سیستم کنترل در شرکت های چند کسب و کاره :: از مدیریت عملکرد به مدیریت استراتژیک

This article discusses the role of control systems in multibusiness companies. The focus is on formulation and implementation of corporate and business unit strategies. Three widely used categories of control models are discussed: (1) models for performance management, (2) models for value-based management, and (3) models for strategic management. The discussion is based upon central normative texts and examples from applications in Nordic companies. The description and discussion of the control models and their features should facilitate decision-making on the design and use of control systems in multibusiness companies.

The primary mission of management is to formulate and implement value-creating strategies. In a multibusiness company, with its many complex activities, that task is especially demanding, as was shown by several European firms that were compelled to change abruptly and drastically during the 1990s — among them BMW, British Aerospace and Ericsson. At these companies, corporate management was forced to downsize money-losing business units (BUs) within their core business. Low stock valuations have made other multibusiness companies prey for corporate raiders, resulting in divestitures and split-ups. Even economically successful multibusiness companies are now questioned by analysts and debaters in a way which formerly was unthinkable. One explanation is that many multibusiness companies find it hard to adjust to the demands of globalization and increasing competition. According to Goold et al. (1994), many problems of adjustment are due to internal routines and processes. In particular, companies seem to find it difficult to formulate a value-creating corporate strategy and to co-ordinate their BUs so that the strategy is implemented and synergies are exploited. A multibusiness company and its corporate management thus face the challenge of creating a control system which aligns the capabilities of the BUs with the competitive marketplace as well as the corporate strategy. Although the problems of many multibusiness companies have contributed to a renewed interest in the role of control systems in the formulation and implementation of strategies, these issues have been under discussion for a long time. One of the pioneers in this area is Robert N. Anthony; the framework which he presented in 1965 is still central today in defining the role of a control system (Anthony, 1965). Under the heading of ‘Management Control’, Anthony discussed the use of financially based models, with measures like Return on Investment (ROI) and Residual Income (RI), in establishing systems of planning and follow-up. In practice, however, it has proven difficult to use these models to create control systems for multibusiness management which further the long-term objectives of owners and the strategies chosen by corporate management. Critics have held that these models focus excessively on short-term financial performance and that for this reason management does not always act in accordance with the long-term interests of the owners. Researchers with backgrounds in financial economics have reacted by elaborating on the original models and re-launching them under names like Economic Value Added (EVA)1 and Cash Flow Return on Investment (CFROI). Any ‘agent problems’ are supposed to be handled by rewarding management through the use of so-called value-based models, which their proponents claim will accurately reflect the creation of shareholder value. In parallel with this development, a growing number of scholars are becoming highly skeptical of all types of financially oriented control models. According to these critics, the focus on financial measures provides a one-sided picture of corporate operations, making effective co-ordination difficult (Johnson and Kaplan, 1987). Models such as the Balanced Scorecard and the Performance Pyramid were developed partly to remedy this shortcoming by providing better co-ordination of corporate business through a combination of financial and non-financial measures. These measures, it is emphasized, are to be clearly linked to the corporate and business strategies of the firm and should thus be a valuable aid in its strategic management (Olve et al., 1999). The presentation will be based on the following three categories of control models for multibusiness management: (1) models for performance management, (2) models for value-based management, and (3) models for strategic management. All were advanced by scholars and practitioners as possible solutions to problems of co-ordination in complex organizations with multiple business units. Empirical studies have also shown that models of this kind have gained wide acceptance at multibusiness companies (Kald and Nilsson, 2000). The purpose of this article is to analyze the fundamental similarities and differences among the three categories of control models and the possible implications for multibusiness management.

