Public Organization Review: A Global Journal 3: 3–27 (2003)

# 2003 Kluwer Academic Publishers. Manufactured in The Netherlands.

Administrative Reform Policy: The Challenges

of Turning Symbols into Practice1
TOM CHRISTENSEN [email protected]
Department of Political Science, University of Oslo

PER LÆGREID [email protected]

Department of Administration and Organization Theory and Rokkan-centre, University of Bergen

Key words: administrative reforms, symbols

Abstract
Administrative reform policies are often characterized by a large number of reform symbols. While
these are generally aimed at furthering the legitimacy of the political leadership, they can have the
opposite effect and generate problems in implementing reform practice. This paper starts by
discussing the theory of reform symbols and relates it to theories of instrumental, negotiation and
cultural features of reforms. We illustrate this discussion by analyzing interview data collected among
the central political and administrative elite in Norway. The analysis shows the relevance of reform
symbols even in Norway, a country with a reluctant reform tradition and few incentives for reform. The
respondents often perceived reform symbols as negative, something that creates problems for the
political leadership in reconciling symbols with practice and leads it to engage in ‘‘double-talk’’.
Political and administrative leaders on different levels have differing attitudes towards administrative
reform symbols, reflecting different roles and perspectives. The conclusion is that symbols are a main
feature of administrative reform but also that reforms are not only symbols. Symbols matter, but
turning symbols into practice is not an easy task to do.

Introduction

Administrative reform policies around the world generally have a strong

symbolical flavor. In contrast to the substantive outcomes of policy-making
processes the symbolic behavior are processes of interpretation and of
conveying signals, moods and impressions. They are opportunities for public
statements and appeals and for exhibiting proper behavior (Lægreid and Olsen,
1986). The predominance of symbols in administrative reform policy is related to
the general and indirect focus on processes, personnel and structures in this
policy field, in contrast to the more substantive policy areas, which normally have
a more distinct, specific and direct focus. Administrative reform policies and
programs are often presented with hype, rituals, myths, ceremonies, metaphors,
and rhetoric of norms and values, surrounding them, all central features of
symbols (Hood, 1998; March and Olsen, 1989; Power, 1997). They are portrayed
4 T. CHRISTENSEN, P. LÆGREID

as fresh starts along promising new paths that will yield substantial improve-
ments. This tendency is understandable in an increasingly complex world, where
the connection between reform policies and practice may be fairly ambiguous.
While the use of symbols does have the potential to arouse support for the
reforms, they may also create opposition and conflicts, and may lead nowhere in
an instrumental sense because they are built on sand. Since leaders through
administrative reform policies often promise action, they must at least turn some
reform symbols into practice.
The aim of this article is two-fold. First, we will discuss in a theoretical
framework what it means to have administrative reform policies dominated by
symbols and how this perspective relates to other analytical perspectives on
public reform. What are the typical expressive features of administrative reform
symbols? How can the dynamic existing between symbols and various
constraints on reform processes give us new insight into reform implementation
and practice? Can symbols be seen as substitutes for implementation or do they
assist in reform practice? Second, what are, empirically the main preconditions
for and challenges of turning administrative reform symbols into practice? What
is the role of political and administrative leaders in reform policy processes
dominated by reform symbols? What is the significance of the way such
administrative reform policy processes are organized? And how does the actual
content of administrative reform policy relate to the use of symbols?
To answer this second set of questions we will apply the theoretical framework
and analyze the findings of elite interviews conducted with 58 political and
administrative top leaders in Norway in late 2000. The survey encompassed
respondents from the Bondevik I cabinet (in power from 1997–2000), secretary
generals in the ministries, agency leaders and directors of state corporations.
We will first outline a symbolic perspective on administrative reform policy and
relate it to cultural, instrumental and bargaining perspectives. Second, we describe
the Norwegian public reform efforts and illustrate the symbolic features of the
reforms by reporting the responses of central political and administrative leaders.
Third, we discuss the empirical findings in relation to the perspectives outlined
and conclude by underlining the challenges of turning symbols into practice.

Reform policy symbols—an analytical discussion

Presenting or packaging public reforms using a set of positive normative

concepts is nothing new, whether it involves presenting new rational budget and
economic reforms (Downs and Larkey, 1986) or broad public reforms following
the assumption of office by a new political leadership (March and Olsen, 1983). In
this respect reform language is part of the symbols of political language
(Edelman, 1964), and symbols are, in turn, a vital and important part of political-
administrative processes (Olsen, 1983). Political leaders hope to attract support
for reforms by labeling them in certain positive ways (Feldman and March, 1981:
ADMINISTRATIVE REFORM POLICY 5

177–179; M. Meyer, 1979). Defining reforms as modern and rational often

arouses hopes of increased effectiveness and efficiency in practice (March,
1986, p. 30), for people tend to like political and administrative leaders who
express they can solve societal problems. In reality a political leadership that
claims to do this is usually engaged in a subtle and often tricky act of reconciling
symbolic promises of reform with ‘‘real’’ reform action, often resulting in ‘‘double-
talk’’ and ‘‘hypocrisy’’ (Brunsson, 1989; Meyer and Rowan, 1977). This division
between rhetoric and aspirations (a realm dominated by myths and symbols) on
the one hand and, on the other hand, action or practice has the potential to
increase the influence of leaders but it can also create problems for the leaders,
for civil servants expecting actions and for citizens wanting implementation
(March, 1986; Olsen and Peters, 1996).
The tendency in recent decades for administrative reform policies to become
more symbol-oriented than before may have many explanations. One is simply
that the world is growing more compounded, resulting in more complex political-
administrative apparatuses and public policies and producing politicians who are
more conscious of how reform symbols can be manipulated. One consequence
of this is increasing problems with the organizational or means-ends thinking
behind reforms (Christensen, Lægreid, and Wise, 2002). What are the appropriate
theories behind reforms (Boston et al., 1996)? How do they translate into practice
(Røvik, 1998; Sahlin-Andersson, 1996)? What are the proper problems to be
attended to and can they be sorted or ranked in a simple way? What are the
expected consequences and impacts of using certain reform instruments? Very
often such questions are not easy to answer, particularly when we are talking
about broad administrative reform policies that aim to transform entire political-
administrative systems (Pollitt and Bouckaert, 2000). Given these difficulties, it is
tempting for political leaders to use symbols, wishful thinking and superstitious
learning (March and Olsen, 1976: 57) to try to convince other elites and the
populace that they have full insight into the problems and are treading a clearly
defined and promising reform path. Added to this is the fact that the reform
mandate politicians receive from citizens appears to be increasingly ambiguous,
which hardly creates a solid basis for reform (Olsen, 1988). New Public Manage-
ment (NPM)-related reforms, for example, might be driven by popular dissatisfac-
tion with the inefficiency of the public sector, but citizens and even political parties
often express inconsistent attitudes on this subject (Forster and Snyder, 1989;
Rainey, 1996). They would like to increase efficiency and lower taxes but at the
same time keep institutions, policies and services intact or even raise them to a
higher level. Such attitudes nurture an increased use of reform symbols.
Another aspect of this increasing complexity that contributes to a greater use
of symbols is that actors involved in reforms may be more difficult to control than
before (Olsen and Peters, 1996: 11). The general decline in the authority and
legitimacy of political leaderships combined with greater devolution, decentral-
ization and delegation make the implementation of reform via unambiguous
hierarchical control more problematic than before. Administrative leaders,
6 T. CHRISTENSEN, P. LÆGREID

