Interfaces of control. Technocratic and socio-ideological control in a global management consultancy firm

Accounting, Organizations and Society 29 (2004) 423–444 www.elsevier.com/locate/aos

Interfaces of control. Technocratic and socio-ideological control in a global management consultancy firm§ Mats Alvesson*, Dan Ka¨rreman Department of Business Administration, Lund University, PO Box 7080, S 220 07 Lund, Sweden

Abstract This paper investigates a variety of forms of management control in a large management consultancy company. The very high level of compliance with corporate objectives among employees is highlighted and singled out as a phenomenon worth exploring. The company exhibits a rich variety of various formal control devices focusing on financial issues as well as human resources. The paper shows that these do not fully work according to intention and that their control effects do not comply that well with the bureaucratic- technocratic logic they rest upon. Technocratic control systems are instead identified as non-obvious sources of socio-ideological control. The paper emphasizes the interface between different forms of control and argues for a more symbolic, meaning-focusing view on bureaucratic and output measurement control. # 2003 Published by Elsevier Ltd.

Introduction Forms of management and organizational control are a common theme in organization and business studies. In fact, the whole idea of management accounting, for example, is founded on the belief that management control is possible, important, and, indeed, necessary. Although the literature on organizational and management control suggests a wide array of forms of control, it is common to emphasize a main form of control, either in the form of a particular organizational structure or in the form of a specific mode of control dominating. Mintzberg’s (1983a, 1983b) five types are well-known, as are Edwards’ § Paper presented at conference on ‘‘Information Flows in Knowledge-Intensive Firms’’, Bocconi University, Milan, 27 November 2001. * Corresponding author. E-mail address: [email protected] (M. Alvesson).

0361-3682/$ - see front matter # 2003 Published by Elsevier Ltd. doi:10.1016/S0361-3682(03)00034-5

(1979) three and a variety of versions setting up one authoritarian-hierarchical-tayloristic against a normative-market-self-governance version (e.g. Clegg et al, 1996; Friedman, 1977; Govindarajan, 1988; Govindarajan & Fisher, 1990; Spekle´ 2001). One influential variation of the last theme is to contrast structural control with cultural control. Earlier enthusiastic versions of this include Peters and Waterman (1982), but also more sceptical researchers like Kunda (1992) have emphasized the significance of normative control—or engineering culture—as a particular form of control. There is a strong tendency to emphasis an either-or orientation in the literature. The first author of the present paper is no exception, having paid particular attention to cultural control (Alvesson, 1993, 1995). To concentrate on one type of phenomenon may often be motivated by the need to have a focus. Theoretical interest as well as what seems to be vital or interesting in a particular empirical site can create a rationale for

424

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

investigating a particular form of control. It is sometimes also argued that organizations typically are structured in particular configurations and tend to move towards balanced and ‘pure’ forms (Mintzberg, 1983a). In a study of a major American high technology company, Van Maanen and Kunda (1989) emphasize that for many employees ‘‘. . .‘culture’ replaces ‘structure’ as an organizing principle and is used both to explain and guide action’’ (p. 72). They consider that the formal organization is not, per se, particularly important. However, a case can be made for understanding various forms of control as simultaneously active, at least in reasonably complex organizations. Indeed, Kunda (1992) maintains that certain elements of bureaucratic control remain in place, even when normative control is predominant. In the company studied by Kunda, control of culture is supplemented by bureaucratic and utilitarian control. To use Mintzberg’s (1983a) typology, a battery of control forms seems to be used in many contemporary organizations, typically blending standardized output measures with standard operating procedures (standardized work procedures) and a professionalized work force (standardized knowledge). In such settings it might be counter-productive to assume the existence of a dominating form. Indeed, different control forms may be linked to, and supporting and sustaining each other, rather than subdued and marginalized by a dominant form. On a conceptual level, it can be argued that contrasting forms of control, while analytically distinct, do not necessary exclude each other. On the contrary, it can be argued that structural forms of control also have a significant cultural dimension (Ferner, 2000; Ranson, Hinings, & Greenwood, 1980). From an anthropological point of view, it could be argued that structural forms of control are cultural phenomena themselves, that have and take on specific meaning depending on cultural contexts (taken for granted in Western culture, signifying modernism and/or imperialism in other cultures). Structural forms never exist in a culture-free context, nor is this ideational level necessarily tightly connected to a particular social form. Structural forms of control focusing on behaviour and/or measurable output

are supported and/or complemented, sometimes perhaps challenged or contradicted by, cultural ideas and values. The present analysis focuses on the interface between certain structural arrangements and the ideologies and norms targeting how people are supposed to think, feel and act. A key theme is how a high level of compliance—including a readiness to work very long hours and accept very ambitious objectives in terms of keeping deadlines and accomplishing high margins in project work— is achieved. How different forms of technocratic control interact and merge with socio-ideological control is investigated. The empirical domain addressed is the management consulting industry and profession, in particular as it is manifest in very large, international companies whose services include advice-giving and strategic planning as well as the implementation of suggested solutions, often involving a large group of consultants over a long period of time. How a fairly strict form of control is accomplished in one such company is an interesting question which is capable of giving some input into how we can understand modes of management control within the particular empirical domain investigated, but also in more general theoretical terms.

Forms of management control Although managers do other things, the exercise of control is a dominant part of the manager’s job (Mintzberg, 1989; Tengblad, 2001). Management control has been defined in numerous ways but most definitions seem to agree that management control includes the exercise of power (influence) in order to secure sufficient resources, and mobilize and orchestrate individual and collective action towards (more or less) given ends (c.f. Langfield-Smith, 1997; Spekle´, 2001 for reviews). Management control typically includes an apparatus for specifying, monitoring and evaluating individual and collective action. It focuses worker behaviour, output and/or the minds of the employees. Sometimes it attempts to focus on all three.

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

Managerial activity that attempts to control behaviour typically includes designing and supervising work processes. This is usually carried out in a way that attempts to make work processes as simple and transparent as possible, thereby lowering knowledge thresholds (and the price of labour). In this sense, the long dominance of Taylorism, broadly defined, as the managerial approach is unsurprising. Its approach is to construct work contexts where workers discretion is minimized, to the extent that they only can do the prescribed thing with a minimum of effort and movement. Management control practices targeting minds, through norms, emotions, beliefs and values, are intended to affect behaviour indirectly. This type of control has been discussed under the label of normative (Etzioni, 1960; Kunda, 1992), culturalideological (Alvesson, 1993), and concertive (Barker, 1993) control. In this article we will use the label socio-ideological control for attempts to control worker mind-sets and technocratic control for attempts to directly control worker behaviour (c.f. Alvesson & Ka¨rreman, 2001a). In the technocratic type, management works primarily with plans, arrangements and systems focusing behaviour and/our measurable outputs. In the socio-ideological version, social relations, identity formation and ideology are basic ingredients. Technocratic forms of control Most studies of social control of work and organization have traditionally focused on the objective, behavioural aspects of control. In a review article, Simpson (1985) for example, drawing heavily on Edwards (1979), identifies five modes of control over work: simple, technical, bureaucratic, occupational control, and worker self-control modes of control. Simple control refers to the personal exercise of power of the boss (owner, supervisor) over the worker. Technical control is embedded in the technology of work. Bureaucratic control is carried out through rules, policies, formal incentives and other impersonal devices. In occupational control, a profession defines appropriate/no appropriate work behaviour. Finally, worker self-control labels control

425

contexts where the producers themselves have a high degree of discretion. Until some time ago, this ‘‘objectivistic’’ view on control was dominant. In his synthesis of research on organizational structure, Mintzberg’s (1983a) five coordination mechanisms concern objective, external control forms and hardly anything on the significance of the whole sphere of culture and ideology. Ouchi (1979:840), for example, noted that ‘‘present organization theory. . . concentrates on the bureaucratic form to the exclusion of all else’’. When ideology is considered, it is basically as legitimating these forms of ‘‘objective’’ social control (Burris, 1989) or as a major control device in very special organizations, such as the ‘‘missionary’’ one (Mintzberg, 1983b). In the context of management accounting research, management control has largely been interpreted in technocratic terms, as either exercised through output control (through focus on various key performance indicators, such as profit, sales, and quality measurements) or through behavioural control (such as direct supervision, rules, standard operating procedures, and business policies) (Govinadarajan, 1988; Govindarajan & Fisher, 1990; Langfield-Smith, 1997; Lind, 2001; Spekle´, 2001). The insight that management accounting seeks not only to affect behaviour but also consciousness to some extent transcends a narrow technocratic focus (Puxty, 1993). But as Langfield-Smith (1997) points out, most research on management control in management accounting research has failed to acknowledge the distinction between ‘‘the existence and the use of controls’’ (p. 226) as well as addressing forms of control that extend beyond the familiar terrain of output and formal behaviour. However, as pointed out above, and elaborated below, management control has a broader theoretical and practical spectrum. Socio-ideological forms of control Managers do not only exercise control through prescribing behaviour or desired outputs. Managers also often seek to enact a particular form of organizational experience for others (Alvesson & Willmott, 2002; Deetz, 1992; Mumby, 1988; Smircich & Morgan, 1982).

