Operation of management control practices as a package—A case study on control system variety in a growth firm context

Management Accounting Research 19 (2008) 324–343

Operation of management control practices as a package—A case study on control system variety in a growth firm context Mikko Sandelin ∗ Helsinki School of Economics, Department of Accounting and Finance, P.O. Box 1210, FIN-00101 Helsinki, Finland

Abstract This empirical case study examines the operation of management control practices as a package in a growth firm context by paying particular attention to the couplings among cultural, personnel, action and results controls. The analysis focuses on two different management control packages in the face of similar contingencies at different points of time. The paper argues that the functionality of a control package depends on internal consistency, specifically on the reciprocal linkages of design and use between a primary mode of control and other control elements. Moreover, it argues that control package variety is driven by the way in which the management responds to functional demands. Two different control packages are considered equifinal to the extent of limited operational complexity, whereas an accounting-centric control package is also sufficient in the face of increasing levels of operational complexity. © 2008 Elsevier Ltd. All rights reserved. Keywords: Management control package; Control system variety; Internal consistency; Functional demands; Equifinality; Growth firm

1. Introduction This empirical case study examines the operation of management control practices as a package in a growth firm context. While it has been acknowledged that ‘soft’ and informal modes of control typically characterize small firms (e.g. Bruns and Waterhouse, 1975; Merchant, 1981; Flamholtz, 1983; Chenhall, 2003; Merchant and Van der Stede, 2007), an increasing body of literature now suggests that growth renders management control systems (MCS), especially management accounting systems (MAS), more formal (Granlund and Taipaleenmäki, 2005; Moores and Yuen, 2001) and that formal MAS facilitate the growth of a firm (Davila and Foster, 2005; Sandino, 2007; Greiner, 1998). In the growth firm context, the more comprehensive management control package has been investigated to a limited extent. Lukka and Granlund (2003) have paid particular attention to organizational culture and Collier (2005) has focused on socialization practices as primary control mechanisms, whereas other studies have examined MAS and formal standard operating procedures (Granlund and Taipaleenmäki, 2005; Davila and Foster, 2005; Sandino, 2007). The main thrust here is that an equally good final state can be achieved by various control system designs in the face of similar contingencies (Huikku, 2007; Merchant and Otley, 2007; Ferreira and Otley, 2005; Gerdin, 2005; Spekle, 2001; Otley, 1999; Chapman, 1997; Fisher, 1995). The need for coordination and control can be met by several alternative ∗

Tel.: +358 9 4313 8457; fax: +358 9 4313 8678. E-mail address: [email protected].

1044-5005/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.mar.2008.08.002

M. Sandelin / Management Accounting Research 19 (2008) 324–343

325

management control system designs (Gerdin, 2005). Contingency theory, as it builds on the assumption that variables are related to each other in a one-to-one manner, seeks optimal control system designs in specific circumstances at the cost of system variety (Spekle, 2001; Gerdin, 2005; Merchant and Van der Stede, 2006). Moreover, a long-held view in organizational design literature suggests that multiple means of control do not only complement each other but may also operate as substitutes (Galbraith, 1973; Mintzberg, 1983; Fisher, 1995; Abernethy and Chua, 1996; Ferreira and Otley, 2005; Huikku, 2007). The potential for achieving the same final state by various configurations of control elements and systems in the face of similar contingencies is referred to as equifinality (Doty et al., 1993; Gresov and Drazin, 1997). In other words, organization designers have the latitude to decide about how to achieve organizational goals. This notion of equifinality has received only marginal attention in empirical management accounting and control studies. Two recent studies, however, suggest that equal control of activities can be achieved either by an informal control practice (Huikku, 2007) or by different formal control systems (Gerdin, 2005). This study goes further and adopts a holistic approach to management control and seeks to increase our understanding of the simultaneous operation of multiple control practices at the firm level and addresses the different, potentially equifinal, control configurations that they form. The motivation for adopting a holistic perspective on management control stems from the fact that empirical management control studies have produced unclear findings and conflicting results because too few components have been studied at the cost of more comprehensive and integrative approaches (e.g. Merchant and Otley, 2007; Ferreira and Otley, 2005; Covaleski et al., 2003; Otley, 1999). It is not uncommon for management control studies to focus narrowly on formal systems, and specifically on MAS alone. Frequently, they also implicitly rest on cybernetic tradition and assume that management control elements operate as a system of planning, measurement, evaluation, and feedback for corrective actions. In this tradition, it has been argued that the linkages between formal MCS components, particularly the strength and coherence of the couplings, explain the functionality of a control system (Ferreira and Otley, 2005). However, our knowledge about the interplay between accounting-based and more subjectively constructed control elements is in its infancy (Brown, 2005), although it is increasingly acknowledged that formal MAS are only part of a broader set of organizational controls (Flamholtz, 1983; Flamholtz et al., 1985; Abernethy and Chua, 1996; Alvesson and Kärreman, 2004; Collier, 2005; Merchant and Van der Stede, 2007; Merchant and Otley, 2007). Formal control systems, including MAS, are important mechanisms for reducing uncertainty, although they may have a modest effect in controlling actual work (Kärreman et al., 2002). On the other hand, formal control practices may significantly shape social order and provide organizational culture with material significance and meaning (Alvesson and Kärreman, 2004; Ahrens and Mollina, 2007). While the control elements may play multiple roles, subtle linkages seem to make them operate as a ‘package’ (Abernethy and Chua, 1996). To date, however, the couplings between control elements have been exposed to limited academic scrutiny (Brown, 2005; Brown et al., 2008). An examination of the couplings between MAS and other control elements therefore has the potential to contribute to our understanding of the features explaining the functionality of control packages, but also of the need for formal control practices such as MAS. Many alternative conceptual frames are available for studying control elements as a package. Those by Otley (1999) and Ferreira and Otley (2005) are grounded in the systems view of organizations and are intended for examining combinations of formal control techniques, or at least of practices mediating information, thus ignoring more subtle ways of motivating and coordinating organizational behaviors. Also, Simons’ (1995) framework focuses on formal, information-based control practices and is primarily applicable to investigation of how managers use rather than design MCS in an attempt to control strategy. Moreover, the organizational control framework of Flamholtz (1983, 2005), Flamholtz et al. (1985) seems to lack specificity although it is informative about the need to align organizational control elements. This study therefore draws primarily on Merchant’s (1998), Merchant and Van der Stede (2007) object-ofcontrol framework because it focuses on the spectrum of control practices, but still provides sufficient rigidity with the specific control objects of culture, personnel, action, and results. Adopting a broad perspective on management control implies that organizational culture and structure are means rather than mere premises of control (cf. Flamholtz, 1983; Flamholtz et al., 1985; Merchant and Van der Stede, 2007). While the mere existence of any organizational culture and/or structure affects organizational behaviors, they are also objects of deliberate managerial actions. They are designed and changed to affect the behavior of organizational members in the attainment of organizational strategies and goals, irrespective of whether these strategies are ex ante planned or emergent (Mintzberg et al., 1998). Hence, the distinction between culture and cultural control is that the

