Journal of Managtment Studies 24:5 September 1987 0022-2380 $3.50 MAPPING MANAGEMENT STYLES IN EMPLOYEE RELATIONS JOHN PURCEZL Templeton College, Oxford ABSTRACT In an attempt to clarify the meaning of management style in employee relations and go beyond the limitations of the frames of reference debate, two dimensions of style are identified. Individualism refers to the extent to which personnel policies are focused on the rights and capabilities of individual workers. Collectivism concerns the extent to which management policy is directed towards inhibiting or encourag- ing the development of collective representation by employees and allowing employees a collective voice in management decision-making. Style is a deliberate choice linked to business policy. Organizations may choose to focus on one or both aspects. Not all firms have a distinctive preferred management style. Survey evidence is analysed and a number of unresolved issues raised which need to be tackled if management style is to become a useful concept in the study of strategic choice in the management of industrial and employee relations. INTRODUCTION Most students of industrial relations can readily identify well-known companies and contrast the differences between them in their employment policies and practices. Companies like IBM, Ford and ICI are frequently quoted as examples of diversity which cannot be explained wholly by reference to structural variables such as size, product markets and technology. Within the constraints of these factors, it is argued, senior managers in companies (or more likely large companies), can exercise a degree of choice in the way they treat their employees. The choices made and the underlying rationale is often described as management style. There is, however, little agreement on what is meant by the term, how style can be analysed and categorized, whether ‘style’ is synonymous with behaviour and whether it is the same as, or linked to manage- ment strategy. Does every organization have a managerial style by virtue of the need to employ people, or is the concept restricted only to those organizations where senior managers have taken strategic, considered decisions on the way employees are to be treated? Address for reprints: John Purcell, Templeton College, Oxford OX1 5NY, UK
534 JOHN PURCELL The purpose of this article is to explore the contours of management style by reference to two key dimensions and to raise certain problems concerned with classification and the units of analysis. Some data is presented drawn from the management philosophy statements collected in 1985[‘] as part of a large survey of 107 multi-plant private sector enterprises, 97 divisions within these companies and 176 establishments owned by these firms. As a word of warning, it is necessary to note that the term management style is imprecise and often confuses rather than clarifies. A great deal more research is required both in refining models and testing them, but this cannot be done until the basic parameters are identified. THE MEANING OF MANAGEMENT STYLE Many writers have sought to distinguish between different types or patterns in industrial relations among dimensions of conflict-co-operation. An important analytical distinction must be drawn, however, between classifications of the outcome of the interaction of management and labour and the attitudes, beliefs, or frames of reference of the parties, most notably management, in determining the style that they wish to pursue. Clearly, frames of reference are influenced in varying degrees by the historical experience of managing employees in the firm and by wider social and political values. If management style exists at all beyond historically determined reactive gestures implicit in the interaction of management and labour, we have to allow for an element of choice which might be more or less constrained. Thus the study of management style is not primarily an analysis of outcomes but of originating philosophies and policies which influence action. Management style is different from management attitudes. We cannot make the assumption that attitudes necessarily translate into action or behaviour. A good example here is the work of Poole and his colleagues (1982) who posed generalized statements in their attitude survey of managers about, among other things, the role and power of trade unions. We know from this that most managers are – or were at the time of the survey – distrustful of trade unions and the role they played but we do not know what, if any, policies or preferences they had on how unionized employees should be, or were, managed in their own firms. Similarly, the existence of certain structures, for example of negotia- tion and consultation cannot be taken, as MacInnes (1985) has shown, to indicate the widespread adoption of a pluralist frame of reference. Our concern is in part with how management chooses to respond to the existence of structures of collective bargaining, whether to oppose or support them. If a preference exists (and there is no reason why it should) it must be, to a degree at least, translated into policy or determined behaviour for us to be able to say that the company has a distinctive management style. It is not necessary for such policies always to lead to the desired outcome, and indeed given power relations
