Trust and Control in Network Relations: A Study of a Public Sector Setting

• Author: Gustaf Kastberg
• Date: 01 February 2016
• DOI: http://doi.org/10.1111/faam.12079
• Published date: 01 February 2016

Financial Accountability & Management, 32(1), February 2016, 0267-4424

Trust and Control in Network

Relations: A Study of a Public Sector

Setting

GUSTAF KASTBERG∗

Abstract: This study contributes to the literature by addressing the role of trust

and control in public sector settings in which several organizations cooperate; its aim

is to advance the understanding of trust and control in relations in which several

parties are involved. The empirical study presented is a longitudinal case study of

the relationships between organizational units within a municipality and a shared

service center (SSC) that provides IT and administrative services. The main finding

is that, compared to actors in previous studies of dyadic relationships in the private

sector, actors in network relations seem to consider trust a risky option. It is less

tempting to rely on a certain party when that party in turn is entangled with other

parties. This situation causes a more intensive use of formal control. Another result

is that increased trust between two parties might lead to more emphasis on formal

control by a third party.

Keywords: trust, control, interorganizational relations, public sector

INTRODUCTION

In the public and private sectors, more and more aspects of organizational

activities are dependent on external parties (Broadbent and Guthrie, 2008),

creating interorganizational relations (IORs). Shared service centers (SSCs),

outsourcing, and strategic alliances are examples of this trend. The public sector

is increasingly organized and controlled by establishing intraorganizational

boundaries that work in the same way as interorganizational relationships

(Peters, 1998). Models such as the purchaser-provider split and voucher systems

have, for example, been introduced to varying degrees in different countries as

part of New Public Management (NPM; Hood, 1991). What characterizes these

∗The author is from the School of Public Administration, Gothenburg University, Sweden.

Address for correspondence: Gustaf Kastberg, School of Public Administration, Gothen-

burg University, Gothenburg, Sweden.

e-mail: gustaf.kastberg@spa.gu.se

C

2016 John Wiley & Sons Ltd 33

34 KASTBERG

models is that they are neither purely market-based nor purely hierarchical

arrangements. These ‘impure’ IORs have been labeled hybrids (Kurunm¨

aki

and Miller, 2011) in order to capture the way they combine various logics of

organization. Despite a growing body of literature focusing on private sector

IORs, less attention has been directed toward such arrangements in the public

sector (Cristofoli et al., 2010).

Consequently, questions have specifically been raised regarding both the

private sector (Otley, 1994; Hopwood, 1996; Tomkins, 2001; and Dekker, 2004)

and the public sector (Kastberg, 2008; Grafton et al., 2011; and Kurunm¨

aki

and Miller, 2011) regarding how the coordination and control of activities

between organizations should take place, which forms of coordination are the

most effective, and the effects various options have. In recent times, however,

attention in the private sector oriented literature has been directed not only

toward what can be described as traditional, formal methods of control but also

toward the role that trust plays in relations between parties (Williamson, 1975;

Das and Teng, 1998; Doz, 1996; Gulati, 1995; Mouritsen and Thrane, 2006;

Chua and Mahama, 2007; Gulati and Nickerson, 2008; and Vosselman and van

der Meer-Kooistra, 2009). The interest in the role of trust in private sector IOR

settings is not reflected in the public sector oriented literature, which contains

only a few explicit discussions of trust and control (C¨

aker and Siverbo, 2011).

Various attitudes toward the relationship between control and trust appear

in the literature. Some scholars maintain that a greater degree of control has

a negative impact on trust between parties (Ghoshal and Moran, 2006; Dekker,

2004; Power, 1997; and Mulder et al., 2006). However, researchers have also

pointed out a positive link between trust and control in the sense that one

reinforces the other (Gulati, 1995; Tomkins, 2001; Poppo and Zenger, 2002;

Vosselman and van der Meer-Kooistra, 2009; and Langfield-Smith, 2008) and

that different forms of control affect trust in different ways (Das and Teng,

2001; and Emsley and Kidon, 2007).

That the literature does not provide a clear-cut picture of the relationship

between trust and control and that the results of previous studies are

contradictory have led to calls for further study (Chua and Mahama, 2007;

Dekker, 2004; Bijlsma-Frankema and Costa, 2005; Vosselman and van der Meer-

Kooistra, 2009; and Emsley and Kidon, 2007). Requests have also arisen that

new studies expand the focus to networks of relations rather than concentrating

on one relation between two parties (Chua and Mahama, 2007; Caglio and

Ditillo, 2008; and H˚

akansson et al., 2010). It has been proposed that the relation

between two parties might be affected by a third or fourth party in a way that has

consequences both for the levels of trust and for the choice of control mechanisms

(Tomkins, 2001). The aim of this study is therefore to advance the understanding of trust

and control in relations in which several parties are involved. The empirical focus is on

network constellations with ‘a number of positions and nodes’ (Tomkins, 2001).

This aim is specified further in the concluding section of the presentation of the

C

2016 John Wiley & Sons Ltd