‘New Public Managers’ in Europe: changes and trends · 2010-12-03 · 3 ‘New Public Managers’ in Europe: changes and trends Sandra van Thiel, Bram Steijn and Marine Allix

1

‘New Public Managers’ in Europe:

changes and trends

Sandra van Thiel, Bram Steijn and Marine Allix

Paper to be presented at the IRSPM 9 Conference in Milan, 5-8 April 2005

Please do not cite **** work in progress **** please do not cite **** work in progress **** Correspondence to: Dr. S. van Thiel Public Administration Erasmus University Rotterdam P.O. Box 1738 (room M8-27) 3000 DR Rotterdam The Netherlands E-mail: [email protected]

2

3

‘New Public Managers’ in Europe: changes and trends

Sandra van Thiel, Bram Steijn and Marine Allix

March 31st, 2005

Introduction

In most western countries administrative and managerial reforms in the public sector have

affected the profile, labour position and functioning of top civil servants since the 1980s. This

paper describes two important changes. First, senior civil servants have become public managers;

as a consequence the conditions of their work have changed (cf. Noordegraaf, Ringeling &

Zwetsloot, 1995). Second, a new category of ‘new public managers’ has arisen; the top

executives of quasi-autonomous agencies (Pollitt & Talbot, 2004). Based on the motives for the

creation of the latter category, we expect a number of differences between the job profile of the

two categories of new public managers. Therefore, after a description of their respective profiles,

we will investigate these expectations. We will base our analysis mainly on our own surveys and

secondary sources from the Netherlands. However, as these developments are occurring in other

countries as well, we will also use sources from other countries to illustrate our main points.

The outline of this paper is as follows. First we will briefly sketch the main characteristics

of personnel management reforms that have occurred since the early 1980s in different western

European countries. Next, we will go into the motives for the establishment of quasi-autonomous

bodies and speculate on the effects thereof for the position and job profile of top executives,

compared to the top public managers in ministries. Then we will use information from different

surveys to describe the profile of top public managers and top executives, and investigate the

differences between them. Data collection on this topic is still in its early stages, so we will not

be able to offer information about all characteristics, but we believe that there is enough data to

give a crosscutting, representative image of what is happening. In the conclusions of this paper

we will formulate some questions for future research into the new public managers and how they

function.

4

Personnel effects of public sector reform

Public sector reform occurred in most western countries from the 1980s on. To size down and

reduce budgets, governments decided to restructure their administrative organization and change

the way in which they were working (for an overview see e.g. Kickert, 1995; Pollitt & Bouckaert,

2000; OECD, 2002; Christensen & Laegreid, 2003). Clearly, these reforms have had their effects

on the position and functioning of the managers of these organisations. Within government

bureaucracies, new management techniques were introduced, usually originating from the private

sector like performance indicators and human resource management (cf. Bovaird & Loeffler,

2003). Of course, this has lead to changes in the way senior managers have to manage their

organisations and departments. Next to restructuring and abandoning tasks, parts of the

(executive) organization and employees were put at arms’ length – resulting in a more

autonomous position for the managers involved. As a result, a plethora of public bodies and

executive agencies were created (Greve, Flinders & Van Thiel, 1999; Pollitt & Talbot, 2003).

Although this paper deals only with reform at central level, other government levels were

affected in similar ways (Kersting & Vetter, 2003).

To some extent these reforms led to a reduction of the number of civil servants working in

ministries, but such numbers are obscured by the fact that in some cases the employees of the

newly established executive agencies remained civil servants after their transfer (Talbot,

2003:10). For example, in the Dutch case employees of agencies (agentschappen) remain civil

servants. In 2002, almost a third of all civil servants at the Dutch national level were working in

these agencies. It is expected that in 2006 that figure will increase to almost 80% (Van Thiel &

Van Oosteroom, 2004). Moreover Dutch ZBOs, another type of public body, employ more

people (over 120.000 in 1992; Van Thiel, 2001) than all ministries together. Approximately 60%

of the 430 ZBOs in 2001 were public law based organisations and often former units from

ministries (Van Thiel & Van Buren, 2002). Some of the employees are civil servants, others not.

Exact figures are not available. The Dutch example shows that shifts in the numbers of civil

servants are difficult to isolate. That is probably why official counts of the total number of public

sector employment show little overall changes in the last decade (table 1).

5

Table 1. Employment in ministries and agencies at central level, 2001 in FTEs Ministries Agencies Total public sector % Change 1991-2001Austria 11,321 150,238 438,264b +0.1%Belgium 55,791 18,319 76,439b +0.4%Czech Republic*a 10,558 3,766 673,438 n.a.Finland 5,357 110,736 n.a. n.a.France 2,212,573b,c n.a. 4,344,934a,c +2%Germany* 23,042 478,978 n.a. n.a.Greece 22,903 9,187 n.a. n.a.Hungary 7,664 10,597 772,502 -2%Ireland 19,799 10,292 259,243 +29%Italy*a 280,779 n.a. n.a. n.a.Luxembourg 956 4,008 n.a. n.a.Netherlands 107,165 24,055 727,849b +7%Norway 3,389 n.a. 111,046b -1%Poland*b 9,064 4,435 2,151,044 +3%Portugal 481,921 n.a. n.a. n.a.Spain 208,175 n.a. n.a. n.a.Sweden 204,000 154,900 204,000 -11%Switzerland* 33,791 n.a. n.a. n.a.* Headcount only a 1999 b 2000 n.a. not available c Including teachers etc. Source: PUMA (2002)

We would therefore expect to find the effects of public sector reform on personnel less in their

numbers, but more in the conditions of their job and the way in which public servants are

expected to operate. It is our contention that as a result of public sector reform two major changes

with respect to the managerial function have occurred (cf. figure 1). First, senior civil servants

have become public managers (cf. Laegreid, 2003:145). And second, a new type of public

manager has been created: the directors or executives of executive agencies and other types of

public body. Together these two categories form what we will call New Public Managers (cf.

Barlow et al., 1996:7). Next we will describe these changes, after which we will focus on a

number of differences that we expect to exist between the two categories of managers.

6

Before public sector reform After public sector reform

Two types of New Public Managers

Senior Civil servants Public managers (in ministries)

Top executives (in executive agencies etc.)

Figure 1. Effects of public sector reform on civil servants

Transforming civil servants

To make governments more efficient, it was believed that they had to become more like a

business or market. As a result, the boundaries between the public and private domain were

lifted. Market-type mechanisms were introduced like internal markets in the public sector (e.g.

hospitals and schools competing with each other over budgets). Also, co-operation with private

sector organisations increased, both in public-private partnerships and through outsourcing of

public tasks. This required new skills and routines of senior civil servants (Laegreid, 2003). Next

to policy-making, they had to be able to negotiate with private organisations, draw up contracts

and employ business strategies for their own organisation. To facilitate these new skills and

routines, senior civil servants were given more flexibility (managerial autonomy) and

responsibility (accountability requirements) in their work. This is reflected in the reform of

personnel policies in the public sector. These reforms focussed on three issues: (i) a reduction of

security of permanent tenure by for example appointing top officials on a temporary basis and (ii)

often on performance-related contracts because (iii) performance and managerial competence

became more important criteria in decisions to hire and fire top managers in the public sector,

like in the private sector (Pollitt & Bouckaert, 2000:72-77).

The implementation of these changes depends on the type of civil service system. In this

respect, two ideal type systems can be distinguished: career based and position based (PUMA,

2003). Table 2 presents a very global summary of these systems and the aims of the changes that

have been planned or implemented. Individual countries will differ from this picture.

