Content uploaded by Andreja Pegan
Author content
All content in this area was uploaded by Andreja Pegan on Aug 17, 2023
Content may be subject to copyright.
Available via license: CC BY-NC-ND
Content may be subject to copyright.
Citation
Pegan, A. (2017). ‘The Bureaucratic Growth of the European Union’, Journal of Contemporary
European Research. 13(2), 1208-1234
First published at www.jcer.net
Journal of Contemporary
European Research
Volume 13, Issue 2 (2017)
Research Article
The Bureaucratic Growth of the European
Union
Andreja Pegan, Trinity College Dublin
Volume 13 Issue 2 (2017) Andreja Pegan
1209
Abstract
This article explores supranational activism on the example of bureaucratic growth in the institutions
of the EU - operationalised as the number of staff members. The main question is to what extent are
the numbers of staff employed in the EU institutions and bodies dependent on member states and
their decisions to increase the functions of the EU through either treaty changes or the ratifications
of accession treaties. To answer this question, two hypotheses are tested based on time-series data
collected from the EC/EU annual budgets (1959-2016). The first hypothesis predicts an incremental
growth of staff in EU institutions and bodies. The second hypothesis tests whether member states
decisions to reform treaties and ratify accession treaties affect the number of staff members in EU
institutions and bodies. Results show that the growth of EU staff has not been gradual and that
decisions which are under the control of member states partially impact the growth of EU staff.
Decisions and willingness of member states as well as EU institutions are needed to affect the
number of staff members in the EU.
Keywords
European Union, Bureaucratic Growth, Staff, Bureaucracy
This article examines the growth of the European Union’s (EU) bureaucracy, its pattern, and
determinants. Bureaucratic growth is one of several aspects taken by bureaucratisation as a process,
which leads to an increasing role of bureaucracy – a hierarchical, formal, and rule bound structure.
Studies on the EU have dealt with bureaucratisation in respect to bureaucracy’s involvement in
policy-making and the centralisation of power towards technocratic centres in Brussels (Radaelli
1999; Ellinas and Suleiman 2012; Häge 2013; Christiansen et al. 2014; Högenauer and Neuhold
2015). This article contributes to the literature on bureaucratisation in the EU by looking at
bureaucratic growth defined as the quantitative increase of staff (Marshall 1985: 83). While data on
the number of staff members in the EU institutions are freely available, no study has so far
addressed this topic in a scholarly manner. This analysis represents a first attempt to clarify the
pattern and determinants of the staff growth in the EU – a subject which has too often been left to
journalistic and political sensationalism at the expenses of an informed EU electorate.
The main question, here examined, is to what extent the numbers of staff members in the EU
institutions and bodies are dependent on their member states’ decisions to confer new functions on
the EU, through either treaties resulting from intergovernmental conferences or membership
enlargements. Why should we then study the staff numbers of the EU? According to polls, one third
of EU nationals perceives the EU administration as the biggest expenditure in the EU budget and at
the same time wishes it represented the smallest cost (European Commission 2015). While
administration represented the highest expenditure of the European Community (EC) in the 1950s
(approximately 35.5 per cent), its cost dropped in 1962 and stabilized in the 1970s (see Figure 1).
1
Since then, the expenditure for administration has varied from an annual minimum of 4.3 per cent
(1986) to a maximum of 6.0 per cent (1975, 1981, 2006, and 2011) (own calculations based on data
Volume 13 Issue 2 (2017) Andreja Pegan
1210
from EC/EU budgets). Overall, the reforms of the EU administration do not affect the EU budget in
ways that reforms in agriculture or cohesion policy funds would, as the largest budget items.
Nevertheless, the attention that citizens attribute to the topic, as lastly witnessed in the Brexit
referendum campaign of 2016, and its prominence in national media call for scholarly attention.
Studying bureaucratic growth in the case of the EU is also important from a fiscal point of view and
the willingness of EU member states to finance the EU budget. In the aftermath of the 2008 financial
crisis and the debt crisis that followed, many European governments adopted legislation limiting
public spending, including staff cuts in national administrations. In line with austerity measures,
several member states, all net contributors to the EU’s budget, demanded to cut its staff by as much
as 12 per cent (Brand 2012; Council of the European Union 2012: 3). In 2013, a decision to reduce
staff by 5 per cent before 2018 was adopted amid inter-institutional tensions.
The article is part of a special issue dedicated to supranational activism in the EU. It contributes to
the special issue in so far as it explores the pattern and causes for bureaucratic growth to provide
evidence of supranational activity. The assumption is that the causes underlying bureaucratic growth
(EU’s membership enlargements and the extension of competences through founding, budgetary,
and reform EC/EU treaties) determine the supranational (controlled by EU institutions) and/or
intergovernmental (controlled by EU member states) character of the EU. If enlargements and
treaties, which have to be approved by EU member states as intergovernmental actors, affect the
growth of the EU staff, then bureaucratic growth is not an indicator of supranational activity, but
rather an outcome of intergovernmentalism. Findings show that member states decisions on
enlargement and treaty changes combined with the willingness of EU institutions are necessary to
affect the number of staff members working in the EU. The analysis is organized around six sections.
The following section provides an overview of the staff organization in the EU. Then the theoretical
framework, hypotheses and analysis are presented and discussed.
EU STAFF
In 2016, the EU employed 39,715 staff members (Table 1). Their rights and conditions of
employment are laid down in the Staff Regulations of the EU (2013). Since 1956, when the first staff
regulation was adopted, these have been reformed 131 times before the last amendment in 2013
(Regulation 1023/2013). The Staff Regulations are divided into two parts to separate regulations
applying to the civil servants of the EU (Staff Regulations for Officials) and regulations applying to all
other employees (Conditions of Employment of Other Servants: CEOS). The civil servants of the EU
are permanent officials with the highest level of job security and privileges. Article 1 of the CEOS
determine five categories of other staff: temporary, contract, local staff, special advisers, and
parliamentary assistants. The administration heading of the EU annual budget covers expenses for
civil servants and temporary staff (Figure 1). All other staff are financially covered by budgets
pertaining to individual institutions. The work of EU staff is divided into three categories:
administrators (AD) with university diploma, assistants (AST) with post-secondary diploma, and
assistants-secretaries (AST/SC) with secondary education. Three quarters of all EU staff are
employed by the European Commission, European Parliament, and Council. This study analyses the
evolution of staff belonging to permanent and temporary categories (either AD, AST, or AST/SC),
whose remuneration are included in the administration heading of the EU budget rather than under
the headings of individual institutions.
Volume 13 Issue 2 (2017) Andreja Pegan
1211
Table 1: Overview of staff in the EU
Institution
1959
1969
1979
1989
1999
2009
2016
Average
growth rate
1959-2016
EU (all)
N
2,591
9,068
16,162
23,483
30,814
39,919
39,715
5.2 %
% EU
100
100
100
100
100
100
100
Commission
N
1,930
7,707
11,649
16,309
21,603
26,209
24,044
5.0 %
% EU
74.0
85.0
72.0
69.0
70.0
65.0
61.0
Council
N
264
569
1,547
2,165
2,621
3,512
3,040
4.7 %
% EU
10.0
6.2
9.5
9.2
8.5
8.8
7.7
EP
N
315
528
1,977
3,405
4,101
5,800
6,762
6.0 %
% EU
12.0
5.8
12.0
14.5
13.0
14.5
17.0
ECJ
N
80
110
315
733
961
1,931
2,073
6.3 %
% EU
3.0
1.2
1.9
3.0
3.0
4.8
5.2
ECA
N
24
214
377
552
880
889
12.8 %
% EU
0.2
1.0
1.6
1.8
2.0
2.2
EESC
N
129
325
494
737
704
670
5.5 %
% EU
1.4
2.0
2.1
2.3
1.7
1.7
CoR
N
221
502
496
4.6 %
% EU
0.7
1.2
1.2
Ombudsman
N
17
63
66
9.2 %
% EU
0.1
0.3
0.2
EDPS
N
37
47
8.9 %
% EU
0.1
0.1
EEAS
N
1,628
0.0 %
% EU
4.1
Source: EC/EU Budgets and own calculations. Note: Average growth rates ():
, where stands for
the number of staff members in the selected institution in a given year. Abbreviations: European Commission
(Commission), Council of the EU (Council), European Parliament (EP), European Court of Justice (ECJ), European Court of
Auditors (ECA), European Economic and Social Committee (EESC), Committee of Regions (CoR), European Data Protection
Supervisor (EDPS), European External Action Service (EEAS).
