International

Review of
Administrative
Sciences
EU agencies: what is common and what is distinctive
compared with national-level public agencies
Dario Barbieri and Edoardo Ongaro
Abstract
The number and relevance of EU agencies have rapidly increased over the years:
EU agencies nowadays constitute an important part of the EU institutional landscape. The article investigates the EU agencies through categories of analysis well
established in studies of public management focused on the phenomenon of
agencies at the country level: structural disaggregation, autonomy, and contractualization. It emerges that EU agencies are relatively homogeneous, an aspect that
differentiates European agencies from the highly heterogeneous world of
national-level agencies. The main features of the EU agencies are examined, the
European type of agency is identified and defined, and the way the EU agency
model differs from country-level agencies is analysed. Research agendas on the
reform of the European Union might benefit from systematic investigation of EU
agencies: theoretical frameworks drawn from the public management field can
provide a significant contribution in this respect.

Points for practitioners

EU agencies are no longer residual organizations: they are a significant component of the functioning of the EU system and policy networks. By investigating the
features of such agencies through the conceptual lenses of public management,
and through comparison with the (much more investigated) national-level agencies, the article provides an outline of EU agencies in terms of structural relations
with the other EU institutions, autonomy, and modalities of steering and control.
Reforms of the EU through the establishment or revamping of agencies could
benefit from the systematic consideration of such features.

Dario Barbieri and Edoardo Ongaro, Professors, Universit Bocconi and SDA Bocconi School of
Management, Milan, Italy.
Copyright 2008 IIAS, SAGE Publications (Los Angeles, London, New Delhi and Singapore)
Vol 74(3):395420 [DOI:10.1177/0020852308095310]

396 International Review of Administrative Sciences 74(3)

Keywords: contractualization, European Union, public agencies,

semi-autonomous organizations, steering and control

Introduction
The employment of semi-autonomous organizations for the execution of public
functions is a phenomenon in many respects as old as public administration, at least
at the country level (Hood and Schuppert, 1988; Van Thiel, 2001). The problems of
governance of these bodies have a long history in studies of public management and
public administration (Wettenhall, 2005). Agencies are an important subject of reform
talk (OECD, 2002) and scientific analysis (Thynne, 2004). However, providing a definition of public agency has proved to be quite a difficult task. In an initial attempt at
defining the phenomenon, Pollitt et al. (2001) identify an agency as a public organization at arms length from the main administration (there is a degree of structural
disaggregation between the agency and the parent administration), carrying out
public tasks assigned on a non-competitive basis, having a core staff of public
servants, being in principle financed by the state budget, and subjected to at least
some administrative law procedures. The issue of the governance structure of these
disaggregated bodies and in particular the analysis of the agency phenomenon
through the central dimensions of: (1) the nature and degree of autonomy of these
organizations, and (2) the modalities whereby the steering and control over agencies
is wielded are central to this debate. Prominent research agendas in public management shaped around such basic concepts are providing important findings about
the phenomenon of agencies at the national level (Lgreid et al., 2005; Pollitt and
Talbot, 2004; Pollitt et al., 2004; Verhoest et al., 2004).
What is the potential of such research agendas if applied to European Union (EU)
agencies? It is the aim of this article to go along this path and investigate the EU
agencies through categories of analysis that are now well established in studies of
public management focused on the phenomenon of agencies at the country level. By
applying such categories of analysis (outlined in the final part of this section and illustrated in detail in the section EU agencies: dimensions of analysis and case evidence),
it emerges that EU agencies are relatively uniform when compared with the highly
variegated world of national-level agencies. The finding is quite puzzling, since there
seems to be a broad agreement by scholars investigating the phenomenon of agencies at the national level about the high heterogeneity of such bodies (Caulfield,
2004; Fedele et al., 2007; Lgreid et al., 2005; Ongaro, 2006; Pollitt and Talbot,
2004; Pollitt et al., 2004; Thynne, 2004; Verhoest et al., 2004): this aspect as well as
the characteristics of EU agencies more broadly are examined in this article.
First of all, some features of the peculiar nature of EU governance the broader
context in which agencies operate are outlined. Conceptualizing the EU governance is a huge task (see, inter alia, McCormick, 2005, for a general introduction; and
Nugent, 2006, ch. 21, for a thoughtful synthesis). For the purposes of the present
contribution, as a first step the perspective of the intergovernmental/supranational
(dynamic) balance of the EU (Nugent, 2006: 55761) will be adopted: intergovernmentalism refers to arrangements whereby nation states, in situations and conditions
they can control, cooperate with one another on matters of common interest; supra-

Barbieri and Ongaro EU agencies 397

nationalism refers to situations in which states work with one another in a manner
that does not allow them to retain complete control over further developments.
Factors that point to the supranational characteristics of the EU are the significant
functions attributed to the European Commission (including its policy initiation role),
the increasing powers of the European Parliament, the role performed over the entire
history of European integration by the European Court of Justice. Factors that point to
the intergovernmental characteristics of the EU include the leading role of the
European Council in terms of taking virtually all major decisions on the general direction and policy priorities of the EU, and the key legislative role of the Council of
Ministers (and the practice of attempting to reach consensus, even where majority
voting is permissible, whenever a state declares that it has an important interest at
stake). EU agencies are problematic to assess according to the intergovernmental/
supranational balance: they are non-majoritarian institutions established through
public acts of delegation by the EU Commission and in this sense they are supranational bodies that have in the EU Commission their main parent administration, and
that in some respects condition from the outside the member states (supranational
influence). At the same time, however, in other important respects EU agencies are
instrumental to forms of cooperation led by member states.
Another important issue concerns the functions performed by the EU agencies
within the policy process. EU agencies are mainly confined to the implementation
phase, though some of them may wield a role also in policy formulation. In an
attempt at applying Kingdons model of the policy process (Kingdon, 1994) to the
European Union, Richardson (2001) conceptualizes the EU policy process in four
stages agenda-setting, policy formulation (alternative specification, in the original
wording of Kingdon), policy decision, and policy implementation. All EU agencies are
involved in the implementation phase of an increasing number of EU policies; some
of them have an influence also on the policy formulation phase (as an example, see
the case of the European Environment Agency, EEA, examined in Appendix 2),
though not in the sense that they exercise formal decision-making powers, that are
the exclusive competence of treaty-based EU institutions (under the so-called Meroni
doctrine), but because they provide data, information, and (mainly informal) proposals that may eventually influence the actual formulation of the content of the
European public policy (formal policy decisions are adopted by the competent EU
institutions).
The administrative features of such bodies represent another important aspect. EU
agencies share many of the features of the EU bureaucracy (Stevens and Stevens,
2001; on the Commission, see Hooghe, 2001, and Page, 1997), though with one
important difference: their locations are scattered throughout Europe. Thus the international composition of the staff of these bodies and their condition as expatriates
on a permanent basis (at least for a significant portion of the staff) elements in
common with the other EU institutions interact with a living place which is different
for each agency, and is not Brussels (thus the equivalents of the anthropological
studies conducted by Abeles, Bellier and McDonald, 1993 cited in Stevens and
Stevens, 2001 and various subsequent publications on EU officials have still to be
produced for these relatively new bodies).
EU agencies have grown rapidly in number and competences during the 1990s

398 International Review of Administrative Sciences 74(3)

Table 1 List of EU agencies (first pillar)
Agency

Established

Location

Policy sector(s)

CdT
Translation Centre for the
Bodies of the European Union

1994

Luxembourg
(Luxembourg)

Internal
administration

Cedefop
European Centre for Development
of Vocational Training

1975

Thessaloniki
(Greece)

Social policies
Employees training

CFCA
European Fisheries Control Agency

2005

Vigo
(Spain)

Internal market

CPVO
Community Plant Variety Office

1994

Angers
(France)

Intellectual property
Products
communitarian
movement

EAR
European Agency for
Reconstruction

2000

Thessaloniki
(Greece)

External relations
Reconstruction
projects
management

EASA
European Aviation Safety Agency

2002

Koeln
(Germany)

Transports
Internal market
enhancement

ECDC
European Centre for Disease
Prevention and Control

2004

Stockholm
(Sweden)

