Theoretical Framework

2 Theoretical framework
2.1 Governance
2.1.1 A broad definition of a popular concept
How can public-sector activities be coordinated internally and externally if

hierarchy, with its dependence on clear boundaries, has weakened? How
can public managers be prevented from becoming confused about their
roles and intervention capacity? These questions are dealt with under the
banner of ‘governance’. Governance has become a buzzword, not only
among public-administration scholars, but also among practitioners. The
term even risks becoming so general that it becomes meaningless and, as
Peters warns, “a tautology: something happened, and therefore governance
occurred”.26 Governance “has become a rather fuzzy term that can be ap-
plied to almost everything and therefore describes and explains nothing”.27
This critique is understandable: the term governance is defined in doz-
ens of different, and in some respects contrasting ways. Only some of the
most influential examples will be mentioned here. Kettl defines govern-
ment as the structure and function of public institutions, and governance as
the way government gets its job done. 28 In his book ‘Understanding Gov-
ernance’, Rhodes distinguishes six uses of the term governance: 29
- Governance as the minimal state: the use of markets and quasi-
markets to deliver ‘public’ services’;
- Governance as corporate governance: this is mainly about transpar-
ency, integrity and accountability, by means of control;
- Governance as the new public management: the introduction of pri-
vate sector management methods and incentive structures such as
market competition to the public sector;
26 Peters (2000: 35): Globalisation, institutions and governance.

27 Jessop (2002: 4): Governance, governance failure and metagovernance.
28 Kettl (2002: xi): The transformation of governance.
29 Rhodes (1997: 47-53): Understanding governance.
10 2 Theoretical framework
- Governance as ‘good governance’: a ‘marriage of the new public

management with liberal democracy’;
- Governance as a socio-cybernetic system: interdependence among
social-political-administrative actors; governance is the result of in-
teractive social-political forms of governing;
- Governance as self-organising networks: networks develop their own
policies and shape their environments.
Stoker identifies similar definitions that are in his view complemen-
tary.30 Farazmand stresses the international dimension of governance, as
well as the fact that concepts like ‘good governance’ are highly normative
and value-laden.31 From a socio-cybernetic viewpoint, Kooiman defines
governance as “the emerging pattern arising from governing.”32 Lowndes
and Skelcher add an actor’s perspective: Modes of governance are “the
outcome of social processes but also provide the medium through which
actors interpret and act to shape their reality”.33 Frederickson prefers to de-
fine governance as “the attempt to comprehend the lateral and institutional
relations in administration in the context of the disarticulated state”34
What unites most of the definitions presented above is that governance
is more than ‘what governments do to get their jobs done’: the term gov-
ernance refers to the relations between public-sector actors and societal ac-
tors when addressing public issues. This relational aspect of governance
has stimulated many scholars to use the term governance as a synonym for
what others call ‘network governance’.35 Rhodes’ list of six approaches to
governance however also includes other types of relations than only net-
work relations, namely hierarchical relations and market-style relations.
Such a broad perspective is required for this research, because the research
topic is the manageability of combinations of hierarchies, networks and
markets. A strong argument for a broad use of the term governance is
30 Stoker (1998): Governance as theory: five propositions.

31 Farazmand (2004: 6): Sound governance in the age of globalization: a concep-
tual framework.
32 Kooiman (ed.) (1993): Modern governance.
33 Lowndes and Skelcher (1998: 318): The dynamics of Multi-Organizational
Partnerships: an Analysis of Changing Styles of Governance.
34 Frederickson and Smith (2003: 226): The public administration theory primer.
35 Also called ‘New Governance’: Social coordination based on the logic of co-
steering and network. See also Lee (2003: 2, Conceptualizing the New Gov-
ernance: A new institution of social coordination), and Voss (2007: 36) where
governance is ‘societal self-regulation’, in contrast to hierarchy (ibid., p.18).
2.1 Governance 11
given by Schuppert: with a narrow governance definition it is impossible

to include the historically most successful form of governance, namely hi-
erarchical-bureaucratic governing.36 Only with a broad governance concept
it is possible to analyse the challenge of governance which the Danish pub-
lic manager Wolf has phrased as follows:
“What is important is to look beyond the fine-tuning of government ma-
chinery and use the wide angle to capture the way in which we organize
our societies and the interaction between government, market, civil society
and individual citizens.”37
The definition of governance should not only emphasise the relational
aspect but also the institutional aspect; public managers, who are the prin-
cipal governance actors in this research, work in and with institutions.
Mayntz’s definition includes this wide angle:
“Governance is the totality of all co-existing forms of collective coordina-
tion of societal issues, from the institutionalised societal self-regulation via
several forms of cooperation between governmental and private actors, to
the official duties of state actors”38
Therefore, in this research the following broad definition of governance
is used:
Governance is the totality of interactions, in which government,

other public bodies, private sector and civil society participate, aim-
ing at solving societal problems or creating societal opportunities.
In this definition, the institutional dimension is included: the relations

between these actors are influenced by their institutions (in a broad sense:
their organisations, values and norms, and procedures). Kooiman distin-
guishes three orders of governance: first order governance (problem solv-
ing and opportunity creation), second order governance (care for institu-
tions) and third order governance or metagovernance.39 Most research on
governance focuses on the nature of the relations between public-sector
organisations and their societal environment. This belongs to Kooiman’s
first order governance (the macro-level mentioned in Section 1.1). How
governance modes emerge and are organised inside public-sector organisa-
36 Schuppert (2007: 8): Was ist und wozu governance?

37 Wolf (2007: 691): Trends in public administration.
38 Mayntz (2004: 66): Governance im modernen Staat. (translated from German
by the author).
39 Kooiman (2003: 182): Governing as governance.
tions (the institutional aspect: second order governance, or the meso and
micro level) in relation to first order governance, is still relatively less ill-
researched. A case study about the failed introduction of interactive poli-
cymaking in the Dutch Ministry of the Environment highlights a possible
internal failure factor: tensions between governance approaches inside the
organisation, on various dimensions (strategy, structure, competencies,
processes and procedures, to name a few).40 The study suggests that public-
sector organisations apply different governance approaches for internal
matters, such as human resources management, and for external matters,
such as tackling societal problems. This not only adds to an unsatisfactory
performance, but also leads to credibility problems.
The next question is how to make this governance concept practicable.
Is it possible to distinguish or rather construct a small number of govern-
ance approaches that are distinct and together have a large explanatory po-
tential?
2.1.2 Three ideal-types of governance and hybrid forms
Governance styles can be defined as “the processes of decision-making

and implementation, including the manner in which the organisations in-
volved relate to each other”.41 Although many governance styles have been
distinguished, they are usually grouped into three ‘ideal-types’ of govern-
ance, that are considered to have played a role in Western administrations
since the 1950s: hierarchical, market and network governance.42
Before the ‘discovery’ of policy networks and the mechanisms of dif-
ferent types of network governance, social coordination was considered to
40 Meuleman (2003): The Pegasus Principle: reinventing a credible public sec-

tor.
41 Van Kersbergen and van Waarden (2004: 143): ‘Governance’ as a bridge
between disciplines.
42 Kaufman et al. (1986: Guidance, control and evaluation in the public sector);
Thorelli (1986: Networks: Between markets and hierarchies); Peters (1998:
Managing Horizontal Government); Lowndes and Skelcher (1998: The dy-
namics of Multi-Organisational Partnerships); Thompson et al. (1991: Mar-
kets, hierarchies and networks); Thompson (2003: Between hierarchies and
markets); Powell (1991: Neither market nor hierarchy: network forms of or-
ganisation); Kooiman (2003: Governing as governance), Considine and Lewis
(2003: Bureaucracy, Network or Enterprise?); Kickert (2003: Beneath consen-
sual corporatism); Schout and Jordan (2005: Coordinated European govern-
ance).
2.1 Governance 13
take place in two distinct forms: hierarchies and markets.43 Market coordi-
nation was the second ideal type that was described after Weber’s bureau-
cratic ideal type had become the prototype for a classical hierarchy. Net-
works were, for a long time, considered a hybrid form of these ideal-types.
We find the fiercest defenders of the idea of a hierarchy-market dichotomy
among economists. In a publication of 2007, Ruys et al. argue that market
contracting (market governance) is the ‘original state of affairs’, and call
the opposite vertical integration (hierarchical governance), while all gov-
ernance styles between these extremes are called hybrid relationships.44
However, economists were also among the first to argue that networks
form a separate type of social coordination.45
Already in 1986, Thorelli stated that the network form is a distinct form
of societal coordination, and not ‘just’ a hybrid form that combines hierar-
chy and markets. 46 Meanwhile, there is a huge public administration litera-
ture based on the idea that network governance, after hierarchical and
market governance, has become the third ideal-type. Network governance
as an alternative to hierarchical or market governance is not only accepted
in public management, but is also widely applied in knowledge-intensive
businesses47 and in private enterprises in general.48 The network concept
has become so popular, that sometimes a new dichotomy emerges, namely
hierarchy versus networks, while market governance is neglected.49 Others
43 Thompson (2003: 37) makes a useful distinction between ‘coordination’

(alignment of the elements in a system) and ‘governance’ (the regulation of
their alignment). He places them on a continuum: coordination simply brings
together elements in an ordered pattern, and governance does this by direction
and design. Hierarchies, networks and markets can be used as coordination
mechanisms and as governance structures as well.
44 Ruys et al. (2007): Modes of governance in the Dutch social housing sector.
45 E.g. Powell (1991: Neither market nor hierarchy: Network forms of organiza-
tion), Thompson (2003: Between hierarchies and markets: The Logic and
Limits of Network Forms of Organization).
46 Thorelli (1986): Networks: Between markets and hierarchies.
47 Roobeek (2007): The networking landscape. Navigation for the route to net-
working organisations.
48 E.g. Assens and Baroncelli (2004, Marché, Réseau, Hiérarchie : à la recherche
de l'organisation idéale) and Larson (1992, Network dyads in entrepreneurial
setting).
49 E.g. Koffijberg (2005: Getijden van beleid: omslagpunten in de volkshuisvest-
ing), Kalders et al. (2004: Overheid in spagaat. Over spanningen tussen verti-
cale en horizontale sturing).
reserve the term governance for what they call the ‘new modes of govern-
ance’ (i.e. market and network50) – which is a contradictio in terminis: hi-
erarchy must then also be a governance style, namely the ‘old’ mode of
governance. Peters bridges this contradiction by bringing network and
market governance approaches under the umbrella of ‘informal govern-
ance’, besides the ‘formal governance’ of hierarchies.51 Notwithstanding
the overwhelming empirical evidence that the trichotomy hierarchy-
network-market has more analytical power than the hierarchy-market di-
chotomy or the hierarchy-network dichotomy, there is still a dispute
among scholars about whether this trichotomy makes sense. This dispute is
emotional and value-laden; not surprisingly, because different world views
or belief systems collide in this debate. In Section 2.4 we will discuss this
cultural dimension more in-depth.
Before concluding that the three ideal types hierarchical, market and
network governance cover the whole field, we have to answer two ques-
tions. The first is: should we distinguish successful hybrid forms of gov-
ernance as separate governance styles? The second question is: how plau-
sible is it that these three styles together, and in combination, cover all
governance approaches?
Hybrid forms of governance
Governance hybrids do not necessarily relate to so-called ‘hybrid organisa-

tions’ exclusively. The term hybrid organisations was coined in 1995 by In
‘t Veld.52 Usually they are somewhere on a gradient between pure govern-
ment agencies and pure commercial firms; they operate in a ‘twilight zone’
between public and private53. They can be defined as organisations gov-
erned by two or more ‘pure’ modes of governance.54 Hybrid organisations,
50 E.g. European Commission (2002: 7): Report from the Commission on Euro-
pean Governance. Rhodes is ambivalent too: he defines hierarchy as one of
the governance structures besides market and network (Rhodes, 1997: 47,
Understanding Governance) and elsewhere in the same book defines govern-
ance in a network connotation, as an alternative to markets and hierarchies
(Rhodes, 1997: 53).
51 Peters (2005: 1): Forms of informal governance: Searching for efficiency and
democracy.
52 In ‘t Veld (2005): Spelen met vuur. Over hybride organisaties.
53 Jörgensen (1999: 570): The public sector in an in-between time: Searching for
new public values.
54 Ruys et al. (2007): Modes of governance in the Dutch social housing sector.
2.1 Governance 15
as intercultural entities, are able to bridge fragmented and decoupled rela-

tions in the public sphere.55 Hybrid organisations are not new: they have
been in existence for quite some time. The British and Dutch East India
Companies of the 17th century are often mentioned as early examples.56
Kickert even estimates that currently most organisations in the public
sphere in Western Europe are hybrid organisations.57
As stated above, network governance was originally regarded as a hy-
brid of hierarchical and market governance, until it was distinguished as a
separate form. Meanwhile, many other candidates for promotion to the
‘Ideal Type League’ have appeared. Six hybrid forms of governance that
are often mentioned are:
- Oligopolies (a market form of coordination with important network
characteristics, that is not restricted to the private sector);
- Public private partnerships (also a combination of market and net-
work governance);
- Chain management (a form of network governance concentrating on
functional instead of social relations between actors, which borrows
its structure from hierarchical thinking);
- The Open Method of Coordination of the European Commission;
- The related concepts of self-regulation and self-organisation;
- An emerging type with mainly network and market features: ‘bazaar
governance’.
Oligopolies
An oligopoly is a market characterised by a small number of organisations

who realize that they are interdependent in their pricing and output poli-
cies. The number of organisations is small enough to give each of them
some market power.58 The behaviour in an oligopoly is based upon self-
interest (autonomy, which refers to market governance), but the fact that
actors realize their interdependency, makes them behave with more empa-
thy and moderation than in more open markets. This leads them to con-
sider each other like actors do in a network approach. Relations in oligopo-
lies are usually bilateral. They become multilateral when they have a
55 In ‘t Veld (1997: 148): Noorderlicht. Over scheiding en samenballing.

56 Wettenhall (2003: 237): Exploring types of public sector organizations.
57 Kickert (2001: 135): Public management of hybrid organizations.
58 Definition OECD (http://stats.oecd.org/glossary/detail.asp?ID=3152, retrieved
on 27 September 2007).
formal agreement: this is known as a cartel. A situation where there is a

single seller in a market is called a monopoly.
Public private partnerships
A public private partnership (PPP) is a non-hierarchical system of govern-

ance in which public and private actors form a common venture that serves
certain policy goals, such as public services or infrastructure development.
Börzel and Risse distinguish two basic steering modes of PPP: bargaining
(related to market governance) and persuasion or arguing (related to net-
work governance). They formulate four distinct types of PPP: cooptation,
delegation, co-regulation and self-regulation in the shadow of hierarchy.59
Klijn and Teisman criticise the PPP practice in the Netherlands during the
1990s and early 2000s. Although PPP became popular among politicians,
in praxis even in PPP arrangements, policymaking “continues to be based
on self-referential organisational decisions, rather than on joint inter-
organisational policymaking.”60
Chain management
Chain management is a form of governance similar to network govern-

ance. The actors in the chain are interdependent, because of the functional
interdependence of the processes that are linked. The governance of chains
differs from network governance: a network is defined by social relations,
a chain only by functional relations.61 The chain concept can be useful be-
cause it brings a certain order in the relations between actors in processes.
The restriction to functional relations however is a risk: a chain perspective
gives only part of the whole picture.62
59 Börzel and Risse (2002): Public-Private Partnerships: Effective and legitimate

tools of international governance?
60 Teisman and Klijn (2002: 197): Partnership arrangements: Governmental
rhetoric or governance scheme?
61 Kort, van Twist and in ‘t Veld (2000: 30): Over ontwerp en management van
processen in ketens.
62 Kort, van Twist and in ‘t Veld (2000: 38): Over ontwerp en management van
processen in ketens.
2.1 Governance 17
Open method of Coordination
Another governance style hybrid is the so-called Open Method of Coordi-

nation (OMC) of the European Union, which was identified at the Lisbon
European Council in 2000.63 The OMC implies:64
- Fixing guidelines for the Union and specific timetables for achieving
set goals in the Member States;
- Establishing indicators and benchmarks as a means of comparing best
practice;
- Translating the European guidelines into national policy reform ac-
tions which are integrated into national action plans (NAPs); and
- Periodic monitoring, evaluation and peer review, organised as mutual
learning processes.
This approach is characterised as a ‘soft’ approach compared to hierar-
chical governance,65 and codifies practices taken from market governance
thinking such as benchmarking, target-setting and peer reviewing.66 The
OMC can also be seen as a form of network governance: it aims at linking
both public and private actors in joint determination of policy.67 However,
this participatory dimension has met much scepticism. It is often claimed
that the OMC is a highly technocratic process involving selected actors in
a closed policy network.68
Self-regulation and self-organisation
Two concepts that are often mentioned in governance literature are self-
regulation and self-organisation. Self-regulation takes the perspective of
the politico-administrative system, self-organisation the perspective of so-
63 Laffan and Shaw (2005): New modes of governance. Classifying and map-
ping OMC in different policy arenas.
64 European Council (2000): Presidency Conclusions of the European Council in
Lisbon, 23 an 24 March 2000.
65 Héritier (2002): New modes of governance in Europe: policymaking without
legislation?
66 Hodson and Maher (2001: 719): The Open Method as a new mode of govern-
ance.
67 Peters (2005: 8): Forms of informal governance: Searching for efficiency and
democracy.
68 E.g. Smismans (2006: 18): New modes of governance and the participatory
myth.
cietal organisations.69 Klijn and Koppenjan define self-regulation as an in-

terventionist form of steering by stimulating and sometimes compelling ac-
tors to take care of the quality of output themselves, through reward rules
and product rules.70 Fuchs argues that self-organisation of society relies
largely on exclusion, competition and heteronomy.71 The ‘self’ in both
forms refers to individual autonomy. This links these concepts to market
governance, with a touch of hierarchy (self-regulation is always regulated
self-regulation72) and a strong network flavour (self-organisation builds on
voluntary cooperation and trust).
The term self-organisation is also used in Luhmann’s theory of social
systems. Self-organising systems are autopoietic: they produce and repro-
duce the elements they consist of, with the help of those elements them-
selves. Autpoiesis is a biological model of living systems. The autopoietic
approach to public administration stresses the limits of both the hierarchi-
cal ‘command and control’ approach and the market governance approach
of public management as a neutral and transferable technology. 73 Kickert
has warned that applying such a natural scientific model to a social science
is hazardous, but can also inspire new ideas. 74 75
Bazaar governance
A new form of self-organisation that seems to emerge is what Demil and

Lecocq have coined bazaar governance. 76 It was first recognised in the
‘market’ of open source software. How the internet encyclopaedia
Wikipedia is ‘governed’ is another example of bazaar governance. It is
characterised by low levels of control (hierarchy), weak incentives inten-
sity (market) and a network that does not build on trust: community mem-
69 Puppis et al. (2004: 9): Selbstregulierung und organisation. Schlussbericht.