Volume 13 Issue 2 (2017) Andreja Pegan
1212
THEORETICAL FRAMEWORK
To analyse the bureaucratic growth of the EU as a form of supranational activity, the following
section reviews three explanations for bureaucratic growth (interest, function, and normative based
explanations) and their applicability to the growth of staff in the EU. The assumption is that the
causes underlying bureaucratic growth determine its supranational character (controlled by EU
institutions) or inter-governmental activity (controlled by member states). The main hypothesis is
that bureaucratic growth results from increased functions decided by member states.
Bureaucratic growth and its causes
The causes of bureaucratisation have been addressed first by Max Weber (1968). According to him,
the function of bureaucracy is to organize and solve problems. As a new problem or demand arises
in society, bureaucracy will grow to solve that problem. Weber also argued that bureaucratic
organization will prevail, since it represents the most efficient and rational way of organization. The
contemporary views of public administration theorists are more nuanced, as empirical evidence
indicates bureaucracies grow due to many (inter-related) processes and events (Peters 2001: 15).
INTEREST-BASED GROWTH
Bureaucratic growth has often been studied using rational-economic models (North 1991; Pollack
2007; Shepsle 2008). Agents, either politicians or bureaucrats, are conceived as individuals who
benefit from bureaucratization (Niskanen 1971; Miller and Moe 1983; Meltzer and Richard 1983;
Meyer 1987; Vaubel et al. 2007; Hibou 2015). For example, unelected officials use their information
advantage over elected officials to increase the budget allocated to their bureaus (Niskanen 1971).
Elected politicians, on the other hand, allow an increase of public spending in order to satisfy
electoral promises (Wilson 1992, p. 209). However, the interest of agents in bureaucratic growth is
not self-evident. Peters (2001: 13) has pointed out that the interests of bureaucrats can be to reduce
bureaucratisation, since more expenditures or more staff can create managerial difficulties.
Similarly, an elected official might want to reduce bureaucracy to calm the public inertia concerning
an inefficient bureaucratised government. Nonetheless, rational-economic models conceive
bureaucratisation as a process controlled by agents.
In the case of the staffing policy of the EU it is difficult to identify all interested parties and apply a
rational approach theorising individual interests at the micro level. While there is a tacit agreement
among the institutions of the EU to not interfere in each other’s budget planning, less is known
about the individual positions of member states and the weight they put on administrative issues.
While it is acknowledged that the staff of the EU have an interest in growth, the argument raised by
Peters (2001) shows that this is not necessarily the case. As the purpose of this article is to analyse
the pattern of bureaucratic growth in the EU, and, given difficulties in identifying all parties and their
interests in the staffing policy of the EU, the analysis adopts a macro perspective. A macro
perspective represents a starting point to uncover the basic mechanisms of bureaucratic growth also
because the number of EU staff has never been analysed before. To analyse the change in the
numerical size of EU staff, the article leans on theoretical approaches which emphasise the
environment and its structure rather than the agent. Two views on the environment have emerged.
The first view considers bureaucratic growth as the result of normative arguments or ideas from the
larger societal environment (Christensen 1997; Bulmer 2009; Jenson and Mérand 2010). This article
leans on the normative view to formulate the first hypothesis regarding the staff growth trend in EU
Volume 13 Issue 2 (2017) Andreja Pegan
1213
institutions. The second view assumes bureaucratic growth has a functional property, i.e. growth
fulfils needs arising from the immediate organisational environment (Wilensky 1975; Schmitter
2005). The article leans on this theoretical approach to formulate the second hypothesis on the
causes of growth. Empirically, the second hypothesis builds on the first one.
NORMATIVE-BASED GROWTH
The normative idea advances the argument that bureaucracies respond to environmental demands
on conformity (Olsen and March 1989; Jenson and Mérand 2010). The existence and development of
bureaucracies are rooted in ideological arguments for their being and their claims to (functional)
domains, rather than their actual functions or activities. Theoretically, this process is known as
institutional decoupling, i.e. the existence of institutions is only loosely coupled with their purpose
(Meyer and Rowan 1977; Westphal and Zajac 1994).
2
The normative idea of bureaucratisation
conforms to the theoretical notion of the logic of appropriateness (Olsen and March 1989).
According to this notion the behaviour of institutions results from norms and ideas that are
perceived as appropriate and legitimate in the wider society. On the contrary to interest based
theories, behaviour is not the outcome of purposive actions or the product of conscious decisions by
instrumentally orientated actors (DiMaggio and Powell 1991: 8). In regards to bureaucratisation, this
has two implications. First, bureaucracies can either grow or decline, since both behaviours are
legitimate. Second, perceptions over legitimacy take longer time to evolve and change (Meyer and
Rowan 1977). Hence, bureaucratisation (growth or decline) is stable and gradual rather than radical:
H1: The number of EU staff members grows gradually.
The EU was set up in the aftermath of the Second World War, which coincides with the general
growth pattern of international organizations (Vaubel et al. 2007; Šabič 2008). Moreover, as the EU
integration progressed, the idea of its growth became legitimised through the greater scope of EU
collaboration. Hence, it is expected that staff of the EU institutions have grown based on EU’s
legitimacy. However, at least since the financial crisis in 2008, and the EU’s ability to respond to the
crisis questioned, dissatisfaction with EU bureaucracy has steadily been growing. A clear
manifestation of this dissatisfaction, and thus the weakening of EU’s legitimate growth, is the 2013
rule to reduce staff of the EU institutions by 5 per cent (Regulation 1023/2013). This is expected to
have weakened EU’s legitimacy to grow. Hence, a decrease of staff can be expected from the 2010s
onwards. In case a change in the perceptions on the EU’s legitimate growth had taken effect before,
a decline should be visible at other moments of legitimacy crisis for the EC/EU, such as the ‘Empty
chair crisis’ between 1965-66, when France refused to take its seat in the Council of Ministers, the
economic stagnation or the ‘Eurosclerosis’ during the 1970s and 1980s, the crisis of the Santer
Commission in 1999, and failed referenda on EU treaties (in Ireland in 2001 and 2008, France and
the Netherlands in 2005). Events that change the direction of subsequent institutional developments
are conceptualised as critical junctures or moments giving rise to a new, but lengthily path
dependent process (e.g. staffing decline) (Pierson 2004; Mahoney and Thelen 2010).
