The Organisation of Industry

Author(s): G. B. Richardson
Source: The Economic Journal , Sep., 1972, Vol. 82, No. 327 (Sep., 1972), pp. 883-896
Published by: Oxford University Press on behalf of the Royal Economic Society
Stable URL: https://www.jstor.org/stable/2230256

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Royal Economic Society and Oxford University Press are collaborating with JSTOR to digitize,
preserve and extend access to The Economic Journal

This content downloaded from

130.238.7.40 on :ffff:ffff on Thu, 01 Jan 1976 12:34:56 UTC
All use subject to https://about.jstor.org/terms
 THE ORGANISATION OF INDUSTRY1
I
I WAS once in the habit of telling pupils that firms might be envisaged as
islands of planned co-ordination in a sea of market relations. This now
seems to me a highly misleading account of the way in which industry is in
fact organised. The underlying idea, of course, was of the existence of two
ways in which economic activity could be co-ordinated, the one, conscious
planning, holding sway within firms, the other, the price mechanism,
operating spontaneously on the relations between firms and between firms
and their customers. The theory of the firm, I argued, had as its central
core an elaboration of the logic of this conscious planning; the theory of
markets analysed the working of the price mechanism under a variety of
alternative structural arrangements.
I imagine that this account of things might be acceptable, as a harmless
first approximation, to a large number of economists. And yet there are
two aspects ofit that should trouble us. In the first place it raises a question,
properly central to any theory of economic organisation, which it does not
answer; and, secondly, it ignores the existence of a whole species of indus
trial activity which, on the face of it, is relevant to the manner in which co
ordination is achieved. Let us deal with each of these matters in turn.
Our simple picture of the capitalist economy was in terms of a division of
labour between the firm and the market, between co-ordination that is
planned and co-ordination that is spontaneous. But what then is the
principle of this division? What kinds of co-ordination have to be secured
through conscious direction within firms and what can be left to the working
of the invisible hand? One might reasonably maintain that this was a key
question-perhaps the key question-in the theory of industrial organisation,
the most important matter that the Divine Maker of market economies on
the first day of creation would have to decide. And yet, as I hope soon to
show, it is a matter upon which our standard theories, which merely assume
but do not explain a division between firm and market, throw little light.
Let me now turn to the species of industrial activity that our simple
story, based as it is on a dichotomy between firm and market, leaves out of
account. What I have in mind is the dense network of co-operation and
affiliation by which firms are inter-related. Our theoretical firms are indeed
islands, being characteristically well-defined autonomous units buying and
selling at arms' length in markets. Such co-operation as takes place
between them is normally studied as a manifestation of the desire to
restrict competi tion and features in chapters about price agreements and
market sharing.
1
I am grateful to Mr. J. F. Wright, Mr. L. Hannah and Mr. J. A. Kay, each of whom gave
helpful comments on a draft of this article.
883

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
884 THE ECONOMIC JOURNAL [SEPT.

But if the student closes his textbook and takes up a business history, or the
financial pages of a newspaper, or a report of the Monopolies Commission,
he will be presented with a very different picture. Firm A, he may find, is
a joint subsidiary of firms B and C, has technical agreements with D and E,
sub-contracts work to F, is in marketing association with G-and so on.
So complex and ramified are these arrangements, indeed, that the skills of a
genealogist rather than an economist might often seem appropriate for their
disentanglement.1 But does all this matter? Theories necessarily abstract
and it is always easy to point to things they leave out of account. I hope to
show that the excluded phenomena in this case are of importance and that
by looking at industrial reality in terms of a sharp dichotomy between firm
and market we obtain a distorted view of how the system works. Before
doing so, however, I wish to dwell a little longer on the several forms that
co-operation and affiliation may take; although the arrangements to be
described are no doubt well known to the reader, explicit mention may never
theless help to draw attention to their variety and extent.

