Public Administration

Full text

Public Sector Performance

Public Sector Performance
Social and Cultural Planning Offi ce
The Hague, September 2004
An international comparison of
education, health care, law and order
and public administration

Social and Cultural Planning Offi ce of the Netherlands
The Social and Cultural Planning Offi ce of the Netherlands (SCP) was established by Royal Decree
of March 30, 1973 with the following terms of reference:
a. to carry out research designed to produce a coherent picture of the state of social and cultural
welfare in the Netherlands and likely developments in this area;
b. to contribute to the appropriate selection of policy objectives and to provide an assessment of
the advantages and disadvantages of the various means of achieving those ends;
c. to seek information on the way in which interdepartmental policy on social and cultural welfare
is implemented with a view to assessing its implementation.
The work of the Social and Cultural Planning Offi ce of the Netherlands focuses especially on problems
coming under the responsibility of more than one Ministry.
As Coordinating Minister for social and cultural welfare, the Minister for Health, Welfare and Sport
is responsible for the policies pursued by the Social and Cultural Planning Offi ce. With regard to the
main lines of such policies the Minister consults the Ministers of General Affairs, Justice, Home Affairs,
Education, Culture and Science, Finance, Housing, Physical Planning and Environmental Protection,
Economic Affairs, Agriculture, Nature Management and Fisheries and Social Affairs and Employment.
© Social and Cultural Planning Offi ce, The Hague 2004
SCP-publication 2004/8
Cover design: Bureau Stijlzorg, Utrecht
DTP: Mantext, Moerkapelle
Distribution in the USA & Canada:
Transaction Publishers, New Brunswick (USA)
ISBN 90-377-0184-1
NUR 740
Social and Cultural Planning Offi ce
Parnassusplein 5
2511 VX The Hague
The Netherlands
Tel. (+31) (070) 340 70 00
Fax (+31) (070) 340 70 44
Website: www.scp.nl
E-mail: info@scp.nl

v
Contents
Preface 1
Summary 3
1 Introduction 27
1.1 The Lisbon Agenda and the public sector 27
1.2 Aim and scope of the report 32
1.3 Outline of the report 35
Annex 1.1 Measuring public sector performance 36
Annex 1.2 Participants in the project 40
Annex 1.3 Members of advisory committee 42
2 Demography and the economy 43
2.1 Introduction 43
2.2 Key data 43
2.3 Public sector 50
2.4 Macro-economic performance 57
3 Education 74
3.1 Introduction 74
3.2 Education systems 79
3.3 Use of resources 87
3.4 Enrolment and graduation 91
3.5 Cost price and productivity 100
3.6 Quality and effectiveness 103
3.7 Further analysis of effectiveness 112
4 Health care 120
4.1 Introduction 120
4.2 Health care systems 126
4.3 Use of resources 135
4.4 Patient care 150
4.5 Cost price and productivity 157
4.6 Quality of care 163
4.7 Effectiveness 167

vi Contents
5 Law and order 185
5.1 Introduction 185
5.2 Organisation of the criminal justice system 189
5.3 Use of resources 196
5.4 From crime to punishment 200
5.5 Productivity, quality and effectiveness: offi cial records 213
5.6 Quality and effectiveness: the public’s view 223
6 Public administration 234
6.1 Introduction 234
6.2 Administrative systems 235
6.3 Use of resources 243
6.4 Administrative processes 247
6.5 Performance 257
7 Performance of the public sector 271
7.1 Introduction 271
7.2 Grouping welfare state by type 271
7.3 Overall performance for education, health care and law and order 276
7.4 Grouping countries on the basis of performance 279
7.5 Overall performance of the public service sector 285
7.6 Some concluding remarks 291
References 293
Annexes
A Data sources and data corrections 303
B Technical notes 305
C Explanatory notes (on the internet: www.scp.nl/9037701841) 309
D Source of data on staff and expenditure for law and order
(www.scp.nl/9037701841) 309
E Data corresponding to fi gures (www.scp.nl/9037701841) 309
Publications of the SCP in English 310

234
6 Public administration
Steven Van de Walle, Miekatrien Sterck, Wouter van Dooren, Geert Bouckaert en Evert Pommer1
6.1 Introduction
After the rapid expansion of the welfare state in the 1950s and 1960s, the public
sector has been under considerable pressure in the past few decades. Declining
public confi dence in government institutions and growing demands on public
fi nances have prompted governments to initiate measures to trim the public sector
and make it more effi cient and effective. Reform strategies adopted can be catalogued
as: Maintain, Modernise, Marketise and Minimise (Pollit and Bouckaert 2004).
‘Maintain’ involves tightening up traditional control mechanisms. The exist-
ing system is stretched, for example, by placing linear restrictions on expenditure
(the ‘cheese slice method’), with no downward revision of policy targets. Another
example of the Maintain strategy is more detailed control of expenditure programs.
Although this strategy causes less disruption in the functioning of government
organisations, it is probably not adequate to tackle existing fi nancial and legitimacy
problems of the public sector.
‘Modernise’ involves organising alternative structures and processes of govern-
ment policy making. However, any modernisation operation must be consistent
with traditional values of public service provision. The public sector is intrinsically
different from the private sector, and any fundamental reform has to take account
of these differences. The focus of reforms is to improve management (managerial
modernisation) and/or to foster participation by citizens and user groups (participa-
tory modernisation).
‘Marketise’, the third strategy, involves introducing a private-sector focus to the
public sector and its values. It does not mean that services are privatised. The aim is
still primarily to reform the public sector, not to reduce its scope and public outlays.
Techniques common to the private sector are transplanted wholesale to the public
sector. In doing so, the unique character of public sector services is implicitly called
into question. One example of this strategy has been the introduction of internal
competition (competitive tendering) in New Zealand and the United Kingdom.
1 This chapter is mainly based on Van de Walle et al., 2004. The fi rst four authors are
employed at the Public Management Institute (‘Instituut voor de Overheid’) of Leuven
University.

235Public administration
‘Minimise’ – reducing the public sector – involves privatising functions that have tra-
ditionally been in the domain of the public sector. The railways in the United King-
dom are perhaps the most notorious example of privatisation, resulting in a poorer
safety record and higher rail fares. Privatisation has been much more successful in
other sectors (such as telecommunications).
Many monitoring systems have been put into place as part of these reforms, in
order to chart the performance of the public sector. This is an essential prerequi-
site for the success of new management techniques, such as performance budgets,
performance contracts and strategic plans. Such national systems produce a wealth
of information, but they do not extend beyond the confi nes of a particular admin-
istrative system. National boundaries are rarely crossed. It is in fact diffi cult to
consolidate the data, as different countries use different defi nitions for the variables
concerned. The data available on public adminsitration mainly concerns subjective
perceptions of performance, rather than actual measurements of performance. This
lack of objective data makes it impossible to draw generalised conclusions as to what
is the optimal administrative system, even if we wanted to. This chapter therefore
aims above all to point out the possibilities and limitations associated with the com-
parison of administrative systems at European level, drawing from empirical data on
public administration performance.
The public administration includes all those activities directed at policymaking,
legislation and management of the public sector. Activities producing individual
services for citizens, like health care and education, are not the domain of public
administration. However, in practice the demarcation between policy, legislation
and management on the one hand and concrete services provided to individual
citizens on the other hand, is not always easy to draw. Moreover, data available on
public sector performance usually do not allow public administration activities to be
identifi ed separately. Consequently, public administration and activities performed
by the public administration must often be measured by related concepts, like ‘gov-
ernment’ and ‘general public services’.
The chapter is structured as follows. An overall comparison of administrative
systems (Section 6.2), covering administrative culture, the degree of decentraliza-
tion and the trend towards the autonomisation of government organisations is
followed by a survey of resources claimed by the public administration (Section 6.3).
Section 6.4 looks at administrative processes in the public sector, including fi nancial
management, human resources management and e-government. The fi nal section
examines the quality of the government and confi dence in the Civil Service.
6.2 Administrative systems
6.2.1 Administrative culture
Differences in administrative culture have a major impact both on fundamental
choices concerning the structure of the public sector, and on the daily functioning of
the government apparatus. Administrative culture forms part of a wider political and

