Kieran McKeown Auth. Marxist Political Economy and Marxist Urban Sociology A Review and Elaboration of Recent Developments

MARXIST POLITICAL ECONOMY AND MARXIST
URBAN SOCIOLOGY
Marxist Political
EconoInY and Marxist
Urban Sociology
A Review and Elaboration of Recent Developments
Kieran McKeown
Social Research Centre
National Institute/or Higher Education
Limerick, Ireland
Palgrave Macmillan
ISBN 978-1-349-18569-6 ISBN 978-1-349-18567-2 (eBook)
DOI 10.1007/978-1-349-18567-2
o Kieran McKeown, 1987

Soficovcr rcprint of thc hardcovcr 1st edition 1987 978-0-333-41382-1
All righLSreserved. For inronnation. write:
Scholarly& Referern:z Division,
St. Martin·sPress, Inc., 175 Fifth Avenue, New York, NY 100[0
First published in the United Statcsof America in 1987
ISBN 978-0-312-51794-6
Library of Congress Cataloging.in.Publication Data
McKeown. Kieran, 1951-
Marxist political economy and Mar:l:ist urban
sociology.
Bibliography: p.
Includes inde:l:.
1. Marxian economics. 2. Business cycles.
3. Marxianschoolofsociology. 4. Sociology,
Urban. i. Title.
HB97.5.M365 [987 335.4 86- [5544
ISBN 978·0-312-51794-6
To Grace, Calm, Clara and Catriona
Contents
List a/Tables and Figures Xlll
Acknowledgements xv
General Introduction XVll
PART I MARXIAN POLITICAL ECONOMY
1 Marx's Labour Theory of Value 3

1.0 Introduction 3
1.1 Marx's Labour Theory of Value 3
1.2 The Transformation Problem 7
1.3 Marx's Solution to the Transformation Problem 9
1.4 The Errors in Marx's Transformation Solution 12
1.5 The Correct Solution to the Transformation
Problem 14
1.6 The Sraffian Approach to Prices and Profit 18
1. 7 Conclusion 23
2 Marxian Crisis Theory 24

2.0 Introduction 24
2.1 The Law of the Falling Rate of Profit 26
2.1.1 Introduction 26
2.1.2 Marx's Formulation ofthe Law 27
2.1.3 The Possibility of a Fall in the Maximum Rate
of Profit 29
2.1.4 The Possibility of a Fall in the Actual Rate of
Profit 30
2.1.5 Technical Change and the Rate of Profit 33
2.1.6 Conclusion 35
vii
viii Contents
2.2 Realisation Crises 35

2.2.2 The Possibility of a Realisation Crisis 36
2.2.3 Realisation Crisis Caused by Absence of
Macroeconomic Planning 38
2.2.4 Realisation Crisis Caused by Technical
Innovation. 42
2.2.5 Realisation Crisis Caused by Variable Weather
Conditions 44
2.2.6 Realisation Crisis Caused by Scarcity of
Labour 46
2.2.7 Conclusion 47
2.3 Crises Caused by Rising Real Wages 47
2.3.2 The Difference between Labour Power and a
Commodity 48
2.3.3 The Mechanism Regulating the Level of Wages 50
2.3.4 The Long-term Trend in Wages 50
2.3.5 The Role of Trade Unions in Wage
Determination 53
2.3.6 Conclusion 54
2.4 Conclusion 55
3 Marx's Theory of Rent 56

3.0 Introduction 56
3.1 Differential Rent I 58
3.2 Differential Rent II 61
3.3 Monopoly Rent I 64
3.4 Monopoly Rent II 67
3.5 Conclusion 68
4 Marx's Theory of Productive and Unproductive Labour 70

4.0 Introduction 70
4.1 'Definition I' of Productive and Unproductive
Labour 70
4.2 'Definition II' of Productive and Unproductive
Labour 72
4.3 Conclusion 76
Contents IX
PART n FRENCH MARXIST URBAN SOCIOLOGY: THE

CASE OF MANUEL CASTELLS AND JEAN LOJKINE
5 The Different Marxist Traditions of CasteUs and Lojkine:

Althusserianism and State Monopoly Capitalism 81
5.0 Introduction 81
5.1 The Althusserian Framework 81
5.2 The Theory of State Monopoly Capitalism 85
5.3 Conclusion 88
6 The Marxist Urban Sociology of Manuel Castells 89

6.0 Introduction 89
6.1 The Critique of Urban Sociology 90
6.1.2 Science' and 'Ideology' in Urban Sociology 90
6.1.3 Urbanism 92
6.1.4 Urbanisation 94
6.1.5 Conclusion 95
6.2. Collective Consumption 95
6.2.2 Defining an Urban Area as an Area of
Collective Consumption 96
6.2.3 The Effects of Collective Consumption 98
6.2.4 Conclusion 101
6.3 The Urban System 101
6.4 Urban Politics 104
6.5 Urban Planning 105
6.5.2 The Typology of Urban Actors 106
6.5.3 The Causes of and Constraints upon Urban
Planning. 106
6.6 Urban Social Movements 109
6.6.2 The Definition of an Urban Social Movement 109
6.6.3 A Framework for the Analysis of Urban Social
Movements III
6.7 Empirical Research 114
x Contents
6.7.2 Research on Industrial Location in the Paris

region 115
6.7.3 Research on Urban Renewal in Paris 118
6.7.4 Research on Protests against Urban Renewal in
Paris 121
6.7.5 Research on Urban Protests in Montreal 123
6.7.6 Research on Urban Development in Dunkirk 125
6.7.7 Research on Immigrant Workers in Western
Europe 134
6.7.8 Research on Urban Protests in Paris 136
6.8 Conclusion 139
7 The Marxist Urban Sociology of Jean Lojkine 141

7.0 Introduction 141
7.1 The Definition of an Urban Area 142
7.2 The Collective Means of Consumption 143
7.2.2 The 'Collective' Nature of the Collective Means
of Consumption 143
7.2.3 The 'Unproductive' Nature of the Collective
Means of Consumption 144
7.2.4 The Unprofitable Nature of the Collective
Means of Consumption 147
7.2.5 The 'Durable', 'Immobile' and 'Indivisible'
Nature of the Collective Means of
Consumption 148
7.3 The Three Problems associated with Capitalist
Urban Development 150
7.3.2 The Problem of Financing Urban Expenditure 150
7.3.3 The Problem of the Locational Strategy of
Capitalist Firms 154
7.3.4 The Problem of Urban Land Rent 154
7.4 Urban Planning 160
7.5 Social Movements 163
7.6 Empirical Research 167
7.6.2 Research on Urban Policy in Paris in the
Period 1945 - 72 167
Contents xi
7.6.3 Research on Urban Policy in Lyon in

Period 1945~ 72 174
7.7 Conclusion 180
PART 01 AN ALTERNATIVE APPROACH TO MARXIST
URBAN SOCIOLOGY
8 From Marxist to Post-Marxist Urban Sociology 185

8.0 Introduction 185
8.1 The Marxist Urban Sociology of Manuel Castells
and Jean Lojkine 185
8.l.l Introduction 185
8.1.2 The Definition of an Urban Area Contained in
the Writings of Castelis and Lojkine 185
8.1.3 The Analysis of Collective Consumption in the
Writings of Castelis and Lojkine 187
8.1.4 The Analysis of Urban Planning in the
8.1.5 The Analysis of Urban Movements in the
8.2. A Conflict Approach to Urban Sociology
The Constrained-Actor Model 192
8.3 Urban Areas and Urban Processes: The Field of
Urban Sociology 195
8.4 The Production of Urban Areas: The Process of
Property Development 196
8.5 State Property Development 197
8.6 Private Property Development 200
8.6.2 Constraints upon the Interests of Private
Property Developers 200
8.6.3 Constraints upon the Interests of Landowners 204
8.6.4 Constraints upon the Interests of the State
Planning Authority 205
8.6.5 Summary 207
8.7 Production in an Urban Area 208
8.7.2 The Supply of Labour in an Urban Area 208
8.7.3 The Demand for Labour in an Urban Area 211
8.7.4 Summary 213
xu Contents
8.8 Consumption in an Urban Area 214

8.8.2 A Typology of Consumption Goods 215
8.8.3 The Spatial Dimension of Consumption 217
8.8.4 The Dwelling Unit 220
8.8.5 Tenure Groups and Changes in the Site Value
of Dwelling Units 222
8.8.6 Individual and Collective Reaction to Changes
in Consumption 223
8.8.7 Resignation 225
8.8.8 Migration 225
8.8.9 Collective Action 226
8.8.10 Summary 229
8.9 Conclusion: New Directions for Marxist Urban
Sociology 229
Notes 231
Bibliography 253
Name Index 274
Subject Index 277
List of Tables and Figures
TABLES
4.1 Marx's 'Definition I' of productive and unproductive labour 73
4.2 Marx's 'Definition II' of productive and unproductive labour 75
6.1 Castells's concept of the urban system 103
6.2 The sub-elements in Castells's urban system 104
6.3 Castells's typology of urban actors 106
6.4 Castells's typology of practices and their effects 110
6.5 Castells's typology of industrial establishments 117
6.6 The index of differentiation for 23 renewal areas in Paris on
selected variables 119
6.7 Correlations between the areal size of renewal operations and
selected variables in Paris 120
6.8 Some characteristics of Dunkirk and the Nord Region
according to Castells 128
6.9 Castell's description of the fractions of the capitalist class and
their interests in Dunkirk 130
6.10 Castells's description of the fractions of the working class and
their interests in Dunkirk 131
6.11 Castells's typology of urban protests in Paris, 1968-1973 137
7.1 Lojkine's analysis of different types of rent in different urban
situations 156
8.1 The constraints and conflicts generated in the process of
private property development: sources of constraints on actors
in the development process 202
8.2 A typology of consumption goods 216
8.3 A typology of the reactions of residents to a negative
consumption good in their area 224
Tables in Notes
Nl Number of citations per year of the works of Manuel Castells

and Jean Lojkine and the average number of citations of each
cited author in the Social Science Citation Index (1969-84) 231
N2 A typology ofland 238
N3 A typology of private and public goods 250
xiii
XIV List of Tables and Figures
FIGURES
1.1 The relationships between physical value and price quantities 22

2.1 The relationship between wages and profit for different
techniques of production 34
2.2 The shares of profits and wages in net output 51
2.3 Marx's view of the effect of an increase in net output upon the
shares of profits and wages 53
3.1 The relationship between output and type ofland 59
3.2 The relationship between output, type ofland and rent 61
3.3 The relationship between output, technique of production and
rent 63
3.4 The incidence of monopoly rent I on different types ofland 65
6.1 The determinants of urban culture according to Castells and
Wirth 93
8.1 The process of property development 197
8.2 The externality gradient of a negative communal consumption
good 218
8.3 The externality field of a negative communal consumption
good 218
8.4 The externality gradient of a consumption good with both
positive and negative effects 219
8.5 The externality field of a consumption good with both positive
and negative effects 219
Figures in Notes
NI The growth of population and resources according to Malthus 237

N2 Reasons for household relocation 249
Acknowledgements
I am very grateful to Chris Pickvance of the University of Kent at
Canterbury for his painstaking and constructive comments on earlier
drafts of this book. He is not, of course, responsible for any errors that
remain nor does he necessarily share all the views expressed in it. I would
also like to thank Mary O'Malley for her careful typing at every stage in
the preparation of the book.
KIERAN McKEOWN
xv
General Introduction
Marxist urban sociology, though a relatively recent development, is
currently a major school within urban sociology (see Lebas, 1982, for a
comprehensive bibliography of the literature constituting this school).
The school emerged in the late 1960s largely as a result of growing
academic disenchantment with the prevailing approaches to urban
sociology. The new approach - sometimes referred to as 'the political
economy approach' (Harloe, 1979, p. 5; Lebas, 1981, p. xi; 1982, p. 20)-
is now entering its second decade and this is an appropriate time to
evaluate it. This book focuses upon the writings of Manuel Castells and
Jean Lojkine, who have made centrally important contributions to this
new body of literature while at the same time representing different
approaches within it. Thus their work represents, to some extent, both
the unity and the diversity of the new school.
Manuel Castells and Jean Lojkine are sociologists whose writing has
attempted to apply a Marxist perspective to the analysis of urban areas.
Throughout the 1970s and early 1980s their writings have generated
considerable and growing interest within the social sciences generally. I
In the present study their writings will be analysed and discussed in
detail. Their work will be located in their respective traditions of
Marxism and evaluated from the point of view of Marx's political
economy. Having identified inadequacies in their understanding and
application of Marx's political economy, the book will attempt to
provide an alternative to the approach of Castells and Lojkine. This will
be done by redefining the field of urban sociology in terms of three basic
processes, namely, the process by which urban (that is, built-up) areas
are produced, the process by which they are used for the production of
commodities and the process by which they are used for final consump-
tion. These three processes will be analysed using a constrained-actor
model of society. This results in an analysis of urban processes which,
although it is not strictly Marxist, nevertheless maintains the Marxian
emphasis upon the role of interests and conflicts in social interaction.
The critique of the work of Castells and Lojkine which is presented in
this book is made from a particular point of view based upon the premise
that the validity of a Marxist analysis of urban areas is ultimately
dependent upon the validity of Marxism itself. Since, in turn, the validity
of Marxism is dependent upon Marxist political economy, the critique
of Castells and Lojkine will focus upon the major issues of political
xvii
XVlll General Introduction
economy which are suggested, implicitly or explicitly, by their work.

The book is written in three parts. Part I, containing Chapters I, 2, 3
and 4, provides an overview of Marxian political economy. Marx's
theories of value, crisis, rent and productive labour are analysed in
detail. It will be shown that significant difficulties exist with virtually
every substantive aspect of Marx's political economy. This finding will
be the basis for appraising the way in which Castells and Lojkine use this
political economy.
Part II, containing Chapters 5, 6 and 7, examines the French Marxist
urban sociology of Manuel Castells and Jean Lojkine. The examination
begins in Chapter 5 by locating the work of Castells and Lojkine within
their respective traditions of Marxism. In the case of Castells that
tradition is normally referred to as 'Althusserianism' and in the case of
Lojkine it is 'State Monopoly Capitalism'. The purpose ofthis chapter is
less to evaluate these traditions than to provide the reader with the
necessary background information and terminology for understanding
the work of Castells and Lojkine. The main thrust and focus of the
critique is contained in Chapters 6 and 7 where the entire works of
Castells and Lojkine are discussed successively. In each ofthese chapters
the strategy is to outline and evaluate the conceptual framework of each
author and follow this with an assessment of his empirical research.
Part III, containing Chapter 8, presents an alternative approach to
Marxist urban sociology. This alternative, which avoids the major
difficulties associated with the work of CastelIs and Lojkine, begins from
the premise that urban areas are essentially built-up areas. A con-
strained-actor model of society is then applied to understand how these
built-up areas are produced in a capitalist society and how, in turn, they
are used both for the production of commodities and for final
consumption. This new formulation of the field of urban sociology
contains an analysis of urban space which tries to advance our
understanding of the constraints and conflicts generated by the
production and use of urban areas in a capitalist society.
Part I
Marxian Political Economy

1 Marx's Labour Theory of
Value
1.0 INTRODUCTION
The purpose of this chapter is to provide an over-all assessment of

Marx's labour theory of value. The analysis begins by outlining the main
elements of Marx's labour theory of value and by defining its central
concepts (section 1.1). Marx regarded these concepts as essential to
understanding prices and profit in a capitalist society and it is for this
reason that the problem arises of transforming values and surplus value
into prices and profit. The precise nature of this problem will be outlined
in section 1.2, along with Marx's attempt to solve it (section 1.3), and the
errors associated with that solution (section 1.4). The correct solution to
the transformation problem will then be presented in section 1.5
The analysis will then show that, despite the formal correctness of the
transformation procedure, the labour theory of value does not provide
the only or even the best explanation of the formation of prices and
profit within a capitalist society. The reason for this is that there is an
alterative to the labour theory of value which treats the same issues in a
more rigorous and less problematic way. This is the Sraffian approach
whose main features will be outlined in section 1.6.1t will be seen that the
weight of the arguments is in favour of the adoption of the Sraffian
approach and the rejection of the labour theory of value.
1.1 MARX'S LABOUR THEORY OF VALUE
The purpose of Marx's labour theory of value is to explain how the

wealth of a capitalist society is produced and distributed. Since, in
Marx's view, this wealth consists of commodities he draws the logical
inference that 'our investigation must therefore begin with the analysis
of a commodity' (Marx, 1974a, p.43.).
Marx defines a commodity as a combination of two independent
characteristics: on the one hand it is a use value, and on the other it is an
exchange value. Commodities are use values in the sense that their
consumption satisfies certain wants, and are exchange values in the sense
that they are products oflabour and exchange according to the quantity
3
4 Marxian Political Economy
of sociallyl necessary abstract labour embodied in them. Every com-

modity, according to Marx, must have these two characteristics (Marx,
1974a, Chapter 1, section 1, pp.43-8).
Marx pointed out that labour, or rather labour power (to use his
preferred term - see notably Marx 1975, section 7, pp. 43 - 6)2 was also a
commodity within capitalist society and was therefore also a use value
(in the sense that its performance led, directly or indirectly, to the
satisfaction of some want) and an exchange value (in the sense that it too
was the product of the socially necessary abstract labour required to
produce it and therefore exchanged accordingly). Marx argued that the
exchange value oflabour power was equal to the bundle of commodities
forming the wage since these were necessary for the subsistence, and
hence the production, of labour power (see section 2.3).
The problem which Marx then confronts is to show how it is possible
for the exploitation oflabour to occur when every commodity (including
labour power) is exchanged according to the quantity of socially
necessary abstract labour required to produce it. This problem arises
because Marx characterises capitalist production in the following way:
M cf{p -lEI - C - M'
where:
M = the amount of money which the capitalist owns at the
beginning of the production process;
CLP
MP = the commodities of labour power and means of production
purchased by the capitalist at the beginning of the production
process;
lEI = the production process;
C = the new commodity
M' = the money which the capitalist receives from the sale of the
new commodity;
M' = M + 11M>.
The typical process of capitalist production, as Marx saw it, is that, a
capitalist invests money (M) with a view to making a profit. He does this
by purchasing a certain quantity of labour power and means of
production based upon the technical requirements and the scale of the
production process. These two sets of commodities (cffp) are then put to
production (lEI) to produce a given output (C). This output is then sold
and the capitalist recoups his initial investment (M) plus a profit or
surplus (11M). If there was no surplus at the end of the production
Marx's Labour Theory of Value 5
process, then the entire exercise would be pointless from the capitalist's
point of view. Thus the pursuit of profit is the underlying rationale of all
capitalist production.
Marx's problem is to explain the origin and nature of this surplus in a
way which is consistent with the proposition that all commodities
exchange in proportion to their value. His explanation is that the surplus
arising in production is due to the commodity labour power which he
regarded as a 'special commodity' (Marx, 1974a, p. 164). The special
quality oflabour power derives from the fact that its use value can create
a quantity of value which is greater than its exchange value (in contrast
to every other commodity whose use value can only create a quantity of
value which is equal to its exchange value). Marx explained this special
quality of labour power in the following way: the capitalist purchases
labour power at its value and then proceeds to consume its use value by
putting it to work (in the same way as he/she will consume the use value
of the means of production by putting them to work). If the capitalist is
to make a profit from the purchase of labour power he/she must ensure
that the value of the output created is greater than the exchange value of
that labour power. Marx emphasised that capitalists will always ensure,
through various forms of coercion, that the value created by the use
value oflabour power will always be greater than its exchange value. It is
this difference between what Marx terms 'the value oflabour power and
the value which that labour power creates' (ibid., p. 188) which is the
origin of profit. The size of the profit, in turn, is determined by the
productivity of labour and by the length and intensity of the working
day.
This is the theroretical core of Marx's theory of exploitation. This
theory highlights two important points about capitalist exploitation.
The first is that capitalists receive an unearned income, that is, a profit
which is based solely upon their ownership of the means of production,
in contrast to workers whose income has to be earned through work.
The second point is that the income which the capitalist receives consists
in the surplus labour which the worker is obliged to perform over and
above that which is necessary to produce the wage. Thus, in his theory of
exploitation, Marx was not content to point out simply that workers
earned their share of society's wealth while capitalists did not, but in
addition, that the capitalists' share was nothing less than the surplus
labour of the workers.
This basic conceptual framework is then elaborated by introducing
the concepts of constant capital (C), variable capital (V) and surplus
value (S) where:

C = the value of the means of production used in the production
process;3
V = the value of the labour power used in the production process;
S = the value of the surplus output produced in the production process.
Marx explained the connections between these three concepts by
asserting that, in the production process, the value of constant capital is
simply transferred to the final output while the value of variable capital,
in addition to being transferred to the final output, also creates
additional value, equal to surplus value, which, itself, is also transferred
to the final output. Thus, the value of any output is equal to Ci + Vi + Si'
where iis any commodity between 1 and n (see Marx, 1974c, Chapter 8,
p. 150).4 Thus, if two commodities have the same value they can both be
exchanged for one another or for the same amount of money. The
amount of money, following Marx's analysis (see Marx, 1974a, pp. 74ft)
will be a quantity of gold whose production required exactly the same
amount of direct and indirect labour to produce as either of the two
commodities with which it exchanges. s
Marx continues the analysis by deriving two important ratios. The
first is:
~ = the organic composition of capital in sector i where i is any

Vi sector between 1 and n.
This ratio is a measure of the capital or labour intensity of a particular
sector (Marx, 1974a, Chapter 25, pp. 574ff; 1974c, Chapter 8, pp. 145ft)
and is assumed to vary between sectors so that:
where i and j refer to any sector between 1 and n.