FUNCTION-BASED GROWTH
The functional explanation of bureaucratic growth has a clear rational component, since growth is
the response to actual activities, such as economic output (Peters 2001: 2-8; Wilensky 1975) and
requirements to provide for previously unmet public needs (Wilson 1992: 210). This logic has been
used to explain the proliferation of international organizations and their bureaucracies in the
Volume 13 Issue 2 (2017) Andreja Pegan
1214
aftermath of the Second World War. Due to technical progress and globalization, international
organizations have advanced the economic wealth of states and helped them manage increased
economic interdependences (Haas 1958; Schmitter 2005). Accordingly, the pooling of resources in
one area should gradually lead to the transfer of competences in other areas as well. Hence,
bureaucracy is set to grow to accommodate newly acquired tasks.
In the case of the EU, new functions can be interpreted as the increase of institutional and policy
competences of the EC/EU and the entry of new member states. Both processes are the result of
treaties and are subjects to national ratifications, which means that member states have veto
power.
3
Treaties – new or founding (e.g. TEC/EURATOM and TEU/TFEU) as well as reform or those
amending already functioning treaties (e.g. Merger Treaty, Single European Act, Amsterdam Treaty,
Nice Treaty, and Lisbon Treaty) - have changed the institutional balance in the EU, extended its
powers, and consequently its workload. Therefore, it is hypothesised that whenever a treaty
changes the institutional balance and/or extends the powers of the EU, new staff will be recruited to
cope with the increasing workload. Given the availability of data, all the treaties, which have
affected either the policy or the institutional functions of the EC/EU are included in this study. These
are summarised in Table 2. Besides treaties, membership enlargements or the accession of new
member states can also be considered an expansion of functions, since needs to accommodate new
languages and manage more national interests are greater. The EU has expanded its membership
seven times as shown in Table 3. All of these enlargements are included in this study. The hypothesis
is:
H2: The numbers of staff in the EU grow as a result of newly acquired functions arising from treaties
changing its policy and institutional competences and membership enlargements.
To distinguish functions, the article refers to competences, when implying functions arising from
treaties; and to enlargement, when referring to functions, arising from the entry of a new member
states.
Table 2 EU Competences
Treaty
Entry into force
Examples of policy and institutional
competences
Treaty establishing the European
Economic Community (TEC) and Treaty
establishing the European Atomic Energy
Community (Euratom)
(Rome Treaties, founding treaties)
1 January 1958
Establishment of the European Economic
Community and Euratom
Single market for goods, labour, services,
and capital across
Common agriculture and transport policies
Consultative power
Treaty amending Certain Budgetary
Provisions of the Treaties establishing the
European Communities and of the Treaty
establishing a Single Council and a Single
Commission of the European
Communities (Luxembourg Treaty,
budgetary treaty)
1 January 1971
Budgetary treaty
EP gains budgetary powers on non-
compulsory expenditures
Treaty establishing a Single Council and a
Single Commission of the European
1 July 1967
The institutions of the EESC, EEC and
Volume 13 Issue 2 (2017) Andreja Pegan
1215
Treaty
Entry into force
Examples of policy and institutional
competences
Communities
(Merger Treaty, reform treaty)
Euratom are merged
Treaty amending Certain Financial
Provisions of the Treaty establishing the
European Communities and of the Treaty
establishing a Single Council and a Single
Commission of the European
Communities
(Brussels treaty, budgetary treaty)
1 June 1977
Budgetary powers
EP can reject the budget as a whole
Establishment of the ECA
Single European ACT
(SEA, reform treaty)
1 July 1987
Reform of legislative process which would
ease the completion of the single market
Introduced the European Political
Community (later CFSP)
Cooperation and assent procedure
Treaty of the European Union
(TEU; Maastricht Treaty; founding
treaty), Treaty on the Functioning of the
EU (TFEU)
1 November
1993
Three pillar structure (single market, CFSP,
and JHA)
Economic and Monetary Union (including
euro)
Codecision (15 legal bases)
Treaty of Amsterdam
(reform treaty)
1 May 1999
Employment policy
Extension of codecision (32 legal bases)
Treaty of Nice
(reform treaty)
1 February 2003
Extension of codecision (37 legal bases)
Reforms for enlargement
Treaty of Lisbon
(reform treaty)
1 January 2009
Extension of codecision (85 legal bases)
Strengthened role of the Commission in
economic governance
ECJ becomes fully competent in JHA
Establishment of EEAS
Source: http://eur-lex.europa.eu/collection/eu-law/treaties.html (22 August 2016)
Volume 13 Issue 2 (2017) Andreja Pegan
1216
Table 3 EU Enlargements
Year
N
New member states
1973
9
UK, Ireland, Denmark
1981
10
Greece
1986
12
Spain, Portugal
1995
15
Austrian, Finland, Sweden
2004
25
Slovenia, Cyprus, Malta, Hungary, Czech Republic, Slovakia,
Latvia, Lithuania, Estonia, Poland
2007
27
Bulgaria, Romania
2013
28
Croatia
DATA AND METHODS
To provide evidence for the hypotheses, descriptive methods (H1) and regression analysis (H2) are
used on time-series data. The dependent variable is bureaucratic growth. The independent variables
are functions arising from treaties and membership enlargements. A summary of statistics is
provided in Table 4.
Table 4 Descriptive statistics
Variables
Label
Observations
Min
Max
Enlargement
enlar
58
6
28
Competences/treaty reforms
comp
58
0
1
Annual EC/EU Budget (million EUR)*
budget
58
48.4
144,000
Trend
trend
58
1
58
trend2
58
1
3364
Population in the EU (million)**
eu_pop
56
171.6
508.5
Number of staff members in the…***
EU
58
2,589
41,895
Commission
58
1,930
26,279
Council
58
264
3,572
EP
58
315
6,786
ECJ
58
80
2,073
Volume 13 Issue 2 (2017) Andreja Pegan
1217
Variables
Label
Observations
Min
Max
EECA
57
42
753
ECA
49
19
891
CoR
21
221
537
Ombudsman
21
13
67
EDPS
15
15
47
EEAS
7
1625
1670
Notes: * European Commission (2009). EU Budget Financial Report 2008. Annex 2. Luxembourg:
Publication office of the EU. ** Eurostat data (retrieved 29 June 2016), *** EC/EU annual budgets (final
adoption) from the Official Journals of the EU.
Dependent variable
Scholars have operationalised the growth of bureaucracy in several ways: staff number, job titles,
and reorganizations (e.g. changes of levels of supervision or hierarchy and management, promoting
criteria, and personnel procedures). In this article, bureaucratic growth is treated in terms of the
change in the number of employees. This choice has been made, since staff numbers can be traced
for several institutions and bodies of the EU from their establishment to the present. It is an
accurate and standardized measure across institutions, which does not take different forms across
EU institutions and bodies at different times. By looking at a standardized indicator, the article aims
at identifying the link between specific events and the bureaucratisation in the EU’s institutions as a
whole. Even though the growth of staff is only one aspect of bureaucratisation, it is expected that
results can be generalized (see Vaubel et al. 2007: 13).
Data on the number of staff members were collected from the establishment plans included in the
EC/EU annual adopted budgets. All of them are available in the Official Journals of the EU.
4
The
collected data allow to observe the size and growth of permanent and temporary EU staff from 1959
up until 2016 and cover the following institutions and bodies: Council of the EU, European
Commission, European Parliament (EP), European Court of Auditors (ECA), European Court of Justice
(ECJ), European Committee of the Regions (CoR), European Data Protection Supervisor (EDPS),
European External Action Service (EEAS), European Economic and Social Committee (EESC), and
European Ombudsman.
5
Given the year of establishment of each institution or body, there are 58
observations for the Council, EP, Commission, and, ECJ; 56 for the EESC, 49 for the ECA, 21 for the
CoR and Ombudsman, 15 for the EDPS, and 7 for the EEAS. The number of staff members does not
distinguish between the AD, AST, and AST/SC categories.
