Hierarchical organisations have gained an image of being rigid and bureaucratic. However they are capable of delivering your product/service efficiently and consistently. In turbid environments rigidity can be a problem, however.
Modern approaches such as the matrix structure also present difficulties. Because of the difficulties and complexity of operating multidimensional structures, more modern structures have emerged, such as network structures.
Mintzberg wrote an article "The Structuring of organisations". In it he says that "in sharp contrast to [...] contingency theory, [...] organisations can select their situations in accordance with their structural designs just as much as they can select their designs in accordance with their situations".For example, diversified firms may divisionalise, but there is evidence that divisionalised firms may diversify more. Bureaucracies can be encouraged to form by stable environments, but again bureaucracies have a habit of trying to stabilise their environments. Entrepeneurial forms, which operate in dynamic environments, need to maintain flexible structures, but they also seek out and try to remain in dynamic environments in which they can outmanoeuvre the bureaucracies.
Mintzberg defined six basic parts of the organisation
1. The Operating Core (where the work of producing the product/service gets done)
2. The Strategic Apex (the home of top management)
3. The middle line (all managers that stand between the strategic apex and the operating core)
4. The technostructure (left blob, where organisational systems are designed)
5. The support staff (right blob, all the staff who support the organisation outside its normal workflow, eg HR, PR etc in a traditional firm - provided it wasn't an HR or PR firm where that would be at the operating core)
6. The ideology (a halo of beliefs and traditions surrounding the whole)
Within the organisation, various methods of co-ordination exist to ensure that output is produced. Mintzberg defined 6 of those, too
1. Mutual adjustment. This is where the people in the operating core achieve co-ordination by the simple process of informal communication.Used in simple organisations, the most obvious way to co-ordinate. But also the most complex, and often used where extremely difficult problems need solving.
2. Direct supervision. This is where the managers at the strategic apex directly control what the operational core does. This comes into play when an organisation is too big to be suitable for mutual adjustment alone. A team manager may manage his players this way, or the entrepeneurial boss of a small firm may manage his staff this way.
The above two are ad-hoc in nature. There are also four more formal methods of co-ordination
3. Standardisation of work processes, where the technostructure produce and provide well defined business processes that staff in the operational core are expected to follow. These processes are tailored to the job in hand.
4. Standardisation of outputs. This is where the technostructure produce and provide an overall well defined process which details the interfaces between jobs in the operational core, defining what must happen to the inputs to produce the output and how the outputs of one stage are passed to the inputs of the next.
5. Standardisation of skills. This could be where staff in the operating core have well defined skills (surgeon, anaethetist etc) which they will have learned outside of the organisation and already know how to interact and co-ordinate with other expert staff as they know what to expect of them.
6. Standardisation of norms. This is where the operating core workers share a common set of beliefs and can achieve co-ordination based on it. An example might be a religious order where every member shares a belief in the importance of attracting converts.
Your firm divides up the work to produce its outputs amongst its individuals and teams. This is known as division of labour. Mintzberg says that these methods of co-ordination serve to knit together the divided labour of the organisation and serve as the most basic elements of structure - the glue that holds the organisation together.
The design for this "knitting" could be considered to have ten parameters which influence the division of labour and subsequent co-ordination. These fall into four groups.
Group 1. Design of individual positioins
Job Specialisation
This is where you decide what each person will do - how specialised will each job be, how many distinct tasks should it contains and how much control over those tasks should the person who does the job have? This allows you to design the division of labour of the organisation. Jobs with a few narrow tasks are horizontally specialised. Those with many broad ones are horizontally enlarged. Jobs that involve little control by those who do them are vertically specialised, those which are thoroughly controlled by the worker are vertically enlarged.
Unskilled jobs are generally those which are specialised horizontally and vertically. Many professional jobs are horizontally specialised (narrow range of tasks) but vertically enlarged (lots of control over how you do them).
Behaviour formalisation
Worker behaviour is formalised to reduce its variability, ultimately to predict and control it. Thus behaviour formalisation is a means to achieve vertical specialisation.
Bureaucracies tend to rely highly on formalisation of behaviour. Mintzberg regards any structure that relies on standardization for co-ordination to be bureaucratic, not just standardisation of processes.