326

M. Sandelin / Management Accounting Research 19 (2008) 324–343

latter considers the means of shaping culture1 and the former the outcome that affects organizational behaviors. In addition, management control may include numerical measurement techniques, such as MAS, but also practices that are based on subjective judgments and evaluation (e.g. Alvesson and Kärreman, 2004; Collier, 2005; Ahrens and Mollina, 2007). While Merchant’s object-of-control framework captures the richness of control practices, it lacks specificity in explaining the couplings between control elements. As this study specifically seeks to examine these couplings, it only mobilizes Merchant’s framework as a medium for analysis. The remainder of the paper is structured as follows. The next section describes the research site and method. The following section analyses empirical evidence by presenting two illustrative case examples of the management control package. These cases are grounded on the same firm at different periods of time. Each case description is followed by a short analysis of the couplings between control elements. The subsequent section discusses empirical findings on the basis of comparative case analysis and the final section draws the conclusions. 2. Research method and site There is little prior theory of the linkages between formal and informal control elements. As the research objective is explorative in this sense, it favors the case study method at the cost of the less in-depth methods of survey studies. On the other hand, findings about less orthodox forms of management accounting and control in the relatively small and entrepreneurial firms (e.g. Perren and Grant, 2000; Greenhalgh, 2000; Collier, 2005) encourage adoption of a sensitive, deep-probing research method that also enables the tracing of subjectively constructed forms of control. Thus, a case study appears to be the most promising mode of enquiry because it makes possible a comprehensive approach to the study of controls in use (Otley and Berry, 1994; Ahrens and Dent, 1998). However, as this study examines two distinct periods of firm evolution, it relies on a comparative logic of analysis, which is likely to reveal subtle similarities and differences between cases and lead to more sophisticated understanding—to categories and concepts unanticipated by the investigator (Eisenhardt, 1989). Thus, this mode of analysis has the potential to refine management control theory about the functionality and appropriateness of management control packages (Lukka, 2005; Keating, 1995; Scapens, 1990). Theme interviews with organizational participants and stakeholders are the primary data source. Data triangulation was sought throughout the data collection process. On the one hand, people were interviewed from various organizational levels and functions, and on the other, interviewees consisted of current employees and those who had already left the company. Altogether, 22 semi-structured interviews were conducted among the key informants before the data became saturated. The data amounted to 34 h of tape-recorded interviews that were subsequently transcribed. While individuals tend to impose order retrospectively on phenomena, an attempt was made to eliminate this type of bias by letting informants discuss freely, by seeking facts rather than opinions in retrospect, and by reflecting interviews on archival data2 . Moreover, confirmation on individual ‘accounts’ was sought by posing similar questions to multiple informants (cf. Cardinal et al., 2004; Miller et al., 1997; Abernethy and Chua, 1996). Only the general aspects of the study design were determined in advance and data collection was undertaken prior to the final definition of research question because of the relatively unknown research phenomena (Yin, 1993). Hence, the data analysis was an iterative process between theory and data, in which patterns emerged and theoretical insights crystallized after several readings, particularly around three distinct periods along the evolution of the firm. At the ends of the continuum control practices appeared as the most promising ones in the theoretical sense because the contingencies were similar but the designs of control packages varied to a large extent. In explaining this variation, the study relies on the interpretive mode of theorizing (Lukka, 2005). The research site of the study, Mobitel Ltd.3 , was chosen because it was an entrepreneurial firm seeking growth on immature ‘high-technology’ markets. Inherent ambiguity and uncertainty of this type of firm (Granlund and Taipaleenmäki, 2005), as well as growth, were expected to characterize management control dynamic phenomena.

1 Culture can be shaped e.g. by codes of conduct, group-based rewards, intra-organizational transfers, physical arrangements and tone at the top (Merchant and Van der Stede, 2007). 2 Archival data included IPO prospectus, investor presentations, interim and annual reports, second hand interview material in public press and the firm’s internal documentation and presentations. 3 The name is pseudonym in order to secure anonymity of the case firm.

M. Sandelin / Management Accounting Research 19 (2008) 324–343

327

On the other hand, the relatively small size, although the firm was growing, was expected to make the linkages between various forms of control more visible and pronounced than in a large and complex organization in which multiple levels of management hierarchy and bureaucratic processes tend to hide informal practices (Mitchell and Reid, 2000; Collier, 2005). The major events, contingencies, and performance data during the firm evolution are presented in the appendix. 3. Case study 3.1. Control practices as an enabling platform—Case 1 The firm was established at the beginning of 1998 in the merger of three entrepreneurial firms. At the same time, a new CEO with a managerial background in a large telecommunications company was hired and the CEOs of the merged firms took positions in the senior management. The firm had a steady cash-flow business from Internet access services. As the firm grew, it hired a CFO at the beginning of 1999. Before the entry of the CFO, who had earlier worked for years in a profitable family business and subsequently as a management consultant, an external agency took care of bookkeeping. The firm has been venture-capital backed since its foundation and it has sought to compete with the lowest price. 3.1.1. Cultural controls and culture From the beginning, the personnel were highly committed to the firm, as each employee owned a relatively significant share of the firm. The interests of the organizational members and the venture capitalist coincided—each party wanted the firm to increase the shareholder value. The venture capitalist described the managerial freedom as follows: “We were not so interested in knowing what was going on within the firm. We were interested in entry and exit prices.” In fact, the senior management had freedom to decide how to develop the firm. The choice was made for growth instead of profitability, as a senior manager commented: “Growth was our priority but we also discussed profitability.” As the firm grew and it hired new managers, they were made shareholders. New employees at the grass root level had a base salary, but they were also committed to the long-term development of the firm with share options. By means of broad ownership and group rewards, the senior management maintained an ‘entrepreneurial’ culture. The personnel were extremely committed to the firm. There was no real need to monitor working time. “Personnel were willing to work even seven days per week and they were ready to cancel holidays if any business challenges occurred,” a line manager commented. The ownership-based entrepreneurial culture provided solid ground for running and developing the business because “everybody tried to seek the best for the firm in order to maximize the value of their personal ownership,” as one business unit manager stated. Shared values and beliefs culminated in technological advancement which guided managerial behavior, as another business unit manager recalled: “We made decisions on the basis of technology. Once something was technologically possible, we developed a product rapidly and made it available to customers [in the portal]. We did not think in advance whether there was demand for such a product.” Technological opportunities dominated managerial thinking. Financial and demand concerns were secondary, as the Chief Technical Officer (CTO) explained: “We developed and built things [products] which were extremely difficult to figure out how they could have generated cash [in-] flows.” Managerial actions focused primarily on discovering new applications for existing technologies and developing new technological solutions. On some occasions, however, the management planned technology development. In these exceptional cases, development efforts were driven by economic concerns in contrast to more general patterns of simply thinking and acting in terms of technological possibilities. The CFO illustrated a planned project:

328

M. Sandelin / Management Accounting Research 19 (2008) 324–343

“The propeller-heads [entrepreneurial managers] had an idea to develop a ‘service of all time’ with which customers could download e-mails onto mobile phone. There was to be a monthly charge for the service or a usage-based charge. [. . .] They developed ‘sms-gateway’ technology on that basis.” Once the e-mail service was ready, only some pioneering customers ordered it. The service was too sophisticated at the time, despite of the weak demand signals detected by the firm. The sms-gateway technology seemed to be merely a costly project without commercial use. However, the information engineering staff had drawn logos and transmitted them via sms-gateway technology to their mobile phones during the technology development process. A line manager noticed these logos and presented them to the senior managers. Soon the logos were available in the portal at a price of less than one euro per download. The CFO commented on the subsequent events as follows: “We were too slow to advertise it. Suddenly, without any advertisement expenses, rumor about the logos spread among customers and we sold almost a million logos in a month.” The demand prospects for e-mail service and related ex ante planning of activities had little significance in guiding ultimate behavior and financial success. Instead, keen interest in technologies across the organization facilitated the emergence of logos. And trying out logos as a commercial product was driven by the freedom of action that originated from the ownership. The culture guided action throughout the organization. In addition, action itself was a characteristic of the culture, as a senior manager described: “We were doers rather than thinkers. [. . .] Some people are good at planning and making theory, but it does not help too much because they are unable to arrive at any conclusion. I mean that it is better to have little theory and to get things done than get nothing done.” Action orientation was also reflected in how the management decided to capitalize upon the logo business. The CEO explained the early steps in taking the logo innovation to the international markets at the end of 1999: “Instead of planning, we just flew over to check in which countries firms were technologically ready, where there was willingness, and where good guys were ready to proceed rapidly.” Formal plans, budgets, and standard operating procedures were replaced by rapid and intuitive actions. By taking bold action, the senior management acted as role models for line managers and employees. They shared responsibility and provided support to each other. All of this was based on the ownership culture, as the CEO explained: “Flexibility came from the ownership. Once you had own money in question, you were ready to be flexible. Our backbone was like spaghetti bent to meet the situation at hand.” Although organizational culture was the primary mode of control and it was deliberately shaped by means of ownership, incentives and the ‘role model’-like behavior of senior managers, the senior management also employed other control practices in an attempt to facilitate the profitable growth of the firm. 3.1.2. Control practices complementing cultural control orientation Personnel controls were deployed from the beginning because growth constantly required more staff. As the firm lacked a human resource manager, the senior managers or the heads of the support functions recruited new employees. A senior manager commented on the evaluation of the candidates as follows: “People were mainly hired but first they needed to be whipped and kicked a little, just like in a gang, as you’ve seen in Italian mafia films. That was the school. It did not matter whether an employee had an academic degree or not.” The CEO further clarified the criteria applied in the personnel selection: “I learned quite rapidly that track record and work experience were not appropriate criteria because those people were not able to adapt to our chaotic culture.” The managers evaluated the candidates on the basis of subjective judgment. Although technological skills and knowledge were sought, the ultimate evaluation was conducted against the relatively chaotic organizational atmosphere where continuity and stability were replaced by bold and intuitive actions. “The ‘old economy gurus’ were unable to

