MANAGING ORGANlSATlONAL CULTURE: FANTASY OR REALITY?
Emmanuel Ogbonna, Cardiff Business School
IS CULTURE MANAGEABLE?
Before proceeding to examine the literature on managing culture, it is useful to briefly define the concept and its major constituents. There is no consensus on the definition of culture. Indeed, it has been noted that in organisation theory, there are as many definitions of culture as there are experts on the subject (Ogbonna, 1990). It would appear that the definition adopted is bound up with a particular researcher‘s view of the world and the preferred methodology (Burrell and Morgan, 1979; Smircich, 1983). Yet a definition is necessary to operationalise the concept and to distinguish it from similar organisational concepts.
Culture is defined in this article as the interweaving of the individual into a community and the collective programming of the mind that distinguishes members of one known group from another. It is the values, norms, beliefs and customs that an individual holds in common with members of a social unit or group. Schein (1985) identifies three levels of the cultural phenomenon in organisations as follows:
- on the surface are the overt behaviours and other physical manifestations (artifacts and creations);
- below this level is a sense of what ought to be (values);
- at the deepest level are those things that are taken for granted as ’correct’ ways ofcoping with the environment (basic assumptions).
It is the pattern of basic assumptions which are taken for granted that Schein describes as culture. This conception of culture is important for discussions of whether culture can be managed and will be taken up later.
The debate on managing culture is characterised by two extreme arguments rooted in the very conception of the subject. Smircich (1983) identifies two distinct approaches based on Burrell and Morgan’s (1979) paradigmatic framework culture as something an organisation is and culture as something an organisation has. These two stances, which appear to be mutually exclusive, have dominated not just the research on organisational culture but also explorations of the very nature of the concept.
When viewed as something an organisation has, culture becomes a powerful organisational tool. It shapes behaviour, gives organisational members a sense of identity and establishes recognised and accepted premises for decision-making. For those researchers who see culture as what an organisation is, the concept is inseparable from organisations. To be sure, ‘organisation is culture and culture is organisation’, hence there is little point in trying to control this socially constructed phenomenon which is embedded in the very roots of organisational existence. This point was illustrated by Weick when in a response to an article on managing corporate culture, he was reported to have stated: ‘organisations don‘t have cultures, they are cultures, and this is why culture is so difficult to change’ (Siehl1985125).
In a similar vein, Fombrun (1983151) has concluded that ‘managing corporate culture is ‘. . . an awesome if not impossible task’, and Uttal (1983:72) provides a summary which captures the feeling of many on the issue:
MANAGING ORGANISATIONAL CULTURE: FANTASY OR REALITY?
For all the hype, corporate culture is real and powerful. It’s also hard to change and you won’t find much support for doing so inside or outside your company. If you run up against the culture when trying to redirect strategy, attempt to dodge; if you must meddle with culture directly, tread carefully and with modest expectations.
Other researchers go further, arguing that culture simply exists and cannot be created or managed by individuals.’ The point being made by these researchers is that no matter how managers try, they cannot change or manage the subconscious assumptions and values which guide people’s behaviour. This is because, by definition, these assumptions exist without the individual’s awareness. As Krefting and Frost (1985:156) so eloquently put it: If organisational culture is funnelled through the unconscious and is therefore not always orderly, then it is unlikely that efforts to manage such a culture can be precisely predicted or tightly controlled.
However, it is quite plausible to argue that the extreme stance of this view is logically
inconsistent as well as empirically suspect. This is because culture, like reality, is a social phenomenon which depends on human action and interaction (Berger and Luckmann, 1966). For example, what appears today to be an unquestionable way of behaving was once an alternative mode of behaviour, and once chosen, it is reinforced and passed down to new members who may come to view it as the appropriate and unquestionable response. In organisational settings, it has been argued that the entrepreneur is the creator of meaning through symbols, ideologies, rituals, beliefs and myths (Pettigrew, 1979 and 1985). By implication, if the value system guiding members’ behaviour is no longer appropriate, it must be replaced by that which is more appropriate by manipulating the elements identified above. That culture is a learning process capable of being ‘unlearned‘ is a view consistent with that of Schein (1984 and 1985) who is probably the most widely cited writer on the concept. Schein’s conception of culture is that of adaptive agent which enables organisational participants learn to cope with problems of survival. This line of thinking suggests that where a culture has outlived its usefulness as a means of sustaining a people’s behavioural pattern, they learn (or soon develop) alternative ways of legitimising the old or new behaviours which may eventually translate into a new set of beliefs and values. There are, after all, many examples of ’cultural change’ in the literature, including forced change?