236 Public administration
social culture.2 Hofstede’s dimensions are probably the best-known categorisation
of administrative cultures (Hofstede 1980), although other attempts have been made
(Mamadouh 1999). It is clearly no simple matter to group countries on the basis of
their administrative culture.
Loughlin (1994) groups countries on the basis of broad philosophical and cultural
traditions. He distinguishes an Anglo-Saxon (minimal state), a Germanic-organicist
and a French Napoleonic state tradition. The Scandinavian type is a mix of the fi rst
two. For Pollitt and Bouckaert (2004), administrative culture is that which is consid-
ered normal and acceptable in an organisation, and they have mapped the extent of a
Rechtsstaat vs. public interest tradition in a country. In the Rechtsstaat (rule of law) model,
the state is the central integrating force in society, and administrative law takes a
prominent place in this tradition. In the public interest model the state assumes a less
prominent role in society and is regarded as something of a necessary evil. Account-
ability is more important here than legality. Of course, Rechtsstaat and public interest
are not extremes on a continuum, and in some countries, such as the Netherlands,
Finland and Sweden, a trend towards other models can be observed. Hooghe (2002)
used four dimensions developed by Page (1995) – cohesion, autonomy from political
control, caste-like character and non-permeability of external interest – to construct
an index of ‘Weberian bureaucratic tradition’ (strong, medium, weak), indicating to
what degree a national administrative culture corresponds to the Weberian model
(strong cohesion, large degree of autonomy from political control, strong caste-
like character of the bureaucracy and low permeability of external interests). Her
research focused on the European Commission, where differences in administrative
cultures are of course more pronounced.
Hajnal (2003) analysed public administration education programmes in Europe.
European countries are clustered in three groups. A ‘legal’ group of countries, where
there is a strong focus on law in public administration education, a ‘public’ group
where the unique public and political character of the public administration is rec-
ognised, and a ‘corporate’ group, where (business) management techniques take a
central place in the curriculum.
2 See Peters (1989) for a detailed discussion of the phenomenon of ‘administrative
culture’.

237Public administration
Table 6.1 Administrative cultures in EU-15 countries
administrative culture
(Pollitt and Bouckaert
2004)
state traditions
(Loughlin 1994)
type of national admin-
istration : Weberian
bureaucratic tradition
(Hooghe 2002)
clusters in PA
education
(Hajnal
2003)
AT – Germanic-organicist weak weberian –
BE recht sstaat French-Napoleonic (until
1988), Germanic-organi-
cist (after 1988)
weak weberian public
DE recht sstaat Germanic-organicist medium weberian –
DK – Scandinavian (mixture
of Anglo-Saxon and Ger-
manic)
medium weberian corporate
ES – French-Napoleonic (until
1978), Germanic-organi-
cist (after 1978)
medium weberian public
FI tending to rech tsstaat weak weberian –
FR predominantly
rechtsstaat
French-Napoleonic strong weberian public
GR – French-Napoleonic weak weberian legal
IE – Anglo-Saxon (minimal
state)
strong weberian corporate
IT rechtsst aat French-Napoleonic weak weberian legal
LU – weak weberian –
NL originally very legalistic,
but has changed to
pluralistic/consensual
Germanic-organicist medium weberian corporate
PT – French-Napoleonic medium weberian legal
SE originally legalistic but
has changed to cor-
poratist
Scandinavian (mixture
of Anglo-Saxon and Ger-
manic)
medium weberian public
UK public interest Anglo-Saxon (minimal
state)
strong weberian –
- : not available
The four categorisations discussed here each have their own focus, so it is diffi cult to
obtain a clear picture. The categorisation of some countries seems fairly coherent, but the
absence of clear indicators means it still entails some risk. The Anglo-Saxon tradition dif-
fers considerably from the continental tradition. This is refl ected, among other things, in
the fact that many public servants in the United Kingdom are generalists, while in Ger-
many they tend to have a legal background. The large number of studies of cultural differ-
ences among European public servants shows how important it is to take a more in-depth
look at this subject. Obviously, the evolution towards a European Administrative Space will
be affected by different views on the role of the public administration in society.

238 Public administration
6.2.2 Degree of local decentralization
Another important characteristic of administrative systems is the degree of decen-
tralization. Decentralization has many faces. One is functional decentralization,
whereby resources and powers are devolved to semi-autonomous institutions. Then
there is territorial decentralization, which increases the role of other tiers of govern-
ment, such as regions and local authorities. We focus here on decentralization to the
local level. Autonomised institutions are examined in the next section. There appear
to be major differences between countries in the degree of decentralization.
Decentralization is usually approached from a fi nancial perspective, with a focus
on devolving public resources. Figure 6.1 shows the proportion of the government
budget spent by local authorities. Three groups of countries may be distinguished.
Firstly, the Scandinavian countries, with a very strong local sector (accounting
for over 30% of spending). The outsider here is Luxembourg. The second cluster
comprises a number of Central European countries (20%-30%). One notable country
included in this group is France, which generally has a centralist image. Finally, we
have a number of countries with a small local sector, mainly Southern European
countries, and also Ireland and Belgium.
BE
DK
DE
GR
ES FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
0
5
10
15
20
25
30
35
40
Figure 6.1 Spending by local authorities as % of total government expenditure
Source: Council of Europe (1997)
These fi gures give an fi rst impression. However, the relative share in total public
spending is too coarse a measure of decentralization. Can we call it decentralization
when a local authority merely acts as the agent of a higher level of government? What
autonomy do local authorities have? Figures on spending must always be viewed in
the light of autonomy. Indicators of autonomy include spending freedom and free-
dom to collect resources. The latter concerns the freedom to increase or cut back
certain fl ows of income, and the former an authority’s freedom to spend its income
as it sees fi t. We can examine these concepts on the basis of local authorities’ income

239Public administration
structure. Spending freedom and freedom of resources can be established by analys-
ing a breakdown of local authority income.
– Taxes: high freedom of resources, high spending freedom
– Loans: high freedom of resources, high spending freedom
– Shared taxation: low freedom of resources, high spending freedom
– General grants: low freedom of resources, high spending freedom
– Fees & charges: high freedom of resources, low spending freedom3
– Targeted grants: low freedom of resources, low spending freedom
Figure 6.2 combines the level of local government expenditure with spending free-
dom, as a measure of autonomy, showing the product of the percentage of the gov-
ernment budget spent by local authorities (see Figure 6.1) and the percentage over
which the local authority has a high degree of spending freedom.
BE
DK
DE
GR
ES FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
Figure 6.2 Autonomy of local authorities (% of total government budget)
Source: Council of Europe (1997)
0
5
10
15
20
25
30
The groups identifi ed above remain largely intact. A number of atypical positions
are confi rmed. France, for instance, despite its centralist image, tends towards the
Scandinavian model. Ireland has only a limited degree of decentralization, and thus
differs strongly from the United Kingdom, with which it is often associated. The
Netherlands fi nances local authorities to a large extent by means of targeted grants
and is therefore less decentralised according to this measure. It ends up closer to
Belgium, as was to be expected. Local administration in the Netherlands and
3 This income fl ow can be increased by raising either performance or charges. Spending
freedom is limited, since charges do not always cover costs, so no new policy freedom
is created.

240 Public administration
Belgium has developed since the early nineteenth century on the foundations laid
during the Dutch Republic.
Another indicator of the degree of decentralization is the distribution of public
servants among the different tiers of government. Table 6.2 shows the proportion
of public servants employed at the various levels of administration. It clearly shows
a shift in staff employed by central government to staff on the payroll of local and
regional authorities. The proportion of public servants working in central govern-
ment is declining, while the proportion of staff working at the local and regional
level is on the increase. The Scandinavian countries (Finland (23.4%) and Sweden
(17.3%)) and most of the federal countries (Australia (12.1%), Germany (11.5%),
Canada (17.1%) and the United States (13.5%) ) have small central governments.
Belgium is an exception, on 34.3%. In unitarian states such as France (51.6%), the
Netherlands (74.2%) and Italy (57.9%), the proportion of public servants working in
central government is signifi cantly higher. In 2000, as many as 90.9% of public serv-
ants worked in central government in New Zealand.
Table 6.2 Percentages of public servants working in each tier of government
central government other level of government
1990 1994 2000 level 1994 2000
Belgium 53.7 b39.9 c34.3 regional 14.1c14,8
local 46.0 c50,8
Germany 21.6 11.9 11.5 länder 51.0 52.2
municipalities 37.1 36.3
France 55.0 48.7 51.6dsub national 30.7 25.3d
Italy 63.0 63.0 57.9 dregional 23.0 26.8 d
municipalities 14.0 15.3d
Netherlands 70.1 73.2 74.2 aregional 5.2 4.7a
municipalities 21.6 21.1a
Finland 24.3 25.2 23.4amunicipalities 74.8 76.6a
Sweden 26.7 17.3 n.a. regional 24.6 n.a.
municipalities 58.1 n.a.
UK 47.7 47.7 47.6 dlocal 52.3 52.4 d
Australia 15.0 14.6 12.1
estate 73.3 77.1e
local 12.1 10.8e
New Zealand 90.1 89.7 90.9 local 10.3 9.1
Canada 17.9 17.1 13.2 provincial 44.1 51.9
local 38.9 35.0
USA 16.7 15.2 13.5 state 22.6 23.1
local 61.1 63.4
Notes: a: fi gures are for 1999; b: fi gures are for 1989; c: fi gures are for 1995d: fi gures are for 1997;
e: fi gures are for 1998
Source: Pollit t and Bouckaert (2004)