The second ratio is:
~ = the rate of surplus value or exploitation in sector i, where i

Vi is any sector between 1 and n.
This ratio is a measure of the quantity of surplus value produced per unit
oflabour power. It can be raised by increasing 'absolute surplus value'
(that is, by increasing the length and intensity of the working day:
increasing the numerator) or by increasing 'relative surplus value' (that
is, by increasing the level of productivity in the wage-goods sector:

reducing the denominator) (see Marx, 1974a, p.299). Marx assumed
(though see Desai, 1979, p. 51) that the rate of surplus value was uniform
throughout all sectors of the economy since, he claimed, the reproduc-
tion of labour power is the same for all workers (that is, a uniform
subsistence wage) while the length and intensity of the working day also
tends to uniformity. Thus:
~=!L
Vi Vj
where i and j refer to any sectors between I and n.

The important relationship between the organic composItion of
capital and the rate of surplus value is that those sectors with a high
organic composition of capital will produce less surplus value than those
sectors with a low organic composition of capital. This result, which
follows consistently from the preceding analysis, is highly significant
because, as will now be seen, it gives rise to the transformation problem.
1.2 THE TRANSFORMA nON PROBLEM
The analysis up to this point has centred around the concepts of the
value of a commodity and the surplus value generated in the course of
producing that commodity. Marx regarded this analysis as a fundamen-
tal prerequisite to the explanation of profit and prices in capitalist
society. This is indicated implicitly in the layout of Capital (broadly
speaking from values and surplus value in Volume I to prices and profit
in Volume III) and explicitly by the following quotation: 'If you cannot
explain profit upon this supposition (that is, that on average, com-
modities are sold at their real values and that profits are derived from
selling them at their values) then you cannot explain it at all' (Marx,
1975, p.42; see also 1974c, pp.43, 157).
The logic of this dualism in his economic analysis (between value and
surplus value on the one hand and price and profit on the other) required
Marx to show that there was, in fact, a systematic and quantifiable
relationship between these two levels of reality.6 To this end he
attempted to transform value and surplus value into price and profit,
thereby proving that the latter were derived from, and determined by,
the former. His attempt to do this gave rise to the transformation
problem.
This problem is the pons asinorum of Marx's economic analysis

(Robinson, 1973, p.61) and has been variously described as 'a bare
contradiction' (Bohm-Bawerk, 1949, p. 30), 'a quantitative incongruity'
(Bortkiewicz, 1952, p. 5), 'a great contradiction' (Blaug, 1962, p. 230)
and 'an interesting theoretical problem' (Sweezy, 1942, p. 69) in Marx's
work. Marx anticipated the problem as early as 1862 (see Marx and
Engels, 1955; see also Marx, 1969, Chapter 10, section 4 (a), pp. 173ft)
before he ever wrote Volumes I and III of Capital between 1863 and 1867
(see Marx, 1974a, I 974c), though he never provided a satisfactory
solution to it. It is now necessary to outline precisely the nature of the
problem which Marx encountered when he attempted to transform
value and surplus value into price and profit.
Marx defined the price of production of a commodity as equal to 'cost
price plus profit' (Marx, 1974c, pp. 36, 165). In turn he defined the rate
of profit as:
where i refers to any sector or commodity between I and n (see Marx,

1947c, pp.49ft).
It follows from this, if Marx's labour theory of value is correct, that
the value of commodity i ( = Ci + Vi + Si) must be proportionate to the
price of commodity i ( = (ci + Vi) (1 + r) ). However, this is not in fact the
case since, with equal rates of exploitation (S;/Vi = Sj/Vj) and variable rates
of profit between sectors (ri 1 rj), the value of commodities will be
greater than their prices in those sectors with a low organic composition
of capital (c)vji) and will be less than their price in those sectors with a
high organic composition of capital. In other words, Marx's analysis
leads to the result that prices are not proportional to their values which
means that there is not a consistent relationship between values and
prices. This raises the problem of the connection of values to prices - in
short, the transformation problem.
In algebraic terms, this can be expressed as follows:
if: Si = Sj
Vi Vj
then: ri1rj1r
then: Ci + Vi + Si is not proportional to (Ci + i) (I + r)
Marx acknowleged this problem by pointing out that prices could

only be proportional to values if the rate of profit varied between sectors
according to the amount of surplus value produced in each sector.
Alternatively, prices would be proportional to values if the organic
composition of capital was the same in every sector.
Marx appreciated that this result was highly anomalous since
'differences in the average rate of profit in the various branches of
industry do not exist in reality and could not exist without abolishing the
entire system of production' (Marx, 1974c, Part II, Chapter 8, p. 153).
Indeed, at this point Marx contemplated the possibility - largely for
rhetorical effect given his intended and subsequent argument - that 'the
theory of value is incompatible with the actual process' (ibid.)
It is now necessary to examine Marx's solution to this problem.
1.3 MARX'S SOLUTION TO THE TRANSFORMATION

PROBLEM
There are essentially two components involved in the transformation of

value and surplus value into price and profit. The first component
involves transforming the values of constant and variable capital into
their corresponding cost price; the second component involves transfor-
ming surplus value into profit. Marx largely neglected the first
component: in particular, he did not indicate precisely how the value of
the inputs of constant and variable capital might be transformed into
their corresponding prices even though he recognised that they needed
to be transformed (Marx, 1974c, pp. 159-64). In his actual numerical
examples he assumed that the value and price of these inputs were the
same.
Marx concentrated most of his energies on the second component
since, in his formulation of the transformation problem, this is the major
source of the difficulty. In his solution he focused on the problem of how
to ensure that a uniform rate of profit existed throughout the economy
in a way which was consistent with the basic proposition that this profit
was derived from, and determined by, surplus value.
Marx's solution involved changing his definition of the rate of profit
and working out the ramifications of this change. Thus, in place of the
definition of the rate of profit in the previous section:
r = __
Si_
Ci + Vi
Marx put forward the following:
I
i=l
Si
S
r = =---
n
c+ V
I + I
i=! Ci
i=! Vi
This definition of the general rate of profit in terms of the arithmetic

average rate of profit seemed to offer Marx the solution he required
because it appears to clarify the relationship between surplus value and
profit as well as the relationship between values and prices. According to
this definition, the aggregate quantity of surplus value equals the
aggregate quantity of profit, although the quantity of surplus value
produced in each sector is not necessarily equal to the quantity of profit
appropriated in that sector (except in the singular case where the organic
composition of capital in a sector is equal to the social average). This fact
can, following the above notation, be written as follows:
it! Si = [r it! (Ci+ V) ]
but
Marx seems to have conceptualised the relationship between surplus

value and profit along the following lines: surplus value is produced in
each sector and contributes to the total fund of surplus value. This, in
turn, is distributed as profit, not according to the amount produced in
each sector, but according to the total capital invested in each sector.
Baumol captures Marx's point when the describes the production and
distribution of surplus value in each sector as follows: 'It takes from each
according to its workforce, and returns to each according to its total
investment' (Baumol, 1974, p. 53). Marx used the analogy of the joint
stock company (Marx, 1974c, p. 158) to illustrate the point that each
sector or capitalist (the two can be treated as synonymous in this
context) receives a share of profit in relation to its total investment,
though the analogy breaks down in relation to the contribution of each
to the total fund of surplus value.
The actual mechanism which ensures that surplus value is distributed
as profit is competition (Marx, 1974c, p. 158, 174ft). The competitive
process ensures that a uniform rate of profit operates throughout the
economy which regulates the distribution of surplus value irrespective of

the amount which each produces.
Marx claimed that the particular relationship between surplus value
and profit was discovered by him and 'revealed for the first time' (Marx,
1974c, p. 168) in Chapter 9 of Volume III of Capital. Moreover, it was a
discovery which, as Marx saw it, was particularly difficult since the rate
of profit acts unintentionally to 'obscure and mystify the actual origin of
surplus value' (Marx, 1974c, p. 167). This is because the capitalist
calculates the rate of profit on the basis of his total investment (Ci + Vi)
whereas surplus value originates from only part of that investment (v;).
Thus the transformation of surplus value into profit (and, by implica-
tion, the transformation of values into prices) is 'a complete mystery to
the individual capitalist; all the more so, since no bourgeois theorists, the
political economists, have so far revealed it' (Marx, 1974c, p. 170).
Marx's redefinition of the rate of profit also implied that the price of
each commodity (or output of each sector) is not proportional to its
value, that is,
Ci + Vi + Si is not proportional to (Ci + v;) (1 + r)
The reason for this is that those commodities which are produced in
sectors with a low organic composition of capital (low C;/Vi) will have
prices which are less than their value (since they produce more surplus
value than they receive as profit), while those commodities which are
produced in sectors with a high organic composition of capital (high C;/
Vi) will have prices which are greater than their value (since they produce
less surplus value than they receive as profit).
Marx argued that the inequalities between values and prices would
'balance out one another' (Marx, 1974c, p. 157) and 'compensate one
another' (ibid, p. 161) in the sense that the aggregate value of all
commodities would equal the aggregate price of all commodities, even
though the value and price of each individual commodity would be
unequal.
This proposition can be written, following the above notation, as
follows:
n
I (Ci+Vi+S i )
i= I
but
Ci + Vi + Si =1= (C i + Vi) (l + r)
Marx described this as a 'generallaw' (Marx 1974c, p. 161) which acts

as a 'prevailing tendency' (ibid.) and in 'a very complicated and
approximate manner as a never ascertainable average of ceaseless
fluctuations' (ibid.).
It is now necessary to examine the validity of Marx's solution to the
transformation problem.
1.4 THE ERRORS IN MARX'S TRANSFORMATION

SOLUTION
Any solution to the transformation problem effectively involves trans-

forming each of the components of the value of a commodity
(c i + Vi + Si) into its corresponding price components. Marx's strategy,
as has been seen, was first to transform surplus value into profit and then
to transform constant capital and viable capital into cost price (though
he tended to neglect this latter aspect). In assessing Marx's solution, it is
necessary to evaluate both of these aspects.
Marx defined the rate of profit as S/( C + V). This formula gives the
value (not the price) rate of profit since each ofthe aggregates C, Vand S
are values. However, the real and actual rate of profit in capitalist society
is the price rate of profit since this is the one upon which capitalists base
their economic calculations and decisions (see Steedman, 1973, p.37;
Hodgson 1974b, p.363). It follows, therefore, that Marx's formula is
only correct if the values of S, C and V are proportional to their price.
This is the implicit assumption in Marx's formula.
Proceeding from this formula, Marx then derives the result that values
and prices are not proportional (as a result of the way in which the rate of
profit distributes surplus value according to the total capital invested in
each sector/commodity). This result clearly contradicts Marx's initial
assumption concerning the proportionality of values and prices. His
entire argument is therefore inconsistent. Steedman succinctly summar-
ises this point as follows: 'Marx's argument is internally inconsistent. He
assumes that S/( C + V) is the rate of profit but then derives the result
that prices diverge from values, which means precisely in general, that S/
(C + V) is not the rate of profit (Steedman, 1977, p. 31; see also Desai,
1979, p: 69).
Marx did not recognise this as an inconsistency per se. This was
because, even though prices deviated from values, the sum of these
deviations was equal to zero, and therefore the aggregate formula for the
rate of profit appeared to be valid. This argument is invalid, however,
since it involves circular reasoning. It is circular because the only reason

why the deviations sum to zero is that this is necessarily built into the
formula S/( C + V): Marx's formula assumes the total values of constant
capital ( I.
;=1
Ci), variable capital ( I.
i= 1
Vi) and surplus value ( I.
i= 1
Si) equal
their respective total prices (see Bortkeiwicz, 1952, p. 10). It is therefore

arithmetically inevitable that the deviations of price from value must
sum to zero, simply because of the way that Marx defined the rate of
profit. The former cannot therefore, under pain of circular reasoning, be
used to justify the latter.
The more general point about Marx's attempt to transform surplus
value into profit is that he defined, wrongly, the rate of profit in value
terms, rather than in price terms. As a result of this his explanation of the
distribution of surplus value as profit lacks any micro-foundation in the
actual behaviour of capitalists. This is the most serious error in his entire
transformation procedure.
The second aspect of the transformation solution involves transform-
ing input values (Ci + Vi) into cost prices. Marx repeatedly drew attention
to this aspect of the solution (Marx, I 974c, pp. 160, 164-5; 206-7) but
he never actually implemented it in any of his numerical examples; nor
did he indicate how it might be undertaken. As regards this aspect of
Marx's solution, therefore, it is more accurately regarded as incomplete
rather than erroneous (see Morishima and Catephores, 1978,
pp. 160-61). There is, however, another, more general, difficulty with
Marx's solution to the transformation problem. This difficulty relates to
the fact that Marx attempted to transform the various components of
the value of a commodity to their corresponding price components in a
successive rather than in a simultaneous manner. Bortkiewicz first made
this criticism of Marx's approach (echoing Marshall's criticism of
Ricardo) and gave it the term 'successivism' (Bortkiewicz, 1952, p. 24).
The main criticism here is that Marx did not fully appreciate (at least in
the case of the transformation problem) that the various components of
the value and price of a commodity determine each other mutually and
simultaneously rather than 'successively in a long chain of causation'
(ibid.). For this reason he attempted to first transform surplus value into
profit and then to transform constant and variable capital into cost
prices rather than determining both simultaneously. To do the latter it
would have been necessary to use simultaneous equations which were
then being used for the first time in economic analysis. However, Marx
was unfamiliar with simultaneous equations and this is probably the
main reason why he failed to solve the transformation problem (see
Bortkiewicz, 1952, p.24; Hodgson, 1973, p.363; Morishima 1974a,

pp. 611-13).7 As will be seen presently, all 'correct' solutions to the
transformation problem require, following Bortkiewicz, the use of
simultaneous equations. Bortkiewicz's solution will now be examined.
1.5 THE CORRECT SOLUTION TO THE TRANSFORMATION

PROBLEM
The identification of the errors in Marx's solution to the transformation

problem clearly indicates that a correct solution to this problem could be
established if the values of constant capital, variable capital, and surplus
value, were simultaneously transformed into cost price and profit
respectively. In 1907 (13 years after the publication of the third volume
of Capital) Ladislaus von Bortkiewicz produced a correct solution to the
problem along these lines (see Bortkiewicz, 1949, 1952). His solution
relies, as he acknowledged (1952, p. 22), on the earlier work of Dmitriev
(1974), a Russian mathematical economist.
Bortkiewicz begins with a three-sector model of the capitalist
economy - similar, in this respect, to Marx's model. His model is written
algebraically, using three input-output equations, to describe each of the
three sectors. The model, using the above notation, is as follows:
Sector I (Capital Goods): CI + VI + SI = CI + C2 + C3
Sector II (Wage Goods): C2 + V2 + S2 = VI + V2 + V3
Sector III (Luxury Goods): C3 + V3 + S3 = SI + S2 + S3
In order to transform this economy from various values to prices, it is
necessary that every element be transformed simultaneously. In order
to do this it is necessary to multiply each of these value elements by a
coefficient which converts its value to its corresponding price. Each
coefficient, referred to as a price coefficent, is a ratio of the price to the
value of each element. When the price coefficient of an element is
multiplied by the value of that element, it will yield the price of that
element. Borkiewicz proposes the following price coefficients:
x = the price coefficient of constant capital
y = the price coefficient of variable capital
z = the price coefficient of surplus value
In addition, in order to write these value equations as price equations, it
is also necessary to introduce the rate of profit (in price terms) which is
denoted by r.
Each of these three price coefficients, together with the rate of profit,
provide all of the necessary requirements for rewriting the above value
equations as price equations. These price equations can be written as
follows:
Sector I (Capital Goods): c\x + VIY + r(c\x + VIY) = c\x + C2X + C3X
Sector II (Wage goods): C2X + V2Y + r(c2x + V2Y) = V\Y + V2Y + Y3.Y
Sector III (Luxury Goods): C3X + V3.Y + r(c3x + V3.Y) = S\z + S2Z + sJZ
In these equations, the price coefficients (x, y, z) and the rate of profit
(r) have been introduced in order to make it formally possible to derive
the price system from the value system. However, x, y, z, and rare
unknowns in the sense that they have been added to the known datum
(C\C2C3V\V2V3S\S2S3) and can only be specified by being expressed in terms
of the known data. To do this by simultaneous equations it is necessary
that the number of unknowns equal the number of equations. In this
case there are only three equations, but four unknowns (x,y,z,r).
Bortkiewicz solved this problem by assuming that the value of the
output of Sector III was equal to its price so that its price coefficient was
equal to one (z = I). In this way the number of equations is equal to the
number of unknowns. From these equations it is possible to derive
formulas which specify the three unknowns (x,y,r) in terms of the other
known quantities (C\C2C3V\V2V3S\S2S3) which can then be inserted into the
three-sector model (Sweezy, 1942, pp. 116-120; Bortkiewitz, 1973,
pp. 202-4; Howard and King, 1975, pp.166-7).
There are two important implications which follow from Bort-
kiewicz's solution to the transformation problem. The first concerns the
degree of arbitrariness involved in that solution and the second concerns
the determination of the rate of profit. Each of these will now be
discussed respectively.
Bortkiewicz's solution demonstrates that it is possible to derive prices
and profit from values and surplus value. However, as has been seen, this
solution required one arbitrary assumption to ensure that the transfor-
med prices are absolute prices. Bortkiewicz assumed that the value of the
output of Sector III was equal to its price (that is, z = I). Since Sector III
is entirely devoted to producing the luxury good, gold, and since its
output is equal to the total quantity of surplus value produced in the
economy, it follows that total surplus value (200) is equal to total profit
(200). The important point to note about this assumption is that it does
not 'prove' that total surplus value must necessarily be equal to total
profit (as Marx believed) since this equality is only arrived at by
assuming it.
A number of subsequent solutions have been put forward along the