leaders of commercial state units and private commercial actors all have more
influence than before (Christensen and Lægreid, 2001a); governments and public
agencies increasingly lean on professional expertise and consulting firms when
engaging in public reform; and interest and ad hoc groups have become more
active. What is more, the general increase in the knowledge basis of the public
and its greater access to information impinging on reform processes has the
potential to undermine leadership control, while increased complexity may also
mean that reforms become more path-dependent over time and between
institutions and levels, again constraining hierarchical control (Olsen, 1988). The
often ambiguous balance and tension between increased complexity in reform
processes and policies, peoples’ increasing knowledge about policies and their
symbolically oriented instrumental expectations towards the political leaders,
potentially make the leaders’ role enactment more problematic.
The increased globalization of reform ideas and measures also enhances the
symbolic aspect of reforms (Spybey, 1996). Some hold that public sectors all
over the world are experiencing the same type of problems and that there are
some ‘‘technical’’ global remedies for these, like NPM, which focus on efficiency.
Another viewpoint, and the one emphasized here, is that the number of
international institutions functioning as symbolic reform entrepreneurs is
increasing (Sahlin-Andersson, 2001). Not least important is the emergence of
transnational networks or discourse communities of people exchanging ideas
about perceptions, solutions and methods in the field (Bislev and Salskov-
Iversen, 2001), in particular those related to the UN complex of organizations (the
UNDP, the World Bank and the IMF), to the OECD (e.g., PUMA) but also a myriad
of networks of academics, practitioners and ‘‘think tanks’’. These networks
‘‘sell’’ reform packages to different countries expressing symbols and they use
the success of these packages to increase their own status. Administrative
reforms are thus enhanced through the transnational exchange of ideas
(Marcussen, 2001). Part of the success of this process is that political and
administrative leaders in some countries take these symbols on board and report
reforms in their own countries as success stories. Such mutual affecting
processes may produce a symbiotic relationship between national governments
and reform ‘‘entrepreneurs’’, as has happened in Australia and New Zealand, for
instance (Aberbach and Christensen, 2001; Sahlin-Andersson, 2001).
The background to making reform symbols global is that the reform elements
have to be decontextualized (Røvik, 1996)—i.e., defined as having an unlimited
reach and application—so as to ‘‘fit’’ countries with a wide variety of historical
traditions and political-administrative structures. Different countries thus develop
isomorphic features (structural similarity) because the institutional environment—
consisting of global reform symbols—increases the pressure to reform. In other
words, while the reform features adopted are apparently similar, in reality there is
a relatively loose coupling between reforms and effects on behavior (Brunsson,
1989). Administrative reforms are viewed as symbolic responses to environ-
mental expectations. When goals are ambiguous and organizational technology
ADMINISTRATIVE REFORM POLICY 7

poorly understood or when the environment creates symbolic uncertainty, as is

often the case in administrative reforms, the reformers model their reforms on
other organizations perceived as more legitimate or successful (DiMaggio and
Powell, 1991; March and Olsen, 1976). Public sector organizations change
because they want to appear ‘‘modern’’ and the result is a trend towards
convergence in the way policy-makers talk about reform. But the relationship
between this new way of talking about administrative reforms and actual change
in practice is tenuous at best (Premfors, 1998).
The advantages of using broad symbols in administrative reform policies
are obvious. First, when an organization faces a problem with unclear causes
or ambiguous solutions, symbolic reforms may yield few costs in the short
term. Second, they increase the legitimacy basis of the political leadership,
adding to the legitimacy that stems from instrumental achievements. People
believe or would like to believe what political leaders tell them about the
rationality and potential of public reform (March, 1986). Moreover, people
take it often for granted, at least in the beginning, that plans and talk are the
same as reality or practice, without being willing or able to control what is
actually happening, something that is indeed very difficult in a complex
political-administrative system. Brunsson (1989) emphasizes that for the
leadership to succeed in this endeavor it needs to engage in double-talk. It
has to remain distant and decouple from the ‘‘production system’’, where
practice is happening, and stick to the broad reform symbols. It can
emphasize talk in certain periods and action in others, structure the internal
organization into units that attend to talk and action respectively, or respond
to some parts of the environment with talk and others with instrumental
action. This shows that symbols and talk related to reforms must be
balanced against action.
All this suggests that administrative reform policies might belong more to the
world of symbols than to the world of practice (Brunsson and Olsen, 1993). They
may thus affect people’s image of a public agency even if the performance of that
agency does not change. It is often easier to influence the way outsiders perceive
a public organization than to alter the views of its members.
From a symbolic perspective administrative reform is important for the
development of meaning and the interpretation of experiences. Thus, an
important task for political leaders is to launch reform symbols, formulate visions
and ideas about the public sector and talk about what represents relevant
questions and proper solutions. From this point of view, the reforms can be
considered to be opinion building (Lægreid and Roness, 1999). They represent an
attempt to assert control over bureaucracy, with political leaders sending the
message that the governing are in control while administrators are able to
communicate that they are responding to the politicians and doing something
about perceptions of lagging performance (Perry, 1992; Tolbert and Zucker,
1983; Pollitt and Bouckaert, 2000). In sum, administrative reform policies, seen
from a purely symbolic perspective, is part of the ritual and myth that helps to
8 T. CHRISTENSEN, P. LÆGREID

preserve the legitimacy of the system of governance, but it may have few
consequences for performance.
There are also potential disadvantages in symbol-oriented administrative
reform policies. Political leaders may be seen as having ambiguous plans and as
trying to compensate for this by using symbols. They may be accused of
promising more action and implementation than they deliver or they may
encounter problems with subordinates who cannot distinguish between talk and
action and take symbols at face value and look for ‘‘real’’ reform (Brunsson,
1989). All this can potentially create problems both within the public apparatus
and in its relationship with the outside world, eventually undermining the
legitimacy of the political leadership, depending on how much they are leaning on
reform symbols.
The theoretical perspective laid out raises several questions when confronted
with empirical data. The most problematic one concerns when a reform or reform
policy is symbolic and when it is not. Generally, we see reform policies as
encompassing more or less symbols, i.e., characterized by different degrees and
types of symbols, and concentrate on when symbols might dominate reform
policies. An indirect answer to the question is that the symbolic elements of
reforms will be enhanced when certain preconditions are fulfilled. Generally we
will expect that reform policies tend to be more dominated by symbols in
situations where the meaning of reform is more important than the substance of
reform and where reform content and meanings are loosely coupled (Bolman and
Deal, 1987). In ambiguous reforms with unclear goals and uncertain technologies
the symbolic elements will be most obvious. Reform policies will be more
dominated by symbols when reforms are broad and comprehensive aiming at
changing large and complex structures or systems; complex policies and
practices; when tasks and functions are difficult to operationalize and the
knowledge basis in a policy area is rather weak; and when the policy field has a
tradition of conflicts and strong emotions running. A more direct answer would be
to make an analysis of the arguments of political and administrative leaders in
reform policies. If the organizational thinking is characterized by ambiguous
goals, means and effect expectations, combined with strong expressive efforts to
portrait the reform policies as solving several problems, something that quite
easily can be seen as inconsistent and unrealistic, one can say that the reform
policies potentially may be dominated by symbols. Another direct answer is just
to ask the leaders whether they see the reform policies they are involved as
dominated by symbols. We combine these three ways of grasping reform
symbols in our analysis.
It’s important to stress that we see the roles of political leaders as multi-
functional, containing many different elements, of whom the symbolic element is
focused in this analysis. We also assume that not all reform policy symbols are
translated into programs, projects or activities that eventually lead to practice
and effects, neither that reform policy symbols and practice are completely
decoupled. This leaves the political leaders facing the problems of turning some
ADMINISTRATIVE REFORM POLICY 9

reform symbols into practice. Sometimes symbolic actions are viewed as

falsification and manipulation. Image crafting substitutes for performance and
policy benefits. But the relevance of symbolic behavior for the study of
administrative reforms is much more comprehensive. Symbolic actions can
contribute to find meaning and order in confusing and ambiguous situations, and
we should not underestimate the importance of the indirect influence on the
administrative practice through influencing believes, normative structures and
interpretations. And we, of course, realize that there are many other elements that
are influencing implementation and practice in policy fields than reform symbols,
like the direct substantive outcome of the reform processes and also broader
ongoing change processes. The focus here is, however, on reform policy
symbols and their coupling to practice.
The reality of administrative reforms probably lies somewhere between
symbolic interpretations and cultural and instrumental perspectives, so we will
now turn to the links between administrative reform symbols, cultural features,
instrumental implementation and bargaining.