7

Table 2. Characteristics of civil service systems and their reforms Career based Position based Recruitment Recruitment based on scholastic

background, by examination Open procedures, applications by candidates from all backgrounds and sectors

Management of senior civil services

Centralized management, pre-structured career paths

Decentralized management, some-times on departmental basis

Aim Coherent civil service; one culture, easy communication and high internal mobility

Wide choice of candidates, promoting competition, innovation and adapta-bility

Appointment Appointment to civil service is permanent, contracts are related to positions

Contracts specify individual appoint-ments for fixed terms, sometimes linked to organizational objectives

Disadvantages • Lack of competition for top positions;

• Bias towards generalist skills; • Alienation from society and

work floor

• Need for intricate appointment procedures;

• Lack of common culture among top executives;

• Weak mobility across organizations;

• In case of departmental basis; small pool of good candidates

Aims of reforms • Opening up recruitment process, increase competition for top executive positions;

• Introduce cultural change to stimulate adaptability;

• Seeking to introduce manage-ment by objectives

• Establish competitive and trans-parent appointment procedures based on merit;

• Re-centralization of management of top executives, e.g. keeping database of potential executives and central training programmes to establish a common culture;

• Review of promotion and mobility system to encourage mobility across organization

Countries (size of senior civil service)

France (5360) Italy (4800) Spain (276)

Belgium (450) Finland (200) The Netherlands (739) United Kingdom (3500)

Based on: PUMA, 2003

In countries with a career based senior civil service top positions are held by members of an

elite, who share a common background and socialisation. They change jobs relatively frequently,

but never leave the civil service altogether. The main objective of reform in countries with this

8

system is to make top executives more responsive and accountable for their performance. The

introduction of competition and performance related pay is expected to help achieve those goals.

In countries with a position based senior civil service, vacant senior positions are open to

a wide range of candidates, from the civil service and the market. Personnel management is

usually decentralized i.e. left to the discretion of departments. As a result, it can be rather

fragmented, preventing mobility across organizations; top officials with specialist skills never

leave their policy sector. Reform in these countries tends to focus on more attention for generalist

skills and a common culture, to establish a larger pool of qualified ‘leaders’ for the public sector

as a whole; see for example the Algemene Bestuursdienst in The Netherlands, which we will

discuss later on.

To some degree the reforms have led to a convergence of civil service systems. For

example, in all the countries in table 2 the recruitment process has become more competitive and

transparent (PUMA, 2003:12). However, most reforms are still only planned or being

implemented as we write this paper. “[P]ersonnel changes seldom, if ever, came first on the

reform agenda. It was much more common for them to follow – sometimes at a considerable

distance – innovations in financial management, organizational structures and management

techniques” (Pollitt & Bouckaert, 2000:72).

Countries such as Belgium, Finland, The Netherlands, Spain and the United Kingdom

have made reform of the senior civil service part of the general public sector reform while other

countries like France and Italy have undertaken specific reform programmes targeted at the senior

civil service (PUMA, 2003:6). Despite their differences, however, all these countries have con-

centrated on the implementation of performance-oriented management. This has led to a

decentralisation of managerial decision-making and a stronger emphasis on the performance and

responsibility of individual top managers. Appointments on (temporary) contracts have become

more common. Often these contracts are related to performance requirements; public managers

are expected to meet their targets. In some cases, salary and job continuation may depend upon

performance. Because of such changes, leadership, change management and human resource

management are or have become important issues in public sector management (Farnham et al.,

2005). The extent to which these changes have been implemented fully is dependent upon

9

country-specific characteristics, like the legal system, the traditional civil service system and the

spread of public sector reform in general (cf. Page & Wright, 1999).1

The new category: top executives

In the wake of public sector reform many public tasks were delegated to organisations at arms’

length of the government. Several reasons were given for such decisions (see Van Thiel, 2001:9-

12 on the Dutch case; Pollitt & Bouckaert, 2000:chapter 7). We will mention three of those

reasons here that are expected to have a direct impact on the position of the top executives who

manage these executive bodies.

First, in some cases it was considered desirable to separate administration from policy.

Political interference with individual decisions was not wanted in the case of for example

payment of benefits or regarding regulatory tasks. Independence would ensure impartiality, also

in the case of tribunals.

A second motive for establishing executive agencies was the expectation that because of

their managerial autonomy these organisations would operate much more efficiently than

traditional government organisations. For example, many of these new bodies are allowed to use

an accrual accounting system, which enables them to retain savings (cf. Pollitt et al., 2004).

A third argument refers to the expectation that the organisations at arms’ length will be

more customer-oriented. Not only will such bodies be closer to citizens and therefore more

responsive to customer needs, but it was also argued that like in a market customers could exert

more direct influence to voice their dissatisfaction in case of poor performance (cf. LeGrand &

Bartlett, 1993). A nice example of this line of reasoning can be found in the establishment of the

so-called ‘centres of responsibility’ in France as part of the Rocard-project, which was launched

in 1989. Public managers were made more aware of the relationship between objectives and

running costs through the use of contracts. These contracts gave them greater management

flexibility in operational matters, in exchange for a commitment to set up management control

systems and achieve certain agreed-upon objectives. Through the use of contracts, managers

received more autonomy but also more responsibility. Nowadays, there are 30 of these centres in

France (Allix & Van Thiel, 2005).

1 In turn, the existing legal arrangements and the civil service force have influenced the trajectory of public sector reform (cf. Pollitt Bouckaert, 2000).

10

It seems likely that the aforementioned expectations bear consequences on the job profile

and position top executives of quasi-autonomous agencies. First, top executives will have to be

managers first and foremost, and be involved in policy-making and politics much less than public

managers in ministries. This calls for extensive management training and expertise, preferably in

the private sector because that is where most of the newly adopted business techniques (accrual

accounting, financial management control, performance indicators, human resource management)

originate. Second, top executives are held accountable for their performance not only by the

principal (i.e. minister), but also by other stakeholders like clients. These stakeholders can be

represented on for example boards. Boards would therefore be expected to play an important role

in the functioning of top executives (cf. Cornforth, 2001). Finally, we would expect that the

position of top executives would be more comparable to private sector positions than public

managers’ position. That would fit with the idea that quasi-autonomous organisations operate

more like a (private) business. This would imply no permanent tenure or other advantages of the

civil service legal position. But it would also give the top executives more flexibility when it

comes to determining for example salaries or bonuses (i.e. higher salary).

Expected differences between public managers and top executives

While the two categories of ‘new public managers’ share a number of characteristics, such as

their managerial autonomy and focus on performance targets, we would also expect some

differences. For example, policy-making and politics are expected to play a much less dominant

role in the work of top executives than public managers. Top executives will be primarily

managing, almost like in a commercial business. Moreover, they operate at a greater distance

from the government, which enables them greater managerial flexibility. This should be reflected

in their labour position (tenure, salary, etc.). Finally, this could also be reflected in the degree of

politicisation of top executives for instance their membership of political parties. We have

summed our expectations up below:

1. Top executives will on average have had more working experience in the private sector than

public managers (career).

2. The educational background of top executives will on average be more directed towards

managerial studies than in the case of public managers (training background).

3. Top executives will on average spent more time on managerial issues than public managers.

11

4. On average, top executives will have a legal position less comparable to traditional civil

servants than public managers (e.g. permanent tenure, limited salary increments).

5. Top executives will on average receive performance related pay more often than public

managers.

6. Top executives will on average be less often member of a political party than public

managers (politicisation).

The following sections will investigate these expectations consecutively.