M. Sandelin / Management Accounting Research 19 (2008) 324–343

329

adapt to these settings—perhaps the fluidity stressed them,” a senior manager commented. Moreover, the management did not hesitate to fire an employee if s/he did not fit to the organizational culture. There was not, however, a need for significant lay-offs. The systematic though informal deployment of cultural criteria in personnel selection and share-based incentives granted to employees homogenized the atmosphere within the firm. A business unit manager described the personnel as follows: “The most positive character of our personnel was the willingness to work crazily. They had intrinsic motivation to develop new things [technologies and products]. [. . .] This was the first challenge for many of our employees. However, you were able to make them to exceed themselves time after time.” Although the average age of the staff was about 27 years, the management considered it a strength rather than a weakness. The young personnel had no reference points about how matters should be organized in a firm. Thus, they were apt to adapt to high velocity. The culture-driven mode of personnel control, subjective evaluation, was also mobilized in assigning tasks on the basis of job performance, as a senior manager commented: “In practice they ultimately showed what they were capable of. [. . .] Of course we did not assign overall responsibilities to a guy who was better at details.” The relatively young and technologically motivated personnel formed a coherent resource for the firm. However, the open-plan office further encouraged culture-driven behavior. Besides efficient information dissemination between organizational members, this particular workplace design facilitated the adoption of cultural values by providing a natural platform for socializing with other employees. On the other hand, the open-plan office induced peer-control, as individual actions were transparent to others. Work was conducted in teams by which the management sought to enhance the sense of individual accountability to other members of the team. In the absence of team rewards, the accountability within teams was quite modest. However, motivation and especially commitment were not particularly problematic, as the CTO described: “Even though the content of tasks changed significantly, the employee turnover was low.” The share-based modes of compensation and the challenging content of tasks provided sufficient incentives for the personnel. Action controls operated upon the open-plan office. The firm had been divided into four business units according to customers and services. Each unit had a responsible manager authorized to make operative decisions. In these units centralized decision-making was sought by means of standard operating procedures, such as approval policies and pre-action reviews. These modes of control were informal, conducted as part of daily discussions in the open-plan office. Moreover, the most important form of action control was personal supervision. The management, by ‘walking the talk’, assigned tasks and provided decision-aid to the teams, as a line manager described: “We had quite an army of hackers. . . You only needed to say to them ‘find out if this or that makes any sense.’ Then these guys worked on it for half a day or a day. Then they said ‘this makes sense and it takes a week to copy it.’ After a short discussion I said, ‘ok, copy it.’ Then we [managers] started to build up a business plan on it.” Operative planning and scheduling of activities prior to action were untypical for the firm. Neither were specific job instructions prepared. These control practices were seen as secondary because events unfolded at a rapid pace inside and outside the firm. The operative management let employees to take the initiative and coached them about how to proceed as new information was available. The line manager continued: “We had 20–30 projects going on . . . Many of them proceeded so that we had worked on a project for about three weeks until we suddenly observed a better solution. We terminated the project and decided to copy an alternative solution.” Line managers challenged teams to consider technological advancement, but placed less emphasis on the financial consequences related to the new ideas. Business planning on technological ideas was a managerial responsibility. Financial evaluation of the new ideas and their fit with other businesses was conducted in discussions between the line managers and the business unit management. Consequently, technological and financial criteria for appropriate action were separated from each other at the operative level. On the other hand, while the senior management sought profitable growth, it considered new product development to be crucial for the firm’s success. This critical function

330

M. Sandelin / Management Accounting Research 19 (2008) 324–343

was controlled, beyond personnel and cultural modes of control, by ensuring that there were enough employees and time available for developing technologies. The CFO commented on his perception as follows: “Somewhat similar issues were considered in multiple teams and that’s why we had so many employees. [. . .] Never had it [the number of employees] been considered primarily from the efficiency perspective.” Besides new product and technology development, the actions related to business processes of a routine type were controlled by more formal means and automated whenever possible. However, irrespective of how formal action controls were, managers conducted subjective performance evaluation and provided constructive feedback, occasionally rewarding exceptionally good performance. On the other hand, at the top of the firm, the management group formed an important body for coordinating and controlling actions, as the CEO described: “We shared and updated information regularly on each Monday morning. We had management group meetings for two hours and we discussed what was going on.” The weekly management group meeting consisted of the business unit managers, the CEO, the CFO, and the heads of technology and operations. The action-reviews and ideas presented by the business unit managers were evaluated and discussed more broadly. Besides directly controlling business unit initiatives and actions, the management group had another control role. The CFO described the management group as follows: “The size and composition of the management group varied constantly.” A senior manager further clarified the rationale underlying the dynamic management group: “. . . in a larger scope [of the whole firm], one man would have not been able to control it. In this sense, the management model of constantly changing the composition of the management group was reasonable.” To obtain a comprehensive understanding of actions within the firm, the senior management invited the line managers to attend the meetings. Knowledge was thereby circulated within the firm. Due to possible personal limitations at the lower organizational levels, important pieces of information could have been missed despite the open-plan office. Results controls were introduced by the CFO. The operative non-financial indicators, such as website downloads, time spent on website, and spoken gsm and fixed line minutes per customer were monitored on a daily basis. These data, however, were mobilized primarily for learning purposes, as the CFO explained: “. . . we monitored the number of subscriptions, gross and net, in order to understand how to keep customers happy and loyal.” The frequent monitoring of non-financial results directed managerial attention to the quality of services and the viability of new products and thus, it provided decision support for terminating or revising a product concept. Nonfinancial indicators were not, however, standardized performance targets tied to compensation. Instead they were used in forecasting turnover. Nevertheless, the non-financial information was problematic in budgeting, as the CFO pointed out: “The general problem with them [non-financial indicators] was poor linkages with book keeping. It was damn difficult to figure out the financial implications and linkages.” It was unclear how the many products developed on the basis of technology push would generate revenues. The non-financial information of the product or service volumes (e.g. in the portal) was not directly associated with revenues and some products were based on termination fees that depended on changes in the telecommunication regulation. A particular challenge in budgeting was the assignment of overhead costs to product or service lines because the ‘true’ cost drivers were based on bytes (e.g. rented network capacity). However, through trial and error the CFO learned to approximate the ‘hype’ in budgeted sales and underestimates in variable costs and also to allocate overheads to business units at the aggregate level. However, the budgets had little significance in guiding or restricting behavior at the operational level, particularly in controlling costs, as the CFO explained: “Cost control, well . . . I had to take disciplinary action quite often, to yell loud enough. It worked for a while but then, I saw it so clearly, quite rapidly there were this and that, invoices not directly related to business . . . coffee machines, new computers, software licenses and upgrades, extremely high traveling expenses, etc.” The budgets brought financial issues to the attention of senior management only, to whom the CFO also reported invoicing data, the number of new customers, and the cash balance on a weekly basis. This information emphasized the liquidity of the firm and set boundaries for growth. Although the senior management was exposed to financial data