Reform symbols and cultural, instrumental and bargaining perspectives

According to a cultural-institutional perspective, the evolution of political-

administrative organizations is gradual and incremental, is based on traditions
and creates certain informal normative features that give an organization a
distinctive institutional character or ‘‘soul’’ (Selznick, 1957). These features are
seen as constraints on the reform processes and policies, insofar as reforms that
are compatible with an organization’s cultural traditions are more often furthered
or accepted than those that are incompatible (Brunsson and Olsen, 1993).
What is the relationship between reform symbols and the historical-cultural
features of political-administrative systems? Will cultural features influence the
coupling or decoupling of symbols and reform action? One view precedes from
the assumption that reform symbols emerge from and feed on the cultural
features of public institutions (Christensen, 1991). Symbols help to show, in a
vivid way, what the main cultural trajectory of a system is and hence the limits of
reform. They embody the appropriateness of reform (March, 1994: 58, 101–102).
This may seem paradoxical, since reforms can also change historical traditions
and the culture of a public organization, but this contextualization of reforms
through symbols may often mean that institutional features are developed or
‘‘stretched’’ to accommodate the need for change (Røvik, 1996). Public
organizations are not rigid, and this is one way that they can develop or
gradually change.
Another view stresses that reform symbols are needed to fight cultural
traditions in political-administrative systems—without changing the culture you
can’t change the practice. Symbols are appropriate vehicles for exaggeration
and can be used to convey the message that existing cultural features are old-
10 T. CHRISTENSEN, P. LÆGREID

fashioned while reforms are modern, innovative and rational (Christensen, 1994;
March and Olsen, 1983). Public organizations are often defined as being in a
state of crisis, either economically or concerning legitimacy, or both. In such
instances symbols are used to demonstrate that reform is unavoidable, i.e., to
show there is simply no alternative (Christensen and Lægreid, 2001). In this
respect reform symbols can be employed to ‘‘reprogram’’ public institutions at
‘‘critical junctures’’, i.e., to establish a new starting point of a trajectory (Aberbach
and Christensen, 2001), or else they may have the more modest function of
changing the attitudes of political and administrative leaders towards reform.
According to an instrumental perspective, an active administrative reform
process and policy are characterized by careful planning, relatively unambiguous
goals, a clear definition of alternatives, insight into the consequences of various
reform measures, a conscious choice between alternatives and effects that
coincide with the main goals (Christensen, Lægreid, and Wise, 2002). Political
and administrative leaders are supposed to dominate the reform process, decide
on how to organize it, control participation, define mandates, include relevant
problems and solutions, etc. This means that there will be a close connection
between reform symbols, reform decisions, reform implementation and reform
practice. Reform is thus seen to be both necessary and possible (March and
Olsen, 1983).
At first glance, the relationship between reform symbols and the instrumental
implementation of reforms may seem rather loose, but we will argue that symbols
may also have instrumental features. First, reform symbols can be seen as a kind
of substitute for reform implementation and action (Brunsson, 1989). Political and
other actors may have problems with plans, goals, instruments, or controls of
reform processes and therefore try to compensate by emphasizing the symbols
and myths of reform. Unable to handle the nuts-and-bolts of reform, they cover
up for this inadequacy with reform symbols that are ‘‘saturated’’ with positive
normative values. They aim to gain political support based on the reform symbols
and to have a broader political effect, expecting that people will care more about
the promises of reform policies than about finding out what really happens to
reform in practice. Symbols are viewed as manipulation and means to justify
decisions, mobilize support and quieting opposition and are thus important
features in the exercise of power (Edleman, 1967; Pfeffer, 1981).
Second, reform symbols and reform implementation can be seen as tightly
coupled in the sense that symbols facilitate the reform process and reform
practice. Political leaders may develop a kind of dual strategy; so that even when
they have relatively unambiguous reform plans they lend them additional support
in the form of reform symbols. These symbols may persuade other actors to
accept the reform policies more easily, either because they accept the authority
of the leaders or would like to be associated with them, or because the symbols
make the rationale of the reforms more understandable. Either way the symbols
improve the chances of implementation. The strategy of coupling reform symbols
and reform practice can be facilitated when reform practice itself produces
ADMINISTRATIVE REFORM POLICY 11

symbols (March, 1986: 19, 33–34). Thus, it may be easier to accept reform
symbols and rhetoric if leaders can refer directly to similar successful reform
experiences, either at home or abroad.
Third, reform policy symbols may have a direct impact on reform implementa-
tion and practice, rather than functioning solely as normative constructions.
Reform practice can use reform symbols as guidelines, for instance, and symbols
may provide the impetus to change or adjust reform practice (March, 1986: 25).
Because symbols are often fairly abstract, they can pave the way for more
standardization in reform practice. Some of the theories of the new institution-
alism are preoccupied with how reform symbols as reform ‘‘prescriptions’’ are
related to reform practice (Røvik, 1998). Reform prescriptions, as institutionalized
standards, can, when entering an organization, be adjusted to the practice but
still influence reform action. They can, either as a whole or in parts, be translated
or edited so that they can meld with existing practice and eventually create new
reform practice. Or else reform prescriptions can function as viruses, i.e., their
influence on practice will show up only after a certain time lag, so that short-term
failures turn out to be long-term successes (March and Olsen, 1983: 287). The
virus can over time break down the organizations’ defence mechanisms, defeat
the institutional blocks and infuse the organization (Helgesen, 2001). One aspect
of this is a phenomenon that March and Olsen (1983: 288) label ‘‘civic
education,’’ i.e., political leaders may present reform ideas that are not
immediately accepted but in the long run will have an influence on practice
once actors have warmed to and learned more about them.
According to a negotiation perspective, or what March and Olsen (1983:
283–284) label Realpolitik, reform processes will be characterized by negotiations
between different actors in heterogeneous systems. Public organizations are
quite often heterogeneous concerning interests, beliefs, values, structure and
demography, as is their environment, and this too will influence the reform
process and policy. The hierarchical authority of political and administrative
leaders is not easy to exert under such conditions, for there is often disagreement
about the definition of problems and solutions, and compromises have to be
found.
How may heterogeneity and negotiations influence the connection between
reform policy symbols and reform practice? One could argue that symbols may
help facilitate the implementation of reforms under conditions of heterogeneity.
Symbols can help to build bridges between differing interests, because they are
rather general and are approaching conflicts by developing shared values and
negotiate meaning (Bolman and Deal, 1987). Political leaders can quite
deliberately direct different types of reform symbols at different actors with the
aim of bringing them on board, i.e., there are potentially many symbols for the
same reform policy (Christensen, 1994). They can appeal to administrative
leaders with symbols of continuity; to various interest groups with promises of
participation and influence; and to the media and public in general with promises
of innovation, rationality and efficiency.
12 T. CHRISTENSEN, P. LÆGREID

But reform processes can quite easily take the form of a tug-of-war between
actors employing different reform symbols—what one might call a confrontation
between competing types of appropriateness (Christensen and Røvik, 1999:
165)—potentially creating conflicts and effectively preventing change and
reform activities. A cabinet may be comprised of members with differing reform
rhetoric; ministries and agencies may have quite different opinions on reform and
support these with symbols; and interest groups—civil servants’ unions
included—may use symbols to fight the reform plans of the political
leadership. In reform bargains there is an inbuilt tendency for the politicians to
delegate blame but not credit and for bureaucrats to accept credit but not blame
(Hood, 2001).

Administrative reform policies in Norway—symbolic features

We would expect administrative reforms to be a mixture of symbolic features,

cultural and instrumental components and bargaining elements. The symbolic
flavor will normally be less dominant in administrative reforms that are narrow in
scope, less ideological and more pragmatic and incremental. This makes Norway
an interesting case, since it was a late and reluctant reformer in the New Public
Management movement (Olsen, 1996). Although reform efforts have gathered
more pace during the last five years, Norway is a deviant case compared with
most other western OECD countries, and we can thus use Norway as a ‘‘critical’’
case for studying symbols in administrative reforms. Our basic assumption is that
symbols will not be of major significance in Norwegian administrative reforms,
owing to the country’s particular reform style, scope, intensity and tempo. Thus, if
we find strong symbolic elements in this reform context then they should be even
stronger in other settings with more radical, comprehensive and ideological
reforming traditions.