same lines to that proposed by Bortkiewicz (see notably, Winternitz,
1948; Meek, 1956; Seton, 1976) and each reveals that no solution to the
transformation problem is entirely free from arbitrariness. In the
solutions put forward by Winternitz and Meek it is assumed (again
arbitrarily) that total value is equal to total price (in the case of
Winternitz) and that total surplus value is equal to total profit (in the
case of Meek). Seton in his discussion of these various assumptions
refers to them as 'invariance postulates' (Seton, 1976, p. 165) essentially
because they postulate (or assume) that some quantity of the value
system is to remain invariant when transformed into the price system.
The significant point about these invariance postulates is that, as Seton
points out, 'there does not seem to be an objective basis for choosing any
particular invariance postulate in preference to all others, and to that
extent the transformation problem may be said to fall short of complete
determinancy' (ibid., p. 167).
The second implication of Bortkiewicz's solution to the transforma-
tion problem is sometimes referred to as the 'Bortkiewicz Corollary' (see
for example, Sweezy, 1942, pp. 123-5; Hodgson 1974a, pp. 366-8) and
concerns the determination of the rate of profit. Bortkiewicz's argument
is that Marx's original formula for the rate of profit (SIC+ V) would still
be incorrect even if its constituent elements (C, V,S) were transformed
into prices (Cx, Vy, Sz). In other words:
S
r =1'--
C+V
and
Sz
r =1'----
Cx + Vy
The problem with Marx's formula (and his underlying argument) can
be seen from a re-examination of the precise way in which the general
rate of profit is formed in capitalist society and the way in which this, in
turn, is enforced through the competitive process.
In any model of a competitive capitalist economy, the rate of profit
will be uniform in every sector and will be equal to the ratio of net output
to total inputs (all valued in price terms). It follows, therefore, that in the
above three-sector model, the rate of profit in each sector will be the
same (r, = r2 = r3) and will be defined as:
(c\x + C2X + C3X) - (C\X + V\Y)
(C\X + V\Y)
(V\Y + V2Y + V3Y) - (C2X + V2Y)
(C2X + V2Y)
(S\Z + S2Z + S3Z) - (C3X + V3Y)
(C3 X + V3Y)
Marx seems to have inferred from the fact that since profit is produced
in each sector that, therefore, each sector must play the same role in
determining the general rate of profit. That this is not so can be seen by
inspecting the inputs and outputs of each sector and the way in which
they affect the rate of profit.
The level of net output in any sector, assuming a given state of
technology, will be determined by the level of inputs. It follows therefore
that the determinants of the general rate of profit must be found among
the inputs (c\x + v,y; C2X + V2Y; C3X + V3Y). More specifically, the rate of
profit which will prevail in every sector of the economy will be
determined by those inputs which enter into each and every sector since
any change in the price of these inputs will affect the rate of profit in
every sector in the same way. Conversely, if an input does not enter into
every sector then any change in its price would not affect the rate of
profit in every sector; it could not therefore be regarded as a determinant
of the general rate of profit.
An inspection of the inputs in the above three-sector model reveals
that all of the inputs are produced in Sectors I and II; there are no inputs
which are produced in Sector III. It is thus possible to affirm
categorically that Sector III has no effect upon the general rate of profit
since its output does not enter as an input into any other sector (not even
itself). The rate of profit is therefore established outside this sector and
imposed upon it through the competitive process. This fact alone
provides adequate proof that Marx's formula for the rate of profit must
be wrong since it includes the inputs and outputs of this sector in its
calculation when in fact they should be excluded (see Morishima and
Catephores, 1978, pp. 161-2). More generally, the rate of profit in a
competitive capitalist economy is determined by those inputs which
enter directly and indirectly into the production of every other
commodity. In the terminology of Sraffa, it is determined by 'basic'
commodities (as opposed to 'non-basic' commodities) because 'basics
have an essential part in the determination of prices and the rate of

profits, while non-basics have none' (Sraffa, 1960, p. 54; see section 1.6).
This is the important corollary of Bortkiewicz's solution.
The Bortkiewicz solution (and all similar solutions) to the transfor-
mation problem show that it is possible, under certain assumptions, to
derive prices and profit from values and surplus value. However it does
not prove that the price of a commodity is determined by its value or that
profit is determined by surplus value. On the contrary, the transforma-
tion solution proves that there is no constant or systematic relationship
between value quantities and price quantities. Since it was precisely
systematic connections of this type which Marx hoped to establish by the
labour theory of value, it follows that the correct solution to the
transformation problem does not in fact vindicate Marx's theory. The
labour theory of value is thus an inadequate theory of profit and prices.
There is, however, an alternative to the labour theory of value which
shows clearly how means of production, wages, profits and prices are all
interrelated in a capitalist economy. This alternative, which was first
elaborated by Piero Sraffa (1960), has now become widespread in the
field of Marxist political economy (see, for example, Morishima, 1973;
Steedman, 1977; Morishima and Catephores, 1978; Abraham-Frois and
Berrebi, 1979; Bose, 1980; Steedman, et al. 1981; Hodgson, 1982;
Roemer, 1981, 1982; Elster, 1985). Some of its main tenets will now be
examined.
1.6 THE SRAFFIAN APPROACH TO PROFIT AND PRICES
Sraffa's book is concerned with the relationships between prices, profits

and wages in a competitive capitalist economy. It is thus concerned with
the same issues that are central to Marx, although it is also concerned, as
the subtitle of the book indicates, with 'a critique of economic theory'.
Sraffa, like Marx, characterises capitalism as a system where com-
modities are produced in order to make a profit. The production of
commodities, according to Sraffa, can be seen as an input-output system
where each commodity is the output which results from the input of
other commodities, both capital and labour. In the course of the
production process profits are produced and these, along with the inputs
of commodities and labour, determine the price of the output. Each and
every commodity is produced in this way so that the entire system of
production can be described by a matrix of n equations, each equation
corresponding to the inputs and output of one of the n commodities in

the system.
In order to write these equations, the following notation (which differs
slightly from that used by Sraffa) will be adopted:
Aij = the quantity of commodity i which enters into the production of
commodity j;
•
= A refers, therefore, to the total means of production
i~1 Aij
employed in the economy;
P,!, = the money price of a unit of commodity i;
•
.L P,!, = pm refers, therefore, to the sum of the price ofa unit of
1= I
each commodity;
ai = the quantity oflabour employed in the production of commodity i;
•
L a = a refers, therefore, to the total quantity of labour
i
i= I
employed in the production of all commodities;
m = the money wage rate (per unit of labour);
r = the rate of profit.
The equation for each commodity can now be written as follows:
P,!, = P,!, All + Pi A21 + Pj A31 + ... + P':: A.I + mal (1 + r)
Pi = Pi Al2 + 1"; A22 + Pj A32 + ... + P':: An2 + ma2(1 + r)
P':: = Pi A ln + pm A2n + Pj A 3n + ... + P':: Ann + man (1 + r)
This matrix can be written more summarily, as follows:
pm = pm A + ma (1 + r)
The content and implications of these equations will now be explored
by examining some of their basic features. The first feature is that these
equations describe a system of production where commodities are used
to produce other commodities. In other words, some commodities enter
as inputs into other commodities (possibly including themselves), which
in turn enter into the production of other commodities, and so on. Sraffa
refers to such commodities as 'basics' because they enter 'directly or
indirectly into the production of all commodities' (Sraffa, 1960, p. 8).
However, there are also some commodities which do not enter into the
production of other commodities and Sraffa refers to these as 'non-
basics' (ibid.). Two important implications follow from this distinction.

The first is that a basic commodity affects the price of every commodity
because it enters, either directly or indirectly, into the production of
every commodity. A non-basic, by contrast, has no effect upon the price
of any other commodity because it does not enter into the production of
any other commodity. In other words, the price of a basic commodity is
mutually dependent upon the price of every other commodity whereas
the price of a non-basic (or luxury) commodity is independent of the
price of any other commodity. The second implication is that only basic
commodities have an effect upon the rate of profit since they enter as
inputs into the production of every other commodity. A change in the
price of a non-basic will have no effect upon the general rate of profit and
will only affect the rate of profit in those sectors where it enters as an
input. Sraffa neatly summarises these two implications by pointing out
that 'the chief economic implication of the distinction' is that 'basics
have an essential part in the determination of prices and the rate of
profits, while non-basics have none' (Sraffa, 1960, p. 54).
The second feature of the above equations is that profit in a
competitive capitalist economy is distributed according to the total
investment of each capitalist in means of production (pm A) and labour
(ma). In other words there is an equal rate of return (r) on every unit of
invested capital (pmA + ma). However, the amount of capital invested
can only be established if the price of the means of production (Pj and
the price of labour (m) is known, since neither of the latter can be
aggregated without knowing their price. Thus the existence of a uniform
rate of profit implies the existence of a price for each commodity
(including labour). Conversely every price, as the above equations
reveal, implies the existence of a uniform rate of profit. It follows that, in
a capitalist economy, the rate of profit can only be known if prices are
known and prices can only be known if the rate of profit is known
(Sraffa, 1960, p. 6). In other words both are mutually determined and
determining so that any explanation of the rate of profit implies an
explanation of prices and vice versa. This feature of the Sraffa system
has two important implications: one for neoclassical theory and one for
Marx's labour theory of value. The implication for neoclassical theory is
that the rate of profit cannot be correctly explained by the marginal
productivity of capital since the latter cannot be identified independent-
ly of the rate of profit. Since this is an essential element of the
neoclassical theory of distribution it follows that this theory is inherently
circular and hence untenable (see Robinson, 1965; Bhaduri, 1969;
Garegnani, 1972; Nell, 1972; Nuti, 1972; Dobb, 1972; 1973, Ch. 9). The
implication for Marx's labour theory of value is that the explanation of

profit cannot be separated from the explanation of prices even though
some Marxists have attempted to treat them separately in order to
'prove' that Marx's theory of profit is still valid even ifhis theory of price
is faulty (see, for example, Medio, 1972, 1977; Armstrong, Glyn and
Harrison, 1978). However Sraffa's analysis clearly shows that these two
issues are inextricably bound together so that a faulty explanation of
prices must also be a faulty explanation of profit and vice versa. 8
The third feature of the Sraffian approach is that the proximate
determinants of prices, the rate of profit and the wage rate can be seen
from an inspection of the above equations. Thus, the price of a
commodity is determined by the physical quantity of inputs (A) and
labour (a) and by the wage rate (m) and the corresponding rate of profit
(r). In turn the rate of profit is determined by the physical quantity of
inputs (A and a) and by the wage rate (m) while the wage rate is
determined by the same physical inputs (A and a) and the rate of profit
(r). These, however, are only the proximate determinants, and it is
necessary also to consider their ultimate determinants. The ultimate
determinants of the physical quantity of inputs of commodities and
labour are the prevailing technical and social conditions of production
(that is, technology, natural resources, workers' skills, work practices,
intensity of the working day, and so on) while the ultimate determinants
of the rate of profit and the corresponding wage rate hinge upon the
respective power of workers and capitalists vis-a-vis the distribution of
the net output.
The fourth feature of Sraffa's system of equations is that the quantity
of inputs (A, a) can all be expressed in terms of labour (a) or indeed in
terms of any commodity (A). This implies that the Marxian value of any
commodity (that is, the quantity oflabour embodied in it) can be derived
and calculated from the data specifying the physical amount of
commodities (A) and labour (a) needed to produce that commodity by
simply expressing the latter in terms oflabour (a). In turn the amount of
surplus value can be calculated if, additionally, the wage rate (m) is
known. Thus, Marx's three crucial value quantities (constant capital,
variable capital and surplus value) can all be calculated, either for one
commodity or for the economy as a whole, if the physical quantity of
commodities and labour and the wage rate are known. The reverse
process (that is, converting value quantities into physical quantities) is
not, however, possible since the value of constant capital cannot be
reduced to the physical quantities from which it is derived.
This has two implications from the point of view of Marx's labour
theory of value. The first is that Sraffa's approach is more fundamental

than Marx's approach in the sense that the physical inputs of
commodities and labour and the wage rate determine not only the rate of
profit and prices; they also determine values and surplus value. In other
words, values are not, as Marx thought, the most basic elements which
determine the price of a commodity; the more basic elements are the
physical quantities of commodities and labour which go to produce it.
The second implication is that, if one is interested in explaining the rate
of profit and prices, as Marx was, then the correct approach is not to
start with values and then transform them into prices, as Marx did, but
to start with the physical quantity of commodities and labour and the
wage rate, and then show how the rate of profit and prices can be derived
from this datum (Samuelson, 1971, pp.419-21; Hodgson, 1977b,
pp. 101- 5; see Steedman, 1977, chapters 3 and 4). These considerations
can be summarised, following Steedman (1975a, p.78; 1977, p.48) in
Figure 1.1, which shows the relationship between physical quantities,
value quantities and price quantities.
Value quantities of:

Co nsta nt cap i ta I
Variable capital
Physical quantities of: Surplus value
Commodities
Labour
and
the wage rate Price quantities of:
Commodities
and
the rate of profit
Figure 1.1 The relationships between physical, value and price quantities
Arrows (a) and (b) indicate that both value quantities and price
quantities can only be derived from physical quantities. Arrow (c) shows
that price quantities can be derived from value quantities if Bortkie-
wicz's transformation procedure is used (see section 1.5 above) even
though, as has been seen (ibid.), there is no systematic relationship
between values and prices, or between surplus value and profit. Arrow
(b) shows the most direct way of explaining prices and profit and, for the
reasons just cited, it is also the best way.
These are the main features of the Sraffian approach. As has been
seen, this approach is concerned with the sames issues as Marx. The
important difference between them is that the Sraffian approach is free

from the difficulties associated with the labour theory of value. Thus the
weight of arguments 9 is strongly in fabour of the Sraffian approach and
the rejection of the labour theory of value.
1.7 CONCLUSION
Marx's labour theory of value has been examined in detail in this

chapter. It was seen, in the course of examining the transformation
problem and its solution, that this theory did not in fact show, as Marx
had intended, that value and surplus value were related to price and
profit in a constant and systematic way. The work of Sraffa was also
examined in some detail and it was seen that his work deals with the same
issues that are central to Marx, albeit in a more rigorous and less
problematic way. There is therefore, as a consequence of this analysis, a
compelling case for the rejection of the labour theory of value and the
adoption of the Sraffian approach. Consequently, the Sraffian approach
will be adopted, where relevant, throughout the remainder of this work.
2 Marxian Crisis Theory
2.0 INTRODUCTION
The concept of crisis is central to the entire Marxian framework even

though Marx never completed and unified a systematic theory of crisis
(see Sweezy, 1942, pp. 133-4; Howard and King, 1975, pp.210-211;
Alcaly, 1978, p. 17). In this chapter three types of crisis are identified and
discussed. These are: the law of the falling rate of profit (section 2.1),
realisation crisis (section 2.2) and rising real wages (section 2.3).
A number offeatures should be noted about these three types of crisis.
First of all Marx only identified two possible sources of crisis; the law of
the falling rate of profit and realisation crisis. There is no evidence from
his writings that he ever realistically considered that rising real wages
would give rise to a crisis within capitalism. However, recent evidence
(Glyn and Sutcliffe, 1972;lBoddy and Crotty, 1975;2 Weisskopf, 1979;3
Heap, 19804) suggests that rising real wages may be an important factor
in explaining the profitability crisis within contemporary capitalism.
For this reason there is a growing tendency for Marxists to distinguish
three variants of Marxian crisis theory (see, for example Alcaly, 1975,
1978; Shaikh, 1978, see p. 376; Weisskopf, 1978, 1979;) in contrast to the
more traditional two variants (see, for example, Sweezy, 1942; Howard
and King, 1975, 1976; Sowell, 1976; Elster, 1985, pp. 154-65).
The second notable feature about these three types of crisis is that
each poses a threat to the stability and growth of the capitalist system
essentially because they reduce the rate of profit. However, each affects
the rate of profit in different ways and this fact raises complex
econometric problems. The reason for this is that it may be extremely
difficult to disentangle the different chains of causation which are
operative in a situation where there is a crisis of profitability, although
Weisskopf(l979) appears to have successfully overcome this problem in
his analysis of the US economy.
A third feature is that some of these types of crisis are secular (that is,
long-run) while others are cyclical (that is, short-run).5 In the case of the
law of the falling rate of profit, it is clear that this is a secular crisis since
the underlying factors which determine it are, in Marx's view, structural-
ly irreversible. In other words, the law of the falling rate of profit is seen
as endemic in the capitalist system because of the type of technical
24
Marxian Crisis Theory 25
innovations which capitalists adopt. In the case of realisation crises,

these are cyclical in nature since their proximate causes are inherently
temporary, albeit recurring. It was in the context of a discussion of
realisation crises that Marx affirmed that 'permanent crises do not exist'
(Marx, 1969, p. 497). The case of rising real wages is perhaps less clear
cut, although it would seem wiser to regard it as a cyclical rather than a
secular crisis. This is because the growth of trade union power which is
associated with rising real wages cannot strictly speaking, be regarded as
a structurally inevitable and irreversible feature of the capitalist system.
The significance of Marxian crisis theory, in the present context,
derives from the fact that both Manuel Castells and Jean Lojkine, whose
work is examined in detail in Chapters 6 and 7 respectively, resort to
different versions of Marxian crisis theory in order to explain, inter alia,
the causes and consequences of state intervention in the area of collective
consumption. The present analysis has the advantage of providing a
precise clarification of Marxian crisis theory as a prelude to assessing the
validity and reliability of these explanations.
The analysis of Marxian crisis theory begins with the law of the falling
rate of profit. This is treated in section 2.1. The basic analytical
procedure is to present Marx's own explanation of the 'law' (subsection
2.1.2.) and then to assess it in terms of its logical coherence and empirical
accuracy (subsections 2.1.3 to 2.1.5). It will be seen, to anticipate, that
Marx's 'law' does not have the law-like quality which he attributed to it.
Moreover, while Marx's law can be shown to have an internal coherence
under certain specified (though unrealistic) conditions, it is not realis-
tically conceivable (as long as capitalists continued to be rational
profit-maximisers) that the rate of profit could fall as a result of the
mechanisms postulated by this variant of crisis theory. Thus the
explanatory role of this variant of Marxian crisis theory is negligible.
The analysis of realisation crisis is undertaken in section 2.2. Marx's
disparate references to realisation crisis are brought together and
moulded to form a coherent theory. In the process, some references
(notably those which could be adduced to show that Marx was a crude
underconsumptionist) have had to be excised from the analysis because
they are untenable per se, and, in addition, are inconsistent with the main
thrust of Marx's theory. The discussion is divided, following Marx, into
two parts; the first and shorter part is concerned with the ultimate causes
which make a realisation crisis possible (subsection 2.2.2), while the
second part is concerned with the proximate causes which make this
possibility an actuality (subsections 2.2.3 to 2.2.7). Four proximate
causes are identified and discussed; the absence of macroeconomic
planning (subsection 2.2.3), technical innovation (subsection 2.2.4),

