JPART 21:285306

The Loyalties of Top Public Administrators

Gjalt de Graaf
VU University Amsterdam

ABSTRACT

Images of the modern public administrator clash with yesteryears neutral public servants
obediently carrying out the orders of elected politicians. Partly influenced by the literature on
New Public Management, people often argue that public administrators today should
ensure quality services, give value for public funds, be responsive, operate strategically,
uphold organizations reputations, and the like. They thus seem to serve many masters, not
just politicians, but do the various masters interests harmonize or contradict? The research
question of this article is: Where do the loyalties of public administrators lie? The answer
involves all the potential objects of their loyalties: colleagues, the public good, moral
imperatives, the law, their organizations, the organizations clients, and elected officials. What
is the composition of loyalty for top public administrators? What are their conceptions of
loyalty? With the use of Q-methodology, we identify and describe four distinct types of
public administrators within the context of loyalty: (a) by-the-book professionals,
(b) societys neutral servants, (c) the personally grounded, and (d) open and principled
independents. These conceptions matter because they indicate how administrators behave
and make decisions.

INTRODUCTION
Much has been written recently on the modern and professional public administrator
(Noordegraaf, Steen, and Frissen 2006) who enjoys much discretionary freedom (Meier
1993; Shumavon and Hibbeln 1986). Partly inuenced by the literature on New Public
Management (NPM) (e.g., Maesschalck 2004) or New Public Service (e.g., Denhardt
and Denhardt 2000; Light 1999), it has been argued that public administrators should give
citizens value for money, deliver quality, be more responsive, operate strategically,
improve the reputation of their organizations, and so on (cf. Clarke and Newman 1997;
Noordegraaf 2004; Pollitt and Bouckaert 2000).
The images of the new public administrator clash with the classic images of the old
one (e.g., Weber 1946): the public administrator who neutrally and obediently carries out
orders of elected politicians. Since Weber, many interesting studies have been done on the
separation between administration and politics (e.g., Aberbach, Putnam, and Rockman
1981; Denhardt and Denhardt 2000; Peters 2001; Svara 1999); One of the most important

The author would like to thank the members of the research group Integrity of Governance of the VU University
Amsterdam for their invaluable comments. He is also indebted to the three anonymous reviewers for their
constructive advice. Address correspondence to the author at g.de.graaf@fsw.vu.nl.

doi:10.1093/jopart/muq028
Advance Access publication on June 21, 2010
The Author 2010. Published by Oxford University Press on behalf of the Journal of Public Administration Research
and Theory, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org
286 Journal of Public Administration Research and Theory

and enduring theoretical constructs in public administration is the politics-administration

dichotomy model (Svara 1998, 51). In this literature, it becomes clear that public admin-
istrators today serve many masters, not just politicians. Do any of the interests of their
masters contradict each other? And if there are loyalty conicts, whose interests are con-
sidered more important by administrators? As Selden, Brewer, and Brudney (1999, 172)
have noted: Popular stereotypes and scholarly depictions do not provide a clear under-
standing of how public administrators perceive their roles and responsibilities or how they
use their considerable discretionary powers.
Among the various objects of loyaltycolleagues, the public good, administrators
consciences, administrators organizations, the law, the organizations clients, and elected
ofcials (Bovens 1998, chap. 9 and 12; Denhardt and Denhardt 2000; t Hart and Wille
2002; Nieuwenkamp 2001)where do the loyalties of top public administrators lie? This
is our main research question and its answer is likely to be a mix of all possible loyalties.
This mix is also likely to lead to loyalty conicts: what loyalty conicts do top public
administrators experience?
Our results describe four conceptions of loyalty. Administrators conceptualizations
of their loyalties are morally important: they indicate how administrators behave and make
decisions (cf. de Graaf 2003, 2005a; Fletcher 1993). Furthermore, having different con-
ceptions means having different loyalty dilemmas and conicts.
Before presenting the results of the Q-study, we describe conceptions of loyalty with
respect to existing literature. We then discuss the main concepts and theoretical background
of the study, followed by a description of Q-methodology and the research background.
Next we describe the empirical results and present the four loyalty conceptions of public
administrators. In the last section, we discuss the results of the study.

LITERATURE ON PUBLIC ADMINISTRATORS LOYALTIES AND ROLES

Questioning Webers Ideal Type
The classic public administrator as envisioned by Max Weber (e.g., 1946) is neutral and
predictable and has unquestioned obedience and loyalty to the elected political authorities
in order to serve the public good (Weber 1946). According to Weber, administration
should be removed from politics and should serve as the neutral servants of their political
masters (Fry and Nigro 1996, 37). In this view, the role of a politician is to give direction
to policies; the bureaucrats role is not to engage in political decision making but to execute
the orders of their political superiors (Fry and Nigro 1996). This image of the administrator,
dominant in the rst half of the twentieth century, has been questioned both normatively
and empirically. We can refer, for example, to the many important studies on the relation-
ship between administrators and politicians and the proper role of administrators in the
political process (e.g., Aberbach, Putnam, and Rockman 1981; Denhardt and Denhardt
2000; Peters 2001; Peters and Pierre 2001; Svara 1999, 2001). Because this study is
not on the administrative-political dichotomy but on loyalties and loyalty conicts per-
ceived by administrators, rather than summarize this literature, we will only point out some
objections to Webers ideal type that explicitly deal with loyalty issues.
Public Choice is a clear example of a school with mainly empirical objections to
Webers ideal type. The school states that people in generalwhich of course includes
politicians and administratorsare motivated by self-interest. The loyalty of public
 de Graaf Loyalties of Top Public Administrators 287

administrators thus lies primarily with themselves and their own interests, not the public
good or elected ofcials (Dunleavy 1991). The principal-agent theory is an outgrowth of the
assumption that all actors are self-interested; trying to align the interests of principals and
agents.
Other scholars have argued on empirical grounds that the image of Webers public
administrator is simply outdated. The ever-changing organizational context in America and
bureaucracies worldwide changes role perceptions and loyalties (Bovens 1998). The com-
plexity of societal problems, professionalism of public administratorship, media inuence,
and citizens expectations of responsiveness from administrators (Rosenthal 1999), or
transformed constituencies and increased policy volatility (Durant 1991; Rourke 1991),
all play a role in the changing context. Denhardt (1988) has mentioned that public admin-
istrators have become more involved in making as opposed to merely executing policy; or,
as t Hart and Wille (2002) have phrased it, complementarity and teamwork among elected
ofcials and administrators are important elements of modern discourse on the relationship
between politics and administration. OLeary (2006), in her important study of Guerilla
Government, has discussed the reality of bureaucratic politics and pointed to three oft-cited
books on the policymaking inuence of career public servants: Kaufmans (1960) The For-
est Ranger, Hirschmans (1970) Exit, Voice and Loyalty, and Lipksys (1980) Street-Level
Bureaucracy. Clarke and Newman (1997) have argued that the expertise and education of
public administrators have risen and that some sense of professionalism has developed.
Professionalism implies a certain expertise and values which give direction to their actions.
Clarke and Newman (1997, 7) have called professional knowledge the engine of social
progress and improvement which would enhance the public good. Professionalism,
whatever its form, changes administrators perceptions of role, which in turn reshapes
loyalties.
Normative questions about public administrators with limitless obedience to superiors
and ofcials (e.g., Eichmanns bureaucracy) have also been raised in the literature (e.g.,
Arendt 1994; Bauman 1991) and in the more recent work on administrative evil by Adams
and Balfour (1998). Administrators, they have argued, should not be neutral cogs in the
wheel but inuence the decisions they make based on their own conscience. It has been
argued that unquestioned loyalty to the established authorities, public administrators, and
police in World War II played a crucial role in heinous crimes. Responsibility and moral
agency, they have concluded, should not be exclusively placed in the organization as
a whole or at the top; an irreducible individual moral agency remains for every member.
Uncritical loyalty to ofcials is dangerous and can lead to morally disastrous conduct.
In NPM literature, the strict loyalty of public administrators to elected ofcials has
also been questioned normatively. Loyalties of public administrators should lie with values
like efciency and effectiveness (cf. Clarke and Newman 1997; Denhardt and Denhardt
2000).