70 Klijn and Koppenjan (2004: 219): Managing uncertainties in networks.
71 Fuchs (2002: 63): Concepts of social self-organization.
72 Mayntz (2003:4): From government to governance: Political steering in mod-
ern societies.
73 Brans and Rossbach (1997: 425, 435): The autopoiesis of administrative sys-
tems.
74 Kickert (1991): Autopoiesis and the Science of Administration: Essence,
Sense and Nonsense.
75 See also: In ‘t Veld et al. (eds.) (1991): Autopoiesis and configuration theory:
new approaches to societal steering.
76 Demil and Lecocq (2006): Neither market nor hierarchy nor network: The
emergence of bazaar governance.
2.1 Governance 19
bers seldom know each other and may enter or leave the network unno-
ticed. This form of governance scores low on key features of all three
ideal-typical governance styles (authority, trust and price) but seems, in es-
sence, a special mixture of network governance (co-creation) and market
governance (individual autonomy).
2.1.3 Are hierarchies, networks and markets ‘all-inclusive’?
The hybrid forms we have mentioned above are indeed mixtures of hierar-
chical, network and market governance. Benz differentiates ‘negotiation’
as a fourth governance style, besides hierarchy, network and ‘competition’
(market).77 However, in the broad definition of governance we use in this
research, each of the three ideal types has its own form of negotiation. Hi-
erarchical negotiation is based on hierarchical positions and instruments;
network negotiation is characterized by deliberation and attempts to reach
mutual gains, and market negotiation is bargaining based on price and
competition. Therefore, distinguishing ‘negotiation’ as a fourth style of
governance seems not necessary and may even blur the clarity we achieve
by using ideal types that encompass many dimensions of governance, in-
cluding the style of negotiation.
Considine defines ‘corporate governance’ as a fourth style of govern-
ance, between ‘procedural governance (hierarchy) and market govern-
ance.78 However, the focus on management and targets suggests that corpo-
rate governance should be placed in the category of market governance.
The six definitions of governance given by Rhodes79 are all congruent
with combinations of hierarchical, network and market forms of govern-
ing. Another example in which the three styles are used to present a sim-
plified but complete picture of governance approaches is the ‘Global Sce-
narios 2025’ report of Shell: The three ‘utopias’ of the corner stones of
their model are a ‘state centric world’, a ‘civil society centric world’, and a
‘market centric world’.80 Security, coercion and regulation are key features
of the first; social cohesion, justice and the force of community determine
the second, and efficiency and market incentives characterise the third so-
77 Benz (2006: 35): Eigendynamik von Governance in der Verwaltung.

78 Considine (2006: 7): The power of networks: Institutional transformations in
the global era?
79 Rhodes (1997: 47-53): Understanding governance.
80 Shell (2005: 42): The Shell Global Scenarios to 2025.
cietal force. Trade-offs between two of the three forces are considered the
most plausible scenarios for (global) societal development:
- The combination of hierarchy and network leads to the ‘Flags’ sce-
nario. In this scenario, hierarchy protects communities against a dan-
gerous outside world. ‘Gated communities’ are a typical expression.
- A trade-off between hierarchy and market results in the ‘Low Trust
Globalisation’ scenario. Checks and control, which are top down
measures, protect the economy. It is a legalistic scenario.
- The combination of market and network leads to the third scenario:
‘Open Doors’. This is a pragmatic free market scenario, characterised
by incentives and building bridges, open standards, and open borders,
for example.
The ‘Open Doors’ scenario bears a strong resemblance to the current
societal governance culture in the Western World. Figure 2 shows that the
six types of hybrid governance that we described above, should all be pic-
tured in this scenario. This does not mean that the other scenarios do not
reflect existing patterns. Gated communities (‘Flags’) can be found on the
level of nations (economic protectionism), but also on the level of organi-
sations. The Dutch Ministry of Agriculture during the 1980s and early
1990s was a gated community: Employees of the Environment Ministry
labelled the Agriculture Ministry the ‘Kremlin’. There was little or no
room in that ministry for differing opinions.81 ‘Low Trust Globalisation’ is
related to New Public Management (see 2.2.2): it works with ‘carrots and
sticks’, and combines flexibility with top down control.
An important argument supporting the trichotomy concept comes from
cultural theory. This is dealt with in Section 2.4.1. Finally, empirical re-
search by Considine and Lewis has shown that public officers indeed ex-
perience that there are three separate styles. For them, hierarchy is weakly
related to both the other styles, and the market and network styles have a
strong negative correlation.82
It may now be concluded that hybrid forms of governance may have
analytical value and should therefore be used in the analysis of governance
cases. It is also possible to conclude that the use of the three ideal-types hi-
erarchy, network and market, provided that they not are presented as
monolithic constructs but as sets of related characteristics with a distinct
internal logic, can provide a basic analytical tool for understanding gov-
81 Kickert (1997: 744): Public governance in the Netherlands.

82 Considine and Lewis (2003): Bureaucracy, network or enterprise? Comparing
models of governance in Australia, Britain, the Netherlands and New Zealand.
2.2 Hierarchical, market and network governance 21
ernance. The concepts of hierarchical, network and market governance to-

gether offer a complete enough analytical framework for explaining the
conflicts and synergies within and between governance approaches.
Fig. 2. Hybrid forms of governance and the three 2025 Shell scenarios
2.2 Hierarchical, market and network governance
Since we have now ‘cleared the ground’, we can take a closer look at each
of the three ideal-types.
2.2.1 Hierarchical governance
During the second half of the 20th century in all OECD countries, funda-
mental changes took place in the organisation of the state and its institu-
tions and in the relations between the state and society. The ideal type of
bureaucracy developed by the German sociologist Max Weber83 became
the role model for public administration in the 1950s and 1960s. Weberian
83 Weber (1952): The essentials of bureaucratic organization: An ideal-type con-

struction.
bureaucracy changed its shape in the 1970s, a decade characterised by a

central top-down planning euphoria – but it did not change its fundamen-
tals. Van Gunsteren described the 1970s public sector as a group of organi-
sations that relied on “regulations, obedient organisations and people, and
if necessary on force”84. In these organisations, functionaries worked
within a system of clear hierarchical relations, under uniform rules.85 The
five main points of Weber’s model are:86
- A carefully defined division of tasks;
- Authority is impersonal, vested in rules that govern official business;
- Employees are recruited based upon proven or at least potential com-
petence;
- Secure jobs and salaries, and promotion according to seniority or
merit;
- A disciplined hierarchy in which officials are subject to the authority
of their superiors.
Weber’s vision of bureaucracy as a rational and objective machine was
based on ideas of efficiency drawn from the Prussian army and the mecha-
nisation of the industrial revolution.87 He believed that efficiency and ra-
tionality would lead to the development of essentially similar bureaucratic
structures the world over. He thus ignored differences that arose from the
political, social and cultural environments of these organisations.88
The Weberian bureaucracy has laid out the basic pattern for the current
public administration in Western democracies. This pattern is hierarchical
governance. The hierarchical mode of governance has developed in
Europe to “replace arbitrary authoritarianism and nepotism”. It “provided a
way for standardising government tasks”.89 It was believed, consistent with
a mechanistic scientific model, that organisations can be built and made to
function as a machine.90 Mintzberg91 refers to it as the machine bureauc-
racy. The primarily hierarchical public-sector organisation typically has
84 Van Gunsteren (1976: preface): The quest for control.

85 Parris (1969: 22): Constitutional bureaucracy. The development of British
central administration since the eighteenth century.
86 Heywood (1997: 220): Politics.
87 Morgan (1986: 21-22): Images of organization.
88 Heywood (1997:347): Politics.
89 Herbst (1976: 16): Alternatives to hierarchies.
90 Herbst (1976: 16): Alternatives to hierarchies.
91 Mintzberg (1993): Structures in Fives: Designing Effective Organizations.
employees with a law background: they are used to reduce complexity by

splitting complex problems into smaller sub-problems92.
The dominance of the Weberian ideal-type in the governance mix was
already criticized in the early 1970s. The American scholar Cleveland ex-
pected that:
“…the organisations that get things done will no longer be hierarchical
pyramids with most of the control at the top. They will be systems – inter-
laced webs of tensions in which control is loose, power diffused, and cen-
tres of decision plural. “Decision-making” will become an increasingly in-
tricate process of multilateral brokerage both inside and outside the
organisation which thinks it has the responsibility for making, or at least
announcing, the decision. Because organisations will be horizontal, the
way they are governed is likely to be more collegial, consensual, and con-
sultative. The bigger the problems to be tackled, the more real power is dif-
fused and the larger the number of persons who can exercise it – if they
work at it”.93
Thus, a new view of public administration emerged, that was not inter-
nally, but externally oriented. Following this view, public administration
was, two decades, later defined as “the whole of mediation institutions that
mobilize human resources in the service of the state in a given territory”.94
The hierarchical structure of Weberian bureaucracy implies a monocen-
tric system: one power centre that governs a civil service system. It is an
instrumentalist approach: public administration with its hierarchical struc-
ture and a hierarchy of value systems is the instrument of governing. How-
ever, empirical research in the 1970s led to the conclusion that societal
problem solving is a continuous process that is multi- and inter-
organisational, and that this necessitates a polycentric system.95 Hierarchy
looses most of its meaning when policymaking is seen as co-production of
interdependent policy centres within and outside of the civil service.
The reactions of the public sector to this critique were mixed. It ad-
dressed a stereotypical Weberian bureaucracy, which in reality did not ex-
ist. However dominant the rational hierarchical paradigm was, public man-
agers had to do their job in a political, social and cultural context that was
92 Schutter (2007: 40): Was ist und wozu Governance?

93 Cleveland (1972: 13): The Future Executive: A Guide For Tomorrow’s Man-
agers. Cited in Frederickson (2004: 2).
94 Morgan and Perry (1988): Re-orienting the comparative study of civil service
systems. Cited in Bekke et al. (1996: 2).
95 Hanf and Scharpf (1978): Interorganizational policy-making; limits to central
coordination and control. Cited by Toonen (1993: 253): Bestuur en beleid.
pluralist and corporatist (like in Germany), and consensus-oriented (like in

the Netherlands). Success of the civil service partially relied on co-
operation with societal partners. Herbst described four assumptions of hi-
erarchical organisations that do not match with complex, dynamic envi-
ronments: 96
- A task can nearly always be decomposed into smaller and smaller in-
dependent parts;
- An organisation has a simple inflexible structure which can be visual-
ised in an organigram with lines of responsibility;
- Organisations are of a uniform type;
- Organisational units have a single, exclusive boundary.
In Germany, the classical Weberian bureaucratic model with its strong
emphasis on legality and proper fulfilment of regulatory functions has re-
mained very popular.97 Therefore, especially at the federal level, there has
been little support for changes. Federal administration was (and is) mainly
limited to law making and not concerned with service delivery. This re-
duced the need for administrative reform. However, there were some re-
forms in the 1960s, which aimed at decentralisation as well as recentralisa-
tion. From the 1970s, the reform objective became to simplify the
administration. This objective was citizen-oriented: better delivery of pub-
lic services.98
Hierarchical governance lost some of its attraction in the 1980s when
the market governance movement New Public Management (NPM) be-
came the focus of both public administration scholars and practitioners
(see also 2.2.2, Market governance). During the 1990s, there was a revival
of the hierarchical approach. Most OECD countries introduced the reform
concept ‘New Public Governance’ (NPG), partly to replace New Public
Management, and partly as an addition to the management movement.
The banner ‘public governance’ contains an interesting dichotomy. In
the majority of (European) public administration literature, the term ‘pub-
lic governance’ is used synonymously with the term ‘network govern-
ance.99 However, business administration literature100, finance specialists in
96 Herbst (1976: 23-28): Alternatives to hierarchy.

97 Kickert and Stillman (2005: 657): The future of European Public Administra-
tion Sciences. Part III: Germany.
98 Naschold et al. (1994): Neue Städte braucht das Land. (cited by Pollit and
Bouckaert (2003: 238).
99 E.g. Bovaird (2005): Public governance: balancing stakeholder power in a
network society. Kickert (1997): Public Governance in the Netherlands: An
ministries and organisations like the World Bank use the term ‘public gov-
ernance’ as an umbrella for what they also call ‘government govern-
ance’101, or, rather normatively, ‘good governance’102. This, in contrast to
network governance, is essentially a hierarchical approach. Government is
considered the key player. Societal actors are influencers of policy imple-
mentation and they form a basis for criteria to assess the results of these
policies103: they are not equal partners of the public sector. Government
governance promotes accountability as a solution for the problem that the
new service arrangements of government with external parties leads to
higher risks for politicians.104 The mergence of this approach was a reac-
tion to societal issues such as the deficient accountability, transparency and
control of the public sector. Its focus on accountability, transparency and
integrity was caused by various financial scandals regarding mismanage-
ment and abuse of public money.105 The core idea is that stakeholders,
within and outside the public sector, benefit from good (internal) control
and good accountability106.
Government governance has (like New Public Management) an Anglo-
Saxon origin.107 It is more tailor-made for typical public sector issues than
NPM was. According to Hajer, government governance more or less com-
bines NPM-thinking with democratic principles such as participation, jus-
tice and equality.108 Therefore, its primary concern was to improve the
compliance with relevant laws and regulations, and financial management
Alternative to Anglo-American 'Managerialism'.

100 E.g. Bossert (2003: 14): Public Governance. Leidraad voor goed bestuur en
management.
101 Ministry of Finance of the Netherlands (2000: 8): Government Governance.
Corporate governance in the public sector, why and how?
102 www.worldbank.org .
103 Van den Berg et al. (2001: 11): Professional Judgement. Handen en Voeten.
Vormgeving van public governance in de praktijk van de gemeentelijke
overheid.
104 Sol (2003: 2): Government governance and beyond. Reconciling flexibility
and accountability in labour market policy in the Netherlands.
105 Van den Berg et al. (2001: 13): Professional Judgement.
106 Bossert (2003: 16): Public Governance. Leidraad voor goed bestuur en
management.
107 Ministry of Finance of the Netherlands (2000: 6): Government Governance.
108 Hajer et al (2004: 14): Nieuwe vormen van Governance. Een essay over
nieuwe vormen van bestuur.
inside the public sector. Thus, government governance typically addresses

internal organisational issues.
As we will see in Section 2.2.5, hierarchical governance in its different
forms still plays a dominant role in Western public-administration organi-
sations, and for good reasons. To quote Peters: “Although analysts have
denigrated hierarchy and praised alternatives such as networks and mar-
kets, one must remember that there are still virtues in hierarchies”. 109 UK
urban regeneration practice provides another example: in the ‘mix’ of
market, hierarchy and network, hierarchy is more persuasive than net-
work.110
To conclude: Hierarchical governance, applied inside and outside 21st
century Western public administration, accounts for top-down decision-
making, strict internal and external accountability procedures, a hierarchi-
cal organisation structure, an emphasis on project management rather than
on process management, strategy styles of a planning and design type, and
a strong preference for legal measures.
2.2.2 Market governance
From the 1980s, the ‘managerial’111 and market-oriented reform movement

‘New Public Management’ (NPM) that was born in a time of economic re-
cession112, stimulated the development of what has become known as mar-
ket governance. The term ‘market’ is a metaphor, which refers to market
mechanisms and market thinking, not to be confused with the economic
market. ‘Market governance’ is a (public) governance style, whereas ‘gov-
ernance of the market’ would mean governance of players active in the
private market. Market governance is a way of thinking and acting that is
used in both the public and private sector, and in hybrid organisations. The
emergence of NPM must be seen against the background of substantial fi-
109 Peters (1998: 301): Managing horizontal government: The politics of co-
ordination.
110 Davies (2002: 301): The governance of urban regeneration: a critique of the
‘governing without government’ thesis.
111 The fact that New Public Management encouraged management thinking in
the public sector does not mean that management was not an issue until then.
However when NPM became influential, management techniques began to
dominate other competencies which top civil servants had developed.
112 Bovaird and Löffler (2001: 5): Emerging trends in public management and
governance.
nancial problems that governments had to deal with in the 1980s. Other in-
centives were the political scandals arising from the complicated inter-
twinement of government and several business sectors, such as the ship-
ping business in the Netherlands. These scandals weakened ties between
government and private sector. NMP therefore had a dual focus: on service
(market thinking) and on accountability (hierarchical thinking).113
The core belief of NPM is that incorporating efficiency principles, pro-
cedures and measures from the private sector, and market mechanisms
leads to better performance of public administration.114 Customer orienta-
tion is another important characteristic of NPM.115 Public choice theory is
central to the NPM model: It implies that
“voters are guided by economic self-interest, interest groups are rent seek-
ing, politicians are entrepreneurs interested in power and perks, and bu-
reaucrats believe in budget maximization and bureau expansion”116.
Starting in New Zealand and spreading through other Anglo-Saxon
countries117 to other countries and international organisations like the
World Bank118, it did not take long before a NPM-set of administrative
doctrines figured prominently in the reform agenda of most OECD coun-
tries.119 Osborne and Gaebler’s Reinventing Government120 became the
worldwide icon of the movement. Accoding to Kettl121, the impact of the
book was surprising: the authors were a journalist and a former city man-
ager respectively, not academics. The academic Hood, who described
NPM as an “uneasy combination of individualism and hierarchism”, dis-
tinguished seven main characteristics of NPM:122
113 Hernes (2005: 5): Four ideal-type organizational responses to New Public
Management reforms and some consequences.
114 Levy (2003): Critical success factors in public management reform: the case
of the European Commission.
115 Kickert (2001: 136): Public management of hybrid organizations.
116 Haque (2007: 180): Revisiting the New public Management.
117 It is plausible that NPM started in Anglo-Saxon countries because their ‘pub-
lic service’ model of administration is inherently more open to market think-
ing than the European continental ‘Rechtsstaat’ model.
118 Kettl (2002: 21): The transformation of governance.
119 Hood (2003: 269): From public bureaucracy state to re-regulated public ser-
vice: The paradox of British public sector reform.
120 Osborne and Gaebler (1992): Reinventing government.
122 Hood (1996: 269): Exploring variations in public management reform of the
1980s.
- Hands-on professional management of public organisation;