variable weather conditions (subsection 2.2.5), and scarcity of labour
(subsection 2.2.6). It will be observed that the first two of these
proximate causes are endogenous while the final two are exogenous.
This variant of Marx's crisis theory represents a possible (indeed
probable) explanation of some of the crises of capitalism since it is based
upon an accurate characterisation of some features of the capitalist
economy. The analysis of rising real wages in section 2.3 represents a
departure from Marx in that he never considered it possible for real
wages to rise to a level where they would, ceteris paribus, threaten
profits. Indeed the analysis in this section shows that Marx held a
subsistence theory of wages (subsection 2.3.2 to 2.3.4) which, if this
theory is to have any semblance of meaning, must imply that the share of
wages in net output is likely to fall (given Marx's view of technical
change) and that the share of profits (and, by corollary, the rate of
profit) must rise. The crucial factor in this context is trade unions
(subsection 2.3.5) whose powers have increased considerably since the
time Marx was writing. This analysis, in conjunction with some recent
evidence on the profitability crisis in some capitalist countries, suggests
that this form of crisis may be of crucial importance.
It should be noted, finally, that the analysis below is conducted in
terms of values. This is because most of the issues being considered are
independent of whether one's economic categories are expressed in
terms of values or prices. The retention of values has, moreover, the
advantage of addressing and assessing the issues in precisely the way that
Marx formulated them. However, these considerations in no way
diminish from the seriousness of the critique of the labour theory of
value presented in the preceding chapter.
2.1 THE LAW OF THE FALLING RATE OF PROFIT
2.1.1 Introduction
Marx regarded his 'law of the tendency of the rate of profit to fall'
(Marx, 1974c, Chapter 13) as the most importance source of crisis in
capitalism. This is evidenced by his assertion, in the Grundrisse, that 'this
is in every respect the most important law of modem political economy,
and the most essential for understanding the most difficult relations. It is
the most important law from the historical standpoint' (Marx, 1973,
p.748).
It is not surprising that Marx should take this view given that the
raison d'etre of capitalist production is profit. By corollary, a persistent

decline in the rate of profit will lead, ceteris paribus, to an underutilisa-
tion of existing capital equipment and to a decline in new investment
which, in turn, will cause a decline in output and employment. This
effectively creates a situation where there is unused (and hence excess)
productive capacity in the economy or, as Marx variously put it, an
'over-accumulation of capital' (Marx, 1974c, p.25l) and an 'over-
production of capital' (ibid). (Some contemporary Marxists, notably
those who adhere to the state monopoly capitalism thesis, have slightly
modified Marx's terminology by referring to such situations as a crisis of
'over-accumulation-devalorization' (see section 5.2).
In order to evaluate the validity of Marx's law of the falling rate of
profit, it is first necessary to summarise the law as Marx formulated it in
Capital, Vol. III, Part III (Marx 1974c, pp.211-66). This is done in
subsection 2.1.2. Two issues then arise in respect of the validity of this
law. The first issue is whether there are any circumstances where it would
be possible for the rate of profit to fall in the manner predicted by Marx's
law. This issue is considered in subsection 2.1.3 where it will be seen that,
ifthe actual rate of profit is equal to the maximum rate of profit, then a
rise in the organic composition of capital will cause the rate of profit to
fall. The second issue is whether a fall in the actual rate of profit is
inevitable in the course of capitalist development. In subsection 2.1.4, it
will be shown that it is not inevitable. Finally, a definitive refutation of
Marx's law is presented in 2.1.5.
2.1.2 Marx's Formulation of the Law
Marx's law of the falling rate of profit begins from the assumption that
in the capitalist system of production there will be a predominant ten-
dency for capitalists to continually introduce labour-savingjcapital-
using technical innovations in order to increase productivity, profit,
and, ultimately, capital accumulation. Marx based this assumption on
two related claims. First, he claimed that technical change would result
in an increase in the mass of the means of production (that is, machines,
buildings and so on) relative to the mass (or size) of the labour force.
Marx expressed this by saying that the technical composition of capital
would rise. Second, he claimed that this would, in turn, result in a more
or less corresponding increase in the value of the means of production
(that is, constant capital) relative to the aggregate value oflabour power
(that is, variable capital). Marx expressed this by saying that the organic
(or value) composition of capital would rise.
Marx then argued that as the value composition of capital tended to
rise the rate of profit would tend to fall, irrespective of whether the rate
of exploitation remained 'the same' (Marx 1974c, pp. 212,213) or was
'rising' (ibid, p. 213). The logic of Marx's argument was that, since all
surplus value (and hence profit) could only be derived from living labour
(that is, variable capital), the latter's relative decline vis-a-vis dead
labour (that is, constant capital) implied that surplus value (and hence
profit) must also ultimately decline.
Marx realised, indeed insisted, that although the predominant
tendency would be for the rate of profit to fall as capitalism developed,
there were a number of 'counteracting influences' (Marx, I 974c,
Chapter 14) which could diminish or even halt (albeit temporarily) the
decline in the rate of profit. These tendencies and counter-tendencies can
be seen more clearly in the light of an examination of Marx's formula for
the rate of profit. Marx defined the rate of profit (r) as:
S
r =---
c+ V
where S = total surplus value; C = total constant capital and V = total
variable capital. Marx's formula can be rewritten as:
s
V
r =---
1+ C
V
From this expression, it is apparent that the rate of profit (r) will only fall
if the organic composition of capital (C / V) increases at a faster rate than
the rate of surplus value (S/ V). In other words Marx's law of the falling
rate of profit can be formally stated in the following way: there will be a
tendency within capitalism for the organic composition of capital to rise
at a faster rate than the rate of exploitation such that the net effect, in the
long run, is to reduce the rate of profit (Howard and King, 1976, p. 37).6
This net effect is assumed to be both necessary and inevitable even
though it may encounter certain 'counteracting influences' which may
cause the organic composition of capital to fall (for example, produc-
tivity increases, foreign trade) or cause the rate of exploitation to rise (for
example, a reduction in real wages, an increase in the length and
intensity of the working day, foreign trade, relative overpopulation) or
both.
In assessing the validity of this law, it is necessary to consider two
issues. The first issue is whether there are any circumstances in which a
rising organic composition of capital could lead to a falling rate of profit

despite the counteracting influence of a rising rate of exploitation. The
second issue is whether the rate of profit will inevitably fall in the manner
and for the reasons described by Marx. It will be apparent that the first
issue focuses on the question of whether the outcome predicted by
Marx's law is possible while the second focuses on whether it is
inevitable. These issues will now be considered in tum.
2.1.3 The Possibility of a Fall in the Maximum Rate of Profit
Marx argued that technical change was an integral part of capitalist

development. Technical change, he argued, would take the form of
adopting capital-usingjlabour-saving innovations, the effect of which
would be to increase the organic composition of capital. The effect of
this, in tum, would be that the relative quantity of labour employed
would be reduced leading, ceteris paribus, to a decline in the relative
quantity of surplus value produced and ultimately to a decline in the rate
of profit. Marx recognised that capitalists would attempt to compensate
for this effect by increasing the rate of exploitation (through increases in
productivity and in the length and intensity of the work day), thereby
offsetting the decline in the rate of profit. However, Marx also pointed
out that the possibility of increasing the rate of exploitation in order to
compensate for the effects of the rise in the organic composition of
capital had certain 'insurmountable limits'. As a consequence, he
argued, the rate of profit must ultimately fall if the organic composition
continues to rise.
Marx's analysis up to this point is internally consistent although it is
incomplete in the sense that it fails to specify precisely the limits beyond
which it becomes impossible to increase the rate of exploitation and
hence to specify the point beyond which each increase in the organic
composition of capital will result in a fall in the rate of profit.
The limit beyond which it is not possible to increase the organic
composition of capital without causing a fall in the rate of profit arises
when the rate of profit has reached a maximum. In capitalist society, the
(theoretically) maximum rate of profit is that corresponding to zero
wages. As Sraffa has pointed out 'the notion of a maximum rate of profit
corresponding to a zero wage' (Sraffa, 1960, p.94), was suggested by
Marx in his discussion of the falling rate of profit when he referred to the
fact that a rise in the organic composition of capital would lead to a fall
in the rate of profit even if workers 'could live on air' (Marx, 1974c,
p.247).
This fact can be illustrated by inspecting the formula for the maximum
rate of profit (in value terms). If the rate of profit is at a maximum, this
means that wages must be zero and the entire output of direct labour
must count as surplus. Thus the maximum rate of profit may be defined
as:
r =--
s+ V
max C+V
Any inspection of this formula reveals that if variable capital is zero

(V = 0) and if it is not possible to increase the rate of surplus value by
increasing absolute surplus value (since it is already impossible to
increase relative surplus value because variable capital is zero) then
every increase in constant capital (C) will lead, ceteris paribus, to an
increase in the organic composition of capital (Cj V) and this will lead, in
turn, to a fall in the maximum rate of profit (see Dickinson, 1957; Meek,
1976a, pp. 217-18; Steedman, 1977, Chapter 9).
This result is highly formal in the sense that the possibility of a fall in
the maximum rate of profit does not necessarily imply a fall in the actual
rate of profit. This analysis simply proves that there are certain
circumstances where a rising organic composition of capital would
inevitably lead to a falling rate of profit. However, the fact that this is
theoretically possible does not imply that it is actually possible or likely.
It is therefore necessary to examine if a rising organic composition of
capital and a falling rate of profit are actually possible or likely.
2.1.4 The Possibility of a Fall in the Actual Rate of Profit
It will be apparent from an inspection of Marx's formula for the rate of

profit, that the actual movement of the rate of profit in the course of
capitalist development will be crucially dependent upon the direction of
change in the organic composition of capital and the rate of exploitation;
only if the former increases more rapidly than the latter will there be a
tendency for the rate of profit to fall. Thus, consideration will now be
given to Marx's analysis of the long-term direction of change in the
organic composition of capital and the rate of exploitation, respectively.
Marx asserted that the organic composition of capital would tend to
rise in the course of capitalist development. This assertion involves two
related assumptions. The first is that technical change will tend to be
labour-saving and capital-using, and the second is that this type of
technical change will result in a rising organic composition of capital.
Both of these assumptions will now be critically examined.

Marx's assumption that capitalists, in the pursuit of increased
productivity and profit, will tend to adopt labour-saving/capital using
technical innovations is an assumption about the actual behaviour of
capitalists. When Marx wrote Capital this was probably an accurate
empirical generalisation of capitalist behaviour; and this may explain
Marx's reason for making this assumption. However, this assumption is
very difficult to justify since there would seem to be no good reason for
assuming that capitalists will necessarily and inevitably adopt either
labour-saving or capital-using technical innovations. It simply depends
upon which is the most profitable at the time. Since capitalists are
concerned, in general, with minimising costs and maximising profits, it
follows that the nature of technical innovation will be highly contingent
on such factors as the availability of a cheap and docile labour force,
government grants, new technology, the scale of production, the type of
product being produced, etc. The more general point is that Marx's
assumption about the nature of technical change, whatever its historical
justification, is unwarranted (see Sweezy, 1973, p. 28; Howard and King,
1975, p. 203; Wright, 1978; p. 136).
Marx's second assumption was that capital-using technical change
would lead to a rising organic (or value) composition of capital. His
argument was that technical change would give rise to an increase in the
total quantity of constant capital (thereby causing an increase in what he
termed the technical composition of capital) and to an increase in the
total value of constant capital (thus causing, ceteris paribus, a rise in the
organic composition of capital).
There are basically two difficulties with this argument. The first is that
Marx's concept of the technical composition of capital is extremely
vague, essentially because heterogeneous physical objects are incom-
mensurable, and hence their physical quantity cannot be aggregated (at
least not meaningfully and sensibly). There is, in other words, no way of
establishing whether the 'mass' or 'quantity' of constant capital has
increased or decreased between two points in time except by aggregation
in terms of values or prices. It is thus conceptually unclear what Marx
meant when he claimed that the technical composition of capital would
tend to rise in the course of capitalist development (see Howard and
King, 1976, p. 38, fn 27; Steedman, 1977, pp. 132-6).
The second difficulty is Marx's claim that the value/organic composi-
tion of capital will tend to rise as a result of capital-using technical
change. The difficulty with this claim is that it is (equally) possible for the
organic composition of capital to fall as a result of such technical

change. This possibility follows from the fact that increases in produc-
tivity in those industries which actually produce the means of produc-
tion can result in a given type of machine being produced with less and
less embodied labour time (that is, less and less value). In this case
technical change would lead to a fall rather than a rise in the organic
composition of capital. Although Marx referred to this possibility, he
assumed, without any rigorous justification, that this occurred only in
'isolated cases' (Marx, 1974c, p.236; see also p.212). It is thus not
possible to affirm, a priori, what precise effect technical change will have
on the organic composition of capital (see Purdy, 1973).
The more general point to emerge from this analysis is that Marx's
assertion about the tendency for the organic composition of capital to
rise cannot be shown to be either necessary or inevitable. This, of course,
does not exclude the possibility that it may rise; it simply demonstrates
that there is no inherent or inevitable tendency for it to change (that is,
increase or decrease) in any given direction within capitalism.
The same general result also applies with respect to the rate of
exploitation. Marx argued that a rising organic composition of capital
would lead to a falling rate of profit, irrespective of whether the rate of
exploitation remained 'the same' (Marx, 1974c, pp.212, 213) or was
'rising' (ibid., p. 213). His own assumptions about the future develop-
ment of capitalism (notably the prospect of technical change coupled
with subsistence wages) must have obliged Marx to expect that the rate
of exploitation would continue to rise (see subsection 2.3.4). Given this
expectation therefore, there does not seem to be any realistic reason for
expecting this rise to be constrained by some upper limit (as Marx's law
requires) unless the limiting case of zero wages (see subsection 2.1.3) is
regarded as realistic. Thus, even on the basis of Marx's own expecta-
tions, his assumption that the rise in the rate of exploitation will not
'compensate' for the rise in the organic composition of capital does not
seem to be firmly and realistically grounded. The more general point,
therefore, is that there is no reason to expect the rate of exploitation to
move in the precise direction required by Marx's law.
It has now been established that Marx's expectations about the way in
which the organic composition of capital and the rate of exploitation
must inevitably change in the course of capitalist development, do not
have any solid or realistic basis. The logical implication of this is that his
expectation that the rate of profit would tend to fall in the course of
capitalist development does not have any solid or realistic basis either. 7
2.1.5 Technical Change and the Rate of Profit
The analysis so far has discussed the law of the falling rate of profit
within the parameters set down by Marx. In other words, it has focused
on whether the decisions and actions of capitalists in relation to
technical innovation would be such as to raise the organic composition
of capital and the rate of exploitation in such a way as to lead ultimately
to a fall in the rate of profit. On the basis of this analysis it has been
shown that technical change affects the organic composition of capital
and the rate of exploitation simultaneously, and that the net effect of this
technical change cannot be ascertained a priori (see Bortkiewicz, 1952,
p. 47.) There is, however, a further piece of analysis which also shows
Marx's law to be invalid. This analysis is usually referred to as 'Okishio's
theorem' (Okishio, 1961, though the argument has also been attributed
to Samuelson; see Howard and King 1975, p. 232, fn 31).
The basic argument is that no rational calculating capitalist would
ever knowingly introduce a technical innovation which would cause a
fall in the rate of profit; if, due to unforeseen and unintended
consequences, a new technique reduced the rate of profit, capitalists
would, ceteris paribus, revert to the previous technique. The reasons for
this can be illustrated as follows.
For any given technique of production, there is an inverse relation
between the real wage rate and the rate of profit. This is illustrated in
Figure 2.1 (adapted from Steedman, 1977, pp. 127-8; see also Howard
and King, 1975, pp. 208-9).
The line AB shows the different possible combinations of the rate of
profit and the real wage rate for that technique (Technique I). If an
additional technique (Technique II) is introduced, it too can be
described by its own particular combination of wage and profit rates
represented by the line CD.
The capitalist confronted with any two techniques will have to decide
which is the most profitable at any given wage rate. In this example, it
can be seen that ifthe real wage rate is below w* then Technique I will be
chosen, while if it is above w*; Technique II will be chosen. If the real
wage is equal to w* then the capitalist will be indifferent as to which
technique he adopts since both are equiprofitable at that point.
If technical innovation occurs again in this economy, then at least
some part of the line describing this new technique (Technique III) must
lie to the right of the wage-profit frontier (AED). This is because, for it to
be an innovation, the attainable rate of profit must be higher, for at least
Technique I
c
Technique II
'---___- - - Technique III
w' w" B H 0 Wages

Figure 2.1 The relationship between wages and profit for different techniques
of production
certain real wage rates, than the rate of profit attainable by the preceding
techniques (Techniques I and II). If the line describing this new
technique is GKH then it can be seen that, at any wage below w**, the
new technique (Technique III) will be more profitable than any of the
previous ones, and will therefore, ceteris paribus, be adopted. In other
words, the new Technique III will be more profitable, at any point on the
line GK, for any wage level below w**. If, however, the real wage should
rise above w** then the capitalist will find that this new technique
becomes unprofitable (by comparison with the rates of profit attainable
using Technique II) and he/she will revert to the older Technique II (on
the line CP). The more general point is that no technique will ever be
adopted, if and once it is known that its effect is to reduce the rate of
profit. It is perhaps ironic that Marx should have made precisely this
argument in Capital, Vol. III, Chapter 15 where he discusses the falling
rate of profit. He writes: 'No capitalist ever voluntarily introduces a new
method of production, no matter how much more productive it may be,
and how much it may increase the rate of surplus value, so long as it
reduces the rate of profit' (Marx, 1974c, p.264). However, he sub-
sequently rejected this argument - wrongly - on the ground that the
aggregate effect of innovations was to increase the organic composition
of capital, thereby causing, ceteris paribus, the rate of profit to fall. For
all of the preceding reasons, there is no justification for assuming that
technical innovation will lead to a fall in the rate of profit.
2.1.6 Conclusion
The analysis in this section has shown that there is no 'law' by which it is
either necessary or inevitable for the rate of profit to fall in the course of
capitalist development, as Marx claimed. This is not to deny that such a
fall is a possibility; it simply denies that it is inevitable.
This result is not particularly novel, having already been established
and repeated by numerous writers (see notably, Bortkiewicz, 1952,
pp. 37-50; Steedman, 1972, 1975a, pp. 79-80; 1977, Chapter 9; Hodg-
son, 1974a, 1975; Howard and King, 1975, pp. 203-9; Roemer, 1977,
1978b, 1979, 1981, Chapter 5; Weisskopf, 1979; Van Parijs, 1980; Harris,
1983). However, it is a result which has not yet made an impact on the
recent French Marxist urban sociology of Manuel Castells (see Chapter
6) or Jean Lojkine (see Chapter 7) for whom Marx's law of the falling
rate of profit has considerable importance, particularly in the explana-
tion of collective consumption (see sections 6.2, 7.2 and subsection
7.3.2).
2.2 REALI SAnON CRISES
2.2.1 Introduction
There are various references in Marx's work to realisation crises, though