Profile of new public managers

Demographics of public managers

In most western countries, the average senior manager in the public sector is a man (see table 3),

over 40 (table 4). The French case looks quite different from the other countries, but that is

because teachers in primary and secondary education are included in the employee count at

central level. The Netherlands is one of the countries where the average age of public sector

employees is relatively high compared to other countries, even after correcting for the baby-boom

effect (PUMA, 2002). This will mean that a renewal of the top level will be immanent in the

coming years. The same will be true for Scandinavian countries like Norway and Sweden.

Table 3. Percentage female employees in central administration at different responsibility levels and overall, 2001

Administrative staff Middle managers Senior managers TotalFinland* 49.0 27.0 23.0 47.8France* 48.2 46.5 51.9 49.0Ireland 73.8 47.4 15.6 53.5Japan 29.3 12.3 1.3 n.a.Netherlands* 41.8 25.5 18.6 38.9Norway* n.a. n.a. n.a. 43.0Sweden 80.0 48.0 21.0 47.7Switzerland 28.2 17.7 6.9 25.4*2000 n.a. not available Source: PUMA (2002)

12

Table 4. Employment ratio by age group in central administrations in 2001 (%) <19 20-29 30-39 40-49 50-59 >60Finland 0.1 14.0 25.7 31.0 25.8 3.5France 0.1 16.7 26.7 32.3 22.7 1.5Ireland 1.3 15.0 27.4 37.1 16.1 3.3Japan 0.2 15.5 25.6 32.4 26.2 0.2Netherlands 0.7 15.0 23.0 34.0 25.0 2.3Norway* 0.0 13.3 27.0 27.5 25.3 6.8Portugal n.a. 14.6 31.5 32.9 16.1 4.9Sweden 0.1 9.7 23.2 26.2 33.7 7.2Switzerland 0.1 11.2 26.7 29.0 28.1 4.9Turkey 10.2 24.8 46.3 16.5 1.8 0.4*2000 n.a. not available Source: PUMA (2002)

In the Netherlands, a special agency (the Algemene Bestuursdienst, ABD) deals with the position

of public managers. According to its annual report of 2003, 899 public managers belong to this

category of civil servants. Table 5 shows some information about these public managers.

Table 5. Demographics of Dutch public managers Characteristic Number% older than 51 yrs 59%Mean age* 51,5 yrs% women* 12%% newly appointed women 16%% ‘outside’ appointments 16%Source: ABD (2003) * 2002

As can be seen in table 5, the average Dutch public manager is a man and relatively old.2 The

number of public managers with an ethnic minority status is negliable. This last number is also

stable and not rising (ABD, 2003). The percentage of women, however, is slightly increasing.

From another survey (by the Home Office, see below) we know that 96% of senior managers

have a high education degree. At central level most public managers have a degree in social

sciences, law or economics (cf. also Nelissen et al., 1996).

In 2003 the percentage of outside appointments, i.e. how many public managers are

appointed from outside the public service, was 16%, which was considerably higher compared to 2 There are hardly any (statistically) significant differences between civil servants at central and decentral level with respect to characteristics such as age, gender, educational background, political affiliation, prior work experience and job mobility (Nelissen, Bressers & Engelen, 1996). Survey was held in 1995, N=226 at central level.

13

2002 (9%), and also much higher than a decade or so earlier (when almost all appointments were

internal). In fact, appointing new public managers from outside the service is an explicit goal of

the ABD (with 15% as the target that the agency is striving for). Not all these outside appoint-

ments, however, concerns people with a private sector background, as the number also includes

newly appointed employees with an earlier career within the local government or within non-

profit organizations. In sum, although the percentage of Dutch public managers with a private

sector background is increasing, it is still a small group.

Data from other countries show similar demographic profiles of public managers. For

example, in Italy the top public managers are men (only 5% women in highest ranks) over 56

years old (in 71% of the cases; Cassese, 1999). Almost all of them (95%) have a university

degree, predominantly in law or political sciences. Most of highest-ranking public managers, the

directors general, (93%) come from the centre and southern regions. The northern part of Italy is

therefore seriously underrepresented in the top of central administration (Cassese, 1999:56).

Apart from the geographical origin, there is however little homogeneity among public managers.

They rise through the ranks (vertical mobility only) during their career, and stay within the public

domain all their working life (cf. career based system). There is a strong focus on legal and

formal aspects of public services. Highly political aspects of the job are not popular among public

managers, neither are pure managerial tasks.

In the German case, the number of women in the administrative elite at federal and Land

level is slightly higher than found so far, but they still form a small minority: 8.6% (Derlien,

2003).3 Almost all senior managers (96.9%) have a university degree, with half of them even

having a doctorate (51.9%). The legal sciences are predominant in the educational background

(52.7%), followed by economics (19.8%). Former East German regions are seriously

underrepresented (0%). More and more senior managers have had experience in other parts of the

labour market next to their pubic service career (32.2%). This applies however mainly to the

younger generations; 41.1% of the current higher civil service still has an uninterrupted career in

public service. Only 4.8% originates from outside the public service. After early retirement (in

their late fifties) it is however not uncommon for senior managers to take up a position in the

private sector (banks, companies).

3 N=139, survey held in 1999.

14

All in all, there is a rather basic stability over time (1970-1994) in German bureaucrats’

role understanding, recruitment and values (Derlien, 2003:402). This is probably due to the fact

that until 1994 NPM reforms occurred mainly at the local level, not the federal (where 14

ministries employed 18,000 public servants in 1999, making up for 13% of the total public

service).

Demographics of top executives

The demographics of top executives in quasi-autonomous agencies do not differ very much from

that of public officials in general. Table 6 shows the results of surveys we have held in The

Netherlands, among public bodies (ZBOs) and contract agencies (Van Thiel et al., 2004), and in

French public establishments (EPs), independent administrative authorities (AAIs) and centres of

national competence (SCNs; Allix, 2005).4 For Denmark we have used two studies (Kettl, Pollitt

and Svara, 2004; Christensen, 2005).

Most directors of quasi-autonomous agencies are men, in their early fifties and holding a

university degree, in the Dutch case usually in law or economics. In France, a diploma from a

‘grande ecole’ is most common, in particular from the National School of Administration (ENA),

often alongside a university degree. Many EP-directors (48.7%) have a degree in engineering,

which fits with the fact that they work in a technology or infrastructure sector (cf. also Page &

Wright, 1999). In the case of AAIs, the average age is higher because a part-time working

‘president’ who has retired from civil service sometimes holds the top position. The slightly

higher number of women in the SCNs could be due to the fact that they are involved in the

cultural domain, which is traditionally more populated with women in higher functions compared

to other sectors.

4 For more information about these types of organization we refer to Greve et al. (1999), Van Thiel (2001) and Allix & Van Thiel (2005).

15

Table 6. Profile of top executives in Dutch, French and Danish quasi-autonomous agencies, 2004

N % female

% University degree

Average age in years

Prior public service

Yrs in current position

Danish agencies

n.a. 15.9 93.8 54.7 15.9 years 4.4

Dutch agencies

19 18.2 84.2 51.1 Yes: 89.9% n.a.

Dutch ZBOs

67 14.3 83.3 51.9 Yes: 76.4% n.a.