GRADUAL GROWTH
There is no universal definition as to what comprises gradual change of staff members. Between
1959 and 2016 the EU staff grew with a rate of 5.2 per cent (standard deviation is 9.5 per cent). In
the same period, they registered five instances of negative growth rate. The average of these is a
staff decrease of 1.5 per cent with a standard deviation of 1.0 per cent. In 2013, the EU decided to
reduce its staff by 5 per cent in a five-year period, which means that on average the EU should be
Volume 13 Issue 2 (2017) Andreja Pegan
1218
able to reduce its staff by at least one per cent a year. Since in the context of EU Staff Regulations it
is easier to hire than to dismiss staff, the decision on what constitutes gradual growth is based on
EU’s negative growth rates of staff or the rate an institution can control staff lay-offs. Based on the
average set out by the EU itself in 2013 and given the average of negative growth rates and its
standard deviation, the yearly benchmark for gradual growth is set to an absolute change of 2.4 per
cent (i.e. average + standard deviation, with downward rounding). However, this is an arbitrary
measure, which is relatively strict given that it is based on the average of negative growth rates and
that the EU’s staff growth rate has varied on average by 5.2 per cent a year.
Independent variables
Functions are defined as the increases of policy and institutional competences and membership
enlargements. The increase of competences is recorded with a dummy variable, which records the
entry into force of a treaty in a given year (see Table 2).
6
The increase of membership is recorded as
a continuous variable of the number of member states in the EU (see Table 3).
7
In addition, the
increase of functions is measured in terms of the population of member states as an alternative
measure to enlargements.
Control variables
All the time series used for this analysis grow over time with an upward trend: the number of staff
increases over time in an approximately linear way. In this circumstance, the spurious regression
problem occurs, which means that unobserved trending factors affect the dependent variable and
can thus be correlated with independent variables (Wooldridge 2009). To avoid this problem, a
trend variable exhibiting quadratic growth (trend, trend2) is included in the regression. The trend
captures the autonomous average annual increase of staff because of reasons that are not directly
observed. Staff increase because of a general increase of the size of the EU in terms of membership
and increased workload due to competences. The problem is that these factors are changing
together over time and in one direction making it harder to disentangle their partial effects on staff.
By including a time trend, we can more precisely estimate what is the partial effect of each
independent variable since coefficients are calculated based on their variation from the trend.
Finally, control variables include the amount of the EU budget (expressed in EUR).
ANALYSIS
Because of serially correlated errors in the time series regression model and highly persistent
developments of the EU institutions captured in the trend variable, statistics and standard errors
from the ordinary least square regressions are not valid. Therefore, the estimation method used
here is the feasible generalised least square (FGLS-Prais Winsten). Alternatively, one can use
dynamic estimation models with lag operators. However, the FGLS procedure is preferred due to the
small sample where losing one observation, as it happens in dynamic models can have large effects
on results. Moreover, modelling the time dependence requires a deeper understanding of the
political dynamics. The latter is a limit for this research.
Regressions are performed on the aggregate level (i.e. the number of all staff in the EU) and on
individual institutions, where the number of observations is at least 49 (Council, EP, Commission,
Volume 13 Issue 2 (2017) Andreja Pegan
1219
ECJ, EESC, and ECA) to secure enough observation for a meaningful test. Regression models are set
up for testing the effect of independent variables alone (enlargement in Model 1 and competences
in Model 2) and together with control variables (Models 3, 4, and 5). Models 3 and 4 test for the
effects of enlargement and competences respectively, while controlling for the EU budget and trend
variables. Model 5 includes both dependent variables (enlargement and competences) and the
control variables. Model 6 replaces the variable for enlargement with the population number in the
EU as an alternative and consistency test to Model 5.
Findings. Pattern (H1)
Data on the number of staff members working in the EU's institutions show the EU's bureaucracy
has been growing continuously until 2011 (Figure 2a). Its growth has not been gradual, which is
shown by a steep growth line of the number of staff members (Figure 2) and by several growth rate
peaks (Figure 3). On average, the EU’s institutions have been growing by 5.2 per cent per year (Table
1). The average growth rate of the staff in the Commission, Council, CoR, and EEAS has been lower
than 5.2 per cent, while the EP, ECJ, ECA, EESC, Ombudsman, and EDPS record higher average rates.
Gradual growth (positive change between zero and 2.40 per cent per year) is observed for short
periods of time (one to five years) in all institutions. On the aggregate level, the longest period of
incremental growth is recorded between 1997-2002 and 2012-2016 (Figure 3a). The EESC (Figure 3g)
has experienced the longest period of gradual growth (seven years between 2008-2015) compared
to all other institutions and bodies. Among the four best staffed institutions and those for which
data is available since 1959, the Council and EP register the most periods of gradual growth (five
periods of gradual growth lasting over two years) and the Commission the least (only two).
8
The ECJ
has undergone 13 instances of incremental growth, however, only three of these were longer than
two years. Among the younger institutions and bodies, the EEAS registers constant gradual growth
(Figure 3k), while the ECA records the highest growth rate in a year (1978, Figure 3f).
Unsurprisingly, the overall growth pattern of the EU bureaucracy is led by the best staffed EU
institutions (Commission, Council, and EP). In the case of the Commission, its relative peak, absolute
maximum, and relative minimum coincide with those of the EU’s overall pattern (all happened in
2010, 1968, and 2011 respectively; Figure 3a, b). This means that any substantial change in the
number of staff members should include change in the number of employees of the Commission.
Between 2011 and 2016, the staff of the Commission have decreased on average by 1.45 per cent
and the highest decline, 4.6 per cent, was recorded in 2011 (Figure 3b). The institution that has thus
far endured the biggest staff cut relative to others is the Council (on average 2.6 per cent between
2011-2016, with a maximum of 11.1 per cent in 2011; Figure 3c).
9
The EP and ECJ record the lowest
drops, 0.7 (2015) and 0.6 (2014) respectively (Figure 3c and Figure 3e). Moreover, while the
Commission and Council have been cutting their staff since 2011, the EP has done so only since
2015.
Volume 13 Issue 2 (2017) Andreja Pegan
1220
Figure 2 Number of staff members
Figure 3 Growth rates
Volume 13 Issue 2 (2017) Andreja Pegan
1221
Causes (H2)
The regression models explaining the number of staff members in the EU institutions show that
variables measuring EU enlargements are consistently statistically significant for all institutions
except for the Commission (Table 6, Models 1, 3, and 5). On the aggregate level, EU membership
enlargement increases on average the number of staff members by 228 to 340 depending on the
model (Table 5, Models 1, 3, and 5). The size effects in the Council and ECJ are similar. The average
change of staff members in the Council varies between 50 to 55 (Table 7, Models 1, 3, and 5) and in
the ECJ it is 45 (Table 9, Models 1, 3, and 5). The effect of enlargements is smaller in the EP than it is
in the Council and ECJ. On average, the EP gains 30 staff members due to one state joining the EU
(Table 8, Model 5). The effects on the EESC and EAC are lower than 15 staff members per
enlargement (Table 10 and Table 11, Models 1, 3, and 5). The increase of population shows the same
statistical significances as the variable on the number of member states. However, the population
increase of one million has a very small effect (Model 6 in Tables 5-11.).