Training
The behaviour required of some tasks is too complex to be proscribed directly by the technostructure's analysts. In this case, training is required for the people in the operational core before they can do their work. In other words, they must acquire some standardised body of knowledge and skills.
Formalisation and training are basically substitutes for one another. Formalisation for the unskilled workers, training for the professional workers. Formalisation takes power from the worker and puts it into the technostructure. Training enables the worker to take power from all the other parts of the organisation and puts it into the hands of the professional workers.
Indoctrination
Socialisation refers to the process by which a new member learns the value system, the norms, and the required behaviour patterns. Much of this happens as new organisational members interact with old, informally and unoficially. Some also takes place officially, eg induction schemes.
Group 2. The superstructure of the organisation
Unit Grouping
Jobs ("positions" in mintzberg-speak) are groupd into units, each under its own manager. These are then clustered into larger units under their own managers and the cycle repeats until the whole organisation comes under a single manager at the strategic apex. Formal power is granted through a hierarchy of authority. Organisational Chart (organigram, from French) only documents the formal power structure, and is often superseded by informal power.
Grouping establishes a system of common supervision among positions and units, requires positions and units to share common resources (and facilitates that), allows p&u to be assessed on common measures of performance, and encourages mutual adjustment amongst units and groups with a tendency to be physically located together or in proximity.
Positions and units can be grouped by function (eg knowledge, process etc) or by market (product/service, client, place), or in other words by means or by ends.
Unit size is also important, but the assumption that no team should be bigger than about 5 or 6 for supervision reasons assumes that direct supervision is the only option. It is not the case. The greater the use of standardisation, the larger the work unit can be. The greater the need for mutual adjustment, the smaller the work unit would normally be.
Group 3. The design of lateral linkages
Planning and Control systems standardise outputs. They allow you to standardise outputs ahead of time. Control allows you to determine whether the standards have been met. Action planning systems focus on before-the-fact determination of outputs, whereas performance control systems refer to monitoring of results after-the-fact.
Liaison Devices
Although mutual adjustment may occur naturally in small work units, it needs to be encouraged across units where grouping has the known tendency to discourage it. This has often been left to chance. Liaison devices are formal parameters of structural design that have been developed to stimulate mutual adjustment across units. Mintzberg's most important four are (in ascending order!):-
Liaison positons - jobs created to co-ordinate the work of two units directly, without having to pass through vertical managerial channels - typically have no power per se and have to rely on negotiation skills
Task forces and standing committees - intitutionalised forms of meetings which bring members of a number of different units together on a more intensive basis
Integrating managers - essentially liaison personnel with formal authority. This is not authority over the units they link, but usually of something important to those units, for example approval of decisions or control over budgets.
Matrix Structure - carries liaison to its natural conclusion. This is a way of balancing two ways of grouping, for example functional with market, or region with product. A dual authority structure exists. Matrixes can be permanent (where the units/positions remain in place all the time) or shifting (as is the case with much project work, where units & positions move around frequently).
Liaison devices are most logically used with work that is horizontally specialised (few narrow tasks), professional, and interdependent. They tend to destroy bureaucratic priority.
Group 4. The design of the decision-making system
Vertical and horizontal decentralisation
Decentralisation:- the sharing of decision making power.
Power at a single point - centralised structure. Power is dispersed - decentralised.
Vertical decentralisation - the delegation of formal power down the hierarchy to line managers.
Horizontal decentralisation - the extent to which formal power is dispensed out of the line managemenent hierarchy to non-managers (operators, workers, support staff).
Decentralisation allows the organisation to respond quickly to local conditions in many places and can serve as a stimulus for motivation, since capable people require significant room to manoeuvre if they are to perform at full capacity.
So thinking back to the six co-ordinating mechanisms
Direct supervision - full horizontal centralisation - all power rests with managers. Vertical centralisation since dependence on direct supervision - this is type 1 decentralisation.
The various types of standardisation co-ordinating mechanisms lead to different degrees of decentralisation. Standardisation of work processes leads to limited horizontal decentralisation - type 2 in Mintzberg's figure 13.6 (reader p257).
Standardisation of output leads to limited vertical decentralisation (type 3).