Herein lies the major locus of the argument of researchers who see cultural change as not only possible but desirable. The emerging theme so far is that changing the ‘deep fabric’ of the organisation is awfully difficult, hence it is sensible to advise companies not to attempt to change their cultures without serious thought. One only has to read Tunstall’s (1983) description of the obstacles to cultural transition encountered by one of the world’s largest corporations, AT&T, to know the nature of the problem involved. Tunstall provides an account which indicates technological, operational and organisational obstacles. Yet changing the corporate culture may be critical to the success of the company, as Tunstall himself argues. Failure to change the culture of the organisation (AT&T) in response to environmental pressures beyond the company’s control would almost certainly have led to disaster. A proposition consistent with this line of thinking is that in some cases organisations may have little choice but to attempt to manage their culture. Robbins (1987368) shifted the argument from whether culture can be managed to the conditions under which it can be managed. He stressed that ‘if managers cannot guide their organisations through planned cultural change, the subject [culture] has limited practical utility and may be of only academic interest’. Robbins goes on to discuss the factors that will determine the success of any attempted change in culture, including the emergence of a crisis, leadership change, the organisational life cycle, age, size, strength of the current culture and absence of subcultures. Furthermore, Kilmann (1982) observes that if culture is not controlled, organisations will spend time and money protecting their outdated assump tions about the business environment. Indeed, where assumptions are not related to the current business environment, the organisation may suffer as a result. Mercer? an IBM employee for fifteen years, takes this point further. Providing an ‘insider account’, he demonstrates how IBMs strong culture is making it difficult for the management to manage the organisation in a changing environment. Bates (1984) has made a similar point, arguing that culture is capable of locking people into their own problems. This is because necessary cultural change may be avoided on sentimental grounds or may even be made difficult by an existing strong culture. Although this point may be held to be a vindication of the argument that culture can and should be managed, it nonetheless exposes the inherent weakness in the underlying attempts to manage culture. This point is crucial and will be taken up later.
Whatever stance researchers take on the issue of managing culture, it is clear they all agree on one thing: culture can have a pervasive effect on organisational outcomes. Krefting and Frost (1985) assert that it constructs the boundaries within which individuals interact in their everyday life and as such limits the processing of information to within human capability. Luhmann (1979) has similarly observed that it has the potential to reduce ambiguity and facilitate social interaction. Handy (1986), taking a more functional view, argues that a strong culture makes a strong organisation, and this may well be so since a strong culture can shape patterns of behaviour in the organisation which may present a competitive advantage (Scholz, 1987). This agreement implies the necessity for the cultural aspect of the organisation tobe kept attuned with other organisational variables such as the environment. Often the organisational environment is dynamic and this presents a challenge to managers in managing corporate culture.
So far this article has taken the view that managing culture is synonymous with changing culture. It should, however, be noted that managing culture need not imply change (Siehl, 1985). For example, it is quite possible to legitimise the prevailing culture while some minor adjustments are made without an apparent change. Many have linked this to altering the climate of the organisation while the deep fabric of the culture remains the same. To be sure, managing organisational culture could involve creating it, changing it, mainfainkg it and abandoning it. That said, the concern of this article is the major issue of changing culture and ‘managing’ and ‘changing’ are used interchangeably.
APPROACHES TO CHANGING CULTURE
Many writers have proposed models which seek to demonstrate how to change culture consistent with the tenet that organisation culture can be managed. One of the most elaborate was posited by Silverzweig and Allen (1976) whose ‘normative systems’ model presents a good summary of the main focus of writers on changing culture. Most other models of culture change are similar to that of Silverzweig and Allen, hence onlv this model will be discussed in some detail although the links with other models will be highlighted. The description of the model will be followed by an examination of some of its major weaknesses. The model is built around the premise that people tend to form shared values when they come into a sustained period of interaction; that there are a set of norms and expectations which whilst not written, constitute a major influence on the behaviours of members; and
finally, that individuals are not merely shaped by their cultures, but are also capable of influencing and shaping the culture of which they are a part. The model thus contends that organisational behaviour can essentially be changed by altering the environment within which the behaviour occurs. The long-term consequence of this becomes a change in the value system guiding behaviour. The normative systems model which consists of four stages is described as adaptive to organisational needs.