Public health

ECHA
European Chemicals Agency

2006

Helsinki
(Finland)

Intellectual property
Internal market

EEA
European Environment Agency

1990

Copenhagen
(Denmark)

Environment

EFSA
European Food Safety Authority

2002

Parma (Italy)

Agriculture
Public health

EIGE
European Institute for Gender Equality

2005

Site to be
decided

Social policies
Human rights

EMCDDA
European Monitoring Centre for
Drugs and Drug Addiction

1993

Lisbon
(Portugal)

Public health,
Social policies
Crime monitoring

EMEA
European Medicines Agency

1995

London
(GB)

Public health
Products
communitarian
movement

EMSA
European Maritime Safety Agency

2003

Lisbon
(Portugal)

Transports
Internal market
enhancement

ENISA
European Network and
Information Security Agency

2004

Heraklion
(Greece)

Information
Internal Market

continues

Barbieri and Ongaro EU agencies 399

Table 1 cont.
Agency

Established

Location

Policy sector(s)

ERA
European Railway Agency

2004

LilleValenciennes
(France)

Transports
Internal market
enhancement

ETF
European Training Foundation

1990

Turin
(Italy)

External relations
Social policies

EUROFOUND
European Foundation for the
Improvement of Living and
Working Conditions

1975

Dublin
(Ireland)

Social policies
Employees
circulation

European GNSS Supervisory

Authority

2004

Site to be
decided

Transports
Internal market
enhancement

FRA
European Union Fundamental
Rights Agency

2007

Vienna
(Austria)

Free citizens
movement
Human rights

FRONTEX
European Agency for the
Management of Operational
Cooperation at the External Borders

2004

Warsaw
(Poland)

Free citizens
movement
Crime monitoring

OHIM
Office for Harmonization in the
Internal Market (Trade Marks
and Designs)

1993

Alicante
(Spain)

Intellectual property
Products
communitarian
movement

OSHA
European Agency for Safety and
Health at Work

1994

Bilbao
(Spain)

Public health
Social policies

Source: http://europa.eu/agencies/index_en.htm (accessed December 2007).

and, especially, the 2000s (see Table 1) and they now represent a significant part of
the communitarian institutional landscape: a part that is still under-investigated. In this
article the main features of the EU agencies are analysed and it is investigated
whether there is a European type of agency and how it differs from the (highly variegated) world of national-level agencies investigated in public management studies.
The main research questions can be formulated as follows: Does a European type of
public agency exist? What are its features? And how does it differ from the world of
the national- and sub-national-level agencies?
The list1 of the agencies that are investigated in this article is reported in Table 1.
The agencies constituted under the second and third pillars of the EU are excluded
from the analysis, as are the ones created under the Euratom Treaty, because of their
peculiar scope and characteristics. There is still no agreed definition of EU agency
(Groenleer, 2006) and there is no definition in official EU documentation, which usually refers to agencies as delegated bodies. EU agencies differ from three other types

400 International Review of Administrative Sciences 74(3)

of European entities commonly associated (in the communitarian institutional setting)

to the concept of delegated bodies. In Krehers classification (1997), the following
bodies cannot be considered EU agencies: internal bodies or departments of the
European Commission (for instance, the Statistical Office of the European Union
Eurostat which is independent from the Commission but is still considered a service,
in the organizational sense, within the same Commission); institutions explicitly listed
in the European Union Treaty (for instance, the European Central Bank, which is
defined by Everson, 1995, as an organization charged with the pursuit of distinct
Constitutional-type normative goals); institutions created on an intergovernmental
basis that is outside the legal framework of the European Union (like the European
University Institute in Florence). In this article, following and expanding the notion
elaborated by Pollitt et al. (2001, 2004), EU agencies are defined as those structurally disaggregated organizational solutions, established by EU institutions, that have the
above-reported features and that occupy a certain portion of space in the [autonomy] [steering & control] conceptual space.
EU agencies have been studied from different theoretical angles. From a political
science perspective, studies have been conducted on the effects of the settlement of
EU agencies on the governance mechanisms of the EU as a whole as well as on the
internal governance mechanisms of the same agencies (Barbieri, 2006b; Everson,
1995, 2005; Flinders, 2004; Hix, 1998; Kreher, 1997; Vos, 2000), occasionally also
through comparisons with the US political system (Shapiro, 1997). EU agencies play a
significant role in the regulatory framework of the EU: many scholars have devoted
their attention to the influence of EU agencies on EU regulatory regimes (Caduff and
Bernauer, 2006; Coen and Thatcher, 2008; Dehousse, 1997; Eberlein and Grande,
2005; Egan, 1998; Gehring and Krapohl, 2007; Lafond, 2001; Majone, 1997, 2003;
Randall, 2006; Thatcher, 2002). Some scholars have focused their analysis on issues
of accountability of EU agencies (Bergstrm and Rotkirch, 2003; Curtin, 2005; Majone
and Everson, 2001; Vos, 2005). Other scholars, from a law studies perspective, have
investigated the influence the establishment of EU agencies has had over the communitarian law (Chiti, 2000, 2004; Dehousse, 2002; Vos, 2003). Another stream of
relevant contributions to the debate on EU agencies comes from theoretical perspectives closer to economics (Borrs et al., 2007; Kassim and Menon, 2003; Krapohl,
2004; Pollack, 1997).
Investigations on EU agencies seem instead to have been neglected in the public
management literature: this article aims to fill this gap and explore the potential of
applying frameworks drawn from the public management camp to the study of EU
agencies. The analysis is conducted by comparison with national-level agencies,
which entails the need to define the term of comparison (what are the features of
national-level agencies?). This step has been carried out: (1) by broadly surveying
literature in public management for identifying the (variable) features according to
some specific categories of analysis (what types of autonomy national-level agencies
have, what mechanisms of steering and control are in operation, etc.); (2) by specifically making reference to a framework of analysis developed for the study of the
Italian public agencies (Fedele et al., 2007), as a suitable template for structuring the
empirical evidence; and (3) by employing as a term of reference the so-called tripod
model, a practitioners model which has been widely accepted as a model to

Barbieri and Ongaro EU agencies 401

describe and interpret the phenomenon of agencies at the national level. The tripod
model was first theorized by Pollitt et al. (2004): they refer to agencies as bodies that
are structurally disaggregated from the parent ministry, that have a focus on the
operations (on the policy delivery phase of the policy cycle), enjoy relevant managerial autonomy and are steered through forms of results-based control. This third point
requires additional explanation: many criticisms have been made of the tripod model
in the terms of worldwide convergence towards such a model not having theoretical nor empirical grounding. The reasons for considering the tripod model are that:
(a) it is instrumental to testing the hypotheses of convergence (i.e. we give another
chance to the convergence hypothesis that maybe in a very specific institutional
setting like the EU might find a chance of not being falsified); (b) it is a model near to
hand to define by comparison of the features of the EU type of agency. The countries
considered in drawing the characteristics of national-level agencies are the United
Kingdom (Carter and Greer, 1993; Greer, 1992, 1994; James, 2001; Pollitt and Talbot,
2004), Sweden (Christensen and Wise, 2003; Pierre, 2004) and Italy (Barbieri, 2006a;
Fedele et al., 2007; Ongaro, 2006), three countries representing substantially different
patterns of public agencies in terms of their features, role in the public sector, and historical development. In the final sections, findings concerning the main research questions and implications for future developments of research on the topic are discussed.

