سیستم کنترل مدیریت عادلانه چگونه می تواند رفتارهای غیر اخلاقی مدیران را کاهش دهد؟

Empirical research has shown that as management control systems (MCS) put heavy pressure on managers to achieve their objectives, they may induce two types of unethical behaviour: creation of budgetary slack and data manipulation. Recently, studies have introduced Organisational Justice theory into the area of management accounting research and shown the positive effects of organisational justice on managers’ attitudes and behaviours. However, few studies have systematically analysed how MCS fairness reduces unethical behaviours and which characteristics of MCS could increase managers’ perception that they are treated fairly. The purpose of the paper is to contribute to this research. It formulates testable hypotheses on how some specific dimensions of MCS can enhance perceived fairness, which in turn helps to reduce the creation of slack and data manipulation through increased Organisational Commitment and Trust in supervisor.

Management control systems (MCS) are widely used in companies. They include all the mechanisms (strategic planning; budgeting; resource allocation; performance measurement, evaluation and reward; responsibility centre allocation; and transfer pricing) managers use to ensure that the behaviours and decisions of their subordinates are consistent with the organization’s objectives and strategies (Anthony and Govindarajan, 2007 and Merchant and Van der Stede, 2007). More specifically, “MCS provide information that is intended to be useful to managers in performing their jobs and to assist organisations in developing and maintaining viable patterns of behaviour” (Otley, 1999, p. 364). Firstly, MCS help managers to make the right decisions by aligning their objectives with the company’s global objectives, and by informing them of their performance so that they can take corrective action if necessary. Secondly, MCS are used to motivate managers: at the beginning of the year, they enable managers to negotiate their objectives and the resources necessary for their achievement; at the end of the year, managers’ performance is assessed by comparing their results with the objectives. Organizations often link managers’ material rewards, such as bonuses, to how well their unit achieves budgeted performance results (Van der Stede, 2000). The ability to meet budgetary objectives is thus a critical factor in managers’ performance evaluation. In addition, managers whose units achieve their budgeted goals may more likely see themselves as “winners” which is a source of psychological rewards such as self-esteem (Merchant & Manzoni, 1989). However, instead of motivating managers and encouraging them to contribute to the achievement of company objectives, MCS may also induce unethical behaviours. In particular, creation of budgetary slack (e.g., Dunk, 1993, Merchant, 1985, Onsi, 1973, Schiff and Lewin, 1970 and Young, 1985), and data manipulation (Bitner and Dolan, 1998, DeFond and Park, 1997, Merchant, 1989, Merchant, 1990, Merchant and Rockness, 1994 and Umapathy, 1987) are two unethical behaviours largely observed in organisations. Slack is generally defined as “resources and effort towards activities that cannot be justified easily in terms of their immediate contribution to organisational objectives” (March, 1988, p. 4). A common type of slack occurs when managers “negotiate highly achievable targets, that is, targets that are deliberately lower than their best-guess forecast about the future” (Merchant & Van der Stede, 2007, p. 183). When created during the management control process, slack is commonly called budgetary slack (Lukka, 1988, Onsi, 1973 and Schiff and Lewin, 1970). MCS provide managers with opportunities for understating revenues and/or overstating costs. Such misrepresentation of managers’ true performance capabilities will frequently be to their advantage since it provides more readily achievable performance targets and increases their chances to obtain monetary (e.g., bonuses) and non-monetary (e.g., favourable evaluations, praise) rewards (Merchant, 1985 and Merchant and Van der Stede, 2007). The second type of unethical behaviour that researchers have examined is data manipulation. Management control systems include performance measurement mechanisms, which organisations use, during the year, to verify that results are in line with objectives and, at year-end, to evaluate managers’ performance. During these performance measurement steps, managers can try to look good by manipulating the performance indicators (Merchant & Van der Stede, 2007, p. 185). Data manipulation comes in two basic forms: falsification and data management. Falsification involves reporting erroneous data, whereas data management, also labelled earnings management, can be defined as “any action on the part of the management which affects reported income and which provides no true economic advantage to the organisation and may, in fact, in the long-term, be detrimental” (Merchant & Rockness, 1994, p. 79). Management accounting research provides considerable evidence that these behaviours exist in organisations (Libby and Lindsay, 2010, Merchant, 1985, Merchant, 1990 and Onsi, 1973). This is a problem because these behaviours distort the planning and performance measurement processes, motivate decisions that are contrary to the organization’s interest, and generate useless costs in the forms of unwarranted rewards. An important issue, yet not much addressed in the literature, is how to design MCS