426

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

They may attempt to define interpretations and meanings than can become widely understood and shared by organization members so that actions are guided by a common definition of the situation. Those with power are able to influence the course of organizational development through control over valued resources and through use of symbols by which organization members mediate their experience (Smircich, 1983:161). In management accounting research, what we have labelled socio-ideological forms of control are more frequently (and misleadingly) labelled informal controls, or in some cases, clan controls (Langfield-Smith, 1997). Ouchi (1979, 1980) has suggested that the clan can serve as an organizational control form in which the level of uncertainty is too great for the classic solutions—the market and the bureaucracy—to function. The clan rests on ‘‘. . .social agreement on a broad range of values and beliefs’’ and ‘‘. . .relies for its control upon a deep level of common agreement between members on what constitutes proper behaviour, and it requires a high level of commitment on the part of each individual to those socially prescribed behaviours’’ (Ouchi, 1979:838). Many writers focusing on ideology come close to what others refer to as clans or cultures, even though the latter concept often is understood as a broader and more complex one. Ideology can be defined as an integrated set of values, ideals and understandings about a particular part of social reality which justify certain commitments and actions (c.f. Beyer, 1981; Geertz, 1973; Weiss & Miller, 1987, etc). Sometimes ideology is seen as control (Czarniawska-Joerges, 1988) but typically a rather broad and general concept is used, which can cover a wide area of possible forms of more specific control. Ideology is often confronted with a single traditional conception of management or control. Becke´rus, Edstro¨m et al. (1988), for example, argue for a view on management based on control through ideology (ideas) rather than through control through instructions. Socio-ideological control can be defined as efforts to persuade people to adapt to certain values, norms and ideas about what is good,

important, praiseworthy, etc in terms of work and organizational life. Ideologies justify certain principles, actions and feelings, and discourage others (Alvesson, 1987; Czarniawska-Joerges, 1988, etc). They are more or less elaborated and deeply held, both on behalf of the person who tries to exercise ideological control and the audience that has been persuaded. When engaged in ideological control efforts, managers more or less consciously and systematically, try to make the employees adhere to the values and ideals which they believe in—or at least the values and ideals which they believe that the company would benefit most from the employees believing in. We can imagine a spectrum of forms of implementation of ideological control. At the one extreme, the agent of ideological control is reproducing ideas that he or she takes for granted and do not reflect upon—in this case the controller is him- or herself also controlled, and it is questionable if this is a case of (conscious) management control. At the other extreme, the agent is manipulating—proposing ideals and values that he/she himself does not believe in as means for influencing others. The latter enterprise probably easily creates backlashes, while there is a risk that people in the long run see through manipulations. In the normal case, ideological control is based on a combination of the calculations and convictions of key actors. On the relationship between technocratic and socio-ideological control The literature typically portrays different forms of control as external and alternative to each other. As indicated in the introduction, most authors aiming to give the broader picture (Mintzberg, Edwards, etc) stress that different organizations and different periods are dominated by different forms of control. Barley and Kunda (1992) talk of cycles of control rhetorics, in which normative and rational forms dominate during different periods in a cyclical way. But also researchers having conducted in depth case studies emphasize a dominating form either focusing on structure and behaviour or corporate culture, ideology and identity (c.f. Kunda, 1992; Martin, 1992; Alvesson,

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

1995). Some draw attention to the interplay between different modes of control. Barker (1993) shows how people’s ideas and norms lead to the construction of rules enacted in a bureaucratic way. Ferner (2000) emphasizes that ‘formal controls are underpinned by personal and cultural control mechanisms’ (p. 524). Some critically oriented, Foucault-inspired accounting research also transcend the distinction between structuralbehavioural and ideological (discursive)-subjective when emphasizing how management accounting and control systems ‘are part of the organization’s (and society’s) regular, routine procedures which embed managers and employees in a disciplinary, punishable web of discourses and practices which go on almost unnoticed and appear as neutral (Hopper & Macintosh, 1993:190, see also Hopwood, 1987; Miller & O’Leary, 1987). Also transcending a simple structure-ideology divide is the point that in certain respects ‘it is not the forecasting ability of a budget that is important, rather it is the desirability of the situation that it helps to create’ (Tinker et al., cited by Puxty, 1993:102). An interesting question is, however, the more precise relationship between different types of control. An alternative idea would be proceed from the assumption that management employs a wide variety of forms of control and it may be difficult to point at a single, dominant one. At least, one could see it as good research question to study the plurality versus domination of a single mode(s) of organizational control. Of course, the way one approaches this question is partly a matter of how broad are the categories one is using. Given a broad view of ideological control, it covers a lot and easily become dominant, but one may also favour dividing the control pie into smaller cakes and then plurality appears. Another interesting question concerns to what extent different forms of control are exercised in different ways, external to each other and perhaps complementing or contradicting each other. Are different forms of control pure or fused. Does for example bureaucratic control mainly function in an ‘objective’ or mechanistic way, or is it heavily dependent on a particular mode of interpretation, based on values, convictions and norms, in order to work? And is any ideational control grounded

427

mainly in the communication of ideological messages, expressed by leaders, explicitly spelling out what is good and valuable? Or is it firmly anchored and expressed in the organizational systems and practices for behavioural and output control? Most of the literature seems to argue that ideological control is exercised in non-bureaucratic and non-output measurement ways. Leadership is typically emphasized in relationship to cultural or ideological control (e.g. Schein, 1985). Ouchi (1980) says that the clan is the alternative when bureaucracy and market (price mechanism) do not work. Based on our case study, we will argue for a different kind of understanding. The analysis proceeds by taking a close look at the various forms of control present in a global management consulting firm. Drawing on observational data, we attempt to demonstrate how control forms interact, occassionally merge and, sometimes, contradict each other. The paper ends with a discussion of how forms of control may interface with, rather than substitute for, each other.

The case Method As pointed out by Jo¨nsson (1998), most management accounting research lacks empirical input, and the empirical material typically drawn upon tends to ‘‘be limited to quick survey studies which fit into the publication requirements of the main stream’’ (p. 411). Jo¨nsson suggests the use a more diverse set of research methods, to be relevant for understanding managerial and organizational work. Consequently, this study draws upon a broad set of practices for data generation. The study has been conducted through interviews as well as observations of a variety of organizational gatherings. Observations includes following a project group at work during two days, observations of training sessions, competence group meetings and the yearly meeting for everybody in managerial positions. We have also covered events where members of the organization have interacted

428

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

externally, as when presenting the company to students. We have also studied internal documents and learning materials such, as manuals for giving feedback, project methodology, etc. We have conducted 52 interviews with 45 persons. People from all parts of the organization have been interviewed: the CEO, partners, consultants on various levels of seniority, support staff, newly recruited organizational members and so on. We have focused on what in ethnographic vocabulary is labelled as ‘studying up’: we have, in relative terms, focused more on experienced people than on fairly recently recruited people, which, in numerical terms dominate the firm. Fieldwork has been conducted in a way that is, as far as possible, open and emergent. Openness is something that one must struggle for, e.g. through thinking critically about and broadening the perspectives, metaphors and vocabularies the researcher is considering before producing an approach to interpretation. The study draws on interpretive principles, but with a slightly criticalsceptical edge. Rather than codifying empirical material, this is looked upon as text, in which one tries to go beyond the ‘surface’ and look for something less obvious, or less easily revealed in a (quick) coding process, and where also the text as a totality is born carefully in mind, which means that variation and contradiction is taken seriously. As our understanding of the field has developed, our lines of inquiry have followed suit. For example, findings and understanding from our first 20 interviews were organized in emergent themes, which were used as input in new interviews, both in terms of questions asked and whom to talk to. In this way we have been able to refine our understanding of the themes that have emerged, without providing excessive a priori closure to field-work practices. On the company Our case company is Global Consulting, one of the words largest management consultancy companies. We have primarily studied the Nordic subsidiary, employing roughly 800 people. The company has grown rapidly, is very profitable and

has generally a good reputation. As one interviewee says, when asked why people consider using the very expensive Global for a task, ‘it is never considered wrong to ask us for an opinion’. The company stands for quality, reliance, professionalism and predictability. The key organizational unit in Global Consulting is always the team. Every project and customer contact is constructed around a particular team. Thus, management control in Global Consulting revolves generally around team management. As a consequence, the individual is typically rather insignificant, at least as an organizational resource. On one level the individual is highly visible in the organization: it is the individual who is evaluated, rewarded and punished. But on another level, the individual is hardly visible at all. Instead, the individual is viewed as perfectly exchangeable and possible to replace, at least on junior and middle levels.