French EPs

40 12.5 97.5* 53.0 5.2 years 3.5

French AAIs

14 13.3 100.0* 62.0 2-4 years (varied)

French SCNs

13 30.0 92.2* 54.0 Yes: 54.0% 6.2

* including a degree from a ‘grande ecole’ n.a. not available Source: Kettl et al. (2004), Van Thiel et al. (2004), Allix (2005)

Top executives in all three countries have extensive experience in the public sector, either in the

parent department, in the same executive organization or another one, or elsewhere in the public

sector. For example, in Denmark most agency directors have worked in the parent department

before being appointed (71.1% of appointments since 1935). Outside appointments are more

likely to be from other parts of the public or even private sector (19%) than from other ministries

(5.8%; Christensen, 2005).

For France, the average number of years spent in previous and the current job is relatively

short. This is caused by the general use of consecutive temporary appointments in public

functions (see below). When we look in more detail we find, however, that most respondents

have had several jobs in the public domain prior to their current one. Particularly in universities

and cultural agencies respondents appear to have made such careers. There appears to be a strong

interdependency between administrative, political and policy-making roles in France (cf. Bezes,

2001). There is homogeneous elite with a strong corporate identity and an interest in self-

perpetuation and self-preservation. This homogeneity is reinforced by the well-known monopoly

of the grandes ecoles, for recruitment opportunities into the most powerful posts. The elite

colonizes all administrative and top political positions within ministries, but also has the ability to

move easily from one sphere to another (administration, party politics and corporate positions)

thanks to strong networks. Apparently, most top executives of French quasi-autonomous bodies

16

are also part of the that elite. Only 7.5% of all French respondents came from the private sector.

In the case of AAIs there is a much higher degree of experience outside of the public sector

(41.6%) though than in the other types of agency. Perhaps this can be explained by the fact that

AAI-executives are usually not appointed by the parent ministry (see also below). The low

number of respondents makes it difficult to generalize such ideas, but the Dutch data could be

considered supportive. The total number of Dutch respondents who moved from the private

sector to a quasi-autonomous agency is 20.9%, ranging between 11.1% for contract agencies and

23.9% for ZBOs; the number is the highest for the type of organization that is the farthest

removed from the parent ministry.

Comparison of demographic profiles

Although the differences are very small, there seems to be some evidence to support our expecta-

tion that agency executives originate more often from the private sector than public managers in

ministries, or at least that they come from outside the parent department more often. Technically,

our expectation is confirmed. However, the majority of top executives still come from the parent

departments, as do senior public managers.

There are more similarities than difference between the two types of new public

managers: they share the same gender (male), age (over 50), training background (higher

education) and career pattern (within public sector). Law and economics are preferred studies for

both types of managers; in France this homogeneity is even stronger due to the important role for

ENA in the training of public servants. Our second expectation is therefore rejected.

In sum, these findings would seem to indicate that even in countries with a position-based

system, there is an administrative elite in charge of ministries and executive agencies.

Job characteristics of public managers

Unfortunately, we cannot fully investigate our third expectation because we have not (yet) been

able to find information on the functioning of top executives. This section will therefore only deal

with information on the way in which public managers function, both in their own perception and

that of subordinates. Of specific interest is our assumption as explained in the introduction that

the job conditions of senior managers have changed under the influence of NPM. Therefore, we

17

will pay attention to the amount of time spent on management issues and/or the managerial skills

of public managers.

Each year, the Dutch Home Office holds a large-scale survey on issues related to the job, desired

mobility and work orientations of employees working in the public sector (cf. Steijn, 2004). Here

we have taken information on the job characteristics of managers. In the 2002 survey 28,312

respondents were interviewed of whom 6,330 were working in sectors relevant for this paper.5 Of

these, 75 can be characterized as high-ranking senior managers.6 For comparison, data about

middle managers and ‘ordinary’ employees is included in table 7 as well.7

Table 7 Job characteristics of senior management, middle management and ordinary employees in selected public sectors in the Netherlands (2004)

Senior management

Middle management

Ordinary employees

Number of working hours (factual) 49 40 35Mean number of subordinates 117 24 3Number of years with current organisation 19 15 14Autonomy in job (5=highest) 4,14 3,94 3,84Level of work-pressure (5=highest) 3,42 3,17 2,92

The seniority of Dutch top officials is quite high. Even more importantly, it becomes clear that

they have a loaded working week (they work more hours than stipulated in their labour contract).

This can probably be partly explained by their level of supervision: on average they have to

supervise more than 100 employees. Not surprisingly, their level of work pressure is substantially

higher compared to other employees (including middle managers). At the same time, their job

autonomy is somewhat (but not that very much!) higher compared to other employees.

Dutch senior managers are more satisfied with their job compared to other workers and

also less often looking for another job (table 8, cf. Nelissen et al., 1996:97-100). If they do,

however, they are more interested in a job in another organisation; a third even preferred a job in

the private sector (overall, this was a preference of only 19% of the employees). In other words,

5 These include the civil service, the police, the Polder Boards, research institutes and academic hospitals. 6 We defined this as employees within the highest three salary scales (16-18) who at the same time stated they had supervision on other employees. 7 Middle managers are defined as employees on salary scales 11-15. Also employees on scales 16-18 without supervision on other employees were defined as middle managers. Ordinary employees are all other employees.

18

if these managers are looking for another job, they are less attached to the organisation and sector

they are currently working in. This is also reflected in their commitment profile. Using Allen &

Meyer’s (1990) famous’ distinction of three different types of commitment, it appears this senior

managers – compared to other employees – have a higher affective and continuance commitment,

but a lower normative commitment. In other words: they feel more ‘attached’ to the organisation

(affective commitment), but also see more options to work somewhere else (continuance

commitment) and put a lower premium on loyalty in itself (normative commitment).

Table 8. Work orientations of senior management, middle management and ordinary employees in selected public sectors in the Netherlands (2004)

Senior management

Middle management

Ordinary employees

Looking for another job? (% yes) 24% 36% 34%Job satisfaction (5=high) 4.24 3.95 3.75Affective commitment (5=high) 3.94 3.66 3.37Continuance commitment (5=high) 3.60 3.32 3.38Normative commitment (5=high) 2.88 3.27 3.94

So, senior managers work hard and are committed to their job. But what do we know about their

performance? A survey in 1995 (Nelissen et al., 1996:93-97) identified personality traits of civil

servants in the Netherlands. At central level most of them see themselves as a statesman (60%,

N=226) rather than an advocate or something else (typology taken from Downs). They are

dedicated to the programmes that they have to carry out (42%). Hardly any one would call

himself a generalist manager (7%). And when asked, most of them (56%) sees his own leadership

more as a participative relationship than a hierarchical one.

A study into top managers of the Danish public sector corroborates this self-image. It

shows that the expertise of top managers is mainly in policy advice and much less on managing

their organizations to reach maximal performance (Kettl et al., 2004). When asked, they rate their

own knowledge of the specific policy domain in which they operate as 74 (on a scale of 100%)

and their abilities in using (new) management techniques at 63 (Kettl et al., 2004: 34). An

evaluation of top managers’ involvement by their subordinates corroborates this opinion (see

table 8). Danish top managers are good at offering advice, maintaining the border between

politicians and administration, and fostering orderly organizational performance. They are,

19

however, less good at (visible) public leadership within the organization and towards

subordinates, and using new management techniques like ICT (Kettl et al., 2004:38).