MAPPING MANAGEMENT STYLES 535 between management and labour (the labour process problem of double contract) we might expect there to be a frequent shortfall between aspiration and outcome. We must also allow for management style to be periodically redefined in the light of changing needs and experience. In some firms under- lying approaches might be radically altered as a new coalition of managers – or owners (for example after privatization of nationalized industries) come to power. Pragmatic, reactive responses to labour problems cannot be classified as management style. Style implies the existence of a distinctive set of guiding principles, written or otherwise, which set parameters to and signposts for management action in the way employees are treated and particular events handled. Management style is therefore akin to business policy and its strategic derivatives. Indeed management style is one of those aspects of wider business policy which ‘state in broad terms both what may and may not be done. . . [and] . . . are more often made as a result of moral, political, aesthetic or personal considerations than as a result of logical or scientific analysis, and are usually made by the owners or the directors of a company rather than by executives at the lower or middle levels’ (Argenti, 1976, p. 63). Not all firms have a business policy defined in the sense of a mission or guiding purpose and many of those which have say nothing about the management of employees, seeing such matters as an operational responsibility of middle management. Indeed, in the extensive literature on business policy and corporate strategy it is extremely rare to find any reference to employees, personnel management or human resource strategy. Thus the study of management style in employee or labour relations is not to be confused with analysis of management practices in each and every firm. It is restricted to those enterprises which, for whatever reason, but often related to the unique contribution of the founding fathers of the firm (Purcell and Sisson, 1983, p. 117) have a guiding set of principles which delineate the boundaries and direction of acceptable management action in dealing with employees. THE DIMENSIONS OF MANAGEMENT STYLE Fox’s use of unitary and pluralist frames of reference had a major influence in initiating the debate on the approach taken by management in industrial relations (1966, 1974) but the concepts are narrower than those implicitly suggested by management style and have a number of limitations. Firstly, wide variations can be found within the unitary frame between those firms which are essentially exploitive of labour and those which emphasize the value of loyalty and commitment. Similarly, pluralism needs to be sub-divided between those maintaining highly structured, often antagonistic, formal relations with employees through their trade unions and those which emphasize dialogue, understanding and co-operation between management and organized labour.
536 JOHN PURCELL Secondly, and more importantly, the very terms unitarist and pluralist do not seem useful as a means of articulating the complexity of management styles since they are, by definition, mutually exclusive. Thirdly, it is often unclear in the way the terms have been used subsequently whether they related to management’s beliefs and policies toward trade unions or, in addition, cover direct relations with employees. In practice, as well as theory, firms can have broad policies or guiding principles toward both the attention to be given to employees and their indi- vidual and group needs, and in regard to trade unions and other types of collective labour organization. Such policies are not necessarily mutually exclusive. Thus two dimensions of management style need to be identified which we might term individualism and collectivism. Individualism This refers to the extent to which the firm gives credence to the feelings and sentiments of each employee and seeks to develop and encourage each employee’s capacity and role at work. Firms which have individualistically centred policies are thus expected to emphasize employees as a resource (often expressed in the clichC ‘our most important resource’) and be concerned with developing and nurturing each person’s talents and worth. We might expect to find strong features of internal labour markets with careful selection at restricted points of entry, internal training schemes, promotion ladders and extensive welfare provisions, including relatively high pay. Payment systems might emphasize merit elements and make use of appraisal and assessment techniques designed to distinguish the contribution of each employee in anything from ‘attitudes to attributes’, as one firm puts it, The line manager’s role in managing people is likely to be emphasized, irrespective of whether the manager is in a production, sales, finance or a personnel position. Attitude surveys may well be used regularly since such organizations are keen to assess the efficacy of their policies. Communication systems are likely to be extensive and developed through a variety of media from newspapers and videos to the personal contact between the manager, the individual and small groups of employees. A typical quote from the 1985 survey reflecting aspects of this style is: Our company’s paramount concerns are recruitment, development, pay and safety of employees. We aim to recruit above average employees who expect to make an above average contribution. We invest time in their development and we very much promote from within the company (corporate level manager). Another respondent at corporate level said: We are quite unusual, operating as leaders of teams. We believe very much in individuals; we trust everybody. We work on the premise that every person
