Executive Summary

A major role of management accounting is to manage the uncertain behaviors of a range of economic actors, such as suppliers and employees. Using principles from theories such as transaction cost economics and agency theory, a range of controls such as incentives and contracts have been developed to manage these uncertainties.

However, an increasing number of researchers believe that controls cannot cope with every uncertainty and that concepts such as trust are also important. Indeed, arguably, trust is already taken into account when designing the management accounting system, but its role has not been widely recognized by management accounting researchers. For example, tight budgetary control implies a lack of trust, which can lead to dysfunctional behavior such as budget gaming. Conversely, loose budgetary control implies too much trust, which can also lead to dysfunctional behavior such as shirking. Achieving an appropriate balance of trust and control is important because it can potentially minimize these dysfunctional behaviors. Thus, while the relationship between trust and control remains a relatively unexplored topic within the management accounting literature, there is considerable potential to improve our understanding of the relationship which, in turn, can lead to better management of organizational uncertainties.

We are, however, a long way from understanding how trust and control coexist, and many basic issues remain unresolved, such as whether trust is simply one form of control or is different from control. There is also some debate as to whether trust and control are substitutes or complements for each other. While early research has suggested that trust and control are substitutes, recent research points to a more complex and complementary relationship.

This paper argues that while both trust and control have the same objective of managing behavioral uncertainties, they achieve it in different ways. Control is a means of influencing another’s behavior, whereas trust is the confidence in predicting that another’s behavior is not intended to harm one. Moreover, arguably, the relationship between trust and control is more complementary than substitutive because information from controls is also used to determine the level of trust. Consequently, as a major provider of control-related information within organizations, the design of the management accounting system also has a major role in determining trust.

However, the research on trust and control has not developed to a point where any optimal configuration of trust and control can be determined. Indeed, many of the seminal papers examining the relationships between trust and control are theoretical and have not been empirically examined. This study aims to empirically examine some of these relationships, but it also aims to explore them further: (a) by understanding how trust and control might work at different levels in the organization, and (b) by differentiating between information that builds trust and destroys trust.

In order to distinguish these relationships, it is necessary to break trust and control into their component parts. In particular, this paper examines information from three different types of control (output controls — for example, budgeting; behavioral controls — for example, contracting; and social controls — for example, shared commitments). The information from these three controls are examined in terms of their effect on two different types of trust (competency trust — for example, technical capabilities; goodwill trust — for example, willingness to harm another’s interests).

Disaggregating trust and control into their component parts is necessary to examine the relationships more closely; as a result, these relationships become increasingly complex. For example, output, behavioral, and social controls vary across organizational levels. Consequently, controls affecting operational managers are likely to be different from those affecting executives. As a result, these different controls are likely to produce information that affects competency trust and goodwill trust differently. Examining each and every one of these complexities is clearly beyond the scope of any single study; thus, we focus on what we perceive to be some of the more important relationships. In particular, this paper focuses on the information produced from output, behavioral, and social controls that affects competency trust at an operational level and goodwill trust at an executive level. This information is initially examined in terms of information that builds trust and, subsequently, in terms of information that destroys trust.

The theory underpinning these relationships culminates in two research propositions. The first research proposition states that all three controls (output, behavioral, and social controls) provide information relevant to building competency trust at an operational level, but only social controls provide information for building goodwill trust at an executive level. The second research proposition states that while it takes less information to build competency trust than to destroy it, the opposite applies to goodwill trust; that is, more information is required to build goodwill trust than to destroy it.

These propositions were examined using a case study of a joint venture between two international airlines. A joint venture, like strategic alliances and outsourcing, is an example of an interorganizational structure that spans two or more organizations. There are several reasons why such structures are more appropriate for studying the relationship between trust and control than traditional, hierarchical organizations. First, the recent proliferation of interorganizational structures such as joint ventures suggests that more research is required, given our lack of understanding about how these structures are managed. Second, trust is more likely to be observed in a joint venture than a traditional, hierarchical organization because partners in joint ventures cannot gain each other’s cooperation by telling each other what to do (as they might their own employees), and they need to rely more on trust.

The findings from the case study broadly confirm the expectations of both research propositions. We found evidence to support the first research proposition that information from output, behavioral, and social controls helped to build competency trust at an operational level but that only information from social controls helped to build goodwill trust at an executive level. Moreover, the reliance on social controls and the infrequent contact between executives meant that the level of goodwill-related information was relatively low and goodwill trust developed slowly.

The second research proposition concerned the different amounts of information needed to build and destroy competency and goodwill trust. Competency trust was not destroyed in this case study; consequently, it is not possible to comment authoritatively about whether more information was needed to destroy competency trust than build it. It is only possible to state that competency trust was robust and unaffected by the tensions that arose in the case study. However, it is possible to comment on the information needed to destroy goodwill trust. The rapid decline of goodwill trust between the partners was based on the relatively small but meaningful amount of information contained in the decisions that one partner made that disadvantaged the other (partner). These findings support the second research proposition that goodwill trust is a fragile concept that takes relatively less information to destroy than build.

In terms of implications for practice, there are several comments to make. First, except for the joint venture contract, the pattern of controls at operational and executive levels mirrored those within each organization, where a greater level of control (especially from output and behavioral controls) exists at the operational level than at the executive level. This is perhaps not surprising because executives determine controls within their own organizations, and they are unlikely to see themselves as a threat to goodwill. However, this situation becomes problematic in joint ventures, where threats to goodwill trust occur between executives of joint venture partners. Consequently, partners need to be mindful of both the possibility of a betrayal of goodwill trust at both executive levels and the need for, as well as the limitations of, social controls to promote goodwill trust. Second, the effect of the joint venture contract to specify partners’ behavior was limited in this case study because the joint venture was open-ended and evolved over time. Consequently, the contract could not possibly specify ex ante all the actions by which partners might disadvantage each other, especially those relating to actions taken outside the context of the joint venture itself. Finally, reliance on information from controls to manage trust probably needs to be supplemented with information systems that are specifically tailored to the management of trust.

All Images & Content Copyright 2006 - Canadian Academic Accounting Association
Powered by WebPal CMS, Palomino Inc.