Role and Role Conceptions in the Literature

In the literature loyalty and loyalty conceptions are closely connected to role, role
conceptions, and role obligations (e.g., Bovens 1998; Cooper 1998; Petter 2005). An
administrative role has been dened by Selden, Brewer, and Brudney (1999, 175) as a
cohesive set of job-related values and attitudes that provides the public administrator a sta-
ble set of expectations about his or her responsibilities. Values play an important part in
288 Journal of Public Administration Research and Theory

this denition, proving that loyalty concept and conicts are also closely connected to the
value concept (de Graaf and van der Wal 2008). A study of public administrators loyalty
conicts is therefore closely tied to the (mostly normative) public administration literature
on values, moral conicts, and ethical dilemmas (e.g., Bowman and Williams 1997; Cooper
1994; Frederickson 1993).
Roles and role perceptions of public administrators have been studied extensively.
Downss (1967) classic Inside Bureaucracy described ve ideal types: climbers, con-
servers, zealots, advocates, and statesmen. Many others, like Long (1952) and Suleiman
(2003), have studied the role of public administrators and their use of discretionary powers.
OKelly and Dubnick (2005) have distinguished several models of bureaucratic behavior
on theoretical grounds and central to administrative ethics in the context of dilemma-
facing situations: the intuitive, blameworthy, virtuous, reexive, adaptive, just, rational,
strategic, and moralist bureaucrat. Empirical studies on the perception of responsibilities or
loyalties and empirical research on how public administrators self-reect on exercising
their considerable discretion are rare, even though OKelly and Dubnick (2005, 393) have
stated that the study of public administration in general should be aimed at analyzing how
public administrators make decisions in the face of dilemmas and in the context of the
structures bureaucracies provide. Selden, Brewer, and Brudney (1999) have provided
an interesting empirical study using Q-methodology on how administrators perceive their
role and responsibilities and found ve role perceptions: stewards of the public interest,
adapted realists, businesslike utilitarians, resigned custodians, and practical idealists.
No single study, to date, has focused directly on the loyalties of top public
administrators.

DEFINITION AND OBJECT OF LOYALTY

The Concept of Loyalty
Before we can answer anything about where the loyalties of public administrators lie, we
must clarify the concept of loyalty. Etymologically, the word traces back to the Latin
legalis, meaning law. Loyalty could then have been expressed by subjects being faithful
to legal objects and furthermore suggests a willingness to sacrice. Serfs and lords come to
mind here. Our denition, however, applies to persons with various objects of loyalty: to
what or whom is a person loyal? Whether it involves sacrice or the law is contextual.
Loyalty has normative, symbolic, and emotional connotations. Like integrity, there
cannot be too much of it. Graham and Keeley (1992, 191) have said: Loyalty is very
popular. It has few opponents. Consider the alternative and the label it evokes: disloyalty,
treachery, betrayal, and treason. Another similarity between integrity and loyalty is that
neither is often empirically studied, perhaps because making the concepts operational is
difcult. And when the concepts are studied, their denitions vary. Loyalty has been de-
ned as a desire to be and remain in the group (Ewin 1992, 419), an allegiance to a concept
outside the self (Hoffmann 2006, 2314), a disposition to accept rather than to criticize
(Birch 1975, 75), an emotional attachment and reaction to its object (Ewin 1993, 389),
a devotion to a cause (Royce 1908, 16), or being true to obligations (Vandekerckhove
and Commers 2004, 22).
In his classic Functions of the Executive, Barnard pointed to the importance of loyalty.
Barnard concluded that: There are a number of words and phrases in common use with
 de Graaf Loyalties of Top Public Administrators 289

reference to organization that reach back to the factor individual willingness. Loyalty,
solidarity, esprit de corps, strength of organization, are the chief .... Thus loyalty
is regarded as not necessarily related to position, rank, fame, remuneration, or ability.
It is vaguely recognized as an essential condition of organization (Barnard 1938, 84).
Most studies in organizational science that do study loyalty empirically have used
Hirschmans (1970) theory and denition (e.g., Golden 1992, adding the dimension of
neglect; Lee and Whitford 2008, adding the dimension of pay). In his Exit, Voice
and Loyalty, the object of loyalty is the organization, giving rise to the term organizational
loyalty. When we read in the literature, for example, that whistleblowing violates the
obligation to loyalty, it refers to this type of loyalty. Because we consider public admin-
istrators various objects of loyalty in our research, this particular construct of organiza-
tional loyalty is clearly inadequate.
As can be deducted from the different denitions, sometimes loyalty is described as an
attitude, sometimes behavior (Withey and Cooper 1992). It makes more sense to treat the
concept in this study as an attitude, a dedicationnoting, however, that dedication always
inuences behavior. In doing so, we do not specify loyalties as merely ideational: they
affect work and behavior in daily practices, mostly by prioritizing signals, issues, and meet-
ings (cf. Fletcher 1993). According to Jeurissen (1997), for example, loyalty is an attitude
aimed at an object, which involves a predisposition to act, is durable, and contains an
element of preference.
Pursuant to the above discussion, we will dene loyalty as the willing and practical
dedication of a person to an object (cf. Stoker 2005, 273). Amidst all the ambiguities, we
agree with most organizational scholars that (1) the subject identies the object of loyalty
(the cause) and (2) the subject behaves in a way that promotes the interests of the object of
loyalty (Fletcher 1993).

Objects of Loyalties
What are the possible objects of loyalty for public administrators? Several attempts to an-
swer this have been made by administration scholars. Waldo (1988) has identied 12 sour-
ces and types of ethical obligations of public servants (a list that, in his opinion, could be
expanded indenitely). They include obligations to the Constitution; the law, nation, or
country; democracy; organizational-bureaucratic norms; profession and professionalism;
family and friends; the self; middle range collectives; the public interest; humanity;
and religion or God.
Bovens (1998) has analytically distinguished ve conceptions of bureaucratic respon-
sibility. The object of loyalty changes in each (table 1).

Table 1
Five Conceptions of Administrative Responsibilities
Type Objects of Loyalty
Hierarchical Superiors and orders
Personal Conscience and personal ethics
Social Peers and social norms
Professional Profession and professional ethics
Civic Citizens and civic values
Taken from Bovens (1998, 149).
290 Journal of Public Administration Research and Theory

About the hierarchical conception, Bovens writes: One could call the idea that func-
tionaries owe their superior strict obedience, their organization complete loyalty, and the
outside world utmost discretion, the orthodox or classical conception of bureaucratic re-
sponsibility (Bovens 1998, 149). Often the terms neutrality or impartiality are used for this
view of responsibility. Since democratic control of public organizations is important, so is
the idea of primacy of politics, the classical conception. Does loyalty of administrators
then lie simply with political leadership? No, it could not be this straightforward. Bovens
(1998, 155):
If control by democratic organs is the most important consideration, one would rather expect
the loyalty of the civil servant to lie rst and foremost with the (majority of the) popular
representatives and only secondary with the political leadership, and only after that with his
own ofcial superiors and departmental organization.
Furthermore, as we saw earlier, the classical model leaves no room for individual
accountability.
To address this, Bovens distinguishes other conceptions of bureaucratic loyalty. In the
personal conception, the emphasis is on the beliefs and personal values of the administrator.
In the social conception, loyalty lies with colleagues, friends, clients, family, subordinates,
and acquaintances. In the professional conception, loyalty is driven by their profession and
professional rules. Bovens notes that the latter is on the rise and has the advantage of in-
troducing external moral considerations into the organization. It could thus assist in coun-
teracting phenomena like groupthink, peer pressure, and crimes of obedience. The fth and
nal conception is civic: loyalty to citizens. It allows disobedience to political leaders based
on appealing to general public interests, such as preserving the rule of law or preventing
large-scale waste of resources. The conceptions are forces that work on an individual
functionary in the case of conicts of loyalties. A strong centripetal force often emanates
from the hierarchical and social conceptions of loyalty. (Bovens 1998, 165).
Based on an extensive literature study, Petter (2005) distinguished eight themes or
responsibility types connected to specic values: (1) moral responsibility (personal moral-
ity, public service, public interest), (2) professional responsibility (competence, expertise,
good practice), (3) hierarchical responsibility (compliance, oversight, productivity), (4)
scal responsibility (efciency, economy), (5) legal responsibility (rule of law, constitu-
tional rights), (6) leadership responsibility (leadership, prudence), (7) consumer responsi-
bility (exibility, effectiveness), and (8) public responsibility (social equity, openness,
responsiveness). Each connects to a specic potential problem area.