Methodology
The research design is based on the following steps: first, drawing on literature, the
categories of analysis have been identified (structural disaggregation, autonomy, and
contractualization) and specific indicators for each category operationalized (see
Table 2, column 1, and the following section). The indicators have been identified in
such a way as to make it possible to use data drawn from secondary sources (public
data, including: the statutes of agencies; the official evaluation reports issued by the
Commission and the evaluation reports made by external and independent scholars
on behalf of the Commission and of the European Parliament; multi-annual reports
and work plans; accounting, budgeting, and auditing documents issued by the same
organizations as well as official documents discharged by the Commission and the
European Parliament; Court of Auditors official reports). This has paved the way for
the second step: the application of the indicators to the entire population (as defined
in the previous section) of EU agencies (see Table 2). Then (the third step), in order to
deepen the understanding of the actual operations of EU agencies, and to be able to
go beyond the analysis of the legal-formal (however relevant) features of EU agencies, a multiple case study on two agencies, based on interviews and on-the-field
investigation, has been developed (Appendices 1 and 2). In this way, the main framework of analysis based on secondary sources has been integrated with in-the-field
analysis of the cases of the European Medicines Agency (EMEA) and the European
Environment Agency (EEA).
The selection of these two agencies is due to their different characteristics: EMEA
is an authorization agency responsible for granting permissions, EEA is a coordination/information agency collecting, analysing and redistributing information.
The EMEA has a relevant executive role in the delivery of the policy (in this case, the

402 International Review of Administrative Sciences 74(3)

pharmaceutical policy, though the formal power to grant authorizations remains with
the Commission), while the EEA is an independent information collecting agency (on
the distinction between executive and independent information collection agencies,
see Kreher, 1997) that may exercise a role in the policy formulation phase, as policy
facilitator and, maybe, also as policy impetus agency (Everson, 1995; Everson et al.,
1999).

EU agencies: dimensions of analysis and case evidence

The concepts employed in the analysis are those of structural disaggregation,
autonomy, and contractualization; contractualization is to be interpreted as a subcomponent of the broader category of steering and control the reason for focusing on contractualization is that it is a qualifying element of the tripod model. The
three concepts will be subsequently illustrated.
In Talbot (2004), structural disaggregation is described as a purchaserprovider
split, or as a policyoperations or policy formulationpolicy implementation division
of labour, that is, a form of functional specialization. Contractualization refers to the
relationship established between the purchaser and the provider, or the actor(s) formulating the policy and the actor(s) in charge of the operations: a sort of contract (in
organizational terms) defines what the agency should do, in return for receiving a
stated amount of resources. The contractual relationship should entail a resultsoriented focus in which results are specified in terms of outputs and outcomes. Those
two categories of analysis must be complemented by the notion of autonomization
(Pollitt et al., 2004). Autonomization assumes that establishing a separate body is not
sufficient if it is not given enough autonomy to perform its functions. The relative
autonomy entails the need to regulate the relationship between the agency and its
parent administration (namely the Commission). This can be done through conctractualization, as well as through re-regulation, i.e. establishing a new and different set
of rules from the ones in operation for the parent administration (Talbot, 2004). The
role of EU agencies is primarily focused on the provision of inputs to the policymaking process and the management of the implementation phase, though in some
cases it is limited to technical support to the European Commission by means of scientific advice or information delivery, or to a purely administrative role (as in the case
of the CDT and EAR).
Turning to the operationalization of such concepts, structural disaggregation
refers to the characteristics of the separation between the agency and the parent
administration and is measured in terms of: institutional distance between the agency
and the Commission (which increases if the agency has legal status, whether it
resolves upon its statute, and whether it has a corporate board); degree of complexity of the tasks performed by the agency (how disaggregation is fulfilled: the variety
of policy areas in which the agency operates, the heterogeneity of the outputs and
functions performed); institutional specialization of the operations (this concept refers
to the existence of a clear separation between the policy level and the operational
level).
Autonomy presents three dimensions: financial, managerial, and strategic (or
policy autonomy). Financial autonomy refers to the capacity to acquire extra financial

Barbieri and Ongaro EU agencies 403

resources, in addition to Commission funding: the higher the level of extra funds
generated, the greater the financial autonomy. Managerial autonomy measures the
autonomy of the agency with reference to the management of its financial, human,
and organizational resources, in terms of the autonomy of the chief executive (the
executive director, in EU terminology) in determining the staff size, the agency organizational structure, and the appointment of agency managers. Indicators of managerial autonomy are also the chance for the agency staff not to be subjected to
community regulation, and the opportunity for the agency to deliberate on financial
transactions and expenditure levels. Strategic autonomy refers to the influence of the
agency on the formulation of the policy targets and on the selection of instruments
to implement the policy.
The degree of contractualization refers to the existence of the contract between
the Commission and the agency, its features and the monitoring and rewarding tools
inherent in the contract. The notion of contractualization is used in order to analyse
whether the Commission can steer the agency through the contract. The indicators
measuring the degree of contractualization have been defined as follows: existence
of the contract, contents, set of tasks regulated by the contract; existence of a link
between targets and human resources management (in terms of evaluation and
possibility of removal of management if the targets of the agency are not achieved);
existence of a link between the achievement of the target and the provision of additional resources to the agency; existence of a system for monitoring the contract.
The agencies operative (at the time of writing) in the EU constitute the population
object of the investigation (Table 1): for each of the above features it has been
specified whether it is absent (no, in the table), it fully exists (yes in the table) or it
partly exists (partly in the table). The difficulties in measuring such dimensions have
led to consider the third option partly beyond the yesno dichotomy, in order to
graduate the evaluation in some instances.
It emerges that, contrary to what happens at the national level (e.g. see Fedele et
al., 2007, for the same framework of analysis applied to the Italian case), in many
cases there is a significant degree of homogeneity (all agencies present similar
features) at a level of analysis at which a significant heterogeneity in country-level
agencies is detectable. At a more detailed level, it would very likely be possible to discriminate between agencies (e.g. presence/absence of a member of the board
appointed by the Commission, and whether s/he is an official of the Commission or
appointed by the Commission): in this work the aim is to describe whether a type of
EU agency exists or not (e.g. all EU agencies have a board, while at the national level
there is differentiation regarding this characteristic). For this reason it is interesting to
highlight that most of the indicators used are homogeneous for all agencies and few
indicators point to differences among EU agencies. The results are summarized in
Table 2 and Table 3 and are described in the next paragraph.
Before presenting the analysis and comparison with national-level agencies, some
common features that characterize the EU agencies are described (drawing on
Barbieri, 2006b). The mandate of the agencies and the range of their powers are
delimited and identified in the founding statute: EU agencies have legal personality
and in all instances are managed by an executive director. The executive director is
responsible for the management of the agencys activities, in line with the agencys

Agency
EUROCdT Cedefop CFCA CPVO EAR EASA ECDC ECHA EEA EFSA EMCDDA EMEA EMSA ENISA ERA ETF FOUND FRA FRONTEX OHIM OSHA
Indicator
Structural disaggregation
A. INSTITUTIONAL DISTANCE
A-1. The agency has legal status
A-2. The agency resolves upon its Statute
A-3. Corporate board
B. TASK COMPLEXITY
B-1. The agency operates in more than one
policy area
B-2. The agency delivers heterogeneous
outputs/services
B-3. The agency wields a plurality of public
functions (regulation, control, production)
C. INSTITUTIONAL SPECIALIZATION on operation
C-1. The separation between Policy (ministry)
and Operation (agency) is clear

Yes
No
Yes

No

No

Yes

No

No

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

No

No No

No

No

Yes

No

No

No

No

Yes

Yes

Yes

Yes

No

No

No

Yes

No

Yes

Yes

No

No No

No

No

Yes

Yes

No

Yes

No

Yes

Yes No

No

No

Yes

Yes

No

Autonomy
D. FINANCIAL AUTONOMY
D-1. The agency can acquire further resources
No
through contracts with other administrations,
outside Commission funding
D-2. The agency can sell its products/services
D-3. The agency can borrow on capital market
E. MANAGERIAL AUTONOMY
E-1. The chief executive can determine the size of
agency staff
E-2. The chief executive can determine the agency
organization structure
E-3. The agency staff is not subjected to Community
regulation
E-4. The chief executive appoints the agency
managers
E-5. The agency can deliberate on financial
transactions and expenditure levels

Yes

No

Yes

Yes

No

Yes

No

Yes

Yes

No

Yes

No
No
Partly
Partly
No
No

Partly

continues

404 International Review of Administrative Sciences 74(3)

Table 2 Structural disaggregation, autonomy and contractualization dimensions of EU agencies