Governing bodies: partner, organizational and project control At Global management control is exercised from a number of distinct levels or domains. First, control is exercised from a partner level, which in this context covers the dominant stakeholder’s ways and means to orchestrate the company’s efforts in line with their collective interests. This includes an interest in financial and operating performance, but also strategic issues, such as change and developments of the business model. Other forms of control are exercised through HRM practices and methodologies for project work aimed to promote the correct behaviour. There is also control being exercised at the level of projects, which includes everyday practices to secure that projects deliver what they are expected to deliver, within the budget and on time. People in the company also talk about these arrangements in terms of culture: as a ‘control culture’—referring to the detailed monitoring of costs and revenues—as a ‘feedback culture’ and as a ‘delivery culture’. All these ‘cultures’ can be described as sets of understandings and values intertwined with formal structures and monitoring practices.

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

The role of partners: frame control The company is, as most other large companies with a fairly low ratio of partner—total workforce, strongly ‘partner-centric’. The partners have—or are at least perceived to have—a strong control over the business. The CEO describes his task as monitoring market opportunities and trying to match resources with these, to relocate resources from a market segment with over time diminishing importance to one with potential. This means mainly to see to that we have partners at the right place. If you get that segment to work, the remaining part of the forces to group themselves in the right ways. So if we move a partner from one area to another, a number of other resources—competence—will follow this move. The employees will not follow automatically or instantly, but if one partner moves to another area, businesses and sales will follow and gradually about 20 people will join him or her. Partners control mainly through sales and assignments and deciding upon the terms—in particular the margin—for projects. We can talk about sales-based control, meaning that part of the control is exercised via the promise made to the customer. Partners have a strong incentive to sell. They are assessed on a number of criteria, but the volumes they have sold during a year seem to be the major variable. It determines how many ‘units’—i.e. shares in the company—they are allowed to purchase next year, which in its turn is the determinator, together with the overall profits of the company, on their future income. There are, within the company, complaints about partners being too one dimensionally sales-oriented and even sometimes greedy. One interviewee talks about ‘a horrible prioritisation’ on selling projects even at times when the company had difficulties in recruiting sufficient number of personnel and that there was an agreement that recruiting and retaining was the critical issue. The partner group exercises control mainly through getting assignments and contracts with clients, making the client agreement the frame of

429

control for subordinates. Project managers must agree on the terms and to do the job with a particular amount of resources allowing for a particular (high) margin. The partner group exercises control over subordinates through the client in combination with demand for and close monitoring of project margin. This combination can be labelled frame control, in which the project manager, in particular, is ‘boxed in’ a frequently fairly tight delivery and margin contract. This is not exclusive for the company here studied. O’Shea & Madison (1998) report how employees at one consultancy company felt depressed every time they won a new contract: ‘how in hell will we be able to deliver all this within the budget’, they felt. Partners are experienced as rather distanced from the broader groups of employees. Junior employees feel that ‘they are very small’ in relationship to partners and do not know what their superiors do. Partners are also described as demanding and sometimes closer to a counterpart in a business relationship than as co-workers. They are also, however, seen as open and receptive to good ideas. There is some internal mystification of partners, partly a consequence of their centrality in selling and ability to control the careers of people, but the perceived distance and lack of knowledge about what partners do and how they are seem to fuel fantasies amongst the junior employees. Both partners and juniors agree on there being little interaction between top and bottom. The scope of leadership is thus limited. Partners control through formal position, and a combined bureaucracy-market mediated form of control, not through being ‘leaders’. Nevertheless, partners do have a strong grip of business, managing a workforce partly through controlling sales and business contracts, monitoring project work and performances and controlling the career options of the people eager to move up along the career ladder. Organizational control: hierarchical levels, division of labour, outputs and the HRM system The organizational dimension of control at Global is characterized by a fine-tuned hierarchy, standardized work procedures, output measurements and the HRM system.

430

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

Hierarchy is highly visible and pronounced at Global. It is perceived to express genuine difference, both in terms of say-so and power, and in terms of experience and competence. The steps include analysts, consultants, managers and partners. People tend to take these career steps and the titles very seriously, meaning that the level of social differentiation is strong and people are focused on their relative position. The hierarchy is further elaborated with the practice of designating total number of years at the position as a marker of seniority. Work processes are standardized in work methodologies. There exists a unified package of methods that consultants are expected to use in projects. Due to the fact that most projects differ somewhat from each other, the methodology is not expected to be used as a prescription, but as a resource. Global have invested heavily in systems for knowledge management. Knowledge management is primarily seen as a way of taking advantage of the scale and scope of the accumulated experience generated within the firm. Another key aspect is to diminish the importance of the individual and of personal experience. The idea is that ideas and experiences developed in one project can be recycled and reused with minimal adaption in other projects. The premise is that experience can be codified and rationalized in a way that suits database storage and retrieval. In the company there is a very strong focus on outputs and the delivery of assignments on time. Work procedures and management styles are typically geared towards giving top priority to the task at hand. This focus on the task is typically viewed as a strength since it contributes to make Global known as a reliable and trustworthy business partner that delivers what is promised. Irrespective of what kind of internal issues that are planned and booked since a long time—if we are needed at the client’s place this is what guides us. And this may create a hell internally, to put it straight, but this priority is very strong’. (partner) The company is well known for its capacity to deliver on time. This is an outcome of the amount

of resources in the company making it possible to put in extra people if needed, but perhaps even more important is a broadly shared mentality around the significance of delivery on time. The company has a strong ‘delivery culture’ making holding deadlines almost a sacred ideal: Something that penetrates the culture here and that you always steer towards is the deadline. A deadline is a deadline is a deadline, and you may almost die before you don’t deliver on deadline. This ‘holy’ nature of deadlines means that quality or ambition level may be adjusted accordingly. According to one interviewee, in projects ‘quality is the absolute most important and then finances may come in secondly’. In the same interview, he also expresses a different position and says that In a systems development project, the system needs to work and there you can’t cheat but if it is a matter of organizational development or personnel education then you can always modify how deeply and how broadly you do things and then you can cheat a bit so that you keep the plan. Deadlines are an important vehicle for control in two ways. One is related to customer control, locating the locus of control to the client and the company’s—the project members—obligation to fulfil the contract. The other is related to financial margin. Through the emphasis on deadline, people are under strong pressure to complete the project so that costs for people’s calendar time are under control and their labour are available for new assignments. As people are not paid for overtime—minor compensations such as a free travel or a nice dinner are not that costly—the strong emphasis on deadline/delivery facilitates the accomplishment of high margins in projects. Delivery culture is thus much more—and less for that matter—than being client oriented. This is illustrated by our observation of a project group working until midnight to complete an internal development project, i.e. without a client waiting for delivery.

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

Financial control is tight and elaborate. Project managers put much effort in budgets, as will be developed in more detail below. Budgets are monitored and reviewed in a systematic fashion. Several persons work exclusively as project controllers. Organizational members typically views the systems as successful, although there are sentiments that voices concern that the systems may be excessive: ‘‘We have very good financial control. We can account for every hour and every penny. We also tightly control the business development process. There are not many pennies that vanishes without being accounted for. The backside is that our fiscal focus tend to hamper creativity and freedom. And you end up in red tape and policies instead of thinking for yourself. And that’s our worst enemy.’’ (Partner) The HRM system at Global consists of systems for recruitment, promotion, evaluation and development. Recruitment: selectivity and standardization. Almost everybody working at Global has an academic degree. Consultants are mainly recruited directly from the larger Swedish Universities. All employees, and even partners, are expected to take part in various recruitment efforts, such as presenting the company at Universities, interviewing job seekers, and generally looking out for people to hire. Recruitment is part of the control structure, as it contributes in producing a fairly homogenous work force, which presumably shares certain predispositions and thus is inclined to act in anticipated ways. Promotion and career structure. The Firm is generally understood to be a career company. Initial advancement is expected to be swift for the individual. Employees are expected to advance within 12–18 months to analyst and within 2–4 years to consultants. They then become ‘‘managers’’, i.e. they get this title and gradually are functioning as project managers. After the manager level, advancement becomes more difficult. Those not being promoted are expected to leave the company in due course. The career structure

431

makes expectations on individuals explicit and the pace of promotion regulates, and is seen as directly linked to, individuals perceived degrees of success. Thus, promotion and career structure provide control in two ways: by prescribing the individual’s preferred career path and grading its relative success. Appraisal and evaluation systems. The employees at Global are under constant performance evaluation. Sometimes people in the company talk about a ‘‘feedback culture’’, but evaluation and feedback are fairly strictly formalized. Evaluation is organized in two main processes. First, employees are evaluated in relation to their individual development. This process is labelled c-mapping and is carried out three to four times every year. The employee’s nearest boss, usually the project leader of the project on which the employee currently works, primarily evaluates the employee. The general idea is that everybody should be evaluated in similar ways and according to similar criteria. Thus, there are several tools available—policy documents, forms, and standardized software—to ensure that everybody is treated in a fair and unbiased way. Second, employees are ranked by their superiors in a process labelled banding. Banding occurs once a year. The employee is ranked in category 1, 2 or 3. Band 1 is reserved for top performers. Category 2 is regarded as acceptable performance. Band 3 is a warning signal, indicating that you are perceived to be an under-achiever. Banding influences compensation: salary influences, career development and perks. Development. The people at Global invest a lot of money, time and other resources in the development of the individual, including training, access to workshops and seminars. They are also members of so-called competence groups—a competence group is a number of consultants, organized around a theme (for example e-commerce or performance management) that gathers regularly and exchanges knowledge and experiences about the theme. Junior members are always paired with a senior person, who operates as his or her counsellor—which is also the official title for what is generally referred to as mentors.