Loyalty Conflicts and the Thick-Thin Distinction

As soon as multiple loyalties of public administrators are distinguished, loyalty conicts are
unavoidable. Since each loyalty is closely tied to specic values, these conicts can in part
be described as value conicts or ethical dilemmas (Bowman and Williams 1997; Cooper
1994; Frederickson 1993; OKelly and Dubnick 2005). Most literature on how adminis-
trators deal with dilemmas in their public service roles has been more empirical in nature
than the literature on the exercise of discretionary powers in modern bureaucracies. Like
the empirical research of this article, it has focused on understanding the nature of admin-
istrators dilemmas. The world of administrators is seen as one of multiple, diverse and
often conicting expectations (Dubnick and Romzek 1993). (OKelly and Dubnick 2005,
 de Graaf Loyalties of Top Public Administrators 291

395). OKelly and Dubnick (2005, 396) state that the tick-thin distinction (Williams
1985) runs to the heart of administrators dilemmas.
Thin moral reasons are universal, general moral principles, and applied without ref-
erence to specic contexts and situations, as in: administrators should rst and foremost
be loyal to political electives. Thick moral reasons, on the other hand, are context-specic
reasons, a tractable morality (de Graaf 2003, 2005a).
The thick-thin distinction is similar to the distinction between the micro- and macro-
morality often used in corruption literature (de Graaf and Huberts 2008). Micro-morality
has to do with connections to people in our social circles (family, friends). Moral obliga-
tions in our personal lives are characterized by reciprocity: we help friends and family just
as we expect them to help us. Macro-morality, in contrast, emphasizes the universal. It is
the product of the process (described by Nelson 1949) of universalizing morality and claims
the legitimacy of its norms on institutions of the law, a universal system of formal norms.
Macro-morality is characterized by the complementarity of rights and duties as the primal
modus of social ties.
The thick-thin distinction will be used here to describe the origins of administrators
loyalties. OKelly and Dubnick (2005, 398) have said:
The thick-thin distinction is very helpful in clarifying moral philosophy and thought, denoting
as it does dilemmas people face as they are torn between, on the one hand, allegiances to
communities and peers and, on the other hand, duties towards all humanity or toward xed
principles of action established, a priori, to any hard case that might emerge.
Given the close conceptual relationship between specic values and objects of loy-
alties, the distinction can be used to understand loyalty dilemmas and the justications,
background, and evolution of loyalties.

RESEARCH METHODOLOGY: Q
Here we study and describe the loyalty conceptions and conicts of Dutch top public
administrators. The context is relevant: Aberbach et al. (1981) already noticed clear dif-
ferences between the United States and Europe on the degree of intertwinement between
political and administrative elites (Lee and Raadschelders 2008, 423). The Dutch context
in brief is the Netherlands is a decentralized unitary state, with parliamentary sovereignty
and cabinet government. Only the Members of Parliament are directly elected by the voters,
not the cabinet and its ministers. After parliamentary elections, a cabinet formation takes
place, resulting in a coalition. Almost all cabinet ministers head a department, which are
staffed by career public administrators (no party political appointments). The idea of the
loyal, politically neutral civil servant recruited on the basis of Weberian merits criteria is
an entrenched norm (t Hart and Wille 2006, 126). Dutch politico-administrative relations
comply with the functional village model in Peterss typology (Peters 1997): politicians and
administrators share a policy role along functional lines (Brans et al. 2006, 61). The Nether-
lands is often described as a consensual democracy (Lijphart 1999). For a good overview of
the Dutch administrative system, see Andeweg and Irwin (2005) or Toonen (1990); for
more on the specic relationship between ministers and top ofcials in the Netherlands,
see t Hart and Wille (2006).
There are various ways to empirically study where the loyalties of public administra-
tors lie; here Q-methodology was deemed most suitable because Q-study results are clusters
292 Journal of Public Administration Research and Theory

that are functional rather than logical (de Graaf and van Exel 2009; van Exel, de Graaf, and
Brouwer 2006). In other words, the clusters are not logically constructed by the researcher,
they result from the empirical data; they are operant. Q-methodology can reveal a charac-
teristic independently of the distribution of that characteristic relative to other character-
istics in a population. Unlike surveys, which provide patterns of variables, Q-methodology
provides patterns of persons, in this case, administrators. Q-methodology is a mixed
qualitative-quantitative small-sample method that provides a scientic foundation for the
systematic study of subjectivity, such as peoples opinions, attitudes, preferences, and so
on (Brown 1980, 1993; Watts and Stenner 2005). Q-methodology has been linked to
Webers theory before (Brown 2000; Stephenson 1962). On the type of research question
of this article, Petter (2005, 211) has said:
For instance, do administrators who evaluate subordinates based on hierarchical
responsibility also emphasize other low-autonomy perspectives? Or do they balance it by also
stressing an outcome focus, such as public responsibility? A Q-methodology study may reveal
patterns.
Q-methodology was applied to this study through four steps (discussed below):
selection of relevant statements (Q-set), selection of respondents (P-set), respondents
ranking of statements (Q-sort), and interpretation of the results (Q-analysis).1

The Q-Set
In a Q-methodological study, people typically are presented with a sample of statements
about some topic (here, issues concerning the loyalties of public administrators), called the
Q-set. We compiled a list of all the quotes on loyalties, responsibilities, and role concep-
tions found in academic and popular literature. The original list contained more than 600
quotes. According to Brown (1980), it is best to use a structure for selection of a represen-
tative miniature of such a list. Whatever structure is used, it forces the investigator to select
statements widely different from one another in order to make the Q-set broadly represen-
tative (Brown 1980). Irrespective of the structure, it is the subject that eventually gives
meaning to the statements by sorting them (Brown 1993).2
A Q-set of 42 statements was chosen (see Appendix) by rst discarding overlapping
statements and second, applying the following schedule to the statements:
Table 2 is based on Bovenss (1998) objects and Petters (2005) responsibilities of
administrators loyalties.3 We ensured that each theoretically relevant category was rep-
resented in the 42 statements in both the object and background of different loyalties. The
logic of the Q-set was not used to classify the respondents. As stated above, Q-study results
are functional rather than logical clusters; that is, the researcher does not logically construct
the clusters. They result from the empirical data; they are operant.

1 The main source for Q-methodology is Stephenson (1953). Within the social sciences, Brown (1980) is a classic.
2 The limited number of comparative studies that have been carried out indicate that different sets of statements
structured in different ways can nevertheless be expected to converge on the same conclusions (Thomas and Baas
19921993).
3 The models supplement each other and largely overlap. The rst four conceptions of Bovens overlap with the
hierarchical, moral, leadership, and professional loyalties of Petter. Bovenss fth conception, civic responsibilities,
was specied in Petters societal, legal, and customer loyalties. Petters eighth responsibility, scal, is now seen as part
of the professional loyalty.
 de Graaf Loyalties of Top Public Administrators 293

Table 2
Logic of the Q-set
Thick Thin
Hierarchical loyalty
Personal loyalty
Social loyalty
Professional loyalty
Societal loyalty
Legal loyalty
Customer loyalty
Note: Adapted from Petter (2005) and Bovens (1998).