Table 2 cont.
Agency
EUROCdT Cedefop CFCA CPVO EAR EASA ECDC ECHA EEA EFSA EMCDDA EMEA EMSA ENISA ERA ETF FOUND FRA FRONTEX OHIM OSHA
Autonomy (cont.)
F. STRATEGIC AUTONOMY
F-1. The agency influences the formulation of
policy targets
F-2. The agency can determine the instruments to
implement policies

No

No

No

No

Partly Partly Partly Partly Partly

Partly

No

Partly

No

Partly No

Partly

Partly

No

No

Partly

Yes

Partly

b)
Partly
No
No
Yes
No
Yes

Barbieri and Ongaro EU agencies 405

Contractualization
G. DEGREE OF CONTRACTUALIZATION OF
COMMISSIONAGENCY RELATIONSHIP
G-1. Formal existence of a contract
G-2. Contents of the contract:
a) Targets and resources
b) General (generic) goals
c) Regulation of the relationship
G-3. The entire set of tasks of the agency are
regulated by the contract
G-4. Agreement on contract is an input to the
budgeting process
G-5. Targets in the contract are an input for
HRM/evaluation of the management
G-6. Failure in achieving contracted targets is one
cause of removal of top management
G-7. Agency is provided with additional resources
on achievement of targets
G-8. Existence of a system for the monitoring of
the contract

No

Structural disaggregation
Indicator
A. INSTITUTIONAL DISTANCE

B. TASK COMPLEXITY
C. INSTITUTIONAL SPECIALIZATIONS
on operation
Autonomy
Indicator
D. FINANCIAL AUTONOMY

E. MANAGERIAL AUTONOMY
F. STRATEGIC AUTONOMY

EU

Tripod Model

UK

Sweden

Italy

High
(though agencies do not resolve upon their
statutes uniform values for all agencies)
Low
(agencies operate only in one policy area)
High
(uniform values for all agencies)

High

Mixed

High

Mixed

Low

Mixed-Low

Mixed

Mixed

High

Mixed-High Mixed-Low Mixed-High

Low
(though a few agencies can acquire further
resources outside Commission funding)
Low
(uniform values for all agencies)
Low
(concerning the formulation of policy targets)
High
(concerning the determination of policy instruments)

Contractualization
Indicator
G. DEGREE OF CONTRACTUALIZATION
(PERFORMANCE CONTRACTING)

Low
(uniform values for all agencies)

Low

Mixed

High

Mixed

High

Mixed-High

High

Mixed-High

Low

Mixed

High

Low

High

High

Mixed-Low

Low

406 International Review of Administrative Sciences 74(3)

Table 3 Comparison of structural disaggregation, autonomy and contractualization dimensions of EU, tripod ideal-type, UK, Swedish and Italian
agencies

Barbieri and Ongaro EU agencies 407

mandate. The executive director should neither solicit nor accept instructions from
any government or other EU body, except in cases provided for by the regulation
establishing the agency. The executive director is usually appointed by the management board on proposal of the Commission, even if some exemptions exist. The
management boards (for an overview, see Table 4) are composed mainly of representatives of the European institutions and member states. In the case of agencies
with a social remit, there is a larger representation of social partners. In some cases,
representatives of the professional sectors are appointed, without voting right. Direct
representatives of the European Parliament are not present: in some cases, independent experts appointed by the European Parliament sit on the management
board. The management board can dismiss the director and establishes the internal
rules of the agency. The management board approves the annual programme of
activities, the annual activity report and proposes the discharge of the budget to the
European Parliament. Advisory Committees in some cases are present in the organizational structures of the EU agencies: they are usually composed of independent
experts and they are required to deliver opinions to the executive director on draft
versions of the acts of the agency.
The activities of the EU agencies are monitored and subjected to rules issued
by the European Parliament, the Council and the European Commission. The Commission plays the role of the main (executive) controller of the activities of the
agencies. From a legal point of view, the connection between the pre-decision and
decision phases of the policy process (involving the Commission, the Parliament and
the Council) and the operational, policy implementation level (formally assigned to
the delegated bodies) is strictly regulated. EU agencies do not hold formal decision
powers concerning the formulation of the policy (though in some cases they can
influence the definition of policy targets, i.e. they influence the alternative specification
phase of the policy formulation process (Kingdon, 1994)). External financial controls
and judicial controls are also present.

Case analysis: differences between EU and national-level agencies

This section examines the features of the EU agencies (Table 2) and compares them
with the features of national-level public agencies (Italy, Sweden, and the United
Kingdom) as well as with the ideal-typical features of the tripod model (findings of
this comparison are summarized in Table 3). The tripod model is derived from the
conceptualization elaborated by Pollitt et al. (2004). The findings of the multiple case
study on the two agencies EMEA and EEA are reported in Appendices 1 and 2
such findings are employed for complementing the analysis with in-depth investigation based on first-hand empirical evidence.
Disaggregation
A premise: given that in a number of instances entirely new tasks were assigned to
agencies established ex novo, in these cases tasks have not been disaggregated
from the parent administration (the Commission) to agencies. However, the notion of
disaggregation is employed as an umbrella concept, even though in some instances

408 International Review of Administrative Sciences 74(3)

Table 4 Composition of management board, advisory forums and scientific committees
of EU agencies
Agency

Management board

Advisory forum

CdT

1 r. per member State

2 r. of the Commission
1 r. from each body or
institution calling upon
the centres services

Cedefop

Per member State:

1 r. of the government,
1 r. of employers
organization, 1 r. of
employees organization.
4 r. of the Commission

CFCA

1 r. per member State,

4 r. of the Commission,
4 r., without right to vote,
of the fishing industry

CPVO

1 r. per member State,

1 r. of the Commission

EAR

1 r. per member State,

2 r. of the Commission

EASA

1 r. per member State,

1 r. of the Commission

Representatives from
aviation personnel,
manufacturers,
commercial and
general aviation
operators, maintenance
industry, training
organizations and
air sports

ECDC

1 r. per member State,

3 r. of the Commission,
2 experts appointed by
the Parliament

1 r. per member
States National
Authority

ECHA

1 r. per member State

Forum for exchange

of information on
enforcement activities
(1 r. per member
State and 5 experts)

EEA

1 r. per member State,

6 r. of the Commission

EFSA

15 members appointed
by the Council (in
cooperation with
Parliament and
Commission)

Scientific committee

1 r. per Regional
Fisheries Council

Committee for Risk

Assessment (1 r. per
member State),
Committee for
Socio-economic Analysis (1
r. per member State),
member State Committee
(1 r. per member State)
Up to 20 independent
scientists from the member
States

1 r. per member
State

Nine Chairpersons of
EFSAs Scientific Panels
plus six independent
scientists

continues

Barbieri and Ongaro EU agencies 409

Table 4 cont.
Agency

Management board

EMCDDA

1 r. per member State,

2 r. of the Commission,
2 experts appointed by
the Parliament

EMEA

1 r. per member State,

2 r. of the patients
organization, 1 r. of the
doctors organization,
1 r. of the veterinarians
organization, 2 r. of the
Commission, 2 experts
appointed by the
Parliament, , 1 r. per
each EEAEFTA State

EMSA

1 r. per member State,

4 r. of the Commission,
4 r., without right to vote,
of the professional sectors

ENISA

1 r. per member State,

3 r. of the Commission,
3 r. of the stakeholders

ERA

1 r. per member State,

4 r. of the Commission,
6 r., without right to vote,
of the professional sectors

ETF

1 r. per member State,

2 r. of the Commission

Advisory forum

Scientific committee

CHMP (Committee for

Medical Products for
Human Use): 1 r. per
member State, 1 r. per
each EEAEFTA State, up
to 5 experts co-opted
by CHMP; CVMP
(Committee for Medicinal
Products for Veterinary
Use): as CHMP; HCMP
(Committee for Herbal
Medicinal Products): 1 r. per
member State, 1 r. per each
EEAEFTA State; COMP
(Committee for Orphan
Medicinal Products): 1 r. per
member State, 1 r. per each
EEAEFTA State, 3 experts
appointed by the
Commission on EMEAs
proposal, 3 r. of the
patients organization, 1 r.
of the Commission; PDCO
(Paediatric Committee): 5
members of CHMP, 1 r. per
member State, 3 r. patients
organizations, 3 r. of healthcare professionals

