1. Introduction
Efficiency is a concept widely used by economists, engineers,
organization theorists, consultants, politicians, managers and others. It figures large in
the many vocabularies that abound in the world today and it seems that 'efficiency' is one
of the focuses of Western culture.
Efficiency has met with enthousiasm as well as critique. An early
advocate of efficiency is Frederick Taylor (1911). (1)
Shortly after, John Dewey made critical remarks on scientific management but considered
efficiency to be a "servant of freedom" (Middle Works, Vol. 10, p. 119).
Kotarbinsky (1968) defended efficiency from a praxiological point of view. Organization
theorist Mintzberg (1990, p. 330-332), for whom efficiency is a "dirty word",
and economist Heilbroner (1977, p. 92), who calls the "worship of efficiency"
one of our faults, (2) are very critical. (3)
Other people say that efficient designs can lead to greater human freedom, with the
implication, I think, that such designs can also become prisons. Cameron (1986) is
also critical. But in spite of these criticisms, many people still sing the praises of
efficiency. An example is Flint (1991) who, writing on the philosophy and practice of
auditing, suggests that we have an obligation to be efficient.
The opposite reactions which efficiency elicits confront us with an
unhappy situation, which needs to be clarified. This can be done only when we have more
insight in efficiency: "What kind of concept is it?"; "What does it
mean?"; What are the presuppositions of its use?". Because knowledge of
historical roots increases our understanding, I will, after some preliminary remarks about
its meaning, first look briefly at efficiency from the perspective of the history of ideas
(see section 2). After that, an important presupposition of efficiency will be analysed
(see section 3). Finally, I will discuss the wretched situation of praise and disapproval
(section 4).
2. Historical roots
Efficiency (or inefficiency) is used nowadays in two ways. First of
all, it functions as a seemingly value-neutral qualitative (by description) (4)
or quantitative (by measurement) characterisation of processes, machines or practices. If
used quantitatively the measurements are often based on 'mathematical' definitions of
efficiency which are assumed to be widely applicable (Charnes 1978, 1981). Secondly,
efficiency is also used in a normative sense to evaluate situations, processes, etc.
Herbert Simon gives the following, well known, definition which stresses the second use:
"In its broadest sense, to be efficient simply
means to take the shortest path, the cheapest means, toward the attainment of the desired
goals".(Simon 1976, p.14, 172) (5)
However, from a historical perspective this meaning is rather recent,
because the term was not used in this way until the 19th century. (6)
Looking for the roots of efficiency makes it necessary to attend to causality, economics
and the field of technology. (7) In what comes next I will
shortly attend to these subjects.
Causality and economy
Philosophers like Bacon (1561-1626) and Hobbes (1588-1679) used
'efficient' to denote any cause that produces an effect. This dates back to medieval
thought and Aristotle's theory of causation, with its four causes: the formal cause, the
material cause, the moving or efficient cause, and the final cause. Not only in nature
were efficient causes to be present, and every antecedent cause which produces some effect
is said to be an efficient one.
A new meaning of efficiency can be found in the works of the British
philosopher Jeremy Bentham (1748-1832). New is that 'efficient' functions as a predicate
for causes that produce, or produce to a certain degree, a desired effect. (8)
Something similar is shown by Charles Babbage (1792-1871), well known for his design of a
mechanical computer. In his study on the economy of machinery and manufactures, which
appeared in 1832, a machine is considered to be 'efficient' when it produces the desired
effect (1832, p. 264). This machine efficiency should be distinguished from its economy,
i.e. the economy of human time it implies and the economy of raw materials needed. The
relation of efficiency to a desired effect is also present in John Stuart Mill's
(1806-1873) work. Compared to Bentham and Babagge, however, he introduced another new
element: the costs. In his writings on political economy (1842), Mill speaks of the
efficiency of labour and of power, such as steam or water, in the context of the costs of
production. One can find expressions like the "general efficiency of all
labour", meaning the production of the same quantity as before with less labour (Mill
Vol. III, p.481), and the "greatest efficiency of labour" (Mill Vol. II, p. 141)
to be attained by the devision of labour, by the sofisticated use of machines, and of
skilled people (who cost more) only when necessary. Therefore, in Mill's thought the
concept of 'efficiency', besides the production of the desired effect, also refers to the
costs involved. By saying this, I am not suggesting that previously costs were not
considered. What is new is the connection between costs and the concept of efficiency.