The P-Set
The set of respondents is usually not randomly chosen but theoretically structured (Brown
1980), and all viewpoints should be included. All top-level Dutch public administrators are
automatically members of ABD (Algemene Bestuursdienst; the Directorate for the Senior
Public Service), the professional organization for top administrators. Since its foundation in
1995, public administrators ofcially work for the ABD rather than a particular ministry.
Total ABD membership is 857. At its highest level are 61 administrators spread throughout
15 ministries, 52 of which are men and 9 women.4 By letter, we requested all for interview.
We also invited 50 other ABD members from different ministries (Interior; Health, Well-
being and Sport; Defense and Foreign Affairs) to ensure variety. From this group, all 26
women were selected for gender leveling and 24 men (6 from each ministry) were selected
randomly. Of the total 111 invitees (76 men and 35 women), 38 agreed to interview.5,6

The Q-Sort
By Q-sorting, people give subjective meaning to the set of statements and so reveal their
subjective viewpoint. Stephenson7 has presented Q-methodology as an inversion of con-
ventional factor analysis, in the sense that it correlates persons instead of tests (i.e., by-
person factor analysis). If each individual had unique likes and dislikes, their Q-sorts would
not correlate. If, however, signicant clusters of correlations exist, they can be factorized,
described as common viewpoints, and individuals can be mapped to a particular factor.
Using a quasi-normal distribution, respondents were asked to rank-order the 42 state-
ments from their own point of view according to some preference, judgment, or feeling
(gure 1).8 The two statements he or she agreed with most were put on the right (for a score
of 13); the two he or she disagreed with most on the left (23). The statements they felt

4 Based on Internet data of March 5, 2007, at http://www.algemenebestuursdienst.nl.

5 Discounting undeliverable invitations (many addresses were invalid), the response rate was a respectable 42%.
6 The interviews were held by drs. Angelique Hornstra and drs. G. J. de Graaf.
7 William Stephenson, the inventor of Q-methodology, served as the last assistant to Charles Spearman, the inventor
of conventional factor analysis (Brown 1997).
8 Even though a forced distribution was used, some deviations were tolerated. If the Q-sorters found the forced
distribution too much unlike their positions, they were allowed to slightly vary the number of statements they were
supposed to have in a category.
294 Journal of Public Administration Research and Theory

Figure 1
Fixed Distribution of the Q-set

indifferent about (or did not understand) were put in the middle (the 0 category). The nal
distribution was the Q-sort. The Q-sorts were factor analyzed with the objective of reveal-
ing a limited number of corresponding viewpoints.
After the statement sorting, we held a second interview with each respondent to gain
insight into the reasoning behind their choices, for example, Why did you put these two
statements in the 13 category? Do you see an issue concerning loyalty that is missing in
the statements? This helped with the nal analysis of the different factors.

Q-Analysis
The individual Q-sorts were factor analyzed using PQMethod 2.119 (extraction method:
centroid; rotation method: varimax) in order to reveal the distinct ways in which the state-
ments were rank ordered. The analysis led to four factors (or loyalty conceptions) A, B, C,
and D. For each factor, a composite sort was computed based on the rankings of the re-
spondents loading on that factor and their correlation coefcient with the factor as weight.
This idealized Q-sort represents the way in which a person loading 100% on that factor
would have ranked the 42 statements (see Appendix).
Each factor was interpreted and described using the characterizing and distinguishing
statements and the explanations of respondents loading on the factor. A statement is char-
acterizing by its position in the outer columns of the idealized Q-sort of the factor and is
distinguishing if the position is statistically signicantly different from its position in the
idealized Q-sorts of all other factors. Respondents explanations (which were transcribed
literally during post-Q-sort interviews) are cited in italics to illustrate administrators way
of thinking and support the description of that viewpoint. Corresponding statement num-
bers from the Q-set are noted in parentheses.

RESULTS: THE FOUR LOYALTY TYPES

Even though all administrators subscribe to the notion that of being loyal to their ministers,
the loyalty is interpreted differently, resulting in different kinds of loyalty conicts. Four
loyalty conceptions were elicited from the analysis and are presented below in the form of
a label and narrative (cf. de Graaf 2005b; van Exel, de Graaf, and Brouwer 2006).

9 Dedicated software and manual can be downloaded from http://www.rz.unibw-muenchen.de/p41bsmk/qmethod.

 de Graaf Loyalties of Top Public Administrators 295

Factor A: By-the-Book Professionals

Public administrators of the by-the-book professional type of loyalty want foremost to serve
society and its citizens. They believe this can best be done by being loyal to their own
profession and professional rules. This is clearly not a loyalty to the stakeholders of their
own organization or colleagues. Their moral reasoning is generally thin.
Factor A administrators work for the government because they want to serve society
(#7): That has always been my motivation. For me working in the private sector was
not an option. It is a sense of duty. Being loyal to your minister is your task, but once
the minister harms the public interest, the loyalty stops. That means that they aim for the so-
cietal effects of their work. Administrators who are merely focused on their own output are not
good public administrators (#22): You have to be effective in society, not just effective for
your department. Administrators should be focused on improving things in society. You have
to worry about outcome; output is just focused on the short term, something to strive for
immediately. We are here for the citizen, rst and foremost. Administrative output is about
great policy business, which can be bad for society. You have to be loyal to the minister, but
there are boundaries. Being there for the citizen is meant in a broad, abstract sense and thus
denotes a thin relationship, not an actual relationship with a citizen or stakeholder: There is
too much talk about being closer to the people and so on. My conviction is that people are at
a distance from us and thats the way it should be. You have to defend what you do, be ac-
cessible and approachable, but the relationship should not be too close or too personal.
Public values are clearly held above administrators own political values (#13): If
theres tension between public values and my own political values, the public values pre-
vail. That does not mean that administrators should have a clear picture of the good so-
ciety; elected politicians are responsible for that. However, blindly following the minister
is not something a professional should do (#34): If the minister asks for something
inappropriate or irresponsible, I wont do it.
Society can best be served by being a professional public administrator, which is not
about pride, status, or peer connectivity: This is a professional institution in the service of
the government. Its all about quality. Colleagues sometimes have specic interests in
mind or specic assignments that are in their way. I dont do that. Being professional also
means that an administrator should not misbehave in private life (#29): Administrators
should set a good example in society for being impartial. Thats the only way to make sure
that people look up to you. In private life you also have to watch that, especially top ad-
ministrators. As an administrator you cant afford to let it all out on weekends. You
always have to behave decently. Thats part of being a public administrator.
Professionalism and professional rules are important (#32): Professionalism and loy-
alty are the leading values in my job. Im a professional. You have to be loyal to the rules of
your profession. Im not someone who just follows orders. My profession, knowledge, and
conscience determine the borders of my job. Professionalism is the key word for public
administrators. You have to be loyal to the rules of your profession. Professionalism is
important and that doesnt just mean knowledge, it also means that you behave decently
and follow the rules of the game. Your behavior should be guided by the public interest.
Bending the rules to help societal stakeholders or specic citizens is unprofessional (#30)
and can even lead to conict: Rules form the playing eld. Only politics can change that.
I do not condone being exible to satisfy all our stakeholders. Societal partners are
important, but a deal is a deal. If someone wants something different, the rules should
296 Journal of Public Administration Research and Theory

be changed. The smallest bit of arbitrariness or privilege-granting is wrong. It under-

mines the principles of our constitutional state. Rather than hindering effectiveness and
efciency (#37), rules are a necessary part of any good and fair government. These admin-
istrators, however, are not neutral rule followers (#21) and a professional administrator is
not a slave to the rules. I dont just stamp pieces of paper!