The content of comprehensive reform programs

Over the last two decades there have been few reasons for reform in Norway and
the driving forces behind reform have been relatively weak (Christensen,
Lægreid, and Wise, 2001). Despite this, Norway experienced more reform
activities in the 1990s, particularly sector-specific reforms involving the
devolution of public enterprises (Christensen and Lægreid, 2001b).
Discussion of NPM reforms began in Norway in the mid-1980s, but the
Norwegian reform programs then were more a loose collection of on-going
reform measures and new reform ideas than a consistent, coordinated and
unified strategic plan for changing the administrative apparatus (Lægreid and
Roness, 2003). The rhetoric in the modernization programs reflected international
trends in administrative policy, but it was not very specific concerning reform
ADMINISTRATIVE REFORM POLICY 13

measures. The reform programs promised better service, better utilization of

resources, higher productivity, more frequent attainment of goals, improved
quality, better working conditions for employees, improved political control and
extended user participation. Even though a vision of market and management-
based practices was presented in administrative policy documents, its
implementation remained optional (Christensen and Lægreid, 1998; Naschold
1996: 66). The assumption was that the pursuit of economic performance should
not affect negatively other values like democratic representation, a state based
on the rule of law and professional quality. Concerns about the implications of the
reforms for democracy and inequality modified the reform efforts (Lægreid,
2001a).
The reform programs of various Norwegian governments thus exhibited more
similarities than differences. They were pragmatic and cooperative rather than
ideological and confrontational, and they were directed more at enhancing
internal productivity and increasing efficiency than at rolling back the state.
Reform strategies tended to be sector-based rather than comprehensive. The
reform style has been maintenance, adjustment and supplementation more than
radical change. Some elements, like the management components of NPM, have
been implemented to a greater degree than the market elements (Christensen
and Lægreid, 1998).
The first NPM-related reform to be implemented in the central civil service was
Management by Objectives and Results (MBOR), which was made mandatory in
1991. The aim of this reform was to strengthen political control by making goals
and means less ambiguous, focusing on results, introducing a monitoring system
and making greater use of incentives. MBOR has been further developed in
recent years and been more broadly applied. It is now more closely connected to
the state budget and audit system, less rigid and more adapted to the special
features of different public organizations. Performance auditing has been
developed and strengthened since the mid-1990s (Christensen, Lægreid, and
Roness, 2002) and a comprehensive model for performance management was
installed into the new Government Financial Regulations in 1996 (Helgesen,
2001). Norway also introduced a pay-for-performance system for top adminis-
trative leaders in the early 1990s. This system was implemented more reluctantly
than anticipated, partly because of cultural resistance (Lægreid, 2001).
Traditionally, state commercial units, like the government enterprises in the
communications sector, have been relatively tightly controlled in Norway. During
the last 10 years, about 50 units changed their organizational status in the
direction of more devolution, which accounts for new forms of state-owned
corporations with more market competition and commercial freedom and less
political control (Christensen and Lægreid, 2001b). Still, there has been
reluctance to privatize, although Norway has experienced some recent break-
throughs. Telenor, the national telephone company, sold 17% of its shares to
private investors in 2000, and Statoil, the large public oil company, was also
partly privatized in 2001.
14 T. CHRISTENSEN, P. LÆGREID

There are thus important distinctions between the Norwegian reform programs
and the main trends in the international administrative policy reform movement
(Olsen, 1996). First, visions of the decline of the public sector through
privatization scarcely featured in the Norwegian reform programs (Christensen
and Lægreid, 1998). Second, international criticism that the growth of powerful
special interest organizations had prevented political institutions from addressing
broader national concerns did not receive much support in Norway. Third, the
changes in practice turned out to be more pragmatic and more constrained by
established administrative cultures than in many other countries. In other words,
Norway has chosen a moderate reform path, trying to balance devolution with
central political control and making only moderate changes in the execution of
central political and administrative functions (Christensen and Lægreid, 2001a).
This makes Norway a deviant case, with increasing public expenditure, growing
numbers of public employees and a comparatively low profile for NPM discourse
and initiatives. This pattern should be seen in the light of Norway’s current
economic well being, a strong welfare state tradition with a firm footing among
the electorate, strong public-sector trade unions and polity features resulting in
weak minority governments. A large number of small steps and more accelerated
reform efforts during the last five years have, however, produced a public sector
that is significantly more flexible and more complex than 15 years ago and
encompasses management models from the private sector that supplement the
traditional Weberian-inspired public administration model.
A mixture of symbols and rhetoric, on the one hand, and pragmatic and
instrumental means and measures, on the other, have characterized the reform
programs in Norway. The three latest programs illustrate this. The Bondevik I
(center) government’s program in 1999 of simplifying Norway included three main
initiatives. First, simplifying government regulations in the business sector by
making rules and regulations less complicated to comply with. Second,
enhancing citizen- and user-oriented development of the public administration.
In order to make it simpler for the citizen to understand and communicate with
the relevant public offices the government decided that all public agencies
should have a service declaration by the end of the year 2000 (Stene, 2001).
Third, simplifying government regulations concerning the municipalities via more
decentralization and allowing more resources to be spent on service production
and fewer on reporting and compliance with central government regulations. The
catchwords were management, democracy, user-orientation, transparency,
readjustment and service.
The main ideas of this program were adopted by the Stoltenberg (Labor)
government that followed, which stated three targets for its program of 2000 for
innovation of the public sector in Norway. First, that organization and service
delivery should be based on user needs. Second, that resources must be
transferred from administration to service production and third, that there should
be a more effective use of resources and greater latitude at all levels. The
overriding aim of the program was the revitalization, renewal, adjustment and
ADMINISTRATIVE REFORM POLICY 15

increased efficiency of the public administration and its adaptation to the needs
of citizens and to economic considerations. Added to this was a long list of more
concrete sector-specific reform initiatives, like establishing ‘‘one-stop-shops’’ in
all municipalities; e-government; and the reorganization of the police adminis-
tration, the court system, the health care system and the universities. One
difference between the two programs is that the Bondevik I government to a
greater extent focuses on the public as citizens and on the development of the
democracy while the labor government emphasizes more the public as users and
customers and the need to improve the service delivery (Skivenes, 2001).
The current Bondevik II government (center-conservative) more aggressively
and ideologically oriented tries to further market principles and competition. Lead
by the minister of labor and government administration, a professor in economy,
the government emphasize more structural devolution of state-owned enter-
prises, including partial privatization, more clear division and organizational
separation of roles and functions (owner, regulator, purchaser, provider), more
out-sourcing of services and more delegation of authority to subordinate levels
and institutions. The slogan ‘‘From words to action’’ indicates a wish to weaken
the symbolic elements of administrative reform and to enhance the substantive
outcome.
Our interviews with the ministers of the Bondevik I government and top
civil servants were conducted 9 months after they had resigned and the
Stoltenberg government came into office. We will now describe their beliefs,
attitudes and experiences concerning the reform programs and administrative
policy.

Reform programs as symbols—general experiences

The first question posed to our elite respondents was whether, based on their
experience, they saw broad and comprehensive reform programs primarily as
symbols. This concerns primarily the content of reforms, but also the behavior of
central actors like the executive politicians. 80% of our elite respondents
reported that based on their experiences they saw the reform programs primarily
as symbols and the tendency was to view this as negative. They said it was
difficult to implement broad and far-reaching reform programs and that such
programs were often dominated by a rhetoric that used similar symbols for all
reform programs. They saw the reform symbols of changing governments as
arousing unrealistic expectations of actual reforms in the public apparatus. The
reform symbols were connected to policies and programs that had ambiguous
goals and unclear organizational thinking in general, that did not sufficiently
clarify the tensions between actors and interests and that did not appear to be
well co-ordinated. But 20% of the respondents spoke positively about the
administrative reform symbols, emphasizing that reform programs and their
symbols were one way of communicating political intentions in a complex world.
16 T. CHRISTENSEN, P. LÆGREID