that avoid or, at least, limit managers’ propensity to adopt such behaviours. However, over the past decade, a few management control researchers have started to use the insights provided by research on organisational justice (Cohen-Charash and Spector, 2001 and Colquitt et al., 2001) to show that these unethical behaviours can be reduced when MCS are perceived to be fair (Little et al., 2002, Staley and Magner, 2007 and Wentzel, 2004). Therefore, our purpose is to combine the results produced by research in both management accounting and organisational justice in order to build a general framework (Figure 1) that identifies: (1) which MCS’ characteristics increase their perceived fairness, and (2) how MCS perceived justice1 reduces managers’ propensity to create slack and manipulate data.The contribution of such an endeavour is to help academics, as well as practitioners, better understand how to design and implement effective MCS. The paper is organised as follows. In “Literature review”, we examine the literature to clarify what is known to date concerning the relationships between MCS components, MCS perceived justice and managers’ propensity to create budgetary slack or manipulate data. In “Designing fair MCS to reduce unethical behaviours: identifying key variables”, we build a set of propositions concerning: (1) the most important MCS characteristics that are likely to increase managers’ perceived MCS justice and (2) how justice operates, that is to say, the reasons why MCS that are perceived as fair may reduce managers’ propensity to create slack or manipulate data. Finally, in “Conclusions and future directions”, we conclude with an exploration of managerial implications and suggestions for future research.

Building on management accounting literature and on organisational justice literature, this paper has tried to develop a framework showing how fair management control systems can reduce managers’ unethical behaviours. Perceived fairness is central to this framework. The contribution of this theoretical framework is twofold. Firstly, it shows that four features play a key role in enhancing MCS fairness: participation in target-setting; application of the controllability principle; quality of feedback; and use of multiple performance measures. Although some of these dimensions have already been examined in management accounting studies, no attempt has been made to date to link these characteristics and examine their interrelated effects on managers’ perception of MCS fairness. In addition, the majority of these studies have focused on one single dimension of organisational justice at a time. In particular, much attention has been devoted to procedural justice. However, research has shown that the different types of justice interact to affect managers’ general perception of fairness (Ambrose and Arnaud, 2005 and Cropanzano and Ambrose, 2001). Our framework is in line with this multidimensional approach of organisational justice and identifies the key variables of a MCS best able to increase one or several dimensions of MCS perceived justice. Secondly, the framework identifies two key consequences of MCS perceived justice that explain how the latter can help to reduce unethical behaviours: organisational commitment and trust in one’s supervisor. The purpose of MCS is to motivate managers to achieve their objectives. However, when too much emphasis is put on goal achievement, managers may be tempted to create budgetary slack and manipulate data. What our framework shows is that attention paid to MCS fairness will affect managers’ attitudes both towards their company as a whole and towards their supervisors. Being committed to their company and trusting their superiors are two attitudes that reduce managers’ propensity to adopt unethical behaviours detrimental to the organization’s interests. In terms of managerial implications, our framework provides organisations with guidelines on how to improve MCS perceived fairness and, more particularly, indicates that there are two aspects to pay attention to: the MCS design and the way it is implemented. Firstly, organisations must design their MCS carefully. For example, the objectives set for each manager should take into account elements that s/he can really influence (controllability principle); procedures should be defined to organise discussions between managers and their supervisors at the target-setting stage; similarly, there should be a formalised feedback throughout the year and the performance evaluation system should not include financial indicators exclusively. Secondly, managers’ perception of justice will also depend on how the MCS is actually implemented. Defining procedures for discussions between a manager and his supervisor during the target-setting stage is not enough; the supervisor has to truly play the game and engage in honest and open communication. Similarly, when partial results are issued during the year, discussions between the managers and their supervisors need to take place in order to analyse the situation jointly and define corrective action if necessary. Finally, at year-end evaluation, supervisors should carefully consider how uncontrollable factors might have affected managers’ results so as to neutralise them, specifically when they have negatively affected a manager’s performance. Keeping a fair MCS alive is not only a question of defining rules and procedures: managers must fully appreciate the importance of fairness and be willing to involve themselves, knowing that fairness is a key