30 experts representing
the relevant
stakeholders

Around 120 experts

appointed by the
management board
from member States,
eligible country, from
the social partners at
the European level

continues

410 International Review of Administrative Sciences 74(3)

Table 4 cont.
Agency

Management board

EUROFOUND

Per member State:

1 r. of the government,
1 r. of the employers
organization, 1 r. of the
employees organization.
2 r. of the Commission
1 independent person
appointed each by:
member States, Council
of Europe. 2 r. of the
Commission
1 r. per member State,
2 r. of the Commission
1 r. per member State,
1 r. of the Commission
Per member State:
1 r. of the government,
1 r. of the employers
organization. 1 r. of the
employees organization,
3 r. of the Commission

FRA

FRONTEX
OHIM
OSHA

Advisory forum

Scientific committee

Independent experts
from the member
States

Three advisory groups

(made up of members
of the management
board and others
nominated by the
Agency): Risk
Observatory, Working
Environment
Information, and
Communication and
Promotion

Note: R in second column stands for representative. In the table only the advisory forums and
scientific committees made up of external members and/or representatives of member States and
stakeholders are listed.

tasks were new and no public institution at the EU level was in charge of performing
them before they were assigned to an EU agency.
A. Institutional distance A-1. All agencies have legal status and are legally distinct
from the Commission. As a consequence, they can operate under the law of the state
in which they are located and are legally responsible for their actions in front of the
legal bodies of the country. Another consequence is that agencies can sign contracts
and acquire goods under their responsibility in the states where they operate: the
distance from the Commission is in this respect relevant. Even if they operate legally
in the member states, however, agencies are responsible to EU bodies for their acts.
National-level agencies present a more varied picture, as they may not have legal
status, and the degree of separation between the department and the agency can
vary widely.
A-2. Agencies do not resolve upon their statute. At the national level it is quite
common that agencies do not resolve upon their statutes, though it may occur: there

Barbieri and Ongaro EU agencies 411

is high heterogeneity at the country level concerning this point. The tripod model
does not require an agency to resolve upon its statute.
A-3. All agencies have a corporate board and a director; in some cases, there is
also an advisory forum and/or a scientific committee. The presence of a management
board, with the compulsory representation of the member states, and in almost all
instances the presence of an appointee of the Commission (and in some cases the
presence of experts nominated by the Parliament) seem to point at some degree of
institutional distance between the agency and the (multiple) parent administrations.
In some cases, stakeholders are represented in the management board. Nationallevel agencies present a wider differentiation.
B. Task complexity B-1. Each agency operates in one and only one policy area
which applies to most country-level agencies, though with exceptions. The tripod
model foresees a clear focus of agencies on the operations in one policy area, and
this is the case for all EU agencies. However, in a number of instances, more than one
Directorate General (DG) of the EU Commission performs as the parent administration. For instance, in addition to the relationships with the DG Budget, the DG Internal
Audit Service and the DG Personnel and Administration (like all agencies), the EMEA
is linked to the DG Enterprise and Industry, its main parent administration, but also to
the DG Enlargement and the DG External Relations, and it has interconnections also
with other DGs, such as DG Agriculture and Rural Development, DG Employment,
Social Affairs and Equal Opportunities, DG Health and Consumer Protection.
B-2/3. It emerges that agencies can deliver both homogeneous (e.g. Cedefop,
ENISA) and heterogeneous (e.g. EFSA, EMSA) outputs and services and execute either
one category or, more often, more than one of the main functions identified. This evidence counteracts a requirement of the tripod model, if we interpret it as being
characterized by specialization on a single task but it confirms the trend at the
national level, where in many cases agencies perform a multiplicity of functions (with
UK Next Steps agencies more focused on relatively well-specified tasks than Italian or
Swedish agencies).
C. Specialization in operations C-1. In almost all cases, there is a clear separation
between policy and operations. This evident split is formalized in the principle of
separation between decision powers (assigned to the Commission and other EU
institutions) and policy delivery tasks (assigned to the agencies). The specialization in
operations is a feature of the tripod model. National agencies, however, are much
more variable creatures.
Autonomy
D. Financial autonomy D-1. Some agencies can acquire further resources outside EU Commission funding through contracts with other administrations. Only a
few agencies are given the opportunity to finance themselves by means of fees
charged to customers; this is the case with CPVO, EMEA and OHIM. In budgetary
terms, agencies fall under the so-called non-compulsory part of the European
Commissions budget: for this reason, in addition to its power of discharge, the

412 International Review of Administrative Sciences 74(3)

European Parliament may also impose strict accountability requirements, such as

reporting procedures and attendance at parliamentary committees. Country-level
agencies are much more varied concerning this dimension.
D-2/3. EU agencies cannot sell their products on the market nor can they raise
loans. The rationale for this strict rule is that it represents a mechanism to maintain the
financial equilibrium of the overall budget, which might be affected by the opportunity to operate on the financial markets. In this respect, the financial autonomy of EU
agencies can be considered to be low. At the national level the degree of financial
autonomy varies: NPM-oriented agencies (typified by the UK Next Steps agencies)
or agencies totally disaggregated from the parent ministries (see the Swedish case;
Pierre, 2004) usually enjoy higher levels of financial autonomy than in other countries,
such as Italy (Fedele et al., 2007), but the majority of agencies even in this latter case
can extend their funding by selling products/services.
E. Managerial autonomy The managers of EU agencies have limitations in determining the size of the staff of the agency, which is approved by the Commission
together with the agency budget. Moreover, the director and the management
board have limited scope in determining the organizational structure of the agency
or in deliberating on financial transactions and expenditure levels; there are limitations
also to the appointment of agency managers. All in all, managerial autonomy seems
to be low compared to agencies in Sweden, the UK, and also Italy, and for sure it is
low compared to the tripod model. Indeed, it appears that EU agencies have no
actual difference in terms of their managerial autonomy from what would be achievable by internally decentralized units of the European Commission (managerial delegation; see Levy, 2003, for an analysis of decentralization and centralization as an
effect of management reform in the European Commission).
F. Strategic (policy) autonomy EU agencies have basically no autonomy in defining policy targets, though they can wield indirect influence on policy formulation, e.g.
by selecting and shaping the format of the information on which policy decisions are
based, as in the case of the environment agency (EEA), but they have some or significant autonomy in defining policy instruments. The country-level world of agencies
is varied concerning this aspect (ranging from agencies often involved also in policy
formulation, as in Sweden, to agencies usually more confined to policy implementation, but with exceptions, as in Italy).
Contractualization
G. Degree of contractualization of the relationship between European Commission and agencies Using the term contract to describe the relationships
between the Commission and the agencies could be considered improper since the
term contract implies a set of objectives to be reached and rules to be respected to
get back predetermined rewards: in the EU setting such a type of document is not
detectable. What is required of all agencies is compliance to certain budgetary procedures, including the production of documents and written communication concerning objectives and performance requirements, as well as follow-up procedures

Barbieri and Ongaro EU agencies 413

mainly involving the Commission and the Court of Auditors, but increasingly also the
European Parliament and the Council.2
Annual reports or work plans can be considered as kinds of contracts affecting the
budgetary process or the evaluation of the human resources, but, even if they can
affect the budgetary procedure and can have serious consequences on the amounts
of funds assigned to the agency, there is no direct and formal relation between
targets, results and punishments/rewards, hence there is no performance contract.
The director is evaluated on the basis of the targets reached and can even be
removed if s/he fails in achieving the targets, but performance management systems,
where they exist, are internal to the agency and they are not employed in an external contractual relationship between the Commission and the agency. EU agencies in
this respect are not tripod model agencies; indeed, the very limited degree of conctractualization, and especially of performance contracting, is a key element of differentiation of EU agencies from the tripod model (a differentiation which is quite
common also in national-level agencies: many studies report about the relative
absence or patchy diffusion of performance contracts, with the partial exceptions of
the UK and a few other countries).