241Public administration
6.2.3 Autonomisation of government organizations
This section takes a close look at the trend towards granting government organisa-
tions autonomous status, otherwise known as autonomisation. Unlike territorial
decentralization to the local level, this is a relatively recent strategy. However, it has
formed the core of many public sector reforms in Europe. Autonomisation is based
on the hypothesis that greater autonomy for ‘front offi ce’ agencies leads them to
work more effi ciently and effectively. The Scandinavian countries have a fairly long
tradition of autonomisation. However, in most countries the trend did not set in
until the advent of New Public Management in the 1980s (Hood 1991).
Autonomisation can take various forms. There are differences in terms both of
legal status, and of the degree of policy and management autonomy. Here, however,
we distinguish between three major types of autonomous organisation: arm’s length
agencies, public law agencies and mixed agencies. Arm’s length agencies have been
hived off from their parent organisation, but remain answerable to the minister.
They might still form part of the parent organisation, or they might have their own
separate legal status. Examples include the United Kingdom’s executive agencies,
and the Netherlands’ baten-lastenagentschappen. Public law agencies are bodies set up
under public law which are institutionally separate from the parent organisation and
have their own board. Examples include the Netherlands’ zelfstandige bestuursorganen
(zbos), the United Kingdom’s non-departmental public bodies and New Zealand’s
crown entities. The third group of agencies are private-law bodies that have a public
function. Examples include public enterprises and non-profi t organisations. These
are therefore sometimes referred to as mixed agencies.
Table 6.3 reviews the three forms of autonomisation in fourteen countries,
indicating which forms exist in each country, whether they form part of the central
budget, and the size of their budgets in relation to the total government budget. This
information is helpful to indicate the degree of autonomy and the relative impor-
tance of the different forms of autonomous organisations. The data have been drawn
from a survey by the oecd and the World Bank (oecd and World Bank 2003).

242 Public administration
Table 6.3 Forms of autonomisation
arm’s length agencies4public law agencies5mixed agencies6
part of central
budget?
% of overall
government
budget
part of central
budget?
% of overall
government
budget
part of central
budget?
% of overall
government
budget
BE yes, entirely <10% yes, partially 10-20% – –
DK – – yes, entirely 10-20% yes, partially –
DE yes, entirely 10-20% yes, partially 10-20% no <10%
FR yes, entirely – no – no –
GR yes, entirely – yes, partially – no –
IE yes, entirely – yes, par tially <10% no <10%
ES yes, entirely – yes, partially – yes, partially –
PT yes, entirely 40-60% yes, entirely 40-60% no <10%
UK yes, entirely – yes, entirely – yes, entirely –
NL yes, entirely <10% yes, partially 10-20% yes, par tially <10%
HU yes, entirely – yes, partially – yes, partially –
CZ yes, par tially – no – no –
AU yes, entirely 10-20% yes, partially <10% no <10%
NZ yes, entirely <10% yes, entirely 40-60% no <10%
-: not available
Source: OECD and World Bank (2003)
4 “Arm’s length agencies have no separate entity from the state; the fi nancial framework
is predefi ned; most are funded through allocations from the state budget, and their
budget is annually reviewed through the state budget process; no own accounts; e.g.
semi-autonomous bodies in New Zealand, executive agencies in the uk (oecd and
World Bank 2003).”
5 “Public Law Agencies: 100 per cent public ownership; partially or completely institu-
tionally separate from ministries; can be partially separate or fully separate legal bodies;
function mostly under public law; not commercially oriented; most PLAs are tax-rev-
enue fi nanced, and their budget is part of the general budget law; own accounts; e.g.
public law ZBOs in the Netherlands, crown entities in New Zealand, most non-depart-
mental public bodies in the uk, Swedish agencies and boards (oecd and World Bank
2003).”
6 “Mixed agencies function mostly under private law, usually with a full separate legal
identity from the state; commercially oriented; usually mostly sales revenue fi nanced
and can carry forward surpluses, borrow and lend; their budgets are separate from
those of ministries; own accounts; e.g. private law ZBOs in the Netherlands (oecd and
World Bank 2003).”

243Public administration
In most countries, the entire budget of arm’s length agencies is part of the budget
of the central or federal government. This is due to the fact that these agencies are
still answerable to central government, even though they have been placed at arm’s
length. Their share of the total government budget ranges from less than 10% to
20%. One exception is Portugal, where the budget of arm’s length agencies accounts
for 40% to 60% of the total budget. The budgets of public law agencies are usually
only partially included in the central government budget. Their proportion of the
total government budget ranges from less than 10% to 20%. In Portugal and New
Zealand, public law agencies account from 40% to 60% of the budget.
Finally, mixed agencies do not form part of the central budget in the majority of
the countries studied. In most cases, their share of the total government budget is
below 10%.
6.3 Use of resources
Public administration includes policy making, legislating policy and management
of the public sector. The un classifi cation of functions of government distinguishes
general public services, including legislative, executive, fi nancial, fi scal and foreign
activities, apart from other typical public functions like defence, public order and
safety.7 This classifi cation fi ts very well the usual defi nition of public administration
activities. Unfortunately, this information is only available for eu-15 countries. The
new member states will follow this classifi cation from 2004.
The share of expenditure on general public services in gdp varies between 1,3%
(United Kingdom) and 4,8% (France). Other countries are scattered evenly between
these extreme values (Figure 6.3). Different levels of expenditure are partly explained
by differences in over-all public spending levels (Figure 2.6). Expressed as a share of
gdp, public expenditure is high in Sweden and Denmark and low in Ireland and the
uk. The same pattern holds for expenditure on general public services.
7 The various functions are: (1) general public services, (2) defence, (3) public order and
safety, (4) economic affairs, (5) environment protection, (6) housing and community
amenities, (7) health, (8) education, (9) recreation, culture and religion, (10) social
protection.

244 Public administration
BE
DK
DE
GR
ES FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
Figure 6.3 Expenditure on general public services of EU-15 countries, 2001 (percentage
of GDP)a
a Excluding debt interest payments (classified as property income consolidated) and foreign transfers (classified as
others transfers consolidated).
Source: Eurostat
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
5,0
The share of general public services in total government expenditure varies widely
between eu-15 countries (Figure 6.4). Luxembourg and France spend well over 9% of
total government expenditure on general public services, while the United Kingdom
spends only 3%. A closer view shows a striking similarity between Figures 6.3 and
6.4. Obviously, countries spending much on general public services (Figure 6.3) also
devote a larger part of total government expenditure to produce general public serv-
ices (Figure 6.4).

245Public administration
BE
DK
DE
GR
ES FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
Figure 6.4 Government expenditure on general public services of EU-15 countries, 2001
(percentage of total government expenditure)a
a Excluding debt interest payments (classified as property income consolidated) and foreign transfers (classified as
others transfers consolidated).
Source: Eurostat
0
1
2
3
4
5
6
7
8
9
10
In Figure 2.12 (Chapter 2) the share of employees in the public administration,
expressed as a percentage of total employment, appeared to range from less then 1%
in Cyprus to more then 8 percent in Belgium. Because public administration pro-
duces public services, it’s also relevant to relate employment in public administra-
tion to total population (Figure 6.5).
Figure 6.5 Employees in public administration per 1.000 population, 2000
Source: OECD, ILO, NATO, European Sourcebook, US Sourcebook, AIV (SCP revision)
0
5
10
15
20
25
30
35
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CY
CZ
EE
LV
LT
HU
MT
PL
SI
SK
AU
CA
NZ
US

246 Public administration
The number of employees per 1.000 inhabitants ranges from 4 in Cyprus to 33 in the
United Status. In eu-15 countries the ratio of public administration staff is low in
Greece, Italy and Ireland and high in Belgium, Denmark and France. The staff-ratio
ends up between 15 and 25 in the other eu-15 countries. Most new member states
end up between 10 and 20 employees per 1.000 inhabitants, except for Cyprus and
Poland (about 5) and the Czech republic (just over 20).
The ranking of eu-15 countries according to Figure 6.3 (expenditure on Civil
service related to gdp) and to Figure 6.5 (civil servants related to inhabitants) differs,
although the relation is positive. On top of the civil expenditure ranking are France,
Sweden and Austria and on top of the civil staff ranking are Belgium, Denmark and
France. Inversely, at the bottom of the civil expenditure ranking are the United King-
dom, Ireland and Spain and at the bottom of the civil staff ranking are Greece, Italy
and Ireland. These differences can be attributed partly to differences in share of cost
of staff in total production of general public services (Figure 6.6). For example, Sweden
scores high on civil expenditure but low on share spent on staff; Belgium scores mod-
erate on civil expenditure but high on share spent on staff and the United Kingdom
scores low on civil expenditure but rather high on share spent on staff.
Figure 6.6 Composition of government expenditure on general public services of EU-15
countries, 2001 (percentage)a
a Excluding debt interest payments (classified as property income consolidated) and foreign transfers (classified as
others transfers consolidated).
Source: Eurostat; SCP revision
BE DK DE GR ES FR IE IT LU NL AT PT FI SE UK
0
10
20
30
40
50
60
70
80
90
100
staff consumption and taxes investment other
Unfortunately, no comparable information is available on wages in public admin-
istration, but there is some information on public sector wages in general. These
public sector wages include earnings in health care, education, defence and law
and order. The level of public sector spending in general depends heavily on costs
of staff. In most countries, mean public sector wages surpass mean wages of all