- Explicit standards and measures of performance;
- Greater emphasis on output controls;
- Shift to segregation of units in the public sector;
- Shift to greater competition in the public sector;
- Stress on private-sector styles of management practices;
- Stress on greater discipline and parsimony in public-sector resource
use.
In the Netherlands, as well as in the USA and Great Britain, rightwing
politicians pressed for a NPM type of reform123. Germany was one of the
last Western-European countries to enter the NPM-movement on the fed-
eral level. One explanation is that the German administration has a number
of institutional features (like multi-level government, decentralisation,
multi-functional local government system and subsidiarity) that were al-
ready ‘NPM-proof’ before NPM started as a reform movement. This, plus
its good international reputation of performing in terms of legally correct
and reliable conduct, served as a cognitive and normative barrier against
an easy adaptation of NPM.124 Only after local, regional and state
(‘Länder’) reforms were carried out, and a change of government took
place in 1998, was a comprehensive federal reform program was
launched.125 This was also triggered by skyrocketing public debts caused
by the German unification.126 The reform program of 1999 builds on the
Clinton/Gore rhetoric of NPM, with a vision of ‘an administration which
performs better and costs less.’127
Although New Public Management was adapted globally128, there have
been many differences129. Countries such as Denmark and the Netherlands
123 Ingraham (1996: 247): The reform agenda for national civil service systems:
external stress and internal strains.
124 Wollman (2003): Public-sector reform in Germany: between continuity and
change; Oppen (2002): From ‘New public management’ to ‘New public gov-
ernance’. Restructuring the public administration of tasks in Germany.
125 Bundesregierung (1999): ‘Moderner Staat – moderne Verwaltung’.
126 Wollmann (2003): Public-sector reform in Germany, between continuity and
change – in international perspective.
127 Ibidem.
128 Osborne and Gaebler (1992: 328): Reinventing government.
129 Hood (1996: 270): Exploring variations in public management reform in the
1980s.
had already developed variable pay (an instrument to enhance competition

of civil servants within administration130), in the 1980s, whereas the Ger-
man public sector escaped from such reform.131 Kickert argues that, from a
comparative perspective, the German national administration has an ‘ex-
ceptional’ ability to escape reforms.132 Only in 2005 did Germany issue a
draft Law that aimed at making light forms of variable pay possible.133 Fac-
tors other than macro-economic performance and political preference also
affected the degree of emphasis to NPM, such as the size of the admini-
stration. The Netherlands, a country with a ‘medium’ NPM emphasis, had
a medium sized government at that time. Downsizing the government was
a more prominent issue there than in Germany: a ‘low’ NPM emphasis
country, already with a small government. 134
Gradually the NPM movement began to receive criticism. Techniques
that flourished in the private sector sometimes showed to be inappropriate
for the public sector. Many NPM reform activities attempted to create a
degree of flexibility and discretion that conflicted sharply with the rigidi-
ties created by complicated civil service laws and regulations.135 NPM was
responsible for at least three structural problems (discovered in an interna-
tional survey on NPM in local governments):136
- Quality management often degenerates to a simple instrument of le-
gitimising the administration;
- Ideologically driven privatisation programs end up in short-term, non-
durable solutions;
- Outcome-orientation often falls back to the traditional hierarchical
steering concept.
130 Variable pay is an instrument taken from the private sector. It implies a shift
from intrinsical motivation (as in Weberian hierarchy) of civil servants to-
wards extrinsical motivation (stimulus-response thinking, which is a common
approach in the private sector). In ‘t Veld (personal communication).
131 Hood (1996: 274, 272): Exploring variations in public management reform in
the 1980s.
132 Kickert (2005): Distinctiveness in the Study of Public Management in Europe.
133 Bundesregierung (2005): Entwurf Strukturreformgesetz.
134 Hood (1996: 280): Exploring variations in public management reform in the
1980s.
135 Ingraham (1996: 262-263): The reform agenda for national civil service sys-
tems: external stress and internal strains.
136 Naschold et al. (1997): International trend of local government modernisation.
An assessment for the mid-1990s.
A fourth problem is that NPM suggests that the private sector is, by
definition, superior to the public sector. This contributed to a low morale
in public administration.137 Furthermore, market thinking may threaten de-
mocratic processes; just as network governance does (see hereafter).138
Finally, who considers citizens as customers, restricts the task of gov-
ernments to providing services and products, whereas citizens are also sub-
jects, voters and nationals, in which roles they expect more than only ser-
vice provision.139
Despite the current critique on NPM, market governance has remained
a very influential ideal-type governance style inside Western public ad-
ministration. In its ideal-typical form it accounts for decentralisation, the
creation of agencies, and furthering the autonomy140 of existing agencies
and other decentralised units. It stimulates the formation of hybrid organi-
sations (mixtures of public-sector and private-sector organisations), and
emphasises the management competencies of staff, instead of policymak-
ing competences. It promotes competition instead of co-operation, stimu-
lates benchmarking, stimulates contract management (although contracts
are also used in hierarchical and network governance, but not so predomi-
nantly), and advocates output instead of outcome.
2.2.3 Network governance
Since the 2nd World War, due to the perceived success of the Weberian
model, governments and their administration came to be increasingly
guided by principles of hierarchy and standardisation. In contrast, at the
same time Western society in general developed into an open democratic
system in which networks play an important role.141 This divergence was
problematic because the notion of a network in which actors with different
interests bargain, is fundamentally different from the mono-rational mode
137 Savoie (2000: 8-9): Governance in the twenty-first century: introducing the
topic.
138 Perry (2007): Democracy and the new public service.
139 Ringeling (2001: 34): Rare klanten hoor, die klanten van de overhead.
140 Verhoest et al. (2004: 116, The study of organisational autonomy: A concep-
tual review) distinguish six dimensions of autonomy in market governance:
managerial autonomy, policy autonomy, structural autonomy, financial auton-
omy, legal autonomy and interventional autonomy.
141 Raadschelders and Rutgers (1996: 92): The evolution of civil service systems.
of governance by a single controller.142 Long before the concept of an

emerging ‘network society’143 became popular in the 1990s, politicians and
societal stakeholders promoted the involvement of more parties in policy-
preparation and decision-making since the 1970s. Scholars and practitio-
ners claimed that the classic hierarchical paradigm had to be replaced – at
least for multi-actor, multilevel policymaking - with a network or arena
paradigm.144 Less ‘command and control’ and more ‘open processes’ were
expected to improve the effectiveness of the governmental organisations.145
The ‘IT-revolution’ of the last decades has also increased the importance
of networks in social life.
Against this background, from the 1980s (in the Netherlands) and
1990s (in most other western European public administrations), the ideal-
type network governance developed as an alternative to hierarchical top-
down steering by government, and to market governance as promoted un-
der the banner of New Public Management. It provides a third alternative
between top-down planning and the anarchy of the market. Whereas mar-
ket governance offered the alternative of deregulation and privatisation to
the inefficiency of hierarchical governance in our complex societies, net-
work governance offered the alternative of horizontal coordination.146
Network governance is considered to offer advantages for learning and in-
novation in an ever-changing environment.147 Powell describes network
governance as “more a marriage than a one-night stand, but there is no
marriage license, no common household, no pooling of assets”.148
Network governance can be defined as the ‘management’ of complex
networks, consisting of many different actors from the national, regional
and local government, from political groups and from societal groups
(pressure, action and interest groups, societal institutions, private and busi-
142 Hanf and Scharpf (1978): Interorganizational policy-making; limits to central

coordination and control. Kickert (1980): Organization of decision-making
(cited in Kickert (1991)).
143 Castells (1996): The rise of the network society.
144 Dirven et al. (1998: 14-15): Stuur of overstuur: Over bestuurlijke
wisselwerkingen tussen overheid en samenleving.
145 E.g. Ringeling (1997: 16-17): Sturing van het milieubeleid.
146 Börzel (1998: 264): Organizing Babylon – on the different conceptions of pol-
icy networks.
147 Jessop (2002: 8): Governance, governance failure and metagovernance.
148 Powell (1991: 269): Neither market nor hierarchy: network forms of organiza-
tion.
ness organisations).149 150 Public sector reform programmes of the early

2000s concentrate on this objective to a substantial extent. Network gov-
ernance has also become popular in public management consulting.151
However, this does not imply that network governance has been widely
implemented. The introduction of network governance in the public sector
is sometimes a symbolic rather than a serious attempt (in Germany, the
term ‘Ankündigungs-politiken’152 has been coined to describe this phe-
nomenon)153. The reason for this may be that politicians tend to (ab)use
participatory approaches as a way of increasing support.154 An example is
the introduction of co-regulation with societal partners by the European
Commission. This has been analysed as an attempt to develop more hierar-
chical power, against the will of the member states.155 In the Netherlands,
it was found that policy processes that begin with a network approach end
with a classical top-down approach that destroys the trustful relations. This
creates an often-seen network governance paradox: when government only
half-heartedly invites citizens and societal partners to participate in poli-
cymaking, the result may be a decrease of citizens’ trust in government.156
Meyer identified four key elements of network regulation: Trust, dura-
bility, strategic dependency, and institutionalisation.157 Trust is a more ef-
fective means of dealing with knowledge-intensive organisations than
price and authority.158 Empathy should be added as key element: the will
149 Kickert (1997: 735): Public governance in the Netherlands.

150 High et al. (2005, Understanding informal institutions) argue that ‘network
management’ is contains a contradiction: networks are characterised by per-
sonal and informal relationships, which makes them in to a certain extent re-
sistant to ‘management’.
151 Eggers and Goldsmith (2004): Government by network. The new public man-
agement imperative.
152 ‘Policy by proclamation’.
153 Hesse (2003: 199): Stability Turned Rigidity. Paradoxes in German Public
Sector Development.
154 Mayer et al. (2005: 197): Interactive policy development: Undermining or
sustaining democracy?
155 E.g. Héritier (2002): New modes of governance in Europe: Policy making
without legislating?
156 Meuleman (2003: 19): The Pegasus Principle.
157 Meyer (2003), cited in W. Meyer and Baltes (2004: 37): Network failures –
How realistic is durable cooperation in global governance?
158 Adler (2001: 215): Market, hierarchy, and trust: The knowledge economy and
the future of capitalism.
and ability to understand the interests of other network partners and to act
accordingly. Thorelli adds power, a “cousin” of trust, as a central concept
in networks: the ability to influence the decisions of others.159
Klijn and Koppenjan distinguish five other characteristics of network
governance: 160
- Mutual dependence of actors which leads to sustainable relations be-
tween them;
- In the course of interactions, rules are formed which regulate actor
behaviour;
- Policy processes are complex and not entirely predictable because of
the variety of actors, perceptions and strategies;
- Policy is the result of complex interactions between actors who par-
ticipate in concrete games in a network;
- Network co-operation is not devoid of problems and needs process
and conflict management, and risk reduction.
According to Considine there are three domains in which network gov-
ernance is making an important contribution to public governance: interor-
ganisational networks (linking public and private organisations), inter-
actor networks (linking leaders and advocates), and inter-agency networks
(linking various agents in the provision of services).161
The ideal-type network governance currently accounts for interactive
policymaking, informal networks such as expertise networks in public ad-
ministration, and covenants.
Network governance and knowledge based network organisations
The term network governance implies a focus on a certain type (namely in-
terdependent) type of relations, for example between a ministry and socie-
tal stakeholders. A network organisation will use, or be a partner in net-
work governance, but differs from other organisations participating in
network governance in the sense that its internal organisation is based on
networking as the main (or even only) coordination and organisation prin-
ciple. This type of organisations has been successful in knowledge-
intensive businesses, which are sometimes called ‘knowledge-based net-
159 Thorelli (1986: 38): Networks: between markets and hierarchies.

160 Based on Klijn and Koppenjan (2000: 142): Public management and policy
networks. Foundations of a network approach to governance.
161 Considine (2002: 7): Joined at the lip? What does network research tell us
about governance?
work organisations’162. Networks work well in knowledge-rich environ-

ments because they have superior information-processing capabilities; they
are also more adaptable and flexible than hierarchies because of their loose
coupling and openness to information.163 Such organisations have in com-
mon that they focus on content in stead of power, and are designed in a
way that their highly professional employees have maximum freedom of
operation. The Dutch consultancy organisation ‘The Vision Web’ that
emerged in the late 1990s was an extreme example. There was no internal
hierarchy (to the extent that employees decided on their own salaries), no
managers; everything was based on trust, people and identity.164
Like all styles of governance, network governance also has its typical
weak points.165 Networks are instable constructions that tend to either dis-
integrate, or convert into a formal organisation. They are not very efficient
compared to markets and hierarchies. Furthermore, the advantage (com-
pared to hierarchies) that networks are open, can also be a threat to another
key element, namely trust, because trust relates to team building in a net-
work (and therefore, closure is important). Sørensen argues that network
governance marginalises politicians and thereby weakens democracy.166 It
‘stretches’ democracy and raises issues regarding equity, accountability
and democratic legitimacy.167 Depending on how democracy is defined,
this may be problematic. Klijn and Koppenjan differentiate between an in-
strumental vision (democracy is an efficient way of decision making) and a
substantive view (democracy is a societal ideal, a value in itself). Propo-
nents of participatory or direct democracy usually take latter view.168
A final weakness of network governance worth mentioning here is in-
herent to networks. People with a higher than average number of ‘links’
with others, play an important role in networks. These ‘hubs’ guarantee
high speed communication. However, if such hubs are removed, networks
162 Term coined by Roobeek.

163 Achrol and Kotler (1999: 146-147): Marketing in the network economy.
164 Derix (2000): The Vision Web. Op reis naar ’s werelds spannendste
ondernemingsvorm.
165 Meyer and Baltes (2004: 42): Network failures – How realistic is durable co-
operation in global governance?
166 Sørensen (2006): Meta-governance: The changing roles of politicians in proc-
esses of democratic governance.
167 Bogason and Musso (2006: 3): The democratic prospects of network govern-
ance.
168 Klijn and Koppenjan (2000: 376-377): Politicians and interactive decision
making: Spoilsports or playmakers.
may break down into isolated pieces.169 Moreover, hubs establish a kind of
hierarchy in a network, and actors who find themselves in such a key posi-
tion, have an advantage over other players.170 This makes the selection of
participants in network governance processes problematic. In addition,
there are no commonly accepted procedures for selection, and the risk is
that privileged actors join in technocratic decision making which may re-
sult in a decrease of citizens’ participation compared to the classical repre-
sentative forms of democratic decision-making.171
2.2.4 Forms of network governance
It is no surprise that a whole range of sub-forms of network governance

have developed: large numbers of Western-European public administration
and political science scholars have concentrated on the governance impli-
cations of Castells’ emerging ‘network society’. Each of the sub-forms
highlights one or more characteristics of network governance. They share
the normative idea that networks are the best way of societal coordination.
Participatory governance (or interactive policymaking) is a normative
concept172 that promotes individual and collective participation of state and
non-state actors in policy-making and implementation.173 Because of the
focus on participation this concept is usually compared with hierarchical
governance (which is in principle not participatory), and not with market
governance.174 During the 1990s, interactive policymaking became en
vogue in the Netherlands.175 Around the Millennium, a revival of hierarchi-
cal thinking in the (national) political arena resulted in the end of several
169 Barabási (2003): Linked. How everything is connected to everything else and
what it means for business, science and everyday life.
170 Thompson (2004: 413): Is all the world a complex network?
171 Von Blumenthal (2005: 1165): Governance – eine kritische Zwissenbilanz.
172 Evans (2007): Methodological problems in evaluating democratic participa-
tion in local politics.
173 E.g. Lovan et al. (eds.)(2004): Participatory governance: Planning, conflict
mediation and public decision-making in civil society. Laws et al. (2001, Pub-
lic Entrepreneurship Networks) argue that policy development “increasingly
occurs in an intermediate arena that is neither governmental nor private”.
174 See for example Pröpper and Steenbreek (1998): Interactieve beleidsvoering.
Typering, ervaringen en dilemma´s.
175 Van der Arend (2007): Pleitbezorgers, procesmanagers en participanten.
experiments with this form of network governance.176 Nodal governance177

is a structured form of participatory governance in which state and non-
state actors co-operate to provide public services. It operates through net-
works and partnerships of governance ‘nodes’ that include state agencies,
businesses and NGOs.178
Bang’s concept of culture governance179 developed in the consensus
oriented society of Denmark and is a difficult mixture: it “represents a new
form of top-down steering; it is neither hierarchical nor bureaucratic but
empowering and self-disciplining”. Not only does this seem to neglect
some of the inherent incompatibilities between hierarchies and networks,
the normative character also includes that this approach is considered to be
a generally applicable style combination.
Community governance is a form of governance in which governments
appeal to citizens to form communities that look after their own affairs,
such as safety. This form is related to community policing (see Chapter
5).180
Whereas participatory and nodal governance position government as a
key actor in processes of solving societal problems, deliberative govern-
ance 181 is ‘deliberately’ anti-statist and focuses on societal processes of de-
liberation as the crucial ‘modern’ problem-solving mechanism. It is based
on the idealistic view of democracy, distinguished by Klijn and Koppen-
jans.
Other approaches consider dealing with complexity as the main chal-
lenge for (network) governance. They do not concentrate on the interac-
tions between actors. Reflexive governance is a form of network govern-
ance that builds on Beck’s notion of reflexive modernisation:182 complexity
and uncertainty require that governance approaches are reflexive, i.e. the
governance approach and the ‘governors’ itself are influenced, or should
176 Meuleman (2003): The Pegasus Principle.