II
Perhaps the simplest form of inter-firm co-operation is that of a trading
relationship between two or more parties which is stable enough to make
demand expectations more reliable and thereby to facilitate production
planning. The relationship may acquire its stability merely from goodwill
or from more formal arrangements such as long-term contracts or share
holding. Thus, for example, the Metal Box Company used to obtain a
discount from its tin plate suppliers in return for undertaking to buy a
certain proportion of its requirements from them, and the same company
owned 25% of the share capital of the firm supplying it with paints and
lacquers. In the same way Imperial Tobacco owned shares in British Sidac,
which made cellophane wrapping, and in Bunzl, which supplied filter tips.
Occasionally shareholdings of this kind may be simply investments held for
their direct financial yield, but more generally they give stability to relation
ships through which the activities of the parties are co-ordinated both quanti-
1 The sceptical reader might care to look up a few cases in the reports of the Monopolies Com

mission. The following example is found in the report on cigarette filter tips. Cigarette Com
ponents Ltd. made filter tips for Imperial Tobacco and Gallaher using machines hired from
these companies. It has foreign subsidiaries, some wholly and some partially owned. It was both
licensee and licensor of various patents one of which was held by the Celfil Trust, registered in
Liechtenstein, with regard to the ultimate control of which Cigarette Components told the Mono
polies Commission they could only surmise. Nevertheless, this patent was of key importance in
that the Celfil licensees, of which Cigarette Components was only one, were bound by price and
market sharing arrangement. Cigarette Components was itself owned by Bunzl Ltd., in which
Imperial Tobacco had a small shareholding. The raw material for the tips is cellulose acetate tow
which was made by Ectona Fibres Ltd., a company in which Bunzl had a 40% interest and a
subsidiary of Eastman Kodak 60%, Agreements had been made providing that, should Bunzl lose
control of Cigarette Components, then Eastman could buy out their shares in Ectona . . . etc.,
etc.

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
1972) THE ORGANISATION OF INDUSTRY 885

tatively and qualitatively. Not only is it made easier to adjust the quantity
of, say, lacquer to the quantity of cans which it is used to coat but the
speci fication and development of the lacquers can be made appropriate to
the use to be made of them. And in the synthetic fibre industry likewise,
linkages between firms at the various stages-polymer manufacture, yarn
spinning and finishing, textile weaving-help bring about the co-ordinated
development of products and processes. The habit of working with models
which assume a fixed list of goods may have the unfortunate result of causing
us to think of co-ordination merely in terms of the balancing of quantities
of inputs and outputs and thus leave the need for qualitative co-ordination
out of account.
Co-operation may frequently take place within the framework provided by
sub-contracting. An indication of the importance of this arrangement is
provided by the fact that about a quarter of the output of the Swedish
engineering industry is made up of sub-contracted components, while for
Japan the corresponding figure is about a third and in that country's auto
mobile industry almost a half. Sub-contracting on an international basis,
moreover, is said to be becoming more widespread and now a dense network
ofarrangements links the industries of different countries.1 Now the fact that
work has been sub-contracted does not by itself imply the existence of much
co-operation between the parties to the arrangement. The plumbing work
on a building contract may be sub-contracted on the basis of competitive
tenders for the individual job. Frequently, however, the relationship
between the parties acquires a degree of stability which is important for two
reasons. It is necessary, in the first place, to induce sub-contractors to
assume the risks inherent in a rather narrow specialisation in skills and equip
ment; and, secondly, it permits continuing co-operation between those con
cerned in the development of specifications, processes and designs.
Co-operation also takes place between firms that rely on each other for
manufacture or marketing and its fullest manifestation is perhaps to be found
in the operations of companies such as Marks and Spencer and British Home
Stores. Nominally, these firms would be classified as retail chains, but in
reality they are the engineers or architects of complex and extended patterns
of co-ordinated activity. Not only do Marks and Spencer tell their suppliers
how much they wish to buy from them, and thus promote a quantitative
adjustment of supply to demand, they concern themselves equally with the
specification and development of both processes and products. They
decide, for example, the design of a garment, specify the cloth to be used and
control the processes even to laying down the types of needles to be used in
knitting and sewing. In the same way they co-operate with Ranks and
Spillers in order to work out the best kind of flour for their cakes and do not
neglect to specify the number of cherries and walnuts to go into them.

1
See the Economic Bulletinfor Europe, Vol. 21, No. I.
No. 327.-voL. 82.

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
886 THE ECONOMIC JOURNAL [SEPT.