A major step in the normative systems model involves analysing the prevailing culture and establishing the specific objectives that will lead to the desired culture. This is consistent with many other approaches to changing culture, notably those of Kilmann (19821, Oreilly (1989), Graves (1986), and Sathe (1985). To elicit the desired culture, Silverzweig and Allen recommend an extensive survey of organisational members which they hope will reveal existing norms and those that members would like the organisation to develop, and Kilmann offers a more informal method of achieving the same objective. The gap between the desired behaviour and the prevailing norms is described as the ’norm gap’ and this becomes the focus of performance objectives aimed at achieving desired behavioural standards. The objectives thus form a base for measuring progress while simultaneously motivating and co-ordinating actions towards the desired norms.
The second step in the normative systems model involves giving organisational members the opportunity to experience the present culture and participate in establishing the desired culture. This goes beyond Kilmann’s suggestion of asking people in an informal way to simply list the attributes they would like to see in the desired culture. Silverzweig and Allen demonstrate the need to involve people in what they describe as a ‘normative systems workshop’, an exercise that gives people limited opportunity to question their old beliefs and experience the positive aspects of the culture being encouraged.
The third step is referred to as ’systems installation’. This follows involvement in workshops and is described as an important stage in the change programme. To be
successful, installation requires a systematic approach to focussing on the prevailing norms to be changed. The role of leadership at this stage is highlighted and there should be emphasis on encouraging successful managers to act as role models with the hope that other employees will emulate their behaviour.
The final step is recommended to become a permanent aspect of the organisation.
Described as ‘on-going evaluation and renewal’, Silverzweig and Allen stress the measurement and effectiveness of this phase of the process. Where the desired culture is not achieved, they recommend that the change programme be renewed in one or more areas of the model.
For our purpose the most important parts of the model are the ’critical’ factors which
Silverzweig and Allen suggest are capable of bringing about a successful change in culture. These factors, which must be properly directed, include leadership, influence of subculture, information and communication systems, performance and reward systems, organisational policies and procedures, training and orientation, supervision, and results orientation. As will be demonstrated with examples, these factors are widely cited in the literature as being crucial to managing culture, and their influence as conceived by Silverzweig and Allen is explored before a critique of the model is presented.
The first of the critical factors is what Silverzweig and Allen call leadership modelling
behaviour. Their argument is that since leaders influence the behaviour of their subordinates, they should be perceived not only to be preaching behavioural change but also to be practising what they preach. Most examples of cultural change in the literature tend to highlight the impact of leadership not only because of leaders’ role-modelling behaviour but the wider impact that leaders‘ motivation and commitment could have on the change process. This is especially so when the appointment of a new leader precedes the change programme. For instance, the appointment of Lord King as the chief executive of British Airways is said to be the key factor behind the change in culture that followed (Thomas, 1985) and the new managing director at Rank Xerox (UK) was credited with the success of the company’s change programme (Chapman, 1988). By implication, organisations wishing to change their cultures should consider hiring new leaders and both new and existing leaders should be encouraged to lead by example with the hope that their subordinates will
emulate their behaviours.
Another important area is said to be the influence of work-team cultures. Here it is held that every work-team in an organisation has a culture, hence the organisational culture becomes a summation of the subcultures of all the work-teams. The organisation, it is claimed, should help work-teams to develop positive subcultures consistent with the wider organisational culture. The current movement towards greater flexibility could be conceived as an attempt by companies to break down work-team demarcations and encourage stronger identification with the values of the wider organisation.