Contrary to enlargement, there is no evidence that an increase of competences affects the number
of staff members in EU institutions and bodies, i.e. the variable measuring the introduction of new
treaties is not statistically significant in any of the models. Except for the ECJ, the trend variable is
significant in all models and it has the greatest effect. This means that most of the changes in the
number of staff members happen as a result of unobserved factors. For example, trending growth is
responsible for an increase of 765 staff members in the EU, 626 in the Commission, 85 in the
Council, 77 in the EP, 24 in the EESC, and 16 in the ECA (Tables 5-11, Model 5). The results also show
a statistically significant and consistent effect of the total EU budget on the number of staff at the
aggregate level and the Commission, but no other institution or body. However, the increase of the
annual budget by one million euro has a negligible effect.
Table 5 Prais-Winsten estimates for EU staff (aggregate)
VARIABLES
Number of staff members in EU institutions and bodies
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
339.8***
228.1***
213.2**
(-87.8)
(-76.8)
(-80.8)
comp
-324.7
-280.5
-157.2
-194.5
(-3371.3)
(-227.9.3)
(243.7)
(234.9)
budget
0.073***
0.090***
0.071***
0.061**
(-0.025)
(0.026)
(0.025)
(0.023)
trend
763.3***
7552***
7654***
656.2***
(-53.5)
(66.0)
(55.5)
(-43.0)
trend2
-5.687***
-4.923***
-5.549***
-3.928***
(-1.428)
(-1.581)
(1.459)
(1.350)
eu_pop
2.17e-05***
Volume 13 Issue 2 (2017) Andreja Pegan
1222
VARIABLES
Number of staff members in EU institutions and bodies
(6.15e-06)
Constant
339.8***
21281.0
228.1***
1057.0
-403.7
-2,566*
(-87.8)
(17,506)
(-76.8)
(794.2)
(982.0)
-2,566*
Observations
58
58
58
58
58
56
R-squared
0.0726
0.936
0.97
0.98
rho
0.996
-0.0246
0.791
0.687
0.640
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
Table 6 Prais-Winsten estimates for staff in the European Commission
VARIABLES
Number of staff members in the European Commission
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
151.8**
62.94
56.2
(-74.4)
(-50.4)
(55.7)
comp
-146.3
-116.8
-85.37
-86.98
(249.2)
(195.9)
(209.6)
(208.2)
budget
0.048**
0.055***
0.047**
0.044**
(-0.020)
(0.020)
(-0.020)
(-0.020)
trend
624.8***
622.0***
626.2***
591.5***
(-48.7)
(49.8)
(50.1)
(46.4)
trend2
-5.897***
-5.735***
-5.843***
-5.331***
(-1.183)
(1.191)
(-1.206)
(1.174)
eu_pop
7.06e-06
(5.51e-06)
Constant
10,982
13,360
252.6
673.2
294.8
-605.6
(8,592)
(10,071)
(-799.6)
(680.6)
(851.3)
(1,234)
Observations
58
58
58
58
58
56
R-squared
0.891
0.878
0.885
0.914
Volume 13 Issue 2 (2017) Andreja Pegan
1223
VARIABLES
Number of staff members in the European Commission
rho
0.991
0.993
0.783
0.794
0.790
0.753
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
Table 7 Prais-Winsten estimates for staff in the Council of the EU
VARIABLES
Number of staff members in the Council of the EU
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
55.3***
51.7***
50.3***
(-6.4)
(-4.9)
(5.5)
comp
-47.2
-43.4
-16.3
-25.6
(34.2)
(26.5)
(17.7)
(17.5)
budget
0.001
0.009*
0.001
0.002
(-0.002)
(0.005)
(0.002)
(0.002)
trend
85.1***
82.0***
85.2***
67.4***
(-10.5)
(11.4)
(10.5)
(9.3)
trend2
-0.902***
-0.856***
-0.893***
-0.760***
(-0.197)
(0.275)
(-0.204)
(-0.227)
eu_pop
4.67e-06***
(7.89e-07)
Constant
842.9
1,714
-266.2*
72664
-254.4*
-753.4***
(-742.6)
(1,231)
(-143.1)
(116.6)
(143.4)
(169.4)
Observations
58
58
58
58
58
56
R-squared
0.313
0.78
0.588
0.770
0.822
rho
0.987
0.991
0.87
0.88
0.875
0.842
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
Volume 13 Issue 2 (2017) Andreja Pegan
1224
Table 8 Prais-Winsten estimates for staff in the European Parliament
VARIABLES
Number of staff members in the European Parliament
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
49.9***
25.0**
29.3***
(-13.2)
(-11.7)
(10.5)
comp
31.4
36.0
52.2
49.37
(42.5)
(41.18)
(41.1)
(40.1)
budget
0.006
0.0108**
0.006
0.005
(-0.005)
(0.004)
(0.004)
(0.004)
trend
77.9***
75.1***
77.0***
63.7***
(-20.1)
(21.4)
(20.4)
(21.2)
trend2
0.246
0.245
0.232
0.367
(-0.344)
(0.373)
(0.347)
(0.363)
eu_pop
3.44e-06***
(8.20e-07)
Constant
2,670
-15,378***
-50.71
122.5
-70.02
-496.2***
(2,509)
(3,247)
(-171.9)
(138.6)
(168.0)
(184.9)
Observations
58
57
58
58
58
56
R-squared
0.013
0.739
0.692
0.734
0.738
rho
0.996
1.006
0.901
0.911
0.906
0.905
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
Volume 13 Issue 2 (2017) Andreja Pegan
1225
Table 9 Prais-Winsten estimates for staff in the European Court of Justice
VARIABLES
Number of staff members in the European Court of Justice
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
44.7***
44.7***
45.2***
(-3.2)
(-3.4)
(3.6)
comp
-23.0
-19.2
5.2
-5.7
(-29.2)
(23.2)
(5.7)
(10.3)
budget
5.44E-05
0.007
5.51e-05
0.002
(-1.36E-03)
(0.005)
(0.001)
(0.002)
trend
3.315
-0.274
3.249
-13.32**
(-3.191)
(4.048)
(3.178)
(5.154)
trend2
0.255***
0.303
0.253***
0.368***
(-0.082)
(0.190)
(0.083)
(0.111)
eu_pop
3.49e-06**
(1.34e-06)
Constant
-819.9***
-1,271**
-209.7***
84.58***
-212.2***
-479.1**
(-185.6)
(537.8)
(-30.72)
(22.50)
(31.76)
(215.3)
Observations
57
57
58
58
58
56
R-squared
0.872
0.031
0.956
0.804
0.957
0.913
rho
1.018
1.017
0.882
0.839
0.881
0.828
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
Volume 13 Issue 2 (2017) Andreja Pegan
1226
Table 10 Prais-Winsten estimates for staff in the European Economic and Social Council
VARIABLES
Number of staff members in the European Economic and Social Council
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
14.2***
13.0***
12.5**
(-3.9)
(-4.8)
(5.1)
comp
-13.2
-13.14*
-6.49
-9.2
(8.5)
(7.8)
(6.6)
(-6.3)
budget
-0.001
0.002
-0.001
-0.001
(-0.002)
(0.002)
(-0.002)
(0.002)
trend
24.0***
23.1***
24.0***
18.1***
(-3.57)
(3.04)
(3.58)
(3.36)
trend2
-0.217***
-0.242***
-0.215**
-0.158**
(-0.081)
(-0.081)
(0.082)
(0.079)
eu_pop
1.05e-06***
(3.75e-07)
Constant
187.8
392.8
-124.4**
-46.74
-121.5**
-207.5***
(-150.3)
(239.5)
(-54.26)
(37.88)
(54.6)
(65.26)
Observations
57
57
57
57
57
56
R-squared
0.092
0.456
0.718
0.464
0.591
rho
0.96
0.977
0.861
0.722
0.860
0.813
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
Volume 13 Issue 2 (2017) Andreja Pegan
1227
Table 11 Prais-Winsten estimates for staff in the European Court of Auditors
VARIABLES
Number of staff members in the European Court of Auditors
Model 1
Model 2
Model 3
Model 4
Model 5
Model 6
enlarg
14.2***
11.2***
10.4***
(-1.5)
(-1.8)
(2.0)
comp
-15.8
-14.5
-8.0
-11.5
(13.2)
(10.1)
(8.9)
(9.2)
budget
0.001
0.002*
0.001
0.001
(-0.001)
(0.001)
(0.001)
(0.001)
trend
16.1***
14.9***
16.3***
11.1**
(-4.3)
(4.8)
(4.4)
(5.5)
trend2
-0.063
-0.043
-0.060
-0.010
(-0.065)
(0.089)
(0.067)
(0.088)
eu_pop
6.76e-07
(4.40e-07)
Constant
216.7
453.3
-231.2***
-146.9***
-227.4***
-239.5***
(-252.3)
(399.3)
(-54.83)
(49.82)
(54.36)
(79.44)
Observations
49
49
49
49
49
48
R-squared
0.179
0.827
0.718
0.821
0.749
rho
0.99
0.994
0.852
0.869
0.860
0.871
Notes: Semi-robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1
DISCUSSION
Based on the findings presented above, there is limited support for a gradual growth of EU staff (H1)
and only partial support for growth as a result of new EU functions (H2). The number of EU staff
members has changed in cycles of sudden growths followed by short periods of incremental growth.