247Public administration
employees in the economy (Figure 6.7). The ratio between public sector wages and
all wages is almost unity in France, Finland, Sweden and Czech Republic. In all other
countries, especially in Greece, this ratio is in favour of public sector employees. On
average, public sector employees earned in 2002 about 37% more then employees
in general. In part, this salary gap can be explained by the fact that government
employees are on average better educated than market sector workers, given job
requirements, for example in education and health care.
Figure 6.7 Ratio of public sector wages and all wages, 2002 (or latest available year)
Source: Eurostat; SCP revision
0,0
0,5
1,0
1,5
2,0
2,5
3,0
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CY
CZ
EE
LV
LT
HU
MT
PL
SI
SK
AU
CA
NZ
US
6.4 Administrative processes
The previous section considered various input variables for public administration. The
government absorbs productive resources to perform its tasks and achieve certain
effects in society. This chapter reviews administrative processes associated with the
machinery that allows government to pursue its policies. Policy implementation is sup-
ported by administrative processes such as fi nancial management, human resources
management and information technology. This section looks at trends in fi nancial
management (Section 6.4.1), human resources management (Section 6.4.2), and e-gov-
ernment (Section 6.4.3). The scope of this section extends to the public sector in gen-
eral, but the government – and the public administration in particular - is responsible
for the quality of fi nancial management, human resources and openness in the public
sector. This creates a horizontal persepctive in public sector policy.
6.4.1 Financial management
Various factors have prompted public authorities to modernise their budget cycle.
The fi nancial reform agenda consists of three major components: greater fi nancial
responsibility for management, results-based budgets and multi-year budgets.

248 Public administration
Financial responsibility for management
Approval of the budget constitutes part of the legislative power’s control function
over the executive power. The New Public Management movement has shifted the
focus from traditional a priori control to control in retrospect, and placed more
emphasis on results and greater fi nancial responsibility for management. This sec-
tion indicates the extent to which governments in Europe are following this trend.
One indicator of the degree of parliamentary control over the budget and of man-
agement freedom is the degree of detail to which the budget is appropriated. Five
levels of appropriation can be distinguished, corresponding to a declining degree of
discretionary power and freedom for agency management:
– aggregated allocation at the level of broad programmes or outcomes;
– aggregated allocation at the level of government departments;
– aggregated allocation for staff and programmes at the programme level, appro-
priation per programme;
– appropriation for staff expenditure and programme expenditure; and
– disaggregated allocation within programmes.
The more aggregated the appropriation, the more freedom management has
to change the budget allocation, giving it more fl exibility to make adjustments in
order to achieve the desired results. However, a higher level of appropriation usually
goes hand in hand with a greater demand for policy information. This information
should allow parliament to guide programmes towards the results envisaged.
Another indicator of the degree of parliamentary control and the degree of man-
agement freedom is end of year fl exibility. This allows managers to carry any surplus
budget allocation over to next year’s budget. Although the principle of end of year
fl exibility is contrary to the traditional one-year budget principle, it does offer incen-
tives to use resources more effi ciently and can counteract wastage of resources at the
end of the year (‘December fever’). Budgets no longer have to be used up completely.
End of year fl exibility might be subject to certain conditions, such as a maximum
transferable amount, approval by the fi nance minister or parliamentary approval.
Table 6.4 shows the level of parliamentary appropriation (X axis) and the degree
of end of year fl exibility (Y axis). When it comes to parliamentary appropriation, a
number of scenarios are possible:
− Aggregate budget amounts for broad programme or outcome areas: envelope for
each policy area or policy objective;
− Aggregate amounts for Ministries: envelope for each Ministry;
− Aggregate amounts of personnel and aggregate amounts at the Ministry level for
programmes: staff envelope and programme envelope for each Ministry;
− Appropriations at the programme level: envelope for each programme;
− Appropriations split between personnel and programme spending at the pro-
gramme level: staff envelope and programme envelope for each programme;
− Disaggregated appropriations at the programme level: detailed allocation to
budget items within each programme.

249Public administration
End of year fl exibility means that unused resources may be carried over to the fol-
lowing year. The following possibilities exist:
− “Yes, without limit”: unrestricted transfer of operational resources;
− “Yes, up to a maximum percentage”: limited transfer of operational resources;
− “Yes, as approved on a case by case basis by the Ministry of Finance or the Central
Budget Authority”: transfer of operational resources possible only with approval
of the central department responsible for administering the budget;
− “Yes, on a case by case basis according to the underlying statute”: transfer of
operational resources possible only if specifi ed in a statute;
− “Yes, with notifi cation of the legislature”: transfer of operational resources pos-
sible only after parliament has been notifi ed;
− “Yes, with the approval of the legislature”: transfer of operational resources pos-
sible only with parliamentary approval;
− “No”: no transfer of operational resources possible.
The countries with the highest degree of management freedom are in the bottom
left corner of Table 6.4, those with the least discretionary powers in the top right.
The table is based partially on a recent survey of budget practices by the oecd and
the World Bank (oecd and World Bank 2003; Scheers and Sterck et al. 2003). Table
6.4 includes data for the eu-15, the new member states and other oecd countries
such as the United States, Australia and New Zealand.
Table 6.4 Level of budget appropriation and end of year fl exibility in the OECD
level of appropriation
aggre-
gated
pro-
gramme
outcomes
aggre-
gated at
Ministerial
level
aggregated
staff & pro-
gramme
at pro-
gramme
level
at
pro-
gramme
level
staff
& pro-
gramme
expendi-
ture
disaggre-
gated
at pro-
gramme
level
end-year fl exibility
no GR, IE ES US, SI
yes, with parliamentar y
approval
UK NZ
according to statute BE
yes, if approved by Min
Finance
NL AT, HU
yes, up to maximum SE
yes, without limit AU DK FI
Source: OECD and World Bank (2003)

250 Public administration
In some countries, parliamentary appropriation occurs at an aggregated level and it
is possible to carry over unused budget resources at the end of the year, albeit some-
times only under certain conditions. This group includes Australia, Denmark, the
Netherlands and the United Kingdom. Diametrically opposed to this group is a clus-
ter of countries with less management freedom, where there is no end of year fl ex-
ibility and parliamentary appropriation is very detailed (the us, Slovenia, Spain). In
a number of countries, end of year fl exibility is allowed subject to certain conditions,
such as a maximum limit, parliamentary approval or the approval of the fi nance
minister. This is the case, for example, in Belgium, Austria, Hungary, Portugal and
Sweden. These countries combine conditional end of year fl exibility with a relatively
detailed level of parliamentary appropriation. A number of countries have both full
end of year fl exibility and disaggregated appropriation (Finland, Germany and Italy).
A number of others have a highly aggregated budget, but no end of year fl exibility
(Greece and Ireland).
The third indicator of parliamentary control is the legislature’s infl uence on the
budget. To what extent does the budget ultimately approved by parliament differ
from the budget originally submitted by the government? In the majority of the
countries studied, the budget submitted by the executive is approved without major
amendments (less than 3%). This is the case in Denmark, Germany, Spain, France,
Italy, the Netherlands, Austria, Portugal, Finland, the Czech Republic, Hungary,
Slovenia and the United States. Some countries in the oecd-World Bank survey even
reported that parliament approved the budget without any amendments (Belgium,
Greece, Ireland, Sweden, the United Kingdom, Australia, New Zealand and Canada)
(oecd and World Bank 2003).
Results-based budgets
The growing importance of effi ciency and effectiveness has prompted governments
to focus more and more on results as the basis for their budgets. Information on
output and outcome is included in the budget. A number of countries are in fact
moving towards accrual budgeting, a system of costs and benefi ts. Table 6.5 depicts
the trend towards more result-based budgeting. First, we have classifi ed countries on
the basis of the use of output and outcome information in budgets. We then went on
to classify them on the basis of their method of reporting receipts and outlays: cash
or accrual. In a cash-based budget system, the actual cash receipts and expenditure for
the budget period are estimated. An accrual budgeting system focuses on resources used
and the associated costs, and claims on receipts that arise during the budget period.
Table 6.5 shows the percentage of spending programmes in the budget for which
performance information is given (from 0% to 100%). The Y axis shows the method
of charging (from full cash to full accrual). The table is based partially on a recent
survey of budget practices by the oecd and the World Bank (oecd and World Bank
2003; Scheers and Sterck et al. 2003). It includes data for the eu-15, the new member
states and other oecd countries such as the United States, Australia and New Zealand.