Marks and Spencer have laboratories in which, for example, there is develop
ment work on uses of nylon, polyester and acrylic fibres. Yet all this
orchestration of development, manufacture and marketing takes place
without any shareholding by Marks and Spencer in its suppliers and without
even long-term contracts.
Mention should be made, finally, of co-operative arrangements
specifically contrived to pool or to transfer technology. Surely the field of
technical agreements between enterprises is one of the under-developed areas
of economics. These agreements are commonly based on the licensing or
pooling of patents but they provide in a quite general manner for the
provision or exchange of know-how through the transfer of information,
drawings, tools and personnel. At the same time they are often associated
with the acceptance by the parties to them of a variety of restrictions on their
commercial freedom-that is to say with price agreements, market sharing
and the like.
This brief description of the varieties of inter-firm co-operation
purports to do no more than exemplify the phenomenon. But how is such
co-opera tion to be defined ? And how in particular are we to distinguish
between co-operation on the one hand and market transactions on the
other? The essence of co-operative arrangements such as those we have
reviewed would seem to be the fact that the parties to them accept some
degree of obligation and therefore give some degree of assurance-with
respect to their future conduct. But there is certainly room for infinite
variation in the scope of such assurances and in the degree of formality with
which they are expressed. The blanket manufacturer who takes a large order
from Marks and Spencer commits himself by taking the appropriate
investment and organisational decisions; and he does so in the expectation
that this company will continue to put business in his way. In this instance,
the purchasing company gives no formal assurance but its past behaviour
provides suppliers with reason to expect that they can normally rely on
getting further orders on acceptable terms. The qualification " normally "
is, of course, important, and the supplier is aware that the continuation of
orders is conditional on a sustained demand for blankets, satisfaction with
the quality of his manufacture and so on. In a case such as this any formal
specification of the terms and conditions of the assurance given by the
supplier would scarcely be practicable and the function of goodwill and
reputation is to render it unnecessary.
Where buyer and seller accept no obligation with respect to their future
conduct, however loose and implicit the obligation might be, then co operation
does not take place and we can refer to a pure market transaction. Here there
is no continuing association, no give and take, but an isolated act of
purchase and sale such, for example, as takes place on an organised market for
financial securities. The pure market transaction is therefore a limiting case,
the ingredient of co-operation being very commonly present, in some degree,
in the relationship between buyer and seller. Thus although

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
1972] THE ORGANISATION OF INDUSTRY 887

I shall have occasion to refer to co-operation and market transactions as

distinct and alternative modes of co-ordinating economic activity, we must
not imagine that reality exhibits a sharp line of distinction; what confronts
us is a continuum passing from transactions, such as those on organised
commodity markets, where the co-operative element is minimal, through
intermediate areas in which there are linkages of traditional connection and
goodwill, and finally to those complex and inter-locking clusters, groups and
alliances which represent co-operation fully and formally developed. And
just as the presence of co-operation is a matter of degree, so also is the sover
eignty that any nominally independent firm is able to exercise on a de facto
basis, for the substance of autonomy may often have been given up to a
customer or a licensor. A good alliance, Bismarck affirmed, should have a
horse and a rider, and, whether or not one agrees with him, there is little
doubt that in the relations between firms as well as nation states, the condition
is often met.

III
It is time to revert to the main line of our argument. I had suggested
that theories of the firm and of markets normally provide no explanation of
the principle of the division of labour between firms and markets and of the
roles within a capitalist economy of planned and spontaneous co-ordination.
And I also maintained that these theories did not account for the existence
of inter-firm co-operation and affiliation. It is upon the first of these two
deficiencies that I now wish to concentrate.
Probably the simplest answer to the question as to the division oflabour
between firm and market would be to say that firms make products and
market forces determine how much of each product is made. But such an
answer is quite useless. If "products " are thought of as items of final
expenditure such as cars or socks, then it is clear that very many different
firms are concerned with the various stages of their production, not only in
the sense that firms buy in components and semi-manufactures from other
firms but also in that there may be a separation of manufacture and marketing
(as in the case of Marks and Spencer and its suppliers) or of development and
manufacture (as in the case of licensors and licencees). If, alternatively, we
simply define " products " as what firms do, then the statement that firms
make products is a tautology which, however convenient, cannot be the basis
of any account of the division of labour between firm and market.
It is worth observing that we learn nothing about this division oflabour
from the formal theory of the firm. And this is perhaps not surprising as the
theory, in its bare bones, is little more than an application of the logic of
choice to a particular set of problems. It may be that the theory indeed
makes it more difficult to answer our question in that, in order the better to
exhibit this logic of choice, it is formulated on the assumption of " given

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
888 THE ECONOMIC JOURNAL [SEPT.

