Accounting for control and trust building in interfirm transactional relationships

Abstract

This paper theorises the accounting-control-trust nexus in interfirm transactional relationships. In the context of such relationships, accounting has predominantly been conceptualised as a control technology. However, in our paper we analyse stable and durable relationships as being the results of interaction between control and trust building. Such an analysis calls for an additional conceptualisation of accounting as a trust building technology. Furthermore, we explain the interaction between accounting for control and accounting for trust building in the context of a process of embedded agency.

Accounting for control is underpinned by a governance structure, which includes accounting structures that have the potential to act as safeguarding and incentivizing devices with the aim of aligning long-term interests. The governance structure is flanked by trust from the institutional environment of the transactional relationship. Accounting for trust building originates from voluntary local decisions to show commitment to the relationship. It takes the form of ad hoc calculations and accounts and becomes a device for relational signalling. Adequate interaction between accounting for control and accounting for trust building results in stable and durable interfirm transactional relationships.

Introduction

Ever since Otley observed that “the scope of the activity of management control is enlarged and it no longer confines within the legal boundaries of the organization” (Otley, 1994, p. 293) and Hopwood urged “the examination of some of the accounting and informational consequences of more explicit concerns with the management of the supply chain and a more conscious questioning of what activities reside within and without the enterprise” (Hopwood, 1996, p. 590), many accounting studies have dealt with interfirm transactional relationships. These studies take a rationalist approach, a constructivist approach or a combination of the two. Rationalist approaches to accounting and control in interfirm transactional relationships theorise and examine the instrumental relationship between control problems and control devices designed and used to manage these problems. A few of these studies adopt a contingency approach without explicitly and substantially incorporating an organisational economics analysis (e.g. Cooper and Slagmulder, 2004, Kajüter and Kulmala, 2005, Mahana, 2006). But, most of them – at least partially – draw upon organisational economics; in particular on transaction cost economics (e.g. Anderson and Dekker, 2005, Anderson et al., 2000, Dekker, 2003, Dekker, 2004, Donada and Nogatchewsky, 2006, Gietzmann, 1996, Langfield-Smith and Smith, 2003, Nicholson et al., 2006, Sartorius and Kirsten, 2005, Van den Bogaard and Speklé, 2003, Van der Meer-Kooistra and Vosselman, 2000). The accounting studies that take a constructivist approach (e.g. Cuganesan and Lee, 2006, Franses and Garnsey, 1996, Håkansson and Lind, 2004, Mouritsen et al., 2001, Mouritsen and Thrane, 2006, Seal et al., 1999) highlight the role of accounting in constituting and shaping interfirm relationships. Some other studies combine the constructivist approach with a rationalist approach (e.g. Chua and Mahama, 2007, Free, 2007, Thrane and Hald, 2006). These studies show the instrumental and the social to be intertwined. Furthermore, a few studies draw upon branches of institutional theory other than transaction cost economics; for example, structuration theory (Seal, Berry, & Cullen, 2004) and evolutionary theory (Coad & Cullen, 2006).

A number of studies touch upon the concept of trust and its relation to control. In a rationalist approach, control and trust are both instrumental in absorbing uncertainty and behavioural risks (Dekker, 2004, Emsley and Kidon, 2007, Langfield-Smith and Smith, 2003, Tomkins, 2001, Van der Meer-Kooistra and Vosselman, 2000). Control and trust are predominantly seen as two distinct concepts, either substitutive of or complementary to each other (Dekker, 2004). The substitution perspective suggests that trust and control are inversely related. This implies that more control results in less trust and, vice versa, more trust results in less control. Van der Meer-Kooistra and Vosselman (2000), for example, consider trust to be a substitute to control in interfirm relationships characterised by transactions of a high complexity and surrounded by a high level of uncertainty. In such relationships trust efficiently replaces control in transactional relationships. The complementarity perspective perceives control and trust as mutually reinforcing. Tomkins (2001), in exploring the interaction between accounting information and trust in relationships with different degrees of complexity, suggests that at an earlier stage of the interfirm transactional relationship control may add to trust; in other words, more control leads to more trust. He rejects the rather simple inverse relationship between control and trust. He claims that at an earlier stage of the relationship the accounting information exchange that follows from the use of controls may produce more positive expectations about future contributions to the relationship. In this way, control could add to trust. In a similar vein, Emsley and Kidon (2007) examine the relationship between control-related information and trust in a joint venture of international airlines. Their study reveals that control-related information affected the level of trust, either in a positive or in a negative way. It could either help to build trust or destroy trust.

In a constructivist approach, control and trust are viewed as active forces in a relationship or network, where they help to mediate, shape and construct that very relationship or network. Mouritsen and Thrane (2006), for instance, view control practices (labelled as management control technologies) as carriers or sources of trust, thereby linking up with the notion of a complementary relationship. They provide evidence that there is trust when proper management control technologies are in place. Trust differs from management control technologies in the sense that it is not an object, but a quasi-object. In their view, trust is an aspiration circulating in the interfirm relationship or network. It is a problematising device that becomes important when it is absent. It then mobilises control that facilitates ongoing business in the network. In this way, trust is shown to emerge through management control technologies including accounting.

