Is hierarchy the natural order for organisations or can a looser structure be just as effective? Andrew Wall examines the options

The clinical psychologist was missing at the meeting where we discussed the resettlement of people with learning disabilities into the community. On enquiry she was found to be at the estate agent's discussing possible houses for our clients - a job given to her by the rest of the team.

This was my first meeting as acting manager for this team and I was exasperated. I wanted the clinical psychologist, the most highly paid member of the group, to use her professional skills to equip the clients to live a more independent life. Surely someone else could run round the estate agents? The group were surprised at my reaction and became rather condescending. 'You don't understand, Andrew. We don't have a hierarchy here. We work as a team.'

Here was the classic confusion between organisational structure and managerial style. Structure determines the roles that will maximise the effectiveness of the organisation. It clarifies who does what, what skills are required, and how to deploy them. Managerial style is about how the people in the roles behave.

Is the traditional, hierarchical approach to organisations still valid or would a more fluid, flexible approach get better results?

The managerial textbooks have had a field-day with matrix management, the systems approach, contingency theory, networks and so on. This branding of traditional methods of hierarchy - what was thought to be the 'natural order' - as somehow outmoded has left the average manager confused and a little demoralised.

But what claims can hierarchy make to being the 'natural order'? Why is this easily understood and common approach to organisations deemed rigid and unresponsive to needs?

Let's go back to basics. A lone caveman did the best he could to protect and feed himself. When another one came along, a decision had to be made about whether to make friends with him. It soon became apparent that there were mutual advantages. Several jobs were made easier by an extra pair of hands, and one man could sleep while the other hunted or kept watch.

From this beginning, the idea of allocating duties and providing support to each other arose. As the groups got bigger, so did the need to allocate work efficiently - this was determined by the skills each person possessed. Some had higher order skills than others - gathering fuel for the fire was much simpler than treating a sprained joint. Those with the rarer skills soon saw that they had some potential to improve their status.

This is not a bedtime story. All organisations are based on these simple principles. Professionals with an exclusive body of knowledge are higher in the hierarchy. Similarly, those who command the resources which those professionals require are also nearer the top of the chain of command.

Many people would argue that alternatives should be found for hierarchies because this 'natural order' is too primitive a system for today's complexities. The criticisms are well known: bureaucracy is the manifestation of hierarchy, leading to long chains of command and elaborate processes. This, in turn, slows down enterprises and diminishes job satisfaction as workers, bound as they are by procedures, have no chance to influence affairs. Such organisations will be unable to respond rapidly to change - and that, we all know, is an absolute requirement today.

But such anti-hierarchy arguments can manipulate the truth. After all, it is equally possible to depict the more modern organisation in a pejorative manner. Indeed, anyone who has examined an organisation will quickly come across staff who say: 'I don't know how I fit into this organisation', 'My so-called boss never shows any interest in my work', 'We sit around in meetings all day and nothing seems to get done', 'I was blamed for something which was nothing to do with me.' All these remarks reflect an organisation where the structure is unclear to those within it and whose objectives, therefore, are uncertain.

Unfortunately, matrix management and its close relative, project management, are particularly prone to these difficulties. People from various departments are gathered around a task under a project leader who is not their usual boss. This may have some point if the task is specific and time-limited - such as commissioning a new building. But if it is the normal manner of working, confusion results. The basic question: 'Who is my boss?' becomes ambiguous and means that the normal processes of supervision, mentoring and appraisal cannot be satisfactorily performed. Such a situation also leads to divided loyalties.

Examples of this can be seen on many wards today. The clinical director, usually a doctor, is in charge of a clinical area of work; in hospital this will be a group of wards or a clinical department. The staff working there are accountable to him or her.

But many of those staff are professionals in their own right. What exactly is the relationship to their own head of profession in that trust? We know that only about 25 per cent of nursing directors actually manage the nurse workforce. What does this do to professional standards and the need to develop each nurse's professional skills? Claiming that trained nurses are autonomous professionals accountable only to themselves and the UKCC is side-stepping managerial reality. The smaller professions, such as physiotherapy, occupational therapy and speech therapy, are even more isolated from their professional colleagues.

It is often said that peer groups such as hospital consultants operate without hierarchy. On examination, however, this appears to be at considerable bureaucratic cost, with an elaboration of committees and procedures. There is a basic confusion here - autonomy in clinical judgement has been muddled with accountability to the organisation as a whole.

The argument in favour of the 'ward team' approach is that the clinical team is centred on a given set of patients and is responsive to their needs. It can also be held accountable for results. The argument against this is that these teams can become too independent of the organisation,

not sharing corporate objectives and operating divisive systems.

So which is the better arrangement? The key issue is accountability. Can that accountability be assured in a way which feels natural to its participants and which fulfils the obligations it has to its patients and to the public more generally?

My argument is that hierarchy is the most easily understood organisational model. People feel it is natural. Look at popular culture - it always depicts organisations in a hierarchical fashion. Basically most of us want a boss - we want to know who we are answerable to and where we fit into things.

The personality of an organisation comes from the prevailing style of management. And there is no doubt that staff and patients alike are affected by this. Research several decades ago by Professor Reg Revans in Manchester suggests that where staff are happy, patients get better more quickly.1

Can an organisation have an unambiguous structure and be happy? No organisational structure that needs a lot of explaining to its participants can be a good thing. This is where many of the matrix types - or what I call 'bubble managerial structures', floating around in space not attached to anything - fall down: they are too confusing. They cannot answer the simple questions such as who is in charge of what.

The boundaries of authority and responsibility are constantly being redefined. Surely this is a waste of time.

But in arguing for the 'old-fashioned' chain of command I do not accept that it is unresponsive or repressive. Staff at the bottom of the structure may well feel more valued if their work is recognised by the person at the top. The trouble is that managers are so nervous at being dubbed paternalistic that they will often ignore staff they pass in the corridor.

This idea that managers can lead from behind is ludicrous. Leadership is about being out in front, giving directions and being the first to take risks. It need not rule out seeking the advice of the rest of the group. In today's NHS, the line of accountability has been clearly spelled out. It won't work if groups within that structure are bamboozled into woolly and sentimental concepts about structure. Urwick, writing 50 years ago, suggested that poor organisational design was 'illogical, cruel, wasteful and inefficient'.2

A clear acknowledgement that sound organisations are based on hierarchy should avoid the worst of Urwick's fears.