432

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

Project management and control Global are typically involved in two basic types of projects: small projects that consist of one team (3–6 person) and Global projects that can include as many as 50 consultants in one single project. Obviously, project management and control differs in scale and scope between these extreme types, but the process and the tools are basically the same for both types of projects. Projects emerge as a consequence of some sort of customer demand, typically stimulated by partner’s efforts to sell more or less ready-made efforts, so called market offerings. Partners are responsible for negotiating fees with customers. Global use three types of fees: variable fees, fixed fees and value billing. Variable fees puts the risk at the client, who is billed on the basis of actual hours worked on the project. Fixed fees put the risk at Global who suggest a price for the whole project and thus take any additional cost and profit. Value billing is a profit sharing scheme, where Global is guaranteed a minimum sum and then have a share of the costs savings or the revenue generated by the project. Fixed fee is the most common pricing scheme currently. The project manager is always responsible for planning the project, in terms of manpower, resources and cost. Project managers are negotiating their budgets with the partner group who primarily looks at the margin provided by the project since partners have a direct financial stake in the company. If the partner group finds the margin unacceptable, project managers are expected to rework the plan. As a consequence, partners are not only part of the project, as individuals and as resources, but they are also operating as external stakeholders to the project, as a collective. Planning and staffing. Budgets are produced by estimating the time span and the manpower needed. Since the partner group focuses on the margin provided by the project, resources are transformed into cost-estimates, in order to produce an estimated gross result figure. This is basically done by multiplying the manpower needed with its standard cost. There is a worldwide model for this calculation. The model uses head count and a standard price per individual cost as the

independent variable. The standard price per individual will vary, depending on seniority and level of competence, thus creating an incentive to staff projects with more junior consultants. The staffing is supposed to be made by the HR department that will allocate people according to the needs specified by the project manager. In practice project managers to a high degree decide about the people he or she wants to use. For juniors, it is viewed as important to have good relationships with what they perceive as good project managers so that the latter will choose/offer them for projects that are attractive to work in. Monitoring. Projects are monitored by project managers and project controllers. Monitoring is facilitated by various information systems, which allow project managers (and controllers) to monitor the numbers down to single individuals. Project controllers are external to the project. In essence, they are the partner groups extended arm into the project. Project controllers also provide some support to project managers but mainly operate as a controlling body. Deviations are dealt with as they emerge. There is, to some extent, a certain slack in budgets, which is used to manage expected deviation. Typically, if deviations can be accounted for in a rational and under stable way, it is claimed that there will be no negative effects for the project manager: You always inform your superior’s if it becomes clear that you are going to bust the budget significantly. And when the project is finished you’ll make a fee adjustment. You recalculate the project and point to the deviations, which occurred because of that and that, and so on And sometimes you will have some one from way up in the hierarchy, a global boss, that calls you and wonder why you didn’t provide better forecasts and plans. But I have never have any problems, nor have I heard about anybody else having problems, if you can explain why the project didn’t become a success. Evaluation. Project evaluation is mostly conducted in cooperation with customers, in particular

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

when it comes to evaluating the relative success of the implemented solution. The impression of success is important for market reasons. As a consequence, the team from Global, and the project manager in particular, typically tries to talk down customer expectations initially in the life span of the project in order to exceed them with the delivered solution. From an internal point of view, i.e. besides the client being satisfied, projects are exclusively evaluated in financial terms: to achieve or exceed the projected margin will lead to a star in the protocol. If the project manager has dealt with the subordinates in a way leading them to be frustrated or leave, this is not monitored, although occasionally it may be noted. Projects that have been viewed as particularly valuable form an experience point of view are also documented and added to the company knowledge management system. The knowledge management system thus mainly consists of write-ups (usually in the power point presentation form) of projects, which then is re-used as inputs in market offerings or new projects. Project outcomes thus provide inputs to further frame control. Excesses of control: holy margins, fake time report cards, and the ‘‘ghosting’’ phenomenon As is evident by the account above, the control systems at Global are elaborate and manifold. Since control is exercised in so many ways, from so many directions and on so many different levels, control systems, taken together, sometimes create situations where individuals find themselves with conflicting sets of instructions. As we will argue here, such situations are highly instructive, because they set members, as well as researchers, straight on the priorities in the company. The control system at Global will, in fact, regularly create a situation for organizational members where they almost routinely will provide the control systems with fake numbers. To understand what they are doing, and why they are doing it, will, we claim, (a) provide insight on how control actually is performed at Global, rather than nominally, and (b) provide perspective on how numerical forms of control may be re-interpreted in terms of socio-ideological control.

433

We stumbled on this peculiar phenomenon when we participated in a management conference day, where most of the upper echelons of the hierarchy at Global’s national subsidiary were present. An excerpt from the field notes [written by the second author, who participated in the event] follows below: (The excerpt starts at a demonstration of the new balanced score-card application) Thereby the bottom line is dealt with and the managing partner (MP) directs the attention towards the ‘‘softer’’ performance indicators—he concentrates on personnel turnover and overtime reporting. Ludvig, a senior manager, raises his hand and asks simultaneously: ‘‘What’s that number by the overtime?’’ I glance at him. He looks slightly annoyed. MP takes a step back and says: ‘‘Yes, there could be a lot of talking about that number. It’s not correct of course [the number says that there is 2% reported overtime]. The number creates some problems for us in the relationship with the other offices since they could get the impression that we are not working [overtime]. But we know we do. So the reason is of course that the reporting is incorrect.’’ Ludvig: ‘‘Are you going to do anything about it?’’ ‘‘Yeah, it’s not so easy to know what we ought to do. This started to happen after the overtime compensation was taken away. And, as you remember, we took it away because it could create absurd consequences. There were cases when an analyst made more money than more experienced managers. But the issue is complex.’’ Partner: ‘‘It would be very unfortunate to reintroduce the overtime compensation’’. Voice in the audience: ‘‘The problem is that there are no consequences for those who report too little overtime. It encourages the wrong behaviour since it is easier to write ‘8 hours’ no matter how much you have worked, than to keep track of your hours.’’ A short discussion concerning the time reporting begins, and, essentially, there are no

434

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

more concrete statements. But the partners are throwing glances at each other all the time, and MP, in particular, is frequently looking in the direction where the partners are. It is obvious that the issue already has been raised among the partners, and it is equally obvious that they do not want to discuss the issue in this forum. After lunch there is a division into different groups based on themes suggested by the participants. One theme concerns the ‘‘overtime problem’’. I decide to participate. It turns out to be a small group. After some disorder we end up being four persons in the group; two women, one man and I. I ask them to explain the problem. It comes out that there is a very good reason for Global-consultants to report incorrectly. Their managers demand that they do not report more than eight hours a day; irrespective of the number of hours they actually work. The reason is very simple. The time reported controls the most important ratio/*measure for the evaluation of the project, the project margin. The higher the amount of reported hours, the lower the margin. But this is just an internal ratio. Most projects are sold based on a fixed price. They thus know exactly how much the revenue will amount to. And since there is no overtime compensation they also know how much the expenses will amount to. The actual margin is thus not affected by the time reporting. But a consultant appears much more competent if she or he does not report more than eight hours a day. This, however, does not mean that they are not working more than eight hours. To work more hours than reported is so well known that there is even a name for it: in Globaljargon it is called ‘‘ghosting’’. After having informed me about the situation, Helen and Nina start, in the spirit of good problem solving, to analyse the different components of the problem. They use a large sheet of paper and start to write down points. The headline is: ‘‘Why don’t people report their time?’’ Nina and Helen bring the most obvious points to light:

