Management controls, role stress, and retail store manager performance.

Although it is commonly assumed that informal and formal management controls are linked to retail manager performance, virtually no empirical work has examined this fundamental control relationship. To investigate this linkage, we rely upon a survey of retail store managers and superior ratings of store manager performance. Interestingly, none of the three control mechanisms investigated in this research have a direct effect on manager performance, however, the findings indicate that the two informal controls (i.e., self and social) have an indirect effect on manager performance through their effects on role stress. In contrast, the formal control mechanism (i. e., output control) has no indirect effect. Finally, role stress is directly linked to store manager performance. Managerial implications and directions for future research in retailing are discussed.

INTRODUCTION

Retail control has been a topic of continuing concern (Lusch 1990) for retail executives as they attempt to keep their organizations on the path of achieving high performance results.

The job of keeping a retail store on its path toward targeted performance presents special control problems. Largely this is because in retail organizations no measurable output is created until a transaction with a customer occurs - and the customer resides in the external environment, which is largely uncontrollable. Also retail control is difficult because there is not a well-defined production function where a given set of known inputs can be expected to yield predictable output. We believe because of these special control problems retailers can benefit from better understanding the management control literature, especially the literature that addresses formal and informal controls.

From the earliest writings on management control systems, researchers have assumed that there is a direct link between the management control system-in-use and managerial performance (see Giglioni and Bedian 1974 review). Surprisingly, however, a literature review identified only a handful of empirical studies that investigated this relationship (see Merchant 1981; Futrell, Swan and Todd 1976). Although these studies purported to establish a causal relationship, problems in measure development (i.e., Merchant 1981; Futrell et al. 1976) and exclusive use of self-report data (i.e., Merchant 1981) limit their validity and generalizability.

The purpose of this paper is to examine the direct and mediating links of management controls on managerial performance in retail organizations. In this regard, the focus of this paper is on managers' perceptions of management controls. This paper not only sheds light on the proper types of controls to use in retail organizations, but also overcomes three significant problems associated with previous research. First, the study uses two mutually exclusive, independent data bases. Extent of reliance on various controls and role stress data were collected from the store manager and ratings of store manager performance were collected from his/her immediate superior. This overcomes the problem of exclusively relying upon self-report performance data (e.g., Merchant 1981). Second, previous work was not as sensitive to the measurement properties of key constructs. Although Merchant (1981) departs from accounting convention by using several multi-item scales, many of his scales are single-item measures. Also, while Futrell et al. (1976) uses multi-item scales, it is questionable whether their measures describe the control system-in-use or consequences of various controls.

Third, the study extends control system work by focusing on both formal and informal controls. As Jaworski (1988) notes, virtually all research to date has examined the effects of a single type of control. However, if one subscribes to the position that both formal and informal controls are always present in organizations (see Anthony 1952; Dalton 1971; Hopwood 1974) but that they vary in their degree and quality of use, then it follows that research should focus on both formal and informal controls.

In order to develop these ideas in a systematic fashion, the paper has been organized in five sections. In the first section we introduce the types of controls-in-use in retail settings. This discussion blends research on both informal and formal controls in arguing that several control types can be expected in retail settings. In the second section, we discuss the theory and associated research hypotheses to be tested. In the third section, we discuss the research design including data collection and measurement. The final two sections present the results and discussion.

TYPES OF CONTROLS-IN-USE IN RETAIL SETTINGS

In order to understand the system of control that has evolved in retail settings, it is first necessary to consider what management desires and expects from retail store managers. Although retail performance evaluations may consider the "people" talents of managers and their leadership abilities (see Davidson, Sweeney and Stampfl 1988), management tends to have an overriding concern with bottom-line performance measures (Lusch and Harvey 1983; McCammon and Hammer 1974). Indeed, much of the jargon in the retail area reflects this output-oriented concern. Terms such as sales-per-square-foot, sales-per-dollar of inventory invested, sales-peremployee-hour, and total daily sales are not atypical for performance reports (Lusch and Dunne 1990). Thus, output control-defined as the setting of performance standards, monitoring, and standard evaluation - is widespread in retail settings (see Table I for definitions). In this research we address the first phase of output control - the feedback and corrective step that completes the control cycle - and thus our reliance on the perceptions of output control seems appropriate.

An overall "output-orientation" is not surprising. To use Ouchi's (1979) management control system framework, we can expect management to use more results-oriented measures as it becomes easier to identify desirable output indicators (see Ouchi and Maguire 1975; Rockness and Shields 1984 for empirical support). In a retail sales context, such results or output indicators are easy to identify and used widely in practice. Key output indicators include dollar sales volume, units sold, and gross profit (Ingene and Lusch 1981). This is not to say that these are the only indicators or even the highest quality indicators. Rather they represent industry agreed upon indicators of output or results for the store manager. Relatedly, using commonly agreed upon output measures also facilitates the comparability of store managers across retail settings. Hence, promotions and raises can be tied directly to company-wide competition. Indeed, the corporate office often reminds store managers where they rate relative to other stores by disseminating company-wide reports (Lusch and Dunne, 1990). Nonetheless, perhaps the most important feature of output control is the feedback that is given to managers when performance is poor or inadequate.

Although this drive for results characterizes the retail store managers' evaluation system, it does not provide any indication of formal and/or informal resources that may be accessed by store managers to attain those goals. Here is where the retail store controls-in-use departs from conventional theory. Although Ouchi's (1979) contingency framework acknowledges the presence of multiple controls, it is restrictive since it focuses attention on "one-best" or "one-dominant" control for various organizational contexts. For example, he would expect "output controls" to be dominant when measurement of output is feasible, behavioral controls" to be dominant when the means to achieve results is well known, and "clan controls" to be in use under conditions of low measurability/low process …