Discussion
In this article we make an initial attempt to explore the potential of an important
portion of the public management research agenda on agencies when applied to the
EU agencies. To this purpose, the main features of the EU agencies have been
analysed in terms of the way they are disaggregated from the parent administration(s), the level of autonomy, and the degree of conctractualization, and it has been
investigated whether there is a European type of agency and how it differs from
national-level agencies.
What conclusions can be drawn concerning the research questions? Does an EU
type of agency exist, and what are its features? It emerges that a sort of European
(Union) type of public agency does exist, or at least that some common distinctive
features are recognizable and overall a significant degree of homogeneity is
detectable in the EU agencies, when compared with the highly variable world of
national agencies. What are the main features of the EU type of agency? All EU
agencies have legal status, and they do not resolve upon their statute. The governance structure of the agencies includes the presence of a corporate board and a
director. Each agency operates in one and only one policy area. Data indicate that
those agencies can deliver both homogeneous and heterogeneous outputs and
services and, in almost all cases, there is a clear separation between policy and operations. Decision powers on policy formulation are assigned to the Commission or the
other EU institutions and operative tasks are assigned to agencies. Considering the
aspects of autonomy, agencies have no autonomy in defining policy targets but have
significant autonomy in defining policy instruments. As regards financial autonomy,
even if some agencies can acquire further resources outside the Commission funding, the agencies cannot sell their products on the market nor raise loans, and overall financial autonomy is very constrained. The flexibility in managing personnel and
the organizational design is limited: there is a high degree of uniformity in the orga-

414 International Review of Administrative Sciences 74(3)

nizational structure of EU agencies and there is limited room for changing the structure and the policies of human resources management (see also Vos, 2005). If the
contractualization of the relationship between the agencies and the Commission is
considered, the elements of NPM-ness are very limited (if present at all): there is no
contract in the organizational sense, though there is some form of conctractualization, nor is there a performance contract. Many agencies developed performance
management monitoring systems and financial indicators and time measures, but
primarily for internal use (i.e. not driven by a performance contract, nor finalized at it).
To sum up, there is an EU type of agency, which is not the widely debated tripod
model, but a different model, with a very limited degree of conctractualization, limited
financial autonomy as well as very limited managerial autonomy concerning personnel and the organizational design, though it does have autonomy in the definition of
policy instruments (features of the EU type of agency in a comparative perspective
are reported in Table 3). A finer-grained analysis (which is advocated, see final section) may find differentiation between EU agencies in more specific aspects within
each of the dimensions considered but at the level of these dimensions, there is a
clearly identifiable EU type of agency with recognizable and distinctive features that
differentiate it from the tripod model as well as from national-level agencies.

Conclusion and developments

The results of the investigation point to the existence of a model of EU agency, an
aspect which makes the EU differ from the varied landscape that can be found when
examining national-level agencies; although it should be added that such a type of
EU agency is heavily influenced in its detailed sketching by the tasks actually performed by each agency, by the time period in which an agency has been established
and by its interaction with external stakeholders (see the cases of the EMEA and EEA).
A first set of questions for future research stems from this finding: what explains the
homogeneity of EU agencies? How did such homogeneity emerge over a relatively
long time span (the first EU agencies date back to the 1970s, though most agencies
were established during the 1990s and the 2000s) and given the different circumstances under which they were established? Is the evolving power balance within the
EU the main explanation for homogeneity (and particularly, is the struggle between
member states and the Commission for wielding influence over first pillar agencies a
factor leading to uniformity in the shape of granting low autonomy and applying a
common set of controls to all EU agencies), or are there other factors at work? And
how do they combine in determining relative uniformity?
Shifting from the search for explanations of uniformity to the policy implications of
it, future research work may also address issues of efficiency and effectiveness. A
second set of questions thus includes: does uniformity imply greater efficiency? And
if so (and this point is of course to be demonstrated), what could be the drivers of
increased efficiency (what kind of economies can be gained so that the same outputs
can be delivered with less resources)? And does uniformity imply greater effectiveness? Or is uniformity an impediment to adaptation to the different characteristics of
the various policy sectors in which EU agencies operate?
We advocate a research agenda on EU agencies rooted in public management

Barbieri and Ongaro EU agencies 415

that might contribute to the debate on how to reform EU institutions: EU agencies are
a widely under-investigated phenomenon, or at least under-investigated with the
conceptual lenses of public management. Studies concerning even more underinvestigated issues, such as the performance of EU agencies, the influence that EU
agencies wield over the functioning of policy networks operating at the European
level (where EU agencies often perform as hub of the network, see the cases of the
EMEA and EEA), and the role EU agencies may have for re-launching European integration through an administrative, low politics approach, could provide further
insights for would-be reformers engaged in the process of EU reform (on the issue of
promoting good governance in the EU and the potential of resorting to agencies,
see also Everson, 2005). Such a type of knowledge appears indeed more and more
necessary in the face of the new challenges posed by the turbulent environment in
which EU institutions operate, the enlargement of the Union in 2004 and 2007, and
the increase in the administrative workload of the Commission (and the strategic
challenges it entails for the Commission, see Metcalfe, 1999, 2000a).
Whatever its limitations, the study outlined in this article points to the consideration that the organizational form of agencies can and should also be studied in the
EU setting by resorting to theoretical frameworks drawn from the public management camp. The proposed categories of analysis, even if taken from studies addressing national and sub-national settings, appear to be fitting also for investigation of the
EU agencies. An elaboration of the categories of analysis as well as an extension of
the empirical evidence might be helpful for the development of the research agenda
on EU agencies and their influence on the broader functioning of the European
Union.

Appendix 1. The case of the EMEA (European Medicines Agency)

The agency has been structurally disaggregated from the DG Enterprise and Industry
of the Commission. The EMEA has the main task of assessing medicinal products
submitted by pharmaceutical companies, while the Commission receives the final
technical decision of EMEA on the approval of medicinal products, verifies the fairness
of the administrative procedures, and gives formal authorization to the commercialization of the products. Some categories of pharmaceutical products are compulsorily examined by EMEA, others follow a system of mutual recognition among national
pharmaceutical authorities, though more and more frequently pharmaceutical companies address EMEA for authorization to the commercialization (79 initial marketingauthorization applications were received in 2006, compared to 41 applications
received in 2005). Before the settlement of the EMEA, the communitarian pharmaceutical policy was poorly regulated at the central level and the role of the member
States was prominent, due to the nature of the comitology process as well as to the
prominence of national authorities in the authorization of pharmaceutical products. In
1987, following Directive 83/570, a committee was set up with the purpose of
fostering coordination among the national agencies. It was only since the settlement
of the EMEA that the centralized procedure of authorization was established and has
gradually become more reliable; the creation and extensive utilization of lists of
experts (approximately 2300) involved in the technical evaluation process has gradu-

416 International Review of Administrative Sciences 74(3)

ally increased the reliability of the centralized procedure of scientific assessment (a

total of 257 scientific-advice, protocol-assistance and follow-up requests were finalized in 2006, compared to 191 in 2005). The EMEA relies on external support, developing close professional relationships with national organizations and the epistemic
community (Metcalfe, 2000b). For what concerns financial resources, the EMEA
receives its funds from both the Commission and the pharmaceutical companies.
The level of financial autonomy in terms of acquiring further resources outside
Commission funding is high (approximately 70 percent of the overall budget). The
functioning of the pharmaceutical regulation processes is generally deemed to have
increased after the establishment of the EMEA. In the list of elements supporting this
analysis there is, first, the very small amount of the EMEAs decisions rejected by the
Commission (16 over 304 in the period 19952005), and, second, the increasing
number of centralized procedures since 1995, which points to the EMEA being
attractive compared to national authorization authorities. The professional nature of
the EMEA and its organizational networking as well as its being closely integrated
within the relevant epistemic community are central for it to perform a role as hub
organization inside the networked pharmaceutical regulatory setting. By operating as
the hub of a network of national organizations devoted to pharmaco-vigilance, and
co-opting national experts in its scientific committees, the EMEA has pushed national
agencies to operate in an EU strategic space (in the sense that national agencies
have to collaborate with all the other pharmaceutical agencies throughout Europe)
and it has the potential to foster a degree of collaboration and exchange of practices
at the EU level.