Table 8. Evaluation of top managers involvement by subordinates (100= very high) N=? Preferred rating Actual ratingFosters trust 94 76Sets good example 94 71Visibility in organization 92 74Team leadership 91 73Focuses on efficiency and results 91 83Source: adapted from Kettl et al. (2004:37)

In the late 1980s (1987), the German administrative elite saw itself mainly as a problem solver

(94.5%), an initiator of new projects (95.2%) and an implementer of policies (94.5%). The roles

of facilitator (3.4%), ombudsman (34.2%) and advocate (53.1%) were least popular. This “largely

emphasizes the classical ideal of the neutral expert civil servant who competently translates

political goals into action and impartially serves the commonweal […] As NPM elements, such

as the policy/operations divide between ministries and agencies, are not applicable to a highly

decentralized system like the German federal polity, […] the “intelligence of bureaucracy” would

probably resist such reforms. [U]nder certain conditions, such as during unification after 1990

and since about 1995, mandarins can turn quite managerial” (Derlien, 2003:420-423). According

to Derlien, the increase in political party membership in the 1990s (see below) can be seen as an

indicator of the increased awareness of the need for economy, efficiency and effectiveness

(NPM) as it will help role understanding and streamlining ministries after reforms. There is

however also a risk of party patronage that could damage the high level of public trust in German

bureaucracy. Therefore ethics should become an important topic in personnel management.

Conclusion

A lack of data on the behaviour of directors of executive agencies makes it impossible to draw

conclusions on our third expectation. However, the data on public managers does not appear to

corroborate our basic assumption about the shift in the work of public managers towards more

attention for managerial matters. The self-reports of public managers still value policy advice

skills as most important (cf. also Noordegraaf, 2001 for the Dutch case). There is even a self-

20

acknowledged disinterest in managerial skills. This would suggest that NPM has not led to real

changes in the work profile of senior public managers. The earlier findings on the background

and history of public managers indicated as much already.

However, there are two noteworthy observations. First, the German experiences show that

although public managers do not often originate from the public sector, they often start working

in the private sector after their public career. This effect is reverse to what we expected as a result

of NPM influences. According to Derlien (2003) this trend is facilitated by the fact that more and

more public managers have an economic rather than a legal background. The commitment scores

of Dutch public managers also pointed in this direction; their search for the next job extends

beyond the parent department. The second interesting observation is presented in the Danish case.

Christensen (2005) argues that as a result of the increased need for transparency and accounta-

bility, key elements of NPM, senior public managers’ careers are more often cut short than

before. Both examples show effects of NPM on the public managers’ work patterns that we had

not expected above.

Appointment and legal position

Public managers

The Dutch system for recruitment and appointment of senior manager has always resembled a

position-based system the most. Some important reforms, which were introduced in 1995, made

this even more apparent. Until then, most public managers made their career through a well-

defined path, usually within the same department. This often led to very long careers within

specific segments of the civil service. The 1995 reforms intended to end this compartmenta-

lization. The main elements of reform were:

• All public managers8 now belong to a central agency, called ABD (see before);

• They are no longer appointed as civil servants of a specific department, but instead they are

all appointed under the authority of the Home Office;

• Although all public managers still have permanent tenure, the maximum number of years in a

single position within the civil service is limited to seven years. Public managers are expected

8 At first this applied to all civil servants from salary scale 17 onwards (with 19 being the highest scale). Since 2000 it also applies to those in scale 16 and since 2002 also those in salary scale 15.

21

to have found another position (within the civil service) before the end of that period. This

change has led to a competition between senior managers for jobs in the civil service. As a

result the medium length of stay within a senior management position in 2002 was only 3.2

years (ABD, 2003); and

• Competition for senior management positions within the civil service is open, which in effect

means that people from outside the civil service can also apply for them – which was

impossible in the past. For 2003 this has meant that 16% of the managers appointed within a

senior position came from outside the civil service (cf. section on demographics).

These reforms have clearly changed the position of public managers within the civil service.

Most importantly, senior managerial positions have become more accessible and can be obtained

more easily by people from outside. The reforms also facilitate the development of an ‘esprit de

corps’ of senior managers within the civil service (‘t Hart et al., 2002), which is expected to

increase the emphasis on the managerial aspects of a senior management job and reduce the need

for material knowledge of a certain field. The latter result was a major objective of the reforms.

A final interesting point concerns the effect of the Dutch reforms on the relation between

the civil service and politics. Traditionally, the Netherlands has a system based on the Weberian

notion of an independent bureaucracy; civil servants are supposed to be impartial, objective and

free of political ties. Political cabinets around a minister, like in the UK and Belgium, do not exist

in the Netherlands. According to some, the ABD was seen by politicians as an instrument to

discipline civil servants (‘t Hart, et al., 2002: 302). It is however unclear whether or not this goal

has been reached (see also below on political affiliations of public managers).

Like in the Netherlands, political appointments of senior public managers are frowned

upon in Denmark (Christensen, 2005). In the past fifteen years several incidents of appointing

‘political advisors’ have been discussed in parliament. However, since ministers from all parties

have done so after coming into office, parliament has up to now “never resorted to legislative

regulation but instead preferred to define its limits through a combination of parliamentary

resolutions and operational measures prompted by the enacting coalition, but developed by the

incumbent government and the senior civil service” (Christensen, 2005). Appointments are

therefore still based on merit and from the own ranks (for permanent secretaries: 55.6% of all

appointments since 1935). Despite a slightly growing tendency to appoint senior managers who

have worked in other ministries than their own (30%), minister seldom recruit from private

22

businesses or local governments (0-14.3%). On the other hand, the average length that senior

managers stay in the same function has decreased (from maximum 18 to maximum 9 years).

Christensen does not blame this on the political affiliation of senior managers, but on the fact that

high civil servants are no longer safe from public affairs and media attention; in case of incidents

not only ministers are held accountable but also the top of their departments.

In other countries political appointments of senior managers are common. Take for example the

recent changes in Italy. The Italian senior civil service has always had the characteristics of a

career-based system; promotions were linked to the length in office. However, major changes

were implemented in 1998 and 2002 (Cassese, 2002). A new government can now remove any of

the 55 top public servants when they come into office. Moreover, all appointments are temporary,

for a period between three to five years depending on the type of function. And finally, it is

specified that about 8% of the highest ranks (‘Dirigenza’) has to originate from outside the public

service. The main reason for these changes was to reduce the political patronage from the old

parties – to the benefit of new parties. And despite the high costs of these changes (like paying

for the leave of senior managers who are not re-appointed), it has also led to an increase in the

autonomy and influence of senior managers, which they did not have very much of before

(Cassese, 1999).

The mix of a career based system with political appointments can also be found in France.

The French senior civil service can be seen as a government-wide elite, with ties of common

interests, if not solidarity, among officials from different ministries and organizations. Contrary

to the Netherlands and Italy, senior managers are not bound to one policy sector during their

career, making for stronger ties between ministries (Page & Wright, 1999:8). The influence of

elected politicians on the appointment of senior managers is high, whereby in general preference

is given to members of the so-called grands corps. The interwoven nature of politics and

administration has less to do with political preferences but more with the common (training)

background of individuals. Senior public managers are not neutral though because they

embedded in a complex network of loyalty and solidarity within the bureaucratic system at

different levels. Through these powers senior managers can exert strong influences on, or

organize resistance against, the development and implementation of policies, including NPM

reforms (Page & Wright, 1999).

23

Top executives

Most top executives of quasi-autonomous agencies in the Netherlands are referred to as (general)

director (see table 9), although sometimes their function is labelled as ’president’, ‘secretary’ or

‘chairman’. In two types of French organizations, we have found a diarchic executive. In AAIs, a

general secretary often supports the director or president. In 40% of the French public establish-

ments both a president and a director are in charge of the organization. There is no detailed

information about the division of tasks and responsibilities between these two executives, but it

would seem that the (executive) role of the president is rather limited. Most often, the president is

the (unpaid) chair of the executive team, consisting of several directors and one director general.