They also stressed, however, that it was necessary to link symbols with more
specific reform measures.
As expected, there were differences among groups of respondents in their
attitudes towards and experience of reform symbols. Relatively speaking, about
twice as many cabinet members as secretary generals saw reform symbols
positively. Agency leaders and directors of state corporations were almost totally
negative towards reform symbols. Politicians are often the actors who formulate
the symbols, while the leaders of agencies and commercial units are often
preoccupied with reform practice and real changes and the secretary generals
are caught between a political and administrative logic.
About half the respondents emphasized that one crucial precondition for
succeeding with administrative reforms was that political leaders, both the
cabinet as a whole and individual ministers, should be strongly involved in the
reform process, something that seems increasingly to have been the case in
Norway during the last two decades (Christensen and Lægreid, 2001a). This
increased politicization of the reforms also seems to be the key to understanding
why the use of reform symbols has increased, i.e., the use of symbols reflects
increased participation by politicians facing greater challenges. 20% of the
respondents said political leaders often had the will to reform but that this will
often evaporated when they tried to actually make changes in the political-
administrative system, a phenomenon that is also connected to the relatively
frequent changes of cabinets. They also underlined that ministers often had
problems gaining an overview of reform processes and in trying to organize
reforms in a ‘‘holistic’’ fashion. In other words, despite political support for
reforms, cabinet members had attention and capacity problems (Olsen, 1983).
About one-third of the respondents said that politicians often were not that
interested in public sector reforms, because reforms seldom helped them to win
elections (c.f. March and Olsen, 1983) and attention given to reforms had to be
balanced with that devoted to other and more sector-oriented policy areas. They
also said politicians often had ambiguous motives for reforms and their goals
were not operative enough. It was difficult for them to implement comprehensive
reforms and their rhetoric tended to be overblown compared with the results.
Regarding differences in responses concerning the role of the politicians
among the groups, the ministers more often than the other groups see their own
role as problematic. The response of the secretary generals lay somewhere in the
middle, while the leaders of agencies and state corporations underlined the
importance of the participation and attention of politicians in the process. This
latter group said the reform interaction between political leaders and adminis-
trative units was crucial for success, and if problematic often led to more symbols
and less implementation. This could partly be remedied by political and
administrative reform entrepreneurship. The non-politicians among the respon-
dents tended to stress that the cabinet members lacked knowledge of the
background and history of the reforms and it was therefore important for the civil
service to carry this tradition.
ADMINISTRATIVE REFORM POLICY 17

Responses were also divided concerning the role of the civil service in reform
processes. Slightly less than half of the respondents said the bureaucracy was
often engaged in reform efforts and did not put up much resistance. They also
underlined that, given the relatively frequent changes of government in Norway
over the past few decades, it was the civil service that represented continuity in
the reform process. The success of the reforms was connected to the ability of
the civil service to anchor the reform programs to more specific reform measures,
something stressed particularly by agency and state corporation leaders.
The contrary view, and the one that had the support of a slight majority among
the cabinet members and secretary generals, was that the civil service is
generally not very eager to support changes and simplifications and that
bureaucrats more often support the principles of reform than specific reform
measures. It was also said that the civil service was not very good at organizing
and furthering reforms and that there were frequent turf battles inside and
between ministries.
Overall, the ministers gave the most diverse responses. They were also more
critical of the role of the civil service in the reforms than the administrative leaders
were of the role of the political leadership, while the agency and state corporation
leaders were the least critical of their own role in reforms.

Organizing the reform process—the significance for symbols

We now turn to the behavior of actors that, apart from the political and
administrative leadership in general, have special responsibilities for advancing
reforms. How do our respondents see their roles? Do these actors participate in
furthering the symbols of reforms? In Norway the administrative reform process is
traditionally sectorized, with responsibility for reforms assigned to each sector
ministry or agency, but it is also in some ways centralized, because the Ministry
of Labor and Government Administration (MGA) has a special obligation to plan,
organize and push forward comprehensive reform programs, an obligation that
has become more institutionalised over the last two decades. It is, however, fair
to say that this ministry is rather weak, compared with the sector ministries
(Lægreid and Roness, 2003). To complicate the picture, the Ministry of Finance is
a key player that is not always in tune with the MGA.
The views on the role of MGA in administrative reform policy were pretty
unanimous. About 70% of the respondents had a negative attitude towards the
role of this ministry in the reform process. There are number of reasons for this.
For one thing, the profile of the ministry, which is supposed to lead the reform
process, is seen as ambiguous, and it is accused of not being able to
distinguish between more and less important parts of administrative reform
policies. In practice the ministry has not succeeded in pushing reforms forward,
even though some of the ministers have shown a great political interest in doing
so. The ministry has very few instruments and resources for furthering reform,
18 T. CHRISTENSEN, P. LÆGREID

unlike the Ministry of Finance, and must to a great extent rely on ‘‘soft’’
symbolic means like guiding, coaching, consulting, appealing and advising to
change general values and believes about what is relevant administrative
problems and good solutions. It is also criticized for lacking competence, not
functioning as a reform role model and not following up on its own intentions.
Instead of furthering reforms, it is often seen as a roadblock in reform processes
and as a theoretical ministry of modernization with ‘‘appropriate’’ reform rhetoric
but a weak grip on practical reforms. Another point of criticism is that it is
unrealistic in often trying to push through broadly based reforms. The ministry
itself has been frequently reorganized, something that has a negative impact on
the continuity and implementation of reforms. Overall the ministry seems to
participate in giving administrative reform policies more symbolic features. The
cabinet members had a more positive view of this ministry than the secretary
generals.
The most negative view of the role of the Ministry of Finance in administrative
reform policy was expressed by a group of cabinet members. They saw the
ministry as rather short-sighted, often against reforms and changes and unable
to understand that one needs resources to implement a reform process. The
ministry was also seen as too efficiency-oriented, not very open to new ideas and
too preoccupied with detailed rule steering. Some agency and state corporation
leaders took a more modified attitude, stressing that the Ministry of Finance was
important for the reforms but was often too control-oriented and frequently came
into conflict with the other ministries over reform measures. The most positive
view of the ministry was expressed by some secretary generals, who underlined
that it was important for pushing forward the reforms and that this ministry
collaborated in reform efforts with the other ministries. They also said that the
MGA often made a mistake in not basing its reform proposals more closely on the
logic of the Ministry of Finance.
The role of the civil servants’ unions in administrative reform policy in Norway is
traditionally integrated and important (Olsen, 1983; Roness, 2001). About half of
the respondents had a mixed view of the unions, and they stressed that it was
important for the unions to participate, even if they were against the reforms,
because they were important in legitimating the change process. One-third of the
leaders saw the role of the unions as negative and obstructive, especially if the
reforms were not perceived to be in their interests, so that a major effort was
required to win their support. The rest revealed positive experiences with the
unions in the reform process and reported that the unions participated in a
constructive way, making implementation easier and more effective. Generally
the cabinet members had the most negative attitudes towards the role of the
unions and the secretary generals the most positive.
Summing up, the role of MGA is very much seen by the respondents as
furthering reform myths, based on overselling the reforms, lack of realistic
planning for implementation, ambiguous reform polices and problems with getting
supporting actors, like Ministry of Finance and the civil service unions, on board.
ADMINISTRATIVE REFORM POLICY 19

‘‘A simplified Norway’’

The main administrative reform effort of the Bondevik I government was labeled
‘‘A Simplified Norway’’, intending to make structures, rules and policies simpler,
more efficient and user-friendlier. This program was one of the main themes in
the Prime Minister’s New Year’s speech to the Norwegian people in 1999,
focusing particularly on service declarations for public agencies. Concentrating
on this reform program is a way of exemplifying some of the challenges of turning
reform symbols into practice. Can this program be seen as primarily symbol-
oriented? Did it deliver the promised effects? And what were the problems of
implementing it?
Only one of the ministers said that the program had actually resulted in a
simpler Norway, while about 40% of the cabinet members thought the effect of
the program had been close to zero regarding increased efficiency. They
stressed that the reform program consisted of a lot of hype and symbols, that it
used too many resources and that it was constantly producing more rules instead
of deregulation.
About half the cabinet members had a more mixed view of the program. While
admitting that Norway had not become simpler during their time in office, they
underscored that the program nonetheless contained many positive elements.
They pointed out that a number of important problems had been put on the
agenda that the government had not had time to finish and that their
government had succeeded with some of the more specific reforms. They
also emphasized that the rules in some policy areas had become simpler,
although new rules had also emerged. Transparency in the work of the civil
service had also increased. The secretary generals were more critical towards
this reform program than the ministers, with about two-thirds of them unable to
see any simplifying effects.
The views on the process surrounding the reform program showed some
interesting paradoxes and differences. About 40% of the respondents said the
political intentions behind the program had been very strong even if these had not
translated into much success. It was emphasized that this was a very important
reform for the cabinet. It was anchored in the Prime Minister’s Office and headed
by a committee of ministers. Regular progress reports appeared and the reform
was seen as increasing consciousness. This group of respondents also
underscored that the political leadership was not ambiguous about the intentions
and goals of the reform program.
Quite a different view of the process was expressed by a slightly smaller group
of ministers, who saw a number of factors as obstructing the reform program.
One was capacity problems and resistance by certain central cabinet members;
another was tension among the leadership and resistance from the administrative
leadership to simplifying and losing policy instruments. Another hindrance was
the tendency of the process to fragment, owing to the large number of actors
involved.
20 T. CHRISTENSEN, P. LÆGREID