nowhere does he give a systemic and comprehensive analysis of them.
The main sources are: Theories of Surplus Value, Part 2, Sections 6-14
(1969, pp. 492-535) and Capital, Vol. 2, Chapters 20 and 21 (1 974b,
pp. 396- 527). Although this type of crisis is often referred to as a crisis of
overproduction and underconsumption (see notably Sweezy, 1942,
Chapter 10; Howard and King, 1976, p. 34; Alcaly, 1978, p. 18; Shaikh,
1978, p. 226; Weisskopf, 1978, p. 246) it is more accurately described as a
realisation crisis for two reasons. The first is that it arises because of an
inability by the capitalist to use or sell his/her commodities and hence
realise the profit for which they were originally produced. The second
reason is that overproduction and underconsumption are the effects of
this realisation crisis (rather than the cause) in the same way that they are
the effects of the (alleged) tendency of the rate of profit to fall (see
subsection 2.1.1).
It should be emphasised that the term overproduction, as used by
Marx, refers to the production of a quantity of commodities in excess of
the quantity which can be sold at the going price of production, that is, in
excess of effective demand (see Marx, 1974b, pp. 414-15). 'It is not a
question' Marx emphasises, 'of absolute overproduction - overproduc-
tion as such in relation to absolute need or desire to possess com-
modities' (Marx, 1969, pp. 506,527; see also 1974c, pp. 255-7). Thus it
is perfectly possible (given this terminology) for a crisis of overproduc-
tion to coexist with a famine in a given capitalist country. A similar
reasoning also applies to Marx's use of the term 'overpopulation' (see
section 2.3.3).
Realisation crises are essentially cyclical (that is, short-run) in nature.
As Marx insisted in this context: 'Permanent crises do not exist' (Marx,
1969, p.497).
The major difficulty with any attempt to outline Marx's theory of
realisation crisis is, as numerous writers and critics have pointed out,
that Marx never systematically worked out his theory (Sweezy, 1942,
pp. 133-4; Howard and King, 1975, pp.21O-11, 1976, p.33; Alcaly,
1978, p. 17; Morishima and Catephores, 1978, p. 118,fn1O; Elster, 1985,
pp.154, 161; for a contrary view see Sowell, 1976,pp. 56-7). The present
section is, therefore, an attempt to present a systematic and coherent
account of Marx's theory of realisation crisis. The analysis begins,
following Marx, by indicating those factors which make a realisation
crisis possible (subsection 2.2.2). This is followed by an analysis of the
four proximate causes of realisation crises; the absence of macroecon-
omic planning (subsection 2.2.3), technical innovation (subsection
2.2.4), variable weather conditions (subsection 2.2.5), and scarcity of
labour (subsection 2.2.6).8 Each of these will now be discussed
respectively.
2.2.2 The Possibility of Realisation Crisis
Every realisation crisis expresses itself in the fact that the act of sale is
separated, in time and space, from the act of purchase. 'If purchase and
sale coincided', Marx affirms, 'the possibility of crisis would under the
assumptions made, disappear' (Marx, 1969, p. 508). This separation of
purchase from sale is possible in any system of commodity production,

simple or capitalist, and is the result of the fact that every commodity
must be transformed into money and vice versa, and it is this
'metamorphosis of commodities' (ibid., p. 500) which creates the
possibility of crisis.
Marx contrasts this situation with that of barter, where the possibility
of realisation crises does not exist. 'In Barter this contradiction does not
exist: no one can be a seller without being a buyer - or a buyer without
being a seller' (ibid., p. 509).
The possibility of a realisation crisis can be revealed more fully by
looking at the typical circuit of capital. This can be represented as
follows:
or
M -+ C + V -+ If] -+ C + V + S -+ Ml
where:
M = the amount of money which the capitalist owns at the
beginning of the production process;
CLP
MP = the commodities oflabour power and means of production
purchased by the capitalist at the beginning ofthe produc-
tion process;
C+ V = constant capital plus variable capital;
If] = the production process;
C1 = the new commodity;
C + V + S = constant capital plus variable capital plus surplus value;
Ml = the money which the capitalist receives from the sale ofthe
new commodity;
Ml = M + 6. M = S.
The possibility of a realisation crisis exists in this case if the capitalist is
unable to sell C1 (and hence realise profit) or if, having sold C 1 (for Ml)
(s)he cannot purchase the same quantity of commodities as formerly
(C}Xp) at the same price, and hence cannot reinvest his/her money at the
same rate of profit. In this case, sale and purchase 'become separated'
(ibid., p. 514), 'get bogged down' (ibid., p. 494) and are 'falling asunder'
(ibid., p. 510).
Having established that the possibility of a realisation crisis is
contained in the form of commodity exchange (thereby refuting Say's
law (ibid., p. 493)), Marx immediately emphasises that 'the factors
which turn this possibility of crisis into an actual crisis are not contained
in this form itself: it only implies that the framework for a crisis exists'
(ibid., p. 509). It is now necessary to look at the second part of Marx's
analysis which identifies some of the factors by which this 'possibility
can develop into actuality' (ibid., p. 512).
It is possible, following Marx, to identify at least four proximate
causes of a realisation crisis. 8 These four causes are the absence of
macroeconomic planning, technical innovation, variable weather condi-
tions and scarcity of labour.
Each of these situations can give rise to a crisis of overproduction and
underconsumption. The processes by which this occurs, in each case, will
now be examined respectively.
2.2.3 Realisation Crisis Caused by the Absence of Macroeconomic

Planning
One of the characteristic features of competitive capitalism is that

production is owned and controlled by individual capitalists. These
capitalists, in turn, are independent and autonomous of each other and,
as a result, there is no ex-ante co-ordination or regulation of their
separate investment decisions. In this sense, capitalism does not produce
'according to a plan' (Marx, 1969, p. 529). This fact, Marx pointed out,
can lead to a realisation crisis. To understand Marx's reasons for making
this claim, it is necessary to examine his two-sector model as outlined in
Capital, Vol. II, Chapters 20 and 21 (Marx, 1974b, pp. 396-527) (for a
useful discussion of this part of Marx's work, see Naqvi, 1960).
Marx's model of simple reproduction (that is, a stationary state)
distinguishes between Sector I which produces means of production and
Sector II which produces consumption goods. 9 Following the usual
notation this two-sector model can be written as follows:
Sector I : MI ~ C I + VI ~0~ C I + VI + SI ~ Y I ~ MI
Sector II: M2~C2 + V2~0~C2 + V2 + S2~ Y2~M2
where
MI> M2 = the total quantity of money at the beginning of production in
Sectors I and II;
CI> C2 the quantities of constant capital in Sectors I and II
respectively;
VI> V2 = the quantities of variable capital in Sectors I and II
respectively;
III = the production process;

S10 S2 = the quantity of surplus value produced in Sectors I and II
respectively;
Y10 Y 2 = the value of the output of Sectors I and II respectively;
M10 M1 = the quantity of money at the end of the production process in
Sectors I and II respectively.
A number of features should be noted about this model. First of all it
is, by assumption, a model of simple reproduction, that is, there is zero
economic growth. This means that there is no technological change and
the same quantity of inputs and outputs are produced and reproduced
each year. The second feature is that a number of inter- and intra-
sectoral exchanges have to take place to ensure that production actually
continues. In effect this means that the output of both sectors (YI and
Y2) will have to be distributed in such a way that each has the amount of
means of production and consumption goods which they require to
continue production on the same scale.
In the case of Sector I, its output (YI ) must be such as to ensure that the
input requirements of both sectors for means of production (CI and C2)
are satisfied. If this occurs then:
In the case of Sector II, its output (Y2) must be such as to ensure that the
input requirements of both sectors for consumption goods (VI ,SI, V2,S2)
are satisfied. If this occurs then:
Y 2 = VI + SI + V2 + S2·
These two conditions, taken together, imply that simple reproduction is
only possible if the quantity of consumption goods required by Sector I
(VI + SI) is equal to the quantity of means of production required by
Sector II (C2 ). In other words, the equilibrium condition for simple
reproduction is:
(see Marx, 1974b, Chapter 20)
It is now possible, using this model, to specify one reason why a
realisation crisis is possible within capitalism. This reason derives from
the fact that the output produced by each of the two sectors in this model
is the result of independent and uncoordinated decisions by a large
number of capitalists. The question therefore arises (as Marx rhetorical-
ly put it): 'How is it possible to achieve the necessary balance and
interdependence of the various spheres of production, their dimensions
and the proportions between them, except through the constant

neutralization of a constant disharmony?, (Marx, 1969, p. 529; see also
Marx, 1974b, p.499).
Marx's rhetorical question draws attention to the fact that the correct
proportions which must be maintained between the different depart-
ments of production, if production is to remain in equilibrium, are likely
to be upset simply because there is no ex-ante, calculated, co-ordination
between the various production units. In turn, the calculations which
each capitalist undertakes as to his/her anticipated level of input and
output are likely to be inaccurate, essentially because he/she does not
have access to the plans of other production units upon whom he/she
depends for the purchase of inputs or the sale of outputs. The capitalist is
thereby precluded, as a general rule, from any calculated co-ordination
with other production units. As a result 'disproportionate production'
(Marx, 1969, p.521) is 'not only possible but very probable' (ibid.,
p. 531). This is true, a fortiori, in the case of expanded reproduction, as
will now be seen.
Marx discusses expanded reproduction in Capital, Vol. II, Chapter 21
(Marx, 1974b, pp.493-527). Expanded reproduction refers to the
situation where there is positive economic growth. The difference
between simple and expanded reproduction therefore centres on the use
to which the surplus is put at the end of the production process: in the
case of simple reproduction the surplus is entirely consumed whereas in
the case of extended reproduction the surplus is only partly consumed,
the remainder being used for reinvestment. Following the notation
adopted above, the nature of expanded reproduction in a two-sector
model can be described as follows:
Sector I:
MI~CI + VI~0~CI + VI + (SOl + SCI + SV1)~ YI~M:
Sector II:
M2~C2 + V2~~C2 + V2 + (S02 + SC2 + Sd~ Y2~M~
where, in addition to the usual notation:
= the quantity of surplus value which is consumed by
capitalists in Sectors I and II respectively;
= the quantity of surplus value which is reinvested by
capitalists in constant capital in Sectors I and II
respectively;
= the quantity of surplus value reinvested by capital-
ists in variable capital in Sectors I and II respec-
tively;
(SOl = SCI + SVl) = the

total quantity of surplus value produced in
Sector I;
(S02 + SC2 + SV2) = the total quantity of surplus value produced in
Sector II;
In this model, the total output of Sector I is equal to the total means of
production required by the economy:
Y I = CI + C2 + SCI + SC2
Similarly, the total output of Sector II is equal to the total consumption
goods required by the economy:
Y2 = VI + (SOl + SVI) + V2 + (S02 + Sv2).
From these two statements, it is possible to infer that the equilibrium
condition for expanded reproduction is
C2 + SC2 = VI + SOl + SVi
It is further possible to infer that the over-all rate of growth in the
economy is equal to:
SCI + SC2 +SVI + SV2
C I + C2 + VI + V2
The significance of expanded reproduction in the present context is that

it increases the probability of a realisation crisis by comparison with
simple reproduction. The precise reasons for this will now be explored
by examining the consequences which follow when 'too much' of a
particular commodity is produced in relation to effective demand. (A
similar type of analysis could also be applied to the analagous case where
'too little' is produced in relation to effective demand.)
The existence of too much of a particular commodity (for example,
cotton, yarn) can arise when, in a period of expanded reproduction,
capitalists in one sector only (for example Sector I) invest in new and
additional fixed capital, the net effect of which is that the output of the
commodity in question exceeds that required by other departments as
inputs. The consequence of this is that all of the output of Sector I cannot
be sold, or can only be sold at a loss. In either case, each and every
capitalist is unable to simultaneously cover all his/her costs (wages,
depreciation, interest, rent, and so on) and make the going rate of profit
and, as a result (assuming all costs are paid at the beginning of the
production period) the rate of profit must fall.
This fall in the rate of profit can only be restored in the subsequent
period of production if, ceteris paribus (that is, assuming a closed
economy and no increase in domestic demand), the productive capacity

in this sector is reduced. This can only be done by reducing the labour
force and destroying the excess fixed caphal in the sector. In this way, the
overproduction of one commodity leads to the overproduction of yet
other commodities, that is, wage goods and fixed capital goods. This in
turn necessitates a reduction in capacity in the wage goods sector and in
the sector producing fixed capital goods, and a corresponding reduction
in those departments which supply these with inputs, and so on. In this
way, the partial realisation crisis in one sector of production spreads as a
result of the interdependence of production. Whether this partial crisis
becomes a general crisis depends, in the first place, on the amount by
which the original commodity is overproduced and, in the second place,
on the extent to which this sector is dependent upon other sectors of
production.
The implication of this analysis is that balanced economic growth
within capitalism can only be achieved through rigorous planning based
upon a systematic calculation of the projected level of inputs and
outputs in the various departments in each period of production. The
Russian economist, Michael Turgan-Baranowsky cogently demon-
strated this point in 1901 (see Sweezy, 1942, pp. 158ft). He wrote:
If social production were organized in accordance with a plan, if the
directors of production had complete knowledge of demand and the
power to direct labour and capital from one branch of production to
another, then, however low social production might be, the supply of
commodities could never outstrip the demand (ibid., p. 166).
The more general conclusion is that realisation crises are highly
probable given the unplanned nature of the capitalist economy,
irrespective of whether one assumes simple or expanded reproduction.
2.2.4 Realisation Crisis Caused by Technical Innovation
It is possible, from Marx's work, to identify two ways in which technical

innovation can lead to a realisation crisis. Both cases reveal how
technical innovation can result in the overproduction of a given
commodity so that it cannot be sold at its price of production. In other
words, the link between the commodity and money (C - M) becomes
broken. Both of these cases (analysed separately here but not necessarily
separate in reality) will now be dealt with.
The first case refers to a technical innovation which is adopted by all
capitalists in a particular sector of production. The effect of this
technical innovation is to double output, although there is no correspon-

ding increase in the demand for this output from other sectors. In this
situation there is a crisis of overproduction. Marx cites the following
example of this: 'When spinning machines were invented, there was an
overproduction of yarn in relation to weaving' (Marx, 1969, p.521),
adding that 'this disproportion disappeared when mechanical looms
were introduced into weaving' (ibid.).
By definition this overproduction of spun cotton affects all capitalists
involved in the spinning of cotton since it is the net, combined effect of
their separate actions. The consequence of their overproduction is that
all of the output of spun cotton cannot be sold or can only be sold at a
loss. In either case, each and every capitalist is unable to simultaneously
cover all his/her costs (wages, depreciation, interest, rent, and so on) and
make the going rate of profit. As a result (assuming all costs are paid at
the beginning of the production period) the rate of profit must fall. This
fall in the rate of profit can only be restored if, ceteris paribus (that is,
assuming a closed economy and. no increase in domestic demand), the
productive capacity in the cotton-spinning sector is reduced both by
laying off labour and by destroying the excess fixed capital in this sector.
The actions and reactions of capitalists in this situation will lead to a
reduction in the production capacity of those sectors producing spinning
machines and, in turn, to a reduction in the production capacity of those
sectors which provide the inputs for making spinning machines, and so
on. In this way, the partial realisation crisis in one sector of production
spreads, as a result of the interdependence of production, to other
sectors. Whether this partial crisis becomes a general crisis depends on
the size of the overproduction in the first place and, in the second place,
on the extent to which other lines of production are dependent upon this
sector (see Marx, 1969, pp.523, 529).
The second case, which would seem to be implicit in one of Marx's
statements (Marx, 1969, p.495) refers to a technical innovation in a
particular sector of production which is adopted only by some capitalists
in the current period of production. As a result two methods of
production now coexist: the old method and the new. This new
technology is twice as efficient as the old and hence produces the
commodity for half the price of production of the old technology.
If it is assumed that the total output in the current period of
production is twice that of the preceding period of production (half of it
being produced by the new technology and half by the old) and if it is
further assumed that there is no change in demand, then it will be
apparent that only the output produced by the new technology will be
sold and, by corollary, that produced by the old technology will not
(since the former is cheaper than the latter and is equal to the entire
demand). There is thus an overproduction of commodities and, by
implication, an overproduction of fixed capital (that is, the old
technology). The consequence of this is that the production capacity of
those producing with the old technology is useless.
There are two main consequences of this technical innovation: the
first is that it will result in workers being laid off which will lead to a
decline in the demand for wage goods, leading, in turn, to overcapacity
and overproduction in this sector; second, it will result in a fall in
demand for the output of those sectors which produced the old
technology and in those sectors which produced inputs for this sector,
and so on. These departments will have a similar effect on yet other
sectors whose output they consume. The extent to which a partial crisis
in one sector can develop to a general crisis throughout the economy
depends, as has already been seen, upon the size of the overproduction in
the first place and on the extent to which other sectors are highly
interdependent with it.
2.2.5 Realisation Crisis Caused by Variable Weather Conditions
Marx explicitly acknowledged that 'weather conditions - play such an

important part in modern industry' (Marx, 1969, p. 533). This is because
weather can directly affect the quantity and quality of those com-
modities which are grown and produced on the land. This, in turn, can
give rise to a realisation crisis if too little or too much is produced in
relation to demand. It will now be shown how both these possibilities
can arise. The first case refers to a commodity where, 'as a result of bad
harvests' (ibid, p. 515) too little is produced in relation to demand. This
results, ceteris paribus, is an increase in the price of production of the
commodity since, with the same amount of labour and capital, the
weather conditions adversely affect the productivity of labour. The
effect of this, in turn, can be seen when the commodity enters as an input
into the production of another commodity.JO
If the commodity enters as an input into the production of another
commodity then the capitalist purchasing this input will be unable,
ceteris paribus, to obtain the same quantity of the input as in the
preceding period since its price of production has increased. The
consequence of this (assuming simple reproduction) is that the capitalist
will have to reduce the quantity of some of his inputs by comparison with
the quantity of inputs employed in the preceding production period. If it
is assumed that the price of the commodity in question doubles, and if

the capitalist spends the same share of money (M) on the various inputs
as in the preceding production period, then he will only be able to
purchase and process half the quantity of the raw material by
comparison with that in the preceding period.
The consequence of this is that the capitalist has excess capacity in the
form of fixed capital equipment and labour, since the quantity of raw
material to be processed has been halved. As a result 'a part of fixed
capital stands idle and a part of the workers is thrown out on the streets'
(ibid., p. 516). This in turn causes the rate of profit to fall, assuming that
the capitalist has certain fixed costs and wage costs which must be paid at
the beginning of the production period. By corollary, the price of the
finished product will increase.
The implications of this rise in price will be felt subsequently in those
sectors where this commodity enters as an input. As Marx put it 'the rise
in its price will result in the same disturbances in reproduction in these
spheres' (ibid., p. 516). In short then, a rise in the price of a commodity,
as a result of bad weather, can give rise to a realisation crisis since it
creates a situation where there is relative overproduction and undercon-
sumption of yet other commodities.
The second case refers to the situation where too much of a given
commodity is produced in relation to demand as a result of favourable
weather conditions. The consequences of this are crucially dependent
upon whether one assumes simple reproduction or extended reproduc-
tion.
If one assumes simple reproduction, and if one further assumes that
the favourable weather has the effect of doubling output (and hence
halving the price of production), then the effect of the bumper harvest
will be that only half the output of this commodity will be sold. This, in
turn, means that the revenue of the capitalist will be halved, which has
the effect (assuming that all fixed and wage costs are paid at the
beginning of the production period) of reducing the rate of profit. The
consequence of this is that in the next period of production, the capitalist
will have to reduce the quantity oflabour and capital employed. This, in
turn, will lead to a reduction in the capacity in those sectors producing
wage and capital goods respectively and to a corresponding reduction in
those sectors which supply them with inputs, and so on. In short, the
effect of the favourable weather conditions, under conditions of simple
reproduction, is that it leads to an overproduction of commodities
which, in turn, results in an overproduction of yet other commodities
(that is, fixed capital and wage goods). The full extent of this realisation
crisis will depend on the size of the overproduction of the original