177 Shearing (2001): A nodal conception of governance.
178 Froestad and Shearing (2004): Contested authorities, mobilizing local knowl-
edge and capacity for nodal governance participation.
179 Bang (2004): Culture governance: Governing self-reflexive modernity.
180 Schedler (2006: 121): Networked policing: Towards a public marketing ap-
proach to urban safety.
181 Hajer et al. (2003): Deliberative policy analysis: understanding governance in
the network society.
182 Beck (1994): The reinvention of politics: towards a theory of reflexive mod-
ernisation.
be open for influences by the governance environment. 183 In reflexive

governance, learning is a central issue. Adaptive governance is a way of
designing policies that can adapt to a range of anticipated and unantici-
pated conditions. A key characteristic of adaptive governance is collabora-
tive, flexible and learning-based issue management across different
scales184. Adaptive management is a form of reflexive governance: it is a
structured learning process designed to dealing with uncertainty.185
2.2.5 Post-bureaucratic administration, a myth?
The internal logic of the three ideal-types makes them so attractive that
each of them has been considered a panacea for all administration prob-
lems: hierarchy in the post-War decades, market governance during the
1980s and 1990s, and network governance since the mid-1990s. Bouckaert
distinguishes three phases in the evolution of public administration in
Western states since the 1950s.186 Between 1950 and 1980 hierarchical
governance was central: the public sector should be above all professional
and rational. From 1980-1995 market governance was introduced (private
sector instruments and procedures: New Public Management (NPM)). In
the period of 1995-2000s, New Public Governance emerged: a combina-
tion of network governance and hierarchical governance (the latter under
the banner of ‘good governance’).
Was this evolution also a succession of the relative dominance of gov-
ernance styles over time, from hierarchy to network via market govern-
ance? Sol argues that the idea of succession is a simplification and feeds
the myth of progress.187 It is a simplification, because there have been sig-
nificant differences in the governance history of individual countries, dif-
ferent public-sector organisations inside the same country and even differ-
183 E.g. Voss et al. (eds) (2006): Reflexive governance for sustainable develop-
ment.
184 Description retrieved on 9 September 2007 from http://www.ias.unu.edu/
sub_page.aspx?catID=155&ddlID=299 (The adaptive governance project,
UNU-IAS). See also Nooteboom (2006): Adaptive governance for sustainable
development.
185 Sendzimir et al.(2006): Adaptive management to restore ecological and eco-
nomic resilience in the Tisza river basin.
186 Bouckaert (2004: 27): Die Dynamik von Verwaltungsreformen.
187 Sol (2000): Government governance and beyond. Reconciling flexibility and
accountability in labour market policy in the Netherlands.
ent organisational units within one public-sector organisation. The myth of

progress is criticized by Hesse in the case of Germany.188 He notices that
public administration reform is characterised by symbolic politics rather
then by real reforms. This may be illustrated by the view of the European
Commission on the participation of networks (i.e. in this case: non-
governmental actors). In its White Paper on Governance, the Commission
states: “By making them more open and structuring better their relation
with the Institutions, networks could make a more effective contribution to
EU policies”189, which can be considered “rather paternalistic advice”.190 In
the White Paper, stakeholder involvement is interpreted as a way of pro-
viding the Commission with information and not as a process of dialogue.
If there has not been a succession from hierarchy to network via market
governance, then all three modes of governance should exist together. This
seems to contradict the popular idea that a ‘post-bureaucratic administra-
tion’ is emerging191. The term ‘bureaucracy’ is used for the politico-
administrative system of performance in modern times, which is mani-
fested in a fixed pattern of responsibilities and jurisdictions, and a hierar-
chical pyramid-shaped structure.192 The notion of post-bureaucratic ad-
ministration implies the opposite: abandon the classical bureaucracy,
replace fixed responsibilities with fluid ones and hierarchical structure
with a flat organisation.193 The question is: has there been a metamorphosis
of the politico-administrative system towards a post-bureaucratic type, or
is this wishful thinking?
Many public administration scholars hold that a change has indeed oc-
curred. Their argument has some plausibility. Western societies are said to
have transformed into ‘network societies’.194 In addition, it was widely sig-
nalled that public-sector organisations needed to change as early as the
1970s and 1980s: “Bureaucracies tend to concentrate on organisational
survival rather than on attending problems of governance”, and “Bureau-
188 Hesse (2003: 199): Stability Turned Rigidity. Paradoxes in German Public
Sector Development.
189 European Commission (2001: 18): European Governance. A white Paper.
190 In ‘t Veld (2003: 52): Governance: A new concept leading to policy innova-
tion?
191 Emery, Wyser and Sanchez (2006): Working in a post bureaucratic context:
civil servants’ perceptions of the main challenges involved and their coping
strategies.
192 König (2003: 450): On the typology of public administration.
193 Heckscher (1994): Defining the post-bureaucratic type.
cratic government is a threat to those who see the central position of bu-
reaucracy in modern policy-making as a threat to traditional values. It is
also a threat to those who desire an effective government”.195 The Dutch
scholar Frissen argues that hierarchical governance is decreasing, because
of the IT revolution.196 Bogason and Toonen conclude that hierarchical
control of government over society is not impossible, but is restricted to
politically and technologically simple fields that require simple human
tasks of intervention.197 Moreover, they argue, “Not many areas of human
activity meet these demands”. Others believe, that “the nature of tasks of
governments in contemporary complex societies are confronted with will
not allow for command and control reactions”.198 None of these authors,
however, have produced empirical evidence for the supposedly drastic de-
crease of hierarchical governance.
There are many more scholars who hold that hierarchy has gone, or at
least, is weakened. Kettl identifies a fundamental shift in American public
administration.199 He concludes that the four traditions that together formed
the US public administration all depend upon the opportunity to draw clear
lines defining the roles and responsibilities of each of the players. 200 How-
ever, with entering the 21st century, these boundaries have all become
fuzzy. Traditional boundaries can no longer contain big problems.201 The
fluidity of administrative boundaries in Western administrations has led to
a move from the familiar institutions “to the edges of organisational activ-
ity, negotiations between sovereign bodies, and inter-organisational net-
works (…)”.202 After governments had discovered the limits of the first al-
ternative to hierarchy, namely a retreat from the public domain by
introducing market governance elements such as privatisation and deregu-
lation, the development of network governance was perceived as a new
opportunity.203 According to Kickert204, it was already in the Netherlands in
195 Peters (1981: 82): The problem of bureaucratic government.

196 Frissen (1999): De lege staat.
197 Bogason and Toonen (1998: 224): Introduction: Networks in Public Admini-
stration.
198 Klijn and Koppenjan (2000a: 154 ): Public management and policy networks.
199 Kettl (2002: 26-49): The transformation of governance.
200 The Hamiltonian, Jeffersonian, Madisonian and Wilsonian traditions.
202 Hajer and Wagenaar (2003): Deliberative policy analysis: understanding gov-
ernance in the network society.
203 Kickert et al. (1997: 2): Public Management and Administrative Reform in
the early 1970s (long before the New Public Management movement) rec-
ognised that central top-down steering in a hierarchical organisation does
not work in a network of independent actors with different goals, interests
and positions. Elsewhere he argues that the early attention in the Nether-
lands to a network approach is caused by the historically underlying conti-
nuity of a corporatist-consensual model of deliberation and pragmatic
compromise.205
Therefore, if classical bureaucracy is labelled as out-dated, and if there
was strong societal and political pressure to change the administration,
why would a new ‘post-bureaucratic’ administration not have developed?
Kettl’s ‘fuzzy boundaries’ of course do influence the possibility of design-
ing and applying certain governance style mixtures. However, this does
not mean that any of the three ideal-types has been put ‘out of business’.
On the contrary, it seems that public-sector organisations have escaped
much of a transition thus far. According to Bouckaert, during the last dec-
ades of the 20th century they have developed from a hierarchical Weberian
style towards a Neo-Weberian style that is still hierarchical.206 The Neo-
Weberian style has moved further away from the classical hierarchical
style in Anglo-Saxon states than in continental European states. Bouckaert
distinguishes two types: a Nordic type that emphasises the participation of
the citizen-client in a ‘citizens’ state’ with a balance of rights and obliga-
tions, and a continental version that insists on the professional dimension
of the state, with a citizen who is considered as a client.207
Furthermore, recent research shows that hierarchy is still the primary
governance style in Western public-sector organisations.208 A case study in
Queensland, Australia, showed that its bureaucracies have not changed
into post-bureaucratic organisations, but that public sector organisations
have evolved “from one form of bureaucracy based on political controls
and values, to a form of bureaucracy associated with market controls and
Western Europe.
204 Kickert (1997: 737): Public Governance in the Netherlands: An Alternative to
Anglo-American 'Managerialism'.
205 Kickert (2003: 127): Beneath consensual corporatism: Traditions of govern-
ance in the Netherlands.
206 Bouckaert (2004: 22): Die Dynamik von Verwaltungsreformen.
207 Bouckaert (2003): La réforme de la gestion publique change-t-elle les systè-
mes administratifs ?
208 Hill and Lynn (2005): Is Hierarchical governance in decline? Evidence from
empirical research.
values”.209 An investigation in the UK showed empirically that “King Bu-

reaucracy is actually alive and well and, above all, absolutely necessary if
policy is to be implemented on a national basis”210. Hierarchy is “in good
shape”, and like the whisky brand “Johnny Walker” advertises “Born in
1820, still going strong”211. Schofield argues that it is useful for politicians
to have an “obedient cadre of professional managers who are the interface
between themselves and the citizen”; in addition, managers stay obedient
because this is rewarding for them. Hill and Lynn state that hierarchy is
still “the primary means by which governments govern. The seemingly
‘paradigmatic’ shift away from hierarchical government toward horizontal
governing […] is less fundamental than it is tactical.” They conclude that
what has taken place is the addition of new tools or administrative tech-
nologies that facilitate public (network) governance within hierarchical
systems.212
Similar observations have been made in Germany, where in the case of
control over prisons and over local authority in building administration, hi-
erarchical forms of control moved from the foreground to a ‘shadow’ from
which they still are influential.213 In the Netherlands, politicians are trying
to re-hierarchise the public sector.214 There may have been a temporary de-
crease, but the return of hierarchy with new vigour is confirmed for the
UK by Taylor and Kelly. They found that there has been an increase in
rules and accountability procedures, which has “reduced the rule discretion
of the street-level bureaucrat, which crucially, in terms of Lipsky’s the-
ory215, has reduced the policy-making element of discretion”216.
Another indication that bureaucratic (i.e. hierarchical) governance is
still important is that Western governments invest a lot of energy in a bat-
tle against ‘over-regulation’. ‘Better regulation’ is the normative slogan,
209 Parker and Bradley (2004: 197): Bureaucracy or post-bureaucracy? Public

sector organizations in a changing context.
210 Schofield (2001: 91): The old ways are the best?
211 Schuppert (2007: 8): Was ist und wozu Governance?
212 Hill and Lynn (2005): Is Hierarchical governance in decline? Evidence from
empirical research. The authors synthesized 70 journals and 800 articles on
governance and public management over a 12 year period.
213 Lodge and Wegrich (2005: 221): Control over government: Institutional iso-
morphism and governance dynamics in German public administration.
214 Ringeling (2007: 22): Tussen distantie en betrokkenheid.
215 Lipsky (1980): Street-level bureaucracy.
216 Taylor and Kelly (2006: 639): Professionals, discretion and public sector re-
form in the UK: re-visiting Lipsky.
which the European Commission uses to bring and keep down the ‘admin-
istrative burden’ of legislation for businesses and citizens. Many European
countries developed ambitious deregulation programmes in the early
2000s. These programmes usually have two objectives: creating more
freedom for the private sector and a better acceptance of government poli-
cies by citizens. Belgium labelled the national anti-bureaucracy pro-
gramme the ‘Kafka’ project; also in the Netherlands, ‘Kafka brigades’
have been established. ‘Better regulation’ targets are sometimes formu-
lated in terms of a 25 to 40% decrease of legal texts without differentia-
tion.217 However, (legal) instruments are not neutral devices: they express a
certain idea about social control and ways of exercising it.218
Finally, König states matter-of-factly that no interpretable material is
yet available from which the type of a post-bureaucratic administration
might emerge.219 He adds that even private enterprises (still) have bureau-
cratic features, because of its low transaction costs. How is it then possible
that so many scholars deny that hierarchy is still very influential? One rea-
son may be that they are mainly interested in the non-hierarchical dimen-
sions of governance: all the world is chaos and complexity, and this leads
to a new paradigm for public administrators.220 This focus on complexity
alone can be criticised. In the words of Frederickson:
“Investments in our prevailing institutions, our cities, states and nationals
and their established governments are devaluated, as are the accomplish-
ments of those institutions. Order, stability, and predictability are likewise
undervalued.”221
We can conclude that there is no evidence of an emerging post-
bureaucratic public sector. There are merely complex mixtures of old and
newer forms of governance (Figure 3), and none of them can be considered
a panacea.222 However, the ingredients of the mixtures have always been
217 Hey (2007): Deregulierung und Entbürokratisierung.

218 Lascoumes and Le Gales (2007: 1): Introduction: Understanding public policy
through its instruments.
219 König (2003: 459): On the typology of public administration.
220 Kiel (2005): Embedding chaotic logic into public administration thought:
Requisites for the new paradigm.
221 Frederickson (2004:12): Whatever happened to public administration? Gov-
ernance, governance everywhere.
222 E.g. Nelissen (2002, The administrative capacity of new types of governance)
questions the claim that the new modes of governance are a panacea.
there. Even in primarily hierarchical times, consent was often strived for.223
Hierarchy, or ‘bureaucratic management’ in a Weberian sense continues to
exist in a complicated mixture with market and network thinking, and
sometimes disguised as network or market governance. The ‘dinosaur sce-
nario’, which emphasises that hierarchy is undesirable and not viable, and
that a shift toward market or network governance is inevitable, is an insuf-
ficient explanation for contemporary public-sector governance: “Bureau-
cratic organization and the success criteria in which it is embedded are still
with us.”224
Even in the supposedly emerging ‘network society’ of Castells225, it
seems logical for some public-sector tasks to stick to hierarchy. The finan-
cial and salary department of a ministry should be reliable and not net-
working or entrepreneurial. Besides, bureaucratic procedures can be con-
sidered as a safeguard for effectively dealing with crises, disasters, or
security issues. On the other hand, hierarchy has proven to not being able
to solve multi-actor, multi-sector, multi-level problems: they are too
‘fuzzy’.
Fig. 3. Development of three governance styles in Western European nations
Nevertheless, the arguments and the empirical findings mentioned

above that hierarchy is still an important governance style, may not con-
vince believers in the ‘network governance is everything’ doctrine. Their
vision “assumes that a single, context-free set of principles for organizing
223 Blatter performed empirical research in four transboundary zones in Europe

and North America. He considers these zones as laboratories for institutional
change. Blatter (2003: 505): Beyond hierarchies and networks: institutional
logics and change in transboundary spaces).
224 Olsen (2006: 17-18): Maybe it is time to rediscover bureaucracy.
public administration is functionally and normatively superior.”226 Finally,

they may argue that empirical research cannot prove that they are wrong,
because the (quantitative and qualitative) evidence produced by such re-
search is based on an analytical concept that fails to recognise that the cur-
rent society is up to its capillary vessels a network society. Frissen formu-
lates the network doctrine as follows:
“I am no Darwinist. I think that the world is accidental, that it could have
gone differently, that things do not have a purpose. This is a crucial point.
When one looks at societal developments from a policy perspective, it
makes a big difference if one believes that the world is the result of contin-
gency, or that the world has a logic which is understandable, in which one
can intervene with predictable outcomes. That leads to a very different
steering behaviour.”227
To conclude, the arguments in this section suggest that taking one of
the three ideal types as a panacea leads to overlooking important character-
istics and mechanisms of the functioning of contemporary public-sector
organisations. Only by considering all three forms simultaneously, the
strengths and weaknesses of each of the forms can be assessed.228 There-
fore, a multi-perspective approach (hierarchies, networks and markets exist
together) has more analytical power. Beetham came to the same conclu-
sion when he, in 1991, compared several alternatives to the Weberian
model of hierarchy:
“Each of these conceptions has in its time been presented as the final truth.
It would be more plausible, however, to see them, not as mutually exclu-
sive alternatives, either to the Weberian model or to an another, but as each
emphasizing an essential aspect of organizational reality, all of which need
taking into account and which together necessitate a modification in the
strictly bureaucratic conception of organizational efficiency, rather than its
outright replacement.”229
226 Olsen (2006: 13): Maybe it is time to rediscover bureaucracy.

227 Frissen in an interview in the Dutch Staatscourant of 17 May 2006 (translated
from Dutch by the author).
228 Podolny and Page (1998: 74): Network forms of organization.
229 Beetham (1991: 135): Models of bureaucracy.
2.3 Relations between governance styles 45
2.3 Relations between governance styles
2.3.1 Differences between hierarchical, network and market

governance
The three ideal-typical styles of governance are internally consistent: they

each have a clear and distinct internal logic. Hierarchy produces legal in-
struments because they can be controlled hierarchically. A typical network
governance outcome like a consensus builds on mutual trust, not on hierar-
chical power play, and therefore does not fit into the logic of hierarchical
thinking, nor in the competitive logic of market governance. As we have
argued in 2.2.5, the ‘logic of governance’230 of the ideal-types makes them
so attractive, that scholars and practitioners have seen each of them as a
panacea for all administration problems. Table 1 (see also Annex 1) pre-
sents an overview of 36 differences between the three governance styles
that have been collected to support the analytical framework of this re-
search. They are clustered into five groups (see Section 3.2.2): vision (and
strategy), orientation, structure (including systems), people and results.
Table 1. Differences between the three ideal types of governance

Governance style HIERARCHICAL NETWORK MARKET STYLE
STYLE STYLE
Organis. dimension
VISION/STRATEGY
1. Culture/ Hierarchism Egalitarism Individualism
'Way of life' 231
2. Theoretical Rational, positivist Socio- Rational choice,
background 232 constructivist, public choice,
social config. principal-agent
theory theory
3. Mode of calcul.233 Homo hierarchicus Homo politicus Homo economicus
230 Thompson (2004: 413): Is all the world a complex network?

231 Thompson et al. (1990: xiii): Cultural Theory.
232 Dixon and Dogan (2002: 184-185): Hierarchies, networks and markets: re-
sponses to societal governance failure. Bevir and Rhodes (2001: 1): A decen-
tred theory of governance: Rational choice, institutionalism, and interpreta-
tion. Haque (2007: 180): Revisiting the New Public Management.
233 Jessop (2003): Governance and metagovernance: On reflexivity, requisite va-
riety, and requisite irony.
Table 1. (continued)
Governance style HIERARCHICAL NETWORK MARKET