M. Sandelin / Management Accounting Research 19 (2008) 324–343

331

and the CFO coached them about financial planning, the budgets were perceived more as a means to shape culture than to control activities, as a senior manager described: “Budgets had two purposes: to nail down profit center [business unit] targets and hence, to enhance the internal spirit in terms of making the impossible possible.” The budgets provided direction by setting targets, but had little significance as control mechanism at the operative level because the degree of participation in budgeting was very low within the business units. Consequently, the line managers in the business units were loosely committed to the budgeted targets. However, budgets operated as a formal control system in the sales function in which financial performance targets were set for the salesmen. Their performance was evaluated on a monthly basis and the achievements were linked to the formal reward system—extra cash bonuses. The CEO summarized the significance of the formal control practices in managing the firm as follows: “My principle was not to rule out anything. [. . .] You never knew what’s going to happen . . . you just sat and waited. [. . .] We intentionally tried to avoid rules and formal procedures [in decision-making], but if they emerged anyway, we got rid of them.” Many transactions, e.g. establishing the logo business in Norway, had proven to be a success though the plans and calculations were not formal—they were only in the minds of the senior managers. These implicit plans were realized through bold and intuitive actions. However, the financial numbers were not totally insignificant—they were mobilized as rough guidelines and targets which fuelled growth aspirations, as the CTO recalled: “He [the CEO] said that we are not allowed to do things [businesses] from which the expected turnover is less than [x.x] million euros.” Financial controls were employed at the aggregate level as mediums of directing attention and motivating managers, whereas operational activities within the business units were controlled by mutually reinforcing forms of cultural, personnel, and action control. 3.2. Analysis of the Case 1 Scant financial resources limited the growth of the firm in the beginning. The lowest prices and innovative products, however, resulted in high volumes and cash flows enabling growth—the headcount grew from one to two hundred. Soon after establishment, an external board of directors took over, but controlled the firm loosely. The problem of developing and controlling the firm in the relatively complex environment was left to the discretion of senior management. The coherent organizational culture was largely based on the fact that ownership and control were not separated. The senior management created an ‘ownership culture’ (Merchant and Van der Stede, 2007) by means of direct share ownership and share options and thus sought to establish and commit employees to self-control. The culture of technological intra-entrepreneurship replaced specific task descriptions and comprehensive pre-action planning. The role model provided by senior management further shaped the organizational culture by emphasizing bold and intuitive action. In addition to enabling relatively autonomous behavior, the organizational culture guided search and development actions and provided management with criteria for selecting personnel. Personnel controls, however, also reinforced the culture, as employees with an intrinsic technological motivation and ability to adapt to ‘unstructured’ organizational settings were hired. The relationship between these modes of control was reciprocal, the culture, however, having primacy. The ‘hybrid’ controls, such as the open-plan office, complemented cultural, personnel, and action controls. In the open-plan office, daily operative actions were conducted in teams. These structures introduced action accountability, but they were essentially complemented by personal supervision. The need for personal supervision stemmed from the personnel selection and placement methods, but it also enabled the existence of these practices. In line with the organizational culture, the standard operating procedures, such as the approval policies, were part of the informal daily interaction. Moreover, the coordination of managerial activities relied heavily on personal supervision and mutual adjustment in discussions, which also took place in the senior management group. The cultural, personnel and action controls had reciprocal linkages, which made them harmonious to a large extent. These controls enabled technologydriven new product development, which was considered critical to profitable growth and thus, its performance was ensured and controlled by allowing several teams to conduct similar type of tasks.

332

M. Sandelin / Management Accounting Research 19 (2008) 324–343

Moreover, all except the results controls were based on broad ownership, which created an incentive for contributing to firm-wide performance. Financial results were accordingly measured and budgets prepared only at the aggregate level. On the other hand, while the firm primarily sought growth, it also monitored profitability at the senior management level. The financial concerns were not deliberately extended to the operative level, as the aim was to preserve innovativeness. To sum up, non-financial results controls were only loosely linked with action controls as they provided decision support for product portfolio decisions, whereas financial results were separated from other control elements in order to maintain the focus on technological advancement. However, the financial performance indicators (see appendix) do suggest that the firm was successful in attaining its goals with this specific management control package. 3.3. Prologue for Case 2 The firm went public in the late spring of 2000. With the IPO money, it continued to grow with increasing emphasis on international markets. By the end of 2000, it had established businesses in more than 10 European countries primarily by means of acquisitions that were mainly funded by share exchange. At the end of 2000, the domestic business was profitable; turnover was D 27 million, EBITDA was D 4.5 million, and the operating profit was D 1.7 million, although the firm had faced problems in taking over the acquired companies. As the events unfolded, it became evident that rapid growth in turnover was insufficient to recover takeover costs. The firm drifted into financial troubles in the late summer of 2001. The senior management was replaced by a crisis management team. The crisis-time CEO and CFO sought to keep the firm afloat by negotiating additional funding. Finally these efforts culminated in voluntary debt-restructuring. The crisis-time CEO and CFO left the firm as their primary task of securing its survival had been accomplished. A new CEO and CFO were hired, both with many years of experience in management consulting. The firm continued to offer products and services at the lowest prices. The cash-inflows were modest—the firm had lost reputation and customers during the crisis. In general, the organization was in a state of flux, as the CEO described: “. . . the organization was in a state of great uncertainty: from the perspective of management, responsibilities, and atmosphere.” In other words, at the outset the new CEO and CFO faced a firm resembling that was presented in the first case—there was little administrative infrastructure of any kind in place. The new management withdrew the debt-restructuring plan after a few months, as new venture capitalists invested in the firm. The following section focuses on the subsequent two-year period. 3.4. Control package as a system of accountability—Case 2 During the acute crisis, business development had been less than modest. Various ‘ad hoc’ activities had been undertaken in an attempt to secure financial survival. The biggest individual shareholder of the firm, the venture capitalist who occupied the board with other investors, described the expectations imposed on the new CEO as follows: “Prospects for growth were promising, although competition was intensifying. We did not want to let the firm go bankrupt because we had invested so much money in it. We just wanted it to continue in some ways.” Indeed, the venture capitalists did not set tight restrictions for the CEO. They were primarily interested in increasing the share price. Whether it would happen by means of growth or profitability or both was left to the CEO’s discretion. 3.4.1. Controlling by results The CEO and CFO, in co-operation with the business unit managers, analyzed the cash flows. While no financial calculations were readily available, they estimated cash flows at the aggregate level and perceived the internet and teleoperator products and services as the most promising in financial terms. The senior management perceived the core of the firm as an ‘invoicing machine’. Instead of emphasizing new product development, they focused on efficient delivery of current products that would secure a steady cash flow. Among the secondary businesses was the more innovative mobile entertainment, which formed a separate business unit with promising European-wide operations. The rest of the businesses formed the third unit, which the management wanted to divest or terminate as soon as possible in order to reduce uncertainty and thereby establish simple premises for systematic management, as the CEO remarked:

M. Sandelin / Management Accounting Research 19 (2008) 324–343

333

“The firm was organizationally unsteady [as the boundaries between businesses were fuzzy]. A clear separation of businesses from each other enabled some sort of management: assignment of managerial responsibility and personnel.” Clear boundaries between business units established a foundation for development of a new control system. Because of the CEO’s preference for numbers, the control system was to be based on accounting practices. The idea was to establish a broad financial control system by which financial performance targets would be extended not only to business unit managers, but also to operative managers. All the existing cost centers were discarded, new cost and profit centers were formed, responsible managers were appointed, and the accounting manual was fully renewed. While the accounting function developed financial control mechanisms, the CEO made decisions on the basis of rough financial estimates, as he pointed out: “It was a case by case evaluation. We couldn’t do it in relative terms, from the perspective of the whole firm. I mean we could not think that now we have achieved this much toward our overall target with this project. We did not have such a comprehensive frame. Decision-making was fact-based, but each case was evaluated separately.” The absence of a financial control system and decision uncertainty was not a reason for the CEO to abandon results-focused control. Once the accounting function had accomplished the financial reporting structure, the CEO immediately sought to allocate financial responsibility to the business units. The CFO issued budgeting instructions and the business units started to calculate their revenues and costs. The budgeted operating profit set performance targets for the businesses. The monthly performance evaluation, however, revealed that the budgets had been highly optimistic. The CEO recalled subsequent events as follows: “. . . it soon became evident that revenues were much less than forecasted. Having seen the [actual] numbers, I started dismissals in the business unit [Internet and teleoperator].” For the CEO, the variance between financial targets and actual results was the primary criteria for evaluating the businesses. Although the business unit managers explained unpleasant variances as merely a consequence of external surprises, the CEO regarded such explanations as secondary. Instead, he sought managerial accountability. If he considered a variance significant enough, he did not hesitate to take corrective action—to lay-off personnel. Even though the firm adjusted its operations in the Internet and teleoperator business, the actual numbers indicated that profitability was less than expected two months later. The business had not grown as budgeted—the environment was unpredictable because of increasing customer mobility due to New Communications Market Act. Consequently, a few dozen of employees were fired—more than ever before in the firm’s history. The CEO also decided to facilitate control and efficiency by physically centralizing all the domestic operations and assets into the same premises. He thereby sought to increase his capacity to control employees and other assets more directly and increase integrity within the firm. Along with the centralization efforts, the financial results indicated that the mobile entertainment business, even though growing, was lagging behind its financial performance targets. The CEO dismissed more employees in three waves that followed each other at intervals of about three months. Some international subsidiaries and profit centers were closed as they were burning cash rather than generating revenues. In addition, the CEO terminated development programs that seemed to occupy too much management attention and time. He thus sought to focus management efforts on improving demand prospects for the current products and services. The CEO and CFO primarily applied the budget-actual-variance framework in exercising control over business unit performance. Although profitability was primarily sought by means of operational efficiency, the firm also invested in growth. The investment decisions were simply based on payback times, thus supporting the profitability criterion of securing liquidity, as the CEO described: “Besides profitability, I emphasized fast liquidity. I did not accept a long delay between costs and revenues. This type of thinking drove our investment decisions. Probably we rejected profitable investments because the payback period was too long.” While profitability targets set the direction and liquidity set the boundaries for business development, the firm also monitored operative results carefully. The operative information systems produced valuable non-financial feedback in a timely manner to the operative units. The CFO described the non-financial indicators as follows:

334

M. Sandelin / Management Accounting Research 19 (2008) 324–343

“We monitored spoken minutes in fixed line and gsm calls, net of subscriptions as regarded almost all of our products, and order-delivery period in Internet access services. We calculated averages, absolute changes, and other ratios on a weekly basis in order to be proactive. In this sense the non-financials were extremely important.” These non-financial indicators were not performance targets. Weekly analysis of the ratios aimed at learning about financial results because the linkages between non-financial data and financial results had remained blurry as earlier. Digitized technologies impeded accurate allocation of overheads and thus exact product margins were unknown. The non-financials provided a proxy for estimating revenues and tracing customer satisfaction. With the passage of time, the firm learned more about its cost-structure and the ratios provided important feedback for adjusting actions in a proactive way. 3.4.2. The control practices complementing the results controls Action controls were manifested in administrative structures and constraints. The top management group consisted of the business unit managers, the CEO and CFO, and the heads of the support functions. Its role was “to maintain a shared information base and assess strategic options,” as the CEO explained. Operative decision-making authority was allocated instead to the business unit management groups. In the beginning, the CEO worked with business unit managers in striving to create action plans for the year at hand. He insisted that the managers draw up three alternative plans: best, basic and worst case scenarios. The chosen plan defined performance targets for each unit, although the scenarios also provided guidance for adjusting operations in the case of unexpected changes in the complex and uncertain environment. On the other hand, to establish the sense of accountability and hence to control actions in the business units, the CEO required formal minutes from management meetings at the business unit level. He commented on the rationale: “Without documentation these business items would have been forgotten.” Many of the managers had experienced rapid growth in the firm’s history and were used to taking the initiative without clear responsibility. They had less experience of the disciplined development and management of projects. Thus, the requirement for the formal documentation enhanced the sense of responsibility and accountability for the new ideas. A controller clarified the atmosphere within the business units as follows: “It was quite common to throw business ideas into air [in the business units]. Nobody really carried responsibility for these ideas. In this sense, documentation forced them to put someone in charge and also to draw up plans, at least roughly.” The CFO continued about the commitment to the financial plans: “The team that produced the forecast was also responsible for achieving the numbers in real life. We wanted them to understand this. The sense of accountability in the business units was not as strong as it should have been.” As the operative managers were unfamiliar with results controls, namely the operative budgets, controllers were hired for the business units. Their primary task was to conduct training and thus to enhance participative budgeting and commitment to the performance targets. The CEO’s cost-cutting actions had made it clear to the business unit managers that the performance targets had to be achieved. Consequently, they allocated financial accountability within their units by setting more detailed monthly and quarterly financial performance targets for the cost and profit centers. Moreover, the CEO called for specific actions in the business units, especially follow-up for working time, in order to decrease the emergence of loose ideas and to keep the focus on planned and budgeted actions. The CFO and controllers had namely observed that despite the efficiency claims and specific performance targets, many employees and managers readily engaged in ‘high-flying’ discussions. Despite the CEO’s requirement for working time monitoring, the results were unsatisfactory as new initiatives kept appearing constantly. The cultural tradition of unconstrained working time still affected behaviors. Stronger actions were taken as the CEO described: “I wanted the employees dedicated to operative tasks to be separated from those doing development tasks. As long as they were mixed, the operative staff was disturbed by the development staff. There was too much inefficiency.” Although the CEO did not totally deny growth initiatives, he prioritized profitability. Particularly his perception of a reasonable path for growth was based on systematic planning and organic business development rather than on radical

M. Sandelin / Management Accounting Research 19 (2008) 324–343

335

innovations. Hence, he decided to separate developmental and operational tasks structurally. The CEO’s requirements forced the business unit managers to adopt new modes of thinking and to focus on economic factors in decision-making. Their primary role was not to coach and inspire the search for innovation but to manage financial performance. Financial results gained significance and gradually they became routine in managerial working, particularly in planning, as a line manager described the plan to establish business in a new country: “We calculated initial investment costs including mandatory payments and fees, as well as marketing expenses. Then we kept the product price constant and calculated required sales volumes in order to figure out profitability. On that basis we evaluated the [business] potential of each country [prior to final decision].” In line with the preparation of alternative scenarios, the business unit and line managers included sensitivity analyses in their action plans. Financial planning induced action accountability, which was enhanced by communicating specific task descriptions and expectations to the employees. Moreover, given the importance of financial results, the senior management decided to nourish action accountability by implementing organization-pervasive approval policies, as a controller described: “We hired a person to take care of the development of policies, project proposals, approval practices, and so on. [. . .] We had to get approval from our superiors, a recommendation and signature for any type of purchase.” The acquisition of this specific resource highlights the senior management’s strong desire to control not only results, but also the actions underlying them. Personnel controls were relatively informal, although action and results controls were based on formal modes. The latter two were reflected in task assignment and recruitment practices. There was no explicit compensation for exceeding the performance target. However, if the target was not achieved, the senior management concluded, in the CFO’s words that “competencies did not match task requirements.” Indeed, the senior management was confident of making decisions on the basis of results controls because they had hired and appointed business controllers for the business units. Consequently, it was reasonable to assume that unsatisfactory results were due to managerial competencies rather than poor financial understanding. Once there was a need for hiring employees, the management sought to match competencies with tasks requirements. Competences were systematically evaluated on the basis of a candidate’s track record. There was no time and other resources for detailed hands-on training; new employees were expected to have the competence for the task in question. The CFO clarified personnel selection and placement: “We had a lot of process development work in the firm. Practically it meant that we had to hire many people with academic degrees. [. . .] These people had better qualifications to work in a bigger entity, organize and build up processes, and in a way, to structure activities.” Academic education was expected to reflect familiarity with process thinking and thus, capability to develop the firm infrastructure. Employees were placed strictly according to the competencies, as a senior manager described: “The functional placement of employees and tasks changed radically. For example, many development-oriented people were laid off while staff were simultaneously hired for customer service.” As the comment hints, the senior management emphasized customer service because they saw customer satisfaction as the only solid basis for organic growth. In the digital business, the costs of switching between service providers were minimal and thus the reputation of the firm was highly significant. The senior management allocated more human resources to customer service in an attempt to improve customer satisfaction. Hence, they sought to ensure that tasks were accomplished in this critical function. Moreover, the high employee turnover, as a result of the structural arrangements and strict accountability, resulted in more fact-based thinking within the firm, as a business controller reported: “Once we fired old managers and hired new ones, we got fresh blood here. [. . .] At the operative level people started to talk about facts and to look at reality instead of dreams and beliefs.” Despite the severe financial crisis, one third of the initial employees still worked for the firm. These employees had visions and beliefs about how to perform tasks and occasionally these traditions blurred the advancing accountability culture.