A small middle group of ministers saw the role of the civil service as a mixture
of initiative and passivity. This group stressed that the administrative leaders took
reform initiatives in certain areas, were less enthusiastic in others, and even
ignored some of the most prominent symbols in other areas.
Summing up, when analysing this specific reform program, it seems evident
that it didn’t reach the main goals and it produced quite a lot of reforms symbols
both before it was realized and particularly after. Even though many of the
respondents think that the program lead to many positive processes and results,
they often realized that promising to simplifying Norway is a dangerous business,
particular when it turned out that the number of rules in fact increased somewhat
during the 212– years the program lasted.

Discussion

One of the most striking findings of our empirical data was that a large proportion
of the respondents believed that symbols had a prominent place in administrative
reform policy—both symbols of a general nature and those more specifically
connected to one particular reform program. The majority of the leaders also
viewed these reform symbols as negative and problematic for role enactment. In
addition, many of them thought reforms raised more problems of defining the
reform policy and program in an unambiguous way than of controlling the reform
processes, were critical of the ministry in charge of reforms and supported the
strong involvement of politicians in the reform process.
On other questions the responses were more varied, particularly those
concerning the role of the executive political leadership, administrative leaders
and civil servants’ unions in the reform process. There were also marked
differences in how different groups of respondents viewed administrative reform
symbols and roles. The politicians, employing a political logic, viewed both
reform symbols and their own role in the reform process slightly more positively
than other groups, while they had a more negative attitude to administrative
actors. Conversely, administrative actors, employing an administrative logic, saw
the symbols and to some extent the role of politicians in a negative light. They
were also generally more preoccupied with the operationalization of reform plans
and symbols and reform implementation.
An interesting paradox is that many respondents thought the involvement of
ministers was an important precondition for successful reforms and many
identified political involvement as a typical feature of the latest reform programs.
A high degree of involvement and dedication from political leaders is supposed to
enhance the fulfilment of reforms, even though the majority of the respondents
criticized the reforms as excessively laden with symbols and thus difficult to
implement. So the realities of successful reform processes might be more multi-
dimensional, i.e., the involvement of political leaders, as an ideal from a
hierarchical view on reform processes, is not enough.
ADMINISTRATIVE REFORM POLICY 21

The purpose of reform symbols is often to underline good reform intentions and
convince other actors that these intentions will be quite easy to put into practice,
thus endowing political actors with support and legitimacy (Brunsson, 1989;
March, 1986). Our main empirical results, however, tend to stress the challenges
and problems associated with using reform symbols, rather than showing that
politicians would have been worse off without using symbols. In other words, our
data seem to show a kind of dynamic ‘‘double-talk’’ syndrome in the reform
process. First, reform programs that have catchy slogans, even if there is no well-
developed reform strategy behind these symbols, attract the political leadership
in the executive. Added to this is the problem of ‘‘overselling,’’ meaning that the
reforms look much more impressive ex ante than ex post (Brunsson and Olsen,
1993). To gain support for reforms, reform agents normally have to promise more
than the reforms can fulfil. The result is raised expectations and subsequent
frustration when the reforms are put into practice.
Second, politicians soon discover that there is quite some distance to be
bridged between general ideas and actually planning and designing compre-
hensive reforms. They have problems with the definitional aspect of the process
(March and Olsen, 1976), i.e., in formulating clear intentions and in explaining
how to organize the implementation of those intentions, co-ordinate the reforms,
etc. This is understandable in a complex world, where the competence of
politicians is constantly being challenged, and where their roles seem to be more
detached from grass-roots politics and street-level bureaucracy than before. And
the more reform symbols are decontextualized, the vaguer the path to
implementation becomes.
Third, the problem of defining the reforms means that it is important for
political leaders to control the activities of other actors in the process,
particularly the closest and potentially most supportive ones (Aberbach and
Christensen, 2001). Our data show that this is problematic: the ministry in
charge is seen as not succeeding in its supporting role—because it is too often
preoccupied with reform theory and rhetoric—and the civil service and civil
servants’ unions are seen as only partially supportive to politicians. Added to
this is the fact that politicians tend to find other areas of policy more attractive
and more rewarding and are therefore rather part-time participants in the reform
process (March and Olsen, 1976, 1983; Lægreid and Roness, 1999). The
symbolic elements are, however, weaker and the substantive outcome more
obvious for the administrative reforms which are conducted by the Ministry of
Finance.
Fourth, the problems of definition and control increase the temptation for
politicians to use more symbols as a way of disguising the fact that they cannot
deliver. This, in turn, again raises public expectations of reform practice,
expectations that become ever more unrealistic, thus potentially creating further
political problems. In other words, what we see is the classical dynamic of
‘‘double-talk,’’ as described by Brunsson (1989), but in this case mainly the
politicians’ struggle to cope with this ‘‘hypocrisy’’, rather than the success story.
22 T. CHRISTENSEN, P. LÆGREID

Our empirical results indicate that cultural traditions are ‘‘stretched’’ to

accommodate reform symbols, but also that reform symbols are deliberately
used to challenge traditional administrative cultures. Although the Norwegian
reform style is characterized by pragmatism, cultural compatibility is still a
relevant issue (Christensen and Lægreid, 2001a). Other studies of Norwegian
administrative leaders show quite positive attitudes towards certain aspects of
New Public Management, e.g., Management by Objectives and Results (MBOR),
but adapt it to the tradition of rule-steering in a sort of ‘‘rule-oriented goal-
steering’’ (Christensen and Lægreid, 1998). In other words, certain parts of
reforms, the most compatible ones, are selected and adjusted to or melded with
traditional cultural trajectories.
Our data also show that civil servants are seen as the bearers of cultural
traditions, while the political leadership, which is often in office for a shorter
period of time, has problems fulfilling such a role. Both political and
administrative respondents said it was very important that the bureaucracy
represent continuity in reform and that political leaders’ lack of familiarity with the
history of the civil service and the trajectory of reforms was what produced
reform symbols.
There are also signs of a cultural shift towards a more result-oriented
administrative culture. The most significant effect of MBOR is increased
consciousness and awareness of goals and results among the ministerial
staff, indicating a shift in their models of thought and internal culture
(Christensen and Lægreid, 1998). There has also been a significant change in
the attitudes of civil service unions, which now accept more devolution and
privatization than they did 10 years ago. Perhaps the most important changes
are the cultural changes in the way we talk about public sector. Administrative
reforms are basically a struggle for people’s minds, so the result might be a
shift in understanding, beliefs and interpretations of public sector organization
(Olsen, 1996).
There is a complex and disputed connection between reform symbols and
reform practice. The majority of responses seemed to indicate a number of
different phases in this connection. Reforms often start out with the politicians
having both symbols and plans or intentions, which they try to co-ordinate in a
parallel strategy. This tight coupling becomes problematic after a while, because
the reforms run into various kinds of problems, particularly cognitive ones. The
result is a looser coupling between symbols and practice, where the ministers
use more and more symbols as a substitute for not being able to deliver reform
practice.
A minority of the respondents seemed to take a different view of the
connection between symbols and practice. They said that symbols generally
played a positive role in the reforms, that politicians could gain political support
based on them, that symbols made it easier to implement some reform
measures, that the MGA and the civil service unions played a constructive role
in the reforms and that a specific reform focus had relatively unambiguous
ADMINISTRATIVE REFORM POLICY 23