Technology
The ideas of Mill and Babbage are narrowly tied up with developments
known as The Industrial Revolution. Important fields of activity were the textile
industry and mining. Ideas about machines during that time developed in this context. Of
course, there was reflection on the nature and possibilities of machinery before. An
example is Leibniz' (1646-1716) involvement with a mining project in the Harz region of
Germany (Leibniz 1950). But it took approximately another century for a really systematic
experimentation and theoretical reflection on machines to develop (Klemm 1959). However,
the question why this was the case is far beyond the scope of this article.
The systematic experimentation and reflection just mentioned did not
use standard terminology. For example, Davies Gilbert (1767-1839), who at the end of
1820's was president of the Royal Society of London, called the work done by a steam
engine its efficiency. But there was also the older practical engineering calculation of
the duty of such engines as the weight of water that could be raised one foot by
the consumption of one bushel of coal. An analogous calculation was used for water wheels.
In this case the duty was expressed as a fraction indicating the capacity of the engine to
restore the moving water to the source. Practical experience had already made clear that,
for example, overshot wheels did a better job than undershot wheels and the same applies
to high pressure steam engines compared to low pressure ones. Theoretical understanding
was lacking however. Instances of a more theoretical approach which contributed to the
explanation of such practical insights are Lazare Carnot's Mémoire (1791),
Hachette's Traité des machines (18111, 18284) and Sadi
Carnot's Réflections (1824). Sadi Carnot is especially important because of his
general theory of heat machines dealing with "the most effective employment of
heat" (Carnot 1960, p.47). Such a theory was needed in order to make a systematic
improvement of machines possible. It was this theory and the relatively low performance of
actual machinery which made Diesel, at the end of the 19th century, look for a process
with the "highest heat efficiency" (quoted in Klemm 1959, pp. 342-345). (9)
Gradually the concept of machine efficiency as some kind of fraction,
analogous to the notion of duty mentioned above, developed. In the case of heat machines,
it was the fraction of the work done by the machine/the heat absorbed by the machine.
Another illustration is combustion efficiency as the ratio of the actual amount of heat
released and the maximum amount which theoretically could be released by the complete
combustion of the fuel. The mathematical definitions given in Charnes (1978, 1981) all
take the idea of a ratio for a starting point.
It is not necessary to trace the historical roots of efficiency any
further here. The picture given is clear enough in order to procede with my argument. I
will conclude this section with some general remarks. Machine efficiency (Etp)
is a technical-physical concept related to energy transformation. One can measure or
calculate it using laws of nature and it has an ideal limit as well. Millian efficiency (Ee)
is narrowly related to the costs, for example, of wages or labour. It is an economic
concept. One cannot simply calculate it from laws and there is no ideal situation which
can function as a standard. Therefore, Etp is not identical to Ee
and it is even possible that a system has a high Etp and a low Ee
or vice versa. They are connected when the costs of physical resources, such as
energy, are taken into consideration.
3. A central presupposition of efficiency
The historical analysis of the meaning of efficiency reveals two
imporant features: a) efficiency is defined rather formally; b) it is not a unitary
concept. The formal character supports the mathematical approaches (related to
input/output ratios; Charnes 1978, 1981). That it is not unitary is already shown by the
differences between Etp en Ee. Simon's definition,
which is wide enough to cover both Etp and Ee suggests
that efficiency is a simple, unitary notion, but it is not. This will become clearer when
we look at the use of efficiency in concrete situations.
Take Etp in the case of steam engines. We can look at
the essential mechanics/thermodynamics of the machine and focus on the heat needed for
doing a certain amount of work. But we can also widen our scope, and consider the
efficiency of producing the heat itself or even introduce aspects of safety and
environmental issues. Of course, this complicates the end the machine is designed for. At
the same time, it broadens the range of factors (costs) that are important in relation to
efficiency. In such a situation we are thinking of various 'efficiencies' at the same
time, which can seldom be analysed by applying a common standard.
Apart from the difficulties of weighing the input and output when both
are multiple, my argument illustrates that every application of efficiency involves
decisions about what we think that is important. When efficiency is defined in terms of a
production function (Charnes 1981) such decision making seems to be absent. However, this
view is a too simple. Even if it were possible to determine empirically what in fact
produces the effect, it would still make a non-empirical decision if we choose to confine
efficiency to those factors.