In about half of the cases (s)he co-signs the contract between agency and ministry, together with

the director.

Appointment procedures vary between types of agency. The parent minister always

appoints top executives of Dutch contract agencies and French SCN. However, for bodies at

greater arms’ length the appointment of top executives becomes less and less the prerogative of

the parent minister. In the case of French public establishments 77.5% of the top executives is

appointed by the parent minister (and confirmed in the council of ministers). In other cases, the

prime minister or the President of the Republic is the leading authority, as in the case of AAIs

(52.6%). Still, in total 90% of EP-appointments is still in the hands of the ‘government’. For

Dutch ZBOs, appointments are the prerogative of the boards (54%). The minister appoints only

one out of three ZBO-directors.

24

Table 9. Appointment of top executives in Dutch and French quasi-autonomous agencies, 2004

Dutch agencies

(N=19)

Dutch ZBOs

(N=67)

French EPs

(N=40)

French AAIs

(N=14)

French SCNs

(N=13)Function described as director

95% 70% 80%* 78.5% 84.6%

Appointment by parent minister

84% 35% 77.5% 13.3% 92.0%

Appointment by board n.a. 54% 7.5% n.a. -Permanent tenure, cf. civil servant

89.5% 32.8% - - 100%

Permanent tenure, other labour agreement

n.a. 49.3% - - -

Temporary tenure, based on a contract

- - 57.5% (3-5 yrs)

93.3% (3-5 yrs)

-

*also includes director general n.a. not available Source: Van Thiel et al. (2004), Allix (2005)

As employees of Dutch agencies remain civil servants, they benefit from the accompanying legal

rights such as permanent tenure. For almost half of the ZBO-directors the situation is different;

their tenure, salary and other benefits are laid down in collective labour agreements from the

private or semi-public sector. Nevertheless, most of them still have permanent tenure (82.1%).

The contracts of these top executives are not included in the contract between parent department

and executive agency, which contains the output and budget agreements. Nor is the performance

of the organization decisive for the tenure and rewarding of the director.

The French situation is very different, at least for AAI and EP-directors. They are

appointed on a temporary basis, between three to five years. Nevertheless, they still consider

themselves to be civil servants (EP:90%, AAI:64%). Those who are appointed on a contractual

basis, have contracts stipulating objectives that have to be achieved. In the case of public

establishments, the directors’ appointment is to some degree dependent on achieving the

objectives for 60% of the contracts. However, since not all of these contracts entail actual

performance measures, it is not entirely certain how parent ministers evaluate the directors’

performance. A number of respondents in the survey pointed out that the political aspects of their

work and performance are also very important to the decision about the continuation of their

appointment, in particular their relation (and trust) with the parent minister or ministry.

25

Comparison

We expected top executives to have a less traditional legal position than public managers. For the

Dutch ZBOs this seems to be true; directors are appointed by boards rather than the minister.

Moreover, in half of the cases their legal position is different from that of a civil servant. Also,

the uniform discipline on Dutch public managers that is exerted by the ABD is absent for ZBOs

(but not for agency executives). For other types of public body and in other countries we have not

found the expected differences though. So, there is only partial corroboration of our fourth

expectation.

Interestingly, we have seen that in countries with a career-based system, such as Italy and

France, temporary appointments have become the norm both for public managers and top

executives – but not (yet) in the position based system countries. In some cases (see below) these

appointments are even linked to performance contracts and salaries. This trend fits with the NPM

reforms of personnel management and is apparently applied to all new public managers in these

countries.

Payment

Public managers

Compared to other countries, Dutch public managers receive a high salary (PUMA, 2002).

Between 2001 and 2002 the salaries of Dutch senior managers in the public sector have increased

by almost 12% (cf. table 10). The average income of the 4,589 top officials in a survey by the

Home Office is 90,400 euros. The strongest increase was found in the case of top managers in

municipalities, however their salaries are still low compared to the salaries for hospital directors

and chairmen of universities.

26

Table 10. Top managers’ salaries in The Netherlands, 2002 Number of top

managers Average in-

come in euros% Executives with

salary >120,000 Income increase

2001-2002Ministries 124 115,400 44 +7.8%Agency 92 94,400 12 +6.2%Municipality 1,962 77,100 1 +16.2%Police 79 87,600 5 +9.9%University 70 125,200 54 +9.7%Hospital 194 140,400 61 +10.1%Public ZBOs 142 100,400 21 +8.5%Private ZBOs 24 106,600 17 +12.3%Public sector 4,589 90,400 13 +11.9%Source: Ministerie van BZK, 2004

Compared to earlier years (1999-2000: +5.3%) and compared to the private sector (2001-2002:

+8.5%) the overall increase in salaries is extraordinarily high. The Dutch Home Office attributes

the increase to the difficulties that governments have to find skilled people and on a number of

amalgamations at decentralized level (Ministerie van BZK, 2004). However, the finding that on

average 13% of the top executives in quasi-autonomous agencies earn higher wages than a

minister (who earns 120,000 per year) has triggered a lot of debate (see also below). A committee

has been asked to give advice (see below).

Performance related pay for public managers

In 1981 the Dutch cabinet decided that senior public managers were not allowed to earn a higher

salary than their minister. Hence, the formal part of salaries (‘scale’) is fixed. However, due to

bonuses and surcharges public managers can in practice receive a higher salary.

Since the early 1990s some elements of performance related pay have been introduced in

the Netherlands. In essence this means public managers can rise faster through their salary scales

(up till the highest scale, 19) or can get extra bonuses if they perform ‘outstanding’. It is however

far more difficult to ‘punish’ public managers if they perform poorly.

The extent to which performance related pay for Dutch public managers is implemented is

unclear. The variable part of the salary seems to be dependent on the job: it is higher for more

important jobs. Data from 2002 show that ‘top’ public managers received on average 15,5% of

their salary as variable bonuses (Committee Dijkstal, 2004:6). Recently, however, the Committee

27

Dijkstal proposed a new salary structure for public managers and top executives: 90% of their

salary should be fixed and 10% should be flexible. The flexible part should be dependent on the

fulfilment of performance agreements (contracts) by public manages and top executives. It is

expected that the variable part of their salary will decline compared to the present situation.

Similar developments can be found in other countries. For example, in 2003 the French

government expressed its dissatisfaction with the systems for appointment and payment of top

public managers (Siciliani, 2004:5). A scheme of performance related pay was suggested, but to

this day it has not yet been put into practice. At the moment the salary of a director in the central

administration is on average 115.000 euros, which is considered low compared to other

countries.9 There is however is a bonus system in place for ‘additional work’ with a high

responsibility. In practice, all directors (except at the MINEFI) receive the maximal bonus

annually (Siciliani, 2004:7). It is argued that this is in some way a compensation for the rather

low level of pay. Rouban (1994:97) has characterized the bonus system as a ‘bewildering world

where secrecy is the rule’. In 1981, the socialist Prime Minister Pierre Mauroy decided to

commission a large study in order to bring some light into this dark jungle. The final report was

never published nor given to any government officials. It was said that some findings were so

disturbing that there were only three copies, kept secret for the exclusive use of the Prime

Minister.

In fact, there is no apparent rationality or equity in the French bonus system. There are

between 200 and 300 kinds of different bonuses (of these, 130 are for the military alone). High

civil servants can greatly improve their rewards when employed in ministerial offices where there

is a specific cabinet bonus or when they get a position in better off ministries. Some of these

bonuses were institutionalised at the turn of the century and only survived until now because they

were part of a long tradition. As a whole it was estimated that the inclusion of bonuses increases

the standard deviation of total rewards up to 17% (INSEE 1981).