The information and communications system within the organisation also needs to be supportive of the desired culture. The importance of informing employees of the desired change in culture is emphasised. A practical example of this is provided by Allied Dunbar, where prior to changing the culture of its administrative unit all the employees in the section received written statements detailing the desired values (Carby, 1985). The perceived importance of communication in any attempt to change culture cannot be overstated. Information that is not sufficiently or effectively communicated will tend to reduce the attention the employees would otherwise have devoted to the subject, yet this maybe critical to the success of the culture change. By implication, managers should control the flow of information in the organisation by ensuring that negative information is filtered out or changed while positive information is encouraged.
The other critical factor to the success of the change programme is said to be performance and reward systems. The factors that are appraised, and the behaviours that are recognised and rewarded will provide a mental cue to employees in deciding what actions they should take. To encourage employees’ favourable disposition toward a desired change in organisational culture, reward systems must be set in context; that is, be matched with the desired culture. The recent increases in the use of performance related pay schemes by companies attempting to change their cultures reflect the perceived importance of the reward system in changing culture. Another important area concerns the ’hard’ aspects of the organisation, including organisational policies, structures, budgets and procedures. The contention is that these systems may need to be altered since they may still be supportive of the culture that the organisation is wishing to change. Training and orientation are suggested to be taken seriously by organisations aspiring to change their cultures.
The recommendation is that the training and orientation functions should not only be supportive of the desired culture but also that employees should be
encouraged to practise what they are taught. This is especially so with induction training because most employees are more susceptible to learning at this early stage. The information provided to them must beconsistent with the culture theorganisation wishes todevelop and must be reinforced in the everyday work environment. Sparrow and Pettigrew (1988) in their study of high technology companies show that the companies place considerable emphasis on training workshops which stress the key norms of the desired culture.
The role of first-line supervisors is another area of concern. The suggestion is that they should be given sufficient support since they are best placed to deal with the inevitable barriers between the management and other employees. The supervisors may, for example, through their interaction with both sides minimise staff apathy and promote better understanding.
The last critical factor in Silverzweig and Allen‘s model is not so much an attribute to be controlled as it is a measure of the success or otherwise of the attempt to change the organisation culture. The emphasis here is on the need not only to have specific and measurableobjediveswhichcanbeused tomonitorthe progressof thechange effort but also to have the potential of converting some cynics within the organisation whose commitment will only be gained if they are convinced that change can be and is being achieved.
Although Silverzweig and Allen’s model has formed the intellectual inspiration for many writers on managing organisational culture, there remain important questions on its validity. For example, the suggestion that survey instruments may be relied upon in assessing participants’ understanding of the prevailing organisation culture must cast some doubt on the reliability of the information revealed. Indeed, it would appear that if a questionnaire is used to elicit the desired and prevailing cultures of the organisation, the results may be misleading. This is because the questions asked may reflect only the manager/consultant’s values or preconceived notions of the culture of the organisation and these may not necessarily be consistent with the culture of different subunits or indeed the wider organisation culture.
Similarly, the involvement of staff in the ‘normative systems workshop‘ may not achieve the objective of experiencing the desired culture. It is quite possible that people’s behaviour in a workshop situation may not necessarily reflect their actual work behaviour. For example, employees may be willing to cooperate initially since they are the focus of managerial and research attention. Once this condition changes, they may go back to their familiar routine and by theso-called ‘Hawthorne
Furthermore, the work of other writers who take their intellectual inspiration from the
normative model raises a number of issues on managing culture. For instance, Graves (1986) identifies a number of levels at which culture can be changed and highlights the distinction between changing behaviours and changing deeper level values which by implication is changing culture, and Schein (1985) identifies three levels of culture, emphasising the significance of the deepest level, the taken-for-granted assumptions that people hold. The question remains as to what this distinction in levels really is and to what extent it can be made in practice. For example, is the visible level a part of culture or an outcome of culture?
Indeed, could it not be that models which purport to change culture are simply extending the behavioural and organisational change strategies found in the literature? To illustrate, Sathe’s (1985) model highlights that changing behaviour is an important step in changing culture, and Graves (1986) attempts to delineate cultural from behavioural change, but behind all the rhetoric lies the suspicion that they are simply packages of behavioural change. Of course, the pundits warn of the consequence of not addressing fundamental cultural issues. For example, Sathe (1985) warns of the danger of relying on behavioural modification techniques, insisting that these would only encourage extrinsic motivation which may not lead to a permanent change in culture. However, if permanent change in behaviour is the yardstick for measuring cultural change, then it is possible that some behavioural change techniques can generate such change. Furthermore, Sathe’s line of
reasoning, while sensible, raises an important question in attempting to manage culture. For if we accept that culture can be changed, reliance on techniques which guarantee permanent or deep-rooted change may itself become an impediment to future change.