commodity and on the extent to which this commodity is interdependent
with other sectors (as well as on the weather conditions in subsequent
production periods).
If one assumes expanded reproduction, and if this expansion is able to
absorb the increase in output as a result of the bumper harvest, then no
realisation crisis will arise. Indeed, if the price of production is halved (as
assumed above), this will have the effect, ceteris paribus, of raising the
rate of profit in those sectors which use the commodity as an input which
will further increase the expansion of reproduction.
2.2.6 Realisation Crisis Caused by Scarcity of Labour
There is evidence that Marx regarded relatively low population growth

as a factor which could trigger off a realisation crisis (Marx, 1974c,
pp. 250ff; see also Morishima and Catephores, 1978, pp. 117 -18).
Marx's argument was that if economic growth continued to grow at a
faster rate than population growth then, assuming no surplus labour
and a maximum rate of exploitation, a point would be reached where
'There would be ... overproduction of capital ... as soon as capital
would ... have grown in such a ratio to the labouring population that
neither the absolute working-time supplied by this population, nor the
relative surplus working-time, could be expanded any further' (Marx,
1974c, p.251).
There would seem to be two different mechanisms, suggested by
Marx, by which this disparity between population growth and economic
growth could trigger off a realisation crisis with its consequent
overproduction. The first mechanism is that the shortage of labour
renders it impossible for a certain amount of capital equipment to be
productively utilised, giving rise in effect to an excess of capital or, as
Marx terms it, an 'overproduction of capital' (Marx, 1974c, p.252).
This will lead to a fall in the rate of profit in those departments where this
excess occurs (since the capitalists in these departments will not be able
to cover all their fixed costs and make the going rate of profit) and can
only be reversed by cutting back on the excess capital. This, in tum, will
have repercussions on those departments whose output enters as an
input into this department, thereby leading to further cut-backs. Similar
repercussions will occur in other departments which are part of this
input-output chain.
The second mechanism is that the shortage oflabour leads to a rise in
wages. This rise in wages leads, ceteris paribus, to a reduction in the rate
of profit in those sectors employing this labour, leading, in tum, to a cut-

back in the labour and capital employed in these sectors, with
consequent repercussions in those sectors which are connected, by the
input-output chain, to this sector.
The net effect of the realis~tion crisis, in both of these cases, is to
reduce the amount of capital employed to the point where there is again
a relative surplus population and a return to the previous level of
profitability.
2.2.7 Conclusion
The analysis in this section has attempted to organise most of Marx's

references to realisation crises into a coherent, consistent and cogent
framework. However, in doing this a number of references which
indicate that Marx also had a crude underconsumptionist theory were
edited out (these references are Marx, 1974b, Chapter 16, Section 3,
p. 320, fn 32 and Marx, 1974c, Chapter 15, Section i, pp. 244-5). These
references were edited out because they are based upon the fallacious
Malthusian idea that realisation crises can arise from the poverty of the
masses. (The fallacy in this idea is clearly demonstrated in Blaug, 1962,
pp.165-7, 283-4; Howard and King, 1975, p.216; Shaikh, 1978;
Walker, 1978, pp.50, 51, 181). The advantage of separating Marx's
various references to realisation crises in this way is that it clarifies the
acknowledged confusion in Marx's work on this point and highlights
those aspects which contain the core of a defensible theory of realisation
crisis. In addition this section provides a basis for evaluating Castells's
claims that collective consumption serves the function of solving the
realisation crises of capitalism (see subsection 6.2.3).
2.3 CRISES CAUSED BY RISING REAL WAGES
2.3.1 Introduction
The possibility of crises being caused by rising real wages, as a result of

their effect on profitability, has received some attention from Marxists
in recent years (see notably, Glyn and Sutcliffe, 1972; Boddy and Crotty,
1975; Roemer, 1977, 1978a, 1978b, 1979; Weisskopf, 1978, 1979) though
there is little evidence that Marx ever realistically considered this as a
possible long-term source of crisis. Nevertheless, it does seem necessary
to consider this possible source of crisis in the present context for two
reasons. The first is that there is some empirical evidence which suggests
that this form of crisis is particularly operative in some contemporary
capitalist countries (ibid.). The second reason is to consider whether this
fact can be incorporated into Marx's framework of analysis without
altering some of the basic elements of that framework.
The present section critically examines Marx's theory of wages as
outlined in Capital, Vol. I, Chapters 6, 19 and 25 (Marx, 1974a) and in
Value, Price and Profit (Marx, 1975). The analysis begins by pointing
out some of the important differences between a commodity and labour
power (subsection 2.3.2). The most important difference is that the
mechanism which regulates the production, and hence the value, of
labour power is the industrial reserve army. The nature of this
mechanism is discussed in subsection 2.3.3, followed by a discussion of
its effect upon the long-term trend in real wages in subsection 2.3.4. The
role of the trade unions within this framework will then be considered in
subsection 2.3.5. The over-all conclusion of this section is that Marx's
theory of wages cannot adequately deal with the possibility that real
wages may rise as a result of trade union power and threaten
profitability. It is suggested that a bargaining power theory of wages
might offer a more realistic (if theoretically less determinate) explana-
tion of wages.
2.3.2 The Difference Between Labour Power and a Commodity
Marx insisted, unlike the classicals, that labour power was a commodity
like any other and that its value (that is, the wage) was determined in the
same way as any other commodity (Marx, 1974a, Chapter 6,
pp.164-72; Chapter 19, pp.501-7; 1975, Section 5, pp.50-3). His
argument was developed as follows. The value of any commodity is
determined by the quantity of socially necessary abstract labour
required to produce it (that is, the labour required under the prevailing
technical and social conditions of production) (see section 1.1). The
value of labour power - also a commodity - is similarly determined by
the quantity of socially necessary labour required to produce it and is
equal to a given quantity of wage goods. These wage goods are necessary
to 'produce, develop, maintain and perpetuate' (Marx, 1975, p. 46) the
worker's labour power. They are 'the means of subsistence necessary for
the maintenance of the labourer' (Marx, 1974a, p. 167).
However, on closer inspection, the value of labour power is not
determined in exactly the same way as the value of any other commodity.
It differs in at least two respects. On the one hand the production process
is owned and controlled by the capitalist in the case of a commodity but

by the worker in the case of labour power. On the other hand
exploitation occurs in the course of the production of a commodity but
not in the course of the production oflabour power. However, the more
fundamental difference arises because the mechanisms which regulate
supply and demand are quite different for commodities and labour
power. This can be seen by considering the outcome which typically
occurs when demand exceeds supply (and vice versa) in the case of a
commodity and labour power respectively.
In the case of a commodity, an excess of demand over supply would
lead, according to Marx, to an increase in the price of the commodity
over and above its value, the effect of which would be that the capitalists
producing this commodity would receive a higher than average rate of
profit. This would lead, in turn, via the process of competition between
capitalists, to increased investment in the production of that commodity
up to the point where the rate of profit was restored to the competitive
average. In short, an excess of demand over supply for any commodity
would lead, in the long run, to an increase in the production of that
commodity up to the point where supply and demand are in equilibrium.
In the case of labour power, an excess of demand over supply would
lead to an increase in the price of labour power (that is, wages).
However, unlike the production of a commodity, the excess demand
would not have the effect of making the production of labour power
more profitable (since no profit is produced in the production process);
nor, by corollary, would it lead capitalists to increase the output of
labour power (since capitalists do not produce labour power). What
typically occurs in this case is that the demand itself is reduced (through
the adoption, by capitalists, of capital-using/labour-saving technical
innovations) or supply will be increased (through natural increase or
immigration) or both. The more general point is that the imbalance
between demand and supply in the case oflabour power is not regulated
by changes in the production oflabour power but by changes in all of the
other factors affecting demand and supply. A similar type of reasoning
can also be applied to the case where supply exceeds demand.
The more general point is that labour power is not a commodity like
any other, as Marx claimed, essentially because it is not produced in the
same way as any other commodity. Ironically Marx recognised that the
mechanism regulating the production and value of labour power is
different from the mechanism regulating the production and value of
commodities, yet he did not draw the logical inference that labour power
is not therefore a commodity like any other (see Bortkiewicz, 1952,
pp. 56-7; Lange, 1968, pp. 82-4). It is now necessary to explore more
thoroughly the nature of the mechanism which regulates the production
and value of labour power.
2.3.3 The Mechanism Regulating the Level of Wages
The mechanism which regulates the production and value of labour

power is, as Marx variously terms it, 'a relative surplus population or
industrial reserve army' (1974a, p. 589). This mechanism, he argues, is
the direct outcome of capital accumulation because capital accumula-
tion is characterised by the adoption of capital-using/labour-saving
technical innovations whose net effect is to create a surplus of labour
power (ibid., p.590; 1975, pp.76-7). In other words, the rise in the
organic composition of capital, which Marx believed to be an inevitable
feature of capitalist development, would bring with it a high level of
unemployment. This would result in a relative surplus population in the
sense that it is surplus to the requirements of capital accumulation: 'a
population of greater extent than suffices for the average needs of the
self-expansion of capital, and therefore a surplus population' (Marx,
1974a, p. 590).
Marx described the formation of this relative surplus population as 'a
law of population peculiar to the capitalist mode of production' (Marx,
1974a, pp.591-2). His basic point was that the supply of labour in
capitalist society was not determined solely by natural increase (as
Malthus claimed)11 but also by the nature of technical change within
capitalist society (see Marx, 1973, p.604; 1974a, pp. 591-2; see also
Sweezy, 1942, pp. 83-95;12 Lange, 1968, pp. 83-4).
The significance of this relative surplus population in the present
context is that it is the mechanism which, Marx believed, regulated the
production and value of labour power. This is because unemployed
workers will compete with each other for jobs, the effect of which will be
to push wages down to their minimum level which is just equal to the
value oflabour power. Thus, in Marx's view, 'the general movements of
wages are exclusively regulated by the expansion and contraction of the
industrial reserve army' (Marx, 1974a, p.596).
It is now necessary to examine the implications which follow from this
view, particularly in relation to the long-term trend in wages.
2.3.4 The Long-term Trend in Wages
Marx defined wages as a share of net output. This definition is rendered

explicit in Wages, Price and Profit (Marx, 1975, Section 12, pp. 60-2)
where Marx defined net output (albeit without using that term) as that
which remains after deductions have been made for 'the value of the raw
materials and other means of production used upon it' (Marx, 1975,
p. 60). Net output, he continues, is the 'fund from which both he (the
worker) and the capitalist have to draw their respective shares or
dividends, the only value to be divided into wages and profits' (ibid.,
p.60).
There is consequently an inverse relation between wages and profit for
a given level of net output which can be graphically illustrated as follows:
Net output
A
DI------~
o Wages c 8
Net output
Figure 2.2 The shares of profits and wages in net output
In this diagram, net output is plotted on the two co-ordinates OA and

OB and the line AB is the set of possible combinations of wages and
profits. OC is the share of wages and on is the corresponding share of
profits. Any change in the share of wages, for example, from OC to OCI ,
will result, ceteris paribus, in a fall in the share of profits from OD to
onl.
Marx argued that the share of wages in net output could, in turn, be
divided into two components: the first component refers to the basic
'necessaries absolutely indispensable for living and multiplying' (Marx,
1975, p. 72) (for example, food) and the second component refers to that
portion of the wage which is determined by 'historical tradition and
social habitude' (ibid.) (for example, alcohol and tobacco) (see also
Marx, 1974b, p.407). The first component forms the 'ultimate limit'
(ibid.) below which wages could not fall without endangering the
reproduction of the labour force. It is, in other words, 'the physical

minimum of wages' (ibid., p. 74). It follows, as Marx correctly pointed
out, that the rate of profit corresponding to this basic minimum wage is
'the maximum rate of profit' (ibid.) (not to be confused with the concept
of the maximum rate of profit used in a previous subsection (2.1.3) which
is the rate of profit corresponding to zero wages).
Marx points out that a fully-developed capitalist society, 'in which
capital domineers over the whole process of production' (Marx, 1975,
p. 75), wages will be kept to this basic minimum. In short 'the general
tendency of capitalistic production is not to raise, but to sink the average
standard of wages, or to push the value of labour more or less to its
minimum limit' (ibid., p. 77). It is for this reason that Marx can be said to
have a subsistence theory of wages (see Walker, 1978, p. 158).
While Marx's theory of subsistence wages applies to the long-run he
did allow for the possibility that wages might rise in the short-run. This
would typically occur in a period of rapid economic growth where 'the
demand for labourers may exceed supply and therefore wages may rise'
(Marx, 1974a, p. 575); and the wage bill would continue to rise as long as
'its rise does not interfere with the progress of accumulation' (ibid.,
p. 580). If, however, it did interfere with capital accumulation, economic
growth would be reduced leading to a reduction of wages to their
subsistence level. This was an intrinsically short-run phenomenon which
arose because of the sporadic nature of capitalist economic growth;
Marx, however, never considered a rise in real wages to be a long-run
phenomenon of capitalism.
It is implicit in Marx's analysis that the total wage bill (that is, the real
wage multiplied by the number of workers) will represent a diminishing
share of net output as economic growth proceeds (Howard and King,
1975, pp. 132-5, make a similar point; Robinson, 1967, p.36, and
Sowell, 1960, give a contrary view). This is implicit because of Marx's
views on the nature and consequences of technical change. This is
because the rate of growth of net output would be increased by the
adoption of new capital-using/labour-saving technology while the rate
of growth of the total wage bill would be reduced by a fall in the value of
the real wage as a result of productivity increases in the wage goods
sector (although the latter may be increased by the employment of
additional workers). While the precise outcome of this remains indeter-
minate it would seem, nevertheless, that the thrust of Marx's argument
implies that the share of wages in net output would tend to decrease
while the share of profits (though not necessarily the rate of profit)l3
would tend to increase. This is illustrated in Figure 2.3.
Net output
D~------''''''-
o Wages c 8
Net output
Figure 2.3 Marx's view ofthe effect of an increase in net output upon the shares
of profits and wages
2.3.5 The Role of Trade Unions in Wage Determination
The role of trade unions, according to Marx, was to prevent wages from
falling below their minimum, subsistence, level. It is in this context that
Marx states that 'trade unions work well as centres of resistance against
the encroachments of capital' (Marx, 1975, p.78). On the same page
Marx describes such struggles over wages as merely 'retarding the
downward movement, but not changing its direction' (ibid).
Marx's view of the role of trade unions was probably an accurate
reflection of their role and strength at the time when he was writing.
Moreover, the crucial role which the reserve army of unemployed played
in his explanation of wages precluded him from attributing any
fundamental importance to trade unions. However, trade unions have
now achieved considerably more power than in Marx's time and it
appears that, in at least some capitalist countries (notably in the UK and
in the US) the share of wages in net output has been increasing. These
facts, based upon the researches of Glyn and Sutcliffe (1972)14 in the
UK, Boddy and Crotty (1975)15 and Weisskopf(l979)16 in the US, and
Hill (1979) and Heap (1980)17 in OECD countries, raise a number of
problems for Marx's theory of wages. First of all, it seems to contradict

the prediction, which is implicit in his work, that the share of wages in
net output will tend to decrease. Secondly, it suggests that the reserve
army of the unemployed are not necessarily the only, nor even the main,
factor determining the level of wages; it would seem that trade unions
today playa predominant role in the determination of wages which was
not foreshadowed in Marx's theory. Thirdly, and most important, rising
real wages have been a cause of crisis within capitalism because of their
adverse effect upon the rate of profit. This implies that trade unions may
have a stronger role, and the industrial reserve army a weaker role, in the
determination of wages than that envisaged in Marx's model. It also
suggests that the respective bargaining power of workers and capitalists
may contain a more reliable focus for explaining the long-term trend in
wages and the possibility of crisis within capitalism (see Roemer, 1978).
2.3.6 Conclusion
The analysis in this section has highlighted a number of problems with

Marx's theory of wages. It has shown that his analogy between
commodities and labour power breaks down essentially because both
are produced in very different ways and their values are determined in
very different ways. In addition it would seem that, in many sectors of
the advanced capitalist economy, trade unions have replaced the
industrial reserve army as a crucial determinant of wages. Finally, as
regards the long-term trend in wages which, according to Marx, would
represent a declining share of net output, the currently available
evidence from some capitalist countries is that the share of wages in net
output has increased rather than decreased. All of this suggests that
Marx's theory of subsistence wages can no longer be regarded as a
cogent explanation of the way in which wages are determined or the level
at which they are determined.
This suggests that wages may be more adequately explained in terms
of the relative bargaining strength of workers and capitalists, and that
rises in real wages, as a result of the growing strength of trade unions
could, ceteris paribus, threaten profits and thereby cause a crisis within
capitalism.
2.4 CONCLUSION
This chapter has focused upon the three basic theories of crisis that are
contained in Marx's writings. Although other theories of crisis have
been developed by Marxists in recent years - notably theories of fiscal

crisis (O'Connor, 1973) and of legitimation crisis (Habermas, 1976)-
these were not considered here, essentially because they do not figure in
Marx's work nor do they play an important part in the writings of
Castells and Lojkine. Castells and Lojkine rely, in various ways and to
varying degrees, upon the three variants of Marxian crisis theory
examined above. As a consequence this chapter provides a bench-mark
against which to assess how they have used these theories in their
writings (see Chapters 6 and 7).
It is now necessary to examine another element of Marx's political
economy - the theory of rent - in order to prepare the way further for the
critique of the writings of Castells and Lojkine.
3 Marx's Theory of Rent
3.0 INTRODUCTION
Marx devoted considerable attention to the discussion ofland and rent

