Modes of network governance revisited: Assessing their prevalence, promises, and limitations in the literature

• Published date: 01 November 2023
• Author: Steven Oord,Patrick Kenis,Jörg Raab,Bart Cambré
• Date: 01 November 2023
• DOI: http://doi.org/10.1111/puar.13736

RESEARCH ARTICLE

Modes of network governance revisited: Assessing their

prevalence, promises, and limitations in the literature

Steven van den Oord

1,2,3

| Patrick Kenis

4

| Jörg Raab

3

| Bart Cambré

1,2

1

Department of Management, Faculty of

Business and Economics, University of Antwerp,

Antwerp, Belgium

2

Antwerp Management School,

Antwerp, Belgium

3

Department of Organization Studies, School of

Social and Behavioral Sciences, Tilburg

University, Tilburg, The Netherlands

4

Department of Public Governance, School of

Economics and Management, Tilburg University,

Tilburg, The Netherlands

Correspondence

Steven van den Oord, Department of

Organization Studies, Tilburg School of Social

and Behavioral Sciences, Tilburg University,

Simon Building, room S 607, PO Box 90153,

5000 LE, Tilburg, The Netherlands.

Email: s.vdnoord@tilburguniversity.edu

Abstract

The systematic literature review takes stock of the empirical literature on the

governance of organizational networks. The analysis is based on empirical papers

citing Provan and Kenis (2008) as the seminal article on the governance of

networks. We synthesize key findings on the modes of network governance, con-

tingency factors, and network-level tensions. The review provides insights into

how the contingency theory of network governance has developed into an estab-

lished and recognized research agenda in the last 15 years. We conclude that the

governance of organizational networks as a vocabulary has been adopted in the

management and organization sciences literature to explain organizational net-

works’development, functioning, and effectiveness. However, further theoretical

development and testing are warranted to inform the practice of network gover-

nance, particularly when, how, and why to use institutions and structures of

authority and collaboration to allocate resources and coordinate and control joint

action of groups of organizations.

Evidence for practice

•Practitioners must be aware of the increasing prevalence of situations where

value must be created in concert with other organizations. The organizational

network’s governance mode is central to the success of value creation.

•The main takeaway for practitioners is that there is no best way to organize an

organizational network. Still, we know that network governance modes’adop-

tion and design choices are contingent on the problem/task structure and envi-

ronmental, structural, and relational conditions.

•Practitioners must become mindful of and competent in reading these condi-

tions when developing appropriate governance for organizational networks.

INTRODUCTION

This literature review focuses on what we know about the

different modes or forms of governance of organizational

networks and their effects. It, therefore, looks at network

governance from an organization and management per-

spective. Building on Provan and Kenis (2008), Gulati et al.

(2012), and Puranam (2018), an organizational network is

defined as “a (1) multiagent system of three or more

legally autonomous organizations that are (2) not bound

by authority based on employment relationships but

characterized by (3) a distinct identity derived from a par-

ticular boundary and membership arrangement and

(4) network-level goals toward (5) which the constituent

organizations’efforts are expected to contribute”(van

den Oord, 2023, p. 27). Organizational networks have

become increasingly prevalent in the public and non-

profit domains (Smith, 2020; van der Weert et al., 2022)

but also in various industries and business domains

(Ciabuschi et al., 2020; George et al., 2023; Provan

et al., 2007; Reeves & Pudin, 2022) because they present

an alternative strategy to organizations to deal with

Received: 15 February 2023 Revised: 8 September 2023 Accepted: 26 September 2023

DOI: 10.1111/puar.13736

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium,

provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

© 2023 The Authors. Public Administration Review published by Wiley Periodicals LLC on behalf of American Society for Public Administration.

1564 Public Admin Rev. 2023;83:1564–1598.

wileyonlinelibrary.com/journal/puar

environmental uncertainty and attain a purpose that a

single organization cannot easily achieve independently

(Nowell & Kenis, 2019; Popp et al., 2014; Provan &

Lemaire, 2012).

Each organizational network exhibits a division of

labor, meaning its network-level goal can be separated

into tasks and allocated among several organizations.

Rather than multiple organizations working together

independently, serendipitously, or competing, organiza-

tions joining a network engage in joint efforts to achieve

a set of tasks (Kenis & Raab, 2020). Any organizational net-

work must then find a way of organizing without relying

on formal authority to ensure that organizations collabo-

rate, that their activities are coordinated for a given divi-

sion of labor, that resources are allocated, and that

activities are directed toward achieving the network goals

(Provan & Kenis, 2008). Only by transforming a group of

individual organizations into a goal-directed system

of coordinative action can authority be arranged and col-

laboration be structured to maximize the value of the net-

work for collective and individual outcomes related to the

network purpose (Gulati et al., 2012; Provan et al., 2007;

Provan & Kenis, 2008).