From a rationalist approach, this paper aims to further contribute to the theory on the relationship between control and trust. Until now there has been little theoretical debate on the control-trust nexus. Hence, there is a strong need to move beyond ‘empiricism’ and to do more theoretical work, thereby creating a more comprehensive understanding of the control-trust nexus (Bijlsma-Frankema & Costa, 2005). More specifically, we intend to explore the accounting-control-trust nexus. Not only do we aim to get a richer understanding of the way control and trust (building) in interfirm transactional relationships are interrelated, but we also aim to get a deeper insight into the way accounting is related to control as well as to trust building. In particular, extant literature fails to explore the way accounting is linked to trust building. It perceives accounting to be directly related to control. Accounting, more specifically, is perceived to be related to external measure based control.² It provides financial and non-financial measures and information on the (output of) the (expected) behaviour of parties in the relationship, not only in order to co-ordinate actions but also in order to cope with appropriation concerns (Dekker, 2004, Gulati and Singh, 1998). External measure based controls (or “hard controls”) aim to incentivize, monitor or constrain actions of the parties involved, thus safeguarding against potential opportunistic behaviour by which a party searches self-interest at the expense of the other party in the relationship. So, these controls cope with behavioural uncertainty by making behaviour predictable and transparent.

Although control-related accounting information is claimed to have an impact on the level of trust (e.g. Emsley and Kidon, 2007, Tomkins, 2001), accounting information has not been directly related to trust building. As we argue that it needs both control and trust to absorb behavioural uncertainty and to effectively cope with appropriation concerns, we claim that the scope of accounting should include the building of trust. Therefore, in order to broaden its scope we will not only conceptualise accounting as a control technology, but also as a trust building technology. To that end, trust is regarded to be an outcome of a trust building process at the level of the transactional relationship.³ This outcome implies that a party is willing to accept vulnerability, although this party is not completely sure that the other party will not behave opportunistically. Our notion of trust is consistent with views expressed by other authors. For, although the concept of trust proves to be more ambiguous than that of control, and it could even be stated that the issue of defining trust remains largely unresolved (Das & Teng, 2001), the willingness to accept vulnerability (Mayer et al., 1995, Mc. Evily et al., 2003, Schoorman et al., 2005) may be considered to be the core of trust. Positive expectations about the ability, the benevolence and the integrity of the other party form the input for such a willingness, whereas risk taking in the relationship is the outcome of this willingness. The latter is the behavioural manifestation of trust. Trust thus encompasses perception, volition/intentionality and behaviour (Mc. Evily et al., 2003).

From a rationalist approach, the paper adopts the view that trust and control have a common goal: they are both instrumental to the absorption of behavioural uncertainty. It takes the perspective that in order to absorb behavioural uncertainty, control and trust do not just substitute or complement, but interact. An interaction approach includes substitution as well as complementary connections, but acknowledges that both trust and control are not static. The rather simple ‘and/or’ or ‘and/and’ debate on the relationship between control and trust downplays the dynamics between trust and control. An interaction perspective accounts for the interrelated dynamics of control and trust.^4,5 Control needs trust, and trust needs control. At the same time, control produces trust and trust produces control. In the course of the relationship, control may change, and such change may have direct as well as indirect effects. It may directly affect the absorption of behavioural uncertainty, but may also have a direct effect on the level of trust, with a subsequent (indirect) effect on the absorption of behavioural uncertainty. Furthermore, the paper extends earlier work by relating accounting to control as well as to trust building. Accounting is present in formal control practices stemming from a contractual arrangement as well as in voluntary interactions between individuals that result in trust. In a control context, accounting is argued to be directly instrumental to the alignment of interests of the parties involved, by compensating for negative behavioural expectations (fear for opportunism). Accounting helps to incentivize, safeguard or constrain the actions of parties involved. In the context of trust building, accounting is argued to be directly instrumental to the generation of positive behavioural expectations. Here, accounting serves the process of relational signalling, a process we argue to be at the heart of the trust building process. Accounting, then, helps in stimulating the willingness to accept vulnerability.

The remainder of the paper is organised as follows. In the next section accounting for control is analysed in terms of control practices enabled by a centre. Such a centre is a governance structure in the form of a contractual arrangement that offers safeguarding devices as well as incentive systems, thereby constraining and directing the behaviour of the parties involved (formal control structures). We will show that in order to be effective, the governance structure and incorporated formal control structures need trust that originates outside the relationship, while the formal control structures and the practices that stem from them produce trust at the level of the relationship. The third section contains an analysis of the drivers and the mechanisms of accounting for trust building. We will show that trust building needs a governance structure and incorporated formal control structures. Together, trust and control constitute positive expectations, thus enabling durable and stable relationships. The fourth section examines the interaction between accounting for control and accounting for trust building. In the final section, we draw conclusions and offer suggestions for future research.