The uneven growth of staff members does not provide support for the normative explanation of
growth, where slowly changing ideas determine the pace of bureaucratic growth. Despite an uneven
growth pattern, the EU has increased the number of its staff members starting from 1959. This
meets the article’s expectation that the EU would increase the number of its staff members after its
establishment, since it operated in an environment legitimising its growth.
Volume 13 Issue 2 (2017) Andreja Pegan
1228
Since 2011, the EU institutions and bodies are experiencing for the first time a period of incremental
decline. This can be attributed to the changing conditions in the environment (e.g. economic
recession and slow recovery), which are less favourable to bureaucratic growth and the EU more in
general. While the EC/EU and their bureaucracies have witnessed unfavourable conditions for
growth before the financial crisis in 2008, these had never been formalised into a Staff Regulation
mandating the reduction of EU staff as in 2013. Hence, the 2008 financial crisis resembles a critical
juncture for the staffing level of the EU institution more than any other event. However, it is too
early to determine whether the declining growth rates of EU staff will turn into a long-term pattern,
thus, providing stronger evidence for a new pattern of negative bureaucratic growth and path-
dependency.
While the decline of staff levels has happened in an unfavourable environment for growth, it is
important to note that it has been primary sought and initiated by member states. It is unlikely that
the EU would have decreased its staff numbers following the financial crisis without its member
states requesting it. It can be said that the current decreasing levels show that EU institutions are,
one the one hand, dependent on ideas as to what behaviour is legitimate (growth or decline) and,
on the other hand, constrained with member states decisions (e.g. pressure for staff cuts in the
2010s).
However, an incremental decline does not show in all EU institutions and bodies (for example the
EP), which be a manifestation of tensions among the actors involved in how to achieve the 5 per
cent reduction target. The EP’s opinion on the Commission proposal to reform Staff Regulations is a
clear example of such tensions. In its report, the Legal Affairs Committee of the EP rejected the
Commission proposal for ‘automatic’ reductions across all institutions (European Parliament 2012:
67). According to the MEP and responsible rapporteur Dagmar Roth-Behrendt (S&D), enlargement
and newly acquired competences under the Lisbon Treaty require ‘tailor-made’ solutions per
institution (European Parliament 2012: 69). The staff cuts that started in 2011 show on the one
hand, that the staff of the EU institutions are not completely free from member states demands to
reduce staff. On the other hand, not all the EU institutions have reduced their staff in the same
manner; some have even recorded an increase (ECJ and EDSP, Figure 2e-j), which shows that the EU
institutions respond differently to proposal regulated in EU decisions. This shows the potential that a
study on the positions held by the EU institutions and member states could bring to our
understanding of the EU’s staffing policy. However, considering the sensitivity of the subject, it is
unlikely that the involved actors would divulge their specific interests when negotiating the
administrative budget of the EU.
While the findings of this article show that the staff numbers of the EU increased because of
enlargements, there is no statistical evidence that treaties affect staff size. Given the statistically
insignificant effect registered by treaties, it is difficult to legitimise the opt out from the 5 per cent
reduction target for any institution or body of the EU as argued by the EP Legal Affairs Committee.
Despite regression coefficients showing enlargements affect some institutions more than they do
the EP, the latter is affected the most by enlargement given that it operates in almost all the official
languages of its member states. Hence, if a new language is spoken in an acceding country, the EP
will need to recruit new translators. This is not the case for the Commission for example, which
works in three official languages. The findings presented here can be explained by the fact that the
overall translation and interpretation services make less than 15 per cent of the EP’s total workforce,
hence, making it unlikely to show a big size effect compared to other EU institutions.
10
What strikes the most from the findings are the large increases of staff members which are captured
in the trend variable and, thus, remain unexplained by the theoretical models presented here. This
could attest to the political nature of the question how many staff members the EU institutions
Volume 13 Issue 2 (2017) Andreja Pegan
1229
should employ. It also warns us that reducing staff in the EU institutions is unlikely to happen based
on a halt to enlargement alone and that proposals targeting measures to reform the plans of staff
numbers should be drafted. If the EU does not reach its 5 per cent staff reduction target and if the
current declining trend is overturned, then citizens might further question the ability of the EU to
control its staffing level, which is regularly highlighted by Eurosceptic groups (e.g. New Direction
2015).
In explaining the difference between the effect of enlargement and the one of treaties, one can look
at their respective relationship with the recruitment policy as set out in the Staff Regulations
(Suvarierol 2007; Ban 2013). While there is a clear connection between recruitment and
enlargement (Article 27 of the Staff Regulations 2013 in respect to national proportionality of staff),
no such relationship is stated for recruitment and treaty reforms. It must be noted that the Staff
Regulations are rather vague on how enlargement should affect recruitment. On the one hand, the
hiring process in the EU has to allow recruitment from the ‘broadest possible geographical basis
from among nationals of Member States’ and permit EU institutions to adopt measures addressing
significant imbalances between nationalities where they exist (Article 27, Staff Regulations 2013). On
the other hand, no post shall be reserved for a particular citizenship (Article 27, Staff Regulations
2013). It is clear, that the accession of a new member state creates an imbalance in the staff
composition, which the EU institutions need to address by organising recruitment competitions per
citizenship.
11
Despite evidence showing that EU staff have grown over the years, the decline in 2011, and the
reasons underlying it (e.g. economic crisis and austerity measures), can be interpreted as a critical
juncture or changing point. This means that a new path has been established, which compared to
the past comprises negative staff growth rates in the EU institutions. It remains to be seen whether
the EU will meet the 5 per cent reduction goal before 2018. From the staff plans laid out in the
budgets and the current reduction rate, it seems the target is unlikely to be achieved. While EU staff
have in its history seen several moments of sudden change, most of these have been towards hiring,
which are easier to implement. It is important to note that the 5 per cent target can be met also with
other means. The data presented here does not include all the institutions and bodies of the EU. For
example, it does not consider EU’s decentralised agencies which employed 5,000 staff members in
2015, three quarters of which were temporary employees rather than permanent civil servants
(Wonka and Rittberger 2011).