STYLE STYLE STYLE
Organis. dimension
4. Key concept234 Public goods Public value Public choice
5. Primary virtues235 Reliable Great discretion, Cost-driven

flexible
6. Common Minimising risk Satisfying Maximising
motive236 identity advantage
7. Motive of Fear of punishment Belonging to group Material benefit
subordinate
actors237
8. Roles of Government rules Government is Government
government society partner in a delivers services
network society to society
9. Metaphors238 Machine; stick; iron Brain; sermon; Flux; carrot;
fist word invisible hand
10. Style of Planning and design Learning style; Power style;
strategy239 style; compliance to Chaos style: getting
rules and control coping with competitive
procedures unpredictability; advantage
deliberation
11. Governors’ Use of legitimate Persuasion of Negotiate deals
responses to power to coerce rebels to engage, or with rebels, using
resistance240 rebels to behavioural expel them incentives and
conformity inducements
ORIENTATION
12. Orientation of Top-down, formal, Reciprocity, Bottom-up,
organisations 241 internal informal, open- suspicious,
minded, empathy, external
external
13. Actors are seen Subjects Partners Customers,
as clients
14. Choice of Controlled by Free, ruled by trust Free, ruled by
actors242 written rules and reciprocity price and
negotiation
234 Hartley (2004): Paradigms, prizes and paradoxes in governance and public
management.
STYLE STYLE
Organis. dimension
15. Aim of stock- Anticipating Involving stake- Finding reliable
taking of actors protest/obstruction holders for better contract partners
results and
acceptance
STRUCTURE
16. Structure of Line organisation, Soft structure, with Decentralised,
organisations centralised control a minimum level of semi-autonomous
systems, project rules and units/
teams, stable/fixed regulations agencies/teams;
contracts
17. Unit of decision Public authority Group Individual
making243
235 Considine and Lewis (1999: 468): Governance at ground level: the frontline
bureaucrat in the age of markets and networks.
236 Streeck and Schmitter (1985: 122): Community, market, state – and associa-
tions?
237 Streeck and Schmitter (1985: 122): Community, market, state – and associa-
tions?
238 Morgan (1986/1997): Images of organisation. Jessop (2003): Governance and
metagovernance: On reflexivity, requisite variety, and requisite irony.
239 Mintzberg et al. (1998): Strategy safari. A guided tour through the wilderness
of strategic management. Knill and Lenschow (2005: 583): Compliance,
competition and communication: Different approaches of European govern-
ance and their impact on national institutions.
240 Dixon and Dogan (2002: 184-186): Hierarchies, networks and markets: re-
sponses to societal governance failure.
241 e.g. Jessop (2003): Governance and metagovernance: On reflexivity, requisite
variety, and requisite irony. Streeck and Schmitter (1985: 122): Community,
market, state – and associations?
242 Assens and Baroncelli (2004: 7): Marché, Réseau, Hiérarchie : à la recherche
de l'organisation idéale.
243 Arentsen 2001: 501): Negotiated environmental governance in the Nether-
lands: Logic and illustration.
STYLE STYLE
Organis. dimension
18. Control244 Authority Trust Price
19. Coordination245 Imperatives; ex ante Diplomacy; Competition; ex
coordination self-organised post coordination
coordination
20. Transactions246 Unilateral Multilateral Bi- and
multilateral
21. Flexibility247 Low Medium High
22. Commitment Medium to high Medium to high Low
among parties248
23. Roles of com- Communication Communication for Communication as
munication249 about policy: giving policy: organising policy:
information dialogues incentives,
PR campaigns
24. Roles of Expertise for Knowledge as a Knowledge for
knowledge250 effectiveness of shared good competitive
ruling advantage
25. Access to Partial: Segregated Partial: Fragmented Total, determined
information251 information information by price
244 Davis and Rhodes (2000:18): From hierarchy to contracts and back again: Re-
forming the Australian public service.
245 Thompson (2003: 48): Between hierarchies and markets.
246 Susskind (1999: 6-18): An alternative to Robert’s Rules of order for groups,
organisations and ad hoc assemblies that want to operate by consensus.
247 Powell (1991: 269): Neither market nor hierarchy: network forms of organisa-
tion.
248 Powell (1991: 269): Neither market nor hierarchy: network forms of organisa-
tion.
249 Rijnja and Meuleman (2004: 35): Maken we beleid begrijpelijk of maken we
begrijpelijk beleid?
250 In ‘t Veld (2000): Willingly and knowingly. The roles of knowledge about na-
ture and the environment in policy processes. Adler (2001: 215): Market, hi-
erarchy, and trust: the knowledge economy and the future of capitalism.
251 Assens and Baroncelli (2004 : 7): Marché, Réseau, Hiérarchie : à la recherche
de l'organisation idéale.
STYLE STYLE
Organis. dimension
26. Context252 Stable Continuous change Competitive
PEOPLE
27. Leadership253 Command and Coaching and Delegating,
control supporting enabling
28. Empowerment Low Em powered lower Empowered senior
Inside officials managers
organisation254
29. Relations255 Dependent Interdependent Independent
30. Roles of public ‘Clerks and ‘Explorers’ Efficiency and

managers256 martyrs’ producing market
public value maximisers
31. Competences of Legal, financial, Network Economy,
civil servants project management, moderation, marketing, PR
information process
management management,
communication
32. Values of civil Law of the jungle Community Self-determination
servants257
33. Objectives of Training is Training helps Training
management alternative form of ‘muddling through’ stimulates efficient
development258 control over decisions
subordinates
management.
253 Hersey and Blanchard (1982): Management of organizational behaviors: Util-
izing human resources.
254 Peters (2004: 2): The search for coordination and coherence in public policy.
management.
257 Laske (2006: 32): Measuring hidden dimensions. The art and science of fully
engaging adults.
258 Simon (1997: 13): Administrative behaviour. Termeer (1999: 92): Van sturing
naar configuratiemanagement.
RESULTS
34. Affinity with Crises, disasters, Complex, Routine issues,
problem types259 problems that can be unstructured, multi- non-sensitive
solved by actor issues issues
executing force
35. Typical Ineffectiveness; Never-ending talks, Inefficiency;
failures260 red tape no decisions market failures
36. Typical types of Laws, regulations, Consensus, Services, products,
output and control, procedures, agreements, contracts, out-
outcome261 reports, decisions, covenants sourcing, vol.
compliance, output agreements
2.3.2 Governance styles, complexity and ambiguity
Administrative organisations are, maybe even more than business organi-

sations, characterised by complexity and ambiguity. The ‘garbage can’
model presented by Cohen et al. in 1972 defines an organisation as “a col-
lection of choices looking for problems, issues and feelings looking for de-
cisions and situations in which they might be aired, solutions looking for
issues to which they might be the answer, and decision makers looking for
work”.262
The question is how the use of ideal-typical governance styles can help
to analyse organisational behaviour. The garbage can model with its four
‘streams’ that are trying to find each other, predicts that ‘pure’ hierarchi-
cal, network or market governance is not feasible: a fixation on one of the
styles would block some of the streams of problems, solutions, choice op-
portunities and actors. However, if we consider the three governance styles
as forces influencing these streams a preoccupation with one of the styles
would predict a certain affinity within these streams (Table 2).
259 EEAC (2003): European governance for the environment.

261 De Bruijn and Ten Heuvelhof (1995: Netwerkmanagement); Hood (2003: Ex-
ploring variations in public management reform of the 1980s).
262 Cohen et al. (1972: 2): A garbage can model of organizational choice.
Table 2. Governance styles ‘inside the garbage can’.
Style Hierarchy Network Market

Garbage can
Problems Crises Unstructured Routine problems
problems
Solutions Rules, control Consent, content, Contracts, services
procedures agreements
Choice Focused on Focused on result Focused on
opportunities rule-making via dialogue bargaining
Actors Subordinates Partners Buyers and sellers
As we have seen in the above sections, public administration organisa-
tions, as part of the politico-administrative system, apply complex mix-
tures of three governance styles – hierarchy, network and market – within
a dominant hierarchical frame. Such mixtures are not always productive.
They may undermine each other, as will be illustrated in the next section.
2.3.3 How hierarchical, network and market governance

undermine each other
A mixture of hierarchy, market and network may produce a variety of con-

flicts. Rhodes considers networks, markets and (hierarchical) bureaucracy
as rivalling ways of allocating resources and co-ordinating policy and its
implementation.263 Baltes and Meyer argue that the main source of ‘net-
work failures’ is the “dualistic pressures from both market and hierarchy”
on the network coordination principle.264 A major reason why the conflict
potential is high is, as mentioned before, that the three styles express dif-
ferent types of relations with other parties: dependency (hierarchy), inter-
dependency (network) or independency/autonomy (market).265 A hierarchi-
cal command and control style of leadership will seldom lead to a
consensus (network style) – even if this was the only feasible outcome of a
policy process, that government is not able to ‘steer’ with legal instru-
ments. Decentralisation or outsourcing (a typical market governance strat-
egy) makes actors more autonomous. They will be frustrated when detailed
263 Rhodes (2000: 345): The Governance Narrative: Key Findings and Lessons
from the ESRC'S Whitehall Programme.
264 Meyer and Baltes (2004: 46): Network failures.
control mechanisms are (re)introduced (hierarchical governance). The co-

existence of ‘new modes of governance’ with compulsory regulation, or
hierarchy, is problematic.266
Other examples of possible conflicts between three pairs of governance
styles are given below. Most examples are taken from ‘external’ govern-
ance-mixture conflicts: they emerge in the relation between administration
and societal actors.267 However, they are usually ‘mirrored’ inside the ad-
ministration.
Conflicts between hierarchical and network governance
From the perspective of the classical hierarchical governance style, net-

work governance is problematic because “governments, like the church,
will find networks messy and carp at the mess”.268 Internal competition
with the traditional hierarchical governance style is one of the reasons that
the introduction of network governance sometimes fails. This competition
has led to obstruction from other public-sector organisations or other parts
of the same organisation, and to unreliable behaviour (not keeping prom-
ises, sudden withdrawal of negotiation mandate).269 Network governance
has also met some resistance caused by distrust and irritation, when net-
work governance is a disguise for (re)gaining control and (hierarchical)
steering information.270 Klijn and Koppenjan concluded that experiments
with network governance in the Netherlands often remain marginal and
half-hearted, because government hesitates when abandoning existing rou-
tines and to giving up unilateral power.271
When hierarchical (‘vertical’) and network (‘horizontal’) steering are
applied at the same time by one public administration organisation, para-
doxical situations appear, in which this organisation ends up in a ‘split’.
Kalders et al. investigated nine cases in the Dutch public sector and found
five typical tensions between hierarchical and network governance:272
266 Eberlein and Kerwer (2004: 121): New Governance in the European Union: A
theoretical perspective. Héritier (2002): New Modes of Governance in
Europe.
267 Most governance literature focuses on the external component of governance.
268 Bevir et al. (2003: 206): Comparative Governance: Prospects and Lessons.
269 Meuleman (2003: 39-41; 203): The Pegasus principle.
270 Bauer (2002: 778-779): The EU 'Partnership Principle': Still a Sustainable
Governance Device Across Multiple Administrative Arenas?
271 Klijn and Koppenjan (2000: 155): Public management and policy network.
272 Kalders et al. (2004: 339-343): Overheid in spagaat. Over spanningen tussen
- The ‘double hat’ problem: the administration combines hierarchical

and network governance in a way that is counter-productive. Volun-
tary agreements273 are frustrated by strict accountability procedures
for the same policy issue;
- The ‘steering split’: an actor wants to comply to norms and expecta-
tions that come from hierarchical and network relations simultane-
ously;
- The ‘accountability curve’: a decentralised government is held ac-
countable for the performance of its partner organisations with whom
it does not have hierarchical relations;
- ‘Horizontal disguise’: a network instrument such as a covenant is
used in a hierarchical way, when the central government unilaterally
decides on the rules of the game;
- The ‘vertical reflex’: (a) bottom-up, if decentralised governments ask
central government for more direction, or (b) top-down, if central
government forces decentralised governments to start network coop-
eration with its partners, within a very strict framework.
Network-style ‘interactive decision-making’ can lead to major tensions
and conflicts with hierarchy when elected politicians, who have the formal
authority to take final decisions, reverse a consensual outcome of an ‘in-
teractive’ process.274 Edelenbos and Teisman developed governance
mechanisms that link hierarchical and network principles in a productive
way. 275 However, like Kalders et al. they did not include an analysis of the
third power, the market governance paradigm with drivers like price (cost-
effectiveness) and autonomy. To take an example: one of their mecha-
nisms is improving the management of expectations about the degree of in-
fluence stakeholders will have on formal decision-making. The problem
here may be that hierarchy (rules, formal decision power) may be predict-
able and reliable, but market thinking is not: a government agency with an
autonomous position will be considered still part of government by the
public, but does not have to follow hierarchical instructions anymore. In
other words, management of expectations is feasible in the relation be-
verticale en horizontale sturing.

273 Kalders et al. consider voluntary agreements a network governance instru-
ment, but it seems to be more related to market governance: such agreements
are forms of performance contracts.
274 Kickert, 2003: 126): Beneath consensual corporatism: Traditions of govern-
275 Edelenbos and Teisman (2004): Interactief beleid en besluitvorming.
tween hierarchy and networks, but not when market thinking is also in-
volved.
Conflicts between hierarchical and market governance
A basic contrast between hierarchical governance and market governance

is the respective affinity to centralisation, and to decentralisation.276 The
wish of politicians to exercise hierarchical control over what happens in
policy networks is, according to Sørensen and Torfing, often constrained
by the market governance discourse that aims to establish a rigid boundary
between the ‘steering’ of politicians and the (autonomous) ‘rowing’ of
public administrators.277 From a market governance perspective, hierarchy
is too inflexible. Classical bureaucracies are considered to try to organise
and dominate markets.278 The market-oriented move to decentralisation and
to create more agencies in the Netherlands during the 1980s and 1990s has
had negative implications for ministerial responsibility, the political pri-
macy, and democratic control.279 Considine concludes that accountability
procedures (hierarchy), demands of contracting-out and output-based per-
formance (market) in three Anglo-Saxon countries and the Netherlands
were contradictory.280
Conflicts between market and network governance
Market governance has the potential to conflict with network governance

on the way decisions are made. Competition in a market setting asks for
quick decisions of independent actors, who strive to optimise their own in-
terests. Decision-making may take a lot of time in a network setting.
Moreover, the type of decision, a consensus, may not be the optimal out-
come for actors’ competitiveness. The interdependency of actors in a net-
work governance setting may conflict with the autonomy a market ap-
proach demands. Network governance relies on trust; hierarchical and
276 Levy (2003: 553): Critical success factors in public management reform: the
case of the European Commission.
277 Sørensen and Torfing (2005): Democratic anchorage of governance networks.
278 Machado and Burns (1998): Complex Social Organization: Multiple Organiz-
ing Modes, Structural Incongruence, and Mechanisms of Integration.
279 Kickert (2005: 23): Lessen uit het verleden. Onderzoek naar veranderoperaties
bij de overheid.
280 Considine (2002): The end of the line? Accountable governance in the age of
networks, partnerships and joined-up services.
market attitudes can damage the trust between network partners.

To conclude this section: Governance style conflicts outside and inside
public-sector organisations exist and may produce serious performance
problems. How, and to what extent, can these conflicts be prevented and
mitigated and how can synergy be stimulated inside these organisations?
2.3.4 Combining hierarchical, network and market governance
There is evidence that productive mixtures of elements of hierarchical,

network and market governance are possible. In their analysis of the Aus-
tralian public sector Davis and Rhodes argued that “to mix the three sys-
tems effectively when they conflict with and undermine one another” is an
important challenge. 281 Steurer argues that policy integration needs a hy-
brid administrative approach, combining hierarchical, market and network
models.282 The possible synergy of governance styles can be illustrated
with an analogy to similar control mechanisms in the business sector: price
(market governance), authority (hierarchical governance) and trust (net-
work governance). These control mechanisms in economic transactions be-
tween actors “can be combined in a variety of ways”; (…) “In a so-called
plural form, organisations simultaneously operate distinct control mecha-
nisms for the same function”.283
Such mixtures are situational, and the factor of time plays an important
role. Lowndes and Skelcher gave an empirical example of how governance
style combinations differ in different phases of a process. 284 They distin-
guish four phases in the life cycle of public partnerships in the field of ur-
ban regeneration: pre-partnershop collaboration; partnership creation and
consolidation; partnership termination and succession. In the phase of pre-
partnership collaboration, networking between individuals and organisa-
tions is emphasised. In the phase of partnership creation and consolidation,
hierarchy is used to incorporate some organisations, and to formalize au-
thority in a partnership board and associated staff. In the phase of partner-
281 Davis and Rhodes (2000: 25): From hierarchy to contracts and back again:
Reforming the Australian public service.
282 Steurer (2004: 1): Strategic public management as holistic approach to policy
integration.
283 Bradach and Eccles (1989: 97): Price, authority, and trust: From ideal types to
plural forms.
Partnerships: an Analysis of Changing Modes of Governance.
ship programme delivery, market mechanisms of tendering and contractual

agreements are applied. Hierarchy takes care of regulation and supervision
of contractors, and networking assists in production of bids and manage-
ment of expenditure programmes. In the last phase, partnership termination
and succession, networking between individuals and organisations is used
as a means of maintaining agency commitment, community involvement
and staff employment.
The temporal dimension of successful governance mixtures is also in-
fluenced by other situational factors, such as the type of problems that are
addressed. In 2003, the network of European environment and sustainable
development advisory councils (EEAC) advised the European Commission
and the EU member states to use a heuristic decision scheme for choosing
the best governance styles combination.285 EEAC proposed hierarchical
governance as, in general, best suited for urgent issues; network govern-
ance for complex multi-stakeholder and multi-level issues; and market
governance for emerging issues which, as far as is known, have relatively
little impact on other stakeholders.
Complementarity of hierarchy and network
A case study on policy changes in the Dutch Housing Ministry concludes

that hierarchical and network types of strategies are often situationally
combined.286 The initiative for a network approach often begins with a hi-
erarchical decision. Another example comes from an analysis of partner-
ships between police departments and community development corpora-
tions. It was noticed that networking strategies were used to establish the
hierarchical structures within which action takes place thereafter.287
Complementarity of hierarchy and market
An example of synergy between hierarchical and market governance, seen

in the United States, is that the promotion of autonomy within the public
sector (market governance) was produced by a top-down method that in-
cluded detailed descriptions of the organisations that would be formed
(hierarchical governance).288
285 EEAC (2003: 26): European governance for the environment.

286 Koffijberg (2005: 363): Getijden van beleid: omslagpunten in de
volkshuisvesting.
287 Thacher (2004: 123): Interorganisational partnerships as inchoate hierarchies.
288 Hesse et al. (2003: 14): Paradoxes in Public Sector Reform. An International
Comparison.
2.4 Governance: A cultural perspective 57
Complementarity of network and market
Synergies have also been found between market and network governance.
The introduction of market techniques has resulted in a fragmented institu-
tional infrastructure of the public sector; networks put it back together
again.289 Poppo and Zenger showed empirically, that managers in inter-
organisational relationships may use contracts (market) and ‘relational
governance’ (network) as compliments: this results in more customised
contracts.290
When the ideal-types hierarchical, network and market governance ap-
pear in combinations, how do the movements that foster one ideal-type in-
fluence these mixtures? The introduction of both market governance and
network governance provoked a hierarchical counter-reaction. For exam-
ple, the introduction of market techniques in the Dutch public administra-
tion contributed to the return of traditional Weberian issues in the mid-
1990s: integrity, accountability, supervision, control, trust in government
and reliability of bureaucracy.291 And as a reaction to the emergence of
network governance, hierarchical concepts like ‘ministerial responsibility’,
‘democratic control’ and ‘primacy of politics’ have been reintroduced in
the Netherlands.292 Some of these Weberian issues (for example control)
frustrate horizontal co-operation and others (like integrity, stability, reli-
ability) may be a necessary complement to network techniques.293
2.4 Governance: A cultural perspective
2.4.1 Governance styles as ‘ways of life’
Why are the ideal types sometimes so fiercely defended? Why are discus-
sions between advocates of market governance and of network governance
289 Davis and Rhodes (2000: 21): From hierarchy to contracts and back again:
Reforming the Australian public service.
290 Poppo and Zenger (2002): Do formal contracts and relational governance
function as substitutes or complements?
291 Toonen (2003: 234): Substance came with little hype. Public sector reform in
the Netherlands.
293 Meuleman (2003: 202-203): The Pegasus Principle.
like religious fights? Why do some consider each of the styles as a panacea
for all problems? The answer lies in the fact that governance styles are,
apart from politico-administrative structures, also belief systems. Govern-
ance is a form of social coordination and governance styles reflect specific
sets of shared values and beliefs and certain patterns of interpersonal rela-
tions. This makes them cultures, or at least images of cultures.
Culture is an important aspect of political and public administration
science that was neglected for a long time.294 One of the reasons was that it
did not fit in the dominant paradigm of the post-war period: rational choice
theory. Aaron Wildavsky was one of the main scholars who focused on re-
introducing the importance of culture to political sciences. Together with
Thompson and Ellis295, he distinguished five ‘ways of life’. Three have ac-
tive interpersonal relations, and two do not, but for different reasons. It
seems that the three socially active ways of life align with the three ideal
typical governance styles we have distinguished above: hierarchism (hier-
archy), individualism (market), and egalitarism (network296). Furthermore,
Thompson et al. notice that, like the three governance styles, these three
ways of life compete with each other, often in a hostile way, but on the
other hand require one another, and they therefore continue to co-exist.297
This co-existence often takes the shape of mixtures: “That what we today
define as free societies – those with the rule of law, alternation in office,
and the right to criticise – are a product of the interpenetration of hierarch-
ism, individualism, and egalitarism”.
How do the two other ‘ways of life’ relate to the concept of governance
styles? Fatalism, a ‘no trust’ style that is found in Southern Italy for exam-
ple, and is a risky culture for public administration: “Where fatalism is en-
demic, democracy cannot survive”. Fatalism relates to hierarchism: “Fatal-
ism generates (and is generated by) authoritarian political systems”298.
However, is fatalism a separate governance style? It looks more like socie-
tal behaviour resulting from an extreme hierarchical governance approach.
When governance styles are modes of co-ordination, then fatalism cannot
294 Geva-May (2002: 388): From theory to practice. Policy analysis, cultural bias
and organizational arrangements.
295 Thompson et al. (1990: 1, 5): Cultural Theory.
296 Networks, communities, clans or egalitarism are terms used for the ‘third’
form of coordination, besides hierarchies and markets (Entwistle et al., 2007:
65. The dysfunction of markets, hierarchies and networks in the meta-
governance of partnerships).
297 Thompson et al. (1990: 50, 257): Cultural Theory.
298 Thompson et al. (1990: 256): Cultural Theory.
be a mode of governance, because it, according to Wildavsky299, denies the