production functions " which represent the maximum output obtainable

from different input combinations. However useful this representation of
productive possibilities, it leaves one important class of ingredients out of
account. It abstracts totally from the roles of organisation, knowledge,
experience and skills, and thereby makes it the more difficult to bring these
back into the theoretical foreground in the way needed to construct a theory
of industrial organisation. Of course I realise that production functions
presume a certain level of managerial and material technology. The point
is not that production is thus dependent on the state of the arts but that it has
to be undertaken (as Mrs. Penrose has so very well explained)1 by human
organisations embodying specifically appropriate experience and skill. It is
this circumstance that formal production theory tends to put out of focus,
and justifiably, no doubt, given the character of the optimisation problems
that it is designed to handle; nevertheless, it seems to me that we cannot
hope to construct an adequate theory of industrial organisation and in
particular to answer our question about the division of labour between firm
and market, unless the elements of organisation, knowledge, experience and
skills are brought back to the foreground of our vision.
It is convenient to think of industry as carrying out an indefinitely large
number of activities, activities related to the discovery and estimation of
future wants, to research, development and design, to the execution and
co-ordination of processes of physical transformation, the marketing of goods
and so on. And we have to recognise that these activities have to be carried
out by organisations with appropriate capabilities, or, in other words, with
appropriate knowledge, experience and skills. The capability of an
organisation may depend upon command of some particular material
technology, such as cellulose chemistry, electronics or civil engineering, or
may derive from skills in marketing or knowledge of and reputation in a
particular market. Activities which require the same capability for their
undertaking I shall call similar activities. The notion of capability is no
doubt somewhat vague, but no more so perhap:s than that of, say, liquidity
and, I believe, no less useful. What concerns us here is the fact that
organisations will tend to specialise in activities for which their capabilities
offer some comparative advantage; these activities will, in other words,
generally be similar in the sense in which I have defined the term although
they may nevertheless lead the firm into a variety of markets and a variety of
product lines. Under capitalism, this degree of specialisation will come
about through competition but it seems to me likely to be adopted under any
alternative system for reasons of manifest convenience. Mrs. Penrose has
provided us with excellent accounts of how companies grow in directions
set by their capabilities and how these capabilities themselves slowly expand
and alter.2 Dupont, for example, moved from a basis in nitro-cellulose
1 E.T. Penrose, The Theory qfthe Growth of the Firm (Oxford University Press, 1959).
2 E. T. Penrose, ibid.

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
1972] THE ORGANISATION OF INDUSTRY 889

explosives to cellulose lacquers, artificial leather, plastics, rayon and cello

phane and from a basis in coal tar dyestuffs into a wide range of synthetic
organic chemicals, nylon and synthetic rubber. Similarly, Marks and
Spencer, having acquired marketing and organisational techniques in
relation to clothing were led to apply them to foodstuffs.
There is therefore a strong tendency for the activities grouped within a
firm to be similar, but this need not always be so. In the history of any
business random factors will have left an influence, and the incentive to take
up a particular activity will sometimes be provided, not by the prior posses
sion of an appropriate capability, but by the opportunity of a cheap acquisi
tion, through a family or business connection or because of management's
belief that the profitability of investment in some direction was being gener
ally under-estimated. There is no need to deny, moreover, that a variety
of potential gains are provided by grouping activities irrespective of their
character; risks can be spread, the general managerial capability of the firm
can be kept fully employed and the allocation of finance can be planned from
the centre. None of this is in contradiction with the principle that it will
pay most firms for most of the time to expand into areas of activity for which
their particular capabilities lend them comparative advantage. A firm's
activities may also, on occasions, be more similar than they superficially
appear. If a firm acquired companies irrespective of the character of their
activities we should term it conglomerate; but if the motive for the purchases
were the belief that the companies were being badly managed, the hope
being to restore them to health before re-selling them at a profit, the manage
ment would be exercising a particular capability.

IV
I have argued that organisations tend to specialise in activities which, in
our special sense of the term, are similar. But the organisation of industry
has also to adapt itself to the fact that activities may be complementary. I
shall say that activities are complementary when they represent different
phases of a process of production and require in some way or another to be
co-ordinated. But it is important that this notion of complementarity be
understood to describe, for instance, not only the relationship between the
manufacture of cars and their components, but also the relationship of each
of these to the corresponding activities of research and development and of
marketing. Now it is clear that similarity and complementarity, as I have
defined them, are quite distinct; clutch linings are complementary to
clutches and to cars but, in that they are best made by firms with a capability
in asbestos fabrication, they are similar to drain-pipes and heat-proof suits.
Similarly, the production of porcelain insulators is complementary to that of
electrical switchgear but similar to other ceramic manufacture. And while

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
890 THE ECONOMIC JOURNAL [SEPT.