intentions and produced politically rewarding results. This view was voiced
particularly often by some politicians, illustrating that reforms can be defined in
different ways and are not seen objectively (Olsen, 1988).
There is large variety of responses concerning administrative reform symbols
and practice and they vary substantially according to level, position, role and
experience. This is understandable in situations when meaning and substantive
outcome of reforms are loosely coupled. The same reform may have different
meaning depending of the interpretive framework through which it is viewed, and
the same meaning can be expressed through a variety of reforms. The question is
whether reform symbols can facilitate the implementation of reforms under
conditions of heterogeneity or whether reforms are characterized by a tug-of-war
between actors employing different reform symbols.
A main argument against an interpretation of symbols as facilitating
implementation is the high share of respondents who viewed reform symbols
in a negative way. This could indicate that reform symbols create tension rather
than a favorable climate for turning intentions into practice. It also tells us that the
administrative actors are less preoccupied with symbols and more focused on
reform practice than the politicians. One reason for this is that the administrators
not only operate in a different environment to the politicians, but they are also
responsible for the implementation of reform policy and therefore naturally will
demand more clarity than the political leaders.
Our data do not, however, indicate a high level of conflict between different
actors as a result of the use of reform symbols, nor is there much evidence of
competing symbols. The position of the secretary generals is so close to the
political leadership that it would be very problematic for them to use counter-
symbols. Agency leaders, and particularly state company leaders, have more
freedom to voice criticism using symbols of their own, but seem to do so to a
limited degree. Instead, they tend to urge the politicians to go further down the
road to reform, to give them more autonomy and put more emphasis on
efficiency.
A strategy of using counter-symbols is more evident among the civil
servants’ unions, particular those that are very reluctant to accept some of the
main ideas of NPM (Jacobsen, 1998). Their dilemma is that if they voice their
objections too loudly they may forfeit the chance to participate in and
influence the process. Showing too much opposition to reforms may also have
the adverse effect of damaging the public image of the civil service and hence
their own position.
There are, however, also signs of symbols facilitating practice and implemen-
tation. One is the share of respondents reporting that symbols help to
communicate intentions and win the support of other actors. Another is the
view by some politicians that, even if they are irritated by the unwillingness of the
civil service and its unions to participate in the implementation of reform policy,
they will still use symbols of importance, co-operation and competence to
underline the position of these actors in the process.
24 T. CHRISTENSEN, P. LÆGREID

Conclusion

In this article we have discussed the symbolic features of administrative reform

policy. Even in a country like Norway, which was a reluctant and late NPM
reformer, the symbolic aspects of administrative reform policy are significant. In
modern and complex systems of public administration, political leaders
generally have problems of capacity, knowledge and authority. Administrative
reform is more comprehensive and it has also become more problematic for
the political leadership to meet the requirements of reform processes
characterized by clear intentions/goals, plans and rationality (Lægreid and
Roness, 1999). In such a situation, certain aspects of administrative reform
policy may remain on a symbolic level. In the case of more radical and
comprehensive reforms with ambiguous goals and solutions, the attention gap
of political leaders will probably become even greater and the symbolic
elements of reforms may thus increase and coherence and consistency
decrease. In such a context, the formation of meaning and the legitimization of
behavior and action will be important aspects of the administrative reform
process. Administrative policy leadership will, in such cases, become a
question of creating, maintaining and changing meaning, viewpoints, doctrines
and proposals for reform in an ambiguous and uncertain environment
(Christensen and Lægreid, 1998; Hood and Jackson, 1991; Olsen and Peters,
1996). The loose coupling between talk and action might not necessarily be a
deliberate strategic choice by political leaders, but rather the consequence of
internal organizational characteristics, which produce a time-lag between talk
and change (Forssell, 2001).
Our conclusion is that symbols are a main feature of administrative reforms
policy but also that reforms are not only talk and symbols (Lægreid, 2001b).
Symbols matter. Symbolic behavior and processes of interpretation are
important for the development of identities, loyalties, trust and compliance, but
they can also be used to manipulate and exercise power. Turning symbols into
practice is, however, not an easy task to do. There is a coupling between the
world of reform policy symbols and the world of practice, even if it is not as tight
as might be expected from an instrumental perspective. Substantive policy-
making and symbolic interpretation sometimes reinforce each other and create a
solid base for authority and legitimacy for administrative reform. Other times
there might be a tension between them, and organizations may put up structural
and procedural façade that matches prevailing ideologies, but also shields
practice and leave core technologies of the organization rather untouched. Or
practice can change substantially within broad and relatively robust symbolic
frames. More use of symbols in reform policy does not necessary mean less
implementation, but it might imply different implementation through processes of
transformation, translation and editing (Christensen and Lægreid, 2001a). The
result might well be tighter coupling over time between the talk of reform and
actual change when symbols are seen as an important aspect of cultural change,
ADMINISTRATIVE REFORM POLICY 25

of Realpolitik and of an active, deliberate and planned administrative reform

policy.

Notes

1. An earlier version of this paper was presented at the 6th National Public Management Conference,
Bloomington, Indiana, October 18–20, 2001. We wish to thank the participants at the conference
and one anonymous referee for helpful comments.

References

Aberbach, J.D., and T. Christensen. (2001). ‘‘Radical Reform in New Zealand: Crisis, Windows of
Opportunities, and Rational Actors.’’ Public Administration 79(2), 404–422.
Bislev, S., and D. Salskov-Iversen. (2001). ‘‘Globalization and Discursive Regulation: New Public
Management.’’ Paper presented at the 16th Nordic Academy of Management Meeting, Uppsala,
16–18 August 2001.
Bolman, L.G., and T.E. Deal. (1987). Modern Approaches to Understanding and Managing
Organizations. San Francisco: Jossey Bass.
Boston, J., J. Martin, J. Pallot, and P. Walsh. (1996). Public Management: The New Zealand Model.
Auckland: Oxford University Press.
Brunsson, N. (1989). The Organization of Hypocrisy. Talk, Decisions and Actions in Organizations.
Chichester: Wiley.
Brunsson, N., and J.P. Olsen. (1993). The Reforming Organization. London and New York: Routledge.
Christensen, T. (1991). Virksomhetsplanlegging. Myteskaping eller instrumentell problemløsning?
(Operational planning. Manipulation of myths or instrumental problem-solving?). Oslo: TANO.
Christensen, T. (1994). Politisk styring og faglig uavhengighet (Political control and professional
autonomy). Oslo: Tano Aschehoug.
Christensen, T., and P. Lægreid. (1998). ‘‘Administrative Reform Policy: The Case of Norway.’’
International Review of Administrative Sciences 64, 457–475.
Christensen, T., and P. Lægreid. (eds.) (2001a). New Public Management. The Transformation of Ideas
and Practice. Aldershot: Ashgate.
Christensen, T., and P. Lægreid. (2001b). ‘‘Coping with Modern Leadership Roles—The Problematic
Redefinition of Public Companies.’’ Paper presented at the 2001 Meeting of the IPSA Section of the
Structure of Governance, University of Oklahoma, March 30–31. Also Bergen: LOS-center. Working
Paper 0107.
Christensen, T., P. Lægreid, and P.G. Roness. (2002). ‘‘Increasing Parliamentary Control of the
Executive? New Instruments and Emerging Effects.’’ Journal of Legislative Studies 8, 267–296.
Christensen, T., P. Lægreid, and L.R. Wise. (2001). ‘‘Assessing Public Management Reform in
Norway, Sweden and the United States of America.’’ International Journal of Political Studies
September, 01, 41–70.
Christensen, T.,P. Lægreid, and L.R. Wise. (2002). ‘‘Transforming Administrative Policy.’’ Public
Administration 80, 153–170.
Christensen, T., and K.A. Røvik. (1999). ‘‘The Ambiguity of Appropriateness.’’ In M. Egeberg and P.
Lægreid (eds.), Organizing Political Institutions. Oslo: Scandinavian University Press.
DiMaggio, P.J., and W.W. Powell. (1991). ‘‘The Iron Cage Revisited: Institutional Isomorphism and
Collective Rationality.’’ In W.W. Powell and P.J. DiMaggio (eds.), The New Institutionalism in
Organizational Analysis. Chicago: The University of Chicago Press.
26 T. CHRISTENSEN, P. LÆGREID