251Public administration
Table 6.5 Use of performance information in budget and method of charging
percentage of programs for which per formance information is
contained in the budget documentation
0% 1%-25% 25%-50% 50% -75% 75%-99% 100%
Method of charging
full accrual AU, NZ
cash and accrual IT FI SE
cash and commitment BE DK US, NL
full cash IE, PT,
AT
DE, GR,
CZ
SI ES
Source: OECD and World Bank (2003)
The fi rst cluster consists of countries which have full or partial accrual budgeting
whereby performance information is provided for all programmes (Australia, New
Zealand, Finland and Sweden). The second group comprises countries which include
performance information in the budget documents, but have opted not to make the
move to accrual budgeting (the us, the Netherlands8, Denmark, Slovenia and Spain).
The third group of countries maintain the traditional line-item cash budget geared
mainly to inputs, with little performance information (Ireland, Portugal, Austria,
the Czech Republic and Greece).
A number of positions allocated to countries in Table 6.5 seem questionable. For
instance, according to the oecd-World Bank survey, all of the Netherlands’ budget
programmes include performance information. Reports from the Netherlands’
Court of Audit, however, show that this is not yet the case (Algemene Rekenkamer
2002; Algemene Rekenkamer 2003). There is therefore a gap between aspirations
concerning results-based budgeting and its actual implementation. The positions
of Sweden and Spain should also be put into perspective. However, the oecd and
World Bank survey does not take account of differences between rhetoric and reality.
Multi-year budgets
The third trend in the modernisation of government fi nances has been a move
towards multi-year budgets. Most countries add multi-year forecasts to their budg-
ets to place their annual income and expenditure in a longer-term perspective. In
most cases, these forecasts are purely informative and do not require parliamentary
approval. However, in Italy and the United States, parliament does have to approve
the multi-year budget. Table 6.4 shows which countries have a multi-year budget,
and what exactly its status is.
8 In 2001 the Netherlands announced it intended to introduce accrual budgeting for all
ministries, as it had done previously for baten-lastenagentschappen (executive agencies).
However, in 2003 the Minister of Finance decided to alter this policy towards a partial
implementation of accrual budgeting.

252 Public administration
Table 6.6 Existence and status of multi-year budgets
yes no
does the annual central government
budget documentation submitted to
the legislature contain multi-year
expenditure estimates?
BE, DK, DE, GR, FR, IE, IT, NL, PT,
SE, UK, HU, SI, AU, US, NZ, CA
ES, AT, FI, CZ,
do these expenditure estimates require
authorization by the legislature?
IT, US9AU, AT, BE, CA, CZ, DE,
DK, ES, FI, FR, GR, HU,
IE, NL, NZ, SI, SE, UK
Source: OECD and World Bank (2003)
6.4.2 Human resources management
Personnel policy, or human resources management is another horizontal policy area
within the public sector currently experiencing change. Strategic human resources
policy, competency management, equal opportunities policy and public service moti-
vation are key concepts in the modernisation of human resources management (hrm)
in the public sector. One important choice in public-sector hrm policy is whether
to attempt to ensure that the staff profi le refl ects the composition of the population.
Moreover, the Lisbon agenda aims to raise the employment rate of women from an
average of 51% in 2000 to more than 60% by 2010. This section therefore looks at the
number of female public servants and the age profi le of the public service.
Figure 6.8 shows the proportion of female staff in the public administration and
defence sector.10
9 The multi-year expenditure estimates are approved by Congress, apart from the appro-
priation bills for the budget year
10 isic Rev. 3 (un Classifi cations Registry): isic l – Public administration and defence;
compulsory social security & isic m – Education. Public servants can also be employed
in other categories, M (health and social work) or O (other community, social and
personal service activities).

253Public administration
Figure 6.8 Employment of women in public administration, defence and compulsory
social security, 2002
Source: OECD online database
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CY
CZ
EE
LV
LT
HU
MT
PL
SI
SK
AU
CA
NZ
US
0
10
20
30
40
50
60
70
An average of 47% of staff in the public administration and defence sector are women.
The average is much higher in education, at 69%, while in the health care and social
welfare sector it is no less than 80%. We shall focus our attention here on the propor-
tion of female staff in public administration, defence and compulsory social security.
The proportion of female staff in this sector is well above average in Denmark, Finland
and Sweden. In Hungary and Slovakia, too, there are more female than male staff in
this sector. The United Kingdom and New Zealand also score above average. A number
of countries fl uctuate around the average: Ireland, Germany, France, Poland, the Czech
Republic and Canada. In a number of Southern European countries, such as Spain, Por-
tugal and Italy, the representation of women in the public sector is some 10% lower than
the overall average. The proportion is even lower in Greece. Notably, the proportion of
female staff in the Dutch public administration and defence sector is also well below
average (38%). Figures from Statistics Netherlands confi rm and indeed reinforce this
conclusion, indicating that only 35% of jobs in public administration and compulsory
social security11 are held by women (cbs 2004).
Figure 6.9 shows staff in the public sector by age group. The graph reveals that 40- to
49-year-olds are most strongly represented in the public service. Exceptions are Sweden,
where 50- to 59-year-olds constitute the largest group, and Germany, which has relatively
young public servants. In the majority of countries, there are more older public servants
(over-50s) than young public servants (aged 20 to 29). Ireland and Germany, however,
seem to have enjoyed more success than other countries in attracting young people to the
public service. In these two countries, staff in their 20s account for more than a quarter
of all public servants, and the 18-19 age group is also relatively large.
11 Category 75, sbi 1993.

254 Public administration
Figure 6.9 Staff in central/federal government by age group, 2002
Source: OECD/PMC (2002)
DE GR FR IE LU NL AT FI SE HU AU US
0
5
10
15
20
25
30
35
40
45
50
< 19 years old 20-29 30-39 40-49 50-59 60 and above
The problem of demographic ageing also presents the public sector with a major
challenge. The preponderance of staff aged over 50 in the public service has
increased steadily since 1990. As the baby boom generation (born between 1945
and 1955) retires over the next ten years, the public service may face a growing staff
shortage. Personnel planning and the recruitment of young staff by promoting
the public sector as a good employer are therefore key objectives of current human
resources policy.
6.4.3 e-Government
The development of a knowledge-based society also has implications for the services
and communications of the public sector. There is a general tendency in the public
sector towards automating bureaucratic procedures and processes, and electronic
interaction with citizens. However, e-government is a very broad concept, ranging
from electronic communications, via online services to e-democracy and e-participa-
tion. e-Government is one of the newest forms of modernisation in the public sector,
and it is being followed closely by research centres and international organisations.
In this fi eld, benchmarks are frequently used to compare and rank countries (Jans-
sen et al. 2004). The focus in this section is on three such benchmarks.
The United Nations and the American Society for Public Administration (aspa)
have developed an e-government index based on countries’ offi cial online presence,
their telecommunications infrastructure and their human development capacity.
This benchmark defi nes e-government as utilising the Internet and the World Wide
Web to deliver government information and services to citizens. Table 6.7 ranks
countries on the basis of this benchmark.

255Public administration
Table 6.7 Ranking of EU-15, accession countries and other OECD countries in the e-govern-
ment index compiled by the United Nations and the American Society for Public
Administration
1 US 11 FI 21 LV
2 AU 12 FR 22 LI
3 NZ 13 ES 23 GR
4 UK 14 IT 24 SK
5 CA 15 IE 25 SI
6 NL 16 P T
7 DK 17 AT
8 DE 18 CZ
9 SE 19 EE
10 BE 20 PL
Source: ASPA and UN (2002)
The second benchmark, the Networked Readiness of Nations (nrn) was compiled by
Harvard University’s Centre for International Development (Table 6.8). It uses a
broad defi nition of e-government, which includes e-administration, e-business, the
presence of infrastructure and it know-how. The indicators used in this benchmark
include the number of online services, the number of online transactions, subsidies
for e-government, promotion of it by government and the e-government environ-
ment (e.g. public access to the Internet).
Table 6.8 Ranking of EU-15, accession countries and other OECD countries in the Networked
Readiness of Nations Benchmark
1 FI 11 FR 21 LV
2 US 12 IE 22 PL
3 SE 13 BE 23 SK
4 CA 14 NZ 24 GR
5 UK 15 EE 25 LI
6 DK 16 ES
7 DE 17 IT
8 NL 18 CZ
9 AU 19 PT
10 AT 20 SI
Source: Harvard University Centre for International Development (2003)
The third benchmark is the e-Readiness Ranking (err) 2003 produced by the Econo-
mist Intelligence Unit (Table 6.9). Here, too, a broad defi nition of e-government is
used. Indicators include public expenditure on it as a proportion of gnp, the quality
of Internet connections, fi nancial support for it projects, and the number of pcs and
Internet connections.