possibility of co-ordination.300 A possibly related ‘loose’ (unstructured) and
temporary human behaviour is the phenomenon of collective empathy,
which emerged for example in the form of the collective mourning of mil-
lions of people after Britain’s Princess Diana died in 1997.301 This was,
among other things, a reaction to the perceived rational and hierarchical,
unsympathetic attitude of the British Queen.
Autonomism, the fifth way of life Thompson et al. distinguish, seems an
extreme form of the individualism of market governance. In this way of
life “the individual withdraws from coercive or manipulative social in-
volvement altogether”302. As autonomists or ‘hermits’ accept no social re-
sponsibility it is difficult to see how this way of life can be considered a
governance style: governance requires dealing with public issues.
Clientelism and nepotism are other forms of relational behaviour that
can be found in governance arrangements, especially in developing coun-
tries in which the ruling party represents a clan or a family. Clientelism
and nepotism require a hierarchical governance basis and a strong collec-
tivist (networking) culture. They are not ‘complete’ governance styles like
hierarchy, network or market governance, but rather mixtures or hybrids of
hierarchical and network governance.
In an earlier publication, Wildavsky argues that the often used left-right
dichotomy in political life is full of contradictions and does not hold
against cultural theory.303 Personal preferences – of politicians, civil ser-
vants and citizens - are traceable to elements of the trichotomy of hierarch-
ism, egalitarism and individualism. Therefore, if we consider governance
styles to be grounded in cultures, and even represent the logic of (mixtures
of) cultures (Figure 4), this enforces the argument that it is useful to distin-
guish three ‘ideal-typical’ governance styles.
Also Bevir and Rhodes emphasise the cultural dimension of hierar-
chies, markets and networks, where they propose to treat them alike, “as
meaningful practices created and then constantly recreated through contin-
299 Wildavsky (1987): Choosing preferences by constructing institutions: a cul-

tural theory of preference formation.
300 Tenbensel (2005: 278): Multiple modes of governance.
301 Walter (ed.) (1999): The mourning for Diana.
303 Wildavsky (1987): Choosing preferences by constructing institutions: a cul-
tural theory of preference formation.
gent actions informed by particular webs of belief”304. Ways of life or

‘world views’ “serve as cognitive instruments of actors in order to select
and to interpret events, facts, symbols, etc. [ ] Different world views lead
to different problem definitions and to different interests of actors.”305
Fig. 4. Relations between the five 'ways of life' (cultural theory) and the three
ideal types of governance.
Dixon and Dogan emphasise the incompatibility of these views: Hierar-

chical, network and market styles of governance
“… derive their governance certainties from propositions drawn from spe-
cific methodological families, which reflect particular configurations of
epistemological and ontological perspectives”. They “have incompatible
contentions about what is knowable in the social world and what does or
can exist – the nature of being – in the social world. Thus, they have in-
compatible contentions about the forms of reasoning that should be the ba-
sis for thought and action”. 306
304 Bevir and Rhodes (2001: 25): A decentered theory of governance: Rational
choice, institutionalism, and interpretation.
305 Jachtenfuchs (1994): Theoretical reflections on the efficiency and democracy
of European governance structures.
306 Dixon and Dogan (2002: 191): Hierarchies, networks and markets: Responses
to societal governance failure.
It is now possible to conclude that there is a strong relationship between

the general types of cultures in society (‘ways of life’) and governance
styles and their mixtures. Cultures can be understood on at least four dif-
ferent levels. The first, the level of human society, has already been dis-
cussed. Two other levels will be addressed in the following section: Cul-
tures on the level of nations, and cultures on the level of organisations. The
fourth dimension, cultures on the level of individuals, will be addressed in
Section 7.2.6.
2.4.2 Governance styles and national cultures
In political science and public administration it is not unusual to link na-

tional socio-politico-administrative cultures to certain governance styles.307
For example, in Europe the United Kingdom has a ‘public interest’ model
of administration with a modest role of the state within society. When
market governance arrived on the scene in the 1980s it was only logical
that the UK was the first European state to embrace this approach. Ger-
many has a typical Rechtsstaat administrative culture in which the state is
central in society and the ‘natural’ governance style is hierarchy. The
Netherlands belongs to a third category and has moved away from its his-
torical link with the Rechtsstaat philosophy (introduced by the French dur-
ing the Napoleontic occupation in the 18th century) towards a (or maybe:
back to the older, 17th century) consensual approach. The Netherlands
therefore have a strong affinity with network governance.
Similar forms of governance approaches have developed in countries
with a similar socio-politico-administrative culture. A good example is the
fast and successful transfer of government reform ideas under the new
Public Management banner in all Anglo-Saxon countries. Also Hesse et
al., who compared public sector reform in twelve countries and the Euro-
pean Commission (2003), illustrate the impact of national cultures on gov-
ernance approaches. As Cepiku (2005: 32), in her comparison of the gov-
ernance of territorial (spatial) issues in seven countries, concludes:
“Territorial governance (…) depends on the political and legal framework
of a country, its geographical conditions and the specific policy field to be
addressed.”
307 E.g. Pollit and Bouckaert (2000: 52-54): Public management reform. A com-
parative analysis.
Developing countries show very different cultures. Morgan identified

three types:308
- The integral or hegemonic state, which has a monopoly on resource
allocation;
- The patrimonial state, with a monopoly on power and resources
through a system of patronage: clientelism based on clan, ethnic or
religious criteria;
- The custodial state, in which the political environment is turbulent
and the administration takes on a custodian character.
Also sociologists have distinguished national cultures. According to

Hofstede national cultures distinguish similar people, institutions and or-
ganisations in different countries. 309 Although Hofstede’s work is based on
extensive empirical research, the concept of ‘national cultures’ is contested
by Thompson et al.: “Differences within each country are at least as strik-
ing as the variation among countries”. 310 McSweeny considers Hofstede’s
theory ‘profoundly problematic’, because of its not underpinned central
claim of causality: that national cultures influence how we think. 311 How-
ever, Hofstede’s five dimensions of national cultures (power distance, the
degree of individualism, gender roles, uncertainty avoidance, and long
term orientation) do seem to help understand general differences in na-
tional cultures that public administration scholars have spotted. As stated
earlier, in public administration it is quite common to describe politico-
administrative cultures of nations or groups of nations. The Dutch (low
power distance: consensus-oriented) are different from the Germans (rela-
tively high power distance: hierarchical), and this is reflected in the higher
affinity of the Dutch with network governance. A culturally mixed country
like Trinidad and Tobago has a mixed culture on the individualism-
collectivism dimension, but at the same time a low power distance and low
uncertainty avoidance national culture.312 Mcsweeny’s problem with the
308 Morgan (1996: 230-232): Analysing fields of change: Civil service systems in
developing countries.
309 Hofstede (2001): Culture’s consequences: Comparing values, Behaviours, In-
stitutions and Organisations across Nations.
311 McSweeny (2002): Hofstede’s model of national cultural differences and their
consequences: A triumph of faith – a failure of analysis.
312 Tidwell (2001): Trinidad and Tobago: Customs and Issues Affecting Interna-
tional Business.
lack of causality between national cultures and how nationals think is void,
because in this research not causality is claimed, but a theoretically plausi-
ble and if possible empirically underpinned construct of correlations.
“Reality is socially constructed, but not all constructions have equal claim
to our credibility and certainly some constructions prove more durable than
others. One important test is correspondence with such empirical evidence
as may be available”.313
Relating Hofstede’s five dimensions of national cultures with the three
governance styles hierarchy, network and market, might produce an ana-
lytical framework for understanding general national differences in prefer-
ence of governance styles (Table 3).
Table 3. Expected relations between governance styles and five dimensions of in-
tercultural differences (own composition, based on Hofstede and Hofstede
(2005).314
Hierarchical Network Market

governance governance governance
Accepted power HIGH (Power is LOW (Equal
distance good, privileges, rights, pluralism,
dependence) interdependence)
Individualist/ COLLECTIVIST INDIVIDUALIST
collectivist (Equality, (Freedom,
relationship- autonomy, task-
oriented, oriented,
interdependence) independent)
Uncertainty HIGH (Many LOW (Trust, high
avoidance laws, low citizen citizen
participation) participation)
‘Masculine/ ‘FEMININE’ ‘MASCULINE’
Feminine’ (Consensus, (Performance,
equality, competition,
relationships) equity, recognition)
Long time/short time LONG (Learning, SHORT
orientation life-long personal (Achievement,
networks) meritocracy,
freedom)
313 Pollit and Bouckaert (2000: 23): Public management reform. A comparative
analysis.
314 Hofstede and Hofstede (2005): Cultures and organizations. Software of the
mind.
What Table 3 shows is no surprise. Nations with a relatively low power

distance, low insecurity reduction and a ‘feminine’ culture, like the Neth-
erlands and the Scandinavian countries, also have a tradition of network
governance. Traditionally hierarchical nations like Germany and France
have, compared to these Nordic countries, a higher power distance and a
higher uncertainty avoidance Index. The US, Australia and Great Britain
are ranking 1-3 on Hofstede’s list of most individualistic countries and
they also score highly on ‘masculinity’. This correlates with the fact that
market governance originated in Anglo-Saxon nations.
To conclude: it seems plausible that governance mixtures in different
politico-administrative national systems differ along the lines of ‘average’
national cultures. If we accept the premise that hierarchical, network and
market styles of governing are each suitable for solving different types of
problems, than the problem-solving capacity of nations may also vary. It
may be a bit speculative, but the capability of governments to solve crises
such as a flood disaster may, in hierarchical cultures with a ‘well-oiled
government machine’, be better than in nations with a dominant market
governance approach, implying a small and fragmented administration.
‘Feminine’, consensus-oriented nations with a network style preference
have shown to be incapable at successfully dealing with the influx of cul-
turally very different immigrants. An example is the immigration in the
Netherlands of people from Arabic countries with a hierarchical culture.
They are used to a relatively high power distance, a high level of ‘mascu-
linity’ and a high level of uncertainty avoidance. They are not at all con-
sensus-oriented. Attempts to make government more efficient by the use of
market mechanisms like outsourcing and the creation of independent agen-
cies has met strong (and successful) resistance in nations with a hierarchi-
cal culture like Germany.315
The ‘proof of the pudding’ is the role of nationality in a multi-national
administration such as the European Commission. Hedetoft sees the
broader polity of the European Union as “both an international organisa-
tion and a kind of European proto-state”; it likes to model itself on and be
judged by nation-state criteria of governance”316. A study of Hooghe shows
empirically that top Commission officials believe that Europe’s diversity
should be explicitly recognised in the Commission. She concludes that na-
tional diversity “is and will be a fundamental feature of the European pol-
315 E.g. Jann et al. (2004: 15): Status-Report Verwaltungsreform. Eine Zwischen-
bilanz nach zehn Jahren.
316 Hedetoft (2003: 43): Cultures of states and informal governance in the EU: an
exploratory study of elites, power and identity.
ity for a long time to come”.317 Although this might lead to the expectation
that the Commission applies a variety of governance styles, others have
emphasised that the European Commission is a primarily hierarchical or-
ganisation, “half way between a French ministry and the German Econom-
ics Ministry”.318
From the above it can be concluded that there seems to be something
like a ‘national governance footprint’: the more or less stereotypical ‘aver-
age’ composition of the mixture of hierarchy, network and market govern-
ance. Therefore, it is questionable if governance style(s) mixtures are
transferable from one nation to another (see Section 6.3).
2.4.3 Governance styles and organisational cultures
This research focuses on the exploration of governance style mixtures and

metagovernance as a function of public administration organisations, not
of society as a whole, or a nation. Therefore, organisational cultures and
individual cultures seem to be the most important here. How do cultural
theories help to reveal the capabilities of organisations and individuals to
apply forms of governance? One way of looking at governance styles is
that they are theoretical constructs that are expressions of politico-
administrative organisational cultures. From this perspective,
- Hierarchical governance and market governance are related to a ra-
tional, positivist attitude. The rational public administrator uses a
causal logic and focuses on selecting the (objectively) best means to
achieve agreed-upon results.319 Central in Simon’s classic Administra-
tive Behaviour (1947) is the concept of purposiveness. This “involves
a notion of a hierarchy of decisions – each step downward in the hier-
archy consisting in an implementation of the goals set forth in the
step immediately above”.320
- Market governance builds on rational choice theory, which views ac-
tions of citizens, politicians, and public servants as analogous to the
actions of self-interested producers and consumers.321
317 Hooghe (1999: 417): Consociationalists or Weberians? Top Commission offi-

cials on nationality.
318 Dimitriakopoulos and Page (2003: 317): Paradoxes in EU Administration.
319 Frederickson and Smith (2003: 162): The administration theory primer.
320 Simon (1997: 4): Administrative Behaviour (Fourth Edition).
321 Frederickson and Smith (2003: 185): The administration theory primer.
- Network governance emphasises the boundedness of rationality in

public administration, highlights ambiguity rather than rationality,
and is related to a more socio-constructivist approach and social con-
figuration theory.
Organisational cultures differ not on the level of values, but on the level
of practices322. Fairtlough 323 uses the three active ways of life of Thompson
et al. as metaphors for ‘three ways to get things done’ in organisations: hi-
erarchy (same as Thompson), heterarchy (Thompson’s egalitarian) and re-
sponsible autonomy (Thompson’s individualism). Hierarchy, market and
network governance are forms of societal co-ordination that together, in
mixed forms, define organisational cultures. This has been confirmed for
businesses organisations, public-sector organisations and non-governmen-
tal organisations. 324
A governance style is rooted in, and part of a culture, because it is
based upon a certain comprehensive set of values. For example, hierarchy
is based upon the belief that there should be a ‘subordinate’, market gov-
ernance considers others as ‘customers’ or ‘clients’ and network govern-
ance believes others to be ‘partners’ and ‘co-creators’. Governance styles
also include behaviour rules (such as regulations and control instruments
in hierarchy), and a preference for certain types of coordination mecha-
nisms, like trust in networks and competition in markets.
2.5 Metagovernance as the ‘governance of governance’
2.5.1 Different views on metagovernance
Each governance style has its own distinctive forms of failure.325 Combina-
tions of the three ideal-typical governance styles may lead to conflicts,
competition and to unsatisfactory outcomes. The question is, if it is possi-
ble to design strategies that harness the benefits of these different ap-
proaches, whilst minimising the negative consequences. How feasible is
322 Hofstede (2001): Culture’s consequences: Comparing values, Behaviours, In-

stitutions and Organisations across Nations.
323 Fairtlough (2005): Triarchy Theory (www.triarchypress.com).
324 E.g. Thompson et al. (2001): Markets, hierarchies and networks: The coordi-
nation of social life. Davis and Rhodes (2005): From hierarchy to contracts
and back again: Reforming the Australian public service.
325 Jessop (2004: 228): The political economy of scale and European governance.
2.5 Metagovernance as the ‘governance of governance’ 67
Rhodes’ challenge that “The future will not lie with markets, or hierarchies
or networks but with all three and the trick will not be to manage contracts
or steer networks but to mix the three systems effectively when they con-
flict with and undermine one another.”? 326 If designing and managing gov-
ernance style mixtures was possible to a certain extent, then this would be
of great importance. It would make the ‘toolbox’ of public managers much
richer compared to when they are ‘stuck’ with only one of the governance
styles:
“If you only have a hammer, you tend to see every problem as a
nail.”327
Metagovernance as governance of hierarchies, networks and markets
As early as 1991, Bradach and Eccles distinguished ‘plural forms’ of coor-

dination, in which distinct control mechanisms (price, authority and trust)
may operate simultaneously for the same function (or project) in the same
organisation.328 They point at the new possibilities managers have if they
do not have to rely on one style alone, but can use other styles too. In
2002, Eberlein and Kerwer329 concluded that the major question in the field
of European governance is “how new modes of governance can be recon-
ciled with the need for binding rules”. This brings us to the term metagov-
ernance, which we will define as the ‘governance of governance’, in line
with the common use of the prefix meta.330 Because governance is the re-
sult of dynamic combinations of the three ideal types, metagovernance is
326 Davis and Rhodes (2005: 25, From hierarchy to contracts and back again: Re-
forming the Australian public service). The same formulation appears in
Fleming and Rhodes (2005: 203): Bureaucracy, contracts and networks: The
unholy trinity and the police.
327 Quote attributed to Abraham Maslov (1907-1970). Citation retrieved from
http://www.brainyquote.com on 28 August 2007.
328 Bradach and Eccles (1991: 288): Price, authority and trust.
302 Eberlein and Kerwer (2004: 136): New Governance in the European Union: A
theoretical perspective.
330 Meta (Greek: μετά = "after", "beyond"), is a common English prefix, used to
indicate a concept which is an abstraction from another concept, used to ana-
lyse the latter. "Metaphysics" refers to things beyond physics, and "meta lan-
guage" refers to a type of language or system which describes language. In
epistemology, the prefix meta- is used to mean about (its own category).
Metadata are data about data (who has produced it, when, what format the
data are in and so on). (Source: www.wikipedia.org).
the governing of mixtures of hierarchical, network and market forms of