MAPPING MANAGEMENT STYLES 537 needs to have the opportunity for self management. Hence, the manager’s role is very much as leader of the group. The contrast may be seen by a quote from an unusually frank establishment manager who described his company’s philosophy as: ‘a bit ruthless – not very caring: people are just numbers’. A corporate personnel executive, when asked what the firm’s policy or philosophy was towards the place of employees, said: ‘There is very tight fiscal control. We ensure that the lowest cost produces (szc) and we look at the cost of using manpower.’ The term ‘manpower’ is instructive. Firms which do not emphasize indi- vidualism tend to see employees as a commodity (‘just numbers’) resonant of the nineteenth-century distinction between the master and his hands. The commodity status view of labour, typical of neo-classical economists, gives little priority to security of employment and, at its logical extreme, would emphasize spot contracting as the means for determining labour price – sometimes seen in seasonal or casual labour (see Willman, 1983, pp. 117-35). The external labour market is emphasized and managerial policy tends to focus on overt control, discipline and the priority of profit in the sense of surplus value. Clearly there is a wide range of options between the two extreme positions described here and one area for further work is the development of operational categories along a scale of high or low individualism. Paternalism is a third variant of individualism. Some 20 per cent of the 303 respondents in the 1985 survey used words such as enlightened, benevolent, charitable, caring, humane, paternal, family and welfare to describe their policies. A typical quote is: Benevolent paternalism. There is a caring feeling. We should take care of our staff. This a family company who have expanded and we do not find it natural to be structured (corporate level manager); or an establishment manager who said: The company has generated a caring, paternalistic attitude. It listens to employees. It is tolerant of the employees within the constraints of successful business. The Oxford Dictionary definition of paternalism is revealing: ‘limiting the freedom of the subject by well meant regulations’. Companies with a paternalist approach do not appear to place emphasis on employee development and career progression and other attributes of individualism, nor are they dismissive of a sense of social responsibility toward their staff. Notions of caring, humanity and welfare are emphasized as a means of legitimizing managerial authority and the subordinate position of lower level employees who are given few, if any, expectations of the possibility of changing their work roles and ‘natural’
538 JOHN PURCELL place in the hierarchy. The corollary of paternalism is the deferential worker. Paternalist styles thus fall somewhere between high individualism, with its emphasis on employee development, and low individualism concerned with labour control treating employees as a commodity to be exploited in the pursuit of profit. Figure 1 illustrates these ideas. Degrees oj individualism Low High Commodity status Labour control Paternalism Resource status Employee development Figure 1. Management styles and degrees of individualism Collectivism The second dimension of management style concerns the extent to which the organization recognizes the right of employees to have a say in those aspects of management decision-making which concern them. We could assess this in a number of ways. One possibility is to use a test of trade union recognition for the determination of pay and conditions through collective bargaining. We would distinguish unionized from non-union firms. This is not of itself sufficient, however, since in so doing we make assumptions about the nature of the relationship between the firm and the union and exclude organizations where employees participate in decision-making through non-union structures. A focus on industrial democracy might be more helpful since it would allow for a variety of methods by which managers become to a greater or lesser extent account- able for their actions to employees, and where staff have some say in decision- making. The problem here is that there is no agreed definition of industrial democracy and we are frequently warned of ‘pseudo participation’ as Pateman (1970) put it. The use of the phrase ‘to have a say in’ draws attention to Hirschmann’s (1970) (and latterly Freedman and Medoffs 1984) emphasis on voice mechanisms, but here again there is such a variety of types of voice systems that, apart from non-union, non-participative, no-voice organizations, we cannot progress in differentiating between types or degrees of voice systems. The nub of the problem is that it is necessary to identify two aspects of collectivism. On the one hand, we can identify the existence of employee participation structures and determine the levels in the firm at which they operate. On the other, we need to assess the approach of management in operating these joint structures. Do they seek to minimize or oppose them or actively co-operate with them? The critical feature of the first aspect of collectivism is the existence of democratic structures representing employees. Thus we may ask to what extent employees have the right to elect some of their number to represent their interests and articulate their views in dialogue with management. In short,