The three models of control reflect different conceptions of the relationship between corporate and BU management in a multibusiness company. All three are intended to promote agreement on goals and strategies through a dialogue on preferences and possibilities. All three acknowledge, mobilize and exploit local knowledge in pursuing organizational goals and strategies. How then should companies decide which model to choose? The question is particularly relevant since the three models have been presented as universally applicable. Proponents of value-based management regard it as more advanced than performance management. Others recommend Balanced Scorecards and similar models, claiming that purely financial controls will be misleading. Yet in many boardrooms there must be a preference for controls with a link to external financial reports, i.e. what we call performance management, as best reflecting market expectations. Considering the discussion in the preceding sections, we believe that there will be a general tendency towards the situation depicted in Figure 4. Here, two dimensions are in focus: corporate strategy and organizational level. For different corporate strategies and organizational levels, the figure shows which of the three types of control model can be expected to be most successful. In this context, ‘successful’ means that the control model has a dominant influence on multibusiness management but that elements of other control models may be present. For example, since performance management has long been a well-established method of planning and follow-up, it may be expected to have a certain influence regardless of the corporate strategy or organizational level under discussion. In light of this statement, and of the interplay with other types of informal controls, Figure 4 must be considered a simplification that only describes a number of tendencies. We do believe, however, that it provides a good starting point for companies seeking to analyse and develop their control structure. Our first conclusion is indicated by the corporate strategy axis at the top of Figure 4. Essentially, the better the potential for activity sharing and the realization of synergies (cf. Porter, 1987), the more convincing the case for strategic management. Since activity sharing requires that corporate management be highly involved in the planning and co-ordination of individual BUs, a rich flow of operationally relevant information is necessary (Goold et al., 1994). Conversely, if corporate management want to view their company as a portfolio of separate BUs for which they provide a market-like discipline, it makes sense to focus more strongly on performance management. This will save corporate management from having to develop a thorough knowledge of the possibilities and strategic preferences associated, for instance, with balanced scorecards. With portfolio management, corporate management can be expected to decentralize a major portion of decision-making on business strategy to business unit management. However, the financial follow-up of plans based on these decisions will be extremely comprehensive (cf. Nilsson, 2001). Our second conclusion is that the suitability of the three control models is also affected by the organizational level at which planning and follow-up will take place (cf. Figure 4). This means that with portfolio management a change from performance management to value-based management can be expected at or below the BU level. There may also be a direct transition to strategic management. The reason is that the financial-control model, with its short-term focus, is suitable for the relationship between corporate management and BU management but is not appropriate at lower levels of the organization; the control model needed here is more focused than performance management on the business logic of the unit and should provide links between goals, strategies and operations. Furthermore, value-based management and strategic management are generally considered more useful in linking individual incentives to strategy for managers and employees at fairly low organizational levels. To sum up, here are some implications: • Appropriate models: Multibusiness companies considering the implementation of EVA, balanced scorecards or similar models should identify the level(s) of the organization for which they will be suitable, depending on the corporate strategy pursued. • Corporate knowledge: Different corporate strategies call for different knowledge about the possibilities available. To the right in Figure 4, more extensive knowledge is required since specific strategic risks at the corporate level may be involved; in addition, goals and expectations will represent more specific preferences. At the extreme left, on the other hand, preferences may only reflect current capital-market conditions. Similarly, to the left corporate management may judge only the financial attractiveness of different businesses, while to the right they will usually want to understand the specific business model. Both extremes obviously have their advantages and problems. • Strategic dialogues: Regardless of the control model chosen, communication between corporate management and BU management is intended to achieve agreement on ambitions for the future. We expect a more extensive exchange of information to the right in Figure 4, where we will also tend to find a narrower range of BU industries, more long-term service with the company, and company careers where a deeper understanding of business conditions and models is rewarded. • Information systems: As previously discussed, the infrastructure needed at the right of the diagram will normally be much more extensive. Perhaps the resurgence of interest in company synergies in the 1990s is partly attributable to the availability of new tools, such as data-warehouse software? Finally, we believe that the preceding discussion of the three control models, and the implications noted above, will facilitate decision-making on the design and use of control systems in multibusiness companies. Unlike the proponents of particular control models, we do not subscribe to any single best model for multibusiness management. The essential point is to choose the model or models that are appropriate in view of both the corporate strategy and the information requirements at lower organizational levels. Sometimes more than one control model will be used within the same company; in such cases, considerable attention must be paid to communication between the models. Studies of Swedish companies listed on the stock exchange show that firms successful at this kind of situational adaptation have also performed well in creating shareholder value (Nilsson, 2000). To do so, however, calls for knowledge and understanding of the properties of the various control models and of the situations in which each model is appropriate. This paper is intended to provide a framework for that purpose.