In recent years the level of bonuses has risen more than basic salaries. Between 1988 and

1990 the average level climbed from 38%. Today, bonuses are worth 11.6% of the total civil

servants’ rewards. Inequalities are considerable. The higher the salary is, the higher the bonus is

(as in the Dutch case). The proportion of bonus to total level of reward varies according to the

9 Most information on salaries is not indexed and can refer to different years. What is high in one country in one year, can be low from other countries’ perspectives in the same or other years.

28

ministry. For instance bonuses are worth 25% of the total reward in the finance ministry and 5%

in the ministry of education.

In line with its policy of administrative modernization, the French government has made

some attempt, in recent years, to institute a pay for performance system within the civil service.

At present these reforms are at an experimental stage; schemes have particularly been set up

within the ministry of infrastructure, for directors of services located in regions outside Paris. As

yet, individual performance appraisal determines only a minor part of the overall yearly bonus.

More progress has been achieved however with the introduction of performance related pay for

top executives in the industrial and commercial public establishments (Siciliani, 2004:15).

So far, performance related pay has been applied only in Italy (20% of pay), Finland and

the United Kingdom (variable bonuses; PUMA, 2003). In countries like Norway, performance

pay related instruments have been implemented but their actual use is still incidental and always

with great caution (Laegreid, 2003:163).

Performance related pay for top executives

There are few differences between performance related pay developments for public managers

and top executives. Based on our information, we can conclude that similar schemes have been

implemented for both categories of new public managers in Italy and The Netherlands. In France,

more progress has been made with performance related pay for top executives than for public

managers, but the literature is inconclusive on the actual difference (Siciliani, 2004).

As far as the absolute differences in salary go, table 10 shows that directors of Dutch

ZBOs earn on average a higher salary than the senior public managers in their parent

departments. However, because of the variable bonuses we cannot be entirely certain whether this

formal difference is also true in practice. The Committee Dijkstal (2005) has criticized the

difference in salaries strongly because they do not consider the job of top executives of ZBOs,

universities and hospitals to be more difficult than that of a public manager in a ministry. The

salary of directors of executive agencies is by the way lower than that of the average senior

public manager at central level.

29

Comparison

The information on salaries and the introduction of performance related pay is at this stage

inconclusive. Only for the Netherlands can clear evidence be found that top executives have a

higher salary on average than public managers (in line with our fourth expectation). However, the

complex systems of bonuses make it difficult to make an accurate comparison.

In most western countries performance related pay is discussed and plans for implementa-

tion have been drawn up. However, progress is slow and varied; in the Netherlands most plans

still focus only on public managers because they are more easily disciplined given their legal

status. On the other hand, the French government appears to make more progress with regulating

the pay of top executives in public establishments. Therefore we cannot draw any conclusions

regarding our fifth expectation.

Membership of political parties

Public managers

Membership of political parties, or at least a professed preference for parties, is quite common

among public managers. The highest-ranking Dutch public managers are secretaries-general.

Data on their preference (and assumed membership) for certain political parties show that the

majority (6 out of 13; 46%) favours the social democratic party, a third (4 out of 13) supports the

Christian democrats, leaving a few liberals (2 out of 13) and one democrat.10

Derlien’s (2003) study of German civil servants shows that membership of a political party

has increased strongly among senior managers; from 28% in 1970 to 59.6% in 1995. The

Christian democrats have profited most from that increase (four times as much) while the social

democrats lost support (by about a third). In 1995 (N=103), 40.4% was not a member of any

political party. The remaining 60% belonged to Christian democrats (40%), social democrats

(11%) and liberals (8.6%). There were no public managers in the survey who were a member of

the green party. According to Derlien, their political affiliation determines public managers’

opinions about their own tasks and those of a government in general (discussed above).

10 Taken from: http://planet.nl/planet/show/id=62967/contentid=447925/sc=5fa286

30

Top executives

Finally we turn our attention to activities of top executives outside of their job. We asked two

questions: one about belonging to a political party and one about voluntary work. Table 11

presents the results.

Table 11. Non-professional activities of top executives in Dutch and French quasi-autono-mous agencies, 2004

N Membership of political party Voluntary workDutch agencies 19 32.0% 52.0%Dutch ZBOs 67 25.0% 62.0%French EPs 40 15.0% 67.5%French AAIs 14 14.2% 46.2%French SCNs 13 0.0% 62.0%n.a. not available Source: Van Thiel et al. (2004), Allix (2005)

Political affiliations among top executives are much higher in The Netherlands than France.

Between one third (agencies) and a quarter (ZBOs) of the Dutch top executives is a registered

member of a political party, most often either the social democrat or the liberal party. The

number of French top executives who report not to belong to a political party is much higher. For

example, 70% of the EP-directors and all of the SCN-directors answered no. Those who are

member of a party are equally divided between the socialists and UMP.

The majority of top executives are involved in voluntary work of some nature, usually by

being a board member of an association or foundation in the same policy field, of school boards,

or other public organizations. Also, some respondents report their involvement in local politics in

this category. Based on these responses we can conclude that most directors choose to assume

executive or direction functions outside of their everyday work rather than perform more classical

types of voluntary work.

Comparison

Additional research is needed to complement the current data, but so far public managers appear

to be more politicised than top executives (party membership rates are about twice as high). That

would be in line with our last expectation.

31

Conclusion

The findings presented in this paper are work in progress; we are still looking for data to com-

plete our comparison between the two types of new public managers, from different countries.

However, a number of conclusions can already be drawn – although more research is sometimes

necessary to substantiate them.

First, today’s public managers are much less managers than NPM reforms aimed for.

Public managers still see themselves mainly as the classical neutral policy expert who advises

politicians on decisions and policies. Managerial skills are less valued, and subordinates do not

see their senior manager as a ‘leader’. There was even some hint of resistance among public

managers against NPM reforms (cf. also James, 2003). That could explain why the implementa-

tion of for example performance related pay is not yet achieved in most countries we looked at.

Where progress has been made with performance related pay, it would seem that career-based

countries such as Italy and France have an advantage because of the centralization and uniformity

of recruitment and payment. In position-based countries like the Netherlands, only public

managers have so far been subjected to performance related pay and temporary appointments.

A second important conclusion is that the differences between top executives and public

managers are not as big as we had expected. Both types of new public managers come from an

administrative elite: middle-aged man with a higher education, often in law or economics, who

make their career predominantly in the public domain. Nevertheless, there are some differences,

which seem to be stronger as the distance between the executive agency and the parent

department increased (e.g. in the case of French AAI and Dutch ZBO). These differences regard

their appointment (less often by a minister), their prior experience in the private sector (more

often the case), their legal position (no civil servant status, higher salary) and their smaller degree

of politicisation (i.e. membership of a political party). All in all, we believe that our distinction

between the two types of new public managers is useful and interesting.

The findings in this paper have raised some new questions for research as well. For

example, part of the NPM reforms is the reduction of indefinite appointments. As a result, we

have seen that the average length of time that a public manager works in the same position is

reduced. Christensen (2005) offers another explanation for the reduction of job length though. He

suggests that public managers have – perhaps as a result of NPM – become more vulnerable for

32

political incidents, which would suggest that, like top executives, public managers are held more

accountable for their performance than before. Based on the current evidence we cannot say

however which of the two explanations is more valid. More research is necessary.

Another interesting observation was that in Germany public managers often take on jobs

in the private domain after their public career – rather than the other way around, which would

appear to be more in line with the NPM-aim to hire people with prior management experience.

