift of regulatory competences to IOs is widely held to constitute a desirable compo-
nent of regional or even global democratization (Held 2002; Archibugi 2008), a process
of politicization is a necessary condition for the emergence of a global citizenry that is
willing and capable of holding such new centers of international authority accountable
(Falk 1995; Urbinati 2003).
What is more, politicization is of enormous practical relevance for IOs work. One
reason for this is, first, that politicizationhascomplexrepercussionsforthelikeli-
hood of reaching agreements inside intergovernmental bodies and for the likelihood
of gaining compliance with obligations that derive from such agreements in a two-
level polity (Putnam 1988). Relatedly, public awareness and contestation can sig-
nificantly change the political dynamic between governments and IO bureaucra-
ciesincluding tensions that derive from bureaucratsattempts to interfere in
domestic debates in order to influence governmental policies (e.g., Seabrooke
2007:264;Rauh2012). Lastly, IOs that directly address or even regulate societal
processes need public support in order to fulfil their mandates (Zaum 2006). For
better or worse, politicization increases the organizational need to manage societal
contestation in order to avoid involuntary defections by governments (Odell and
Eichengreen 1998), organizational crises induced by the withdrawal of resources
(Cárdenas 2000;Smith2004), or immediate societal resistance that obstructs oper-
ations on the ground (e.g., Moulin and Nyers 2007). Thus, nonstate actors constitute
relevant social constituencies that have to be addressed and accommodated in order
to make many IOs work smoothly.
Ecker-Ehrhardt M.
Author's personal copy
More specifically, observed politicization suggests that IOs increasingly require
social legitimacy, that is, a Bgeneralized perception or assumption that actions are
desirable, proper, or appropriate^(Suchman 1995: 574). In order to gain lasting support
from societal actors, IOs are in desperate need of normative recognition as legitimate,
given that IOs notoriously lack alternative sources of social power such as economic
incentives or disciplinary force (Hurd 1999;Steffek2007; Symons 2011;Bernstein
2011). While systems of political rule show a general tendency Bto cultivate the belief
in its legitimacy^(Weber 1978: 213), politicization drives organizations to engage in
strategic communication in order to manage legitimacy.
In fact, the strategies and mandates of major public communication departments in
the IO field aim explicitly at generating legitimacy. Tellingly, a core aim of the World
Bank External Affairs department as well as NATOs Public Diplomacy Division is to
increase Bpublic support^for their organizations (WorldBank 2011:41;NATO2009:
2). APECsBCommunications and Public Affairs Strategic Plan for 2014-16^asks its
communications team to Bidentify and highlight APEC success stories and concrete
benefits and distribute these through appropriate vehicles^(APEC 2014: 3). In the
same vain, the OAS Department of Press and Communications has been explicitly
tasked with Bproject[ing] the image of the OAS as a hemispheric forum for policy
discussion with a presence in the political dealings of the member states and a
meaningful role to play in the solution of any crises that occur in the Americas and
the Caribbean^(OAS 2006:26).
While such references to self-legitimation might already suggest its empirical
significance with regard to the internal discourse, what are the testable implications
of its purported causal role in bringing about centralized public communication over
time and across IOs? In the following, I focus on three observations that should provide
conclusive evidence for that claim, each based on how different aspects of politicization
can manifest empirically and plausibly change the organizational demand for self-
legitimation. If self-legitimation drives centralization, I argue, centralization should
empirically arise (1) along with the formation and inclusion of transnational civil
society, (2) in the context of contentious politics, and (3) where IOs have recently
experienced publicly salient scandals. I discuss each of these observable implications of
the proposed politicization-legitimation linkage in turn.
3.1 Transnational civil society
The need for self-legitimation vis-à-vis nonstate audiences should first of all increase
with politicization in the form of a general Bwidening of the audience or clientele
interested and active^(Schmitter 1969, 166) in international governance. Such a
Bwidening^implies what many observers have repeatedly described as a
Btransnationalization^of civil society over recent decades: the formation of societal
demands that frame political issues as Btransnational^and the articulation of such
demands by Btransnational^associations and networks that increasingly get access to
international arenas (Anheier et al. 2001).
Regarding the formation of transnational demands, public education and first-hand
experience have widely expanded the societal understanding of interconnectedness in
terms of Bextended relations connecting them to other peoples, places and
environments^across borders (Szerszynski and Urry 2002: 472, similarly Hannerz
Self-legitimation in the face of politicization: Why international...
Author's personal copy
1990). What is more, such cognitive transnationalization of Bpeoples horizons^
(Rosenau 2003:52)seemstohavehadahighlycontingentimpactonpoliticalattitudes
and discourses by facilitating cosmopolitan narratives of transnational belonging and
normative responsibilities in some parts of transnationalized societies (Norris 2000;
Furia 2005; Mau et al. 2008) as well as ethnocentrism, chauvinism and xenophobia in
other parts of these societies (Beck 2004). Empirical research suggests that such diverse
readings of interconnectedness have become increasingly consequential for IOs by
producing a plurality ofin parts highly conflictingviews on international gover-
nance: In terms of a Bcosmopolitan politicization^it seem to have significantly fostered
societal expectations regarding and support for IOs (Ecker-Ehrhardt 2012); in terms of
aBnationalist backlash,^it has also led to the formation of significant demands for
political Bdemarcation^and the Brenationalization^of political authority (Kriesi et al.
In organizational terms, such demands are articulated by a range of diverse associ-
ations and networks. What some have pointedly termed a Bglobal associational
revolution^(Polachek et al. 1999: 4) includes an extensively researched process of
organizing societal demands directed at IOs transnationally,which adds a significant
degree of complexity into IO operations (e.g., Rosenau 1995; Price 2003; Scholte
2011). Spanning a broad spectrum of ideological leanings (Bob 2012), diverse expec-
tations regarding proper ways of allocating and legitimately organizing political au-
thority further complicate interaction with a pluralizing societal context (Dingwerth
et al. 2015). A global sphere of advocacy organizations has emerged in recent decades,
Binserting themselves into a wide range of decision-making processes on issues from
international security to human rights to the environment^(Florini and Simmons 2000:
3; Raustiala 1997; Keck and Sikkink 1998;Joachim2003). A notable trend to give
institutional access to nonstate actors has significantly intensified the degree to which
transnational demands have become relevant for specific IOs. The inclusion of repre-
sentatives from transnational civil society has been shown to work as a (highly
selective) Btransmission belt^for societal demands from the bottom-up, which has
made questions of societal legitimacy much more prominent in the internal discourses
of IOs (Steffek et al. 2008; Hahn and Holzscheiter 2013; Dingwerth et al. 2015;Gronau
and Schmidtke 2016).
In any event, the flourishing and inclusion of transnational civil society has signif-
icantly increased the organizational need for centralized public communication with the
aim of more effectively managing public legitimacyif only to avoid public delegit-
imation when well organized demands cannot be met and transnational civil society is
mobilized to challenge IO decisions or actions in the public sphere. By implication, if
we are to believe in the role of self-legitimation in motivating public communication in
general and its centralization more specifically, such flourishing and inclusion of
transnational civil society should significantly account for variation in observed cen-
tralization of IO communication over time and across IOs.
Even if access is handled restrictively on many occasions, new opportunitysuch as to file written
statements tabled at important meetingsarguably increase the likelihood that significant parts of the
organization will become aware of transnational demands. Further, institutional access may significantly
empower representatives of nonstate organizations to more convincingly lobby for transnational demands,
making those demands more credible as an eminent challenge to the organizations legitimacy in its societal
environment (Joachim 2003; Steffek et al. 2008;Scholte2011).
Ecker-Ehrhardt M.
Author's personal copy
Hypothesis 1 Centralized public communication is more likely the more transnational
civil society organizes and finds access to IOsinternal processes.
3.2 Contentious politics
Second, we can assume that waves of massive public contention that challenge the
mandates, policies, or procedures of specific IOs lead to internal perceptions of a
popular legitimacy deficit and make it much more imperative to effectively manage
public discourse by improving means of communication. Scholars of social move-
ments have extensively studied a remarkable upsurge in confrontational activism
that has contested IOs in recent decades. Early examples that attracted scholarly
attention included the wave of massive protests that spread all over Europe to
contest NATOs Bdual-track^policy (Rochon 1988)andmuchmoreisolatedfood
riots that contested the implementation of IMF structural adjustment programs
during the 1980s (Walton and Shefner 1994). However, public protest against IOs
reached an unprecedented level of intensity with the emergence of the Bglobal
justice movement,^which focused on global economic institutions like the WTO,
IMF, or World Bank (della Porta 2007). Whereas in the early 1990s less than five
events addressing one of these institutions occurred per year, by 2005 the number
had risen to about 25 (Pianta and Zola 2005). While the Bglobal justice movement^
might be in decline for various reasons (Gibson 2008; Scholl and Freyberg-Inan
2013), it is widely perceived to be a very powerful contester of IO legitimacy that
has had a significant political impact on the respective institutions and international
politics it targets (Reus-Smit 2007; Dingwerth et al. 2015; Gronau and Schmidtke
2016). For example, protest activities in the context of the intergovernmental
negotiations on the OECD Multilateral Agreement on Investment significantly
contributed to its failure in 1999 (Henderson 1999). What is more, the inclusion
of nonstate actors into IO decision-making has been partly attributed to contentious
politics (Steffek et al. 2008;Tallbergetal.2014).