336

M. Sandelin / Management Accounting Research 19 (2008) 324–343

Cultural controls were not the primary mechanisms by which the senior management tried to affect and guide employee behavior. The ownership culture had lost its significance due to employee turnover4 . Organizational culture was not, however, considered insignificant. The senior management had recognized that some employees were not willing or able to accept fact-based management. Consciously or not, these employees kept up old patterns of action, manifested in innovative, ad-hoc decisions and actions. To overcome these somewhat dysfunctional but deeply rooted traditions, the senior management sought to shape organizational culture by sharing financial information, as a controller described: “Once the numbers for a quarter were ready, we had a firm-wide meeting. The CEO represented the actual figures, ratios, and plans. We had a clear aim to share financial information for the broader audience rather than just for the management.” As the controller continued, the senior management succeeded in this to some extent: “In these meetings some people almost fell asleep. However, I was often surprised about how interested the staff from business unit A was in the performance of business unit B. They had detailed questions and they wanted to know the underlying reasons.” Moreover, while interest in financial thinking increased throughout the organization, cultural change was even more evident at the management level. A business unit manager who had been involved in the firm operations from the beginning described the changes in thinking as follows: “We no longer talked about ‘first-mover-advantage’ but about ‘fit-advantage’. The second step taker was able to further develop the product, launch it at a lower price and even to introduce a technologically more sophisticated product than the first mover.” The result oriented performance management model had made managers consider economic factors and financial pros and cons more carefully prior to action. 3.5. Analysis of the Case 2 In the beginning, the firm struggled with cash sufficiency. The senior management sought, by making operations as efficient as possible, to secure liquidity and profitability, which were not, however, externally imposed imperatives. The external board was interested in monitoring share price. Hence, the new CEO and CFO had the freedom to decide about how to develop the firm and organize the control function. The headcount decreased from 250 to 220 in two years. On the basis of digital technologies the firm sought to compete with lowest price in a complex and uncertain though gradually stabilizing environment. Management control was primarily built around financial results. As the endeavor was to make the employees accountable for their actions by setting specific performance targets, the structural arrangements were a necessary precondition for the budgets. Besides the budget-actual-variance framework, result orientation was facilitated by nonfinancial indicators and payback time calculations. Here the payback calculations and action plans operated as pre-action reviews, and thus as action controls. It is not, however, obvious whether such controls should be understood as results or action controls, particularly if they include figures (cf. Merchant and Van der Stede, 2007). Other action controls focused on increasing accountability. Formal documentation and minutes shifted attention of business unit managers from innovating to carrying responsibility for efficiency. By means of approval policies they sought to control efficient and orchestrated actions at the lower organizational levels. Although these control practices were not linked as directly as action plans with the results controls, they complemented the hierarchical structure of result accountability. On the other hand, working time monitoring and structural separation of development and operational staff within business units supported the attainment of efficiency in operations and creativity in development. However, these practices were loosely coupled with other control elements although they evidently facilitated efficiency pursuits by focusing attention. 4 The company had earlier (in late 2000) issued new shares in order to fund acquisitions and thus the value of the shares held by the initial employees had been diluted. In addition, many employees had sold shares in the IPO. More recently, employee turnover had eroded the ownership culture.

M. Sandelin / Management Accounting Research 19 (2008) 324–343

337

Personnel selection and placement methods were affected by results and action controls. Task performance was evaluated on the basis of results and sanctioned accordingly. As this procedure increased results accountability, action planning pointed out the needs for specific competencies and thus, set the criteria for personnel selection and placement. Moreover, financial training in the business units complemented the result orientation although it also enhanced cultural change. For the same purposes of establishing an ‘accounting culture’, the senior management arranged a company wide meeting around interim reports in order to increase employees’ interest in financial results. The design of the control package was based on result orientation. The predominance of results controls determined the design of the most significant action and personnel controls. These controls complemented the result orientation by increasing accountability and also contributed independently to goal achievement by shaping organizational culture. In fact, only action controls were reciprocally linked with results controls, whereas the presence or absence of personnel controls had little impact on results or action controls. This package of controls seemed functional—the firm grew and improved profitability as manifested by the financial results (see appendix). 4. Comparative analysis of the cases and discussion This study has examined operation of management control practices as a package by presenting two case descriptions from the same firm at different time periods. The empirical analysis was based on Merchant’s (1998), Merchant and Van der Stede (2007) object-of-control framework about the diverse forms of management controls. Hence the approach went beyond the controls with information content (Simons, 1995) and the formal MAS on which the prior studies have focused in a growth-firm context (Granlund and Taipaleenmäki, 2005; Davila and Foster, 2005; Sandino, 2007). This section builds on the comparative case analysis and discusses features explaining the functionality of the control packages, the extent to which they seem to be equifinal, and functional demands as explanations for variety in management control packages. The comparative analysis revealed that contingent factors were quite similar between the cases. Although these factors varied, the degree of variation was modest. Hence, the contingent explanation for control package variety was not obvious. Instead, the cases were interpreted as being similar in terms of contingencies and an alternative explanation was sought from the notion of equifinality, which suggests that control systems may look fundamentally different but result in an equally good final state even in the face of similar contingencies. If there is a high degree of conflict in functional demands, two forms of equifinality are possible – suboptimal and configurational5 – depending on the degree of latitude with respect to the structural, here the control package, arrangements (Gresov and Drazin, 1997). In both cases, the senior management had somewhat unconstrained latitude to implement and organize control practices, as is typical for newly founded and family businesses. The venture capitalist(s) were more interested in share prices than in management methods. The management also faced conflicting functional demands—whether to emphasize new product development and innovation, or efficiency and development of firm infrastructure. Therefore, the cases seem to represent configurational equifinality—different combinations of control elements with which the firm achieved both growth and profitability6 . None of the functional demands7 , however, was dominant enough to dictate the best way of organizing the control function. As manifested by the first case, the senior management particularly emphasized new product development. Consequently, results controls were used only at the senior management level in order to preserve culture and protect the innovative, technological core of the firm from administrative bureaucracy. The concerns of profitability and/or efficiency were separated from those of innovation in an attempt to succeed in at least one of the functional demands. The management assumed that rapid revenue growth by means of broad product portfolio and scale benefits would recover costs and enable competition by means of the lowest price, whereas in the latter case cost-efficiency was seen as a precondition for the lowest price. In line with this, the latter case illustrated how efficiency was sought throughout the organization by implementing formal results and action controls, and the development of firm infrastructure gained 5

The other forms of equifinality are ideal profiles (low conflict in functional demands and constrained latitude of structural options) and tradeoff equifinality (low conflict in functional demands and unconstrained latitude of structural options). See Gresov and Drazin (1997) for detailed explanations. 6 Although performance levels were not identical, they nevertheless suggest that the firm was able to respond to the functional demands. 7 There were also other functional demands such as customer satisfaction and acquisition for instance. For the sake of simplicity, only the most dominant demands are considered here.

338

M. Sandelin / Management Accounting Research 19 (2008) 324–343

importance in employee selection and placement. Innovative branches of business were detached from the core of the ‘invoicing machine’ and even within it, efficiency imperative was reinforced by separating development and operative tasks. Two different sets of control practices have many implications. First, the mere presence of different control practices does not explain the functionality of the package of controls. The functionality of different organizational configurations has been argued to depend on internal consistency, in both equifinality (Gresov and Drazin, 1997) and control package literature (Abernethy and Chua, 1996). The concept has remained, however, a black box. Abernethy and Chua (1996) have contended that the control systems operate “as a package when they are internally consistent—that is, they are designed to achieve similar ends” (ibid. p. 573). Their findings suggest that each control element contributes independently and directly to goal attainment; internal consistency echoes independent yet goal-consistent design of control elements. This study is not particularly informative of the process of making a control package internally consistent, although the latter case suggests quite radical rather than incremental adjustment of control elements. However, of greater importance here is that none of the control elements was sufficient alone. In both cases, internal consistency was built into the control system by prioritizing a certain form of control—cultural in the former and results in the latter one. The fact that the result orientation became part of the new culture in the latter case, and the existence of hybrid controls in the former, implies that different controls are often tightly intertwined in the empirical context and sometimes only analytically separable (cf. Merchant and Van der Stede, 2007). Moreover, in both cases the internal consistency of the control packages had little to do with the independent yet goal-consistent operation of control elements, thus suggesting a different logic from that found by Abernethy and Chua (1996). Only the primary mode of control was aimed at facilitating success in the prioritized functional demand. The design of other control elements was affected by the primary mode of control, whereas their use reinforced the particular control orientation. In this sense, the reciprocal relationships between design and use orchestrated the package of controls. The primary control element was given material significance by other control elements (e.g. Alvesson and Kärreman, 2004). Thus, here the logic of internal consistency builds on reciprocal processes: the primary mode of control shapes the design of the control package whereas the use of the secondary modes of control complements the primary one. The use of the secondary modes of control secures the primacy of certain mode of control to the extent that their design is based on the primary one. The evidence suggests that appropriate management control packages are not mere functions of a single control element, such as culture or results, but are based on combinations of control elements that can support a particular control orientation or management philosophy, at least in a growth firm context (cf. Lukka and Granlund, 2003; Granlund and Taipaleenmäki, 2005). While it seems that internal consistency can be achieved through reciprocal linkages between elements within a control package, it is worth noting, however, that loosely coupled control practices may significantly facilitate the goal attainment besides the primary/core control system (Abernethy and Chua, 1996). Hence, it seems that the internal consistency of MCS packages can be conceptualized either as an independent yet goal-consistent design and operation of each control element (Abernethy and Chua, 1996) or as a “primarycontrol-driven-consistency-of-elements” that is based on reciprocal linkages of design and use as pointed out in this study. To sum up, it is argued that a relatively simple and less accounting-centric control package can be functionally equivalent with more formal and accounting-centric designs if the internal consistency holds. This important feature – the internal consistency of management control packages – lends support to the argument about the functionality of a formal control system being dependent on the coherence and strength of the linkages between system elements (Ferreira and Otley, 2005). However, internal consistency requirement is not limited to formal systems such as MAS, but to their linkages with organizational structure and culture (Flamholtz, 1983) and informal but systematically applied control practices as well. This implies that organizational culture (and structure) should not be regarded merely as premises, but also as a form of management control (e.g. Ahrens and Mollina, 2007; Merchant and Van der Stede, 2007; Chenhall, 2003) that complements or is complemented by MAS, for instance. Second, the equifinality of the control packages does not come without limitations. The financial crisis demonstrated that the control package in the first case was sufficient only for the short term. A more specific analysis of the boundary conditions in the first case suggests that as operations spread to the international scene, the control package proved to be insufficient because it was significantly dependent on the presence of human actors and was unable ‘to act at a distance’ (see e.g. Lukka and Granlund, 2003). As many managers were abroad, establishing businesses in new