social coordination.
Metagovernance is a concept positioned ‘above’ the three main govern-
ance styles. It takes a multi-perspective, ‘helicopter view’ approach. It can
be considered as a negotiation process between competitive governance,
authoritative governance and cooperative governance.331 Jessop coined the
term metagovernance in 1997 as “coordinating different forms of govern-
ance and ensuring a minimal coherence among them”.332 Later, he de-
scribed metagovernance as “the organisation of the conditions for govern-
ance”, which involves “the judicious mixing of market, hierarchy, and
networks to achieve the best possible outcomes from the viewpoint of
those engaged in metagovernance.”333 Metagovernance is about rebalanc-
ing market, hierarchy and networks.334 Like Jessop, Rhodes also argues
that governments can choose between the three ‘governing structures’: hi-
erarchies, markets and networks.335 In this research the following more
precise definition will be used: 336
Metagovernance is a means by which to produce some degree of

coordinated governance, by designing and managing sound combi-
nations of hierarchical, market and network governance, to achieve
the best possible outcomes from the viewpoint of those responsible
for the performance of public-sector organisations: public managers
as ‘metagovernors’.
This definition is on the one hand broad: all possible governance style
combinations are included. On the other hand, it is deliberately narrow: In
this research one type of metagovernors is central: line and project manag-
ers of public-sector organisations. Managers of private companies or
NGOs may, as much scholarly literature suggests, experience a rather
331 Arentsen (2001: 512, Negotiated environmental governance in the Nether-

lands: Logic and illustration) therefore calls it ‘negotiated governance’.
332 Jessop (1997: 7): Capitalism and its future: remarks on regulation, govern-
ment and governance.
334 Jessop (2004: 228): The political economy of scale and European governance.
335 Rhodes (1997: 47): Understanding governance.
336 This definition combines definitions by Kooiman (2003: Governing as
governance), Jessop (2003: Governance and metagovernance) and Sørensen
(2004: Democratic governance and the role of public administrators).
similar challenge, although there are limitations. Wolf argues that there are
several functions of metagovernance that cannot be fulfilled by the private
sector, and have to be provided by the public sector:337
- Providing and guaranteeing the constitutional legal framework for
private self-regulation;
- Providing a ‘shadow of hierarchy’ to keep self-regulation geared to-
wards the general good;
- Authorising participatory claims of private actors;
- taking part in providing normative environment and therefore legiti-
macy to goals of private actors;
- Supporting the monitoring of self-regulation;
- Avoiding negative externalities by linking the different sectoral self-
regulation efforts with each other.
To what extent is this definition of metagovernance normative? As an
analytical concept it is just as non-normative as they three ideal types of
governance are. However, as a practical concept (hierarchy, network or
market governance as public management), the ideal types become norma-
tive: each of the styles is used as a ‘best’ way of societal coordination. A
hierarchical manager may judge that expensive cars are always better than
cheap cars, because they are more reliable. A market manager will make a
cost-benefit analysis before he decides which car is better. A network
manager may judge that any car is good, as long as its use is not limited to
superiors or those who are rich. In contrast, metagovernance deliberately
takes a situational view: what is ‘best’ is determined by a range of envi-
ronmental factors. It does have a ‘light’ normative dimension, related to
the underlying concept of governance, which inherently has several nor-
mative assumptions, such as that (1) the intention is to solve collective
problems, (2) not for individual profit but for the common good, and (3)
that it supposes more or less well defined social groups with rights and du-
ties338.
The emerging literature on metagovernance since 2000339 presents two
other schools of thought about metagovernance.
337 Wolf (2003: 1): Some normative considerations on the potential ans limits of
voluntary self-regulation.
338 Risse (2007: 6-8): Regieren in Räumen begrenzter Staatlichkeit. Zur "Reise-
fähigkeit" des Governance-Konzeptes.
339 See Torfing et al. (2003: Nine competing definitions of governance, govern-
ance and networks and meta-governance) for a short overview.
Metagovernance framed as network management
The first and most prominent school of thought considers metagovernance

to be a way of “enhancing coordinated governance” 340. They restrict the
use of the term ‘governance’ to network governance. Jessop341 followed
this line of thought when he called the Open Method of Coordination of
the European Union a form of metagovernance. Metagovernance is a form
of ‘network management’.342 Metagovernance is an “indirect form of top-
down governance that is exercised by influencing processes of self-
governance through various modes of coordination such as framing, facili-
tation and negotiation.” In this approach, metagovernance aims at enhanc-
ing self-governance (with which mainly network governance is meant) by
using methods such as framing, facilitation and negotiation. 343 Hierarchi-
cal, and to a lesser extent market governance are not considered to be dis-
tinct and useful approaches to societal coordination: hierarchy is only used
to increase the success of network governance as the superior style. ‘Self-
governing networks’ are the focus: they are implicitly considered as the
best way to solve (all) societal problems. From that perspective, it is logi-
cal that there can be “too much or too little” metagovernance: Too little
and networks risk becoming undemocratic, too much and “the self-
regulatory capacity can be undermined and the network actors loose inter-
est in and responsibility for the network activities”.344 ‘Democratic me-
tagovernance’ therefore is about governments taking responsibility for
demanding participation from a broader set of actors and guaranteeing that
all major interests are reflected in networks, for example.345
This vision is also shared by scholars of the deliberative governance
school, of whom some have a state declinist attitude. Metagovernance or
340 Sørensen (2006: 100): Metagovernance: The changing roles of politicians in

processes of democratic governance. Sørensen and Torfing (2007: 169: Theo-
retical approaches to metagovernance.
341 Jessop (2004: 49): Multi-level governance and multi-level metagovernance.
342 Kickert et al.(eds.)(1997): Managing complex networks: strategies for the
public sector; Klijn and Edelenbos (2007: 199): Meta-governance as network
management.
343 Hovik and Vabo (2005: 262): Norwegian local councils as democratic me-
tagovernors?
344 Westh Nielsen (2007): Metagovernance in the Global Compact – Regulation
of a global governance network.
345 Hovik (2005: 15): Local network governance for sustainable development in
Norway.
“smart steering” is then a means of combining instruments of network and

market governance.346 It is a way “to sort of try to get to an understanding
(…) of the ability of people to find open spaces (…) in which new forms
of action are possible”. This means: “Forget about government; start with
society, and ask yourself how as society do we steer things?”347 Laws and
van der Heijden noticed that businesses and environmental groups, without
the direct involvement of governmental actors, tackle societal problems
such as environmental problems for example. Section 2.2.5 argued that
there is no empirical evidence that the state does not play a (major) role
anymore, and that hierarchy remains the preferred style of coordination of
public-sector organisations. Even when hierarchical governance is only a
vague ‘shadow’, this may be a trigger for businesses and NGOs to cooper-
ate without direct government involvement. Therefore, it is important “to
bring government back in when analysing governance”348, as Bell and Park
argue.
Similar to the idea that metagovernance is ‘enhancing coordinated gov-
ernance’ is the approach of metagovernance as a mechanism that transfers
power from politicians to public managers.349 This approach is however not
anti-statist, but anti-political. Jayasuriya has explored this for trans-
national organisations like the World Bank. 350 He concludes that the World
Bank – a non-political body – uses ‘metagovernance’ to ‘colonise’ civil
society and “depoliticises social and economic life by distancing the allo-
cation of social goods from the centres of political decision making”.
Metagovernance framed as supervising network and market

governance
This brings us to another school of thought, where metagovernance is a

form of regaining state control (hierarchy) over new forms of governance.
346 Van der Heijden (ed.) (2005: 51): Recombinatie van overheid en samenleving.
Denken over innovatieve beleidsvorming.
347 Citations of respectively David Laws (MIT, USA) and Jurgen van der Heijden
(University of Amsterdam), retrieved on 2-10-2006 from http://www.lafol
lette.wisc.edu/research/environmentalpolicy/mswgmetagovernance.html.
348 Bell and Park (2006: 63): The problematic metagovernance of networks: Wa-
ter reform in New South Wales.
349 Sørensen (2006): Metagovernance: The changing roles of politicians in proc-
esses of democratic governance.
350 Jayasuriya (2003: 6): Workfare for the global poor’: Anti politics and the New
Governance.
It is a new form of hierarchical governance that coordinates network and

market styles of governance. It secures governmental influence, command
and control within network and market style governance regimes.351 Ac-
cording to Bell and Park, “metagovernance maintains a continuing role for
hierarchical government within any governance regime”.352 They argue
that metagovernance should be a response to the “hollowing out of the
state thesis” of governance literature that adopts an anti-statist model as
mentioned above.353 Kelly noticed that the UK central government uses
metagovernance to “retain its authority and exercise central steering
mechanisms in an era of apparent diminishing state power”.354 Damgaard
investigated similar cases in Denmark.355
Let us return to the form of metagovernance that will be investigated in
this research: the organisation and coordination of hierarchies, networks
and markets. Problematic relations between metagovernance and democ-
ratic institutions may also arise here, as Skelcher et al.356 argue. They dis-
tinguish four types of problematic relations, depending on the national
socio-political-administrative context. Mixtures of governance styles can
be (1) incompatible, (2) complementary, (3) part of a transitional process
of governance, and/or (4) used instrumentally. The impact of different na-
tional contexts will be discussed in section 6.3, as well as another crucial
issue: the question if public managers have the requisite metagovernance
skills, capacities357 and capabilities358, which is at this moment an empiri-
cally open question (See section 6.4).
351 Whitehead (2003: 8): ‘In the shadow of hierarchy’: metagovernance, policy
reform and urban generation in the West Midlands.
353 Bell and Park (2006: 64).
354 Kelly (2006: 619): Central regulation of English local authorities: An example
of metagovernance?
355 Damgaard (2006): Lessons on meta-governance from a longitudinal policy
network study.
356 Skelcher et al. (2006): Governance networks, democratic anchorage, and the
impact of national political context.
358 Capabilities are central: capacities and competences derive from capabilities.
(Laske, 2006): Measuring hidden dimensions.
Metagovernance and ‘sound governance’
In ‘Sound Governance’359, Farazmand designs a ‘best of’ hierarchical, net-

work and market governance. The fact that he defines ‘sound governance’
in an abstract way makes it possible to apply the concept in different situa-
tions, in different ways. Farazmand’s concept aims at overcoming the typi-
cal failures of market governance and network forms of governance. In
addition, he advocates the re-introduction of key, ‘sound’ elements of hier-
archical governance, such as the importance of a constitution and of robust
institutions. He describes the characteristics of a ‘sound’ combination of
governance styles. Metagovernance as it is used in my research is how
public managers may achieve such useful, smart governance mixtures.
As shown in Figure 5, all forms of governance and metagovernance
that are mentioned in this chapter are related to each other. It is a ‘family
tree’ of contemporary thinking about governance. This overview illustrates
the ‘conceptual crowd’ on the middle and right side of the figure.
2.5.2 Internal and external metagovernance
Where should metagovernance be located? If it means dealing well with

potential conflicts and synergies between hierarchy, network and market
governance, then metagovernance is something that happens in the relation
between government and societal actors, as well as inside government. In
literature on metagovernance the term is usually applied to the governing
of governance relations between public administration and society, but
Sørensen differentiates between the governance challenge inside the politi-
cal system and between public and private actors.360 She argues that me-
tagovernance should primarily be a task for politicians, and observes that
politicians are hesitant to take up this new task, leaving it to public admin-
istrators – “at severe costs for democracy”.361 The question is which form
of democracy is meant here. Surely, market governance with its autono-
mous agencies has produced questions of democratic control.
359 Farazmand (ed.) (2004): Sound governance. Policy and administrative innova-
tions.
360 Sørensen (2004: 107-108): Democratic governance and the role of public ad-
ministrators.
361 Sørensen (2006: 112): Metagovernance: The changing roles of politicians in
processes of democratic governance.
Fig. 5. Governance and metagovernance: a ‘conceptual crowd’ addressing the

‘new modes of governance’ (own composition)
When public administrators are supervised by elected politicians, they

can do anything that is considered professional, as long as it is accountable
in the eyes of the politicians. Most policy preparation takes place inside
administration, and politicians are usually involved in the beginning and
the end.
Management and organisation inside administration cannot be isolated
from the societal and political context.362 Therefore, good ‘internal’ me-
tagovernance might be a prerequisite for ‘external’ metagovernance. Inside
administration, governance style conflicts arising between different policy
units, between different ministries, between different administrative levels,
and between internally oriented units (like financial and legal departments)
and externally oriented policy units, should be dealt with.363
Internal metagovernance, when it succeeds in producing more tailor-
made, situational governance processes, may enhance the quality of public
administration performance and of democratic institutions and processes.
When metagovernance is defined as state reconfiguration, then it may have
the opposite outcome.
The boundaries between ‘internal’ and ‘external metagovernance are
vague. This research focuses on how public managers deal with govern-
ance conflicts and synergies, which consequently touches both internal and
external metagovernance.
2.5.3 Introducing the public manager as metagovernor
Metagovernance requires a management perspective. ‘Public management’

is broader than management in the private sector. The latter can be defined
as “the search for the best use of resources in pursuit of objectives subject
to change”.364 Private sector management “is about getting things done as
quickly, cheaply and effectively as possible – and usually about getting
things done through other people (‘staff’, ‘the work force’, ‘personnel’,
‘human resources’)”.365 Management in public administration organisations
is not a neutral, technical process, but “intimately and indissolubly en-
362 Kickert (1997: 738): Public Governance in the Netherlands: An Alternative to

Anglo-American 'Managerialism'.
363 E.g. Meuleman (2003): The Pegasus Principle.
364 Keeling (1972): Management in government. Cited by Pollit and Bouckaert
(2000: 19).
365 Pollit and Bouckaert (2000: 9): Public Management Reform. A Comparative
Analysis.
meshed with politics, law and the wider civil society. It is suffused with
value-laden choices and influenced by broader ideologies”.366 In this re-
search we will use Noordegraaf’s broad definition of public management
as “the influencing of common activities in the public domain or by pub-
lic-sector organisations and public managers”367.
Noordegraaf distinguishes four approaches to public management:368
- Business management; result-oriented, coordinated and efficient
managing public-sector organisations as if they are normal enter-
prises;
- Organisation management: leading professional public-sector organi-
sations who operate in complex environments;
- Policy management: organising and influencing policy making proc-
esses in order to address societal problems;
- Political management: influencing political agenda-setting and deci-
sion making.
The first approach will ‘flourish’ best in a market governance environ-
ment. The second and third types have more affinity with network govern-
ance, and the third can be applied following the logic of each of the three
governance styles. From a metagovernance perspective, all four ap-
proaches to public management are useful.
Scholars like Bevir and Rhodes have doubted the ‘manageability’ of
metagovernance.369 The issue is too complex and the possible number of
combinations of conflicting or synergetic elements of hierarchy, network
and market thinking is overwhelming. Moreover “the ability of the state to
manage the mix of hierarchies, markets and networks that have flourished
since the 1980s” risks to undermine the bottom-up orientation of societal
networks.370 Some have argued that applying metagovernance can only be
done by being aware of the complexity, not by applying crucial success
factors.371 Apart from this practical argument, there is also a theoretical
one. Several authors point out that a governance style ‘emerges’ from a
366 Pollit and Bouckaert (2000: 16): Public Management Reform. A Comparative
Analysis.
367 Noordegraaf (2004: 94): Management in het publieke domein. (Translated
from Dutch by the author).
368 Noordegraaf (2004: 95-96): Management in het publieke domein.
369 Bevir and Rhodes (2001: 32-34): A decentred theory of governance.
370 Rhodes (2007: 1257): Understanding Governance – Ten years on.
371 Koffijberg (2005: 365): Getijden van beleid: omslagpunten in de
volkshuisvesting.
certain organisational and environmental context. Styles of governance

“are the outcome of social processes but also provide the medium through
which actors interpret and act to shape their reality”.372 Bevir and Rhodes
argue that “patterns of governance arise as the contingent products of di-
verse actions and political struggles”.373 This view suggests that govern-
ance styles and their combinations are not designed or chosen, but are a
mere result of some ‘invisible governance hand’. This is a risky assump-
tion, because such an ‘invisible hand’ does not work in the free market.
Why then is it claimed to work in government? In their article on shifts in
governance styles during the last decades Van Kersbergen and Van
Waarden point at external factors as causes of these shifts. They suggest
that an important cause may be the fact that governments have to deal with
different problems than they had to one or two decades ago.374 Internal fac-
tors such as the attempts of metagovernance are not mentioned. A compa-
rable view with regard to the feasibility of metagovernance is found with
Bevir and Rhodes. First they observe three approaches to how the state can
manage governance: an instrumental, an interactive and an institutional
approach. The instrumental approach is top-down: the state is central and
can impose imperatives to reach its objectives. The interactive approach
focuses on managing by means of negotiation and diplomacy. The institu-
tional approach concentrates on the use of laws, rules and norms. They
then reject these approaches as too positivist: there is not a set of tech-
niques or strategies for managing governance. The only way to ‘manage’
governance is “to learn by telling and listening to stories”. 375 376
Earlier, Rhodes suggested that governance style (conflict) management
is feasible as well as desirable. In 1997, he took the view that the choice
between hierarchy, network and market as ‘governing structures’, is a mat-
ter “of practicality, that is: under what conditions does each governing
structure work effectively”.377 Thus, he argued, managing the three styles is
about coordinating difficulties. Because markets, hierarchies and networks
are not found in their pure form, it is the mixture of governing structures

373 Bevir and Rhodes (2003: 59): Searching for civil society: Changing patterns
of governance in Britain.
374 Van Kersbergen and van Waarden (2004: 166): ‘Governance’ as a bridge
between disciplines.
375 Bevir and Rhodes (2001: 32-34): A decentered theory of governance.
376 Rhodes (2007: 1257): Understanding Governance – Ten years on.
377 Rhodes (1997: 47): Understanding governance.
that distinguishes services one from the other. These governing structures
may “mix like oil and water.”378
A similar position was taken in an advisory report to the Dutch Envi-
ronment Ministry: The choice between hierarchical, network and market
governance is not the selection of the one right style, but about choosing
the situationally best role for the government, taking into account the char-
acteristics of all three governance styles.379 Lowndes and Skelcher, al-
though they define governance as an ‘emerging outcome’, consider man-
aging the interaction of different modes of governance a challenge for
governance of partnerships between governmental and other parties.380
Lowndes and Skelcher as well as Rhodes have proposed making choices
between governance styles, not making combinations of elements of hier-
archy, network and market governance. However, the latter seems more
realistic: in their pure form the three ideal-types are seldom reported to ex-
ist. Furthermore, others argue that using the characteristics of all three
styles in combination, may lead to the best results, not only for solving
policy problems, but also for creating a successful organisation design.
Such an organisation would be able to use complementary modes of coor-
dination: authority, competition and mutual adjustment.381 Mixing hierar-
chies, networks and markets is analogous to a chemical reaction: “elements
may react in different ways under different circumstances”382. Outcomes of
such governance reactions “will be specific to particular areas and contin-
gent upon prevailing policies”.383 Linker designed a model of situational
steering by public-sector organisations that builds on typical characteristics
of the three governance styles: control (hierarchy), trust (network) and
‘pressure’ (for which he advocates market instruments like performance
contracts and benchmarking).384
378 Rhodes (1997): From marketisation to diplomacy: it’s the mix that matters.
379 VROM-Raad (1998: 52): De sturing van een duurzame samenleving.
381 Assens and Baroncelli (2004): Marché – Réseau – Hiérarchie. Une réflexion
sur les idéaux types organisationels.
382 Exworhty et al. (1999: 20): The NHS: quasi-market, quasi-hierarchy and
quasi-network?
383 Exworthy et al. (1999: 20): The NHS: Quasi-market, quasi-hierarchy and
quasi-network?
384 Linker (2006): Sturing in de rijksdienst: nieuwe en bestaande inzichten vere-
nigd in hét besturingsmodel.
In ‘t Veld raises doubt about the feasibility of creating a desired gov-

ernance mix.385 He illustrates this with the example of the Dutch university
system. In the 1970s, the Dutch universities switched from a hierarchical
towards a participatory style of governance, after students had successfully
lobbied for such a change. However, this change did not result in improved
services to students. The service improved in the 1980s, without a change
of governance arrangement. In the 1990s, the old hierarchical arrangement
was restored without a noticeable impact on the level of service. In an ear-
lier publication, a review of the Dutch Environment Ministry, In ‘t Veld
observed different governance styles inside the same ministry, in different
directorates and also between the top and the lower echelons.386 According
to him, this is inevitable in a professional public administration organisa-
tion. However, he argues, the variety has to be managed, the organisation
has to learn from the variety, and there should be a ‘general’ governance
style on the meta-level in the organisation.
A recent research evaluating the EU chemicals directive (REACH)
shows that a specific governance style mixture was deliberately developed,
consisting of regulated self-regulation and regulated standardisation.387 Pe-
ters argues that “metagovernance is not a given process that will occur
simply through political will, and must be considered as a management
problem”. It is a form of ‘meta-management’, which focuses more on “es-
tablishing parameters for action rather than necessarily determining that
action”.388
The above literature overview does not present a clear answer to what
extent and how metagovernance is feasible. Scholars seem to be torn be-
tween a positive and a negative answer. Rhodes, for example, in 1997 (see
above) argued that it is just a matter of practicality, and in 2001 rejected
the notion “that there is a set of techniques or strategies for managing gov-
ernance”; instead, “no matter what rigour or expertise we bring to bear, all
we can do is tell a story and judge what the future might bring”.389 In order
for this question to be answered, empirical research will be needed. How-
ever, the limited literature does lead to the assumption that, to a certain ex-
385 In ‘t Veld (Personal communication, February 2006).