 It is easier to write eight hours than to report in detail  You want to seem good and effective  The margin is controlling the project  Employees are loyal to the project Meanwhile, Nina is telling me about a colleague who writes her time reports for four months at a time. Helen: ‘‘I write the correct time on the first report. Certainly, there is a manager who tells me to write eight hours instead. I think: ‘Fine, I don’t care’.’’ She continues: ‘‘The whole thing depends upon the fact that there are no incentives to report the correct time. Who are you loyal to, really?’’, she asks rhetorically. She answers the question herself: ‘‘In the first place you are loyal to yourself. In the second place your loyal to the project, and only in the third place are you loyal to the firm.’’ Helen’s loyalty hierarchy is written down. Nina: ‘‘I think people would report the correct time if there were only one possibility to save and allocate time. Now this is handled entirely informally. If I need to take an afternoon off and I have ‘saved time’ so that I can take the time off without problem, then there has to be damn much negotiating to make it work.’’ Helen is unsure if it would help: ‘‘I don’t think there’s going to be any change until there is something in it for the individual.’’ Nina: ‘‘But I think this kind of system would be good for the individual too. ‘‘ Helen: ‘‘I don’t believe the individual would think it’s enough. There must be an incentive. I can’t see the incentive in this.’’ I mention a few words about the time report systems that they have in larger companies. Nina is listening with interest but Helen seems to be sceptical about the benefits of that kind of system. Sven looks thoroughly at the points: ‘‘But there has to be more to this. Why does the margin have to be so high? Why do people have to show off?’’ Sven, Nina and Helen discuss that question for a while. They agree that it is a side effect of the evaluation systems. People want to

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

appear to be good and effective because it increases the chance of getting good evaluations. The project managers, on the other hand, want a high margin, because the projects are evaluated based on their achieved margin. The flipside of the coin is that people work themselves to exhaustion and that they are actually selling themselves with a discount. Sven: ‘‘If we use the hours as the basis for the price, and assume that we work 40 hours a week when we actually work 60 hours, then we are selling ourselves with a 25% discount.’’ The others give a nod of assent. Nina adds: ‘‘But the problem is probably a different one, actually. Do we really want to sort this problem out?’’ In the meantime, our little group is slowly growing. After a while we are about ten persons who enthusiastically participate in the discussion. One of the newcomers says: ‘‘I think something has to be done, but I won’t suggest anything. I got burned on the time reporting thing. In my first project there was a crisis and we worked like crazy. Sixteen hours a day, at a minimum. I tried to report all my hours but the manager couldn’t accept that I reported more than eight hours. Then I worked eight hours. The manager got furious and had me executed in the next evaluation. The others look at him as if they couldn’t believe their ears. ‘‘You worked eight hours? And you’re still in the firm!?’’ ‘‘Who was it?’’ someone asks curiously. ‘‘I don’t want to tell’’, the newcomer replies. Helen throws a sceptical and evaluating glance at him: ‘‘I think you would have been more clever if you had worked.’’ The discussion starts to become technical. It is clear that there are quite a lot of internal systems that are built up around the time reporting. I silently wonder how they work since nobody seems to report the correct time. Helen is expressing this as she once again wants to stress the argument that the individual has to get something out of a possible change: ‘‘Everybody is ghosting. Everybody. And everybody will be continuing until they get a good reason to quit. I will be continuing

435

it they [the project managers] ask me to. And they will because their evaluations depend on the margin, which in its turn depends on the hours. I don’t mind. I don’t give a damn.’’ It is starting to get rather crowded in our little room. Janne, the person who is in charge of this particular activity during the day, has arrived. Except for him, there are few senior persons. They don’t seem too eager to discuss this issue officially. Janne, however, is senior enough to know what it looks like on the other side. He has been working as a project manager and someone asks how he is dealing with this. ‘‘Well, it happens that I say we write eight hours.’’ ‘‘Irrespective of the amount of hours we put in?’’ ‘‘Yes. It has to do with the fact that you don’t want to go to the partners with a margin that is way under the budgeted one. And it’s damn hard to get a project if you don’t have a good margin. So in order to keep the margin you tell people to write eight hours.’’ ‘‘But everybody knows that is can be manipulated’’ ‘‘It doesn’t matter. That’s the way we do it.’’ At the assembly afterwards, representatives of the discussion group present an idea including skipping reporting the number of work hours but after some resistance of senior people, this idea is dropped. From a control point of view the excerpts provide many illuminating insights. First, it demonstrates the hegemonic influence wielded by the partner group, and thus the significance of hierarchy. Although partners may have disagreements in private they operate as a single entity towards the rest of the company, at least in public. This is, in particular, demonstrated in the first excerpt, where it is quite clear that the partner group will not discuss ‘‘the overtime problem’’ outside the partner group, and also, perhaps more vividly, in the last, where partner intervention effectively puts an abrupt end to the discussion on ‘‘the overtime problem’’ at the conference, at least publicly. In fact, ‘‘the overtime problem’’ is in a sense created by the partner’s hegemonic exercise of control, through choosing to almost exclusively

436

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

focus on one (of many) key performance indicators: the project margin. Although project managers and others may have several performance indicators at their disposal, it is quite clear that the partner group pay most, if not all, attention towards the project margin. This focus has ripple effects. Since it is important for project managers to look good, in order to get good evaluations, they are disinclined to do anything that will ‘‘destroy’’ the margin. Since project participants have fixed salaries, and will not be compensated in relation to actual work hours, ‘‘ghosting’’ provide a mechanism for project managers to both preserve the project margin and manage uncertainty through making it possible to extract more work hours, without incurring costs. For project managers, ghosting makes perfect sense. Second, the excerpt demonstrates the extent of subordination and compliance expected by project members. ‘‘Ghosting’’ is clearly acknowledged as a common practice, to the extent that it, at least for some participants, seems to be routine. But ‘‘ghosting’’ is also clearly a practice that is dubious from a moral point of view, since it involves lying and faking. The moral aspect is highlighted by the person who refused to fake the numbers, and insisted on working the reported hours. His manager’s reaction, and the audience reaction in the excerpt is instructive: The manager ‘‘executed’’ him in the evaluation, and the audience is stunned, partly because he, as some one puts it, is still at the firm, partly because, as another one puts it, it was the wrong thing to do: he should has solved the dilemma by reporting the required hours, instead of working the required reported hours. Obedience is the norm. Third, the opportunistic mindset, doubtless at least partially an effect of HRM practices fine tuned towards career success, is literally spelled out in the excerpt. This is, of course, highlighted most visibly by Helen’s ‘‘hierarchy of loyalty’’ but also in the emphasis on making it worthwhile for individuals to report the overtime correctly. It is asserted that nothing will ever happen without the proper incentives for individuals. Ghosting is a practice that fits rather nicely with an opportunistic and instrumental mindset, in particular when supported by the systems of evaluation at Global.

To help the project manager to look good will undoubtedly help your project evaluation and consequently your reputation. Your cost is negligible, and mostly a matter of conscience. As a matter of fact, it might hurt you not to cooperate, as demonstrated by the participant who was ‘‘executed’’ in his evaluation when he refused to fake the numbers. From a narrow management accounting point of view, it could perhaps be argued that fake time-reporting does not matter that much, as it is not affecting financial costs, as neither revenues nor wages are affected by whether an employee does a particular job in 40 or 60 h. But although not registered by the financial measurement system, there are huge costs involved, if a cost is related to the use of ‘real’ resources. On average, the overtime is probably 30–50% and not 2%, but it varies much between different projects. One can here talk of the alternative costs, as the time spent on ‘high-ghosting’ projects to some extent could have been utilized on other projects or on competence development, etc. This is hidden as it may appear as if all the overtime is for ‘free’, and is so for the project manager. There are, of course, also costs following from the high turnover that pressure to work a lot of overtime leads to. Other problems, with more indirect consequences for the company’s profits, are mentioned in the interview excerpt.

