DISCUSSION
On the basis of the findings presented earlier, our study seems to suggest that a
relationship might exist between network structure, mechanisms, and manager(s) (or
better, between the network centralization, the formalization of the network mechanisms
and the presence of the network manager). In order to ensure network success, in
fact, different mechanisms and managerial abilities seem to be necessary, depending on the network structure. In particular, whereas informal and personal relationships
between the network partners seem to be important in all four centrally integrated
networks, and a network manager appears to be paramount for network success, the
success of the multicentrally integrated network seems to be guaranteed by reliance on
formalized rules and procedures and the sharing of responsibilities among the network’s
multiorganizational governing bodies. Thus, we could say that (1) given the context
(and all other things being equal), different structures require different mechanisms for
the network success; (2) given the context (and all other things being equal), different
structures require different abilities to manage successful networks; and (3) given the
context (and all other things being equal), different structures require different
mechanisms and managerial abilities for network success.
In this way, our study paves the way to new considerations about how to ensure the
network success, the importance of formal network mechanisms in successful networks
and the cruciality of the network manager(s) for the success of public networks.
About how to ensure the network success, first our study goes in the direction
suggested by Turrini et al. (2010) and shows that the network success can also depend
by an interaction effect (mediation or moderation) among the predictors of the network
performance. Different networks structures might in fact require different mechanisms
and managerial abilities (or a different combination of them) in order to be successful.
Secondly, and as a consequence of this, our study seems to suggest that there are a
number of ‘best ways’ to govern and manage public networks, rather than just one
‘best way’ of going about it, depending on the possible multiple combinations of the
abovementioned factors, equally leading to good performance.
This consideration leads to a more general reflection about the convenience of taking
a contingent approach at the network management, as recent studies seem to highlight
(Kelman et al. 2011).
Regarding the importance of network mechanisms, the results of our study seem also
to enrich the existing literature and suggest new considerations. The existing literature
stresses the importance of formalized integration, coordination and control mechanisms
to manage partner interaction and guarantee network success (Agranoff 2003; Jennings
and Ewalt 1998; Kickert et al. 1997; Klijn 1996; Mitchell et al. 2002). Our study
shows that such instruments are useful in complex networks, where the partners’
different interests might make network management difficult, but also that they are
superfluous in networks where the governance structure is clear and the partner
integration is high. In these networks, the personal relationships between the network
partners seem to offer a more efficient – and above all more effective – integration and
coordination mechanism. This is in keeping with the traditional behaviour of public
sector organizations, and it can be explained by considering the importance of the
bureaucratic culture for them. It is well known that public sector organizations typically
respond to uncertain and complex situations by establishing rules and procedures, as
formalization is considered the best way to attain the final objectives without sharing
the responsibilities (and risks) of achieving them. In contrast, when situations are clear and easy to manage, public sector organizations can also enjoy the benefits of being
flexible.
Finally, regarding the presence of network managers, our study seems to challenge
the importance of the role of the network manager and suggest some considerations
about it, coherent with the literature emphasizing that it is more the managerial actions
than the persons that matters, as one anonymous reviewer suggested us. The existing
literature has typically stressed the need for someone to act as a network mediator,
facilitator and/or leader in order to ensure that the network is successful (Agranoff and
McGuire 1998; Kickert et al. 1997; Mitchell et al. 2002). Little by little various studies
have shed light on the different skills and abilities that a network manager is expected to
develop in order to manage public networks successfully. We summarized them in the
literature review section. However, given the results of our study, the presence of a
network manager does not always seem to be necessary for network success. The
Spitex-Indigo network has been successful without a prominent network manager, or
perhaps due to the presence of a number of ‘network administrators’ in the form of
institutional and official bodies, which simply run the network by ensuring that the
partners’ activities comply with the existing rules and procedures. These results can
also be explained by considering the bureaucratic approach of public sector organizations.
As is typical in the bureaucratic culture, public networks seem to prefer sharing
tasks and responsibilities among numerous official actors in complex situations, whereas
they appear to favour taking a managerial and entrepreneurial approach in simple ones.
These last two considerations lead to a more general reflection about the nature and
characteristics of public networks. They were conceived as a new and flexible organizational
form that was able to overcome the rigidity of bureaucracy, but in uncertain and
complex situations, public networks seem to be managed in a bureaucratic way. Are we
witnessing the bureaucratization of public networks?