256 Public administration
Table 6.9 Ranking of EU-15, accession countries and other OECD countries in the e-Readiness
Ranking (ERR), 2003
1 SE 7 CA 13 BE
2 DK 8 AU 14 FR
3 US 9 DE 15 IT
4 UK 10 AT 16 PT
5 NL 11 IE 17 ES
6 FI 12 NZ 18 GR
Source: EIU/IBM (2003)
Although the three benchmarks do not use the same indicators, the correlation
between ranking orders in Tables 6.7 through 6.9 would appear to be signifi cant.
Countries that score high on one benchmark also tend to score high on the others.
The only exceptions are Finland and New Zealand, for which varied pictures emerge.
The us has the best average score on all three e-government benchmarks, and
three Scandinavian countries appear in the top fi ve: Sweden (2), Finland (3) and
Denmark (5).12 The United Kingdom comes in fourth place. The majority of coun-
tries have an average score. This group includes a number of countries in continen-
tal Europe: the Netherlands (7), Germany (8), Austria (10), France (11) and Belgium
(13). Canada (6), Australia (9) and New Zealand (12) are also in this group. A small
number of countries score below 5: Ireland (14) and a group of Southern European
countries (Italy (15), Spain (16), Portugal (17) and Greece (18).
The United Nations and aspa Networked Readiness of Nations benchmark also
includes data on the newly acceded eu member states (with the exception of Malta
and Cyprus). These countries score less well than the eu-15 and Canada, the United
States, Australia and New Zealand. The new member countries that have progressed
furthest in the fi eld of e-government are Estonia (1), the Czech Republic (2) and
Slovenia (3), followed by Poland (4), Latvia (5) and Slovakia (6). Lithuania (7) has the
poorest scores.
6.5 Performance
A well-functioning and reliable public administration is an essential factor for
economic growth. However, there are no objective indicators of the effi ciency and
effectiveness of public administration. Because the whole nation takes advantage of
services produced by public administration, they can be classifi ed as social goods:
consumption is non-rival and nobody can be excluded from consumption (Musgrave
and Musgrave 1984). In the absence of individual consumers, the functioning of the
12 The average score differs from the average ranking; especially the ranking of Germany,
Finland and Canada differs: Germany takes position 5 on the base of average scores and
position 9 on the base of average ranking; Finland takes position 3 respectively 6 and
Canada takes position 7 respectively 4.

257Public administration
public administration can only be measured by subjective indicators. These indica-
tors refl ect trust of the population in public administration and confi dence in the
Civil service. Of course, the lack of information about the performance of public
administration can be explained by the very nature of its products: legislation, policy
and management. No ‘natural’ performance indicators are available for these activi-
ties. However, some public administration activities can be regarded as individual
services, like the issue of passports or the entering of transactions in the land regis-
ter. In Kuhry en Veldheer (2004) objective performance indicators are defi ned for the
majority of municipal services. However, we don’t have acccess to similar detailed
data for other countries.
While subjective indicators yield useful information, they must be approached
with caution, for two reasons (Van de Walle et al. 2004). Firstly, a negative attitude
towards the public sector on the part of the general public can lead to a negative per-
ception of its performance. Such an attitude might arise from the general cultural
context, and therefore have little to do with objective performance. Secondly, expec-
tations of the citizenry will affect their perception of public sector performance.
Satisfaction and perception are determined partly by expectations. In countries with
traditionally strong services, people will expect a lot of the public sector. If their
expectations are not met, they will award low performance scores, even if the objec-
tive performance of the public sector is better than in countries where people expect
less of their government.
It is already hard to measure the performance of public administration at all, but
even more diffi cult to make cross-national comparisons of administrative perform-
ance. Only global judgements about the functioning of government and the public
administration can be used to measure the performance of public administrations.
The Dutch government states that citizens may expect an integer, transparent,
credible, responsive, effective and effi cient government in exchange for taxes and
trust they make available (TK 2003). In general, public administrative systems are
founded on the following principles (Vidlakova 1999): trust and credibility, openness
and transparency, accountability, effi ciency and effectiveness. These principles were
an important condition for the new member states to enter the European commu-
nity.
6.5.1 Quality of government
The well functioning of the government is an important dimension of the performance
of public administrations. A major source for measurement of well functioning govern-
ments- called hereafter government quality - is a survey among representatives of the busi-
ness community in a range of countries (imd 2003). More than 4,000 respondents in about
60 countries answered questions about economic performance and government quality.
Four indicators of government quality are selected, based on the principles of well
functioning administrative systems:
− level of bureaucracy: does bureaucracy hinder business activity?

258 Public administration
− level of transparency: is transparency of government policy satisfactory?
− level of effectiveness: are government decisions effectively implemented?
− level of corruption: do bribing and corruption exist in the economy?
The level of corruption is included in the imd index of government effi ciency.
However, because Transparency International publishes a widely used composite
index of corruption, including the imd index, this composite index will be used
instead of the imd index (www.transparency.org).13 On the next indices, zero is the
worst possible score and ten the best possible.
0
1
2
3
4
5
6
7
8
9
10
0 1 2 3 4 5 6 7 8 9 10
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CZ HU
PL
AU
CA
NZ
US
EU-15 new member states non-EU Anglo-Saxon
Figure 6.10 Bureaucracy does not hinder business activity, 1995 and 2003a
a scale: 0-10 (0 = total bureaucracy; 10 = no bureaucracy);
Scores only available for 2003: EE (4,7), SI (2,2), SK (1,9)
Source: IMD (1995, 2003)
-
+
2003
1995
Bureaucracy seems to hinder economic activities more in southern then in northern
European countries (Figure 6.10). In 2003, countries with the most favourable outcome
13 This composite index of Transparency international (ti) is compiled of different
indices, of which the imd index, World economic forum index, World bank index and
Gallup international index. Both corruption indices (imd and ti) are highly correlated
(r > 0,96).

259Public administration
are Finland, Denmark, Luxembourg and Sweden. The ranking of Germany amongst the
less performing southern countries is striking. Differences between the new member
states are also worth mentioning. Between 1995 and 2003 the perception of the level of
bureaucracy shows an upward trend. Especially in New Zealand but also in Germany,
Sweden and the United Kingdom the level of bureaucracy has risen substantially. Luxem-
bourg and Finland counter this trend and have moved to a lower level of bureaucracy.
Transparency of government is judged markedly more positive then is bureaucracy
(Figure 6.11). Even though differences between countries are smaller, roughly the
same ranking emerges. Within the eu-15 area, northern countries generally perform
well and southern countries perform less. But this pattern is less manifest then with
bureaucracy. Compared with its position on bureaucracy, Portugal performs better
on transparency. Obviously, the level of transparency has strongly improved between
1995 and 2003. Only two countries, Germany and the Czech republic, deviate from
this positive trend. Transparency has increased most in Finland, Luxemburg, Spain,
Portugal, Hungary and the United Kingdom.
2
3
4
5
6
7
8
9
10
2 3 4 5 6 7 8 9 10
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CZ
HU
PL
AU
CA NZ
US
EU-15 new member states non-EU Anglo-Saxon
Figure 6.11 Transparency of government is satisfactory, 1995 and 2003
a
a scale: 0-10 (0 = no transparency; 10 = full transparency)
Scores only available for 2003: EE (6,0), SI (4,3), SK (5,1)
Source: IMD (1995, 2003)
2003
1995
-
+

260 Public administration
An important indicator for the quality of the government is effective implementa-
tion of decisions and regulations. Government decisions seem to be implemented
very well in Scandinavian countries (Figure 6.12). Greece, Germany and Italy seem
to experience some problems in implementing government decisions. Differences
between the new member states are smaller, as compared with bureaucracy and
transparency, but their ranking is about the same. The perception of effective imple-
mentation of governmental decisions did not change much between 1998 and 2003.
Some countries improved the implementation of their decisions according to the
business community (Australia, Sweden) but some did not (Netherlands, Canada,
Ireland and the United Kingdom).
2
3
4
5
6
7
8
9
10
2 3 4 5 6 7 8 9 10
EU-15 new member states non-EU Anglo-Saxon
Figure 6.12 Government decisions are effectively implemented, 1998 and 2003a
a scale: 0-10 (0 = not effective; 10 = full effective)
Scores only available for 2003: EE (5,1), SI (3,9), SK (3,6)
Source: IMD (1998, 2003)
2003
1998
-
+
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CZ
HU
PL
AU
CA
NZ
US
Another important dimension of the quality of government is absence of corruption.
The differences between European countries are signifi cant. Confi dence in public
services appears to be linked to perceived corruption. In countries where there is
believed to be a lot of corruption, citizens do not tend to have confi dence in the
administration.