MAPPING MANAGEMENT STYLES 539 do employees have their own collective organization as an alternative authority structure or interest group? Firms with recognized employee collectives include those where trade unions and collective bargaining exists and organizations with works or company councils along the lines of the West German model. It might well include elected employee trustees on pension funds, collective consultation mechanisms and worker directors, whether they are union based or not. Some firms may have collective representation at all levels of the enterprise from the board room to the workplace, while others restrict collectivism to the local level, preferring to operate a highly decentralized strategy. The degree of collectivism must therefore be incorporated in assessments of this aspect of style. We cannot, of course, say that the existence of collective organization necessarily derives from management policy or style. In many cases unions have fought to be recognized against management’s wishes and labour law is important in some countries in imposing requirements on firms to deal with work councils or recognize trade unions. Given the existence of collective employee organizations, management can however choose the extent to which they legitimize, de facto, the collective and joint structures of negotiation and consultation. Thus, the second aspect of collectivism concerns the degree of legitimacy afforded to the collective by management and thereby the extent to which it is accepted or opposed. Many scholars have sought to distinguish between co-operative relations and those marked by conflict or coercion. The essential ingredient is the degree of legitimacy afforded by management to the employee collective, as Walton and McKersie (1965, p. 189) noted in their elaboration of the attitudinal components of the relationship pattern. In their schema, conflict is marked by a denial of legitimacy; containment-aggression by grudging acknowledgement, accommodation by acceptance of the status quo and co- operation by complete legitimacy. Legitimacy is closely associated with trust and that in turn is related to the extent to which the relationship is relatively open and co-operative allowing for the full flow of information, or is marked by adversarial behaviour, for example excessive reliance on rules as a defence mechanism for coping with conflict. Thus we distinguish between constitutional and consultative (or co-operative) approaches. I have defined the constitutional approach in an earlier article (1986) as where: Unions have been recognized for some time and accepted as inevitable. Employee relations policies centre on the need for stability, control and the institutionalization of conflict. Management prerogatives are defended through highly specific collective agreements and careful attention is paid to the administration of agreements at the point of production. The importance of management control is emphasized with the aim of minimizing or neutralizing union constraints on both operational and strategic management. In contrast, the consultative (or co-operative) approach is defined as where:
54.0 JOHN PURCELL Unions (or other employee collective organizations) are recognized. The attempt is made to build ‘constructive’ relationships with these employee organizations and incorporate them into the organizational fabric. Broad ranging discussions are held with extensive information provided on a whole range of decisions and plans including aspects of strategic management. An illustration of a collectivist style drawn for the 1985 survey is given by a corporate level manager: We operate through trade unions. We encourage union membership. There is a highly participative structure, still paternal but recognizing the legitimacy of unions. The interconnection between these two dimensions of style is likely to be complex and we would expect emphasis to be placed on one or the other in certain periods. Thus it is widely argued that in Britain and the USA there is currently a move to emphasize individualism, seen in the shift to human resource management, the growth in merit pay, assessment systems and greater attention paid to recruitment, selection and internal training schemes linked to flexibility. This is associated with an interest in non-union companies which have distinctive employee development or sophisticated human relations policies such as IBM and Hewlett-Packard (Foulkes, 1981). The view is widely held that the adoption of such a style inevitably implies a move away from collec- tivism in those companies with trade unions or other employee organizations or a positive avoidance of trade unions. There is no particular reason why this should be so and, as will be indicated later, there is little evidence that this is occurring. A number of firms are anxious to preserve trade union organization and high membership, as illustrated by the comment of one senior manager that ‘if trade unions did not exist I would have to invent them’. Some firms may choose, if they can, to have no collectivist system and believe that this is essential for the pursuit of individualism, but other companies, for example Japanese firms, often choose to have aspects of both employee development and co-operative collectivism. Thus the interconnections of the two dimensions of style can be represented diagramatically, see figure 2. THE EVIDENCE FOR MANAGERIAL STYLE What evidence is there to support the contention that there are two dimensions to management style and that organizations can and do operate along both dimensions? Deaton (1985) used data drawn from the first Workplace Industrial Relations Survey in 1980 (WIRS 1) to cluster data around the styles identified by Purcell and Sisson (1983). The results were disappointing but the choice of measures