Interpretations Movements of control: interfacing technocratic control, socio-ideological control and self-control On the whole, the level of compliance at Global is high. People generally seem obedient and hardworking. They deliver on time, provide high margins on projects and avoid making clients dissatisfied. It can be seen as an outcome of the fairly elaborate systems and practices for behavioural and outcome control, as a sign of the victory of formal management control. However, we do not think this exclusively or even mainly should be interpreted as the triumph of technocratic control focusing on behaviour and output. There are

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

considerable loopholes in the structures and systems, as reported by the time reporting practices. But also to the extent that the formal systems work more accurately in relationship to their targets, it is not clear to what extent they are precise or powerful enough to assure the prescribed behaviour materializing. A possible consequence of sub optimal behaviour is poor assessment and a low ranking, which could affect the wage negatively in annual revisions. But this is not a fine tuned vehicle for behavioural modification. It could be argued that partners, through two vital sources of control can create obedience: the careful monitoring of results and the right to decide who will become partners. But project managers are not simply dependent on partners in any one-dimensional way: partners rely heavily on the project managers to deliver the goods. And juniors may not feel that they should not necessarily work very hard so that the project manager should appear good in front of partners. Deviations from the prescribed margin may mainly fall back on the project manager, and be of moderate concern to the other project members. A moderate, even high sense of loyalty and the desire to do a good job, and a moderate fear of punishments (delayed rewards) from above, would not be sufficient to account for people sometimes working 60–70 h per week. The hierarchical structure in the company does, however, seemingly hold a strong grip over people. The employees are not that extremely dependent on the employer, but could easily protest with their feet: exit is one possibility. Given that they are highly educated people with generally a high self-esteem and excellent exit possibilities, one could imagine that they could also protest through voice. One could imagine a group of employees complaining to senior people about project managers trying to get them to fake time reporting. Managing a project that does not meet the margin can often be explained. Every project manager must have access to a variety of specific circumstances that can account for less than optimal or below calculated results: unexpected complexities around the task, employees quitting or becoming sick during the project, problematic customers, etc. As one informant says, if you just have a good

437

explanation for additional costs, you are typically excused. But this option is seldom used—which makes the company different from most other companies, which are less religious about missing a deadline or a budgeted margin. As the technocratic control systems hardly in themselves can fully account for the extreme receptivity to and compliance with goals and rules, we need to go further to ask how this is accomplished. The technocratic production of norms and meaning In order to understand the high level of compliance at Global, we must go further and investigate the ideas, expectations, values and identities that are created at the company and contribute to a high level of responsiveness to demands from seniors and management control systems. To some extent, these can be illuminated through the concept of socio-ideological control, but we also feel that this is insufficient and that we need to go beyond it. A number of distinct expressions of socio-ideological control are easy to detect at Global. Meritocratic ideology puts strong imprints on the beliefs and identities of people in the company. This overlaps with performance anxiety, a strong emphasis on career, instrumentalism and elite identity constructions. Titles and promotion speed are strongly valued. All this seems to contribute to an extremely strong inclination to put in long working hours. It is also connected to, indeed a key element behind, a very high level of compliance. In many other organizations, highly educated people with considerable status and self-confidence give some weight to autonomy and professional ideals that at least sometimes lead to tensions in relations to management control arrangements and corporate demands on subordination (e.g. Covaleski et al., 1998; Empson, 2001; Robertson, 1999). At Global there is very little of this. People subordinate themselves to corporate demands and/or the wishes of superiors. In Global there are strong claims about an elite being employed. They charge clients huge amounts of money for their services, being convinced that the client gets a lot for the costs. The large amount of corporate advertising and general

438

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

emphasis on nurturing corporate image contributes to the production of a salient social identity associated with corporate belonging and even a heightened sense of corporatively mediated self. Title, formal position, signs of one’s relative worth, here strongly matter. On the other hand, the fairly elaborated control systems and high pressure on—or internationalization of—subordination to the corporate regime is at odds with an elite identity. There are uncertainties and tensions around worth, leading to a strong inclination to draw upon resources for securing a stable sense of self, such as receiving promotion before one’s peer group and look good in all assessments. Self, socio-ideological and technocratic control thus go together. Rather than assuming that these exist in pure forms, the interfaces of control forms need to be emphasized. Values that Global employees adhere to include money, formal status and brand name—all external and highly visible signs of status and success. Growth is very important and partners have strong incentives to increase the sales. Not much energy or skills are invested in socio-ideological control as a special area of impact. Global exhibits very little of the characteristics of Mintzberg’s (1983b) missionary organization and there is not much of a corporate culture ‘outside’ regimes of bureaucracy and output control. However, the behavioural and output oriented forms of control are supported by ideas, beliefs and values indicating what exists, what is good and what is worth striving for, i.e. ideologies (Therborn, 1980). This tends, however, to be subtle and expressed within the use of technocratic forms of control. We suggest that these forms of control function in a different way than commonly assumed. Their accuracy in terms of measuring and regulating behaviour or accomplishing certain performances through output measurement is limited. So is at least the case with the accounting systems built around or affected by time reporting. This is also the case with the HRM system which hardly is capable of creating a differentiation and filtering system leading to the perfect hierarchy, in which everybody is located according to competence. When pushed in interviews, people’s perception of

promotion and how people are placed in different ranks do not correspond that well with this ordering logic. The impact of the HRM systems and practices is not necessarily only or even mainly about locating people on the right career level and improving them. The impact may be as much in terms of affecting the motivation to perform in ways compliant with the control system. The motivation is thus moulded after the regulative powers of the HRM ingredients. Competence is vague and hard to assess. The same goes for motivation, but interviewees’ reports and also our direct observations indicate very long working weeks. So irrespective of how the company affects people’s competence, people make strong efforts in order to get favourable indications of their value in assessments. The KM system also seems to work partly in other ways than as a rational system for the storing and re-use of knowledge. It does not primarily works through the re-cycling of knowledge or the use of earlier experiences and knowledge inputs as templates for a new project. The KM system emphasizes that people may be contacted for discussions and aid. People are expected to be able to phone or e-mail other people within the company and if these can assist, they do so (Alvesson & Ka¨rreman, 2001a). Similar to the HRM system, its workings are mainly through producing certain orientations and values. There are good reasons to emphasize the symbolism of the technocratic systems and procedures. They communicate certain ideals. They draw attention to certain dimensions and frame consciousness in a particular way. The intensity of HRM activities does not necessarily deliver the goods in any rational sense: a highly competent workforce, carefully distributed according to ability. But all the systems, procedures, acts and talk about feedback, assessments, the frequent promotions, fine-tuned hierarchical differentiation system, and the people improvement activities seen as closely related to these—education, self-improvement efforts—signal the presence of an impressive apparatus for the screening of highly competent people, the improvement of them and the rescreening of them at frequent points in their

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

careers. HRM symbolizes people development and indicates competence. All this HRM means that hierarchy is viewed as a reflection of competence, nothing else. People thus connect their self-esteem and ambitions closely to the HRM systems and the organizational hierarchy. They are more inclined to accept hierarchy and comply with it than in situations where it is not made credible by the HRM apparatus and the meritocratic ideology it simultaneously draws upon and reinforces (Alvesson & Ka¨rreman, 2001b). Normative ideals as templates for self-understanding and identity construction To some extent associated with meritocracy is an orientation of comparison and competition. The company employs a homogenous group of people in terms of age, educational background but also stated motives and attitudes towards work life. These people are then placed in various categories (1, 2 or 3) and promoted with different speed. The combination of the recruitment policy and a meritocratic ideology fuels competitiveness and strong focus on comparisons. This forms an important motive for performing and, in particular, getting signs of one’s relative worth from superiors. As there is a general faith in the assessment and promotion system, and self-esteem is tightly connected to its outcomes, people tend to want to give favourable impressions to their superiors. The contradictory experiences of the individual—highly visible in assessment, rewards and punishments combined with a view of the individual being exchangeable in a standardized and impersonal system—also fuel a desire to secure a positive sense of self. Any signs of success are strongly valued in the light of contradictory experiences. The formal information and control systems and processes in the company interact with this corporately created self in a variety of ways: (1) The systems direct attention and encourage a particular outlook and mentality, in this way the formal control structures, although intended to target behaviour and outcomes, also exercise cultural control, (2) Construction of self is partly accomplished through subordination to control systems and efforts to employ these as resources in securing a self, (3) the

439

socialization and reproduction of a control-, feedback and delivery culture (or a set of cultures) includes a ‘naturalization’ of a set of guidelines for how to relate to corporate practices, that means an acceptance of the control arrangements. The possible conflict between integrity and confirmation of one’s self within a corporately defined domain is, for most employees, not a very serious one, although there is some irritation and need to rationalize one’s willingness to fake. As Helen says, when told to write 8 hours of working day, ‘fine, I think, I don’t care’. An intensive system of control encourages instrumentalist orientations. A feedback system pointing at a particular set of qualities/imperfections may lead the individual to change in accordance with what is prescribed. This is different than if efforts to change are grounded in intrinsic interests or hedonistic motives. The scope and intensity of structures and systems instructing individuals what to do and claiming to carefully monitor their performances and progress all draw attention to means to be used to accomplish fixed goals. There is less explicit focus on values and ideals, on the ‘pure’ ideational arena, i.e. outside formal systems. Generally in the company, a technocratic mentality is encouraged. Incentives are viewed as central. The people at Global say that time reporting is an outcome of incentives. At Global, we find that Economic Man and Woman are at work, at least according to the self-understandings of people in the company. Without clear incentives, people will fake—they will ask subordinates to fake in order to look good for their superiors and the subordinates will respond to the request for cheating in order to please their project manager. Integrity appears not to matter that much. The person that shows reluctance to comply is viewed with surprise by his colleagues. What may in other groups be seen as a moral virtue— honesty, standing up against somebody requesting one to fake—is interpreted almost as a sign of stupidity. The corporate members of Global appear as incentive-driven. Pay and promotion are very important, interviewees say. It is worth pointing to the relative value of these rewards. The pay and speed of promotion in relationship to one’s peer