More importantly, case study evidence already suggests a causal link between
protest activities and the fact that public communication has become an organiza-
tional priority. In her historical analysis of the NATO Information Service, Linda
Risso gives compelling evidence of how the internal perception of Beroded public
supportat a time of important negotiations with the East^led to a far-reaching
Brethinking of the alliances information strategy^(Risso 2014:132f).Inthecaseof
the Asian-Europe Meetings, Gilson has argued that protesters have effectively
Bpushed government circles to issue more public declarations of their intentions
and to disseminate information about ASEM more widely^(Gilson 2011:216).
Similarly, Jennifer Gronau and Henning Schmidtke suggest that protest activism
against the G8 and the IMF has led both organizations to stress self-legitimation as
an operational goalincluding strategies addressing the general public (Gronau
and Schmidtke 2016: 553). Such evidence suggests that we may expect public
contestation to substantially enhance organizational demands for self-legitimation.
In terms of observable implications, I thus expect individual IOs to centralize public
communication as instances of protests that contest mandates, policies or proce-
dures increase.
Self-legitimation in the face of politicization: Why international...
Author's personal copy
Hypothesis 2 Centralized public communication is more likely after waves of public
protest against the IO in question.
3.3 Public scandals
Third, the organizational demand for self-legitimation can be assumed to substantially
increase with the frequency and public salience of scandals, which are defined as
instances of public debate where bureaucratic leaders or staff members are accused of
transgressions, that is, gross misconduct that is widely deemed unethical (cf. Thompson
2000: 12). Empirical cases that immediately come to mind include allegations of
corruption (oil-for-food at the UN), patronage (Paul WolfowitzsBRizagate^at the
World Bank), sexual harassment (Ruud Lubbers at UNCR), or organized sexual
exploitation and abuse (UN peacekeepers in Bosnia, DRC, and elsewhere). One may
add the Kurt Waldheim scandal; Waldheim had left the UN post of Secretary-General
years previously, however, Bthe affair became an international scandal, precisely
because Waldheim had been Secretary-General of the UNholding the organization
retroactively responsible for the selection of a Secretary-General with a highly dubious
moral stature^(Lehmann 2011: 7). Even in the case of Dominique Strauss-Kahn,
criminal charges concerning his behavior outside his official role led to negative
coverage of the IMF and was an BIMF scandal^to some extent.
In these and similar cases, individual behavior has been deemed damaging for the
reputation of the affected IO: in the case of Waldheim, the Bpolitics of embarrassment
worked to undermine the UNs symbolic power in the 1980s and beyond^(Lehmann
2011: 11). While the Wolfowitz presidency might be seen in retrospect as a highly
unsatisfactory Bstopgap^for the World Bank (Bazbauers 2014: 103), the BRizagate^
scandal is arguably a nadir as Bcommentators where talking about development
institutions, sex, and money in the same sentence^(Bedford 2009: xii) for months.
Arguably, the relevance of individual misconduct for organizational reputation varies
depending on the degree to which individual misconduct is perceived to indicate
systematic deficiencies (Brändström and Kuipers 2003). For example, the oil-for-
food-scandal was perceived by insiders as a threat to the whole organization, because
Bthe difficulties encounteredthe managerial weaknesses the failures to accept respon-
sibility, the ethical lapsesare symptomatic of systemic problems running through the
UN Organization^(Paul Volcker in Meyer and Califano 2006: xii). In this context, the
public salience of scandals may substantially rest on a strategically motivated framing
of these as illustrating even larger deficiencies. Those unsatisfied with specific office
holders (or the policies they represent) may play an important role in playing up
allegations in order to exploit organizational crises (Boin et al. 2009).
In any event, case study evidence illustrates how effective scandalization may have a
disruptive impact on organizational life and lead to decisions aimed at managing public
legitimacy. The oil-for-food scandal at the UN persistently attracted public blame and
added significant weight to calls for institutional reforms inside the UNin the US and
elsewherein its aftermath (Meyer and Califano 2006; Sanchez and Urpelainen 2014;
Schlagheck 2016). At the World Bank, BRizagate^led to the immediate resignation of
Paul Wolfowitz within weeks. In the case of the EU, allegations of fraud not only
triggered the resignation of the Santer Commission in March 1999 but Bwas sufficiently
Ecker-Ehrhardt M.
Author's personal copy
traumatic to induce change at various levels^including the further centralization of the
commissions public communication, for example, by strengthening the External
Communication Network (ECN) Bas a central co-ordinating mechanism^(Meyer
2009: 1059). Beyond single-case evidence, a quantitative analysis by Alex
Grigorescu (2007) suggests that public scandals significantly account for different
levels of institutional transparency across IOs.
According to this evidence, profound experiences of public scandalization seem to
substantially increase the organizational need for self-legitimation. By implication,
therefore, we should expect to find scandals to significantly account for the centraliza-
tion of public communication.
Hypothesis 3 Centralized public communication is more likely to arise after political
scandals at the IO in question.
I assess the empirical validity of the proposed politicization-legitimation argument by
means of multivariate regression. My centralization index reflects counted tasks
weighted by degree of departmentalization. This results in a discrete variable that only
includes integer values and substantially deviates from normality, with many zeros and
decreasing density with higher values. This is a characteristic distribution for event-
count dataan outcome that is highly plausible, given that the data-generating process
of centralization includes the counting of codified taskswhich suggests that the
subsequent analysis based on ordinary least squares may be problematic. What is more,
the variance of centralization (= 10.61) significantly exceeds its mean (= 3.55),
indicating that the data fits a negative binomial distribution significantly better than
Poisson (Cameron and Trivedi 1998;LongandFreese2006).
However, the multipli-
cative formula leads to notable Bgaps^in the vector of possible outcomes (including
0,1,2,3,4,5,6,8,10,12). Thus, I carefully checked the results for robustness by using
alternative specifications, namely ordered logit and tobit regression (see the online
appendix for details). To account for heteroscedasticity, I use robust standard errors
clustered on IOs. Year fixed effects are included to control for time trends in the
aggregate levels of centralization and explanatory variables across all models.
Regarding my explanatory variables, I use three sets of measures to capture
important parts of the proposed politicization-legitimation linkage. First, as outlined
above, the transnationalization of civil societyin terms of the formation, organization
and inclusion of demands addressing IOs is assumed to enhance the need for IO self-
legitimation (H1). There is still a lack of valid data to comprehensively test this
expectation, at least with regard to the proper measurement of the degree to which
transnational demands that address specific IOs form and organize. However, as argued
above, the transnationalization of civil society arguably becomes most relevant where
This is also highly plausible for theoretical reasons, as observed overdispersion (variance being larger than
the mean) can be assumed to reasonably reflect interdependencies in the occurrence of task assignments in a
given IO-year. One obvious reason of such clustering is the weighting of counted tasks by departmentalization.
Self-legitimation in the face of politicization: Why international...
Author's personal copy
IO grant access to its representatives. What is more, the inclusion of nonstate actors into
IO decision-making has been widely perceived as reflecting global patterns in the scope
and density of civil society associations in general and as capturing severe inequalities
in mobilization towards international politics across groups and world regions more
specifically (Willetts 2000; Brühl 2010; Hahn and Holzscheiter 2013). To test my first
hypothesis, I therefore draw on data on the inclusion of nonstate actors in IO proce-
dures between 1950 and 2010 as provided by the TransAccess project (Tallberg et al.
2014). The variable Transnat ional Access is a composite index that comprises infor-
mation on four dimensions of institutional access by transnational nonstate actors:
depth (level of involvement), range (range of nonstate actors entitled to participate),
permanence, and legal codification of arrangements.
The variable is lagged by one
year to account for the typical delay in reorganizing international administrations and to
address concerns of reverse causality and selection.
Second, the organizational demand for self-legitimation is assumed to increase with
higher levels of protest activities contesting the IO in question, which leads me to
expect increased levels of centralization in the context of societal protest (H2).
Regarding protest activities, I chose the Associated Press (AP) to identify relevant
events, because global news agencies have a lower threshold for identifying relevant
events and have a wider reach than international newspapers.
Each AP news piece was
carefully screened to determine to what extent it really covered a relevant act of societal
protest directed at the specific IO. Relevant information is captured by two variables:
(a) The dummy variable Protests indicates whether I found any evidence for societal
protest activities addressing the respective IO in the preceding year, yes (B1^)orno
(B0^). While it is blind to different intensities of coverage, it is expected to function as a
rather robust measure of politicization over different levels of overall public attention
for individual IOs (e.g., it should level out differences between the omnipresent UN and
the much less salient OSCE). (b) The count variable Absolute Coverage of Protest is
based on the logged number of identified AP articles on protests addressing a specific
IO for the preceeding year. I assume that this measure captures varying intensities of
public delegitimation of an IO by contentious political activities to a substantial extent.
Again, protest variables are lagged by one year to help address concerns of reverse
Third, the organizational demand for self-legitimation is assumed to increase IO
scandals, defined as instances of public debate where bureaucratic elites or staff
members of an IO are accused of gross misconduct that is widely deemed unethical.