the activity of retailing toothbrushes is complementary to their manufac

ture, it is similar to the activity of retailing soap. This notion of comple
mentarity will require closer definition at a later stage, but it will be conven
ient first to introduce one further (and final) set of conceptual distinctions.
It is clear that complementary activities have to be co-ordinated both
quantitatively and qualitatively. Polymer production has to be matched,
for example, with spinning capacity, both in terms of output volume and
product characteristics, and investment in heavy electrical equipment has
likewise to be appropriate, in scale and type, to the planned construction of
power stations. Now this co-ordination can be effected in three ways; by
direction, by co-operation or through market transactions. Direction is
employed when the activities are subject to a single control and fitted into one
coherent plan. Thus where activities are to be co-ordinated by direction it is
appro priate that they be consolidated in the sense of being undertaken jointly
by one organisation. Co-ordination is achieved through co-operation when
two or more independent organisations agree to match their related plans
in advance. The institutional counterparts to this form of co-ordination are
the complex patterns of co-operation and affiliation which theoretical formu
lations too often tend to ignore. And, finally, co-ordination may come about
spontaneously through market transactions, without benefit of either direc tion
or co-operation or indeed any purposeful intent, as an indirect conse quence of
successive interacting decisions taken in response to changing profit
opportunities. Let us now make use of this somewhat crude cate gorisation to
re-interpret the questions with which we started.

V
What is the appropriate division of labour, we should now ask, between
consolidation, co-operation and market transactions?
If we were able to assume that the scale on which an activity was
undertaken did not affect its efficiency, and further that no special capabilities
were ever required by the firm undertaking it, then there would be no limit
to the extent to which co-ordination could be affected by direction within
one organisation. If production could be set up according to " given "
production functions with constant returns, no firm need ever buy from, or
sell to, or co-operate with any other. Each of them would merely buy
inputs, such as land and labour, and sell directly to consumers-which,
indeed, is what in our model-building they are very often assumed to do.
But, of course, activities do exhibit scale economies and do require specialised
organisational capabilities for their undertaking, the result being that self
sufficiency of this kind is unattainable. The scope for co-ordination by
direction within firms is narrowly circumscribed, in other words, by the exist
ence of scale economies and the fact that complementary activities need not
be similar. The larger the organisation the greater the number of capabilities

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
1972] THE ORGANISATION OF INDUSTRY 891

with which one may conceive it to be endowed and the greater the number of
complementary activities that can, in principle, be made subject to co
ordination through direction; but even if a national economy were to be
run as a single business, it would prove expedient to trade with the rest of
the world. Some co-ordination, that is to say, must be left either to co
operation or to market transactions and it is to the respective roles of each of
these that our attention must now turn. -
Building and brick-making are dissimilar activities and each is undertaken
by large numbers of enterprises. Ideally, the output of bricks ought to be
matched to the volume of complementary construction that makes use of
them and it is through market transactions that we expect this to come about.
Brickmakers, in taking investment and output decisions, estimate future
market trends; and errors in these estimates are registered in stock move ments
and price changes which can lead to corrective actions. As we all know, these
adjustments may work imperfectly and I have myself argued elsewhere 1 that
the model which we often use to represent this type of market is
unsatisfactory. But this is a matter with which we cannot now concern
ourselves. What is important, for our present purposes, is to note that
impersonal co-ordination through market forces is relied upon where there is
reason to expect aggregate demands to be more stable (and hence predictable)
than their component elements. If co-ordination were to be sought through co-
operation, then individual brick-makers would seek to match their investment
and output plans ex ante with individual builders. Broadly speaking, this does
not happen, although traditional links between buyers and sellers, such as are
found in most markets, do introduce an element of this kind. Individual
brick manufacturers rely, for the most part, on having enough customers to
ensure some cancelling out of random fluctuations in their several demands.
And where sales to final consumers are concerned, this reliance on the law of
large numbers becomes all but universal. Thus we rely on markets when there
is no attempt to match complementary activities ex ante by deliberately co-
ordinating the corre sponding plans; salvation is then sought, not through
reciprocal undertak ings, but on that stability with which aggregates, by the
law oflarge numbers, are providentially endowed.
Let us now consider the need to co-ordinate the production of cans with
tin plate or lacquers, of a particular car with a particular brake and a
particular brake lining, of a type of glucose with the particular beer in
which it is to be used, or a cigarette with the appropriate filter tip. Here
we require to match not the aggregate output of a general-purpose
input with the aggregate output for which it is needed, but of particular
activities which, for want of a better word, we might call closely complementary.
The co-ordination, both quantitative and qualitative, needed in these cases
requires the co-operation of those concerned; and it is for this reason that
1 In Information and Investment (Oxford University Press, 1961).