Almost all the models of managing culture emphasise the need to control variables
considered crucial to the success of any attempt to change culture. Of these, leadership has received the widest attention as the single most important factor in changing culture. For example, Katzenbach, a McKinsey partner, was quoted to have stated: ‘the interesting question is whether a company can achieve broad cultural change without new leadership. It is very difficult’ (Thackray, 1986:70). Graves (1986) suggests that changing culture requires change in leadership at thetop, apoint which is consistent with Sathe (1985),Silverzweigand, Allen (19761, Kilmann (19821, Tichy et al. (1986) and Thackray (1986), among many others.
The justification for changing the organisation’s leaders is that they epitomise the prevailing culture; they may hold strong values which block attempts to change culture. However, it should be noted that new leadership is not a guarantee that employees, as many writers assume, will conform to the new values desired by the new leaders. This assumption is flawed in three major respects. The first is that it ignores the influence of informal group norms and other forces outside the control of the individual, or indeed the organisation, and these can be powerful determinants of behaviour, especially when they are in conflict with organisational norms. Secondly, it ignores the political processes in organisations which suggest that all outside movements (including new leaders) pose a threat to existing institutional and cultural arrangements. To be sure, organisational members are unlikely to act to promote insurgent challenges to the status quo.
Finally, as Robbins (1987359) demonstrates, there are conditions for the acceptance of the new leaders’ values: the new leader must establish a clear alternative vision of what the organisation can be; there must be respect for the leader’s ability; and the leader must have the power to enact the alternative vision. In other words, the new leaders must be distant enough so as not to see things through the eyes of their predecessors, yet they must understand the organisation and its politics well enough to be able to develop alternative directions. This paradox sums up the difficulty in
Communication is another variable which has been widely cited as a necessary prerequisite to changing culture. Graves (1986), for example, stresses the need to communicate the level of culture that managers wish to change. Hence organisations are advised that when changing only behaviour employees should be encouraged not to change their fundamental values. Silverzweig and Allen (1976) highlight the importance of information and communication in modifying the existing culture, and Sathe’s (1985) model emphasises the relevance of managerial conviction and credibility in communicating the new patterns of values and beliefs that organisational members are expected to adopt. The problem here is that communication is presented as if it were something that managers can effectively
control and this is patently not the case as unintended meanings are often put across. This problem is compounded by the fact that receivers (employees) would often be biased in their interpretation of information. Communication difficulties are likely to be heightened where there are strong and perhaps conflicting subcultures within the organisation. This is because communication essentially depends on shared meanings, and a multiplicity of cultures is likely to be translated into a multiplicity of meanings which will inevitably reduce the extent to which values are shared.
Having reviewed the key approaches to managing culture suggested in the literature and raised some of their inherent limitations, it is necessary to reconsider the major question posed in this article, namely, is organisational culture manageable? As already indicated, the key argument of researchers who see the change of culture as impossible is that organisational members are seldom aware of their culture because it operates at a subliminal level of human consciousness. It is taken for granted and for the most part perceived as objective reality never to be questioned. However, proponents of the opposite stance imply that it may well be that these researchers dismiss the feasibility of cultural change because they are conceptualising change as occurring at the fundamental level of changing beliefs and basic underlying assumptions. If this is so, then there is a degree of credibility to the argument in that approaching a change in culture from this level is very difficult and time-consuming. However, it appears that the management of culture does not have to be concentrated solely at this level. In fact, judging from the comments of some of
the most widely cited writers on the subject, one has to be sceptical as to the prospect of achieving ‘deeper level’ change. As Deal and Kennedy, the authors of Corporate Cultures, put it: ‘let’s be candid about this, we don’t know this area better than anyone else. Cultural change is still a blackart as far as weareconcerned’ (1982162).