Assuming a much longer tradition of such scandals (compared to protest activities), I
For later years (20112015), the values of the variable Trans natio na l Access are based on a search of more
recent general rules of procedures (the main kind of documents TransAccess based their coding on) and
specific documents on civil-society relations for all those IO bodies selected by Tallberg and colleagues. The
values of TNA Access in the given IO-year equals the value in 2010, unless I found evidence for the release of
such documents (indicating possible changes we could arguably not code with an acceptable level of
consistency with the TransAccess project). Where I found such changes, the variable is set to Bmissing^for
the given and later IO-years.
Exploratory analyses also suggest a more inclusive reporting of protest activities by AP compared to Reuters.
In line with the social movement literature, I assume protest activities addressing IOs to be a rather recent
phenomenon; hence the absence of access to AP content before 1977 seems acceptable. Relevant articles
published by AP were identified in the LexisNexis-database using the keywords demonstrat*and protest*
in conjunction with the mentions of the respective IO.
Ecker-Ehrhardt M.
Author's personal copy
chose a two-step approach: First, I collected available information on public scandals
using a variety of sources, including the New York Times archive and additional mass
media content provided by LexisNexis. Scandals were coded wherever I found evi-
dence for significant misconduct attributed to the IO that has received some minimal
level of public attention (or was reported as such in the respective source). Second,
additional data was collected on how many articles were published by the New York
Times (NYT) covering a specific scandal in order to approximate its degree of public
salience. Again, two variables were constructed: (a) The dummy variable Scandals
indicates whether there was evidence for a public scandal attributed to the respective IO
for the preceding year; (b) the count variable Absolute Coverage of Scandals equals the
logged number of identified NYTarticles on scandals attributed to the respective IO for
the preceding year.
Existing theories and evidence hint at a number of alternative factors,whichare
important to control for in order to fully test the validity of the politicization-
legitimation argument. First, the centralization of public communication could
significantly be affected by the level of sovereignty costs anticipated by member
states: By centralizing communication, member states face a significant loss of
control over how internal negotiations are communicated back home to national
constituencies; this limits the extent to which governments are able to shape
domestic perceptions of how successfully they have negotiated to the benefit of
domestic interests (Putnam 1988). Security issues may be perceived by states as
being much more sensitive in this regard than, for example, technical standardiza-
tion or development aid. Therefore, IOs concerned with security may be expected to
show less centralization than others.
Second, centralization of public communication may reflect expectations of
more effectively governing societal actors by communicative means. In particular
where IOs are designed to function as socializing agents (or Bnorm teachers^)vis-à-
vis societal discourses (Barnett and Finnemore 2004) or to implement projects on
the ground (Zaum 2006), more capable public communication may effectively
reduce the costs of political intervention, thus making centralization a rational
choice of institutional design from the perspective of member statesinterests (cf.
Abbott and Snidal 1998;Avantetal.2010). Regarding varying mandates for
governing social change directly, existing case studies have predominantly demon-
strated the relevance of public communication for teaching norms and knowledge
with regards to human rights and development or health issues (Aghi and McKee
2000; Coldevin 2001; Defourny 2003; Servaes 2007; Mefalopulos 2008;Odugbemi
and Lee 2011). Therefore, I control for the relevance of different issue areas for the
individual IO by employing a set of variables, that measures to relevance of each
issue area for the individual IO. Ranging from 0 to 1, the variables Development,
Environmental, Human Rights, Finance, Technology, Trade, Security Area,
Commodities equal the proportion of a single issue area to all the other issue areas
in which an IO is active. The coding of issue areas is based on data provided by the
replication data set of Tallberg et al. (2014). Further, IO public communication may
be expected to become more centralized over time in those IOs that have an
operational mandate to implement policy programs that aim to create social change
on the ground, i.e., that seek to intervene directly in local communities. This
relevance of local implementation is controlled for by introducing the variable
Self-legitimation in the face of politicization: Why international...
Author's personal copy
Local Implementation. The variable builds on information provided by Tallberg
et al. (2014), who coded the relevance of local activities for specific IOs over time
using the description of tasks in official documents and self-presentations of main
IO bodies.
Third, it may be argued that the centralization of public communication reflects
changing norms regarding legitimate governance on the international level. Such
changes are widely debated as possible drivers of IO opening towards societal
actors in the form of procedural inclusion (Steffek et al. 2008;Liese2010;Tallberg
et al. 2013) as well as complaint mechanisms (Heupel et al. 2015). The prime
suspect with regard to public communication is institutional transparency, which
directly concerns how IOs organize their flow of information from the inside out
(Florini 2007). Democratic member states are widely expected to be the main
drivers of organizational change by Buploading^democratic normssuch as insti-
tutional transparencyto international institutions (Grigorescu 2007). Therefore, I
control for different levels of democracy among IO members by introducing the
variable Democratic Membership,whichistheone-yearlaggedmeanscoreof
democracy institutionalization in a given IO membership. Information on IO mem-
bership from the most recent version 2.3 of the COW-2 International Organizations
Dataset (Pevehouse et al. 2004) was updated for the year 2006 to 2014. The mean
scores of democracy institutionalization of all member states per IO were calculated
using Polity IV data (variable Bpolity2,^ranging from 10 to 10; cf. Marshall et al.
Fourth, I control for the possibility that small IOs might not be able to pursue
certain kinds of reforms and may thus be less likely to reform public communi-
cation. Empirically, public communication offices and the publication of commu-
nication strategies can even be observed in tiny IOs such as EUROMET (with an
annual budget of only 200,000 euros in 2010 according to data provided by
Tallberg and colleagues). However, descriptive analyses suggest minimal thresh-
olds for fully codifying tasks.
Accordingly, I control for a minimum size that
might constrain an IOsability to centralize public communication to some degree
by including the variable Minimal Budget that indicate a budget smaller than one
million euros.
Finally, one may suspect that long-term changes in the development of new
communication technologies and organizational scripts (Meyer et al. 1987;Drori
et al. 2006) have a significant impact on the organization of IO public commu-
nication. However, as mentioned above, year fixed effects are included in all
models to account for changes in these and possible other confounding factors
over time.
To allow for the estimation of data points beyond 2010, I proceeded as follows: Information was gathered
about changing tasks of existing IO bodies or the establishment of newones after 2010. For years in which no
evidence of relevant changes could be found, the values of Local Activities in the given year equals its value in
2010. Where such changes were found, the variable is set to missing.
More specifically, I do not observe the codification of Bresearch^as a communication task in the five
smallest IOs (in terms of budgets in 2010), that is, NEAFC, EUROMET, ACSO, IKSMS, and SACEP. The
smallest IO in the sample codifying research as a communication task is NAFO, with an annual budget of
about 900,000 Euro in 2010, which might indicate that budget constraints effectively prevent small IOs from
defining Bresearch^as a communication task.
Ecker-Ehrhardt M.
Author's personal copy
5 Findings
What conditions lead IOs to centralize public communication? The analysis provides
significant support for the argument that politicization is a main factor that brings about
centralization in IO public communication. Turning to the results of the various
negative-binomial regressions presented in Table 1, the estimated relationships are
consistently positive and highly significant over alternative specifications of models.
To ease the interpretation of results, Table 2reports marginal effects in terms of factor
change and percentage change coefficients. While there is a lack of comprehensive data
on the actual formation and organization of transnational demands across time and IOs,
the results first of all suggest that granting access to those making demands can account
for substantial variation in the centralization of public communication, as expected
(H1). Estimates of Model 9 point to a change in predicted centralization by a factor of
1.41 (or 41%) per standard deviation of Transnational Acces s (Table 2). Fig. 1depicts
the effect of Transnational Access using simulated values of Centralization, which
range from 1.9 to 9.8 over the full range of Transnational Access aremarkable
increase by a factor of 4.7 (Fig. 2).
To illustrate, the IOs that have most comprehen-
sively granted access to nonstate actors recentlyboth in terms of range of organiza-
tions as well as granted opportunities for influencing decisionsinclude the
Organization of American States (OAS), the International Criminal Court (ICC) and
the Council of Europe (CoE); these three IOs have also moved towards the most
centralized public communication (with value of 12 in case of the ICC and OAS, and a
value of 10 in case of the CoE) observed in this sample of 48 IOs. Of course, this is not
conclusive evidence of a causal relationship; however, such correlations over time and
across IOs arguably match the theoretical expectations that attribute observed central-
ization to the increased desire for self-legitimation in cases where transnational civil
society organizes and intrudes into IO decision-making.
Regarding the estimates for observed protest addressing a specific IO, the expected
level of centralization increases more moderatelybut still substantiallyby a factor
of about 1.60 (or 60%) when IO-Years with protest activities (as measured by the
simple indicator variable Protest) are compared to those without protest (Model 6). If
varying levels of protest coverage are taken into account (Model 8), the expected level
of centralization increases by a factor of 1.14 (or 14%) for a standard deviation increase
of observed coverage. The simulated values of Centralization ranges from about 2.6 for
IO-Years without protest coverage to 6.2 for IO-Years with maximum levels of
observed coverage, which equals an increase over the range of observed protest
coverage by a factor of about 2.6 (Fig. 1). To illustrate, in the context of massive
public contestation during the 1980s, NATO repeatedly reorganized its public commu-
nication, first, by upgrading its Press Service to a staff unit (implying with direct access
to the Secretary General) in 1984 and secondly by establishing the NATO Office of
Information and Press (NATIP) in 1988. In the same vein, ASEANspublic
Log-likelihood tests of the dispersion parameter of all models presented indicate that Poisson models would
be inferior because of overdispersion.
