استفاده از اصل کنترل و عملکرد مدیریتی : نقش ادراک نقش

The controllability principle stipulates that the evaluation of a manager should be based only on elements that are under the manager's control. Arguments for and against its application are theoretically well understood, but empirical evidence based on the evaluation of the perceptions of managers and their implications for managerial performance is scarce. By empirically analyzing the effects on managerial performance, this paper explores managers’ responses to the application of the controllability principle. We draw on role theory and analyze how role ambiguity and role conflict mediate this basic relationship. Moreover, we test whether application of the controllability principle equally affects role perceptions of top-level managers and those of lower and middle-level managers. Empirical analysis of survey responses from 440 managers reveals that role perceptions completely mediate the effect of application of the controllability principle on managerial performance. This effect is insignificant in the group of top-level managers, who appear to cope with uncontrollable factors more effectively.

The controllability principle is one of the strongest tenets in management accounting, and is considered to be directly relevant to evaluations of managers’ performance (Bhimani et al., 2008, Merchant and Otley, 2006 and Merchant and Van der Stede, 2007). Mostly known as a normative principle, the controllability principle stipulates that managers should be held accountable only for results that are within their control (Atkinson et al., 2007, Dalton, 1971 and McNally, 1980). Extant literature suggests that non-application of the principle leads to decreased motivation and increased role stress, which result in dysfunctional behavior by individual managers (Dent, 1987, Merchant, 1987 and McNally, 1980). Despite this rationale for the application of the principle, there are compelling arguments as to why organizations should not always fully conform to it. Companies can direct managers’ attention to critical performance areas and gain additional information about their hidden actions by partially disregarding the principle (Ittner and Larcker, 2002, Manzoni, 2002 and Simons, 2007). Given the convincing arguments both for and against applying the controllability principle, prior studies not surprisingly found that companies differ in the extent to which they follow the principle in formal performance evaluations (Drury and El-Shishini, 2005, Huffman and Cain, 2000, Merchant, 1989 and Vancil, 1979). Even though researchers have been interested in the controllability principle for some time, quantitative empirical work investigating the actual consequences of its application is sparse (Drury and El-Shishini, 2005 and Simons, 2007) with respect to both the expected benefits of application and the potential drawbacks of strict application. Considering the theoretical importance of the principle and its practical relevance for companies, this paucity of research is startling. As Giraud et al. (2008) commented, “The position of the managers evaluated has not been much examined” (p. 34). This paper aims to foster a better understanding of responses of individual managers to the application of the controllability principle, and especially the potential negative effect of non-application on managerial performance. Building on role theory (Kahn et al., 1964, Kahn et al., 1966, Katz and Kahn, 1966 and Katz and Kahn, 1978), we relate application of the controllability principle to managers’ role perceptions. We argue that application of the controllability principle affects role ambiguity and role conflict. We assume that role ambiguity and role conflict, in turn, affect managerial performance. Because role theory emphasizes the importance of the hierarchical position of individuals for their role perceptions, we also account for possible moderating effects of this variable on the relationships between application of the controllability principle and role ambiguity and role conflict. Investigation of these effects illuminates the dynamics between application of the controllability principle and managerial performance, which is desirable from a theoretical, as well as practical, point of view. Exploration allows the derivation of recommendations on how to effectively mitigate dysfunctional consequences of non-application when full application is not desirable or feasible. We test our theoretical model with data from 440 managers and find support for most of the proposed hypotheses. We find strong and significant effects of application of the controllability principle on role ambiguity and role conflict—the more superiors apply the controllability principle, the less managers experience role ambiguity and role conflict. These effects are significant even after controlling for managers’ trust in their superior. Moreover, we find that it is in particular a decrease in role ambiguity that has a positive impact on managerial performance. Our analyses reveal that the initial direct effect from application of the controllability principle on managerial performance completely disappears once the indirect effect through role ambiguity and role conflict is taken into account. Cognitive perceptions, therefore, fully mediate the dysfunctional effect of non-application of the principle on managerial performance. Moreover, we find that the relationship between its non-application and role ambiguity is insignificant for the subsample of top-level managers. Compared to lower and middle-level managers, top-level managers seem to be better able to deal with uncontrollable factors. They are also better at withstanding dysfunctional cognitive perceptions that