440

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

group is significant, indicating the role of the symbolism and self-defining elements of pay and promotion. It is important here to emphasize that the orientations of people in working life is not given. It is not self-evident that pay and promotion are central for people or that compliance is the only route to success. Autonomy, strong social relations, interesting work tasks, integrity and selfrespect can play a role. In many companies there are conflicts between professional and corporate ideals and often people are not willing to sacrifice professional values and autonomy for increased financial rewards (Covaleski, et al., 1998). Again in other companies social climate and intrinsic motives for working in the company are emphasized and hierarchy is viewed pejoratively (Alvesson, 1995; 2000; Alvesson, Robertson & Swan, 2001). Conducting control through constructing the (controllable) person Some researchers have investigated how different forms of knowledge and practice make possible the governable individual (Miller & O’Leary, 1987). Our approach is different as we focus not on how institutionalised discourse includes a particular construction that encourages certain plans and arrangements, but seek to investigate how subjectivity ‘empirically’ is being formed. (Miller & O’Leary, like many other Foucauldians, are interested in formal texts constructing subjects, rather than the specific encounters and reactions shaping subjectivity.) It is important to emphasize how motives and selves are constructed in organizations. There are not just once and for all established needs, expectations and preparedness to rank different sources of satisfaction in a clear manner, nor is there a set of rewards to which people respond in a straightforward way. What motivates people is partly a matter of how they construct themselves—governed by various agents in the organizations they work in—and how they construct different advantages and frustrations in their work, and, in particular how they construct the self-work relationship in terms of fit, satisfactions and frus-

trations. How people, in particular relatively young persons, relate to the work world is an open project.1 The desire to ‘succeed’ is not a self evidently dominant idea. It is an outcome of social processes. So is, of course, the idea of being successful in a particular way, given a specific set of criteria and constraints. In Global success is equated with promotion speed, ranking and playing the game within the rule set promulgated by the company: including keeping the deadline and keeping the margins. As the theory of social information processes (Salancik & Pfeffer, 1978) says, different understandings of needs and indications of the elements of significance for job satisfaction are made salient as a result of influences in the social environment. Management, as well as other groups and agencies, may influence people’s understandings of their needs, their workplace situation (job characteristics), etc. In a ‘high-intensive organization’ such as Global, the social information processes are relatively coherent and persistent. The motivation for carrying out tasks and continuing in an employment are not only a matter of instrumental or even ’hedonistic’ motives (such as pleasure from the activity in itself), but the motivation may come from the meaning for the individual, e.g. in terms of the affirmation of his or her identity and collective affiliations (Shamir, 1991). Also moral obligations may serve as motives: ’moral internalization turns constraints into preferences’ (Etzioni, 1988:46). One may act in order to feel in line with one’s cherished beliefs and selfimage and also follow the imperatives of the group with which one identifies. The ’motive’ is then to maintain and enhance self-esteem and a sense of existential security in the world, a motive that in corporate contexts is exploited by and ‘satisfied’ by forms of power offering control as a sense of security (Alvesson & Willmott, 2002; Knights & Willmott, 1989). The operation of this form of power in the case of Global is salient. 1 Of course, poststructuralists suggest a radical de-centering of subjects, an ‘emptying’ of them on any substance such as needs or wants of a fixed nature (e.g. Shotter & Gergen, 1989). This paper goes some way along a de-stabilisation route, but does not adhere to any strong anti-humanist position and does not use poststructuralist vocabulary.

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

Motives are then closely related to identity and again to forms of social regulation. Identity is a certain set of understandings of what is appropriate and natural. Who am I and—by implication—how should I act? are questions answered by the constructions of identity. In Global, as in large parts of contemporary working and business life, identity gets twisted into struggling with questions such as ‘What I am worth?’ and ‘Am I performing according to standards?’. Templates such as keeping deadlines and achieving the budgeted margin must be understood in this context. One aspect of the motivation/identity interface is the inclination to maintain and enhance one’s identity and self-esteem. Another aspect means that identity influences how various driving forces are appreciated. A person who, through education, training and on-going social relations, experiences herself strongly as a professional may as a result give priority to certain values—e.g. autonomy, knowledge-development, specialization—as an outcome of the identity rather than an effect of the pure motivating power of instrumental and hedonistic sources of gratification. In ‘high-intensity’ organizations such as Global— which make strong claims on the employees in terms of time spent on and energy put into work—the identities of the personnel are constructed and reconstructed in a more powerful way than in organizations that have lower profiles and express less distinct claims on the employees. The multitude and variety of formal systems of control—affecting behaviour as well as subjectivity—contribute to this intensity and a heightened subjectivity associated with the workplace and being an organizational person. As the individual stands out so clearly as an object of frequent assessment, feedback and hierarchical relocation through systems and rules focusing on behaviour, performance outcomes, capacity and potential, feelings of personal worth and identity become closely tied to the indications produced by these systems.

Conclusion: interfaces of control In Global there are strong beliefs in the capacities of formal systems of KM, HRM and man-

441

agement accounting to deliver effective control and the accomplishment of what is viewed as desirable behaviour and outputs of various types. However, there is considerable ambiguity regarding what these systems really can accomplish in terms of behavioural and output control. It is interesting to compare the partner cited earlier claiming that ‘We have very good financial control. We can account for every hour and every penny’ with the overwhelming evidence that time reports are faked. Equally interesting is to note how people can claim that the HRM system and fine-tuned organizational hierarchy means that competence is mirrored in the social differentiation system, while still expressing doubts about the qualifications of a number of those promoted to managerial and partner levels. As we have demonstrated above, almost all control practices at Global include aspects of cultural engineering more or less geared towards organizational members’ self-definitions. Evaluations, both formal and informal, include aspects of conduct: for example, whether one has been a team player or not, whether one has been ‘‘constructive’’ or whether one is considered to be a ‘‘showstopper’’; ‘‘constructive’’ here meaning being cheerful rather than negative in ones contribution to discussions, and ‘‘showstopper’’ here roughly meaning working less hours than the rest in the team. Competence groups are occasions for intense socialization, where re-enacting Global values and meanings is almost as dominant as a theme, as acquiring qualifications. Even such a straightforward thing as the formal career structure is used as a template for self-definition. Socio-ideological control is thus intimately tied to bureaucracy and output control. It is not, as claimed by most of the literature on control (e.g. Ouchi, 1979, 1980), an alternative to the latter two, useful in situations where complexity and uncertainty make rules prescribing behaviour and the precise measurement of results impossible. In businesses like Global we have in a sense a lot of complexity and uncertainty, but this has been counteracted with efforts to create a vast bureaucratic and output measuring apparatus. This apparatus works simultaneously badly and well. The ghosting ‘problem’ indicates that the

442

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

accounting system presents a highly distorted version of the use of ‘real’ resources—people’s working time. The uncertainty involved in this kind of knowledge work means that the use of time and energy to solve problems and deliver solutions are difficult to predict, and the technocratic control systems fail to deal with it. Ghosting can be interpreted as a pragmatic response to an obvious problem. It is, however, also a problem: the practice wrecks havoc with the formal control and evaluation systems, it compromises the balance scorecard, and it makes project evaluations unreliable since it typically under-estimates actual work hours, it might contribute to frustration and cynicism and the company may sell itself more cheaply than it thinks. Other problems regard bad handling of junior people—often exposed to a lot of stress and exploitation—and the company being weak in creativity. These are well known negative sides of (machine) bureaucracies. More positively, for the partners and partners-wannabees at least, are the efficiency and business successes of the company. Customers apparently think they get value for money and the partners become rich. It is far beyond the scope of this paper to account for the company’s results, but we can point at the significance of the hard labour and discipline of the employees of Global. Socio-ideological control matters strongly here. Technocratic forms of control work efficiently as hammering out vehicles for messages that find a receptive audience in the selves gradually formed by the control, delivery and feedback cultures of Global. The systems do not function entirely— perhaps not even mainly—through precise bookkeeping of revenues and costs (including how people spend their time), through the ‘objective’ selection, developing and ranking people or through the use of earlier projects as templates for new ones. Instead, through their systematic and intensive communication, they are normatively efficient. They work in relationship to and contribute to the shaping of a specific ‘interpretive community’, positive to techniques and an instrumental logic. The magnitude of and frequency of the control systems then contributes to the production of a regulated social order and individuals. This does not, however, work in a

mechanical way according to an ‘objective’ logic. The clan-like input to this gradually increasing receptivity is important here: selective recruitment, socialization and ritualised, ceremonial forms of control (Ouchi, 1979:838). In particular, selective recruitment and a highly homogenous work force making competition and comparison important ingredients in career aspirations and self-assessments. The ceremonies of Global are couched in technocratic rather than ‘corporate culture’ forms. Still, the symbolism and powers of these forms in shaping thinking, feeling and identifying are central. We can, perhaps, talk of Global as a technocratically framed clan. As such it is probably not unique. We can also identify bureaucracy and output measurements as potential sources of hidden socio-ideological control. Again, this has a potential in broad relevance. This study therefore offers food for the re-thought of established management control ideas and categorizations. In particular, the paper questions common ideas on the existence of pure and alternative forms of control and the assumption that technocratic and socioideological controls are mutually exclusive and function in different organizations and situations. Looking at the cultural dimension of bureaucracy and performance measures—both in terms of what is communicated and how the messages are being interpreted—seems a productive way forward.