Simulations use the program CLARIFY (Tomz et al. 2003) and are based on a code provided by Braidwood
(2012). Please note that Figures 2,3and 4stem from two slightly simplified models (with biannual fixed
effects) to allow the inclusion of all other control variables and clustered standard errors (Figures 2and 4are
based on Model 10, Figure 3on Model 11, both fully reported in the online appendix, Table A.4).
Self-legitimation in the face of politicization: Why international...
Author's personal copy
Tab l e 1 Negative Binomial Regression of Public Communication Centralization
Model 1 2 3 4 5 6 7 8 9
Transna tional Ac cess 1.473** 0.975* 1.225** 1.022* 1.270**
(0.477) (0.402) (0.434) (0.409) (0.443)
Protest (0/1) 0.693*** 0.468** 0.377*
(0.185) (0.144) (0.159)
Scandal (0/1) 0.977*** 0.324* 0.611***
(0.224) (0.127) (0.176)
Protest Coverage 0.344*** 0.228** 0.183*
(0.096) (0.075) (0.080)
Scandal Coverage 0.451*** 0.127* 0.268**
(0.135) (0.059) (0.101)
Democratic Membership 0.040 0.000 0.021 0.001 0.022 0.015 0.031 0.016 0.032
(0.031) (0.027) (0.032) (0.027) (0.032) (0.026) (0.029) (0.026) (0.030)
Local Implementation 0.865 0.780 1.049 0.731 1.090 0.632 0.796 0.602 0.790
(0.579) (0.542) (0.686) (0.550) (0.702) (0.502) (0.537) (0.504) (0.544)
Development 0.812** 0.516# 0.697* 0.525# 0.705* 0.599* 0.717** 0.611* 0.729**
(0.280) (0.299) (0.300) (0.308) (0.302) (0.288) (0.276) (0.289) (0.278)
Environmental 0.426 0.501 0.654 0.489 0.662 0.335 0.439 0.320 0.430
(0.398) (0.437) (0.423) (0.447) (0.426) (0.441) (0.389) (0.446) (0.391)
Human Rights 0.407 1.539 1.472 1.574 1.462 0.821 0.577 0.802 0.549
(0.641) (1.067) (1.071) (1.119) (1.089) (0.664) (0.599) (0.679) (0.614)
Finance 1.009** 0.283 0.551 0.336 0.537 0.704# 0.889* 0.756# 0.915*
(0.389) (0.342) (0.437) (0.356) (0.445) (0.398) (0.375) (0.402) (0.377)
Tec h n ology 2.346* 2.451** 2.245* 2.448** 2.248* 2.512**
2.371** 2.510** 2.371**
Ecker-Ehrhardt M.
Author's personal copy
Tab l e 1 (continued)
Model 1 2 3 4 5 6 7 8 9
(0.932) (0.837) (0.893) (0.842) (0.900) (0.837) (0.897) (0.842) (0.904)
Trade 0.342 0.038 0.118 0.024 0.101 0.161 0.292 0.174 0.296
(0.409) (0.394) (0.426) (0.397) (0.433) (0.402) (0.402) (0.404) (0.404)
Security 0.320 0.494 0.272 0.436 0.260 0.005 0.097 0.071 0.147
(0.556) (0.428) (0.516) (0.455) (0.521) (0.516) (0.510) (0.538) (0.525)
Commodities 0.434 0.528 0.433 0.519 0.418 0.539 0.498 0.529 0.484
(0.415) (0.418) (0.409) (0.422) (0.412) (0.428) (0.407) (0.430) (0.409)
Minimal Budget 1.501* 1.701** 1.724*** 1.699** 1.734*** 1.559* 1.519* 1.552* 1.516*
(0.631) (0.555) (0.512) (0.556) (0.508) (0.639) (0.630) (0.642) (0.631)
Alpha (over-dispersion) 1.211** 1.245** 0.918* 1.207** 0.893* 1.553* 1.323** 1.522* 1.294**
(0.455) (0.479) (0.384) (0.458) (0.384) (0.617) (0.469) (0.595) (0.459)
Log Likelihood 3477.8 2775.1 3546.9 2782.8 3552.9 2734.1 3451.1 2739.8 3457.8
Time span covered 1950- 1977- 1950- 1977- 1950- 1977- 1950- 1977- 1950-
NIO-Years 1644 1291 1644 1291 1644 1291 1644 1291 1644
NIOs 48 48 48 48 48 48 48 48 48
The table reports coefficients from negative-binomial regression models with year fixed effects and robust standard errors clustered over IOs in parentheses. Results for year fixed
effects and constant terms have been omitted
Self-legitimation in the face of politicization: Why international...
Author's personal copy
communication made a significant step towards centralization with the establishment of
the Public Outreach and Civil Society Division in 2009 after ASEAN summits had
been repeatedly targeted by various groups of protesters. In 2007 and 2008, for
example, massive activism contested ASEANspolicyofnon-interferencetowards
the Myanmar government; in 2009, the ASEAN summit even had to be cancelled
due to security concerns after protesters forced their way into the main summit venue.
Regarding instances of scandalization, the coefficient for Scandal relates to a change
in predicted levels of centralization by a factor of 1.8 (or 84%) when comparing IO-
Years without to those with observed scandals according to Model 7. Taking the
varying levels of media coverage of these scandals into account, predicted
Tab l e 2 Estimated Substantive Effects of Politicization Variables
Factor Change in predicted level
of Centralization
%Change in Centralization in predicted
level of Centralization
over range of X per SD increase in X over range of X per SD increase in X
Transnational Access
4.691 1.414 369.1% 41.0%
Protest (0/1)
1.597 1.154 59.7% 15.4%
Scandal (0/1)
1.843 1.074 84.3% 7.4%
Protest Coverage
2.637 1.139 163.7% 13.9%
Scandal Coverage
2.841 1.062 184.1% 6.2%
Based on Models as presented in Table 1
Model 9
Model 6
Model 7
Model 8
Fig. 2 Expected Centralization by Observed Level of Transnational Access (with 95%-confidence interval)
Ecker-Ehrhardt M.
Author's personal copy
centralization is estimated to increase from 2.6 to 4.5 over the observed range of
Scandal Coverage (Fig. 4). This increase relates to an estimated change in centraliza-
tion by a factor of about 2.8. The estimated positive relationship is illustrated by the
IWC, where public communication was structurally enhanced in 2012 following the
public scandalization of bribery. Similarly, after allegations of sexual assault against
Dominique Strauss-Kahn, then IMF managing director in 2011, we observe some
significant changes towards a centralization of tasks in the following years, including
the transformation of the External Relations Department into the Communication
Department in 2012.
Fig. 3 Expected Centralization by Observed Coverage of Protest (with 95%-confidence interval)
Fig. 4 Expected Centralization by Observed Coverage of Scandals (with 95%-confidence interval)
Self-legitimation in the face of politicization: Why international...
Author's personal copy
In sum, my results fully match the expectations as formulated in hypotheses 1 to 3, which
expect more centralized public communication when we find a greater transnationalization
of civil society, public protest activities, and scandals of IO personnel. Hence, there is
substantial empirical supportfortheformulatedpoliticization-legitimationlinkageasa
causal mechanism driving public communication ofIOsingeneral,andthecentralization
of communication in terms of tasks and departmentalization more specifically.
Remarkably, the alternative explanations yield much less support. Estimates for the
variable Security Area are insignificant across full Models (69), which does not provide
any evidence that sovereignty costs play a measurable role in making centralization less
likely. I also find only limited support for the ideathatthepredictedlevelofcentralization
of public communication increases with demand for promoting ideas by means of trans-
national or local campaigns. While IOs active in the areas of development (Development
Area)showconsistentlyhigherlevelsofcentralization, the estimated coefficients for the
variables Human Rights Area and Local Implementation are insignificant. Finally, there is
no empirical evidence for that normative change towards transnational transparency and
accountability drives centralization: Contrary to expectations, when controlling for other
factors, the observed levels of centralization negatively relate to the average level of
domestic democracy as measured by Democratic Membership.Inlightoftheseresults,
the suggested politicization-legitimation mechanism seems to provide a better explanation
for centralized public communication than alternative accounts.
In terms of robustness, alternative specifications of the estimated models using Tobit
and ordered logit regression yield very similar results (see online appendix, Table A.5).
While the variables of politicizationcapturing transnational access, public protest and
public scandalsshow significant relationships with centralization as expected, the
estimates of control variables provide no consistent empirical support for alternative
explanations. Finally, one may suspect that analysis of single dimensions of centraliza-
tionthe Codification of communication tasks and Departmentalization of these tasks
in communication unitsmight qualify results in important ways. For this reason, I run
additional models for each dimension separately fully reported in the online appendix
(Table A.6). In terms of estimated coefficients, however, the results remain fairly stable
for both Codification and Departmentalization. Only results for the variable
Transnational Access differ, in that the estimated coefficients are positive but insignif-
icant across models for Departmentalization. While evidence matches expectations with
regard to an empirical correlation between levels of transnational access and the level of
observed Codification, my data provides no such evidence for the way public commu-
nication is located in the organizational chart of IOs included in the analysis.