Standardisation of skills leads to extreme horizontal decentralisation - a bottom-heavy organisation, type 4.
Mutual adjustment leads to selective horizontal and vertical decentralisation - type 5.
Standardisation of norms leads to organisations allowing members considerable freedom to act, and therefore plain decentralisation (type 6).
The situational factors
Mintzberg says that many factors influence how you go about determining your design parameters and vice versa. For example, the age and size of your organisation, its systems, its environment (stability and complexity) and its power system (Is it tightly controlled from the outside).
Age and size hypotheses
H1. The older the organisation, the more formalised the behaviour (seen it all before syndrome, or even not invented here syndrome)
H2. The larger the organisation, the more formalised its behaviour (seen it all before often)
H3. The larger the organisation, the more elaborate its structure (the big barber shop can afford a specialist to cut children's hair, the smaller one cannot)
H4. The larger the organisation, the larger the size of its average unit (makes sense, with growth this happens)
H5. Structure reflects the age of founding of the industry (interesting. Industries that predate the industrial revolution seem to favour one kind of structure, more modern industries another. Compare the traditional academic hierarchy in universities which is based on ancient practices to that of a modern software company, for example)
Technical systems hypotheses
H6. The more regulating the technical system - that is, the more it controls the work of the operators - the more formalised the operating work and the more bureaucratic the structure of the operating core - eg mass production assembly lines render that work highly routine and predictable, encourage its specialisation and formalisation, and this creates the conditions for bureaucracy
H7. The more complex the technical system, the more elaborate the administrative structure, especially the larger and more professional the support staff, the greater the selective decentralisation (to that staff) and the greater the use of liaison devices to co-ordinate the work of that staff! Phew! Complex machinery requires skilled experts, who have the capability to design, select and even modify it. Those people then are given considerable power to make decisions concerning that machinery and they must therefore be encouraged to use liaison devices to ensure mutual adjustment among them.
H8. The automation of the operating core transforms a bureaucratic administrative structure into an organic one. Unskilled work co-ordinated by standardisation of work processes creates a bureaucratic structure-friendly environment. An obsessive control mentality exists. But where that work is automated, social relationships change - the obsession with control disappears. The managers and analysts needed to look after staff in the operational core are no longer needed in such numbers. Automation results in a reduction of line authority in favour of staff expertise. Organisations get humanised by the automation of their operating work :)
Environmental Hypotheses
H9. The more dynamic the environment, the more organic the structure - the organisational structure cannot stabilise while the environment is not stable.
H10. The more complex the environment, the more decentralised the structure - the prime reason to decentralise a structure is that all the decision making information cannot be held in one brain. Complex environments lead to a requirement to process more information
H11. The more diversified the organisation's markets, the greated the propensity to split it into market-based units, or divisions, given favourable economies of scale - diversification breeds divisionalisation. This reduces the need to co-ordinate across units.
H12. Extreme hostility in its environment drives any organisation to centralise its structure temporarily. - Notwithstanding H10, social science says that when threatened, the tendency for groups (and thus orgs) is to centralise power, in other words fall back and regroup under direct supervision.
H13. Disparities in the environment encourage the organisation to decentralise selectively to differentiated work constellations. What a large amount of management-speak in this hypothesis! But it's simple. Where an organisation faces more than one type of environment (eg turbid and stable) which would normally require different structures, the tendency is to create "work constellations" to deal with each and differentiate the structure. Each constellation is given the power to make decisions related to its own "subenvironment". This is selective decentralisation.
Power Hypotheses
H14. The greater the external control of the organisation, the more centralised and formalised its structure - do you have a parent firm or organisation? If so you tend to centralise power at your strategic apex to formalise behaviour - the CEO is likely to be held responsible for the organisation's actions.
H15. The power needs of the members tend to generate structures that are excessively centralised - everyone wants to increase their own power, right? Or at least to keep others from having power over them. Its human nature. But of course, the authority structure in place can be used to subdue this desire.
H16. Fashion favours the structure of the day (and of the culture), sometimes even when inapppropriate - don't forget the latest trendy management fad! Many organisations will try to adopt design parameters that aren't necessarily appropriate for themselves!