Appendix 2. The case of EEA (European Environment Agency)

The EEA can be defined as a provider of inputs data on the environment to the
DG Environment in order to elaborate policy proposals. Formally, the establishment of
this agency stems from the disaggregation of the data-gathering function (competence of the EEA), while decision-making powers in terms of formulation of policy
proposals on environmental issues remain with the Commission. The debate about
whether the core task of the EEA is data gathering, or (also) data analysis, or even
more largely the elaboration of initial policy proposals is crucial. Schout (1999) points
out that the EEA has frequently gone beyond its data gathering role and [it] has
worked on information for the public: the EEA had, or at least tried to have, a more
policy-oriented role. The Regulation 1641/2003 streamlined the EEAs structure and
made it less divisionalized; the EEAs activities were overall better integrated in the EU
environmental policy-making process. The organizational changes of 1999 and 2004
created a more complex and decentralized structure. There has also been an evolution of the relationship between the EEA and the DG Environment during the years:
the same Commission has officially reported that in the first years after the settlement
of the EEA there was from time to time a divergence of views regarding these priorities and activities, particularly between the Commission and the EEA, but since 2003
there was a fairly clear common understanding of the respective tasks, although
there is still occasional disagreement (Commission of the European Communities,
2003). The EEA has been asked by the European Parliament, the European

Barbieri and Ongaro EU agencies 417

Commission, and the member States not only to report and advise on the state of the
environment and of the technical issues concerning environmental protection and
the sustainable use of natural resources, but also on the effectiveness of key environmental and sectoral policies and their implementation (Office for Official
Publications of the European Communities, 2003). The EEA manages a network of
national and sub-national organizations gathering environmental information, and
verifies the scientific reliability of the data gathered: its funding mainly depends on
the DG Environment. The network of organizations collecting information in the
national states operates to a large extent through projects designed by the EEA and
having the EEA as its main contributor: it is through this funding that the EEA influences which projects are put into the pipeline at the EU level (Martens, 2005). It is
almost universally accepted, inside the Commission, that EEA is playing not only a
subsidiary role in the DG Environment, but is a major player in the EU environmental
policy cycle. The call of the Commission in 2003 for stronger integration of the activities of the EEA within environmental policy may also be interpreted as a recognition
by the Commission of the prominent and in a sense autonomous role performed by
the EEA in the EU scenario. The EEA has increasingly played an effective role in coordinating the national and sub-national organizations dealing with environmental
issues and has assumed an uncontroversial position in the EU environmental policy
field. The europeanization of procedures and techniques and the consequent
uniformity of environmental data, which allows reliable comparisons of the state of
the environment across Europe, is also an important achievement of this agency.

Notes
The article is the joint work of the authors; however, in the final writing the sections Methodology,
Conclusion and developments, Appendix 1, and Appendix 2 have been written by Dario Barbieri;
the sections Introduction, and Discussion have been written by Edoardo Ongaro, while the other
parts have been jointly written. We are grateful to Professor Christopher Pollitt and three
anonymous reviewers for their constructive comments.
1
2

EIGE (European Institute for Gender Equality) and European GNSS Supervisory Authority are
included in Table 1 but, at the time of writing, are not operative.
This kind of performance reporting is compulsory and involves primarily the Commission and
the European Parliament. It is a competence of the Court of Auditors to evaluate how well the
agency has applied the principles of sound financial management economy, efficiency and
effectiveness to the management of its budget. All agencies are subjected to the
assessment of their performances, known as performance audits and value for money audits.
The internal auditor, in each agency, advises the agency on dealing with risks, issues
independent opinions on the quality of management and control systems as well as
recommendations for improving operating conditions.

References
Abeles, M., Bellier, I. and McDonald, M. (1993) Approche Anthropologique de la Commision
Europenne (Brussels, unpublished report to the European Commission; cyclostyled).
Barbieri, D. (2006a) Transnational Networks Meet National Hierarchies: The Cases of the Italian
Competition and Environment Administrations, in M. Egeberg (ed.) Multilevel Union
Administration. London: Palgrave Macmillan.
Barbieri, D. (2006b) Evaluating Delegated Modes of Governance Inside EU: Towards a Model of

418 International Review of Administrative Sciences 74(3)

Communitarian Agency?, in C. Reichard, R. Mussari and S. Kupke (eds) The Governance of
Services of General Interest between State, Market and Society. Berlin: Wissenschaftlicher
Verlag.
Bergstrm, C.F. and Rotkirch, M. (2003) Decentralized Agencies and the IGC: A Question of
Accountability, Swedish Institute for European Policy Studies, 14.
Borrs, S., Koutalakis, C. and Wendler, F. (2007) European Agencies and Input Legitimacy: EFSA,
EMEA and EPO in the Post-delegation Phase, Journal of European Integration 29(5): 583600.
Caduff, L. and Bernauer, T. (2006) Managing Risk and Regulation in European Food Safety
Governance, Review of Policy Research 23(1): 15368.
Carter, N. and Greer, P. (1993) Evaluating Agencies: Next Step and Performance Indicators, Public
Administration 71(3): 40716.
Caulfield, J. (2004) Measuring Autonomy in Social Security Agency: A Four Country Comparison,
Public Administration and Development 24: 13745.
Chiti, E. (2000) The Emergence of a Community Administration: The Case of European Agencies,
Common Market Law Review 37: 30943.
Chiti, E. (2004) Decentralisation and Integration into the Community Administrations: A New
Perspective on European, Agencies, European Law Journal 10(4): 40238.
Christensen, T. and Wise, L.C. (2003) Evaluating Public Management Reform in Government:
Norway, Sweden and the United States, in H. Wollmann (ed.) Evaluation in Public-sector
Reform: Concepts and Practice in International Perspective. Cheltenham: Edward Elgar.
Coen, D. and Thatcher, M. (2008) Network Governance and Delegation: European Networks of
Regulatory Agencies, Journal of Public Policy 28(1): 4971.
Commission of the European Communities (2003) Report from the Commission to the Council:
Review of the European Environment Agency (EEA) (COMM (2003) 800 final). Brussels:
European Commission.
Curtin, D. (2005) Delegation to Non-majoritarian Agencies and Emerging Practices of Public
Accountability, in D. Gerardin and N. Petit, Which Regulatory Authorities for Europe? London:
Routledge.
Dehousse, R. (1997) Regulation by Networks in the European Community, Journal of European
Public Policy 4(2): 24661.
Dehousse, R. (2002) Misfits: EU Law and the Transformation of European Governance, in R.
Dehousse and C. Jrges (eds) Good Governance in Europes Integrated Market. Oxford:
Oxford University Press.
Eberlein, B. and Grande, E. (2005) Beyond Delegation: Transnational Regulatory Regimes and the
EU Regulatory State, Journal of European Public Policy 12(1): 89112.
Egan, M. (1998) Regulatory Strategies, Delegation and European Market Integration, Journal of
European Public Policy 5(3): 485506.
Everson, M. (1995) Independent Agencies: Hierarchy Beaters?, European Law Journal 1(2):
180204.
Everson, M. (2005) Good Governance and European Agencies: The Balance, in D. Geradin, R.
Mnoz and N. Petit (eds) Regulation through Agencies in Europe: A New Paradigm for
European Governance. Cheltenham, UK and Northampton, MA: Edward Elgar.
Everson, M., Majone, G., Metcalfe, J.L. and Schout, A. (1999) The Role of Specialized Agencies in
Decentralising EU Governance, paper presented to the European Commission, Bruxelles.
Fedele P., Galli, D. and Ongaro, E. (2007) Disaggregation, Autonomy and Re-regulation,
Contractualism: Public Agencies in Italy (19922005), Public Management Review 9(4): 55785.
Flinders, M.V. (2004) Distributed Public Governance in the European Union, Journal of European
Public Policy 11(3): 52344.
Gehring, T. and Krapohl, S. (2007) Supranational Regulatory Agencies between Independence and
the EMEA and the Authorization Pharmaceuticals in the European Single Market, Journal of
European Public Policy 14(2): 20826.
Greer, P. (1992) The Next Steps Initiative: An Examination of the Agency Framework Documents,
Public Administration 70(1): 89100.
Greer, P. (1994) Next Steps Origins: Transforming Central Government. Buckingham: Open
University Press.