6 Conclusion
How and why do IOs centralize organizational capacities for public communication?
For all the attention IR scholars have paid to the institutional design of IOs we know
very little about organized public communication. According to the central argument
presented in this paper, IOs have taken substantial steps to centralize public commu-
nication in order to more effectively promote their mandates, policies, and procedures
in the face of increased levels of societal awareness, activism, and contestation.
Politicization drives centralization for two reasons: First, the demand for pro-active
Ecker-Ehrhardt M.
Author's personal copy
self-legitimation increases as transnational civil society organizes and gains access to
IO decision-making. Second, self-legitimation becomes an organizational priority the
more IOs are contested in the form of protest activism or public scandals, which
constitute additional triggers for enhancing capacities for managing public discourse.
The systematic analysis of a novel data set on the centralization of public commu-
nication across a stratified random sample of 48 IOs between 1950 and 2015 provided
some empirical support for the validity of the general argument. Focusing on the
internal explication of communication tasks and the departmentalization of these tasks
in communication units or bodies, this study has shown that observed centralization
substantially varies over time and across IOs. In terms of correlational evidence,
different degrees of transnational access, mediatized protest activities, and scandals
better account for this variation in a multivariate analysis than alternative factors, such
as democratic membership, sovereignty costs, or ambitious mandates to implement
policy goals on the ground.
The results have important implications for future research on IOs in general and IO
public communication in particular. First, the observed variation of centralization
should (re)direct scholarly attention towards organized public communication as an
important feature of IOs. While the opening up of IOs to civil society inclusion and
transparency has been investigated systematically (Steffek et al. 2008; Liese 2010;
Tallberg et al. 2014; Grigorescu 2007), mainstream institutionalism has neglected IO
public communication as an institutional organizational feature. Second, my findings
challenge existing research on IO-civil-society interaction that has disregarded the very
possibility of IOs following a logic of self-legitimation, including those studies that
solely frame public communication as Bpublic information^or a governance-tool for
development (Brüggemann 2008;Coldevin2001). Focusing on one of these logics
carries the danger of downplaying strategic imperatives that might bring IOs to
significantly diverge from normative expectations of institutional transparency (public
information) as well as efficient problem solving for specific policy problems (gover-
nance). Thus, the findings presented above significantly add empirical evidence in
support of recent calls to more systematically engage with self-legitimation as an
organizational imperative of IOs (Dingwerth et al. 2015; Gronau and Schmidtke 2016).
Finally, the results carry important normative implications. Some of the most
prominent visions of legitimate Bliberal^or Bcosmopolitan^orders assume that IOs
have an important role to play by promoting and enacting liberal norms such as human
rights and public accountability (Buchanan and Keohane 2006; Held 2002;Archibugi
2008). However, the findings suggest that IOs might try to avoid public control by pro-
actively Bmanaging^(de)politicization. For example, self-legitimation qua IO public
communication can be expected to symbolically construct compliance with the widely
accepted norms that define legitimate authority (Bernstein 2011). Further, we may
expect to find a significant Bde-coupling^between symbolic appearances and actual
negotiation procedures or operations. Most remarkably, IR scholars have argued that
recent trends of giving access to negotiations to transnational NGOs, or including the
Global South, have been more symbolic than efficacious, as regards actual decision
making and crucial negotiations inside IOs (Steffek et al. 2008). In light of such
critique, enhanced capacities for strategic public communication should cause concerns
that IOs might purposely misrepresent organizational realities in order to create public
supportif only in terms of Bwhite propaganda,^where biased, but correct information
Self-legitimation in the face of politicization: Why international...
Author's personal copy
is strategically deployed in order to persuade an audience largely unaware of its
strategic intentions and partialities (Jowett and O'Donnell 1992). If effective, such
management could substantially undermine the democratic credentials of IO authority
and, ultimately, global governance in general. Additional research is needed, in partic-
ular to better inform normative theorizing about the puzzling role IOs play as objects as
well as managers of (de)politicization and how strategic communication might affect
public control of international authority.
Acknowledgements Research for this article was funded by the German Research Foundation (EC 323/1-2).
Essential research assistance by Manuel Hofmann, Stefan Wiechmann, Florence Wild, Minna Ålander, Roisin
Cronin and Laura Jung is gratefully acknowledged. I wish to give special thanks to the three anonymous
reviewers, and Axel Dreher for detailed comments and recommendations.
Abbott, K. W., & Snidal, D. (1998). Why states act through formal international organizations. Journal of
Conflict Resolution, 42(1), 332.
Aghi, M., & McKee, N. (2000). Involving people, evolving behaviour. Penang and New York: Southbound
and United Nations Children's Fund.
Aldrich, H., & Herker, D. (1977). Boundary spanning roles and organization structure. Academy of
Management Review, 2(2), 217230.
Alleyne, M. D. (2003). Global lies? : Propaganda, the UN, and world order. In Houndmills, Basingstoke,
Hampshire. New York: Palgrave Macmillan.
Altides, C. (2009). Making EU politics public. How the EU institutions develop public communication.
Baden-Baden: Nomos.
Altman, T., & Shore, C. (2014). Paradoxes of 'public diplomacy': Ethnographic perspectives on the European
Union delegations in the antipodes. [article]. The Australian Journal of Anthropology, 25(3), 337356.
Anheier, H. K., Glasius, M., & Kaldor, M. (2001). Global civil society 2001. Oxford: Oxford University Press.
APEC. (2014). APEC Secretariatscommunicationsandpublicaffairsstrategicplan:20142016. Singapore:
Asia-Pacific Economic Cooperation.
Archibugi, D. (2008). The global commonwealth of citizens: Toward cosmopolitan democracy. New Jersey:
Princeton University Press.
Avant, D. D., Finnemore, M., & Sell, S. K. (2010). Who governs the globe?In D. D. Avant, M. Finnemore, &
S. K. Sell (Eds.), Who Governs the Globe?(pp.135). New York: Cambridge University Press.
Barnett, M. N., & Finnemore, M. (2004). Rules for the world. International organizations in global politics.
Ithaca/London: Cornell University Press.
Bazbauers, A. R. (2014) The wolfensohn, wolfowitz, and zoellick presidencies: Revitalising the neoliberal
agenda of the world bank. In Forum for Development Studies, (Vol. 41, pp. 91114, Vol. 1): Taylor &
Beck, U. (2004). Der kosmopolitische Blick oder: Krieg ist Frieden. Suhrkamp: Frankfurt A.M.
Bedford, K. (2009). Developing partnerships: Gender, sexuality, and the reformed world bank. Minneapolis
and London: U of Minnesota Press.
Bennett, W. L., Pickard, V. W., Iozzi, D. P., Schroeder, C. L., Lagos, T., & Caswell, C. E. (2004). Managing
the public sphere: Journalistic construction of the great globalization debate. The Journal of
Communication, 54(3), 437455.
Bernstein, S. (2011). Legitimacy in intergovernmental and non-state global governance. Review of
International Political Economy, 18(1), 1751.
Commission's legitimation strategies. JCMS: Journal of Common Market Studies, 51(2), 194211.
Bob, C. (2012). The global right wing and the clash of world politics (Cambridge studies in contentious
politics). New York: Cambridge University Press.
Ecker-Ehrhardt M.
Author's personal copy
Boin, A., 't Hart, P., & McConnell, A. (2009). Crisis exploitation: Political and policy impacts of framing
contests. Journal of European Public Policy,16(1), 81106.
Boomgaarden, H. G., Schuck, A. R. T., Elenbaas, M., & de Vreese, C. H. (2011). Mapping EU attitudes:
Conceptual and empirical dimensions of Euroscepticism and EU support. European Union Politics,
12(2), 241266. doi:10.1177/1465116510395411.
Braidwood, T. (2012). CLEAR-PLOT: Automating clarify & predicted values for chart generation.
Brändström, A., & Kuipers, S. (2003). From normal incidentsto political crises: Understanding the selective
politicization of policy Failures1. Government and Opposition, 38(3), 279305.
Brüggemann, M. (2008). Europäische Öffentlichkeit durch Öffentlichkeitsarbeit? Die Informationspolitik der
Europäischen Kommission (1ed.). Wiesbaden: VS Verlag für Sozialwissenschaften.
Brühl, T. (2010). Representing the people?NGOs in international negotiations. In J. Steffek& K. Hahn (Eds.),
Evaluating transnational NGOs. Legitimacy, Acountability and representation (pp. 181199).
Houndsmills: Palgrave.
Buchanan, A., & Keohane, R. O. (2006). The legitimacy of global governance institutions. Ethics &
International Affairs, 20(4), 405437.
Cameron, C. A., & Trivedi, P. K. (1998). Regression analysis of count data (Vol. 30, economic society
monographs). Cambridge: Cambridge University Press.
Cárdenas, E. J. (2000). UN financing: Some reflections. European Journal of International Law, 11(1), 6775.
Coldevin, G. (2001). Participatory communication and adult learning for rural development. Journal of
International Communication, 7(2), 5169.
de Wilde, P. (2011). How politicisation affects european integration: Contesting the EU budget in the
Netherlands, Denmark and Ireland. Oslo: Faculty of Social Sciences, University of Oslo.