And Waterman, theco-author of the widely cited In Search of Excellence, commenting on the popularity of the book, was reported to have stated: ‘it’s not a bible you know!’ (Piercy, 19871 1). The implication of such statements is that ‘deeper level’ change is more difficult to achieve and it may be easier to concentrate on the more visible behavioural patterns. When this is interpreted in the light of Schein’s (1985) levels of culture (already highlighted), it becomes apparent that only the more visible levels of culture may be susceptible to conscious manipulation. This in fact raises the question of whether what may be managed is behavioural and not attitudinal change.
The argument is, of course, that behavioural change often precedes cultural change since culture is an adaptive learning process which responds to the needs of the people (Schein, 1984 and 1985). However, there is scepticism as to whether deeper level values (taken-forgranted assumptions) are as susceptible to change as the more visible behavioural patterns. Gagliardi (1986), for instance, questions the logic implied by the work of Schein and many others that as organisational members learn to adapt to their environment they replace their existing values with new ones. Deep-rooted values, he argues, are by definition entwined with the deepest level of human cognition and changing them is often not considered by people since such change invigorates anxiety, and people are often reluctant to try as there is no guarantee that changing values will solve organisational problems.
In other words, deep-rooted values remain intact while behavioural patterns may change in response to environmental pressures. This point is illustrated by a recent study of major UK supermarkets where managements attempted to change organisational culture. Most of the cultural change strategies described in this article were implemented over a five-year period. This was a companywide change programme but particularly directed at front-line staffs who come into direct contact with customers. The result was that check-out operators were more likely to smile and behave deferentially towards the customers but there was no evidence that they have adopted the values behind the desired culture. Indeed, it emerged that the visible manifestations of the desired culture (’smile’ and ’politeness’) were delivered under the increasing surveillance and threat of sanctions that may be visible to any inquisitive shopper (Ogbonna and Wilkinson, 1990).
The above point raises the important issue of the relationship between culture and
organisational control and shows that if culture can be changed, vested interests within the organisation can foster or hinder the change process. In the supermarket study described above, the fact that check-out operators did not share the assumptions of management made it difficult to achieve a change in their basic underlying values. Indeed, far from changing to suit management’s aims, the collectivism which attempts to change culture can breed may become counter-productive as the workers may use this to their advantage (Ray, 1986).
Other writers have noted that the results of managing culture are often unanticipated
(Krefting and Frost, 1985; Ackroyd and Crowdy, 1990), implying that the management of culture may be fraught with unintended organisational consequences. This is because although organisational culture, rather like the culture of societies, can and does change, its direction cannot be precisely controlled. This line of reasoning appears convincing when it is considered that for anyone attempting to change culture to be taken seriously, the person needs to have a vision of what the new culture will be like.
Unfortunately, many managers are often unsure of what they want to change and are unlikely to be able to formulate visions of the future, which in itself is characterised by uncertainty and ambiguity. Indeed, far from analysing and planning the future of their organisations, managers tend to be reactive, dealing with problems as they arise (Mintzberg, 1975). An important practical implication of this work is, therefore, that any manager who is seriously interested in managing culture must be certain and unambiguous about what is to be managed and what outcomes are desired. Specific policies and success levels will vary depending on the definition of culture and the changes desired.
To summarise, the literature on managing organisational culture is confusing. There are clearly two ways in which culture is treated in relation to change. There are those who treat culture as behaviour and there are those who treat it as values and taken-for-granted assumptions. The consequence of this confusion is that there is no convincing conceptual model which clearly demonstrates how change of deeper level values should be attained.
This is despite the fact that much of the literature accepts such change as constituting a change of culture. Instead, what we are presented with are haphazard treatments of cultural change which either equate it to behavioural change, or simply assume that behavioural change will in the long run lead to change in culture. Attempting to change culture then becomes a journey into the unknown which may or may not achieve the desired aim. Whatever the case, it is clear that the concept of culture has lost much of its value as a tool for analysing and interpreting the behaviour of people within organisations. It would appear that many academics, like consultants and managers, have been taken in by the search for mechanisms which will generate behaviours concomitant with the strategies of organisations. Culture appears then to have been reduced to the status of yet another concept which, like many before it, has reached the decline stage of its ‘life cycle’. Although it may never be completely exterminated, it is doubtful whether, as in the 1980s, three of the top management journals will devote special issues to it in the 1990s.