(Marx, 1969; 1974c, pp. 614-813; 1977, pp. 309-21) which suggests that
he regarded it as an important element in the analysis of capitalism.
However, his theory of rent is based entirely upon the analysis of
agricultural production even though the rest of his theory of capitalism
is based upon the analysis of industrial production. In this sense Marx's
analysis of rent is not integrated tightly into the remainder of his
analysis.
The purpose of this chapter is to provide a rigorous formulation of
Marx's theory of rent which will be used subsequently to evaluate the
Marxian concept of rent found in the writings of Jean Lojkine (see
subsection 7.3.4). The formulation presented here will involve severing
Marx's theory of rent from his labour theory of value (because of the
difficulties associated with the latter - see Chapter 1) and recasting it
within the linear equation approach outlined above (see section 1.6).
Marxists have not, until recently, devoted much attention to the
analysis ofland or rent. Those who have considered the topic tend to fall
into two categories (see Gibson and Esfahani, 1983) - those who
maintain an intimate connection between Marx's theory of rent and his
labour theory of value - and those who, abandoning Marx's theory of
value, follow the approach pioneered by Sraffa. The former, sometimes
referred to as Fundamentalists, include, Harvey, 1973, 1982; Harvey
and Chatterjee, 1974; Walker, 1974, 1975; Broadbent, 1975; Breugel,
1975; Byrne and Beirne, 1975; Clarke and Ginsburg, 1975; Edel, 1976;
Ball, 1977, 1980; Markusen, 1978; Murray, 1977, 1978; Massey and
Catalano, 1978; Fine, 1979, 1980; Lauria, 1984. The latter, sometimes
referred to as NeoRicardians, include, Scott, 1976a, 1976b, 1979, 1980,
Chapter 3; Kurz, 1978; Abraham-Frois and Berrebi, 1979, pp. 93-102;
Bose, 1980, Chapter 13; Quadrio-Curzio, 1980; Bandyopadhyay, 1982;
Gibson and Esfahani, 1983; Gibson and McLeod, 1983. The analysis
presented here relies heavily upon the 'NeoRicardian' approach.
The most important feature about land in any system of production is
that it is essential to the production of every commodity but is not itself
56
Marx's Theory of Rent 57
produced. l In fact land cannot be produced at all (with the possible

exception of land reclamation). The significance of this fact is that the
price of land is not the same as the price of any other commodity and
hence cannot be explained in the same way. However the fact that it is a
constituent element in the price of each commodity means that it must be
explained within the framework of the determinants of the price of each
commodity. This means that, although rent will emerge in any system of
production where land is scarce and/or variable in quality, 2 the precise
way in which this occurs will depend upon how prices are determined in
that system of production.
In capitalist society, land is usually owned privately and this will be
assumed throughout the analysis in this chapter. More specifically, it
will be assumed that land is owned by landowners and leased to
capitalists who use it, in turn, to produce commodities. This was the
framework within which Ricardo (1970, Chapter 2; 1966) and Marx
developed their analyses of rent. The results of this analysis will not be
affected if allowance is made for the fact that the state, capitalists, and
even workers, own land, since the conceptual status of income derived
from the ownership ofland (that is, rent) is not in any way affected by the
characteristics of its recipients.
The analysis below differentiates four different types of rent. These
are: differential rent I, differential rent II, monopoly rent I and
monopoly rent II. These four types of rent are all present in Marx's
work, and the present analysis demonstrates that there is a firm
conceptual basis for this fourfold typology, even though there may be
some difficulties in rigorously applying them to empirical situations.
It will be shown (in section 3.1) that differential rent I arises ifland is
scarce and variable in quality and suitable for different uses - a fairly
typical real-world situation. This will be demonstrated in the simplest
possible case where one agricultural commodity is produced on two
different types ofland. This result can then be readily generalised to the
case where n different commodities are produced on n different types of
land. Differential rent II will be shown (in section 3.2) to be inherently
temporary and transient in the sense that it arises when there is a change
in the technology for producing a given commodity, such that the
change is temporarily unknown to all landowners or unavailable to all
capitalists. Monopoly rent I (for which Marx used the term 'absolute
rent') will be seen (in section 3.3) to arise if landowners form a cartel in
order to impose a rent (independently of differential rent I and II) on the
use of all land essential for the production of basic commodities. Finally,
monopoly rent II will be shown (in section 3.4) to be a special type of rent
in the sense that it arises only on land which produces a special type of
commodity, that is, a commodity which can only be produced on one
unique plot (or set of plots) of land and whose output cannot be
increased above a fixed upper limit.
3.1 DIFFERENTIAL RENT I
In a capitalist economy rent arises when land is scarce and/or variable in

quality (quality depending on fertility or situation or both - see Ricardo,
1970, p. 70; Marx, 1974c, p.650; Smith, 1976, p.250). If it is assumed
that corn is the only commodity that is produced in the economy, and if
it is produced by combining the inputs3 of corn and labour (itself
produced by wages which consist of corn) and if it is produced on only
one plot ofland which is also the most fertile (and is designated as type (i)
land)4 then the production conditions of corn on this land can be written
as follows:
Pi(i) = Pi AlI(i) + Pi A 2l (i) + ... P';: (A nl (i) + mal (i) (l + r(i))
where
pm
1,2,n = the money price of a unit of commodity 1, 2, n, where
commodity 1 is corn;
= the quantity of commodity 1, 2, n which enters into the
production of commodity 1 (that is, corn). It follows
from the above assumptions that A2h Anh are zero;
m = the money wage rate per unit of labour;
= the quantity of labour employed in the production of
commodity 1 (that is, corn);
r = the rate of profit;
(i) = the plot ofland on which commodity I is produced and
whose fertility is known to be the best.
The relationship between output and type ofland, which is expressed by
this equation, is illustrated in Figure 3.1. In this diagram the output on
type (i) land is broken down into its constituent parts: the inputs of corn
(pm A(i)) and labour (ma(i») and the quantity of profit (r (pm A(i) + ma(i»))'
The possibility of differential rent I now arises if type (i) land and, by
implication, the corn produced on it, becomes scarce. 5 This scarcity can
be overcome in one of two ways: either by introducing a technical
innovation such that the increased output can be produced on the same
plot or by extending production to an inferior plot. If the second option
Profit
Output Corn
Wages
(i)
Type of Land
Figure 3.1 The relationship betwen output and type of land
is adopted, 6 and if it is assumed that the second plot is less fertile than the
first (and is designated as type (ii) land), then the production conditions
of corn on type (ii) land can be written, using the above notation, as
follows:
Pf<ii) = P'j A l1 (ii) + PT A 21 (ii) + ... 1"; Anl(ii) + mal(ii) (1 + r(ii»·
It follows from the preceding assumptions that the basic difference

between the production of corn on type (i) land, and the production of
corn on type (ii) land is that the same input of commodities (P'j AnI = P'j
Anl(ii» and labour (mal(i) = mal(ii» produce a smaller output on type (ii)
land, which may result in higher prices or lower profits, depending on
the assumptions made about the nature of the commodity being
produced (that is, basic or non-basic) and the wage and profit rates (that
is, whether they are constant or flexible). If it is assumed, following
Ricardo and Marx, that the commodity being produced (in this case
corn) is a basic (see section 1.6) and that the wage rate is constant while
the profit rate is flexible, then it is possible to infer that the effect of
extending production from type (i) to type (ii) land will be to reduce the
general rate of profit in the economy. 7 This reduction in the rate of profit
will first occur for those capitalists producing on type (ii) land since, with
constant real wages and hence constant prices for basics, the diminished
returns experienced by those capitalists producing on type (ii) land (by
comparison with those producing on type (i) land) will inevitably lead to
a lower rate of profit (that is, r(ii) < r(i». This lower rate of profit will then
spread, via competition, to those capitalists producing on type (i) land,
since the capitalists producing on type (ii) land would be willing to pay a
rent for the lease of type (i) land up to the point where the rate of profit
on both types ofland is the same. This rent is equal to differential rent I.
In this way differential rent I emerges on type (i) land independently of
whoever owns the land (see Marx, 1974c, p. 647; see also Benetti, 1976,
p. 12). 8 Thus, the final outcome of extending production from type (i) to
type (ii) land can be written in terms of two equations:
P%) = P'{' A11(i) + Pi A 21 (i) + ... P':: An1(i) + mal(i) (1 + r) +
DRI(i)
P~ii) = P'{' All(ii) + Pi A 21 (ii) + ... P';: An1(ii) + mal(ii) (1 + r)
where, in addition to the usual notation:

DRI(i) = differential rent I on type (i) land.
These two equations are expressed diagrammatically in Figure 3.2. This
diagram, together with the above equations and the analysis underlying
them, illustrates that when land is scarce and/or variable in quality it will
yield a rent, and hence a price, since the price ofland is nothing more
than the discounted annual rent over a number of years. The origin of
this price is the surplus profit which emerges when a given commodity is
produced on a more fertile plot ofland. This price ofland constitutes, in
tum, an income for the owners of that land, and it is in this sense that it is
possible to speak of an inverse relationship between the income of
landowners and the income of capitalists (that is, profits) - as long as real
wages are assumed to be constant.
Differential rent I would also emerge if production, instead of being
extended to land of inferior quality, were extended to land of superior
quality. In this case those capitalists producing on the superior land
would, under the above assumptions, obtain an above-average rate of
profit. However, this would only be temporary since those capitalists
producing on the inferior land would be willing to pay a rent up to the
point where the difference in the rates of profit was abolished. In this
way differential rent I would emerge on the superior land without,
however, any concomitant fall in the rate of profit: the rate of profit
would remain the same. Marx made this point - a correct but minor
point - in criticism of Ricardo, West and Malthus, who assumed that
differential rent I only emerged when production was extended from
superior to inferior land (see Marx 1974c, p.658).9
Marx's analysis of differential rent I is substantially identical to that
developed here which, in tum, is identical with that developed by
Ricardo (though Ricardo does not make an explicit distinction between
Differentia I rent I {
--------~----------,
I
Profit {
~---------+-----------;
ll------t--------j
Output
Com
W.~
(i) (ij)
Type of Land
Figure 3.2 The relationship between output, type of land and rent
differential rent I and II despite its implicit presence in his work (see
notably Ricardo, 1970, Chapter 2).
3.2 DIFFERENTIAL RENT II
The essential features of differential rent II can be brought out most

easily in the simple case of a capitalist economy where corn is the only
commodity being produced (that is, the same model which was used to
explicate the nature of differential rent I). In this economy it can be
shown that differential rent II will arise (over and above differential rent
I) if there is a change in technology such that, ceteris paribus, the rate of
profit is increased on at least one plot, and perfect competition is
hindered, temporarily.
If a new technique is invented for the production of a given
commodity (in this case, corn), this effectively means that there are two
different technologies for the production of that commodity: the old
technology (technique I) and the new technology (technique II). Ifit is
assumed that the new technology is more efficient than the old then it
follows, given the assumptions made in the preceding section (that is,
that the commodity being produced is a basic, and that the real wage is
constant), that the new technology will give rise to a higher rate of profit
than the old. If, in turn, it is assumed that competition between
capitalists is hindered and that (for example) only one capitalist has
access to the new technology, then the increased rate of profit obtained
by the new technology will, ceteris paribus, be appropriated by this

capitalist as surplus profit (that is, a surplus over and above the going
rate of profit). This surplus profit is differential rent II.
lt follows from this that differential rent II can only arise if two
conditions are satisfied. The first is that two different technologies
coexist (one more efficient than the other) for the production on land of
a given commodity. Marx expressed this first condition by affirming that
differential rent II arises w~en 'capitals of different productivity are
invested successively in the same plot of land or side by side in different
plots of land' (Marx, I 974c, p.674).10 The second condition is that
competition between capitalists is not operative so that the surplus profit
obtained by using the more efficient technology is not transformed into
differential rent I and appropriated by the landowner; it is appropriated
by the capitalist as differential rent II.
There are two possible reasons why competition between capitalists
may be hindered, even if only temporarily. The first is that there may not
be a perfect flow of information throughout the system so that a new and
more efficient technology for producing a particular commodity may
not be universally known or universally available to all capitalists.
The second reason, and this is the one cited by Marx, is the leasing
system. This could hinder competition between capitalists in the
situation where one capitalist secures a long lease, in the course of which
a new and more efficient technology is invented and applied by the
leasing capitalist, giving rise to a surplus profit, that is, differential rent
II. Marx referred to this possibility in the following terms:
Rent is fixed when land is leased, and after that the surplus profit
arising from successive investments of capital flows into the pockets of
the tenant as long as the lease lasts. This is why the tenants have
fought for long leases, and, on the other hand, due to the greater
power of the landlords an increase in the number of tenancies at will
has taken place, that is, leases which can be cancelled annually (Marx,
1974c, p.675).
This analysis can be readily expressed in terms of the above notation.
The two equations describing the conditions of production of the
commodity in question (that is, com) using two different techniques of
production (technique I and technique II) can be written as follows:
Pi(I) = Pi A 11(1) + Pi A 21 (I) + ... P: Anl(l) + mal(l) (1 + r)
Pi(II) = Pi A 11 (II) + Pi A21 (II) + ... P: Anl(II) + mal (II) (I + r)
where, in addition to the usual notation, the subscripts:
I = technique I (that is, the old technique of production)

II = technique II (that is, the new technique of production)
On the basis of the assumption that technique II is more efficient than
technique I (that is, that the commodity being produced is a basic and
that the real wage is constant) it follows that, ceteris paribus, the rate of
profit obtained by producing with technique II will be higher than that
obtained by producing with technique I. This difference constitutes
differential rent II for the capitalist producing with technique II. This
fact can be brought out more clearly by rewriting the two equations as
follows:
Pi(l) = Pi All(l) + Pi A 21 (I) + ... r,: An1(I) +mal(I) (I +r)
Pi(II) = Pi A ll(II) + Pi A 21 (II) + ... r,: An1(II) + mal(II) (l + r) +
DR I!.
where, in addition to the usual notation,
DR II = differential rent II.
These equations, and the analysis underlying them, are expressed
diagramatically in Figure 3.3.
differential rent \I {
r--------4--------~
Prof;, 1
Output Com 1 f-----+-----1
w,~ lL...--_--'---_.....L.--_
(i) {ii}
Technique
Figure 3.3 The relationship between output, technique of production and rent
From this analysis it is possible to affirm that differential rent II is a

temporary form of rent, that is, it is an income which goes to the
capitalist (not the landowner) only so long as the conditions of perfect