386 In ‘t Veld (1996: 42): Rapportage onderzoek Besturingsconcepten VROM.
387 Hey et al. (2007): REACH als Beispiel für hybride Formen von Steuerung
und Governance.
388 Peters (2006: 14): The meta-governance of policy networks: Steering at a dis-
tance, but still steering.
389 Bevir and Rhodes (2001): A decentered theory of governance: Rational
choice, institutionalism, and interpretation.
tent, it should be possible to define conditions that make the design and
management of a desired combination of governance styles within a public
administration organisation possible: the should be something like a ‘me-
tagovernor’s rationale’. Jessop argues that a kind of self-reflexive irony is
necessary, “not only for individual governance mechanisms but also for
the commitment to metagovernance itself”: Metagovernors must realize
that their attempts will probably (partly) fail, but should proceed as if suc-
cess were possible.390 Likewise, Svensson and Trommel warn that Rhodes’
argument that it is “the mix that matters”, should not be confused with ar-
guing that “anything goes”.391
There is not yet much research on the limitations of applying metagov-
ernance. Considine and Lewis showed that there is a limit to the degree of
variation in governance styles that can be used inside one organisation:392
some style elements are inherently incompatible, as we have already seen
in Section 2.3.3.
2.5.4 Metagovernance as managing the governance trilemma
Metagovernance is a ‘multiple-choice’ issue because choices have to be

made on a range of governance aspects: strategy type, type of communica-
tion, type of policy instruments, etc., while securing that the resulting mix-
ture is working. It also involves solving three interconnected dilemma‘s:
between hierarchy and network, hierarchy and market, and network and
market. This makes it a triple dilemma or trilemma: a trade-off between
three forces (Figure 6).’Curing’ the trilemma not only requires solving the
dilemma between two of the forces, as the third force must also be dealt
with, or it will endanger the trade-off between the first two forces. A good
example is the often observed trade-off between ‘new modes of govern-
ance’ (network and market) in environmental policy, which is a threat to
the idea that the environment should also be protected by legislation.393
390 Jessop (1998: 44): The rise of governance and the risk of failure.
391 Svensson and Trommel (2004: 10): Avoiding a messy mix. Hybrid govern-
ance in labour reintegration.
392 Considine and Lewis (1999: 475): Governance at ground level: The frontline
bureaucrat in the age of markets and networks.
393 Hey (2003: 140): Environmental governance and the Commission White pa-
per: the wider background of the debate.
The term trilemma is used in economic policy394 and increasingly in

governance literature. Jessop uses the term in governance issues pointing
at situations when “agents are faced with choices such that they undermine
key conditions of their existence and/or their capacities to realise some
overall interest.”395 On the level of global governance Slaughter formulates
a central trilemma between accountability, participation and profit.396 Folke
et al. use similar apexes of the trilemma triangle - legitimacy, participation
and effectiveness – in their analysis of social-ecological systems.397
Lundqvist uses the same terms in his analysis of the multilevel governance
of Swedish water resources in relation to the EU Water Framework Direc-
tive.398 The terms they use are aspects of hierarchy (accountability, legiti-
macy), network (participation), and market governance (profit, effective-
ness), respectively.
Fig. 6. Metagovernance and the governance trilemma (own composition)
394 Obstfeld et al. (2004): The trilemma in history: Trade-offs among exchange
rates, monetary policies, and capital mobility.
395 Jessop (2005): The governance of complexity and the complexity of govern-
ance, revisited.
396 Slaughter (2004): A new world order.
397 Folke et al. (2005): Adaptive governance of social-ecological systems.
398 Lundqvist (2004): Integrating Swedish water resource management: a multi-
level governance trilemma.
How can the trilemma model be used to understand the mechanisms of

metagovernance? A good example is how Shell uses a trilemma in its
‘Global Scenarios to 2025’.399 (See also 2.1.3). The Shell trilemma de-
scribes trade-offs between three competing forces, that offer three alterna-
tives (scenarios) of the future that are laid out as ‘two-wins, one loss’ op-
tions. The three forces that make out the apexes of the triangle are similar
to the three governance styles this research uses: coercion and regulation
(more or less congruent with a hierarchical style), social cohesion and the
force of community (congruent with a network style) and market incen-
tives and efficiency (congruent with a market style). Shell distinguishes
three analytical layers: actors, objectives and forces. Their scenario study
focuses on forces that shape behaviours and expectations, because actors
react to different forces and objectives can be reached by using mixtures of
all three forces.400
Besides the ‘two win, one loss’ mechanism, the Shell model proposes
to use points in the trilemma triangle to show complex policy trade-offs
and social choices, “as competing forces pull towards the three triangle
apexes.” 401
2.5.5 Metagovernance, line, project and process management
Managing the three ideal types of governance requires adaptation of man-

agers to the logic of the styles. In a hierarchical setting, the organisation
form of change processes seldom has the form of an adhocracy402: change
is achieved by using the standard line organisation. Management is primar-
ily line management. More flexible is the project organisation: a temporary
team consisting of representatives of units of the line organisation. Project
management is on the one hand a threat to the line organisation for exam-
ple because it may lead to jealousy of line managers.403 On the other hand,
besides the different structure, project management builds strongly on the
logic of hierarchy. It aims to control resources and typically works in
phases towards a result that has been defined in the beginning.404
399 Shell (2005): The Shell Global Scenarios to 2025.

402 Term coined by Mintzberg (1985): Strategy formation in an adhocracy.
403 In ‘t Veld (personal communication).
404 Table 11 in Section 7.2.2 summarises the differences between project and
process management.
The logic of the ideal type market governance does not prescribe a form
of organisation, although often a project organisation will be chosen be-
cause its flexibility. The logic of network governance requires a network
form of organisation with an emphasis on process management. This is a
flexible form of management, which builds on the idea that a solution for a
certain problem can only be achieved when relevant actors are involved in
the process between definition of the problem to the choice of a solution.405
None of these types of management (line, project or process manage-
ment) is better than the other: it depends on the circumstances, such as the
type of problem406, the relative influence of internal and external actors,
and the organisational culture407. Metagovernance implies mixing the three
forms of management in a situationally optimal way. It seems that a suc-
cessful metagovernance can be a line manager (who has the advantage of
clear defined resources), a project manager (who has the advantage of
flexibility ‘in the shadow of a robust line organisation) or a process man-
ager (who has the advantage of being allowed to bring together all actors
that have vested interests in an issue).
2.5.6 Governance, metagovernance and performance
How are tensions and lack of synergy in governance style mixtures related
to performance of public administration? Although the concept of public
performance measurement has been around for at least a century, it was
New Public Management that actively emphasized the significance of it.408
Performance measures and indicators were meant to improve the effi-
ciency and effectiveness of producing output409 - a market governance
405 De Bruijn et al. (2002, Process management. Why project management fails
in complex decision making processes) present 15 design principles for good
process management.
406 The type of problem is not ‘given’: problems are framed in a certain way and
can be reframed.
407 In a hierarchical organisation, project management will lead to more jealousy
than in a market-type organisation, and process management seems unfeasi-
ble; in a network organisation project management may lead to irritation be-
cause of its relative rigidity (fixed problem definition, fixed project phases,
compared to the process rounds typical for network management).
408 Bouckaert and van Dooren (2003): Progress and regress in performance
measurement systems.
409 Jackson (1988: 11): The management of performance in the public sector.
Boyle (1996: 3): Measuring civil service performance.
term, whereas network governance uses the term outcome. Defining objec-
tives, goals and targets was believed to motivate civil servants and help to
restore public confidence in government.410 Recent research shows that
administrative performance may indeed be treated as a precondition to
trust in governance.411
It has been shown that improving and measuring performance of public
administration is very difficult. Bouckaert described thirteen ‘management
diseases’ that point at possible defects in performance measurement sys-
tems.412 Others raised the question of how the impact of an individual pub-
lic administration organisation on a complex societal problem can be as-
sessed, when other organisations – public, private and non-governmental –
also influence the results. Another problem is that striving for maximum
performance of a single organisation may lead to a sub-optimal contribu-
tion to solving the societal problem completely.413 The causal logic of ra-
tional performance tools that fits well in a combination of hierarchy and
market is too ‘messy’ for a reality in which organisations also work to-
gether in networks. Uusikylä and Valovirta suggest a solution that consists
of three spheres of performance governance: the organisation’s internal
perspective (focusing enabling factors for successful operation), single-
organisational performance targets, and multi-organisational targets. 414
Jackson draws a similar conclusion. He concludes that for improving pub-
lic administration performance, neither hierarchy nor market mechanisms
provide sufficient solutions. Instead of control and competition, co-
operation and participation in networks will help administration to deliver
added value.415 There is a growing literature on performance and network
governance.416 The assumption that a change in the legal status of an or-
ganisation (for example from ministerial department to executive agency, a
410 Jackson (1988: 11). The management of performance in the public sector.
411 Vigoda and Yuval (2003): Managerial quality, administrative performance,
and trust in governance revisited. A follow-up study of causality.
412 Bouckaert (1995): Improving performance measurement.
413 Uusikylä and Valovirta (2004: 1): Three spheres of performance governance.
Spanning the boundaries from single-organization focus towards a partnership
network.
414 Uusikylä and Valovirta (2004: 18): Three spheres of performance governance.
415 Jackson (2001: 5): Public sector added value: Can bureaucracy deliver?
416 E.g. Farquhar (2003): Network governance and performance measurement in
Healthcare; Imperial (2004): Collaboration and performance management in
network settings: lessons from three watershed Governance Efforts.
market governance approach), per definition improves its performance, has

been empirically proven to be not correct.417
According to Skelcher and Mathur, governance arrangements, designed
from elements of what they call the ‘hierarchy-network-market tryptich’,
are contingent on organisational performance. A governance arrangement
is
“a technical fix to a problem of policy delivery, a temporary structure
within which inter-organisational commitments can be generated and de-
livery managed. When efficiency suffers, the structure is changed. […]
Governance arrangements, then, are subsidiary to the imperatives of organ-
isational performance.”418
Thus, it seems plausible that metagovernance, as designing and manag-
ing mixtures of hierarchy, market and network governance, is a prerequi-
site for improvement of public administration performance. It should also
be postulated that performance indicators have to vary with governance
mixtures that are applied. Discussing public-sector performance raises the
question of how (meta)governance can be measured. This will be briefly
discussed in Section 7.3.3.
2.5.7 Metagovernance and individual cultures
Metagovernance requires the ability of metagovernors to look beyond their

own perspective, in order to at least ‘see’ hierarchy, market and network
elements as the building stones of their organisation. They also need to
have the ability to understand tensions and conflicts between elements of
these styles and be able to design and manage mixtures that work well in a
certain context. Finally, politicians and public administrators when acting
as metagovernors must be able to connect their work with the metagovern-
ance tasks of politicians. Hey et al.419 highlight this as an important lesson
in the case of the hybrid (hierarchy-market-network) design of the EU
chemicals policy. Therefore, metagovernance is the outcome of coopera-
tion among many actors.420 Jessop formulates three key dimensions of me-
417 Dunshire et al. (1991): Organizational status and performance: Summary of

the findings.
418 Skelcher and Mathur (2004: 23-24): Governance arrangements and public sec-
tor performance: Reviewing and reformulating the research agenda.
419 Hey et al. (2006: 27): Better regulation by new governance hybrids? Govern-
ance styles and the reform of European chemicals policy.
420 Sørensen (2006: 104): Metagovernance. The changing role of politicians in
tagovernors’ abilities.421 Firstly, requisite variety: deliberate cultivation of a

flexible repertoire of responses. Secondly, a reflexive orientation: regular
re-assessment of to what extent current actions are producing desired out-
comes. Thirdly, self-reflexive ‘irony’: recognition of the likelihood of fail-
ure while proceeding as if success were possible.
With this, Jessop acknowledges that metagovernors must be able to
handle complexity very well. This requires that metagovernors are, among
other things, ‘mindful’ managers. They must be able to deal well with un-
expected events, which includes the counterintuitive act of seeing the sig-
nificance of weak signals and being able to react strongly to such sig-
nals.422 Chapter 7.2.6 will discuss the implications of the individual
personal dimension of governance and metagovernance for management
development programmes.
2.6 Conclusions
In this chapter, we have analysed a large body of governance literature.

The conclusions are:
- Hierarchical governance is, empirically, still a major way of govern-
ing, notwithstanding the abundant literature that claims that ‘every-
thing is network’ or that ‘market concepts are the main solution for
public-sector problems’.
- The evidence that hierarchical, network and market governance ap-
pear in mixtures in contemporary public-sector organisations, is con-
vincing.
- This implies that public managers have to cope with three competing
‘forces’.
- These forces are embedded in cultures, and maybe even more: they
very much resemble the main ‘ways of life’ of cultural theory.
- The question of whether conscious design and management of gov-
ernance style mixtures is feasible, and if so, to what extent, is dis-
puted. Some claim that governance mixtures are merely contingent;
others argue that they are just a matter of practicality.
processes of democratic governance.

421 Jessop (2002): The network society, new forms of governance, and democ-
ratic renewal.
422 Weick and Sutcliffe (2001: 4): Managing the unexpected.
http://www.springer.com/978-3-7908-2053-9

Theoretical Framework

Uploaded by

Copyright:

Available Formats

Theoretical Framework

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Theoretical Framework

Uploaded by

Copyright:

Available Formats

2 Theoretical framework

2.1.1 A broad definition of a popular concept

How can public-sector activities be coordinated internally and externally if

26 Peters (2000: 35): Globalisation, institutions and governance.

- Governance as ‘good governance’: a ‘marriage of the new public

30 Stoker (1998): Governance as theory: five propositions.

given by Schuppert: with a narrow governance definition it is impossible

Governance is the totality of interactions, in which government,

In this definition, the institutional dimension is included: the relations

36 Schuppert (2007: 8): Was ist und wozu governance?

2.1.2 Three ideal-types of governance and hybrid forms

Governance styles can be defined as “the processes of decision-making

40 Meuleman (2003): The Pegasus Principle: reinventing a credible public sec-

43 Thompson (2003: 37) makes a useful distinction between ‘coordination’

Hybrid forms of governance

Governance hybrids do not necessarily relate to so-called ‘hybrid organisa-

as intercultural entities, are able to bridge fragmented and decoupled rela-

An oligopoly is a market characterised by a small number of organisations

55 In ‘t Veld (1997: 148): Noorderlicht. Over scheiding en samenballing.

formal agreement: this is known as a cartel. A situation where there is a

Public private partnerships

A public private partnership (PPP) is a non-hierarchical system of govern-

Chain management is a form of governance similar to network govern-

59 Börzel and Risse (2002): Public-Private Partnerships: Effective and legitimate

Open method of Coordination

Another governance style hybrid is the so-called Open Method of Coordi-

Self-regulation and self-organisation

cietal organisations.69 Klijn and Koppenjan define self-regulation as an in-

A new form of self-organisation that seems to emerge is what Demil and

69 Puppis et al. (2004: 9): Selbstregulierung und organisation. Schlussbericht.

2.1.3 Are hierarchies, networks and markets ‘all-inclusive’?

77 Benz (2006: 35): Eigendynamik von Governance in der Verwaltung.

81 Kickert (1997: 744): Public governance in the Netherlands.

ernance. The concepts of hierarchical, network and market governance to-

2.2 Hierarchical, market and network governance

2.2.1 Hierarchical governance

83 Weber (1952): The essentials of bureaucratic organization: An ideal-type con-

bureaucracy changed its shape in the 1970s, a decade characterised by a

84 Van Gunsteren (1976: preface): The quest for control.

employees with a law background: they are used to reduce complexity by

92 Schutter (2007: 40): Was ist und wozu Governance?

pluralist and corporatist (like in Germany), and consensus-oriented (like in

96 Herbst (1976: 23-28): Alternatives to hierarchy.

Alternative to Anglo-American 'Managerialism'.

inside the public sector. Thus, government governance typically addresses

2.2.2 Market governance

From the 1980s, the ‘managerial’111 and market-oriented reform movement

- Hands-on professional management of public organisation;

had already developed variable pay (an instrument to enhance competition

2.2.3 Network governance

of governance by a single controller.142 Long before the concept of an

142 Hanf and Scharpf (1978): Interorganizational policy-making; limits to central

ness organisations).149 150 Public sector reform programmes of the early

149 Kickert (1997: 735): Public governance in the Netherlands.

Network governance and knowledge based network organisations

159 Thorelli (1986: 38): Networks: between markets and hierarchies.

work organisations’162. Networks work well in knowledge-rich environ-

162 Term coined by Roobeek.

2.2.4 Forms of network governance

It is no surprise that a whole range of sub-forms of network governance