In comparison to EU agencies, most staff in the Commission and Council are civil servants and less
than 3 per cent of staff are temporary employees (Wonka and Rittberger 2011; Murdoch and
Trondal 2013; Murdoch et al. 2015).
12
The share of civil servants in the Commission is an obstacle in
meeting the 5 per cent reduction target, as it is difficult to provide legal justification for their layoffs.
The prevalence of civil servant contracts in the EU is justified by its public mandate to serve citizens,
which requires staff whose conditions of employment incentivise them to behave solely in the EU
interest (see preamble to the Staff Regulations 2013). A change towards reducing the share of civil
servants which would enable easier layoffs is unlikely to occur. The decision to introduce an EU civil
service was in fact taken already at the beginning of the European integration in 1956. Moreover,
the idea of a civil service is a fundamental principle of democratic government elsewhere in Europe
as well (Page 1992).
Besides permanent and temporary employees, EU institutions employ also contract staff (Murdoch
and Trondal 2013). These represent 20 per cent of the workforce (their number are not included in
the EU budget). Compared to civil servants and temporary staff they are easier to downsize, since
their contracts are time-fixed and cannot exceed six years (Article 88b, Staff Regulations; CEOS 2013;
Ward et al. 2001). However, their dismissal or employment ban would significantly limit the
Volume 13 Issue 2 (2017) Andreja Pegan
1230
flexibility of the EU to quickly respond to sudden and temporary staff needs in its institutions. The
Staff Regulation reform of 2013 did not only address the downsizing of the staff, but other aspects
as well. For example, staff salaries were adjusted, a special tax on staff salary was increased,
retirement age was raised to 65, working hours were increased to minimally 40 hours per week, and
allowances were reduced. In addition to the 5 per cent staff reduction target, these reforms can also
contribute towards the de-bureaucratisation of EU’s institutions and the image of a ‘down to earth’
administration. While the data presented here does not include all categories of staff and the fact
that bureaucratic growth can be limited in other ways than reducing the number of staff members, it
is striking that the largest increases of staff members, which are captured in the trend variable,
remain unexplained by the theoretical model presented in this article.
Overall, the results from the regression analysis show that the decisions of member states to
increase the functions of EU institutions determine the growth of staff in the EU only partially. Since
bureaucratic growth is determined by events that are not under the sole control of member states,
the EU cannot be conceived as an intergovernmental actor only. Implicitly, this means that
supranational activism of EU institutions or their autonomous activity is a possible explanation for
how EU staff numbers change. Overall, the findings provide support for the governance system of
the EU, where outcomes reflect the dispersed levels of decision making in the EU (Marks et al.
1996). In other words, the bureaucratic growth of the EU is difficult to explain without looking at
member states and the process within the EU institutions. One does not exist without the other.
CONCLUSION
This article has looked at supranational activism of the EU institutions and bodies in terms of the
bureaucratic growth in their administrations. One aspect of bureaucratic growth – the number of
staff members - has been analysed using budget data. First, the article has looked at the staff growth
rate in the EU and whether this has been gradual (Hypothesis 1). Second, the article has asked to
what extent the EU institutions and bodies are dependent on member states’ decision to reform the
EU through either EU/EC treaties or enlargement (Hypothesis 2). Findings show that over the years
the EU institutions have rapidly acquired new staff and that some of these gains can be explained by
enlargements, but not by treaties. Hence, the theoretical model of this article does not capture in
total the reasons for staff changes in the EU. Quite the opposite, the biggest changes remain
unexplained. Thus, there is no sufficient statistical evidence supporting either of the two
hypotheses. This could attest to the political nature of the question how many staff members the EU
institutions should employ. It also warns us that reducing staff numbers could require specific
proposals and a rethink on what size of staff the EU needs.
With regard to the special issue, supranational activism is treated as the autonomy of EU institutions
from member states. It can be concluded that evidence does not show a convincing argument for
either the autonomy of EU institutions to set their own staffing levels or member states exercising
complete control over EU staff numbers. As a result, this article contributes to the view of the EU as
a system of governance, where bureaucratic growth depends on member states’ decisions, as well
as the willingness of supranational institutions to implement them. Finally, the article has addressed
the issue of staffing levels in the institutions of the EU as an issue of controversy among EU’s
citizens. While public polls show that most of the EU citizens have a negative opinion about the
administrative expenditures of the EU – possibly because Eurosceptic parties have politicised and
dominated the public discourse on this issue – the EU has reduced its workforce since 2011.
Following the financial crisis in 2008, the EU pledged to reduce its staff by 5 per cent before 2018.
Such a pledge is unique in the history of the EU and it might indicate the start of a negative staff
Volume 13 Issue 2 (2017) Andreja Pegan
1231
growth trend. Given that staffing levels attract significant negative attention among the EU public,
meeting the staff reduction target would help the EU to portray itself as a down to earth
administration.
CORRESPONDENCE ADDRESS
Andreja Pegan, Trinity College Dublin, Trinity Research in Social Science, Arts Building, Trinity
College, Dublin 2, Ireland [pegana@tcd.ie]
***
ENDNOTES
1
Expenditures for administration were the highest in the late 1950s and early 1960s, since the two biggest funds of the EU
(agriculture and regional) were not yet introduced.
2
An institution is decoupled from the functional mission of the organisation, even though legitimacy can be based on the
supposition that a form is rationally effective (i.e. efficiency is a norm and thus endogenous to the actors) (Meyer and
Rowan 1977; Westphal and Zajac 1994).
3
A candidate country becomes a member of the EU after signing an accession treaty, which is then ratified by all parties. In
the following, enlargement or a new country becoming an EU member state is also referred to as the adoption of an
accession treaty.
4
Available at <http://eur-lex.europa.eu/budget/www/index-en.htm>. [Accessed 10 June 2016]. The first annual budget
was published in 1958.
5
ECA, EDPS, EEAS, and EESC are for the purpose of the Staff Regulations treated as institutions (Article 1b, Staff
Regulations 2013).
6
Competences can be recorded as separate dummy variables to measure the effect of individual treaty amendments.
However, it is difficult to theorise on the effect of individual treaty amendments. For example, the Nice treaty extended
codecision competences to 37 legal bases and introduced changes to prepare the EU for the 2004 enlargement.
Disentangling these effect is difficult. Therefore, I have chosen to only record the occurrence of a treaty change or not.
Competences were distinguished also given the type of treaty they derived from, i.e. founding, reform or budgetary treaty
(Table 2). However, regressions results show that the type of treaty does not affect the number of staff working for the EU.
7
As policy and institutional competences, the increase of the number of member states does also derive from treaties (i.e.
accession treaties).
8
For the Council, these periods are 1976-1979, 1989-1994, 2000-2003, 2007-2010, and 2012-2016 (Figure 3c). For the EP:
1982-1985, 1988-1991, 1997-2002, 2007-2009, and 2012-2016 (Figure 3d). For the Commission: 1999-2002 and 2012-2016.
9
Among all the institutions and for the whole period considered, the biggest relative cut was performed on the European
Economic and Social Council in 2002, when staff was reduced by 32 per cent or 239 employees.
10
Despite this, the EP is the biggest employer for translators and interpreters in the world.
11
Proportional national representation is with some exceptions respected for the permanent staff of the EU (Belgian
nationals tend to be overrepresented, while British are underrepresented). The national representation of temporary staff
on the other hand is not proportional to the population (see Murdoch et al. 2015: 5).
12
In the EP one fifth of employees are temporary. These are mainly employed by parliamentary party groups, who cannot
provide permanent contracts as dependent on European election outcomes.