M. Sandelin / Management Accounting Research 19 (2008) 324–343

339

countries, there were inadequate managerial resources for conducting personal supervision, for instance, in the parent company. Moreover, as a result of innovations the product portfolio had expanded and the business had also become more complex in this sense. The functional demands were shifting away from product innovations to systematic and coordinated maintenance of a complex businesses portfolio. The control package was not, however, developed and reformed to meet these challenges. The apparent reason for the stability of the control package seems to reside in its functionality. It divested managerial attention of the need to change the control package because the firm was able to grow and improve profitability as well. It also convinced external parties, as the venture capitalist recalled: “It was a good [management control] model. The decisions seemed to be correct as the firm was innovative and agile without any sluggishness and bureaucracy.” Hence, it is argued that informal cultural, personnel, and action controls, if they are internally consistent and hence functional, form a substitute for the need to adopt more formal control systems. This lends support to Huikku’s (2007) findings, which suggest that a systematic, albeit informal control practice replaces, at least partially, the adoption of a formal control practice. Here the contribution is that this substitution effect may also occur at the more comprehensive level between control package elements or sub-systems such as cultural, personnel, action, and results controls. This also suggests that internally consistent combinations of controls may prevent external intervention, e.g. by venture capitalists, in control system design and MAS adoption (cf. Davila and Foster, 2005; Granlund and Taipaleenmäki, 2005) and thus explain change and stability of MAS (e.g. Granlund, 2001; Lukka, 2007). Moreover, the findings imply that appropriate control can be achieved by other means than extensive MAS, even though the number of employees increases (cf. Davila and Foster, 2005). The appropriateness of informal control practices seems to be, as discussed above, conditional on operational complexity. That is, for instance, complexity originating from the geographical dispersion of operations and product portfolio expansion that induces a functional demand for processing greater amounts of information (Chapman, 1997; Galbraith, 1973). It is therefore argued that by keeping the organizational control package relatively simple (Lukka and Granlund, 2003) and hence internally consistent, a growth firm can successfully manage relatively local and limited operational complexity. However, as operational and geographical complexity increases the need for adopting more formal, information mediating control practices increases (Moores and Yuen, 2001). Third, the findings suggest that the design of the control package was mainly affected by the managerial response to functional demands. This in no way excludes the possibility that these functional demands are related to the lifecycle phase, history, managerial preferences, or institutional environment (Gresov and Drazin, 1997). As the latter case demonstrates, it is quite evident that voluntary debt restructuring, although the firm gained new capital, shaped managerial preferences for seeking profitability. However, the argument here is that rather than mere contingencies, the way of responding to the functional demands affects the control package design and adoption of MAS. In the face of similar contingencies, the functional demands can be the same (or differ), but the management may prioritize these demands differently and thus implement different control packages. Finally, some caution is needed in interpreting the findings. Equifinality of the control packages does not follow the traditional reasoning of functional equivalence (cf. Gresov and Drazin, 1997). The control packages are evaluated against overall goal attainment instead of functional performance. On the other hand, equifinality is a theoretical concept; it is rarely a perfect match with empirical settings. Here it is contended, however, that the cases could quite likely be regarded as equifinal even if the empirical settings are not a perfect match. Undoubtedly one can argue that the strategy was different between the cases. Such an argument, however, depends on how abstract concepts such as strategy8 are conceptualized. In both cases the competitive advantage, a widely acknowledged approach to strategy (Porter, 1985), was sought with the lowest price. The fact that efficiency became an imperative in the latter case manifests the pursuit of the lowest cost and thus, operational support for the lowest price. The former case illustrates a different operational logic of revenue growth, i.e. high sales volumes were expected to generate revenues exceeding the ‘fixed operational costs’ which were due to deployment of knowledge and information technology resources. However, if the managerial responses to functional demands are perceived as strategy, these responses and demands should guide

8

See Langfield-Smith (1997, 2007) for a review of various strategy concepts in MCS literature.

340

M. Sandelin / Management Accounting Research 19 (2008) 324–343

our strategy conceptualization efforts9 . Managerially meaningful strategy constructs are likely to provide insights into designs, uses and alterations of management control packages and facilitate the attempt to put management back to accounting (Otley, 2001). On the other hand, operating environments may seem different at first sight. However, during both periods there were changes in telecommunication regulations, mergers and acquisitions affected industry structure, and new opportunities were emerging as the mobile and digital communication devices became more sophisticated. It is acknowledged that the degree of change varies between the cases. However, it seems that environmental changes or uncertainty had little direct impact on functional demands and managerial responses to them and finally, to the designs of the control packages. Rather it seems that managerial idiosyncrasies mediated uncertainty into organizational action and affected MCS design in the former case, whereas in the latter one, although the environment may seem more certain, the CEO faced uncertainty as he commented: “. . . possibly I was throwing babies [profitable products] out with the bathwater.” There was an information deficit but the CEO sought to simplify the management context and build on relatively certain facts. Thus, environmental uncertainty seems inconclusive with respect to control package design. 5. Concluding remarks In recent years there has been little empirical research on the operation of formal MAS and less formal control practices as a package. This study has examined a wide array of control practices and sought to address features explaining the functionality of a control package and the needs for adopting formal MAS in relation to other control elements. This study contributes to our knowledge of control system design by pointing out that rather than mere contingencies, the control package design is driven by functional concerns, particularly the way in which conflicting functional demands are prioritized and/or balanced. Second, the study shows that the control function involves many elements and that functionality of a control package seems to depend on internal consistency between these elements. Third, the study has demonstrated two fairly equifinal control packages and pointed out that equifinality is conditional on the complexity of operations. Finally, the study has shown that management accounting change and stability can be significantly affected by informal control practices if they are able to deliver the desired results. The findings are based on relatively a small, growth oriented ‘new economy firm’, limiting direct generalization to wider population. It should also be noted that the described packages are not necessarily optimal ones and that neither do they suggest any trend of control system evolution. Also equifinality concept has its limitations in an empirical context. However, the theoretical contributions of this study encourage further examination of management control packages, particularly mechanisms of internal consistency, rather than mere collection of control techniques in different settings. On the other hand, equifinality appears to be a possible way to integrate dispersed and sometimes conflicting findings of contingent MCS studies. It also directs attention beyond the contingencies, to the managerial challenges of balancing between conflicting interests and demands. Hence, future studies, instead of examining the existence of MAS merely in relation to the contingent factors, could examine functional demands vis-à-vis MCS packages. Efficiency, customer satisfaction, quality improvement, and new product development among other functional demands are pertinent to the majority of contemporary for-profit organizations. Understanding the act of balancing between these requirements, or more generally the balance between exploitation and exploration (e.g. March, 1991, 2006; Gupta et al., 2006), is not only likely to explain control system variety but also to help practitioners in coping with trade-offs between control elements. Acknowledgements The author is grateful for the constructive comments and suggestions of David Bedford, Seppo Ikäheimo, Jari Huikku, Taru Lehtonen, Teemu Malmi, Juhani Vaivio, two anonymous reviewers, and the Editors of the special issue on earlier versions of the paper. The financial support of the HSE Foundation, the Foundation for Economic Education, the Finnish Foundation for Economic and Technology Sciences, and the Kemira Foundation is also greatly appreciated. 9

Miles and Snow’s (1978) strategy typology captures some functional demands (e.g. new product and market development, efficiency) but ignores many other functional concerns.

