After Glenn Mehltretter’s comments about Kinston and Rowbottom’s article from 1990, I went and got copies, OCRed them, and got Warren’s permission to post them here. This is the first, from 1989. They are useful articles and should be in someone’s database but this journal has never been electronically archived anywhere that I could find.
Warren has developed these articles further in later documents. These are substantially correct from today’s point of view, I think, but still missing some of the changes that they made later.
You may also wish to pick up Rowbottom and Billis’s Organisational Design: The Work Levels Approach from the GO Society website (check in the online books section).
Note that Warren’s 7 Languages of Achievement (or decision making) are something different than the seven work levels.
Also interesting is that Kinston and Rowbottom are writing about seven levels of work, as did Rowbottom and Billis. Jaques at some point allowed for an 8th level of work within organizations, and said that other levels of thinking exist. He was kind of wrong here and kind of right, but it is useful to think of seven levels of work as outlined in this article being within a single domain of work.
Even in situations where there is no hierarchical relationships (no Managerial Authority Hierarchy, in Jaques’s term) you still have work being done at a certain level. just because attorneys in a firm work in a Partnership and not a work hierarchy doesn’t mean that all of them do work at the same work level. So this is useful for a variety of contexts, even those not in MAHs.
Warren Kinston and Ralph Rowbottom. 1989. “Levels of Work: New Applications To Management In Large Organisations”. Journal of Applied Systems Analysis, 16: 19-34. [PDF: 6.3MB]
The introductory comments from this paper:
All organisations show stratification into levels of management associated with some form of accountability between individuals or groups working at these levels. Consultancy research over many years has frequently revealed the existence of too many, or less often too few, levels of management. Confusion about the work expected at each level in the organisation is common. Such mistakes present as intractable day-to-day problems and complaints like ‘too much redtape’, ‘failure to implement policies’, ‘understaffing and ovenvork’. ‘role confusion’, ‘line and staff conflict’, ‘insufficient delegation’. ‘duplication of decisions’ and so on. Research confirms that clarification of the levels of management required by an organisation is essential for its own vigour and for the morale and development of its personnel.
The levels of work scheme was first devised by Elliot Jaques  in collaboration with Wilfred Brown . and has since been investigated and developed by Rowbottom and Billis [30,31]. Billis . Kinston and Rowbottom  and others. A popular exposition was provided by Evans . Jaques’s emphasis has been on the time-span of discretion ofthe longest task in any role. together with the (subjective) mental processes involved in work at each level. Rowbottom and Billis. by contrast, have emphasised the (objective) work output or differential response to need at each level – in other words the differing nature of the mission at each level (Kinston ). All authors emphasise dramatic qualitative shifts in the nature of work as one moves from level to level.
The theory has been created and developed through an iterative scientific inquiry process, with an emphasis on critical scrutiny of concepts, and validation through long-term testing with organisations. It has been used in consultancy projects and workshops for some thousands of managers and professionals in several countries including the U.K., U.S.A.. Australia. Netherlands, Singapore, and South Africa. Applications have been made in diverse fields including industry and commerce (Jacques ; Brown ). social services (Row bottom et al., : Billis et aI., ), health services (Jaques [13); Kinston and Rowbottom : Kinston. [20)), the armed forces (Stamp. ), voluntary agencies (Billis, ) and the civil service (Jaques, ).
Practising managers rapidly take to the theory and experience it as helpful in clarifying their situation. By contrast, academics have felt no such pressure to give the theory the attention it deserves. A major difficulty for them lies in the fact that empiricism needs to give way to design: i.e. applying the theory usually requires involvement with the organisation and opens up existing problems in its hierarchical structure.
This paper is based primarily on findings from the 20-year action-research and organisational consultancy programme to improve organisation and management in the U.K.’s National Health Service (NHS) cited above. The NHS has the stupendous task of operating as a unified enterprise providing a comprehensive range of health services for a population of over 55,000,000. It is the largest employing organisation in Europe. with around 1.000,000 staff and an annual budget of over £20 billion (in 1986-87). Over the years, we have worked with thousands of staff of all kinds and at all levels using a method of research which is collaborative analytic and systemic (Jaques ; Rowbottom ; Kinston . We will first summarise the theory. then use examples from the NHS to illustrate the basic approach. and finally show that viewing the levels as part of a matrix of management has led to clarification of many practical issues.
Image Credit: ADLER typewriter Model n°7 (Frankfurt / Germany). Unknown model date (probably ~1930/40). © Dake. (CC BY-SA 2.5)
Thanks for posting the article! It takes a while to get these articles from the library.
The original research done initially by Elliott on development of pay, which resulted in the bands of the development curves only had seven levels and modes. This was confirmed by other researchers at Bioss. The Levels of Work framework as developed by Gillian Stamp has seven levels.
When Elliott later reinterpreted the levels and constructed the concept of declarative, accumulative, serial and parallell, and the metalevels of these, it was possible to theorize about levels which did not exist in previous research findings.
“Jaques at some point allowed for an 8th level of work within organizations, and said that other levels of thinking exist. He was kind of wrong here and kind of right, but it is useful to think of seven levels of work as outlined in this article being within a single domain of work.”
Forrest- This is a rather bald assertion. Can you explain how Jaques was “kind of wrong and kind of right”?
It’s not that bald: there has always been a good deal of disagreement between Jaques and many of the BIOSS staffers about there being eight levels to corporate management. These aren’t a single question but several: Are there no more than seven levels to work organizations? Are there more than seven levels to mental processing regarding work? Are levels grouped in quintaves? How many quintaves are there? How are quintaves organized? What is the relationship of timespan of discretion or personal time horizon to multiple quintaves? What in the world are we “measuring” here? What does this interact with?
Some folks have solutions that would clearly indicate that Jaques was “kind of right, kind of wrong”, as were the BIOSS staffers who disagreed. Unfortunately, until they are finally published I can only provide Fermat’s bald assertion — “”I have a truly marvelous proof of this proposition which this margin is too narrow to contain.” Except in this case it is “others” and not “I”. You have to love academic review cycles.
There are a number of issues at hand – one set about personal capabilities and the other about organizational levels.
There is nothing in the research that Elliott and others conducted during the bioss years that indicate more than seven levels. The proposal of an eight level does not have the solid research background as the initial seven.
Yes organizations usually have (far) more organizational levels than there are levels of personal capability. The consequences of that is well described by Elliott and other authors. But should the difficulties in adapting the number of organizational levels to the number of levels of personal capabilities drive us to add levels to accomodate the issue?
Interesting response. You’ve begun by stating that the labelled “bald assertion” is really not that bald and end by stating until additional information is published that we will need to be content with the “bald assertion”. Although I initially did not understand your references to quintaves there may be some relevance inasmuch as the eight stratum is concerned.
In its association with the Georgian scale due to the neutral tuning within the quintave system, the eighth degree or octave is slightly widened, which often results in a rise in pitch from the beginning of a song to the end.
From my recollection of a stratum VIII organization Jaques advocated these to be super corporations not unlike General Electric or General Motors. The analogy that they are distinct and a widening of an S VII organization seems appropriate.
Solid article. There’s a lot of good data here, though I did want to let you know something – I am running Fedora with the current beta of Firefox, and the look and feel of your blog is kind of flaky for me. I can read the articles, but the navigation doesn’t function so great.