***
Volume 13 Issue 2 (2017) Andreja Pegan
1232
REFERENCES
Ban, C. (2013). Management and Culture in an Enlarged European Commission. Basingstoke: Palgrave Macmillan.
Brand, C. (2012). Parliament should be spared staff cuts, European Voice. 1 March.
Bulmer, S. (2009). Politics in Time meets the politics of time: historical institutionalism and the EU timescape. Journal of
European Public Policy, 16(2), pp. 307-324.
Christensen, J. G. (1997). Interpreting Administrative Change: Bureaucratic Self-Interest and Institutional Inheritance in
Government. Governance, 10(2), pp. 143-174.
Christiansen, T., Högenauer, A. H. and Neuhold, C. (2014). National Parliaments in the post-Lisbon European Union:
Bureaucratization rather than Democratization?. Comparative European Politics, 12(2), pp. 121–140.
Council of the European Union (2012). Contributions requesting calculations on savings in the context of the Staff
Regulations Review. 13820/12. Brussels, 17 September.
DiMaggio, P. J. and Powell, W. W. (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in
Organizational Fields. American Sociological Review, 48(2), pp. 147–160.
Ellinas, A. A. and Suleiman, E. (2014). The European Commission and Bureaucratic Autonomy. Europe's Custodians. New
York: Cambridge University Press.
European Commission (2015). Europeans and the European Union budget. Standard Eurobarometer 83. Luxembourg.
European Parliament (2012). Report on the proposal for a regulation of the European Parliament and of the Council
amending the Staff Regulations of Officials and the Conditions of Employment of Other Servants of the European Union. A7-
0156/2012. Brussels, 3 May.
Haas, E. (1958). The Uniting of Europe: Political, Social, and Economic Forces, 1950-1957. Palo Alto: Stanford University
Press.
Häge, F. M. (2013). Bureaucrats as Law-Makers: Committee Decision-Making in the EU Council of Ministers. London:
Routledge.
Hibou, B. (2015). The Bureaucratization of the World in the Neoliberal Era: An International and Comparative Perspective.
New York: Palgrave Macmillan.
Högenauer, A.-H. and Neuhold, C. (2015). National Parliaments after Lisbon: Administrations on the Rise. West European
Politics, 38(2), pp. 335-345.
Jenson, J. and Mérand, F. (2010). Sociology, institutionalism and the European Union. Comparative European Politics, 8(1),
pp. 74-92.
Mahoney, J. and Thelen, K. (2010). A Theory of Gradual Institutional Change. In Mahoney, J. and Thelen, K. (eds), Explaining
institutional change. New York: Cambridge University Press, pp. 1-37.
Marks, G., Hooghe, L. and Blank, K. (1996). European Integration from the 1980s: State-Centric v. Multi-level Governance.
Journal of Common Market Studies, 34(3), pp. 341-378.
Marshall, M. W. (1985). Limits to Bureaucratic Growth. Berlin and New York: Walter de Gruyter.
Meltzer, A. H. and Richard, S. F. (1983). Tests of a rational theory of the size of government. Public Choice, 41, pp. 403-418.
Meyer, J. W. and Rowan, B. (1977). Institutionalized Organizations: Formal Structures as Myth and Ceremony. American
Journal of Sociology, 83(2), pp. 340–363.
Meyer, W. M. (1987). The Growth of Public and Private Bureaucracies. Theory and Society, 16(2), pp. 215-235.
Volume 13 Issue 2 (2017) Andreja Pegan
1233
Miller, G. J. and Moe, T. M. (1983). Bureaucrats, Legislators, and the Size of Government. The American Political Science
Review, 83(2), pp. 297-322.
Murdoch, Z. and Trondal, J. (2013). Contracted Government: Unveiling the European Commission’s Contracted Staff. West
European Politics 36, pp. 1–21.
Murdoch Z., Trondal J., and Geys, B. (2015). Representative bureaucracy and seconded national government officials in the
European Commission. Regulation and Governance, 10(4), pp. 335-349.
New Direction (2015). Ending Excess 2. The Extravagance Continues. Brussels: New Direction.
Niskanen, W. A. (1971). Bureaucracy and Representative Government. New Brunswick and London: Aldine Transaction.
North, C. D. (1991). Institutions. The Journal of Economic Perspectives, 5(1), pp. 97-112.
Olsen, J. P. and March, J. G. (1989). Rediscovering Institutions. New York: Free Press.
Page, E. C. (1992). Political Authority and Bureaucratic Power. A Comparative Analysis. Hertfordshire: Harvester
Wheatsheaf.
Peters, B. G. (2001). The Politics of Bureaucracy. London: Routledge.
Pierson, P. (2004). Politics in Time. Princeton and Oxford: Princeton University Press.
Pollack, A. M. (2007) Rational Choice and EU Politics. In K. E. Jørgensen, M. A. Pollack and B. Rosamond (eds), Handbook of
European Politics. Thousand Oaks: Sage, pp. 31-55.
Radaelli, C. M. (1999). Technocracy in the European Union. London: Longman.
Regulation 1023/2013 (EU, Euratom) (2013) of the European Parliament and of the Council of 22 October 2013 amending
the Staff Regulations of Officials of the European Union and the Conditions of Employment of Other Servants of the
European Union. Official Journal of the EU, L 287/15. Brussels.
Šabič, Z. (2008). Building Democratic and Responsible Global Governance: The Role of International Parliamentary
Institutions. Parliamentary Affairs, 61(2), pp. 255-271.
Schmitter, P. C. (2005). Ernst B. Haas and the legacy of neofunctionalism. Journal of European Public Policy, 12(2), pp. 255-
272.
Shepsle, K. A. (2008). Rational Choice Institutionalism. In S. Binder, R. Rhodes and B. Rockman (eds), Oxford Handbook of
Political Institutions. New York: Oxford University Press, pp. 23-38.
Staff Regulations (2013). Regulation No 31 (EEC), 11 (EAEC) laying down the Staff Regulations of Officials (Staff Regulations)
and the Conditions of Employment of Other Servants (CEOS) of the European Economic Community and the European
Atomic Energy Community, last amendment 22 October 2013.
Suvarierol, S. (2007). Beyond the Myth of Nationality: A Study on the Networks of European Commission Officials. Delft:
Eburon.
Vaubel, R., A. Dreher and Soylu, U. (2007). Staff growth in international organizations: A principal-agent problem? An
empirical analysis. Public Choice, 133(3/4), pp. 275-295.
Ward, K., Grimshaw, D., Rubery J. and Beynon, H. (2001). Dilemmas in the management of temporary work agency staff.
Human Resource Management Journal, 11(4), pp. 3-21.
Weber, M. (1968). Economy and Society, G. Roth and C. Wittich (eds). New York: Bedminister Press, Vol. 1, pp. 956-1005.
Westphal, J. D. and Zajac, E. J. (1994). Substance and symbolism in CEOs’ long term incentives plans. Administrative Science
Quarterly, 39, pp. 367–390.
Volume 13 Issue 2 (2017) Andreja Pegan
1234
Wilensky, H. L. (1975). The Welfare State and Equality: Structural and Ideological Roots of Public Expenditure. Berkeley:
University of California Press.
Wilson, J. Q. (1992). On Predicting the Bureaucratization of American Government. In L. B. Hill (ed), The State of Public
Bureaucracy. Armonk and London: M. E. Sharp, pp. 207-215.
Wonka, A. and Rittberger, B. (2011). Perspectives on EU governance: an empirical assessment of the political attitudes of
EU agency professionals. Journal of European Public Policy, 18(6), pp. 888-908.
Wooldridge, J. M. (2009). Introductory Econometrics. A Modern Approach. Boston: Cengage.