result from being confronted with factors beyond their control. Consequently, companies can disregard the controllability principle for their top-level managers more easily than one might have previously assumed. Our findings also suggest that companies can mitigate the dysfunctional effects of its non-application among lower and middle-level managers. Companies should consider using more effective personnel controls, which constitute an important element of management control systems (Merchant and Van der Stede, 2007). Personnel controls comprise (a) the adequate selection of people to achieve a high person–job fit and (b) training to prepare managers for new and stressful jobs. Researchers from the field of organizational behavior find that companies often ignore such controls (Cooper et al., 2001). An ex ante implementation of appropriate systems would help to avoid control problems, and psychological research has in fact provided evidence of the general effectiveness of such measures ( Saunders et al., 1996). Our paper is organized as follows. Section 2 reviews the previous literature on the controllability principle. Section 3 introduces role theory and relates the role episode model to application of the controllability principle. Considering the application of the principle from a role theory perspective, this section develops our hypotheses. Section 4 describes the research method, data collection, and survey instruments, and Section 5 presents the empirical findings. Finally, Section 6 discusses these findings, highlights the limitations of the study, and proposes avenues for future research.

Despite the great interest of academics and practitioners in the controllability principle, few empirical studies have quantitatively examined the consequences of its application in formal performance evaluations. Most of these studies have not gone beyond descriptive analyses or actually examined the consequences of not applying the principle at the level of individual managers. Our paper addresses this oversight. Our analyses show that the effect of applying the controllability principle on managerial performance is not direct but indirect in nature. Managers’ cognitive perceptions of role stress fully mediate this relationship. More specifically, role ambiguity mediates the relationship between application of the principle and managerial performance. When managers perceive the controllability principle as not being applied or only barely applied, they find less clarity in the role expectations imposed on them. Ultimately, this ambiguity lowers their job performance. From a theoretical perspective, our study complements previous empirical work that found the effects of control systems on individual outcomes are not necessarily direct. Instead, individual cognition, and role perceptions in particular, often partly mediate the effects (Burney and Widener, 2007, Hall, 2008, Maas and Matějka, 2009 and Shields et al., 2000). Our moderator analysis reveals that the dysfunctional effect of non-application of the principle on role ambiguity and indirectly on managerial performance is insignificant for top-level managers. Top-level managers seem to be better capable of coping effectively with the uncontrollable factors than lower and middle-level managers. The literature provides some indication as to why this is the case. Managers’ self-image, motivation, and especially attitude toward risk vary with hierarchal level (Kahn et al., 1964 and Sitkin and Pablo, 1992). Moreover, numerous studies emphasize that top-level executives are constantly confronted with high uncertainty (Cannella et al., 2008 and Carpenter and Fredrickson, 2001), implying that they are used to dealing and expect to be dealing with uncontrollable factors. Consequently, the negative effects associated with non-application of the controllability principle do not hold universally, but rather depend on the experience and personality type of the individual managers. On the basis of our empirical findings, we can give a more differentiated answer to the key question of whether companies should always fully apply the controllability principle. First, the question should not be answered independently from the individual manager's position in the corporate hierarchy. A plausible assumption is that including uncontrollable performance measures in the evaluation of top-level managers is particularly appropriate. One advantage of non-application of the controllability principle is that it allows for directing top-level managers’ attention toward critical performance areas which are not fully controllable by them but which are important to the organization. This advantage is particularly notable given the pronounced impact of the decisions and actions of top-level managers on corporate performance (Hambrick and Mason, 1984). Their actions and decisions generally affect corporate welfare much more than actions and decisions of lower level managers. Our results suggest that, for top-level managers, companies can disregard the controllability principle more easily. Apparently, the benefits from such accountability arrangements come at a lower cost than generally assumed in terms of the resultant role stress among managers. Second, important practical implications arise from insight into the cognitive mechanisms that mediate how non-application of the controllability principle affects managerial performance, particularly at lower managerial levels. These mechanisms provide the basis for understanding the potential levers for mitigating dysfunctional managerial behavior (here, lower managerial performance) when companies are not able to fully apply the principle or have a rationale for not doing so. As outlined at the beginning of