Different levels of analysis and perspectives are

applied to study organizational networks (Berthod &

Segato, 2019; Carboni et al., 2019; Lemaire et al., 2019;

Nowell et al., 2019). Provan and Kenis (2008) suggested

combining the network analytical and governance per-

spectives to consider the network as the unit of analysis

and treat it as a differentiated organizational form. By

combining these perspectives, Provan and Kenis (2008)

developed a contingency theory of network governance

to propose which form of network governance fits best in

which situation and context. The basic assumption is that

the better the fit between the form of network gover-

nance and their contingency factors, the more effective

the network is. Network governance concerns the “use of

institutions and structures of authority and collaboration

to allocate resources and to coordinate and control joint

action across the network”(Provan & Kenis, 2008, p. 231).

It “entails the structure and processes that enable organi-

zations to direct, coordinate and allocate resources for

the network and to account for its activities”(Vangen

et al., 2015, p. 1244).

Provan and Kenis (2008) suggest three ideal types of

network governance modes: (1) a shared-participant

(SP) mode, where network members jointly govern the

network (non-brokered/internal); (2) a lead organization

mode, in which one network member governs the net-

work (brokered/internal); and (3) a network administrative

organization (NAO) mode, where a separate organization

is established to govern the network (brokered/external).

Provan and Kenis’network governance modes, first

published in a paper in the Journal of Public Administra-

tion Research and Theory (JPART), have been widely

adopted and cited in the public management and organi-

zation sciences literature. The paper has become one of

the most cited articles published in a public administra-

tion journal in recent years. However, it has also received

a strong reception outside this field.

1

The paper by

Provan and Kenis (2008) is part of the literature on

organizational networks and governance, more broadly

developed since the 1980s in organization and (public)

management, economics, political science, and sociology.

In (public) management, it directly builds on the seminal

paper by Provan and Milward (1995) on the effectiveness

of networks and has strong connections to questions of

network management (cf. Milward & Provan, 2006;

Provan & Lemaire, 2012).

Furthermore, it further develops the framework sug-

gested by Milward and Provan (2006), which guides (net-

work) managers on types, purposes, and essential tasks

such as accountability, legitimacy, conflict, design, and

commitment. Indirectly, it strengthens Agranoff

and McGuire’s framework, suggesting that decisions on

network governance are part of the activation and mobili-

zation tasks and have substantial implications for the

framing and synthesizing management tasks for networks

(McGuire, 2002). The governance modes of Provan and

Kenis (2008) also partake in the broader discussion in

public management on how to govern collaborative

(organizational) arrangements (Ansell & Gash, 2008). Last

but not least, the paper builds on broader questions of

the governance of social and economic systems as they

have been discussed since the 1970s (Ostrom, 1990; Pow-

ell, 1990, 2011; Williamson, 1975), and more recently, are

reinvigorated by network scholars that study social-

ecological systems to find solutions for grand societal

challenges (Bodin, 2017; Vantaggiato & Lubell, 2022).

While building on the discussion on governance through

markets, hierarchies, and networks, the paper by Provan

and Kenis (2008) represents an important further step

and a change of perspective, that is, the idea that there

are different types of and ways to govern networks that

might lead to different outcomes depending on a set of

contingency factors.

The broad reception of the network governance

framework by Provan and Kenis indicates that organiza-

tions jointly produce collective outcomes through net-

work forms of organization (Powell, 2011) rather than

markets and hierarchies in more areas of the economy

and society. The explanation of collective outputs and

outcomes has traditionally been a focus and a strength of

public administration (PA) research, which, therefore,

becomes increasingly relevant outside the core PA

domain. The broad reception of the governance modes

enables us to take on a prospector perspective in analyz-

ing network governance across many scientific fields,

especially concerning the “blending and merging of liter-

ature across fields”(Breslin & Gatrell, 2023, p. 145) and to

investigate how network governance functions, how it

varies, and what the outcomes are in different settings.