Defourny, V. (2003). Public information in the UNESCO: Towards a strategic role. In K. Sriramesh & D.
Vercic (Eds.), The global public relations handbook. Theory, research, and practice (pp. 425440).
Mahwah: Routledge.
della Porta, D. (2007). The global justice movement: Cross-national and transnational perspectives. Boulder:
Dellmuth, L. M. (2016). The knowledge gap in world politics: Assessing the sources of citizen awareness of
the United Nations security council. Review of International Studies, 42(4), 673700. doi:10.1017
Dellmuth, L. M., & Tallberg, J. (2015). The social legitimacy of international organisations: Interest repre-
sentation, institutional performance, and confidence extrapolation in the United Nations. Review of
International Studies, 41(03), 451475.
Dimitrov, R. (2014). Bringing communication up to agency: UNESCO reforms its visibility. Public Relations
Inquiry, 3(3), 293318. doi:10.1177/2046147x14544396.
Dingwerth, K., Lehmann, I., Reichel, E., Weise, T., & Witt, A. (2015). Many Pipers, Many Tunes? Die
Legitimationskommunikation internationaler Organisationen in komplexen Umwelten. In E. d.
Conceição-Heldt, M. Koch, & A. Liese (Eds.), Internationale organisationen: Autonomie, politisierung,
interorganisationale beziehungen und wandel (1st ed., pp. 191217). Baden-Baden: Nomos.
Drori, G. S., Meyer, J. W., & Hwang, H. (2006). Globalization and organization: World society and
organizational change.Oxford:OxfordUniversityPress.
Ecker-Ehrhardt, M. (2012). Cosmopolitan politicization: How perceptions of interdependence foster citizens'
expectations in international institutions. European Journal of International Relations, 18(3), 481508.
Ecker-Ehrhardt, M. (2014). Why parties politicise international institutions: On globalisation backlash and
authority contestation. Review of International Political Economy, 21(6), 12751312.
Falk, R. (1995). On humane governance. Toward a new global politics. The world order models project report
of the global civilization initiative. University Park: The Pennsylvania State University Press.
Florini, A. (Ed.). (2007). The right to know. Transparency for an open world. New York: Columbia University
Florini, A. M., & Simmons, P. J. (2000). What the world needs now? In A. M. Florini (Ed.), The third force:
The rise of transnational civil society (pp. 115). Tokyo and Washington, DC: Japan Center for
International Change and Carnegie En- dowment for International Peace.
Furia, P. A. (2005). Global citizenship, anyone? Cosmopolitanism, privilege and public opinion. Global
Society, 19(4), 331359.
Gibson, J. (2008). The myth of the multitude: The endogenous demise of Alter-globalist politics. Global
Society, 22(2), 253275.
Gilson, J. (2011). Structuring accountability: Civilsociety and the Asia-Europe meeting. In J. A. Scholte (Ed.),
Building global democracy (pp. 206224). Cambridge: Cambridge University Press.
Self-legitimation in the face of politicization: Why international...
Author's personal copy
Grigorescu, A. (2007). Transparency of intergovernmental organizations: The roles of member states,
international bureaucracies and nongovernmental organizations. International Studies Quarterly, 51(3),
Gronau, J. (2015). Die selbstlegitimation internationaler institutionen: G8 und G20 im vergleich (staatlichkeit
im wandel, vol. band 24).CampusVerlag:FrankfurtamMain.
Gronau, J., & Schmidtke, H. (2016). The quest for legitimacy in world politicsinternational institutions
legitimation strategies. Review of International Studies, 42(03), 535557.
Grunig, J. E. (1992). Excellence in public relations and communication management (communication
textbook series public relations). Hillsdale: L. Erlbaum Associates.
Grunig, J. E., & Hunt, T. (1984). Managing Public Relations. New York: Rinehart and Winston.
Hahn, K., & Holzscheiter, A. (2013). The ambivalence of advocacy: Representation and contestation in global
NGO advocacy for child workers and sex workers. Global Society, 27(4), 497520. doi:10.1080
Hannerz, U. (1990). Cosmopolitans and locals in world culture. Theory, culture and society, 7(2), 237251.
Held, D. (2002). Law of states, law of people: Three models of sovereignty. Legal Theory, 8(1), 144.
Henderson, D. (1999). The MAI affair: A story and its lessons. London: Royal Institute for International
Heupel, M., Hirschmann, G., & Zürn, M. (2015). Internationale Organisationen und der Schutz der
Menschenrechte. In E. d. Conceição-Heldt, M. Koch, & A. Liese (Eds.), Internationale
Organisationen: Autonomie, Politisierung, interorganisationale Beziehungen und Wandel (1st ed., pp.
428456). Baden-Baden: Nomos Verlagsgesellschaft mbH & Co. KG.
Hurd, I. (1999). Legitimacy and Authority in International Politics. International Organization, 53(2), 379
Joachim, J. (2003). Framing issues and seizing opportunities: The UN, NGOs, and women's rights.
International Studies Quarterly, 47(2), 247274.
Johnson, T. (2011). Guilt by association: The link between states' influence and the legitimacy of intergov-
ernmental organizations. The Review of International Organizations, 6(1), 5784.
Jowett, G. S., & O'Donnell, V. (1992). Propaganda and Persuasion. London: Sage.
Keck, M., & Sikkink, K. (1998). Activists beyond Borders: Transnational advocacy networks in international
politics. Ithaca: Cornell University Press.
Koremenos, B., Lipson, C., & Snidal, D. (2001). The rational Design of International Institutions.
International Organization, 55(4), 761799.
Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S., & Frey, T. (2008). West Eur o pean po l i tics in the
age of globalization. Cambridge: Cambridge University Press.
Lehmann, I. A. (1999). Peacekeeping and public information: caught in the crossfire (Cass series on
peacekeeping, Vol. 5). London and Portland: F. Cass.
Lehmann, I. A. (2011). The political and cultural dynamics of united nations media scandals: from waldheim
to annan. MEDIA@ LSE electronic working papers no. 22. London: London School of Economics and
Political Science.
L'Etang, J. (2009). Public relations and diplomacy in a globalized world: An issue of public communication.
American Behavioral Scientist, 53(4), 607626. doi:10.1177/0002764209347633.
Liese, A. (2010). Explaining varying degrees of openness in the food and agriculture organization of the
United Nations (FAO). In C. Jönsson, & J. Tallberg (Eds.), Transnational actors in global governance
(pp. 88109). Houndmills: Palgrave Macmillan.
Long, S. J., & Freese, J. (2006). Regression Models for Categorical Dependent Variables Using Stata (2ed.).
College Station: Stata Press.
Marshall, M., Jaggers, K., & Gurr, T. (2016). Polity IV project: Political regime characteristics and
transitions, 18002015. INSCR. Vienna: Center for Systemic Peace.
Mau, S., Mewes, J., & Zimmermann, A. (2008). Cosmopolitan attitudes through transnational practices?
Global Networks: A Journal of Transnational Affairs, 8(1), 124.
Mefalopulos, P. (2008). Development communication sourcebook: Broadening the boundaries of communi-
cation. Washington, DC: World Bank Publications.
Meyer, C. O. (2009). Does European Union politics become mediatized? The case of the European
Commission. Journal of European Public Policy, 16(7), 10471064.
Meyer, J. A., & Califano, M. G. (2006). Good intentions corrupted: The oil-for-food scandal and the threat to
the UN. New York: Public Affairs.
Meyer, J. W., Boli, J., & Thomas, G. M. (1987). Ontology and rationalization in the western cultural account.
In G. M. Thomas & J. W. Meyer (Eds.), Institutional structure: Constituting state, society, and the
individual (pp. 1237). Beverly Hills: Sage.
Ecker-Ehrhardt M.
Author's personal copy
Moulin, C., & Nyers, P. (2007). "We live in a country of UNHCR" - refugee protests and global political
society. International Political Sociology, 1(4), 356372.
NATO. (2009). NATO strategic communications policy. Brussels: North Atlantic Treaty Organization.
Nissen, S. (2010). Kommunikation in der Krise. Entwicklung und Erfolgsbedingungen der EU-
Informationspolitik. Zeitschrift für Politikberatung, 57(4), 453473.
Norris, P. (2000). Global Governance & Cosmopolitan Citizens. In J. S. Nye Jr. & J. D. Donahue (Eds.),
Governance in a globalizing world (pp. 155177). Washington, DC: Brookings Institution Press.
Nullmeier, F., Bigeon, D., Nonhoff, M., Schmidtke, H., & Schneider, S. (2010). Prekäre legitimitäten:
rechtfertigung von herrschaft in der postnationalen konstellation. Campus: Frankfurt A. M.
OAS. (1977). Executive order no. 771. Subject: objectives and activities of the department of public
information. Washington, D.C.: Organization of American States.
OAS. (1982). Orden ejecutiva no. 821. Asunto: oficina del secretario general. Washington, D.C.:
Organization of American States.
OAS. (1992). Executive order no. 921. Subject: creation and structure of the Department of public
information. Washington, D.C.: Organization of American States.
OAS. (1997). Executive order no. 974. Subject: office of external relations. Washington, D.C.: Organization
of American States.
OAS. (2006). Annual report of the secretary general 20052006. Washington, D.C.: Organization of
American States.