Factor B: Societys Neutral Servants

Like the professionals, neutral servants feel a strong loyalty to society in general and to
public values. They believe they can best serve society by being as neutral as possible; in
a democracy elected ofcials should make the decisions. Of all types of top administrators,
they are most loyal to their ministers and have the fewest conicts with them. Their loyalty
is not with their profession; the profession and professional rules were not mentioned by
these respondents. Another difference from by-the-book professionals is that their own
opinions and convictions carry less weight in their work.
Neutral-servant administrators want to serve society (#7); they work with passion for
the government (#24): My whole career has been in public service, a conscious choice.
I nd pleasure in serving society in my regular and additional functions. My goal is to
help society through my job. They are focused on societal effects rather than bureaucratic
output (#22): Thats my greatest horror and annoyance. I work for the central govern-
ment because I want to do something for society. Theres too much attention nowadays on
measuring output, but thats not what it should be all about. It should be about whether
something actually happens. Sometimes bureaucratic output is bad for society. Were
here for the citizen. You have to be completely loyal to your minister. Output is nice
to show quick results. But our work should be organized to achieve lasting societal effects.
Will the citizen be better off? That should be the leading question. Personal values should
not be above an individuals own political values (#13). And it is not so much a personal
opinion of what is right that should guide administrators behavior: Personal values are
important, but you should serve the public interest. When personal values play too big
a role, public trust in government is undermined. That administrators should not base
their decisions on personal values is very important. It is the only way to maintain legal
security and equality before the law. I nd administrators inuenced by political parties
considerations to be gravely in error. The elected political should always have the last
word. The minister has certain goals and it is my task to technically advise on how he can
best reach those goals.
Neutral servants clearly serve the public good in general; their loyalty is not with their
own organization or ministry (#35). They are also least rebellious when ministers ask them
something unlawful (#2): The elected ofcials are responsible, period. However, when
something is bad for citizens, I warn the minister more than once.
The major parting point from by-the-book professionals is that neutral servants nd
their own role less important and think that top civil servants do not have the right to opine
publicly about their occupations affairs (#5). Freedom of speech is inside the walls. You
have to be allowed to say anything to the minister. Outside you have to restrain yourself,
otherwise there is a loyalty problem. Of course I feel that I should be able to talk to my
friends about public affairs. But once I speak as a top administrator, my personal opinion
does not count. As an administrator it is not your decision. You present choices and the
elected ofcials decide. Never reveal outside that you disagree with something. In this
 de Graaf Loyalties of Top Public Administrators 297

respect, it is nonsense to claim that the disqualication of public administrators opinions

cannot be tolerated (#39); again because it is all about being a good advisor, not about the
personal opinion of the administrator. When youre with three administrators at one table
and personal opinions are allowed, the minister gets three different pieces of advice; that
doesnt work. There has to be one objective, qualied opinion. Only then is it administrative
advice. Quality is not in the word opinion; it suggests something personal.

Factor C: The Personally Grounded

The loyalty of these public administrators tends to lie with their private lives and their own
consciences. They often use thick moral arguments. They weigh work loyalties against
those dictated by other societal roles, sometimes leading to loyalty conicts. They are crit-
ical of government and bureaucracy and dislike rules because they impede creativity.
Their private lives matter above all else, including their high-level jobs (#1): The
business at work sometime creates tensions at home, like not giving attention to your wife
and children. But in the end I would always choose for my private life. My family comes
rst. My private life is more important than my job because a job is just one facet of who I
am. My life has many components. They therefore are not twenty-four-hour civil servants
(#24): I cant imagine being that. When I come home I am a mother, wife, and all sorts of
other things. I have many roles and identities. Expecting me to behave as an administrator
outside the ofce is absurd.
Loyalty to their identities and consciences is important (#34, 25): On my policy
terrain, I am the expert. I look at an interest that goes above the minister. Neither
as a human being nor as a professional can I do things that go against my conscience.
Sometimes balancing is necessary but too much balancing means there is something wrong
and it is time to leave. That makes the loyalty dilemma for our profession tough sometimes.
I cannot do something if it will keep me awake at night. My identity is the most important.
Rules and regulations do not lead these administrators jobs; indeed, they have a no-
table dislike of them (#11): Theyre repugnant to the individual mind; they kill its cre-
ativity. Every day it seems theres a new litany of rules. Theyre not interesting and dont
work. You have to follow most rules, but sometimes you have to apply a higher norm.
Whats legal doesnt interest me. Rules are made by people in power and often stem
from their own interests or the latest fads. As a society you have to have some rules,
but in the end I pursue a higher value. Government is schizophrenic. It has an urge
to control everything just to make sure everything is executed correctly. I nd it important
to know the rules, but it cannot be that a system-of-madness determines what I can and
cannot do . . . I see too many colleagues who strictly follow rules. It destroys their creativity.
Nor does the loyalty of this type of administrator lie with stakeholders in their policy
eld. Letting concerned parties know they disagree with policy is beyond the purview of
their jobs (#31): If I think that a minister ought to do something else, its an internal
matter. I keep matters internal. I would not tinker with external powers as a counter-
weight to a ministers decision. For these administrators, elected politicians have the last
word: The primacy is with the politician.

Factor D: Open and Principled Independents

The loyalty of independent public administrators lies clearly with their own principles, the
law, and ofcial rules. They argue mainly in terms of thin guidelines and principles. They
298 Journal of Public Administration Research and Theory

were averse to the statements that began with I [think, feel, believe, etc.]. There is too
little room to express their own consciences, which in any case might lead to loyalty con-
icts. They believe that elected ofcials should determine the general direction of policies
under the limits of the law and ofcial rules. Loyalty conicts also arise when ministers
want something disallowed by law or when expressing themselves in public.
Independent administrators will not quickly go against their conscience: The goal of
my job is not just to do what management wants. Its a mix of my own conscience, external
contacts, etc. Doing something I cant live with means Im on the wrong path, Im em-
bracing the wrong loyalty. In any mature democracy, top public administrators should be
allowed to say what they want in public; thats simply freedom of speech (#5). This some-
time leads to loyalty conicts; the role of the involved citizen may clash with the role of top
civil servant. I feel that an administrator is part of society. He should therefore be al-
lowed to speak to outsiders. You dont make policy behind your desk; you have to talk to
outsiders. I feel we sometimes say too little in public. I do that sometimes, but am not
thanked for that. Of course, you cant say that what the minister wants is bullocks, but you
should be able to publicly express concerns. Politicians determine in the end, but that
doesnt mean I have to be silent about it.
Public administrators may decide for themselves what they believe the best option
is, which doesnt mean they have the right to make the nal decision (#20). Politics decide,
not the civil service. Public administrators should not base their decisions on personal values;
in such cases, citizens could no longer trust that policies and rules are followed uniformly
(#41). The core of our job is our workmanship; choices are part of that. But there are
professional boundaries to that. I do not make decisions about gay marriages or abortion.
If you want to determine that sort of decision, you are in the wrong trade. What follows from
that is that loyalty to the minister is important (#31), but it is not an absolute loyalty. These
administrators will not do something that is against the law, even if asked by the minister:
Thats my civil responsibility. If the minister wants to do that, I wont participate.
Loyalty does not lie above all at home (#1). Administrators should not complain about
the tensions and responsibilities that come with their positions: Whatevers important in
my job, I take home. If it takes a toll on my private life, so be it. Its the sort of thing that
comes with the job.

COMPARING LOYALTY CONCEPTIONS OF TOP DUTCH PUBLIC ADMINISTRATORS TO

OTHER PUBLIC ADMINISTRATORS
The loyalty types of Selden, Brewer, and Brudney (1999)stewards of the public interest,
adapted realists, businesslike utilitarians, resigned custodians, and practical
idealistsare different from those identied in our study. In particular, we found no busi-
nesslike utilitarians who could be described by statements like, Efciency is far more im-
portant to methanbeing fair orsensitive. I feel that being fair is anexcuse foraccepting a lack of
competence and a lowering of standards (Selden, Brewer, and Brudney 1999, 189). Possible
explanations are (1) the authors concept of role (and therefore their Q-set) was broader; (2)
their sample of public administrators differed from ours, which was restricted to top admin-
istrators; (3) contexts have signicantly changed over the past ten years; and/or (4) cultural
differences (the Netherlands versus the United States) affect conceptions of loyalty. This rai-
ses the question of the extent to which Dutch top public administrators conceptions of loyalty
differ from other types of administrators and other nations top administrators.
 de Graaf Loyalties of Top Public Administrators 299