MAPPING MANAGEMENT STYLES 541 Employee development .- 2 – m Paternalism 9 Labour control High Unitary Adversarial Co-operative Collectivism Figure 2. The interconnections between individualism and collectivism such as types of payment system, use of clocking on, and presence of a specialist personnel manager, among others, do not seem adequate proxies for any particular style and the survey was not designed to be used in identifying styles. More recently, Edwards completed a survey of senior plant executives in the manufacturing sector (1985 and this issue). He did not explicitly seek to analyse style but two points emerge from the survey which are relevant. First, he noted that many of the respondents used the term ‘involvement’ to signify the essence of their broad philosophy or policy and that this term was used to signify ‘a desire to achieve a positive degree of commitment to the goals of the enterprise; achieving the co-operation of workers as individuals was more important than getting detailed industrial relations procedures right’ (1985, p. 8) (emphasis added). This is confirmed by case study research carried out in three projects by the Oxford Institute of Employee Relations at Templeton College in the mid-1980s (Purcell and Undy, 1986). A similar move to commitment strategies and styles based on individuals is noted in America by Walton (1985). Second, according to the Edwards survey and confirmed by case studies and earlier work by Batstone (1984), such a move is not associated with the withdrawal of union recognition or collective bargaining rights and shop steward activities. Research into the labour practices of Japanese companies in Britain has emphasized their distinctive approach of combining union recognition with an emphasis on employee development and commitment (Reitsperger, 1984), a clear instance of carefully considered management style linked to business policy for competitive advantage. We are thus on safe ground in asserting that manage- ment styles operate along the two dimensions and that action in one area, toward individualism for example, is not necessarily associated with changes in the collectivism scale.
542 JOHN PURCELL The 1985 survey asked managers at various levels whether the company has an overall policy or philosophy concerning the management of employees and if so whether it is written down and given to the employees. Respondents are then asked ‘How would you describe the policy?’ Such an open ended question should provide some clues as to the essence of the policy, in that we expect managers to mention the most important or dominant feature of their approach. Whole statements can be coded and a word analysis under- taken, noting the frequency of use of certain words. In the 1985 survey well over 80 per cent of managers at all levels claimed to have a policy or general philosophy. Some 60 per cent of enterprise and divisional managers who said there was a policy indicated that the policy was written and a quarter said it was given to employees. Table I shows the word frequency and statement analysis at each level. Statements and individual words were divided into eight categories derived from the data itself, not predetermined. A number of features are worthy of note. Most of the statements related to relationships or styles concerning individual employees and only 2.5 per cent emphasize relations with trade unions or collectives per se. However, the same survey found that many firms had fairly extensive aspects of collectivism. Ninety per cent recognized trade unions for collective bargaining in the manual worker category, three quarters recognized non-manual unions and just over half had a consultative committee covering the whole of the enterprise for one or more category of employee. Two thirds had employee representatives as trustees for the pension fund. Secondly, there was a fairly even spread of responses between those firms emphasizing more resource-based policies of employee development and involvement, those focusing on paternalism and communication and firms who chose to describe their policies in terms of contractual relations or control and performance criteria. Thirdly, it is possible for the first time to compare responses at the corporate level with those of the divisional and establishment managers in the same enterprise. ‘Involvement’ is a word frequently used at enterprise level but overall statement analysis shows this to be the area where the word is often subsumed in more dominant sentiments of paternalism, communication or contractual relations especially at the establishment level. Direct comparison of responses in the same firm, where this is possible, show that philosophy statements only coincide in 17 per cent of cases. If employee development (or high individualism) is mentioned at corporate level it is the least likely to be listed at establishment level once trade union statements have been excluded. It would appear that the progressive or proactive rhetoric of corporate headquarters seen in the categories of employee development and involvement are the least likely to be recognized at the establishment. This is likely to be because they are the hardest to translate into action, whereas aspects of paternalism or contractual relations are descrip- tions, rather than styles, of management action at the level where employees are located.