dimension of their role. Although our theoretical framework contributes to better understand how MCS can limit unethical behaviours, additional work remains to be done, which opens new perspectives for research. Firstly, the model needs to be tested. Although some of the relationships have already been empirically examined, some others have never been investigated. Measurement instruments exist for most variables but some need to be adapted to a management control context. For example, items have been developed to measure participation in budget goals (Milani, 1975 and Kenis, 1979), quality of feedback (Kenis, 1979 and Steelman et al., 2004) or the use of non-financial measures (Lau & Moser, 2008). Recently, Burkert, Fischer, and Schäffer (2011) have designed an instrument to measure the application of the controllability principle. Instruments developed to measure distributive justice (Magner et al., 1992), procedural justice (Price and Mueller, 1986, Colquitt et al., 2001 and McFarlin and Sweeney, 1992) and interactional justice (Bies & Moag, 1986) can be modified slightly to be applied to a management control situation. For instance, Hartmann and Slapničar (2009) have adapted McFarlin and Sweeney’s instrument (McFarlin & Sweeney, 1992) to measure perceived procedural justice in determining targets and evaluating performance. Measuring slack creation or data manipulation is more difficult, since people may not want to show or say that they adopt unethical behaviours. However, some instruments have been used in prior studies. For instance, Merchant (1990) designed an anonymous questionnaire in which managers were asked how frequently they manipulated performance measures, such as shifting funds between accounts to avoid budget overruns or deferring a needed expenditure. In terms of research methodology, cross-sectional surveys can of course be used. However, longitudinal studies could also be relevant to take into account the possible delays with which perceptions appear and evolve. Experimental studies, although usually difficult to implement in managerial contexts, would also offer the possibility to better control for contingencies. Finally, qualitative studies based on interviews or case studies should not be disregarded as they can help better understand how perceptions and attitudes take form. Secondly, other dimensions can be explored in order to refine the framework. A more systematic analysis of the antecedents as well as the consequences of MCS fairness is needed. For example, one can examine the impact on perceived fairness of other MCS components that have been shown, in past research, to influence attitudes and behaviours (e.g., goal clarity, goal achievability, subjectivity). The same can be said regarding the consequences of MCS fairness: how does it affect other behavioural dimensions (e.g., motivation, job-related tension, citizenship) and eventually performance? One should also consider contingent or moderating effects. For example, studies have shown that national culture influences the way individuals react to MCS (e.g., Merchant, Van der Stede, Lin, & Yu, 2011) or form their perceptions of fairness (Beugré, 2007) or of ethics (Ardichvili et al., 2012). Thus, one can assume that national culture may moderate some of the relationships described in our model. As an example, one could test how uncertainty avoidance influences the effect of the controllability principle on unethical behaviours: an individual with a high degree of uncertainty avoidance may find the non-application of the controllability principle more unfair and try to protect him/herself by creating more budgetary slack than a person who is less risk-averse. Many other moderating variables could certainly be examined. Another direction to explore is the dynamics between variables over time. As the management control process takes place over a period (usually a year) and is repeated over time, the effects of MCS characteristics on managers’ attitudes at some point may affect justice perceptions and behaviours subsequently. For example, managers who perceive their performance evaluation to be fair at the end of a year may be less inclined to build budgetary slack when negotiating their objectives for the following year. Thus, it would be interesting to explore whether and how MCS fairness strengthens (or weakens) over time. Research in the management control and behavioural accounting areas have a lot to learn from organisational justice literature. Although most of the impacts of MCS features on unethical behaviours were known, organisational justice theory provides a better understanding of the mechanisms through which MCS may increase or decrease managers’ propensity to adopt those behaviours. Similarly, researchers as well as managers can benefit from a better knowledge of ethics literature (Cropanzano & Stein, 2009). There is a debate about what constitutes (un)ethical behaviours and how to ensure more ethics in organisations. Much is still to be done in the management accounting field on these issues (Merchant & Van der Stede, 2000). As Berry et al., underlined (2005, p. 300), “Ethical perspectives on organisation seem to be markedly absent from the majority of mainstream management control texts”. Finally, researchers in ethics and in organisational justice should take an interest in the MCS. These are widespread practices in many organisations, where outcomes, procedures as well as interactions between people can be studied, and where high stakes can easily motivate unethical behaviours.