A similar study among top municipal administrators and street-level bureaucrats (to be
more precise: local license providers) in the Netherlands (de Graaf 2010; de Graaf and van
der Wal forthcoming) found that top public administrators conceptions of loyalties seem to
be different from other groups of administrators. For example, and perhaps not surprisingly,
client loyalty seems to play a large role in the loyalty conceptions of street-level admin-
istrators, whereas loyalty to stakeholders in the policy eld (the equivalent of client loyalty
for top-level administrators) plays a small role in the loyalty conceptions of top adminis-
trators as described in the previous section; just like among the top municipal administra-
tors (the smallest of all distinguished objects of their loyalty; de Graaf 2010). Some types of
license providers express a strong loyalty to their clients, types that have been identied in
the literature (e.g., Lipsky 1980). In the case of street-level bureaucrats, therefore, dealing
with government clients on a daily basis has a signicant impact on loyalty and engenders
a type of loyalty not seen in top-level administrators.
Hierarchical loyalty seems to be the most important object of loyalty for all three
groups of administrators. To this generalization, we immediately add that the hierarchical
loyalty is interpreted differently within each group (see, e.g., the Results section) as well as
between the groups and is contextually dependent. Furthermore, hierarchical loyalty is
clearly weakest in street-level bureaucrats relative to other groups; indeed, the hierarchical
loyalty of several types of municipal license providers is relatively weak. An explanation
for this could be that street-level bureaucrats, unlike top administrators, rarely deal directly
with elected political superiors, which inuences the way they interpret hierarchical loy-
alty. Since they also have the strongest loyalty to clients they have day-to-day contact with,
we can hypothesize that, in general, the higher the frequency of contact with specic stake-
holders, the stronger the loyalties towards them.
Are the loyalty conceptions of top Dutch administrators different from other Western-
style democracies? And does the form of government inuence the loyalties of top admin-
istrators? As argued previously, context is relevant (Aberbach et al. 1981). Yet, it can be
hypothesized that the nature of the work of top administrators in Western-style countries is
similar to such an extent that the differences between the loyalty conceptions of Dutch top
administrators (as presented here), and, say, American top administrators are smaller than
those between Dutch top administrators and Dutch municipal license providers. Based on
the hypotheses on the differences between the Dutch groups and the international compar-
ative literature on political-administrative relations, we could furthermore hypothesize that
contextual factors of inuence on international differences between loyalty conceptions of
top administrators include the degree of intertwinement between political and administra-
tive elite and the recruitment of top administrators (e.g., Aberbach et al. 1981; Lee and
Raadschelders 2008). Lee and Raadschelders (2008, 243): In terms of recruitment for
top bureaucratic positions, national differences have been found. The United States appears
to emphasize loyalty and political responsiveness to the government in power; the British
model stresses expertise of top-ranking civil servants, while the German model combines
loyalty and expertise. Combining this with the differences between the groups in the Neth-
erlands, we can expect hierarchical loyalty to be more important in the United States and
professional loyalty to be more important in the United Kingdom. Also, we can expect the
strong involvement of the Dutch civil service in the production of policy advice and
limited use of political advisers (Brans et al. 2006, 58) to inuence loyalty conceptions
in the Netherlands.
300 Journal of Public Administration Research and Theory

Because of the aforementioned connection between the ambiguous concepts of loyalty

and accountability, a comparative study could add to the literature on different account-
ability notions within and between Western-style democracies (e.g., Bovens et al. 2008).
The classical model leaves no room for individual accountability.

CONCLUSION AND DISCUSSION

Although all administrators see their elected ministers as important objects of loyalty, no
type adheres to Webers ideal of unquestioned obedience and loyalty. We found differences
in how modern top public administrators weigh their loyalty to the minister vis-a-vis other
loyalty objects; four different loyalty conceptions were identied. By-the-Book Professio-
nals (conception A) want rst to serve society and its citizens. They generally use thin
moral reasoning and are more loyal to their own profession and professional rules than
the other top administrators. Like them, Societys Neutral Servants (conception B) feel
a strong loyalty to society in general and public values. Unlike conception A, however,
they believe they can best serve society by being as neutral as possible and by giving little
weight to their own opinions and convictions; of all types, they are the most loyal to their
minister and have the fewest conicts with them. The Personally Grounded (conception C)
generally employs thick moral arguments. More than any others, their loyalty lies most with
their private lives and individual consciences. They weigh their loyalties against those dic-
tated by other societal roles, sometimes leading to loyalty conicts. They dislike rules be-
cause they impede their creativity. In contrast, the loyalty of the Open and Principled
Independents (conception D) lies largely with their own principles, the law, and ofcial
rules. More than any other administrators, they employ thin guidelines and principles.
The high correlation between the factors ranged from .42 to .63, which usually indi-
cates an underlying consensus among respondents. In our case, it meant that the top admin-
istrators did not view their loyalty conceptions completely differently; the four factors can
be seen as variations on a common theme. To help with pinpointing where the theme lies,
a second-order factor analysis was performed on the four Q-sorts.
The underlying theme is that all administrators subscribe to the notion of being loyal to
their ministers (the hierarchical object of loyalty), as was concluded earlier by t Hart and
Wille (2006). The second-order analysis revealed that another object of loyalty is also im-
portant to all administrators: in the underlying consensual factor, statements concerning
society as the object of loyalty were most important (statements #7 and #13). Most top
administrators stress the importance of the public interest (as opposed to their own political
values) and that they work for the government because they want to serve society. This
indicates a strong Public Service Motivation (PSM) among top Dutch public administrators
(Perry 1996).
Several statements were ordered very similarly by the administrators (contributing to
the rather high correlations). For example, all four types rejected the view that efciency is
more important than impartiality and fairness (see statement #4 and its associated factor
scores in the Appendix). Whatever the inuence on Dutch civil service, these traditional
public values remain central, even at the cost of efciency: Effectiveness is of course
important, but its not what government is about. Government is not a company. There
are demands like transparency and so on. Citizens should be able to trust government. That
means procedural justice and impartiality, including the bureaucracy that it sometimes
brings.
 de Graaf Loyalties of Top Public Administrators 301

Also notable is that none of the administrators felt a strong connection with other ABD
members; their loyalty is therefore not with their particular organization: Does not in-
terest me. All the emails I get, I delete without reading. Theyre nice people, but I have
more to do with the top of the civil service. They happen to be ABD members, but only their
position is important to me. I would function perfectly well without the ABD.
When we look at the objects of loyalty as earlier distinguished in table 2, it is clear that
hierarchical loyalty is the most important of all the loyalty conceptions. Loyalty conception
B has the highest loyalty to political superiors and conception D has arguably the lowest.
The second-order factor analysis revealed that society is also an important object of most
top administrators loyalty (indicating a strong PSM in this group). Also notable is that
loyalty to stakeholders in their policy eld (customer loyalty in table 2) plays the small-
est role. Administrators loyalty to colleagues has an important role in their conceptions;
the other ve objects of loyalty (hierarchical, personal, professional, societal, and legal) all
play a dominating role in at least one of the loyalty conceptions.
Based on a survey of its readers, the Dutch journal re.Public,11 recently claimed that
more than half of the respondents (presumably administrators) indicated that they were not
willing to work for a Partij voor de Vrijheid politician.12 This started a media discussion on
just where the loyalty of Dutch administrators in fact liesthe research question at hand
here. The question was answered by the four descriptions of loyalty conceptions of top
administrators. They described how top public administrators conceptualize their loyalties
and weigh objects of loyalties against each other. Administrators conceptualizations of
their loyalties are morally important: they indicate how administrators behave and make
decisions (cf. de Graaf 2003, 2005a; Fletcher 1993). Furthermore, different conceptions
mean different loyalty dilemmas and conicts, which are in turn indicative of different
solutions.

