How much hierarchy does anyone want?

[Tim Lund]

This is something that occurred to me yesterday when the police reps at the Forest Hill Assembly mentioned that the gangs blighting - other - parts of the borough were quite sophisticated, and even had their own hierarchies. Later, when it came to the 'Our Lewisham, Our Say', with views being sought on where Lewisham should cut its spending, there was the inevitable suggestion that we want to maintain front line services, but cut hierarchy ...

So - does hierarchy help make organisations more effective, or is it just a way to make those at the top feel important? Assuming some hierarchy is a good thing, how much is too much of a good thing, and how could you say how much there is? In fact, in a normal workplace, there will be such a way, which is the average number of people line managers have reporting to them. After the meeting I took this up with one of the senior officers there, and he said that it was, in fact, something Lewisham's CEO, Barry Quirk, is very aware of.

There's something here http://en.wikipedia.org/wiki/Span_of_control on what this number should be - less than 4 is fairly hierarchical, 10 is fairly impressive - if 'flat' organisations are what you want. I don't suppose it's possible to say what the best number is for a local authority, but I'd be interested to know how Lewisham compares with other boroughs. It would give an idea of how much middle management could be lost before losing the benefits of hierarchy.

[maturin]

I've always found this quite useful in any dealings I have had with government, business or media organisations.

To comprehend Factor One, we must picture a civil servant, called A, who finds himself overworked. Whether this overwork is real or imaginary is immaterial. For this real or imagined overwork there are, broadly speaking, three possible remedies. He may resign; he may ask to halve the work with a colleague called B; he may demand the assistance of two subordinates, to be called C and D.

There is probably no instance, however, in history of A choosing any but the third alternative. By resignation he would lose his pension rights. By having B appointed, on his own level in the hierarchy, he would merely bring in a rival for promotion to W's vacancy when W (at long last) retires. So A would rather have C and D, junior men, below him. They will add to his consequence and, by dividing the work into two categories, as between C and D, he will have the merit of being the only man who comprehends them both.

It is essential to realize at this point that C and D are, as it were, inseparable. To appoint C alone would have been impossible. Why? Because C, if by himself, would divide the work with A and so assume almost the equal status that has been refused in the first instance to B; a status the more emphasized if C is A's only possible successor. Subordinates must thus number two or more, each being thus kept in order by fear of the other's promotion.

When C complains in turn of being overworked (as he certainly will) A will, with the concurrence of C, advise the appointment of two assistants to help C. But he can then avert internal friction only by advising the appointment of two more assistants to help D, whose position is much the same. With this recruitment of E, F, G and H the promotion of A is now practically certain.

For more see the inestimable C Northcote Parkinson http://www.heretical.com/miscella/parkinsl.html

[Robin Orton]

I used to be civil servant O, and 'Maturin's' Parkinson quotation is too near the truth for comfort!

Obviously bosses should not have so many people working directly tothem that they cannot get to know each of them personally. Subject to that, the optimum 'span of command' must surely depend on a whole range of factors, for example, the degree of complexity and unpredictability of the job done by the junior person, whether or not there are clear output measures for assessing his or her performance, and how far the organization is driven by detailed objectives passed down from the top. I expect these factors will vary considerably between the various businesses within an individual local authority such as Lewisham, and also between different authorities (depending e.g. on the extent to which particular services have been contracted out to the private or voluntary sector.) I'm not sure that any one single 'span of command' index would be a particularly good measure of different LAs' efficiency.

[Tim Lund]

Robin:

I didn't mean to suggest that there was an easily identified best span of control index, but it should be possible to say which boroughs have the flatter structures. It might then be interesting to compare this with how satisfied people are with the services supplied - although getting a sensible measure for this would be pretty hard - people will always like having a moan about their council, wherever they live. But there has to be something better than the Comprehensive Area Assessment http://oneplace.audit-commission.gov.uk ... ?region=51 where Lewisham gets exactly the same - positive - report as Tower Hamlets, e.g. a Green Flag for

Engaging and empowering local people

.

For more on Tower Hamlets, see http://londonist.com/2010/10/tower_haml ... _elect.php