this paper, not only practical reasons, but also organizational interests often call for an enlargement of managers’ accountabilities and for an erosion of the controllability principle. A notable result, therefore, is that non-application of the principle does not lead to decreased managerial performance unless role stress occurs. The results of our moderator analysis support the view that managers react differently to stress stimuli. They are also consistent with the claim of stress management literature that individuals generally differ in how well they are able to deal with stress (Cooper et al., 2001). As a consequence, companies might want to (a) select and (b) prepare managers more adequately for jobs that expose them to uncontrollable factors. Such measures are referred to as personnel controls (Merchant and Van der Stede, 2007, p. 84), and as they represent important elements of effective management control systems, researchers from the field of organizational behavior are surprised to find that many companies ignore them (Cooper et al., 2001, p. 187). A selection strategy requires a job analysis in combination with an assessment of each manager's characteristics and skills, particularly the ability to deal with stress, as these form the basis for selecting the right manager for the right job (Spector, 2008). Specific personality traits associated with stress resistance could help guide selection and placement decisions. Training represents another suitable personnel control to mitigate resulting dysfunctional effects from non-application of the controllability principle. Sessions helping managers to deal with uncontrollable factors effectively could complement executive education programs, and training might be particularly useful when employees are promoted to a position that has them take over general management responsibilities. Stress researchers have labeled such ex ante attempts at modifying individuals’ responses to stress as secondary interventions ( Cooper et al., 2001). They are important when primary interventions—that is, changes in the working conditions that cause stress—are not possible. Organizations might also introduce stress management programs to buffer managers against levels of stress that inevitably come with relaxation of the controllability principle. So-called tertiary interventions focus on the “treatment” of problems and consist of measures to mitigate stress symptoms after their occurrence. Counseling and employee assistance programs are examples of those interventions ( Cooper et al., 2001). Like the results of other studies, our findings are not without limitations. First, the study relies on survey data, which suffer from restrictions associated with the cross-sectional survey method. Second, we used a newly developed construct for measuring the application of the controllability principle. Even though we drew on existing management accounting literature and followed state-of-the-art guidelines in developing the construct, we recommend that future studies establish the construct's usefulness and validity in a broader context. Third, to assess the application of the controllability principle in management control systems, our study investigates managers’ subjective perceptions of the controllability of their performance measures. It does not address the specific management accounting practices such as cost allocation and transfer pricing methods that lead to forming a certain perception. Research that relates the “objective” reality of management accounting practices to perceptions of managers represents another promising avenue for future research. Fourth, although theory clearly predicts a negative relationship between role conflict and managerial performance, we could not find a significant effect. Other studies that simultaneously analyzed the effects of role ambiguity and role conflict on managerial performance had similar results. Future research should re-investigate this important link by using alternative measures for managerial performance (see Marginson and Bui, 2009, for use of a different proxy) and role conflict (see Siegall, 2000, for a critique of traditional role ambiguity and role conflict constructs). Fifth, despite the use of role theory to guide the development of our structural equation model, we cannot demonstrate cause-and-effect relations empirically. Future tests of the proposed hypotheses in an experimental setting would be beneficial to further validate our findings. Lastly, we focused on two aspects: the dysfunctional effects of not applying the controllability principle (in terms of a decrease in managers’ individual performance) and the influence of one specific moderator variable—hierarchical level—on this relationship. Future research should investigate other moderating variables and, more importantly, the potential organizational advantages of non-application of the controllability principle. Notwithstanding the aforementioned limitations, we believe our study contributes to management accounting theory and practice. Our empirical analysis enhances the understanding of the effects of applying or not applying the controllability principle at different hierarchical levels. The results of this large-scale empirical study should make companies aware of the importance of dealing more effectively with behavioral effects that are caused by accountability arrangements such as application of the controllability principle, particularly at lower and middle managerial levels. Overall, understanding managers’ responses to application of the controllability principle is highly relevant for any organization, as “significant problems can arise if uncontrollables are not dealt with properly” (Merchant, 1998, p. 583).