M. Sandelin / Management Accounting Research 19 (2008) 324–343

341

Appendix

References Abernethy, M.A., Chua, W.F., 1996. A field study of control system “redesign”: the impact of institutional processes on strategic choice. Contemporary Accounting Research 13 (2), 569–606. Ahrens, T., Dent, J.F., 1998. Accounting and organizations: realizing the richness of field research. Journal of Management Accounting Research 10, 1–39. Ahrens, T., Mollina, M., 2007. Organisational control as cultural practice—A shop floor ethnography of a Sheffield steel mill. Accounting, Organizations and Society 32 (4–5), 305–331. Alvesson, M., Kärreman, D., 2004. Interfaces of control. Technocratic and socio-ideological control in a global management consultancy firm. Accounting, Organizations and Society 29, 423–444. Brown, D., 2005. Management Control System Packages. PhD dissertation, University of Technology, Sydney. Brown, D., Malmi, T., Booth, P., 2008. Loosely coupling theory of management control systems. Paper presented in 2008 Global Management Accounting Research Symposium (GMARS). Bruns, W.J., Waterhouse, J.H., 1975. Budgetary control and organizational structure. Journal of Accounting Research 13, 177–203. Cardinal, L.B., Sitkin, S.B., Long, C.P., 2004. Balancing and rebalancing in the creation and evolution of organizational control. Organization Science 15 (4), 411–431. Chapman, C.S., 1997. Reflections on a contingent view of accounting. Accounting, Organizations and Society 22 (2), 189–205. Chenhall, R.H., 2003. Management control system design within its organizational context: findings from contingency-based research and directions for the future. Accounting, Organizations and Society 28 (2/3), 127–168. Collier, P.M., 2005. Entrepreneurial control and the construction of a relevant accounting. Management Accounting Research 16, 321–339. Covaleski, M.A., Evans III, J.H., Luft, J.L., Shields, M.D., 2003. Budgeting research: three theoretical perspectives and criteria for selective integration. Journal of Management Accounting Research, 15. Davila, T., Foster, G., 2005. Management accounting systems adoption decisions: evidence and implications from early-stage/startup companies. The Accounting Review 80, 1039–1068. Doty, D.H., Glick, W.H., Huber, G.P., 1993. Fit, equifinality, and organizational effectiveness: a test of two configurational theories. Academy of Management Journal 36 (6), 1196–1250. Eisenhardt, K., 1989. Building theories from case study research. Academy of Management Review 14 (4), 532–550.

342

M. Sandelin / Management Accounting Research 19 (2008) 324–343

Ferreira, A., Otley, D., 2005. The design and use of management control systems: an extended framework for analysis. Working paper. Fisher, J., 1995. Contingency-based research on management control systems: categorization by level of complexity. Journal of Accounting Literature 14, 24–48. Flamholtz, E.G., 1983. Accounting, budgeting and control systems in their organizational context: theoretical and empirical perspectives. Accounting, Organizations and Society 8 (2/3), 153–169. Flamholtz, E.G., 2005. Strategic organizational development and financial performance: implications for accounting, information, and control. Advances in Management Accounting 14, 139–165. Flamholtz, E.G., Das, T.K., Tsui, A.S., 1985. Toward an integrative framework of organizational control. Accounting, Organizations and Society 10 (1), 35–50. Galbraith, J.R., 1973. Designing Complex Organizations. Addison-Wesley, Reading, MA. Gerdin, J., 2005. Management accounting system design in manufacturing departments: an empirical investigation using a multiple contingencies approach. Accounting, Organizations and Society 30, 99–126. Granlund, M., 2001. Towards explaining stability in and around management accounting systems. Management Accounting Research 12 (2), 141–166. Granlund, M., Taipaleenmäki, J., 2005. Management control and controllership in new economy firms—a life cycle perspective. Management Accounting Research 16, 21–57. Greenhalgh, R.W., 2000. Information and the transnational SME controller. Management Accounting Research 11, 413–426. Greiner, L.E., 1998. Evolution and revolution as organizations grow. Harvard Business Review, 55–64. Gresov, C., Drazin, R., 1997. Equifinality: functional equivalence in organization design. Academy of Management Review 22 (2), 403–428. Gupta, A.K., Smith, K.G., Shalley, C.E., 2006. The interplay between exploration and exploitation. Academy of Management Journal 49 (4), 693–706. Huikku, J., 2007. Explaining the non-adoption of post-completion auditing. European Accounting Review 16 (2), 363–398. Keating, P., 1995. A framework for classifying and evaluating the theoretical contributions of case research in management accounting. Journal of Management Accounting Research, 66–86. Kärreman, D., Sveningsson, S., Alvesson, M., 2002. The return to machine bureaucracy? Management control in the work settings of professionals. International Studies of Management and Organization 32 (2), 70–92. Langfield-Smith, K., 1997. Management control systems and strategy: a critical review. Accounting, Organizations and Society 22, 207–232. Langfield-Smith, K., 2007. A review of quantitative research in management control systems and strategy. In: Chapman, C.S., Hopwood, A.G., Shields, M.D. (Eds.), Handbook of Management Accounting Research—Volume 2. Elsevier, Oxford. Lukka, K., 2005. Approaches to case research in management accounting: the nature of empirical intervention and theory linkage. In: Jönssön, S., Mouritsen, J. (Eds.), Accounting in Scandinavia—The Northern Lights. Kristianstads Boktryckeri, Sweden, Kristianstad. Lukka, K., 2007. Management accounting change and stability: loosely coupled rules and routines in action. Management Accounting Research 18, 76–101. Lukka, K., Granlund, M., 2003. Paradoxes of management and control in a new economy firm. In: Bhimani, A. (Ed.), Management Accounting in the Digital Economy. Oxford University Press. March, J.G., 1991. Exploration and exploitation in organizational learning. Organization Science 2 (1), 71–87. March, J.G., 2006. Rationality, foolishness, and adaptive intelligence. Strategic Management Journal 27, 201–214. Merchant, K.A., 1981. The design of corporate budgeting system: influences on managerial behavior and performance. The Accounting Review 56 (4), 813–829. Merchant, K.A., 1998. Modern Management Control Systems. Prentice Hall, New Jersey. Merchant, K.A., Otley, D., 2007. A review of the literature on control and accountability. In: Chapman, C.S., Hopwood, A.G., Shields, M.D. (Eds.), Handbook of Management Accounting Research—Volume 2. Elsevier, Oxford. Merchant, K.A., Van der Stede, W.A., 2006. Field-based research in accounting: accomplishments and prospects. Behavioral Research in Accounting 18, 117–134. Merchant, K.A., Van der Stede, W.A., 2007. Management Control Systems. Prentice Hall. Miles, R.E., Snow, C.C., 1978. Organizational Strategy, Structure, and Process. McGraw-Hill, New York. Miller, C.C., Cardinal, L.B., Glick, W.H., 1997. Retrospective reports in organizational research: a reexamination of recent evidence. Academy of Management Journal 40, 189–204. Mintzberg, H., 1983. Structures in Fives: Designing Effective Organizations. Prentice-Hall, Englewood Cliffs, NJ. Mintzberg, H., Ahlstrand, B., Lampel, J., 1998. Strategy Safari: A Guided Tour Through the Wilds of Strategic Management. Free Press: Simon & Schuster. Mitchell, F., Reid, G.C., 2000. Editorial. Problems, challenges and opportunities—the small business as a setting for management accounting research. Management Accounting Research 11, 385–390. Moores, K., Yuen, S., 2001. Management accounting systems and organizational configuration: a life-cycle perspective. Accounting, Organizations and Society 26, 351–389. Otley, D., 1999. Performance management: a framework for management control systems research. Management Accounting Research 10, 363–382. Otley, D., 2001. Extending the boundaries of management accounting research: developing systems for performance management. British Accounting Review 33, 243–261. Otley, D., Berry, A.J., 1994. Case study research in management accounting and control. Management Accounting Research 10 (10), 45–65. Perren, L., Grant, P., 2000. The evolution of management accounting routines in small business: a social construction perspective. Management Accounting Research 11, 391–411. Porter, M.E., 1985. Competitive Advantage. Creating and Sustaining Superior Performance. Free Press, New York.

M. Sandelin / Management Accounting Research 19 (2008) 324–343

343

Sandino, T., 2007. Introducing the first management control systems: evidence from the retail sector. The Accounting Review 82 (1), 265–294. Scapens, R.W., 1990. Researching management accounting: the role of case studies. British Accounting Review 22, 259–281. Simons, R., 1995. Levers of Control: How Managers use Innovative Control Systems to Drive Strategic Renewal. Harvard Business School Press, Boston. Spekle, R.F., 2001. Explaining management control structure variety: a transaction cost economics perspective. Accounting, Organizations and Society 26, 419–441. Yin, R.K., 1993. Applications of Case Study Research. SAGE Publications, Thousand Oaks.