On the other hand, the fact that public managers do cross over the boundary between the public

and the private sector fits with the NPM ideology. Therefore, Derlien’s observation needs to be

investigated further. We only looked into the origin of public managers and top executives in this

paper, not where they went working after their current job.

Finally, the difference in salaries between public managers and top executives has lead to

a hot debate, at least in The Netherlands. It remains difficult however to draw hard conclusions

because the complex bonus systems obscure actual payments. More data are necessary.

33

References: Algemene Bestuursdienst (2003). Jaarverslag Bureau ABD 2002. Ministerie van BZK, The Hague. Algemene Bestuursdienst (2004). Kenmerken ABD-populatie 2003. Ministerie van BZK, The Hague. Allen, N.J. & J.P. Meyer (1990). The measurement and antecedents of affective, continuance and

normative commitment to the organization. Journal of Occupational Psychology, 63, 1-18 Allix, M.I.M. (2005). Rapport sur les enquetes aupres des etablissements publics nationaux, des autorites

administratives independantes et des services a competence nationale. [Report]. Rotterdam. Allix, M., & S. van Thiel. (2005). Mapping the field of quasi-autonomous organizations in France and

Italy. International Journal of Public Management, 8(1), 39-55. Barlow, J. D. Farnham, S. Horton & F.F. Ridley. (1996) Comparing public managers. In: Farnham, D., S.

Horton, J. Barlow, & A. Hondeghem. (eds.). (1996). New Public Managers in Europe- Public Servants in Transition. London, Macmillan. [pp.3-25].

Bovaird, T. & E. Loeffler. (eds.). (2003). Public management and governance. London, Routledge. Cassese, S. (1999).Italy’s senior civil service: an ossified world. In: Page, C. E. &. V. Wright. Bureau-

cratic Elites in Western European States. Oxford, New York, Oxford University Press. [pp. 55-64] Cassese, S. (2002). Le nouveau regime de la haute function publique en Italie: une modification

constituelle. Revue Francaise d’Administration Publique, 104, 677-688. Christensen, T. & P. Laegreid. (2003). New Public Management: the transformation of ideas and

practices. Aldershot, Ashgate. Christensen, J.G. (2005, forthcoming). Ministers and mandarins in a parliamentary system. International

journal of Public Administration [special issue] Committee Dijkstal. (2004). http://www.minbzk.nl/contents/pages/8415/commissiedijkstal-rapport.pdf Cornforth, C. (ed.). (2003). The management of public and non-profit organisations: what do boards do?

UK, Open University Press. Derlien, H-U. (2003). Mandarins or managers? The bureaucratic elite in Bonn, 1970 to 1987 and beyond.

Governance, 16(3), 401-428. Farnham, D., S. Horton, J. Barlow, & A. Hondeghem. (eds.). (1996). New Public Managers in Europe-

Public Servants in Transition. London, Macmillan. Farnham D., A. Hondeghem & S. Horton. (eds.). (2005). Staff participation and public management

reform: some international comparisons. Hampshire: Palgrave Greve, C., M.V. Flinders & S. van Thiel. (1999). Quangos: what's in a name? Defining quasi-autonomous

bodies from a comparative perspective. Governance, 12, 129-146. James, O. (2003). The executive agency revolution in Whitehall: public interest versus bureau-shaping

perspectives. [Transforming Government Series]. Longdon, Palgrave MacMillan. Kersting, N. & A. Vetter. (eds.). (2003). Reforming local government in Europe: closing the gap between

democracy and efficiency. Opladen, Leske + Budrich. Kettl, D.F., C.J. Pollitt & J.H. Svara. (2004). Towards a Danish concept of public governance: an

international perspective. [Report to the Danish forum for top executive management, August 2004]. Kickert, W.J.M. (1995). Public management and administrative reform in Western Europe. Cheltenham,

Edward Elger. OECD. (2002). Distributed public governance: agencies, authorities and other government bodies. Paris. Laegreid, P. (2003). Transforming top civil servant systems. In: Christensen, T. & P. Laegreid. (2003).

New Public Management: the transformation of ideas and practices. Aldershot, Ashgate. [pp.145-172] Le Grand, J. & W. Bartlett. (eds.). (1993). Quasi-markets and social policy. London, MacMillan. Ministerie van BZK (2004). Over dienen en verdienen. Adviescommissie beloning en rechtspositie

ambtelijke topstructuur. [Rapport]. Den Haag. Nelissen, N.J.M., M.P.C. Bressers & S.J.F. Engelen. (1996). De opkomst van een nieuwe

overheidsmanager: een onderzoek naar significante topambtenaren. Bussum, Coutinho.

34

Noordegraaf, M., A.B. Ringeling & F.J.M. Zwetsloot. De ambtenaar als publiek ondernemer. Bussum, Coutinho.

Noordegraaf, M. (2001). Attention! Work and behaviour of public managers amidst ambiguity. Delft, Eburon.

Page, C. E. &. V., Wright (1999). Bureaucratic Elites in Western European States. Oxford, New York, Oxford University Press.

Pollitt, C. &. G., Bouckaert (2000). Public management reform - a comparative analysis. Oxford, Oxford University Press.

Pollitt, C. & C. Talbot. (eds.). (2004). Unbundled government: a critical analyis of the global trend to agencies, quangos and contractualization. [Routledge Studies in Public Management.] London, Routledge.

Pollitt, C., C. Talbot, J. Caulfield & A. Smullen. (2004). Agencies: how governments do things through semi-autonomous organizations. Basingstoke, Palgrave MacMillan.

PUMA. (2002). Highlights of public sector pay and employment trends: 2002 update. Human Resources Management (HRM) Working Party Meeting. PUMA/HRM (2002)7. Paris, OECD. [www.oecd.org]

PUMA. (2003). Managing senior management: senior civil service reform in OECD member countries. Background note GOV/PUMA (2003)17. Paris, OECD. [www.oecd.org]

Rouban. (1994). France: Political Argument and Institutional Change. In: C. Hood and G. Peters. (eds.). Rewards at the top. UK, Sage. [pp. 90-105.]

Siciliani, J-L. (2004). La remuneration au merite des directeurs d'administration centrale: mobiliser les directeurs pour conduire le changement - Rapport au Premier Ministre. Paris, La documentation francaise: 45.

Suleiman, N. E. (1995). Les ressorts caches de la reussite francaise. Paris, Editions du Seuil. Talbot,C. (2004). The agency idea: sometimes old, sometimes new, sometimes borrowed, sometimes

untrue. In: Pollitt, C. & C. Talbot. (eds.). Unbundled government: a critical analyis of the global trend to agencies, quangos and contractualization. [Routledge Studies in Public Management.] London, Routledge. {pp.3-21]

‘ t Hart, P. ‘t et al. (2002). Politiek-Ambtelijke verhoudingen in beweging. Boom, Amsterdam. Van Oosteroom, R. & S. van Thiel. (2004). Agentschappen: kruiwagen voor modernisering?

Bestuurskunde, 14(7), 292-300. Van Thiel, S. & M.W. van Buuren. (2001). Ontwikkeling van het aantal zelfstandige bestuursorganen

tussen 1993 en 2000: zijn zbo’s ‘uit’ de mode? Bestuurswetenschappen, 55(5), 386-404. Van Thiel, S. (2001). Quangos: trends, causes and consequences. Aldershot, Ashgate Publishing Ltd. Van Thiel, S., M. Allix & J. Dwarswaard. (2004). Respondentenrapport onderzoek directies verzelfstan-

digde organisaties. Rotterdam, 29 April, 2004.