This investigation should also be fruitful input for the the-

ory development in Public Administration. For example,

PUBLIC ADMINISTRATION REVIEW 1565

we still lack systematic empirical evidence on which

mode is effective under what specific conditions and for

what purposes different governance modes achieve inte-

gration of effort by a group of organizations working

together to accomplish a collective set of tasks. Since

governments increasingly use organizational networks as

an intervention (cf. Valente, 2012) to cope with complex

societal and environmental issues (Gray & Purdy, 2018), it

is imperative to develop further and continuously update

our knowledge on the governance of these networks. In

this context, it is indicative that organizational networks

have found their way as the overarching goal 17: “part-

nership for the goals”into the UN Sustainability Goals

(Kapucu & Beaudet, 2020).

We, therefore, propose it is time to take stock of the

empirical evidence about how organizational networks

govern themselves and further develop Provan and Kenis’

network theorem. This article aims to revisit Provan and

Kenis’contingency theory of network governance by

assessing the accumulated knowledge of the governance

of organizational networks as reported in empirical

research from 2008 to 2022. More explicitly, we set out to

systematically review what network governance modes

have been examined in the empirical literature citing Pro-

van and Kenis (2008) because we are interested in what

institutions and structures of authority and collaboration

organizational networks use to arrange three or more

organizations given contingent conditions, manage

network-level tensions, and adapt network governance

modes over time. Given the prominence of the article,

which at the time of writing has been cited 1837 times

according to Web of Science and 5166 in Google Scholar,

and the establishment of the three modes of network

governance, we deem it a valid strategy to take this paper

as the seed article for our literature search. The following

questions guided our review of the empirical literature:

1. Which network governance modes suggested by Pro-

van and Kenis (2008) have been identified in empirical

research?

2. Which other types of network governance modes are

described in the empirical research?

3. What factors explain the differences in governance

modes?

4. What are the consequences of network governance

modes?

We structure the article into three sections. In the first

section, we describe the scope and method of the system-

atic literature review. In the next section, we take stock of

the contingency theory of network governance. This

section contains three main parts, each centering on a core

construct of the theorem. In the third section, we present

lessons learned from reviewing the empirical literature on

the network governance modes and their evolution, contin-

gency factors, systems fit, network-level tensions, and net-

work effectiveness. We close this last section by outlining

implications for practice and a research agenda presenting

what future direction researchonthegovernanceoforgani-

zational networks would be fruitful. We propose to take this

review as the starting point to develop an open science

database on network governance literature, which forms

the basis to continuously update our knowledge on net-

work governance and update the present review every few

years. We would also like to open this up to other

researchers who want to contribute and thus collectively

systematize the existing knowledge and improve the

knowledge aggregation process.

SCOPE AND METHOD OF REVIEW

We conducted a systematic literature review on the empiri-

cal literature citing the journal article: Provan and Kenis

(2008)

2

.Weoptedforthisscopebecauseweareprimarily

interested in whether and how network governance intro-

ducedbyProvanandKenis(

2008) has been used and

developed since its online publication in 2007. We

extracted, mapped, and assessed 1357 journal articles writ-

ten in English that cited the seed article reported in the ISI

Web of Knowledge database (Clarivate, 2019, covering

15 years [2008–2022]). We refined the extraction to the

Web of Science database’s Social Sciences Citation Index

(SSCI). We applied three inclusion/exclusion criteria to deter-

mine eligibility based on the core constructs of Provan and

Kenis’s contingency theory. A formal definition of Provan

and Kenis’theorem is shown below in Table 2.

Given our scope and interest, we excluded articles that

did not consider the network as the unit of analysis (Provan

et al., 2007;Raabetal.,2013), were primarily conceptual or

descriptive, or mainly contributed by methodological con-

siderations and measurements. Simultaneously, we checked

whether articles focused on “network governance”as a

topic in the title and abstract using Provan and Kenis’sdefi-

nition of network governance. We have summarized our

approach in Table 1. Ultimately, the scope and method of

review led to a final selection of 184 articles that we classi-

fied on the seven core constructs of the network-level struc-

tural contingency theory of network governance by Provan

andKenisasdefinedinTable2.

RESULTS

In this section, we take stock of the contingency theory of

network governance. This section consists of three main

parts. In the first part, we identify which network gover-

nance modes we have found in the literature. In the second

part, we explain the differences in these network gover-

nance modes by focusing on Provan and Kenis’soriginal

contingency factors. The third part centers on the conse-

quences of network governance modes. Based on these

parts, we present what lessons we learned from reviewing

the empirical literature on the network governance modes

and their evolution, contingency factors, network-level

1566 MODES OF NETWORK GOVERNANCE REVISITED