Å®ÉúСÊÓƵ

The Management of Scale: Big Organizations, Big Decisions, Big Mistakes
by David Collingridge, Routledge, pp 202, £35 hbk/ £12.99 pbk

Disasters at Three Mile Island and Chernobyl, the misery produced by
many high-rise buildings, the lamentable performance of large-scale irrigation
schemes in developing countries – all are failures, but do they share common
roots in organisational terms? Do big decisions in big organisations inevitably
produce big mistakes?

When it comes to ‘inflexible’ technologies, David Collingridge thinks
this is the case. For more than 10 years, his work has elaborated the difficulties
of controlling these technologies, which by their nature impede a trial-and-error
approach to innovation. According to Collingridge, inflexibility is associated
with long lead times, meaning that much development takes place before anyone
can detect problems. By that time, capital-intensive technologies with large
unit costs make changes difficult, if not impossible. The need for extensive,
dedicated production infrastructures compounds the problem. Inflexible technologies
are, he states, ‘incapable of control by any organisation, no matter how
²õ³Ù°ù³Ü³¦³Ù³Ü°ù±ð»å’.

Inflexible technologies may be impossible to control, but do some organisational
arrangements favour their adoption? In Collingridge’s view, large-scale,
highly-centralised organisations in which technological innovation takes
place in an environment of consensus, shielded from debate, are the guilty
ones. What we need, he says, is small-scale, trial-and-error, incremental
planning, with debate involving dissenting experts.

‘Invariably’, Collingridge asserts, ‘history reveals that today’s giant
systems have very humble origins, with the entrepreneurs in the early stages
of the technical system never adopting an inflexible project with long lead
time, large unit size, high capital intensity and infrastructure demands.’

Collingridge surveys a range of interesting case histories in considerable
detail, including the space shuttle, nuclear power, North Sea oil, videotext,
large irrigation schemes and system building for high-rises. He argues that
these projects all ‘failed to deliver the promised benefits to those who
chose them in the first place’. All are pronounced to be costly failures,
save North Sea oil. Despite suffering from similar problems to the others,
oil has made substantial profits. This, it seems, was a fluke, caused by
fortuitous shifts in the market price for the commodity.

Certain questions now arise. What exactly does Collingridge mean by
failure? Does he mean an undesirable output from the system in question?
But where should we place the system boundary, and who is to judge what
is undesirable? These are fundamental omissions from the book.

Consider nuclear power. This technology has suffered from well-known
serious accidents, and it has been dogged by severe underestimates of operating
costs and other technical problems. On the other hand, in Britain, for example,
central government has undoubtedly greatly valued its relationship with
the weapons programme, and its role as the source of stable supplies of
energy removed from the danger of disruption by industrial action.

Oil from the North Sea may also have made a great deal of money for
some but it has been costly for others. Collingridge recognises the tragic
human costs, but events like the Piper Alpha disaster were catastrophic
for the insurance industry. He speaks of the ‘inhumanity of these stories’,
where the consequences fell ‘on the heads of innocent people quite powerless
to alter the technology imposed on them by a combination of big business
and government’.

These ideas seem to ring a bell. Nearly 20 years ago, Ernst Schumacher
wrote Small is Beautiful, the well-known ‘study of economics as if people
mattered’. His ideas on decentralisation were influential in debates about
‘appropriate technology’, with their underlying thesis advocating the use
of technologies that are labour-intensive, cheap, environmentally friendly,
small-scale, and use renewable resources.

A similar tone seems to be adopted by Collingridge, who states that
the root cause of many of the ‘failures’ he cites is the application of
‘authoritarian, large-scale, inhuman, expert-ridden technology’. This is
followed by the book’s only quote from Schumacher, with its call for small-scale
technology ‘compatible with man’s need for creativity’. As Collingridge
puts it, ‘a wonderfully refreshing statement of the ideal of technology’.

In my view there is a tendency here to conflate different arguments,
covering organisational, management, and certain moral issues, so leading
to a lack of coherence in the book’s overall message.

As he turns to technical aspects of the model, Collingridge’s work is
reminiscent of the typology drawn up by the American disaster theorist Charles
Perrow. Perrow’s scheme identifies the inevitability of failures with structural
characteristics of the organisational systems associated with high-risk
technologies. Like Perrow, Collingridge’s scheme is subject to the criticism
that human factors and corporate cultures may vary sufficiently from one
organisation to another to distort the theory’s predictions.

Perrow’s work is much concerned with complexity rather than scale. This,
in my view, is the more important consideration in determining the likelihood
of failure, and it is not discussed by Collingridge. He does not consider
other management problems associated with matters of scale, for example,
the difficulty of extrapolating the experience of small to large-scale production.

The arguments in this book, like so much of Collingridge’s work, are
potentially invaluable for senior managers and civil servants concerned
with technological innovation. Sadly, the book’s shortcomings in its lack
of coherence regarding policy matters may inhibit its usefulness.

It does, however, provide a highly stimulating excursion through a range
of issues at the interface of decision making, organisation and technology.
Its publishers suggest the book is appropriate for MBA and undergraduate
students of management. I am sure that is the case, and it will make a
very interesting read for anyone with a relatively casual interest in
technology policy, or someone interested in the failures of a specific technology,
say nuclear power.

Tom Horlick-Jones is director of the London-based organisation Epicentre,
and is a visiting research fellow at the Open University’s Centre for Technology
Strategy at Milton Keynes.