References Alvesson, M. (1987). Organization theory and technocratic consciousness. Berlin/New York: de Gruyter. Alvesson, M. (1993). Cultural-ideological modes of management control. In S. Deetz (Ed.), Communication yearbook. Vol. 16. Newbury Park: Sage. Alvesson, M. (1995). Management of knowledge-intensive firms. Berlin/New York: de Gruyter. Alvesson, M. (2000). Social identity and the problem of loyalty in knowledge-intensive companies. Journal of Management Studies, 37(8), 1101–1123. Alvesson, M., & Ka¨rreman, D. (2001a). Odd couple. Coming to terms with the curious concept of knowledge management. Journal of Management Studies, 38(7), 995–1018. Alvesson, M., & Ka¨rreman, D. (2001b). The perfection or meritocracy or ritual of bureaucracy? HRM in a management consultancy company. Paper presented at 2nd International conference on critical management studies, Manchester July 2001.

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444 Alvesson, M., Robertson, M., & Swan, J. (2001). The best and the brightest. Construction of elite identities in knowledgeintensive companies. Paper presented at 2nd International conference on critical management studies, Manchester July 2001. Alvesson, M., & Willmott, H. (2002). Identity regulation as organizational control. Journal of Management Studies, 39(5), 2002. Barker, J. (1993). Tightening the iron cage: concertive control in self_managing teams. Administrative Science Quarterly, 38, 408–437. Barley, S., & Kunda, G. (1992). Design and devotion: surges of rational and normative ideologies of control in managerial discourse. Administrative Science Quarterly. Becke´rus, A˚., & Edstro¨m, A., et al. (1988). Doktrinskiftet. Nya ideal i svenskt ledarskap. Stockholm: Svenska Dagbladet. Beyer, J. M. (1981). Ideologies, values, and decision-making in organizations. In P. C. Nystrom, & W. H. Starbuck (Eds.), Handbook of organizational design. New York: Oxford University Press. Burris, B. (1989). Technocratic organization and control. Organization Studies, 10, 1–22. Clegg, S., et al. (1996). Management knowledge for the future: innovation, embryos and new paradigms. In S. Clegg, & G. Palmer (Eds.), The politics of management knowledge. London: Sage. Czarniawska-Joerges, B. (1988). Ideological control in nonideological organizations. New York: Praeger. Covaleski, M., et al. (1998). The calculated and the avowed: techniques of discipline and struggles over identity in Big Six public accounting firms. Administrative Science Quarterly, 43, 293–327. Deetz, S. (1992). Democracy in an age of corporate colonization. Albany: State University of New York Press. Edwards, R. (1979). Contested Terrain. The transformation of the workplace in the twentieth century. London: Heinemann. Empson, L. (2001). Fear of exploitation and fear of contamination: impediments to knowledge transer in mergers between professional service firms. Human Relations, 54(7), 839–862. Etzioni, A. (1960). A comparative analysis of complex organizations. New York: The Free Press. Etzioni, A. (1988). The moral dimension. New York: Free Press. Ferner, A. (2000). The underpinnings of ‘bureaucratic control systems: HRM in European multinationals. Journal of Management Studies, 37(4), 521–540. Friedman, A. (1977). Industry and labour. London: Macmillan. Geertz, C. (1973). The interpretation of culture. New York: Basic Books. Govindarajan, V. (1988). A contingency-approach to strategy implementation at the business-unit level: integrating administrative mechanisms with strategy. Academy of Management Journal, 828–853. Govindarajan, V., & Fisher, J. (1990). Strategy, control systems, and resource sharing: effects on business-unit performance. Academy of Management Journal, 259–285. Hopper, T., & Macintosh, N. (1993). Management accounting

443

as disciplinary practice. The case of ITT under Geneen. Management Accounting Research, 4, 181–216. Hopwood, A. (1987). The archeology of accounting systems. Accounting, Organizations and Society, 12, 207–234. Jo¨nsson, S. (1998). Relate management accounting research to managerial work! Accounting, Organization and Society, 23, 411–434. Knights, D., & Willmott, H. (1989). Power and subjectivity at work. Sociology, 23(4), 535–558. Kunda, G. (1992). Engineering culture. Control and commitment in a high-tech corporation. Philadelphia: Temple University Press. Langfield-Smith, K. (1997). Management control systems and strategy: a critical review. Accounting, Organizations, and Society, 22, 207–232. Lind, J. (2001). Control in world class manufacturing—a longitudinal case study. Management Accounting Research, 12, 41–74. Martin, J. (1992). The culture of organizations. Three perspectives. New York: Oxford University Press. Miller, P., & O’Leary, T. (1987). Accounting and the construction of the governable person. Accounting, Organizations and Society, 12, 235–265. Mintzberg, H. (1983a). Structure in fives. Englewood Cliffs: Prentice-Hall. Mintzberg, H. (1983b). Ideology and the missionary organization. In J. B. Quinn, H. Mintzberg, & R. M. James (Eds.), The strategy process. Englewood Cliffs: Prentice-Hall. Mintzberg, H. (1989). Mintzberg on management. New York: The Free Press. Mumby, D. (1988). Communication and power in organizations: discourse, ideology and domination. Norwood, NJ: Ablex. O’Shea, J., & Madison, J. (1998). Dangerous company. Harmondsworth: Penguin. Ouchi, W. G. (1979). A conceptual framework for the design of organizational control mechanisms. Management Science, 25, 833–848. Ouchi, W. G. (1980). Markets, bureaucracies and clans. Administrative Science Quarterly, 25, 129–141. Peters, T. J., & Waterman, R. H. (1982). In search of excellence. New York: Harper & Row. Puxty, A. (1993). The social and organizational context of management accounting. London: Academic Press. Ranson, S., Hinings, C. R., & Greenwood, R. (1980). The structuring of organizational structures. Administrative Science Quarterly, 25, 1–17. Robertson, M. (1999). Sustaining knowledge creation within knowledge-intensive firms. Diss. Warwick Business School. Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 23, 224–253. Schein, E. H. (1985). Organizational culture and leadership. San Francisco: Jossey-Bass. Shamir, B. (1991). Meaning, self and motivation in organizations. Organization Studies, 12, 405–424. Shotter, J., & Gergen, K. (Eds.). (1989). Texts of identity. London: Sage.

444

M. Alvesson, D. Ka¨rreman / Accounting, Organizations and Society 29 (2004) 423–444

Simpson, R. L. (1985). Social control of occupations and work. Annual Review of Sociology, 11, 415–436. Smircich, L. (1983). Studying organizations as cultures. In G. Morgan (Ed.), Beyond method, strategies for social research. Beverly Hills: Sage. Smircich, L., & Morgan, G. (1982). Leadership: the management of meaning. Journal of Applied Behavioural Science, 18, 257–273. Spekle´, R. (2001). Explaining management control structure variety: a transaction cost economics perspective. Accounting, Organizations, and Society, 26, 419–441. Tengblad, S. (2001). Examining the stability of managerial

behavior: a replication of Henry Mintzberg’s classic study 30 years later. GRI-rapport 2001:6 Go¨teborg, GRI. Therborn, G. (1980). The power of ideology and the ideology of power. London: Verso. Van Maanen, J., & Kunda, G. (1989). Real feelings: emotional expression and organizational culture. In B. M. Staw, & L. L. Cummings (Eds.), Research in organizational behaviour, vol 11. Greenwich: JAI Press. Weiss, J., & Miller, L. (1987). The concept of ideology in organizational analysis. Academy of Management Review, 12, 104–116.