APPENDIX

Table A1
Idealized Scores of the Factors of the 42 Statements
A B C D
1. In the end, my private life is more important than my 1 0 3 22
work.
2. When elected ofcials ask something of me that is 1 21 0 2
against the law, and despite my warnings they do not
listen to me, I will not do what they ask.
3. Committing to obligations towards colleagues can lead 0 21 0 21
to cronyism.
4. In my work, efciency is more important than 23 22 22 23
impartiality and fairness.
5. In any mature democracy, top public administrators 21 23 21 3
should be allowed to say what they want to in
public. Thats simply freedom of speech.
Continued

10 May 8, 2009.
11 The Partij voor de Vrijheid is a new anti-Islamic party in the Netherlands headed by Geert Wilders.
302 Journal of Public Administration Research and Theory

Table A1 (continued)
Idealized Scores of the Factors of the 42 Statements
A B C D
6. It is my main duty to mediate between conicting 0 21 21 0
interests and to nd solutions everybody can live
with.
7. I work for the government because I want to serve 3 3 1 1
society.
8. I try to mitigate extreme resistance from societal 0 0 1 1
partners of our organization.
9. Sometimes elected ofcials want something that is 0 0 0 0
practically impossible. What I notice in those cases
is that we as public administrators have a very hard
time saying no.
10. In my work, I worry about the well being of less 0 1 21 0
privileged citizens
11. I nd it very important to know all the rules and 0 22 23 22
regulations and to stick to them.
12. The public administrator whose actions are 1 2 2 1
determined by party political considerations
undermines essential principles and procedures of
democracy.
13. In my work, I should apply public values over my 2 3 1 1
own political values.
14. How I look at things in my work is partly determined 21 0 1 1
by the people who surround me in my private life;
a discussion with friends can inuence my judgment.
15. I have my own opinions on the morality of my actions 1 0 0 21
when at work I cannot leave that up to my political
superiors.
16. Even without religious or philosophical motives, 0 0 0 21
administrators should be allowed to refuse a task if
their conscience dictates so.
17. Public administrators should be accountable to 22 21 22 21
criminal law for their public actions, even when
doing exactly what their political superiors expect
or want them to do.
18. It is my main duty to do what management expects 21 22 21 22
me to.
19. The most important features of good public 1 0 0 0
administrators are that they are loyal to their
organization and discreet in the outside world.
20. A public administrator works in public service. The 21 21 21 22
political ofcial should be his/her rst client, but
public administrators should decide for themselves
what they think the best option is.
21. I know what is legal, not what is right. I stick to what 22 21 21 21
is legal.
22. Good public administrators focus on societal effects. 2 2 0 1
Bad public administrators focus on bureaucratic
output.
Continued
 de Graaf Loyalties of Top Public Administrators 303

Table A1 (continued)
Idealized Scores of the Factors of the 42 Statements
A B C D
23. Even when my personal convictions about the public 1 1 1 0
good are at odds with instructions of the elected
ofcial, I should follow his or her instructions.
24. I feel like a public administrator 24 hours a day. 21 1 22 21
25. I should be able to live with myself, which is why 1 0 2 1
being loyal to my own conscience is the most
important thing for me.
26. Loyalty to my minister can be less important than 21 0 1 0
serving the public interest.
27. Acting with integrity means for public administrators 1 21 21 0
that they act according to their conscience.
28. Public administrators who are strongly focused on 0 1 0 0
their own careers dont necessarily want the best for
their colleagues.
29. Punishable or reprehensible conduct in your private 2 1 1 1
life can sometimes be at odds with your duty to
behave as a good public administrator.
30. Sometimes you have to bend the rules a little when 23 0 1 21
dealing with societal partners of the organization.
31. When I disagree with a policy, I simply tell the 21 22 23 23
concerned parties of our organization.
32. I believe that professionalism and loyalty to 3 0 2 2
professional rules are the leading values in my work.
33. I will not easily betray my colleagues, even when it 21 0 0 21
would bring me into deep trouble.
34. When I nd an assignment irresponsible, loyalty to 2 1 3 2
my own conscience and identity is the deciding factor.
35. I feel a stronger connection with government in 0 2 21 0
general than with my own ministry.
36. A top public administrator should always be careful 1 1 0 0
not to express him- or herself publicly if the minister
dislikes such behavior.
37. The rules that I have to follow in dealing with 22 21 21 0
stakeholders lower my efciency and effectiveness.
38. As a top public administrator, I feel strongly 22 21 22 21
connected to other ABD members.
39. The disqualication of any educated opinion 0 23 1 1
including public administrators cannot be
tolerated in a modern democracy.
40. Political ofcials often have an interest in the short 0 2 2 2
term only, which leads to ad-hoc decision making. It
is the task of public administrators to also look at the
long run.
41. Public administrators should not base their decisions 21 1 0 3
on personal values because in such a case citizens
can no longer trust that policies and rules are
followed uniformly.
42. Good personal relations with my superiors will make 0 1 0 0
me more committed to my work.
304 Journal of Public Administration Research and Theory

REFERENCES
Aberbach, J. D., R. D. Putnam, and B. A. Rockman. 1981. Bureaucrats and politicians in Western
democracies. Cambridge, MA: Harvard Univ. Press.
Adams, G., and D. Balfour. 1998. Unmasking administrative evil. London: Sage.
Andeweg, R. B., and G. A. Irwin. 2005. Governance and politics of the Netherlands. Houndmills, UK:
Palgrave Macmillan.
Arendt, Hannah. 1994. Eichmann in Jerusalem: A report on the banality of evil. New York: Penguin
Group.
Barnard, Chester Irving. 1938. Functions of the executive. Cambridge, MA: Harvard Univ. Press.
Bauman, Zygmunt. 1991. Modernity and the holocaust. Ithaca, NY: Cornell Univ. Press.
Birch, A. H. 1975. Economic models in political science: The case of exit, voice and loyalty. British
Journal of Political Science 5:6982.
Bovens, Mark. 1998. The quest for responsibility. Cambridge, UK: Cambridge Univ. Press.
Bovens, Mark, Thomas Schillemans, and Paul t Hart. 2008. New forms of accountability and EU
governance. Comparative European Politics 5:10420.
Bowman, James, and Russell L. Williams. 1997. Ethics in government. From a winter of despair to a spring
of hope. Public Administration Review 57:51726.
Brans, Marleen, Christophe Pelgrims, and Dieter Hoet. 2006. Comparative observations on tensions
between professional policy advice and political control in the low countries. International Review of
Administrative Sciences 72:5771.
Brown, Steven. 1980. Political subjectivity: Applications of Q-methodology in political science. New
Haven, CT: Yale Univ. Press.
. 1993. A primer on Q methodology. Operant Subjectivity 16:91138.
. 1997. The history and principles of Q methodology in psychology and social sciences. Kent, OH:
Kent State Univ. Available at: http://facstaff.uww.edu/cottlec/Qarchive/Bps.htm.
. 2000. Types of social organization: A conuence of Weber and Stephenson. Paper presented at
the meeting of the International Society for the Scientic Study of Subjectivity. Muncie, IN: Ball
State Univ.
Clarke, J., and J. Newman. 1997. The managerial state. London: Sage.
Cooper, T. L., ed. 1994. Handbook of administrative ethics. New York: Marcel Dekker.
Cooper, Terry. 1998. The responsible administrator. San Francisco, CA: Jossey-Bass.
de Graaf, G. 2010. De loyaliteiten van gemeentelijke topambtenaren. Bestuurswetenschappen 64 (2):
3657.
de Graaf, G., and L. Huberts. 2008. Portraying the nature of corruption. Using an explorative case-study
design. Public Administration Review 64:64053.
de Graaf, G., and Z. van der Wal. 2008. On value differences experienced by sector switchers.
Administration & Society 40:79103.
. 2010. Forthcoming. Public sector professionals. Empirical evidence from Dutch Municipalities.
In Professionals under pressure. Perspectives on professionals and professionalism in public service
delivery, ed. Steijn, B., Noordegraaf, M. Amsterdam Univ. Press.
de Graaf, Gjalt. 2003. Tractable morality. Customer discourses of bankers, veterinarians and charity
workers. Rotterdam, the Netherlands: Erim.
. 2005a. Tractable morality. Journal of Business Ethics 60:115.
. 2005b. Veterinarians discourses on animals and clients. Journal of Agricultural and Envi-
ronmental Ethics 18:55778.
de Graaf, Gjalt, and Job van Exel. 2009. Using Q methodology in administrative ethics. Public Integrity 11
(1): 6378.
Denhardt, K. G. 1988. The ethics of public service, resolving moral dilemmas in public organizations.
Westport: Greenwood.
Denhardt, Robert B., and Janet Vinzant Denhardt. 2000. The new public service: Serving rather than
steering. Public Administration Review 60:54959.
Downs, A. 1967. Inside bureaucracy. Boston, MA: Little Brown.
 de Graaf Loyalties of Top Public Administrators 305