OAS (2012). Executive order no. 0801 rev. 5. Subject: structure of the general secretariat. Washington, D.C.:
Organization of American States.
O'Brian, R., Goetz, A. M., Scholte, J. A., & Williams, M. (2000). Contesting global governance: Multilateral
institutions and global social movements. Cambridge: Cambridge University Press.
Odell, J., & Eichengreen, B. (1998). The United States, the ITO, and the WTO: Exit options, agent slack, and
presidential leadership. In A. O. Krueger (Ed.), The WTO as an international organization (pp. 181206).
Chicago: Chicago University Press.
Odugbemi, S., & Lee, T. (2011). Accountability through public opinion: From inertia to public action.
Washington, DC: World Bank Publications.
Pevehouse, J., Nordstrom, T., & Warnke, K. (2004). The correlates of war 2 international governmental
organizations data version 2.0. Conflict Management and Peace Science, 21(2), 101119.
Pianta, M., & Zola, D. (2005). The rise of global movements, 1970-2005. Paper presented at the ACI Meeting,
Paris, 30 September-1 October 2005.
Polachek, S. W., Robst, J., & Chang, Y.-C. (1999). Liberalism and interdependence: Extending the trade-
conflict model. Journal of Peace Research, 36(4), 405422.
Price, R. (2003). Transnational civil society and advocacy in world politics. Wor l d P oliti c s , 5 5(4), 579606.
Putnam, R. D. (1988). Diplomacy and domestic politics: The logic of two-level games. International
Organisation, 42(3), 427.
Rauh, C. (2012). Politicisation, issue salience, and consumer policies of the European commission: does
public awareness and contestation of supranational matters increase the responsiveness of Europe's
central agenda-setter?. Dissertation. Berlin: Freie Universität Berlin.
Raustiala, K. (1997). States, NGOs, and international environmental institutions. International Studies
Quarterly, 41(4), 719740.
Reus-Smit, C. (2007). International crises of legitimacy. International Politics, 44(23), 157174.
Risso, L. (2014). Propaganda and intelligence in the Cold War: the NATO Information Service (studies in
intelligence). Milton Park, Abingdon, Oxon. New York: Routledge.
Rochon, T. (1988). The politics of the peace movement in Western Europe.Princton: Princeton University Press.
Rosenau, J. N. (1995). Governance in the twenty-first century. Global Governance, 1(1), 1343.
Rosenau, J. N. (2003). Distant proximities. Dynamics beyond globalization. Princeton: Princeton University
Sanchez, T., & Urpelainen, J. (2014). A strategic theory of effective monitoring arrangements for international
institutions. Journal of Theoretical Politics, 26(4), 599628. doi:10.1177/0951629813511550.
Schlagheck, D. M. (2016). Corruption at the United Nations: A Nonprofit, Humanitarian Organization
Confronts Ethical and Financial Misconduct. In C. Funderburk (Ed.), Political Corruption in
Comparative Perspective: Sources, Status and Prospects (pp. 165173). London and New York:
Schmitter, P. C. (1969). Three neo-functionalist hypotheses about international integration. International
Organization, 23(1), 161166.
Scholl, C., & Freyberg-Inan, A. (2013). Hegemony's dirty tricks: Explaining counter-globalization's weakness
in times of neoliberal crisis. Globalizations, 10(4), 619634.
Self-legitimation in the face of politicization: Why international...
Author's personal copy
Scholte, J. A. (2011). Global governance, accountability and civil society. In J. A. Scholte (Ed.), Building
global democracy? Civil society and accountable global governance (pp. 841). Cambridge: Cambridge
University Press.
Seabrooke, L. (2007). Legitimacy gaps in the world economy: Explaining the sources of the IMF's legitimacy
crisis. International Politics, 44(2), 250268.
Servaes, J. (2007). Harnessing the UN system into a common approach on communication for development.
International Communication Gazette, 69(6), 483507.
Smith, C. B. (2004). The politics of USUN reengagement: Achieving gains in a hostile environment.
International Studies Perspectives, 5(2), 197215.
Steffek, J. (2007). Legitimacy in international relations: From state compliance to citizen consensus. In A.
Hurrelmann, S. Schneider, & J. Steffek (Eds.), Legitimacy in the age of global politics (pp. 175192).
Houndmills: Palgrave Macmillan.
Steffek, J., Kissling, C., & Nanz, P. (Eds.). (2008). Civil society Partizipation in European and global
governance. A cure for the democratic deficit? Houndmills: Palgrave Macmillan.
Suchman, M. C. (1995). Managing legitimacy: Strategic and institutional approaches. The Academy of
Management Review, 20(3), 571610.
Symons, J. (2011). The legitimation of international organisations: Examining the identity of the communities
that grant legitimacy. Review of International Studies, 37(05), 25572583.
Szerszynski, B., & Urry, J. (2002). Cultures of cosmopolitanism. The Sociological Review, 50(4), 461481.
Tallberg, J., Sommerer, T., Squatrito, T., & Jönsson, C. (2013). The opening up of international organizations:
Transnational access in global governance. Cambridge: Cambridge University Press.
Tal lbe rg, J ., So mme rer, T., S qua trit o, T., & J öns son , C. (2 014). Explaining the transnational Design of International
Organizations. International Organization, 68(04), 741774. doi:10.1017/S0020818314000149.
Tarrow, S. (2001). Transnational politics: Contention and institutions in international politics. Annual Review
of Political Science, 4(1), 120.
Thompson, J. (2000). Political scandal: Power and visibility in the media age. Cambridge: Polity.
Tomz, M., Wittenberg, J., & King, G. (2003). CLARIFY: Software for interpreting and presenting statistical
results. Journal of Statistical Software, 8(1), 130.
UNGA. (1946). Resolution 13 (I). Organization of the Secretariat. A/RES/13(I). New York City: United
Urbinati, N. (2003). Can Cosmopolitical democracy be democratic? In D. Archibugi (Ed.), Debating
Cosmopolitics (pp. 6785). London: Verso.
Wallace, M., & Singer, D. J. (1970). Intergovernmental organization in the global system, 1815-1964: A
quantitative description. International Organization, 24(02), 239287.
Walton, J., & Shefner, J. (1994). Latin America: Popular protest and the state. In J. Walton (Ed.), Free Markets
&FoodRiots(pp. 97134). Oxford: Blackwell Publishers Ltd..
Weber, M. (1978). Economy and society: An outline of interpretive sociology. Berkeley: Univ of California
Willetts, P. (2000). From" consultative arrangem ents" to" partnership": The changing status of NGOs in
diplomacy at the UN. Global Governance, 6(2), 191212.
WorldBank. (2011). A guide to the world bank. (3rd ed. ed.). Washington, D.C.: World Bank.
Zaum, D. (2006). The Authority of International Administrations in international society. Review of
International Studies, 32(3), 455473.
Zürn, M., Binder, M., & Ecker-Ehrhardt, M. (2012). International authority and its politicization. International
Theory, 4(1), 69106.
Ecker-Ehrhardt M.
Author's personal copy
... Encouraged by electoral gains in recent years, populist politicians have made fierce criticism of IOs part and parcel as well functioning (Kruck and Zangl 2020). IOs themselves increasingly invest in public communication, justifying their operations and policies to a variety of stakeholders, from governments to citizens (Zaum 2013;Ecker-Ehrhardt 2018;Dingwerth et al. 2019;Bexell et al. 2022). Recent years have even seen the emergence of new advocates for global governance, such as global coalitions of city leaders and businesses working with the UNFCCC to address climate change. ...
... Second, a number of studies have foregrounded states' attempts at legitimizing and delegitimizing IOs as a means to further their objectives in world politics (Hurd 2007;Morse and Keohane 2014;Binder and Heupel 2015; Stephen and Zürn 2019). Third, scholars have started to thoroughly scrutinize IOs' strategies of self-legitimation (Steffek 2003;Zaum 2013;Gronau and Schmidtke 2016;Ecker-Ehrhardt 2018;Zürn 2018;Dingwerth et al. 2019;Rocabert et al. 2019;von Billerbeck 2020). ...
... Legitimation is a relational concept, and IOs can be both the subject and object of the communication (Biegón and Gronau 2012, 179). As subjects, IOs engage in self-legitimation, aiming at maximizing their own legitimacy (Zaum 2013;Ecker-Ehrhardt 2018). As objects, IOs are often the targets of communication by national governments, NGOs, and political parties (Stephen and Zürn 2019; Dellmuth and Tallberg 2021). ...
Full-text available
Once staunch advocates of international cooperation, political elites are increasingly divided over the merits of global governance. Populist leaders attack international organizations for undermining national democracy, while mainstream politicians defend their importance for solving transboundary problems. Bridging international relations, comparative politics, and cognitive psychology, Lisa Dellmuth and Jonas Tallberg explore whether, when, and why elite communication shapes the popular legitimacy of international organizations. Based on novel theory, experimental methods, and comparative evidence, they show that elites are influential in shaping how citizens perceive global governance and explain why some elites and messages are more effective than others. The book offers fresh insights into major issues of our day, such as the rise of populism, the power of communication, the backlash against global governance, and the relationship between citizens and elites. It will be of interest to scholars and students of international organisations, and experimental and survey research methods.