261Public administration
2
3
4
5
6
7
8
9
10
2 3 4 5 6 7 8 9 10
EU-15 new member states non-EU Anglo-Saxon
Figure 6.13 Perception of corruption, 1997 and 2003a
a scale: 0-10 (0 =total corruption; 10 = no corruption)
Source: www.transparency.org
2003
1997
-
+
BE
DK
DE
GR
ES
FR
IE
IT
NL
AT
PT
FI
SE
UK
CZ
EE
LV
LT
HU
PL
SI
SK
CA NZ
US
LU AU
Figure 6.13 compares scores on Transparency International’s corruption perceptions
Index for the eu-15 and the new member countries in 1997 and 2003. The Scandi-
navian countries head again, and the central European countries trail behind the
rest. Among the Western European countries, the poor scores of Italy and Greece are
striking, as well as the poor scores of most new member states in eastern Europe.
Belgium, in particular, is experiencing an upward trend. Italy, Lithuania and Latvia
are also making progress, although – on the basis of these results – they still have
a long way to go to reach the current European average. Poland, the Czech Republic
and Greece have declining scores and Latvia, Lithuania and Spain have rising scores.
6.5.2 Openness of government
Freedom of information is a fundamental right in a democracy. The importance of
open and transparent government is universally recognised. Although freedom of
information has been enshrined in law for centuries in some countries, most of the
legislation allowing access to government information has been passed in the last
ten or twenty years, in response to the democratisation of countries in transition,

262 Public administration
scandals, the modernisation of the public sector, the growth of the knowledge-based
society and international pressure. Sweden was a pioneer, passing its Freedom of the
Press Act in 1766. Other early birds were the United States (Freedom of Information
Act 1966), France and Australia. Most industrialised countries took steps towards
legislation on access to government information in the 1980s and early 1990s. One
exception is the United Kingdom, which did not pass its Freedom of Information Act
until 2000. This particular legislation will not come into force until 2005. In Central
and Eastern Europe, the fall of the Communist regimes allowed freedom of informa-
tion to be enshrined in the constitution and the statute books. Most of these coun-
tries passed freedom of information legislation in the late 1990s/early 2000s.
Access to information is a constitutional right in most countries. The openness
of government is also regulated by legislation. Firstly, there are laws that determine
procedures. Freedom of Information Acts stipulate the type of documents that
should be in the public domain, institutions subject to the legislation, response
times, any exceptions and appeal procedures in the event of a refusal. The most
common reasons cited for refusing access to information are national security and
international relations, privacy, commercial confi dentiality, law and order, respect
for the confi dentiality in which information was given and the internal nature of
certain debates among policy makers. Most countries also have specifi c legislation
relating to classifi ed information. The protection of personal data and access to
individual information are also regulated separately in most countries. Legisla-
tion gives people access to their own personal fi les held by both public and private
institutions.
Deadlines for responding to requests for information differ from country to
country. In Sweden, Denmark and Belgium the authorities are obliged to respond
immediately. On average, legislation stipulates a maximum of 19 days for the release
of the information. Spain has the longest deadline, with offi cials having up to three
months to furnish the requested information.
Table 6.10 shows the degree of government openness in the eu-15, the accession
countries and other oecd countries. The fi gures refl ect whether the principle of
transparency is laid down in the constitution, when the fi rst freedom of information
laws were passed, what legislation on transparency is currently in place, the situa-
tion regarding classifi ed information and protection of and access to personal data
and, fi nally, the deadline for responding to requests for information. The data are
based on surveys of the openness of government (Banisar 2003; Mendel 2003).

263Public administration
Table 6.10 Openness of government in the EU-15, the accession countries and other
OECD countries
con-
stitu-
tion
current freedom of Information ;
legislation
legislation on classifi ed
information
legislation on the
protection of and access
to personal data
SE yes Freedom of the Press Act 1976 Secrecy Act 1980 Personal Data Act 1998
DK no Acces s to Public Administration;
Files Act 1985
- Act on Processing of
Personal Data
US no Freedom of Information Act 1966
as amended by the Electronic
Freedom of Information Act 1996
Presidential Records Act Privacy Act of 1974
FR - Law on Access to Administrative
Documents 1978
Loi du 8 juillet 1998
instituant une Commis-
sion ; Consultative du
secret de la défense
nationale
1978 Data Protection Act
NL yes Government Information (Public
Access) Act 1991
- Personal Data Protection
Act of 2000
NZ no Offi cial Information Act 1982 Of fi cial secrets Act 1951 Privacy Act 1993
AU no Freedom of Information Act 1982 266 (2001-2002) Privacy Amendment Act
2000
CA no Access to Information Act 1983
as amended by the Terrorism Act
2001
Security of Informa-
tion Act
Privacy Act of 1985
AT yes Federal L aw on the Duty to Fur-
nish Information 1987
- Data Protection Act 2000
IT no Law n°241 of 7 August 1990 - Data Protection Act of
1996
HU yes 1992 Protection of Per sonal Data
and Disclosure of Data of Public
Interest
Secrecy Act of 1995 1992 Protection of Per-
sonal Data and Disclosure
of Data of Public Interest
ES yes Law on Rules for Public Adminis-
tration 1992
- Personal Data Protection
Act 1999
PT yes Law of Access to Administrative
Documents 1993
Law of State Secrecy
1994
Act on the Protection of
Personal Data 1998
BE yes Wet van 11 april 1994 op de
openbaarheid van bestuur
Law on the Security of
Information 1998
Law on the Protection of
Personal Data 1992
LI yes Law on the Provision of Informa-
tion to the Public 1996
Law on State Secrets
and Offi cial Secrets
1999
Law on Legal Protection of
Personal Data 2003
IE no Freedom of Information (Amend-
ment) Act 2003
Offi cial Secrets Act 963 Data Protection Act 1988
as amended by the Data
Protection (Amendment)
Act of 2003
LV yes Law on the Freedom of Informa-
tion 1998
State Secrets Act 1996
as amended in 2001
Law on Personal Data
Protection

264 Public administration
Table 6.10 Openness of government in the EU-15, the accession countries and other
OECD countries
con-
stitu-
tion
current freedom of Information ;
legislation
legislation on classifi ed
information
legislation on the
protection of and access
to personal data
FI yes Act on the Openness of Govern-
ment Activities 1999 -Personal Data Act 1999
CZ Law on Free Access to Informa-
tion 1999
Protection of Classifi ed
Information Act 1998
2000 Data Protection Act
GR yes Code of Administrative Procedure
1999
- Law on the Protection of
Individuals with regard to
the processing of personal
data
EE yes Public Information Act 2000 State Secrets Act 1999 Data Protection Act
SK yes Act on Free Access to Information
2000
Act on Protecting Classi-
fi ed Information 2001
Act on Protection of
Personal Data
UK no Freedom of Information Act 2000 Offi cial Secrets Act
1989
Data Protection Act 1998
PL yes Law on Access of Public Informa-
tion 2001
Classifi ed Information
Protection Act 1999
Act on Protection of
Personal Data 1997
SI yes Act on Access to Information of
Public Character 2003
Classifi ed Information
Act 2001
Personal Data Protec-
tion Act
-: not available
Source: Banisar (2003); Mendel (2003)
6.5.3 Confi dence in the Civil service
Confi dence in public institutions has been widely researched by political scien-
tists. In general, trust in government is declining in almost all modern welfare
states (Dalton 2004). The World Values Survey (wvs) is the most important source
to investigate cross-national differences in confi dence and trust. This worldwide
survey, conducted by a network of social scientist around the world, is performed on
nationally representative samples of at least 1,000 respondents and includes ques-
tions about confi dence in different public institutions. According Rothstein & Stolle
(2002) three types of public institutions can be distinguished: political (parliament,
government), controlling (media) and implementing (police, justice, school, hospi-
tal) institutions. The civil service is in between those three types of public institu-
tions, but resembles most the controlling type.
Confi dence in the Civil service differs strongly across countries (Figure 6.14).
The eu-15 countries are both scattered at the left side (much confi dence) as well as
the center and right side (less confi dence) of the range of values. The same holds for
the non-eu Anglo-Saxon countries and – to a smaller degree - for the new member
states. Residents of Luxembourg and Ireland are the most positive about their Civil
service and the residents of Greece and Lithuania are the least positive.
(cont.)