Dubnick, Melvin J., and Barbara S. Romzek. 1993. Accountability and the centrality of expectations in
American public administration. In Research in public administration, ed. J. L. Perry, 3778.
Greenwich, CT: JAI Press.
Dunleavy, P. 1991. Democracy, bureaucracy and public choice. Brighton: Harvester Wheatsheaf.
Durant, R. F. 1991. Whither bureaucratic inuence?: A cautionary note. Journal of Public Administration
Research and Theory 1:46176.
Ewin, R. E. 1992. Loyalty and virtues. The Philosophical Quarterly 42:40319.
. 1993. Corporate loyalty: Its objects and its grounds. Journal of Business Ethics 12:38796.
Fletcher, G. P. 1993. Loyalty: An essay on the morality of relationships. Oxford: Oxford Univ. Press.
Frederickson, H. George, ed. 1993. Ethics and public administration. Armonk, NY: M.E. Sharpe.
Fry, Brian R., and Lloyd G. Nigro. 1996. Max Weber, and US public administration: The administrator as
neutral servant. Journal of Management History 2:3746.
Golden, Marissa Martino. 1992. Exit, voice, loyalty, and neglect: Bureaucratic responses to presidential
control during the Reagan administration. Journal of Public Administration Research and Theory
2:2962.
Graham, Jill W., and Michael Keeley. 1992. Hirschmanns loyalty construct. Employee Responsibilities
and Rights Journal 5:191200.
Hirschman, A. O. 1970. Exit, voice and loyalty: Responses to decline in rms, organizations, and states.
Cambridge, UK: Cambridge Univ. Press.
Hoffmann, E. A. 2006. Exit and voice: Organizational loyalty and dispute resolution. Social Forces
84:231230.
Jeurissen, R. 1997. Geen organisatie zonder loyaliteit. Filosoe & Praktijk 18:16982.
Kaufman, H. 1960. The forest ranger. Baltimore, MD: Johns Hopkins Univ. Press.
Lee, Kwang-Hoon, and Jos C. N. Raadschelders. 2008. Political-administrative relations: Impact of and
puzzles in Aberbach, Putnam, and Rockman, 1981. Governance 21:41938.
Lee, Soo-Young, and Andrew B. Whitford. 2008. Exit, voice and loyalty, and pay: Evidence from the
public workforce. Journal of Public Administration Research and Theory 18:64771.
Light, Paul. 1999. The new public service. Washington, DC: Brookings Institution.
Lijphart, Anton. 1999. Patterns of democracy. New Haven, CT: Yale Univ. Press.
Lipsky, Michael. 1980. Street-level bureaucracy: Dilemmas of the individual in public services. New
York: Russel Sage Foundation.
Long, Norton E. 1952. Bureaucracy and constitutionalism. American Political Science Review 46:80818.
Maesschalck, Jeroen. 2004. The impact of new public management reforms on public servants ethics:
Towards a theory. Public Administration 82:46589.
Meier, K. J. 1993. Politics and bureaucracy: Policy making in the fourth branch of government. Pacic
Grove, CA: Brooks/Cole.
Nelson, Benjamin. 1949. The idea of usury. Form tribal brotherhood to universal otherhood. Princeton,
NJ: Princeton Univ. Press.
Nieuwenkamp, Roel. 2001. De prijs van het politieke primaat: Wederzijds vertrouwen en loyaliteit in de
verhouding tussen bewindspersonen en ambtelijke top. Delft, the Netherlands: Eburon.
Noordegraaf, M. 2004. Management in het publieke domein. Issues, instituties en instrumenten. Bussum,
the Netherlands: Countinho.
Noordegraaf, Mirko, Martijn van der Steen, and Paul Frissen. 2006. Andere ambtenaren. Of en hoe andere
ambtenaren ontstaan. Bestuurskunde 15 (3): 29.
OKelly, Ciaran, and Melvin J. Dubnick. 2005. Taking tough choices seriously: Public administration and
individual moral agency. Journal of Public Administration Research and Theory 16:393415.
OLeary, Rosemary. 2006. The ethics of dissent. Managing Guerrilla government. Washington, DC: CQ
Press.
Perry, James L. 1996. Measuring public service motivation: An assessment of construct reliability and
validity. Journal of Public Administration Research and Theory 6:522.
Peters, B. Guy. 1997. Bureaucrats and political appointees in European democracies: Whos who and does
it make any difference? In Modern systems of government, ed. A. Farazmand, 23255. London: Sage.
. 2001. The politics of bureaucracy. London: Routledge.
306 Journal of Public Administration Research and Theory

Peters, B. Guy and J. Pierre, eds. 2001. Politicians, bureaucrats and administrative reform. London:
Routledge.
Petter, J. 2005. Responsible behavior in bureaucrats: An expanded conceptual framework. Public Integrity
8:197217.
Pollitt, C., and G. Bouckaert. 2000. Public management reform. Oxford: Oxford Univ. Press.
Rosenthal, U. 1999. De politiek-ambtelijke rechtsstaat. Over de tweezijdigheid van de politiek-ambtelijke
verhoudingen. Liberaal Reveil 40:14651.
Rourke, F. E. 1991. American bureaucracy in a changing political setting. Journal of Public Adminis-
tration Research and Theory 1:11129.
Royce, Josiah. 1908. The philosophy of loyalty. New York: Macmillan.
Selden, Sally Coleman., Gene A. Brewer, and Jeffrey L. Brudney. 1999. Reconciling competing values in
public administration. Understanding the administrative role concept. Administration & Society
31:171204.
Shumavon, D. H. and H. K. Hibbeln, eds. 1986. Administration discretion and public policy imple-
mentation. New York: Praeger.
Stephenson, W. 1953. The study of behaviour: Q-technique and its methodology. Chicago: Univ. of
Chicago Press.
. 1962. Ideal types. Psychological Record 12:916.
Stoker, K. 2005. Loyalty in public relations: When does it cross the line between virtue and vice? Journal
of Mass Media Ethics 20:26931.
Suleiman, Ezra N. 2003. Dismantling democratic states. Princeton, NJ: Princeton Univ. Press.
Svara, James H. 1998. The politics-administration model as aberration. Public Administration Review
58:518.
. 1999. The shifting boundary between elected ofcials and city managers in large council-manager
cities. Public Administration Review 59:4453.
. 2001. The myth of the dichotomy: Complementarity of politics and administration in the past and
future of public administration. Public Administration Review 61:17683.
t Hart, Paul and A.W. Wille, eds. 2002. Politiek-ambtelijke verhoudingen in beweging. Amsterdam, the
Netherlands: Boom.
. 2006. Ministers and top ofcials in the dutch core executive: Living together, growing apart?
Public Administration 84:12146.
Thomas, D. B., and L. R. Baas. 19921993. The issue of generalization in Q methodology: Reliable
schematics revisited. Operant Subjectivity 16:1836.
Toonen, Theo. 1990. The unitary state as a system of co-governance: The case of the Netherlands. Public
Administration 68:28196.
Vandekerckhove, W., and Ronald Commers. 2004. Whistle blowing and rational loyalty. Journal of
Business Ethics 53:22533.
van Exel, N. J. A., G. de Graaf, and W. B. F. Brouwer. 2006. Everyone dies, so you might as well have fun!
Attitudes of Dutch youths about their health lifestyle. Social Science & Medicine 63:262839.
Waldo, D. 1988. The enterprise of public administration. Novato: Chandler and Sharp.
Watts, S., and P. Stenner. 2005. Doing Q methodology: Theory, method and interpretation. Qualitative
Research in Psychology 2:6791.
Weber, Max. 1946. Politics as a vocation. In From Max Weber: Essays in sociology, eds. H. H. Gerth and
C. Wright Mills, 196244. Oxford Univ. Press.
Williams, Bernard. 1985. Ethics and the limits of philosophy. Cambridge, MA: Harvard Univ. Press.
Withey, Michael J., and William H. Cooper. 1992. Whats loyalty? Employee Responsibilities and Rights
Journal 5:23140.