Downs, G.W., and P.D. Larkey. (1986). The Search for Government Efficiency. From Hubris to
Helplessness. Philadelphia: Temple University Press.
Edelman, M. (1964). The Symbolic Uses of Politics. Urbana: University of Illinois Press.
Feldman, M.S., and J.G. March. (1981). ‘‘Information in Organizations as Signal and Symbol.’’
Administrative Science Quarterly 26, 171–186.
Forssell, A. (2000). ‘‘Reform Theory Meets New Public Management.’’ In T. Christensen and P.
Lægreid (eds.), New Public Management: Transforming Ideas and Practice. London: Ashgate.
Forster, G.D., and S.K. Snyder. (1989). ‘‘Public Attitudes towards Government: Contradictions,
Ambivalence and the Dilemmas of Response.’’ In Report and Recommendations of National
Commission of the Public Service. Washington D.C. pp. 19–42.
Helgesen, S. (2001). ‘‘Management-by-Objective-and-Results: Implementing the 1996 Financial
Management Regulation in Norwegian Central Government Administration.’’ Paper presented at the
17th EGOS Colloquium, Lyon July 5–7.
Hood, C. (1997). The Art of the State. Culture, Rhetoric, and Public Management. Oxford: Oxford
University Press.
Hood, C. (2001). ‘‘Control, Bargains and Cheating: The Politics of Public Service Reform.’’ Paper
presented at the 2001 Annual Meeting of the American Political Science Association, San Francisco,
August 29–September 2.
Hood, C., and M. Jackson. (1991). Administrative Argument. Aldershot: Dartmouth.
Jacobsen, D.R. (1998). Diskurser som begrensende og mulighetsskapende. (Discourses as
constraints and opportunities). Bergen: LOC-center. Working Paper 9832.
Lægreid, P. (2001a). ‘‘Transforming Top Civil Servant Systems.’’ In T. Christensen and P. Lægreid
(eds.), New Public Management. The Transformation of Ideas and Practice. Aldershot: Ashgate.
Lægreid, P. (2001b). ‘‘Administrative Reforms in Scandinavia: Testing the Cooperative Model.’’ In B.C.
Nolan (ed.), Public Sector Reform. An International Perspective. London: Palgrave.
Lægreid, P., and J.P. Olsen. (1984). ‘‘The Storting: A Last Stronghold of the Political Amateur.’’ In E.N.
Suleiman (ed.), Parliaments and Parliamentarians in Democratic Politics. New York: Holmes & Meier.
Lægreid, P., and P.G. Roness. (1999). ‘‘Administrative Reform as Organized Attention.’’ In M. Egeberg
and P. Lægreid (eds.), Organizing Political Institutions. Oslo: Scandinavian University Press.
Lægreid, P., and P.G. Roness. (2003). ‘‘Administrative Reform Programs and Institutional Response in
Norwegian Central Government.’’ In J.J. Hesse, C. Hood, and B.G. Peters (eds.), Paradoxes in
Public Sector Reform. Berlin: Duncker & Humbolt.
March, J.G. (1986). ‘‘How We Talk and How We Act: Administrative Theory and Administrative Life.’’ In
T.J. Sergiovanni and J.E. Corbally (eds.), Leadership and Organizational Culture. New Perspectives
on Administrative Theory and Practice. Urbana and Chicago: University of Illinois Press.
March, J.G. (1994). A Primer on Decision Making. New York: The Free Press.
March, J.G., and J.P. Olsen. (1976). Ambiguity and Choice in Organizations. Bergen: Universitets-
forlaget.
March, J.G., and J.P. Olsen. (1983). ‘‘Organizing Political Life: What Administrative Reorganization
Tells Us About Government.’’ American Political Science Review 77(2), 281–297.
March, J.G., and J.P. Olsen. (1989). Rediscovering Institutions. The Organizational Basis of Politics.
New York: Free Press.
Marcussen, M. (2001). ‘‘The OECD: Transnational Regulation Through the Idea-Game.’’ Paper
presented at the SCANCOR Workshop ‘‘Transnational Regulation and the Transformation of
States.’’ Stanford University, 22–23 June 2001.
Meyer, J.W., and B. Rowan. (1977). ‘‘Institutionalized Organizations: Formal Structure as Myth and
Ceremony.’’ American Journal of Sociology 83 (September), 340–363.
Meyer, M.W. (1979). ‘‘Organizational Structure as Signaling.’’ Pacific Sociological Review 22, 481–
500.
Naschold, F. (1996). New Frontiers in Public Sector Management. Berlin: De Gruyter.
Olsen, J.P. (1983). Organized Democracy. Political Institutions in a Welfare State—the Case of Norway.
Bergen: Scandinavian University Press.
ADMINISTRATIVE REFORM POLICY 27

Olsen, J.P. (1988). ‘‘Administrative Reform and Theories of Organization.’’ In C. Campbell and B.G.
Peters (eds.), Organizing Governance: Governing Organizations. Pittsburgh: University of Pittsburgh
Press.
Olsen, J.P. (1996). ‘‘Norway: Slow Learner—or Another Triumph of the Tortoise?’’ In J.P. Olsen and
B.G. Peters (eds.), Lessons from Experience. Oslo: Scandinavian University Press.
Olsen, J.P., and B.G. Peters. (1996). ‘‘Learning from Experience?’’ In J.P. Olsen and B.G. Peters
(eds.), Lessons from Experience. Oslo: Scandinavian University Press.
Perry, J.L. (1992). ‘‘The Merit Pay Reforms.’’ In P.W. Ingraham and D.H. Rosenbloom (eds.), The
Promise and Paradox of Civil Service Reform. Pittsburgh: University of Pittsburgh Press.
Pfeffer, J. (1981). Power in Organizations. Boston: Pitman.
Pollitt, C., and G. Bouckaert. (2000). Public Management Reform. A Comparative Analysis. Oxford:
Oxford University Press.
Premfors, R. (1998). ‘‘Reshaping the Democratic State: Swedish Experiences in a Comparative
Perspective.’’ Public Adminstration 76(2), 141–160.
Power, M. (1997). The Audit Society. Rituals of Verification. Oxford: Oxford University Press.
Rainey, H.G. ( 1996). ‘‘Public Opinion Toward the Civil Service.’’ In H.A.-G.-M. Bekke, J.L. Perry, and
T.A.J. Toonen (eds.), Civil Service Systems in Comparative Perspective. Bloomington: Indiana
University Press.
Roness, P.G. (2001). ‘‘Transforming State Employees’ Unions.’’ In T. Christensen and P. Lægreid
(eds.), New Public Management. The Transformation of Ideas and Practice. Aldershot: Ashgate.
Røvik, K.A. (1996). ‘‘Deinstitutionalization and the Logic of Fashion.’’ In B. Czarniawska and G. Sevon
(eds.), Translating Organizational Change. New York: de Gruyter.
Røvik, K.A. (1998). Moderne organisasjoner. Trender i organisasjonstenkningen ved tusen
rs-skiftet
(Modern organizations. Trends in the organizational thinking at the turn of the millenium). Bergen:
Fagbokforlaget.
Sahlin-Andersson, K. (1996). ‘‘Imitating by Editing Success’’. In B. Czarniawska and G. Sevon (eds.),
Translating Organizational Change. New York: de Gruyter.
Sahlin-Andersson, K. (2001). ‘‘National, International and Transnational Constructions of New Public
Management.’’ In T. Christensen and P. Lægreid (eds.), New Public Management. The
Transformation of Ideas and Practice. Aldershot: Ashgate.
Selznick, P. (1957). Leadership in Administration. New York: Harper & Row.
Skivenes, M.I. (2001). Forvaltningspolitikk og borgermedvirkning i politisk viljedannelse. (Adminis-
trative policy and citizens’ participation in the formation of political will). Bergen: LOS-center.
Working Paper 0111.
Stene, Ø. (2001). From Idea to Concept and Administrative Reform. Paper presented at the 17th
EGOS Colloquium, Lyon, July 5–7.
Tolbert, P.S., and L.G. Zucker. (1983). ‘‘Institutional Sources of Change in the Formal Structure of
Organizations. The Diffusion of Civil Service Reform 1880–1935.’’ Administrative Science Quarterly
28, 22–39.

Tom Christensen is professor in public policy and administration at the Department of Political
Science, University of Oslo.

Per Lægreid is professor at the Department of Administration and Organization Theory and research
director at the Rokkan center, University of Bergen.