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
892 TIIE ECONOMIC JOURNAL [SEPT.

the motor car companies are in intimate association with component makers,
that Metal Box interests itself in its lacquer suppliers, Imperial Tobacco with
Bunzl and so on. Co-ordination in these cases has to be promoted either
through the consolidation of the activities within organisations with the
necessary spread of capabilities, or through close co-operation, or by means of
institutional arrangements which, by virtue of limited shareholdings and
other forms of affiliation, come somewhere in between.
Here then we have the prime reason for the existence of the complex
networks of co-operation and association the existence of which we noted
earlier. They exist because of the need to co-ordinate closely complementary
but dissimilar activities. This co-ordination cannot be left entirely to
direction within firms because the activities are dissimilar, and cannot be
left to market forces in that it requires not the balancing of the aggregate
supply of something with the aggregate demand for it but rather the match
ing, both qualitative and quantitative, of individual enterprise plans.

VI
It is perhaps easiest to envisage co-ordination in terms of the matching, in
quantity and specification, of intermediate output with final output, but I
have chosen to refer to activities rather than goods in order to show that the
scope is wider. The co-operation between Marks and Spencer and its
suppliers is based most obviously on a division of labour between production
and marketing; but we have seen that it amounts to much more than this
in that Marks and Spencer performs a variety of services in the field of
product development, product specification and process control that may be
beyond the capability of the supplying firms. And one may observe that
inter-firm co-operation is concerned very often with the transfer, exchange
or pooling of technology. Thus a sub-contractor commonly complements
his own capabilities with assistance and advice from the firm he supplies.
New products also frequently require the co-operation of firms with different
capabilities, and it was for this reason that LC.I. originally co-operated with
Courtaulds in the development of nylon spinning and now co-operates with
British Sidac in developing polypropylene film.
It is indeed appropriate to observe that the organisation of industry has
to adapt itself to the need for co-ordination of a rather special kind, for co
ordination, that is to say, between the development of technology and its
exploitation. A full analysis of this important subject cannot be attempted
here but it is relevant to consider those aspects of it that relate to our
principal themes. What then are the respective roles, in relation to this
kind of co-ordination, of direction, co-operation and market transactions ?
Obviously there are reasons why it may be convenient to co-ordinate the
activities of development and manufacture through their consolidation within
a single organisation. Manufacturing activity is technology-producing as

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
1972] THE ORGANISATION OF INDUSTRY 893

well as technology-dependent; in the process of building aircraft or turbo

alternators difficulties are encountered and overcome and the stock of
knowledge and experience is thereby increased. But there are also good
reasons why a firm might not be content to seek the full exploitation of its
development work through its own manufacturing activity. The company
that develops a new product may itself lack sufficient capacity to manu facture
it on the scale needed to meet the demand and may not have time enough to
build up the required additional organisation and material facilities. It
could, of course, seek to acquire appropriate capacity by buying firms that
already possessed it, but this policy might prove unattractive if it entailed
taking over all the other interests to which these firms were com mitted. The
innovating firm might judge that its comparative advantage lay in
developing new products and be reluctant therefore to employ its best
managerial talents in increasing the output of old ones. It would be aware,
moreover, that not only manufacturing but marketing capability would be
needed and might properly consider that it neither possessed nor could readily
acquire this, especially in foreign countries. All these con siderations may lead
firms to seek some indirect exploitation of a product development. And, in the
case of the new process, the incentive might be even stronger in that there
might be a wide variety of fields of production in which the process could
be used.
The indirect exploitation of new technology could be sought, in terms
of our nomenclature, either through market transactions or through co
operation with other firms. But technology is a very special commodity
and the market for it a very special market. It is not always easy, in the
fi.rst place, to stop knowledge becoming a free good. The required
scarcity may have to be created artificially through a legal device, the
patent system, which establishes exclusive rights in the use or the disposal of
new knowledge. Markets may then develop in licences of right. But these
are very special markets in that the commercial freedom of those operating
within them is necessarily restricted. For suppose that A were to sell to B
for a fixed sum a licence to make a new product, but at the same time
retained the unfettered right to continue to produce and sell the product
himself. In this case the long- and short-run marginal costs of production
of the good would, for both parties, be below unit costs (because of the
fixed cost incurred by A in the development work and by Bas a lump sum
paid for the licence) so that un restrained competition would drive prices to
unremunerative levels. It might at first seem that this danger could be
avoided iflicences were charged for as a royalty on sales, which, unlike a
fixed sum, would enter into variable costs. But the licensee might still
require assurance that the licensor, un burdened by this cost element,
would not subsequently set a price disadvan tageous to him or even license
to others on more favourable terms. These dangers could be avoided if the
parties were to bind themselves by price or market-sharing agreements or
simply by the prudent adoption of the policy