265Public administration
Figure 6.14 Confidence in the Civil service, 2000
Source: European Values Survey (2000); World Values Survey (1995-1997)
LU
IE
US
DK
UK
PT
CA
HU
LV
MT
SE
BE
FR
AT
FI
ES
EE
DE
SK
AU
NL
IT
PL
NZ
SI
CZ
LT
GR
0
10
20
30
40
50
60
70
80
90
100
none at all not very much quite a lot a great deal
Confi dence in the Civil service can be compared with confi dence in other institu-
tions delivering public sector services, like health care, safety, justice, education and
social security (Figure 6.15). On average, confi dence in the Civil service amounts to
about 70% of mean confi dence in other public sector institutions. Only in Portugal
and Hungary the Civil service is able to compete with other public sector institu-
tions. The Civil services in Ireland, the United Kingdom, Luxembourg and Latvia
perform relatively well on confi dence, but this is certainly not the case in Slovenia
and Finland.

266 Public administration
Figure 6.15 Confidence in the Civil service as percentage of average rating of public
sector services, 2000
Source: European Values Survey (2000)
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CY
CZ
EE
LV
LT
HU
MT
PL
SI
SK
AU
CA
NZ
US
0
20
40
60
80
100
120
An important issue in some countries is the level of decentralization of public services.
One argument in favour of decentralization is better fi ne-tuning of supply of public
services to demand, because the needs and preferences of citizens are in principle
better known at the local level. One argument against decentralization is the unequal
regional allocation of public services that may result. In general, the population sup-
ports the idea to enlarge the competence of local authorities (Figure 6.16). On average,
about 46% of the residents of all selected countries are in favour of handing down
greater powers to local authorities. A lot of residents don’t have an opinion on this
matter and only a minority of 25% feel decentralization to be a bad idea.

267Public administration
Figure 6.16 Percentage of population that thinks that more power to local authorities is
a good thing, 2000
Source: European Values Survey (2000)
BE
DK
DE
GR
ES
FR
IE
IT
LU
NL
AT
PT
FI
SE
UK
CY
CZ
EE
LV
LT
HU
MT
PL
SI
SK
AU
CA
NZ
US
0
10
20
30
40
50
60
70
80
Opinions differ strongly between countries. Especially in the new member states,
except for Hungary, residents are in favour of shifting power to local authorities.
This is not surprising, given the high level of centralization in the former socialist
republics. Nowadays, the process of decentralization is well under way. In the eu-15,
only residents of the Netherlands are clearly in favour of central government and
don’t like to give more power to local authorities.
6.5.4 Quality, confi dence and expenditure
The public administration produces public services like legislation, public policy and
public management. In general, no clear indicators are available to measure the pro-
duction of this kind of public services. Because of the lack of objective indicators, the
effectiveness of public administration is measured here by using subjective indica-
tors. In previous sections two types of subjective indicators were presented: subjec-
tive judgements of government quality and confi dence in the Civil service. Subjective
quality refers to the perception of governmental bureaucracy, transparency, effec-
tiveness and corruption expressed by representatives of the business community in
different countries (imd 2003). Confi dence refers to the perception of the perform-
ance of the Civil service by a representative sample of the population (wvs 2000).
Effectiveness of public administration can be expressed in subjective terms by relat-
ing expenditure to subjective quality (Figure 6.17) and confi dence (Figure 6.18).
Subjective quality of the government is a composite index of perceived bureaucracy,
transparency, effectiveness and corruption, and is related to expenditure on general
public services per capita.

268 Public administration
Figure 6.17 Subjective quality of government (2003) and expenditure per capita on
general public services, 2001 (nl €)
Source: IMD (2003) and Eurostat
0
1
2
3
4
5
6
7
8
9
10
0 500 1000 1500 2000 2500
UK
ES
PT
IE
GR
DE IT
DK
BE
FI
NL AT
SE
FR
LU
In both fi gures there is a weak relation between expenditure and subjective per-
formance. The most striking observation is the location of Finland and Denmark,
combining fairly moderate expenditure with high subjective quality. Luxembourg is
also characterized by high quality, but pays a considerably higher price.The lowest
spending countries in Figure 6.17 perform in general at a lower level and the high-
est spending countries at a higher level. In Figure 6.18 this weak relation is hardly
observable. As far as there is any relation between expenditure on general public
services and subjective performance, this relation cannot be interpreted as a causal
relation without closer investigation. Other factors related to expenditure as well as
subjective performance can explain the observed relation. In this report, the rela-
tions between expenditure and performance are not checked on possible intervening
factors.
Expenditure on general public services can also be expressed in relative terms, as
percentage of GDP, and related to quality and confi dence. However, in general this
does not change the presented pictures. Of course, Luxemburg moves halfway to the
left (above the Netherlands) and Ireland and Spain change positions in fi gure 6.17.
But the favourable positions of Finland, Denmark and - to a lower degree - the uk in
fi gure 6.17 are preserved, as are the unfavourable positions of Greece, Italy, Ger-
many and France. In fi gure 6.18 only Luxembourg moves halfway to the left and the
remainder of the picture is roughly preserved.

269Public administration
Figure 6.18 Confidence in the Civil service (2000) and expenditure per capita on general
public services, 2001 (nl€)
Source: European Values Survey (2000) and Eurostat
0
10
20
30
40
50
60
70
80
90
100
0 500 1000 1500 2000 2500
UK
ES
PT
IE
GR
DE
IT
DK
BE
FI
NL
AT
SE FR
LU
A closer examination of both plots shows a fairly similar pattern. However, Finland
has fallen from the top (Figure 6.17) to the lower part of the middle group (Figure
6.18). The resemblance of both plots seems to indicate a strong correlation between
subjective quality of government and confi dence in the Civil service. Figure 6.19
demonstrates the relationship bet ween both types of subjective indicators, including
some other countries too. Over half of the countries are located close to the diago-
nal, indicating a close relationship between subjective quality of government and
confi dence in the Civil service. In just over a quarter of all countries this relation-
ship is somewhat weaker, but still strong. Only three countries show a greater gap
between both indicators: Ireland, Australia and Finland. In Ireland, the population
expresses slightly more confi dence in the Civil service then the business community
in the perceived quality of the government. In Australia but especially in Finland it’s
just the opposite: more quality perceived by the business community and less confi -
dence expressed by the population.

270 Public administration
Figure 6.19 Subjective quality of government (2003) and confidence in the Civil service,
2000
Source: European Values Survey (2000); World Values Survey (1995-1997); IMD (2003)
0
10
20
30
40
50
60
70
80
0 1 2 3 4 5 6 7 8
UK
ES
PT
IE
GR
DE
IT
DK
BE
FI
NL
AT
SE
FR
LU
CZ
HU
PL
SK AU
CA
US
EU-15 new member states non-EU Anglo-Saxon

271
7 Performance of the public sector
Bob Kuhry and Evert Pommer
7.1 Introduction
Previous chapters have assessed the performance of twenty-nine countries in several
policy areas: the economy (Chapter 2), education (Chapter 3), health care (Chapter 4),
law and order (Chapter 5) and public administration (Chapter 6). In some cases, the
analysis covered less than twenty-nine countries because of missing data. This con-
cluding chapter aims to synthesise the fi ndings presented in earlier chapters. Section
7.2 attempts to group nations in several classes by taking into account system charac-
teristics in all policy areas considered in the report. Section 7.3 describes aggregated
performance in the fi elds of education, health care and law and order. Section 7.4
attempts to group nations on the basis of aggregate public service sector performance.
Sections 7.5 describes aggregated performance and relates it to aggregated confi dence.
The fi nal section summarizes core results for a number of different regions which,
irrespective of the subject, appear to emerge from all cluster analyses: Northern, West-
ern, Southern en Central Europe and the Anglo-Saxon countries.
7.2 Grouping welfare states by type
Existing typologies of welfare states
In a strand of the literature, the performance of welfare states is linked to their
institutions. A typology aims to explain the performance of national institutions
in the light of their key characteristics. One well-known typology of institutions for
social protection found in welfare states was developed by Esping-Andersen (1990).
In his approach, the defi ning characteristic of welfare states is the generosity and
accessibility of government programs designed to protect the citizenry against loss
of income and poverty. Esping-Andersen proceeds by grouping countries on the
basis of a historical sociological theory, and uses an analytical sociological model to
defi ne three types of welfare state that perform differently in their efforts to provide
social protection: liberal, corporatist and social-democratic. Each type is different
in terms of the regulation of labour markets (primary protection) and the level and
scope of income guarantees (secondary protection). The Social and Cultural Plan-
ning Offi ce – using empirical data – has also classifi ed welfare states on the basis of
social protection offered to citizens. Eastern European countries emerge as a sepa-
rate type of welfare state.1
1 See: scp/cerp 2004.