It is clear that the use of the concept of efficiency presupposes a
decision about the kinds of costs we are willing to recognize. Such decisions
regarding costs are unavoidable and to recognize them is of paramount importance. The
notion of costs should be taken in a broad sense: the sacrifices made (e.g. energy, time,
money, safety, clean air, etc.) in order to obtain the results, the ends pursued. Often,
the costs are obvious. We take them for granted, not seldom because of cultural
influences, oblivious of the fact that use of the concept of efficiency involves decisions
about them. In our society, for example, 'environmental costs' were out of sight for a
long time. Cost decisions are related to what we value, explicitly or implicitly.
Therefore, every application of efficiency is itself value laden, based on values that are
part of our world view. (10)
4. Praise and Disapproval
As such, efficiency functions as a normative criterion for means in
relation to ends, that is for instrumental rationality. (11)
In the 20th century, instrumental reason is praised as well as disapproved. It is critized
by Horkheimer and Adorno. Habermas considers it to be a limited mode of rationality that,
if isolated from communicative rationality, has unhappy consequences. In line with this,
we may impute the problems we have with efficiency to a misunderstanding and misuse of
instrumental reason. (12) This is not an indefensible line
of thought, but more needs be said. The problems arise mainly because the concept of
efficiency itself remains unreflected. When people disagree we should ask what notions of
efficiency are being used, what kinds of costs are considered, what background values they
start from, etc. Their positions probably depend on differences on this level. The same
holds in the case of internal inconsistenies. Heilbroner, for example, critizes
efficiency, but is at the same time anxious because "efficient fission reactors"
may not come on time (Heilbroner 1974, p. 54). If the character of efficiency is not
sufficiently understood, unfruitful clashes of opinion (e.g. "efficiency is a dirty
word" <-> "it is a servant of freedom") or internal inconsistencies
will remain.
I feel supported in my view by the works of the American philosopher
John Dewey. He connected efficiency with different adjectives, such as
"technical" and "social". Technical efficiency is linked to issues
such as the division of labour and social efficiency to making our experiences more
valuable to each other (Middle Works, Vol 9, p. 127). Besides this, he viewed
efficiency as a practical skill, as doing things without waste and confusion (see also
Kotarbinsky 1968). Practices are of many kinds (the artist, the lawyer, the physician, the
mathematican, the chemist, the businessman, the teacher etc.). They all involve
efficiencies, which (and that is crucial) are closely connected with the meaning, the
purpose, of those practices (Later Works, Vol. 17, p. 77-820). Therefore,
efficiency can only be a servant of freedom when we balance the different efficiencies
with each other. Gains in technical efficiency, for example, should evaluated in terms of
the effects on social efficiency and on the skills needed for a meaningful practice.
I think that it is important to be aware of the differentiated contexts
(technical, social, medical care, business, etc.) involved and we must always ask the
question "what is the meaning of the practice we are engaged in?".
For example, if we are judging a school in terms of its efficiency,
then we should always keep the meaning of teaching in mind. Exclusive attention to
technical efficiency in education and administration, using methods of measurement and
standardization, can only lead to an improvement of existing methods, as already Dewey
told us (Middle Works, Vol. 10, p. 118-121). In the case of business, we could
start from the general view that its meaning/purpose is to add value to life by offering
goods to the market. Generally speaking this implies that business should also try to
avoid a decrease of value as far as possible, which brings in environmental issues. Of
course, in concrete situations this must be worked out in more detail, in collaboration
with individual organizations and the people involved. When we are working in a medical
practice one also needs a clear view of its poper meaning. When we want to create an
efficient organization of medical practice this meaning will help us to discern and weigh
different kinds of costs, those for medication, time (of physicians, nurses, the use of a
laboratory), and equipment, but, for example, also patients' time and the burden put on
them need to come in focus.
5. Conclusion
In the introduction I refered to Flint's view that we have an
obligation to be efficient. I think that my argument has made clear that we can conclude
in favour of such an obligation only when we duly consider the contexts/meanings of our
activities. Only then can we overcome the clash of praise or disapproval, of onesided
worship or rejection of efficiency. However, it is not easy to overcome this clash. It
presupposes that we are able to rise beyond the level of personal tastes and favourite
pursuits. In order to realize this we are in need of a worked out philosophy of the
different practices our culture knows of, be it eduation, business or medicine, to name a
few.
|