Barbieri and Ongaro EU agencies 419

Groenleer, M. (2006) The European Commission and Agencies, in D. Spence (ed.) The European
Commission. London: John Harper Publishing.
Hix, S. (1998) The Study of the European Union II: The New Governance Agenda, Journal of
European Public Policy 5(1): 3865.
Hood, C. and Schuppert, G.F. (1988) Delivering Public Services in Western Europe: Sharing Western
European Experience of Para-Government Organization. London: SAGE.
Hooghe, L. (2001) The European Commission and the Integration of Europe: Images of
Governance. Cambridge: Cambridge University Press.
James, O. (2001) Business Models and the Transfer of Businesslike Central Government Agencies,
Governance 14(2): 23352.
Kassim, H. and Menon, A. (2003) The PrincipalAgent Approach and the Study of the European
Union: Promise Unfulfilled?, Journal of European Public Policy 10(1): 12139.
Kingdon, J. (1994) Agendas, Alternatives and Public Policies. Boston, MA: Little, Brown.
Krapohl, S. (2004) Credible Commitment in Non-independent Regulatory Agencies: A Comparative
Analysis of the European Agencies for Pharmaceuticals and Foodstuffs and Pharmaceuticals,
European Law Journal 10(5): 51838.
Kreher, A. (1997) Agencies in the European Community A Step Towards Administrative
Integration in Europe, Journal of European Public Policy 4(2): 22545.
Lgreid, P., Roness, P.G. and Rubecksen, K. (2005) Autonomy and Control in the Norwegian Civil
Service. Does Agency Form Matter?, in T. Christensen and P. Laegreid (eds) Autonomy and
Regulation: Coping with Agencies in the Modern State. Cheltenham: Edward Elgar.
Lafond, F.D. (2001) The Creation of the European Food Authority: Institutional Implications of Risk
Regulation, European Issues 10: 131.
Levy, R. (2003) Confused Expectations: Decentralizing the Management of EU Programmes, Public
Money and Management 23(2): 8392.
McCormick, J. (2005) Understanding the European Union: A Concise Introduction. 3rd edn.
Basingstoke: Palgrave.
Majone, G. (1997) The New European Agencies: Regulation by Information, Journal of European
Public Policy 4(2): 26275.
Majone, G. (2003) The Politics of Regulation and European Regulatory Institutions, in J. Hayward
and A. Menon (eds) Governing Europe. Oxford: Oxford University Press.
Majone, G. and Everson, M. (2001) Institutional Reform: Independent Agencies, Oversight,
Coordination and Procedural Control, in O. de Schutter, M. Lebessis and J. Paterson (eds)
Governance in the European Union. Luxembourg: Office for Official Publications of the
European Communities.
Martens, M. (2005) Double-hatted National Agencies on the European Scene? A Case Study of the
Environmental Policy Network IMPEL, in M. Egeberg (ed.) Multilevel Union Administration.
London: Palgrave Macmillan.
Metcalfe, J.L. (1999) New Challenges in European Public Management: Designing Interorganizational Networks as a Research Methodology, paper presented at the International
Public Management Network IPMN Workshop, Siena.
Metcalfe, J.L. (2000a) Will Organisational Efficiency Produce Effective Governance?, Journal of
Common Market Studies 38(5): 81741.
Metcalfe, J.L. (2000b) Linking Levels of Government: European Integration and Globalization,
International Review of Administrative Science 66: 11942.
Nugent, N. (2006) The Government and Politics of the European Union, 6th edn. Basingstoke:
Palgrave.
OECD (2002) Distributed Public Governance: Agencies, Authorities and Other Government Bodies.
Paris: Organization for Economic Cooperation and Development Public Management Service.
Office for Official Publications of the European Communities (2003) EEA Strategy 20042008.
Luxembourg.
Ongaro, E (ed.), (2006) Le Agenzie Pubbliche: Modelli Istituzionali e Organizzativi. Saveria Mannelli:
Rubbettino.
Page, E. (1997) People Who Run Europe. Oxford: Calrendon Press.
Pierre, J. (2004) Central Agencies in Sweden: A Report from Utopia, in C. Pollitt and C. Talbot (eds)

420 International Review of Administrative Sciences 74(3)

Unbundled Government: A Critical Analysis of the Global Trend to Agencies, Quangos and
Contractualisation. London and New York: Routledge.
Pollack, M.A. (1997) Delegation, Agency, and Agenda Setting in the European Community,
International Organization 51(1): 99134.
Pollitt, C. and Talbot, C (eds) (2004) Unbundled Government: A Critical Analysis of the Global
Trend to Agencies, Quangos and Contractualization. London: Routledge.
Pollitt, C., Bathgate, K., Caulfield, J., Smullen, A. and Talbot, C. (2001) Agency Fever? Analysis of an
International Policy Fashion, Journal of Comparative Policy Analysis 3(3): 27190.
Pollitt, C., Talbot, C., Caulfield, J. and Smullen, A. (2004) Agencies: How Governments Do Things
through Semi-autonomous Organizations. New York: Palgrave Macmillan.
Randall, E. (2006) Not that Soft or Informal: A Response to Eberlein and Grandes Account of
Regulatory Governance in the EU with Special Reference to the European Food Safety Authority
(EFSA), Journal of European Public Policy 13(3): 40219.
Richardson, J. (2001) European Union: Power and Policy-making, 2nd edn (3rd edn 2006). London:
Routledge.
Sandholtz, W. and Stone Sweet, A (eds) (1998) European Integration and Supranational
Governance. Oxford: Oxford University Press.
Schout, A. (1999) The European Environment Agency (EEA): Heading Towards Maturity, in M.
Everson, G. Majone, J.L. Metcalfe and A. Schout The Role of Specialized Agencies in
Decentralising EU Governance, paper presented to the European Commission, Bruxelles.
Shapiro, M. (1997) The Problems of Independent Agencies in the United States and the European
Union, Journal of European Public Policy 2(1): 27891.
Stevens, A. and Stevens, H. (2001) Brussels Bureaucrats? The Administration of the European
Union. Basingstoke: Palgrave.
Talbot, C. (2004) The Agency Idea: Sometimes Old, Sometimes New, Sometimes Borrowed,
Sometimes Untrue, in C. Pollitt and C. Talbot (eds) Unbundled Government. London: Routledge.
Thatcher, M. (2002) Regulation after Delegation: Independent Regulatory Agencies in Europe,
Journal of European Public Policy 9(6): 95472.
Thynne, I. (2004) State Organizations as Agencies: An Identifiable and Meaningful Focus on
Research?, Public Administration and Development 24: 919.
Van Thiel, S. (2001) Quangos: Trends, Causes and Consequences. Aldershot: Ashgate Publishing.
Verhoest, K., Verschuere, B., Peters, B.G. and Bouckaert, G. (2004) Controlling Autonomous Public
Agencies as an Indicator of New Public Management, Management International 9(1): 2535.
Vos, E. (2000) Reforming the European Commission: What Role to Play for EU Agencies?, Common
Market Law Review 37(5): 111334.
Vos E. (2003) Agencies and the European Union, in L. Verhey and T. Zwart (eds) Agencies in
European and Comparative Law. Antwerpen: Intersentia.
Vos, E. (2005) Independence, Accountability and Transparency of European Regulatory Agencies,
in D. Geradin, R. Mnoz and N. Petit (eds) Regulation through Agencies in Europe: A New
Paradigm for European Governance. Cheltenham, UK and Northampton, MA: Edward Elgar.
Wettenhall, R. (2005) Agency and Non-departmental Public Bodies: The Hard and the Soft Lenses
of Agencification Theory, Public Management Review 7(4): 61535.
Wincott, D. (2001) Looking Forward or Harking Back? The Commission and the Reform of
Governance in the European Union, Journal of Common Market Studies 39(5): 897911.