Configurations
The elements of structure tend to cluster naturally in a certain number of ways - Mintzberg calls them Configurations. 6 methods of co-ordination, parts of the organisation, types of decentralisation. They all fit together to describe the essence of six basic configurations - table 13.1 p 263+4 reader. Each of the parts of the organisation can pull the organisation in its own direction. When conditions favour one of these pulls over the others, a particular organisation is drawn to structure itself appropriately.
The pulls create the following configurations
Simple Structure
The name tells it all. One large unit consisting of one or a few top managers, one of whom dominates by the pull to centralise, and a group of operators who do the basic work. Standardisation - absent. No analysts (technostructure absent). Little use of planning, training or liaison devices. Few middle line managers. Organic structure. Typically operates in a dynamic environment as that is often the only place it can outsmart the bureaucracies. But although environment is dynamic, it is also simple. Typically a young organisation, as few old simple structures exist. Time would normally drive a simple structure to bureaucracy and where it doesn't their vulnerability causes them to fail. Many are often small as bureaucracy often comes with size. Classic entrepeneurial firm. Sometimes under crisis conditions, large organisations also revert temporarily to simple structures to allow forceful leaders to save them.
Machine Bureaucracy
This is the offspring of the industrial revolution, says Mintzberg. Jobs became highly specialised and work became highly standardised. Requires a large technostructure which has a good deal of informal power, limited amount of horizontal decentralisation. A large hierarchy of middle line managers, usually structured on a functional basis all the way to the top, so centralised vertically. Environment and production system must be fairly simple. A natural fit with mass production.
Professional Bureaucracy
This is another bureaucratic configuration, but one that relies on the standardisation of skills rather than work processes or outputs. The pull to professionalise dominates. For example, hospitals, universities. Highly decentralised horizontally, power over many decisions, both operating and strategic, flows down the hierarchy to the operating core. Small technostructure since standardisation occurs as a result of training that takes place outside the organisation.
Divisionalised Form
Like the professional bureaucracy in that it is a set of rather independent entities coupled together by a loose administrative structure. In a professional bureaucracy those entities are individuals, here they are units in the middle line, generally called divisions. They exert a dominant pull to balkanise. It is not a complete structure but a partial one superimposed on others. Kind of a structure of structures, where individual divisions have their own structures, probably machine bureaucracies in themselves. This happens where product lines are diversified, in large and mature organisations. Mintzberg says they tend to have run out of opportunities or become bored. Headquarters relies on performance control systems - the standardisation of outputs. A small technostructure maintains these.
Adhocracy
None of these structures suits modern industries, particularly aerospace, film making, consulting or petrochemicals. Innovation is essential. Bureaucratic structures are too inflexible and simple structures are too autocratic. Project structures fuse experts drawn from different specialities into smoothly functioning creative teams. Adhocracy fills this role and is dominated by the experts' pull to collaborate. Co-ordination for mutual adjustment amongst its experts is critical and so liaison devices are used, integrating managers, standing committees, task forces and matrix structures are used here. Experts are grouped in functional units for housekeeping, but deployed in market-based project teams to actually do their work. Power is delegated to these teams. Decentralisation happens selectively in the vertical and horizontal dimensions. This results in power distributed unevenly all over the structure according to expertise and need. The environments of adhocracies are both complex and dynamic. Some adhocracies fail, and others end up converting themselves to a form of bureaucracy.
Missionary
This is what happens when an organisation is dominated by a pull to evangelise. This results in members pulling together, loose division of labour and little job specialisation. There's also little differentiation of the strategic apex from the rest or between divisions and so on. The missionary structure is held together by its standardisation of norms. This is achieved through the design parameter of indoctrination. Once the new member is indoctrinated into the organisation they will identify with the common beliefs and given considerable freedom to make decisions. The missionary has virtually no technostructure. Beyond a certain size it tends to divide itself into smaller units that look much like the parent.
In summary, Mintzberg's structural model is a way to take a very complex subject matter and make it manageable by considering how the various dimensions (coordinating mechanisms, organisational parts, design parameters, situational factors etc) cluster together to form distinct types of organisations. This is more realistic than trying to consider all of the permutations and combinations of these dimensions, or trying to deal with them all in a fragmented way.
No comments:
Post a Comment
Comments are moderated before posting.