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
894 THE ECONOMIC JOURNAL [SEPT.

oflive and let live. But, in one way or another, it seems likely that competi
tion would in some degree have been diminished.1
It would appear, therefore, on the basis of these considerations, that where
the creation and exploitation of technology is co-ordinated through market
transactions-transactions in licences-there will already be some measure
of co-operation between the parties. The co-operation may, of course,
amount to little more than is required not to rock, or at any rate not to sink,
the boat. But there are reasons why it will generally go beyond this.
1
Professor Arrow reaches a different conclusion. The matter is considered in his article
"Economic Welfare and the Allocation of Resources for Invention" published in The Rate and
Direction qf Inventive Activity, (edited by National Bureau of Economic Research, Princeton
University Press, 1962). Professor Arrow maintains that "an incentive to invent can exist even
under perfect competition in the product markets though not, of course, in the ' market' for the
information containing the invention" and that "provided only that suitable royalty payments can
be demanded, an inventor can profit without disturbing the competitive nature of the industry."
The issue is simplest in the case ofa cost-saving invention. Professor Arrow considers a product
made under constant costs both before and after the invention and shows how the inventor can
charge a royalty that makes it just worth while for firms making the product to acquire a licence.
On the face of it one might then conclude that the licensor would have no need to bind himself
not to reduce price below the level that provided licensees with a normal profit or to re-license for a
lesser royalty, for, if he were to do either of these things, existing licensees would make losses, stop
producing and therefore discontinue royalty payments. But this conclusion is valid only under the
highly special assumption of there being no fixed costs. For firms will in general continue in pro
duction so long as price does not fall below variable costs. Thus the licensor could find it in his
interest, having sold as many licences as he could at the higher royalty, to license others at a lower
royalty, or to enter the market himself. He would thus extend the market for the product and in
crease his earnings provided, of course, that price were kept above variable costs and therefore high
enough to induce the original (and by then no doubt aggrieved) licensees to stay in business. It is
true, of course, that in the long run fixed plant would wear out and firms deprived of their quasi-rents
would cease producing, but the fact that an opportunity for exploitation is merely temporary
does not warrant our assuming that it will not be seized. In general the licensor would stand to
gain by" cheating "the licensees in the manner described and the latter would therefore want some
measure of assurance (which need not be formal) that he would not do so. There would be a
market for licences, that is to say, only if the commercial freedom of the licensor were in this way
reduced.
It may be that Professor Arrow would not consider this to represent a significant restriction of
competition; and indeed the important practical issue concerns the manner and degree in which
the parties accept limitations on their freedom of action. I have suggested that the licensor would
be in a position, having licensed other firms, subsequently to deprive them of expected profits. A
firm will therefore seek a licence only ifit believes that this will not happen, but it may consider that
sufficient assurance is provided by the fact that the licensor, in his own long-run interest, will not
wish to acquire the reputation for such sharp practice. Much the same situation obtains in the
context of the relationship between a large purchaser and a small supplier. Marks and Spencer,
having offered attractive enough terms to induce the blanket manufacturer to devote a large pro
portion of his capacity to meet its needs, might subsequently press for a price reduction that left him
with a poor return. The hapless supplier, in the short run at any rate, might have no option but to
give way. But although the purchaser could thus act, it could scarcely be in his own long-run
interest to acquire the reputation for doing so.
The upshot would therefore seem to be this. A market for licences can function only if the
parties to the transactions accept some restraints, but, in certain circumstances, no more restraint
might be required than enlightened self-interest could be depended upon by itself to ensure. In
practice, of course, licensing arrangements are commonly associated with much more-and often
more formal-restraint of trade, the extent of which may or may not be greater than is necessary
for the transfer of technology to take place.