... While previous research has focused on either legitimation (Zaum ed. 2013;Gronau and Schmidtke 2016;Ecker-Ehrhardt 2018;Dingwerth et al. 2020) or delegitimation (Haunss 2007;Copelovitch and Pevehouse 2019;Stephen and Zürn 2019), we demonstrate that this process typically unfolds in an interplay between the two. Moreover, while previous research has tended to focus on discursive, institutional, or behavioral (de)legitimation practices separately, we show how these practices more often than not go together and reinforce each other. ...
Full-text available
This book explores processes of legitimation and delegitimation of global governance institutions (GGIs). How, why, and with what impact on audiences, are GGIs legitimated and delegitimated? The book develops a comprehensive theoretical framework for studying processes of (de)legitimation in global governance and provides broad comparative analyses to uncover patterns of (de)legitimation processes. It covers a diverse set of global and regional governmental and nongovernmental institutions in different policy fields. Variation across these GGIs is explained with reference to institutional setup, policy field characteristics, and broader social structures, as well as to the qualities of agents of (de)legitimation. The approach builds on a mixed-methods research design that uses both quantitative and qualitative new empirical data. Three main interlinked elements of processes of legitimation and delegitimation are at the center of the analysis: the varied practices employed by different state and non-state agents that may boost or challenge the legitimacy of global governance institutions; the normative justifications that these agents draw on when engaging in legitimation and delegitimation practices; and the different audiences that may be impacted by legitimation and delegitimation. This results in a dynamic interplay between legitimation and delegitimation in contestation over the legitimacy of GGIs.
... 716). The importance of discursive strategies is also testified in a study by Ecker-Ehrhardt (2018), who shows that IOs have strongly invested in centralizing public communications in response to politicization, as the message they spread is almost as important as the actual content of the policies. The EU as well began expanding its traditional functional and peace narratives with those embedded in the language of democracy (Sternberg, 2015). ...
... This article sheds light on the perceptions actors have on an important staffing issue in IOs. Second, it adds to the study of the narratives and practices of legitimation used by IOs ( Ecker-Ehrhardt 2018 ;Dingwerth, Schmidtke, and Weise 2020 ;von Billerbeck 2020 ). Third, I contribute to IO legitimacy's empirical research by focusing on an overlooked dimension "administrative legitimacy" ( Murdoch, Connolly, and Kassim 2018 ). ...
Full-text available
The link between bureaucratic representation and the perceived legitimacy of international organizations (IOs) is often cited in the literature. However, we do not know exactly how this works empirically. In this article, I introduce two variables to better understand the bureaucratic representation–IO legitimacy relationship: elite beliefs about geographical representation and self-legitimation practices. The theoretical framework bridges the literature on IO legitimacy in international relations and the literature on representative bureaucracy in public administration. Based on the case of the United Nations Secretariat and semistructured interviews with staff members, human resources experts, and member state representatives, the qualitative analysis points to two conclusions. First, this article presents the various representative bureaucracy–related legitimation practices employed by the bureaucracy at the discursive, institutional, and behavioral levels. Second, bureaucratic representation is perceived as a democratic, fair, and technocratic source of legitimacy by member state representatives. This article adds to the empirical study of IO legitimacy and to recent studies on representative bureaucracy in IOs.
Apesar das diferenças estruturais e regulatórias que caracterizam as organizações internacionais individuais, a doutrina de hoje é unânime em afirmar a capacidade de cada organização de criar seu próprio sistema jurídico interno. A relação de trabalho dentro dessas organizações reproduz claramente muitas das características da relação de trabalho internas a um Estado e, embora haja uma falta de regras internacionais uniformes sobre o emprego, e cada organização internacional tenha absoluta autonomia jurídica e administrativa, é possível identificar certas obrigações e direitos dos funcionários internacionais que se replicam, de forma ampla e homogênea, na maioria das organizações. O artigo analisa resumidamente, em termos gerais, os direitos e deveres dos funcionários nacionais e internacionais, referindo-se aos “Standards of Conducts for the International Civil Service” e examinando os regulamentos internos da Organização Internacional do Trabalho (ILO, na sigla em inglês) e do World Bank (WBG, em inglês) com foco na evolução do dever de confidencialidade, tanto em nível nacional como internacional. Com efeito, apesar da reversão drástica da tendência que tem caracterizado os Estados democráticos, conduzindo a uma reformulação restritiva do sigilo administrativo e do dever de confidencialidade, este dever ainda está muito vivo e continua a desempenhar um papel importante. Não parece, portanto, desarrazoado afirmar que, dada a substancial afinidade entre as disposições regulamentares em matéria de dever de confidencialidade, esta pode ser qualificada como um princípio geral. Por outro lado, o que representa um importante ponto de distinção entre os ordenamentos jurídicos nacionais e internacionais é a lógica que move os diversos legisladores na regulamentação desse dever.
The aim of this article is to introduce a framework that expands our understanding of the power of public opinion in influencing decisions at the multilateral level, with particular focus on the relationship between public opinion and legitimacy of international institutions and IOs. The framework traces how multilateral issues evolve from having high consensus to being highly contested, or vice versa. The progression and pathways become clear when public contestation is disentangled from elite contestation. Further, it argues that the faster pace of issue politicization is aided by the political ecology dominated by social media, as compared to the period when news media were the public’s main source of information on international matters.
Full-text available
This book explores processes of legitimation and delegitimation of global governance institutions (GGIs). How, why, and with what impact on audiences, are GGIs legitimated and delegitimated? The book develops a comprehensive theoretical framework for studying processes of (de)legitimation in global governance and provides broad comparative analyses to uncover patterns of (de)legitimation processes. It covers a diverse set of global and regional governmental and nongovernmental institutions in different policy fields. Variation across these GGIs is explained with reference to institutional setup, policy field characteristics, and broader social structures, as well as to the qualities of agents of (de)legitimation. The approach builds on a mixed-methods research design that uses both quantitative and qualitative new empirical data. Three main interlinked elements of processes of legitimation and delegitimation are at the center of the analysis: the varied practices employed by different state and non-state agents that may boost or challenge the legitimacy of global governance institutions; the normative justifications that these agents draw on when engaging in legitimation and delegitimation practices; and the different audiences that may be impacted by legitimation and delegitimation. This results in a dynamic interplay between legitimation and delegitimation in contestation over the legitimacy of GGIs.
Full-text available
This book explores processes of legitimation and delegitimation of global governance institutions (GGIs). How, why, and with what impact on audiences, are GGIs legitimated and delegitimated? The book develops a comprehensive theoretical framework for studying processes of (de)legitimation in global governance and provides broad comparative analyses to uncover patterns of (de)legitimation processes. It covers a diverse set of global and regional governmental and nongovernmental institutions in different policy fields. Variation across these GGIs is explained with reference to institutional setup, policy field characteristics, and broader social structures, as well as to the qualities of agents of (de)legitimation. The approach builds on a mixed-methods research design that uses both quantitative and qualitative new empirical data. Three main interlinked elements of processes of legitimation and delegitimation are at the center of the analysis: the varied practices employed by different state and non-state agents that may boost or challenge the legitimacy of global governance institutions; the normative justifications that these agents draw on when engaging in legitimation and delegitimation practices; and the different audiences that may be impacted by legitimation and delegitimation. This results in a dynamic interplay between legitimation and delegitimation in contestation over the legitimacy of GGIs.
This introductory chapter sets out the rationale for the book, which offers an analysis of the role of media in shaping governments and particularly the standing of prime ministers and uses Sweden as a case study. It also offers the term “logic of centralization,” as a way to capture the movement upwards of communication in governments. The chapter relates this approach to bodies of literature in political science, public administration, and political communication. The central analytical concern is the empowerment of prime ministers and the reasons behind this development. The book advances the argument that an important contributing reason is media, which presents governments with functional pressures for the adaptation of procedures and practices. This involves an increased centralization of structures and functions, in part driven by the demands of media.KeywordsFunctionInstitutional changeLogic of centralizationMediaPrime minister empowerment
Full-text available
The public politicisation of European integration indicates a growing demand for public communication of supranational politics. This paper highlights that the messages the European Commission sends to its citizens do not meet this demand. A text analysis of almost 45,000 press releases the Commission has issued during 35 years of European integration rather indicates an extremely technocratic style of communication. Benchmarked against large samples of national executive communication, public political media, and scientific discourse, the Commission used and notably continues to use very complex language, specialized jargon, and a nominal style that obfuscates political action. This appears disadvantageous in a politicized context and more research on the reasons for this apparent communication deficit is needed.
International institutions vary widely in terms of key institutional features such as membership, scope, and flexibility. In this 2004 book, Barbara Koremenos, Charles Lipson, and Duncan Snidal argue that this is so because international actors are goal-seeking agents who make specific institutional design choices to solve the particular cooperation problems they face in different issue-areas. Using a Rational Design approach, they explore five features of institutions - membership, scope, centralization, control, and flexibility - and explain their variation in terms of four independent variables that characterize different cooperation problems: distribution, number of actors, enforcement, and uncertainty. The contributors to the volume then evaluate a set of conjectures in specific issue areas ranging from security organizations to trade structures to rules of war to international aviation. Alexander Wendt appraises the entire Rational Design model of evaluating international organizations and the authors respond in a conclusion that sets forth both the advantages and disadvantages of such an approach.