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
1972] THE ORGANISATION OF INDUSTRY 895

Technology cannot always be transferred simply by selling the right to use

a process. It is rarely reducible to mere information to be passed on but
consists also of experience and skills. In terms of Professor Ryle's
celebrated distinction, much of it is" knowledge how" rather than"
knowledge that." Thus when one firm agrees to provide technology to
another it will, in the general case, supply not only licences but also
continuing technical assistance, drawings, designs and tools. At this stage
the relation between the firms becomes clearly co-operative and although,
at its inception, there may be a giver and a receiver, subsequent
development may lead to a more equal exchange of assistance and the
pooling of patents. Arrangements of this kind form an important part of
the networks of co-operation and affiliation to which I have made such
frequent reference.

VII
This article began by referring to a vision of the economy in which firms
featured as islands of planned co-ordination in a sea of market relations.
The deficiencies of this representation of things will by now be clear. Firms
are not islands but are linked together in patterns of co-operation and affilia
tion. Planned co-ordination does not stop at the frontiers of the individual
firm but can be effected through co-operation between firms. The dicho tomy
between firm and market, between directed and spontaneous co ordination, is
misleading; it ignores the institutional fact of inter-firm co operation and
assumes away the distinct method of co-ordination that this can provide.
The analysis I presented made use of the notion of activities, these being
understood to denote not only manufacturing processes but to relate equally
to research, development and marketing. We noted that activities had to
be undertaken by organisations with appropriate capabilities. Activities
that made demands on the same capabilities were said to be similar; those
that had to be matched, in level or specification, were said to be comple
mentary. Firms would find it expedient, for the most part, to concentrate
on similar activities. Where activities were both similar and comple
mentary they could be co-ordinated by direction within an individual busi
ness. Generally, however, this would not be the case and the activities to
be co-ordinated, being dissimilar, would be the responsibility of different
firms. Co-ordination would then have to be brought about either through
co-operation, firms agreeing to match their plans ex ante, or through the
processes of adjustment set in train by the market mechanism. And the
circumstances appropriate to each of these alternatives were briefly dis cussed.
Let me end with two further observations. I have sought to stress the
co-operative element in business relations but by no means take the view that
where there is co-operation, competition is no more. Marks and Spencer

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms
896 THE ECONOMIC JOURNAL [SEPT. 1972]

can drop a supplier; a sub-contractor can seek another principal; technical

agreements have a stated term and the conditions on which they may be
re-negotiated will depend on how the strengths of the parties change and
develop; the licensee of today may become (as the Americans have found in
Japan) the competitor of tomorrow. Firms form partners for the dance but,
when the music stops, they can change them. In these circumstances
competition is still at work even if it has changed its mode of operation.
Theories of industrial organisation, it seems to me, should not try to do
too much. Arguments designed to prove the inevitability of this or that
particular form of organisation are hard to reconcile, not only with the
differences between the capitalist and socialist worlds, but also with the
differences that exist within each of these. We do not find the same
organ isation of industry inJugoslavia and the Soviet Union, or in the
United States andJapan. We ought to think in terms of the substitutability
of industrial structures in the same way as Professor Gerschenkron has
suggested in relation to the prerequisites for economic development. It
will be clear, in some situations, that co-ordination has to be accomplished
by direction, by co-operation or through market transactions, but there will
be many others in which the choice will be difficult but not very important.
In Great Britain, for example, the artificial textile industry is vertically
integrated and the manufacturers maintain that this facilitates co-ordination
of production and development. In the United States, on the other hand,
anti-trust legislation has checked vertical integration, but the same co-
ordination is achieved through close co-operation between individual firms
at each stage. It is important, moreover, not to draw too sharp lines of
distinction between the techniques of co-ordination themselves. Co-
operation may come close to direction when one of the parties is clearly
predominant; and some degree of ex ante matching of plans is to be found
in all markets in which firms place orders in advance. This points,
however, not to the invalidity of our triple distinction but merely to the
need to apply it with discretion.1
G. B. RICHARDSON
St. John's College,
Oxford.
1
In his article," The Nature of the Firm," Economica, 1937, pp. 386-405, R.H. Coase explains
the boundary between firm and market in terms of the relative cost, at the margin, of the kinds of
co-ordination they respectively provide. The explanation that I have provided is not inconsistent
with his but might be taken as giving content to the notion of this relative cost by specifying the
factors that affect it. My own approach differs also in that I distinguish explicitly between inter
firm co-operation and market transactions as modes of co-ordination.

This content downloaded from

130.238.7.40 on Wed, 23 Nov 2022 20:45:03 UTC
All use subject to https://about.jstor.org/terms