competition do not operate. Once, however, the obstacle to competition
is removed, then differential rent II will disappear as such and become
part of differential rent I, and appropriated by landowners.
This analysis of differential rent II is substantially identical to that
developed by Marx: the only difference is that Marx's analysis is much
more detailed. In fact Marx devotes five chapters (1974c, Chapters
40-44) to the analysis of differential rent II in order to show the effect of
using different methods of production when prices of production are
constant (the 'first case' examined in Chapter 41), falling (the 'second
case' examined in Chapter 42) or rising (the 'third case' examined in
Chapter 43). In turn, each of these three cases have three variants
depending on whether the new methods of production have constant,
falling or rising productivity. All of these details are entirely consistent
with the analysis presented here.
3.3 MONOPOLY RENT I
Monopoly rent I (for which Marx used the term 'absolute rent')" arises
when the owners ofland form a cartel and impose a rent on the use of all
land, irrespective of quality, which is necessary for the production of any
non-substitutable basic or non-basic commodity. Thus monopoly rent I
acts like a tax on land which is imposed by landowners, and which falls,
given the preceding assumptions (notably constant real wages), on
capitalists.
There are two basic conditions necessary for the existence of
monopoly rent I. The first is that the landowners who impose it must
form a cartel and the second is that this cartel must have ownership of all
the land necessary for the production of at least one non-substitutable
commodity. The importance of the first condition derives from the fact
that iflandowners do not form a cartel and fix the amount of monopoly
rent I, then some landowners may not impose the rent at all while others
may impose it at different rates. If this were to occur then it is possible,
with more intensive production on the non-monopoly rent I plots, that
monopoly rent I may not be effectively imposed at all. The second
condition is important because if the cartel does not have ownership of
all the land essential to the production of at least one non-substitutable
commodity, then capitalists will cultivate that land which is owned
outside the cartel, thereby avoiding monopoly rent I. Thus both of these
Differential rent I 1
Monopoly rent I
~~~LL~~LLLL~~~
Profit
Com j~---r------i
W,," 1 L......--_~-----'-_
(i) (ii)
Type of Land
Figure 3.4 The incidence of monopoly rent I on different types of land
conditions must be satisfied for monopoly rent I to exist.

The effect of monopoly rent I on wages, profits and prices will depend
upon whether the commodity being produced is a basic or a non-basic
and on the assumptions made about the variability of wages and profits.
These effects will now be analysed, first in the case of basics and second
in the case of non-basics.
If monopoly rent I is imposed upon land producing at least one non-
substitutable basic commodity and if it is assumed that the real wage is
constant (as Marx assumed), then its effect will be to reduce profits by
the amount of the monopoly rent I. This effect is illustrated in Figure 3.4
(which is an adaptation from Figure 3.2).
The real wage may, however, be variable and, in this case, the effect of
monopoly rent I on wages and profits will depend upon the relative
bargaining power of workers vis-iI-vis capitalists and the ability of one to
force the other to bear the cost of monopoly rent I. If, on the other hand,
the commodity is a non-basic, then the effect will be to raise prices by the
amount of the rent, producing no effect on either wages or profits. In this
way monopoly rent I acts entirely like a tax and, a tax on luxuries (that is,
non-basics) raises the price of luxuries by the amount of tax.
Marx's analysis of monopoly rent I is significantly different to the
analysis developed here and is seriously flawed. The reasons for this
emerge from a close examination of his analysis.
Marx based his analysis of monopoly rent I on the case of a plot of

agricultural land which yielded neither differential rent I or II. On this
plot, a landowner imposes a rent, monopoly rent I (see Marx 1974c,
pp.748-60). Marx explained that this occurred because the organic
composition of capital in the agricultural sector is below average for the
economy as a whole. Marx regarded this assumption as crucial since
'whenever this assumption no longer holds, the corresponding form of
rent likewise no longer holds' (Marx, 1974c, p.760; see also p.765).
From this assumption, Marx inferred that the agricultural sector
produced an above-average amount of surplus value. However, con-
trary to the normal competitive conditions of capitalism, the existence of
'landed property' (ibid., p. 761) inhibits the distribution of the above-
average amount of surplus value from the agricultural sector to the rest
of the economy.12 This, he argues, gives rise to 'a surplus-profit which
could be converted into rent' (ibid., p. 761), which is monopoly rent I.
This is the essence of Marx's analysis. There are a number of
difficulties with it, three of which may be noted. The first is that Marx
did not know that the organic composition of capital in the agricultural
sector was higher than in the rest of the economy (a fact he curiously
acknowledges - Marx, 1974c, p. 760). Thus the basic premise of Marx's
argument is based upon an assumption, not a fact, and no evidence is
adduced to either confirm or refute this assumption. The second
difficulty is that landowners are equally ignorant about the organic
composition of capital in the agricultural sector, and hence are also
ignorant of the above-average quantity of surplus value which is
allegedly produced in that sector. Consequently they cannot act upon
this information (even if it were correct) although Marx's argument
requires them to do so. The third and more general difficulty relates to
the labour theory of value which underlies this analysis. The difficulty is
that an above-average amount of surplus value is not necessarily
associated with an above-average amount of profit; it may even be
associated with a negative amount of profit (see Chapter 1). Consequen-
tly, the existence of an above-average amount of surplus value in the
agricultural sector (even if correct) cannot be taken as proof of the
existence of above-average profits, which are supposed to be the basis of
monopoly rent I. These three reasons provide a compelling case for
rejecting Marx's analysis of monopoly rent I. However, it has also been
shown in this section that the concept of monopoly rent I is amenable to
a clear and rigorous formulation and this formulation should therefore
be adopted.
3.4 MONOPOLY RENT II
It is possible, following Marx, to distinguish a second type of monopoly

rent which, for convenience, I shall term monopoly rent II. Marx
referred to this possibility when he cited the case of 'a vineyard
producing wine of very extraordinary quality which can be produced
only in relatively small quantities' (Marx, 1974c, p. 775; Adam Smith
also examined this case - see Smith, 1976, Chapter 9, pp. 259-60). As
will be seen, Marx's analysis of monopoly rent II (though he did not use
this term) is substantively identical to that presented here.
In order to understand the nature of monopoly rent II it is necessary
to make a distinction between two different types of commodity:
(a) reproducible commodities (and their substitutes) whose output can
be increased in response to an increase in demand. Most com-
modities produced in capitalist economies (in fact in every econ-
omy) are reproducibles and it is this set of commodities which is the
central focus of Marxian political economy. The important charac-
teristic of reproducibles in the present context is that their market
price tends to equal their price of production;
(b) non-reproducible commodities (which have no substitutes) whose
output cannot be increased in response to an increase in demand.
Such commodities include antiques, oil paintings by 'grand masters'
and rare wines. The important characteristic of non-reproducibles,
in the present context, is that their market price tends to exceed their
price of production.
The significance of this distinction in the present context is that
monopoly rent II arises only on land which produces non-reproducible
commodities. Marx's example of 'a vineyard producing wine of very
extraordinary quality' represents a typical example of a non-reproduci-
ble commodity. This is because an increase in demand for this
commodity simply results in an increase in its market price with no
increase in output. As a result, its market price will tend to exceed its
price of production. Marx refers to this market price as 'a monopoly
price' (Marx, 1974c, p. 775) because it is determined by 'the purchasers'
eagerness to buy and ability to pay, independent of the price determined
by the general price of production' (ibid.).
Monopoly rent II is determined entirely by this difference between
market price and price of production. This is because the commodity
(for example, wine) can only be produced on this land (that is, the
vineyard) and it (or a suitable substitute) cannot be produced on any

other land. Since this land is limited it follows, ceteris paribus (that is,
assuming the land cannot be cultivated more intensively) that an
increase in the demand will lead to an increase in monopoly rent II. The
reason for this is that the increase in the market price for this commodity
creates the possibility of a higher rate of profit for the production of this
commodity by comparison with that obtainable in other spheres of
production. However, competition between capitalists will ensure that a
rent (that is, monopoly rent II) is paid to the owner of this land up to the
point where the difference in the rates of profit disappears. In this way,
the difference between market price and price of production is
converted, via competition between capitalists, to monopoly rent II for
landowners.
If it is assumed that the land in question (that is, the vineyard) is not
just uniquely suited, but only suited, to the production of wine, then it
can be affirmed that the rent on this land (that is, monopoly rent II) will
be determined independently of the differential rent (lor II) on any other
land. The landowner may also levy a monopoly rent I on this land,
irrespective of whether or not this is levied on other land.
The effect of monopoly rent II on wages, profits and prices depends
upon whether the land in question produces a basic (for example, oil) or
a non-basic (for example, wine) and on the assumptions made about the
variability of wages and profits. Its effects are identical to those
produced by monopoly rent I, as outlined in the preceding section.
3.5 CONCLUSION
The analysis in the preceding sections has illustrated how the price of
land is determined under four different sets of conditions, thereby giving
rise to four different types of rent. The basic insight conveyed by this
analysis is that the price of land - which is 'nothing but the capitalized
and therefore anticipated rent' (Marx, 1974c, p. 808; see also Keiper, et
al., 1961) - is an income which landowners receive from capitalists for
the use of land (assuming that wages are constant and that neither
workers nor capitalists own land).
The price of any plot of land is thus conceptually decomposable into
its various rental components (that is, differential rent I and II,
monopoly rent I and II) some of which may be zero. Thus differential
rent I exists when the output per unit of input on a plot ofland is greater
than the output per unit of input on those plots which yield no rent. In
practice this means that, in a competitive capitalist economy, the price of

a plot of land is the capitalised value of differential rent I. Differential
rent II, by contrast, exists when the difference in output between two
otherwise identical plots of land is due entirely to the different types of
technology employed on them. In this situation, the land which employs
the more efficient technology, that is, yields a higher net output, will
yield differential rent II. In the urban context, this new technology may
be a new and taller building (for example, a larger office block) or an old
but renewed one (for example, as a result of gentrification). Monopoly
rent I arises when landowners form a cartel to raise the price of certain
plots above that prevailing for identical plots elsewhere. In the urban
context, this type of rent can arise through land-use zoning, particularly
when the zoning is exclusionary (see Williams and Norman, 1971) and
its effect is to raise the price ofland above that which would prevail in a
competitive market. Finally, monopoly rent II exists when (a) the land is
uniquely (and perhaps only) useful for producing a particular com-
modity and (b) the quantity of the commodity being produced cannot be
increased above a fixed upper limit and (c) there is no suitable substitute
for this commodity within the system of production. In the urban, as in
the agricultural context, there are few examples of this type of rent.
However, a residential area in a city which was highly 'exclusive' and for
which there was no other substitute in terms of 'exclusivity' could have
an element of monopoly rent II in its land prices.
The analysis in this chapter, in short, has attempted to place Marx's
rent categories on a firm conceptual footing so that they can be usefully
applied to both rural and urban contexts. The chapter also provides a
basis for evaluating how the categories of rent are used in the work of
Jean Lojkine (see section 7.3.4).
The next chapter will now examine one final element in Marx's
political economy, namely, the distinction between productive and
unproductive labour.
4 Marx's Theory of
Productive and
Unproductive Labour
4.0 INTRODUCTION
Marx considered the distinction between productive and unproductive

labour to be of the 'greatest importance' (Marx, 1969, p.396) even
though his writing on the topic is rather confused. Nevertheless, the
distinction continues to be used in Marxist writing and is given some
prominence in Lojkine's recent attempt to develop a Marxist urban
sociology (see section 7.2). For this reason it is essential to establish the
precise meaning of the distinction.
This chapter provides a thorough exegesis of Marx's use of the
distinction and reveals that there are two different and inconsistent
usages implicit in his work. The first usage is outlined and examined in
section 4.1. This usage, referred to for convenience as 'Definition 1',1 is
then shown to give rise to certain anomalies which provide the basis for a
more general definition. In section 4.2 Marx's second definition of
productive and unproductive labour - 'Definition 11'2 - is examined and
explained. Two fundamental problems associated with this definition
are then identified which render it untenable. It is concluded, therefore,
in section 4.3, that the most consistent, tenable and heuristically useful
formulation of the distinction is Marx's slightly modified first definition.
4.1 'DEFINITION I' OF PRODUCTIVE AND UNPRODUCTIVE

LABOUR
The basic criterion for distinguishing between productive and un-

productive labour, according to this definition is whether it contributes
to capital accumulation. Labour which contributes to capital accumula-
tion is productive; labour which does not is unproductive. This criterion
is explicit in the following definition:
Productive labour, in its meaning for capitalist production, is wage-
labour which, exchanged against the variable part of capital (the part
of the capital that is spent on wages), reproduces not only this part of
70
Marx's Theory of Productive/Unproductive Labour 71
capital (or the value of its own labour power) but in addition produces
surplus value for the capitalist. . . only that wage-labour is produc-
tive which produces capital' (Marx, 1969, p. 152).
Unproductive labour, by corollary, is defined as 'labour which is not

exchanged with capital, but directly with revenue, that is, with wages or
profit' (ibid., p. 157). In other words, iflabour results directly in capital
accumulation it is productive,3 otherwise it is unproductive.
The distinction between productive and unproductive labour implies
a further distinction between a capitalist sector (where labour is
exploited and accumulated as capital) and a non-capitalist sector (where
labour, irrespective of whether it is exploited, is not accumulated as
capital).4 The examples which Marx used of, inter alia, actors, tailors,
cooks, musicians, dancers, poets and prostitutes, illustrate that the
'same labour' (Marx, 1969, p.401) is productive if performed in the
capitalist sector and unproductive if performed in the non-capitalist
sector. In other words, only labour performed in the capitalist sector is
productive since it results directly in capital accumulation; labour
performed in the non-capitalist sector is unproductive because it does
not directly result in capital accumulation.
The distinction between productive and unproductive labour is
further reinforced by the distinction between capital and revenue.
Capital refers to the quantity of profits 5 which is exchanged against dead
and living labour for the purpose of capital accumulation. Revenue by
contrast refers to that quantity of profits and wages which is exchanged
entirely for consumption goods. Thus, labour which is exchanged
against capital is productive labour, whereas labour which is exchanged
against revenue is unproductive (ibid., p. 157).
Marx insisted, contrary to Smith's view,6 that the distinction between
productive and unproductive labour (as used in 'Definition 1') had
'absolutely nothing to do with the determinate content of the labour'
(ibid., p.401). The distinction depends rather upon the 'social relations
of production, within which the labour is realized' (ibid., p 157). Since
the social relations of capitalism are characterised by the exploitation of
labour and its accumulation as capital, it follows that this is the basic
criterion for differentiating between productive and unproductive
labour.
Combining these different dimensions of the distinction, Marx's first
definition of productive labour may be formulated as follows: produc-
tive labour is labour which is exchanged against capital, is performed in
the capitalist sector and results directly in capital accumulation;
unproductive labour, by corollary, is defined as labour which is
exchanged against revenue, is performed in the non-capitalist sector and

does not give rise to capital accumulation.
This definition of productive and unproductive labour gives rise to
one important anomaly which must now be considered: the case of
labour which is exchanged against revenue, is performed in the non-
capitalist sector, but nevertheless results, not directly, but indirectly, in
capital accumulation. This anomaly, which Marx did not consider, is
typified by the case of state expenditure on real wages (that is, on
education, health, housing, social welfare, and so on). This case is
anomalous whenever the net effect of state expenditure on real wages is
either to reduce the total cost of the wage bill for capitalists or to increase
the productivity oflabour employed in the capitalist sector. In either of
these cases the effect of state expenditure upon real wages is to increase
the rate of profit and hence the rate of capital accumulation in the
capitalist sector.
The significance of this is that labour which is employed by the state, is
exchanged against revenue (that is, taxes) and is performed in the non-
capitalist sector (that is, the state sector), may nevertheless have the
effect of increasing the rate of profit in the capitalist sector. It could
therefore, under these conditions, be classified as productive labour,
according to Marx's first definition. 7 Thus, it is necessary to modify
Marx's definition by stating that labour is productive if it directly or
indirectly increases the rate of profit and capital accumulation; conver-
sely, it is unproductive if it directly or indirectly decreases the rate of
profit and capital accumulation. In Sraffian terms, labour is productive
if it produces 'basics' (that is, commodities which enter directly or
indirectly into the production of other commodities) and is unproduc-
tive if it produces 'non-basics' (that is, commodities which do not enter
directly or indirectly into the production of other commodities) (see
Sraffa, 1960, p. 54 and section 1.6; see also Hunt, 1979, pp. 311,320 ff).
This criterion provides a clear dividing line between the two types of
labour which is at once consistent with Marx's first definition while
nevertheless free from its anomalies. The main features of this refor-
mulated version of 'Definition I' are summarised in Table 4.1.
4.2 'DEFINITION II' OF PRODUCTIVE AND

UNPRODUCTIVE LABOUR
The distinctive feature of Marx's second definition is his characterisa-

tion of circulation costs as unproductive. In order to understand this
Table 4.1 Marx's 'Definition I' of productive and unproductive labour
Productive Unproductive
Criterion Labour which, directly or Labour which, directly or

indirectly, maintains or indirectly, decreases the level
increases the level of capital of capital accumulation (i.e.,
accumulation (i.e., basics) non-basics).
Mechanism (i) Capital exchanged against Revenue exchanged against

labour in the capitalist labour in the non-capitalist
sector sector whose net effect is to
(ii) Revenue exchanged reduce capital accumulation.
against labour in the non-
capitalist sector whose net
effect is to increase capital
accumulation.
Examples Respectively, Labour involved in

(i) All labour performed in performing the legislative,
capitalist enterprises. judicial and executive
(ii) Labour performed in the functions of the state.
state provision of certain
types of collective
consumption.
definition it is necessary to recall Marx's characterisation of the typical

process, or cycle, of capitalist production (see section 2.2.2). Marx
characterised the process as follows:
M - cffp - P - C' - M'
(I) (2) (3)
where M represents the amount of money at the beginning of the

production cycle; Cfi~ is the labour power and means of production
necessary for production; P represents the production process proper;
C1 represents the new commodity at the end of the productive process;
Ml is the monetised form of the new commodity and is greater than M
thus indicating the amount of the surplus created in the production
process.
This process of production, according to Marx, takes place in three
stages (see Marx, 1974b, p.25). The 'first stage' (1) is the stage of
circulation where money is transformed into commodities. The 'second
stage' (2) is, according to Marx, the stage of 'productive consumption'

(ibid., p. 25) where a new commodity is produced from the combination
oflabour and means of production. The 'third stage' (3) is the final stage
of circulation, and the inverse of the first stage, where the new
commodity is converted back into money.
Marx attempted to show that the two stages of circulation were
unproductive, and that only the production stage was productive. He
put forward the 'general law' that 'all costs of circulation which arise
only from changes in the forms of commodities do not add to their value'
(ibid., p. 152). Circulation costs (such as accounting costs, advertising
costs, packaging costs, storage costs, wholesaling costs, retailing costs,
and so on) are 'an unavoidable evil' (ibid., p. 153) and 'intrinsically'
(ibid., p. 134) unproductive, because they are expenses which must be
met if surplus value is to be realised. In the aggregate they are 'a
deduction from surplus value' (ibid., p. 152) and 'unproductive expenses
as far as society is concerned' (ibid., p. 140).8
Marx further explained why he regarded circulation costs as un-
productive:
a certain amount oflabour-power and labour-time must be expended
in the process of circulation. . . A part of the variable capital must be
laid out in the purchase of this labour-power functioning only in
circulation. This advance of capital creates neither product nor value.
It reduces pro tanto the dimensions in which the advanced capital
functions productively. It is as though one part of the product were
transformed into a machine which buys and sells the rest of the
product. This machine brings about a reduction of the product (ibid.,
p.138).
According to most commentators (Baran, 1957; Gillman, 1957;
Gough, 1972; Harrison, 1973; Hunt, 1979) the precise reason why Marx
classified circulation costs as unproductive is that they are 'costs which
are occasioned solely by the fact that the goods are being produced and
distributed under the capitalist mode of production' (Harrison, 1973,
p. 73). In other words, 'the critical distinction is between those activities
necessary to production in general, and those activities peculiar to
commodity production' (Gough, 1972, p.57; see also Marx, 1974b,
pp.146-52).
The implication of this is that those costs which are technically
necessary for the production of a commodity in any mode of production
are productive, while those which are peculiar to the capitalist mode of
production are unproductive. Since circulation costs are assumed to be
Table 4.2 Marx's 'Definition II' of productive and unproductive labour
Productive Unproductive
Criterion All labour whose performance All labour whose performance

is technically necessary in any is necessary only in the
mode of production capitalist mode of production.
Example Wage costs as well as the cost of Circulation costs and certain
produced and unproduced supervisory costs.
means of production.
peculiar to the capitalist mode of production they are defined, therefore,

as unproductive.
This definition of productive and unproductive labour is schematical-
ly outlined in Table 4.2.
Marx's second definition of productive and unproductive labour
faces two fundamental problems. The first is that it is inconsistent with
the previous definition (,Definition 1') and by implication with other,
more important, aspects of Marx's work. The second problem is that it is
impossible to apply empirically. Both of these problems will now be
considered.
Marx's second definition is inconsistent with his first since circulation
costs are unproductive according to one (the second) but productive
according to the other (the first). This is because, according to the first
definition, the labour involved in the circulation process produces
commodities (that is, use values and exchange values) which are
financed by capital and produced in the capitalist sector and hence are
vehicles of capitalist exploitation and accumulation.
Marx seems to have been aware of this inconsistency and attempted to
resolve it with two different arguments. He argues, 'in the first place' that
the labour performed in the process of circulation' is used up in a mere
function of circulation. It cannot be used for anything else, not for
productive labour' (Marx, 1974b, p. 135). However, this argument is
tautological since it fails to show why the 'function' of circulation is
'unproductive' (in the first sense). Marx argues 'in the second place' that
'society does not appropriate any extra product or value' (ibid.) as a
result of circulation. In fact, however, the opposite would appear to be
the case (unless it is assumed, erroneously, that 'immaterial' com-
modities cannot result in capital accumulation). The services produced
in the process of circulation are commodities like any other, are
produced and financed in the capitalist sector, and result in capital

accumulation and are thus productive (according to the first definition).
There is thus an inconsistency between Marx's two definitions of
productive and unproductive labour. 9
The second problem with Marx's definition is that it is impossible to
apply empirically. This is because the criterion implied by Marx's second
definition involves a comparison between that labour (that is, produc-
tive labour) which is technically indispensable for the production of a
given output, using the same technique, in any mode of production and
that labour (that is, unproductive labour) which is specific to, and
necessitated solely by, the particular features of the capitalist mode of
production. The problem with this criterion is that it assumes that it is
possible to establish the amount of labour which is technically
indispensible for a given output, independently of the mode of
production in which the production takes place. This assumption is
untenable because in a situation where there is choice of technique, no
one technique (or method of production) can be regarded as technically
indispensable except in the context of a particular set of criteria which
are likely to be specific to each particular mode of production. For example,
in the capitalist mode of production, the criteria of cost minimisation
and profit maximisation would probably make one technique of
production 'technically' indispensable, though only in terms of those
criteria. It is conceivable that in a different mode of production, certain
criteria (such as the optimising oflabour inputs, or machinery inputs, or
raw material inputs, or the optimising of spatial and social equity) would
make altogether different techniques of production 'technically' indis-
pensable. This means that techniques of production will tend to be
specific to each mode of production - thereby ruling out the possibility
of empirically applying the criteria implicit in Marx's second definition
(see Harrison, 1973, pp. 74-5).10 It would seem therefore, on the basis of
these arguments, that Marx's 'Definition II' should be rejected.
4.3 CONCLUSION
The analysis in this chapter has argued that Marx's second definition of
productive and unproductive labour ought to be rejected and that his
first definition ought to be retained. It has also argued that the first
definition ought to be modified to allow for the possibility that certain
forms of state expenditure (for example, expenditure on infrastructure,
education, health, social welfare, and so on) may have a positive effect
on capital accumulation. The value of this analysis in the present context

is that it provides a basis for evaluating how this distinction is used in the
writings of Jean Lojkine and thereby assessing his claim that certain
forms of state expenditure (which he refers to as the 'collective means of
consumption') are unproductive (see section 7.2.3).
Part II
French Marxist Urban

Sociology: The Case of
Manuel Castells and Jean
Lojkine
5 The Different Marxist
Traditions of Manuel
Castells and Jean Lojkine:
Althusserianism and State
Monopoly Capitalism
5.0 INTRODUCTION
The works of Manuel Castells and Jean Lojkine are written within
different Marxist traditions and it is necessary to understand these
traditions in order to assess their work. Castells's work - at least that
written prior to 1983 (see Castells, 1983a, pp. 289-336) - is located
within the Althusserian framework, while that of Lojkine is located
within the framework of state monopoly capitalism (although in one
place Castells attempts to combine elements of both frameworks - see
Castells and Godard, 1974). Each of these frameworks will be briefly
outlined as they impinge upon the work of Castells and Lojkine
respectively.
5.1 THE ALTHUSSERIAN FRAMEWORK
The A1thusserian school of Marxism is the name usually given to the

works of Althusser (1969; 1971; Althusser and Balibar, 1970), Balibar
(Althusser and Balibar, 1970) and Poulantzas (1972,1973,1975, 1976).
Castells, in numerous places, openly acknowledges his debt to the
writings of this school (Castells, 1969a, p.420; 1976a, p.43; 1976c,
p.149; 1977a, pp. ix, 125). This 'debt' is evident throughout all of
Castells's work but particularly in his concept of the urban system and in
his theories of urban politics, urban planning and urban social
movements (See sections 6.3, 6.4, 6.5 and 6.6 respectively). In those parts
of his work the Althusserian penchant for taxonomies of structures,
elements and sub-elements is particularly evident.
The Althusserian framework can be seen as an attempt to provide an
81
82 French Marxist Urban Sociology
alternative to the crude determinism which is implicit in the Marxian

distinction between the base and the superstructure and the Marxian
claim (as expressed in Marx, 1976) that the history of any society can be
understood solely by reference to changes occurring in the economic
base. l Althusser rejects this 'spatial metaphor of the edifice' (Althusser,
1971, p. 130) by asserting that the superstructure has 'relative auton-
omy' (ibid.). As a result, he claims that every mode of production should
be characterised, not in terms of its base and its superstructure, but in
terms of three relatively autonomous elements: the economic, the
political and the ideological. These elements (or instances, levels or
systems as the terms are interchangeably used) are said to be combined
in a specific way according to each mode of production. It is the 'specific
combination' (Poulantzas, 1973, p. 13) of instances which permits the
various modes of production to be differentiated.
Every mode of production is said to have one instance which is
'dominant' (Poulantzas, 1973, p.14) and it is this instance which
characterises the mode of production in question and determines its
laws. In the capitalist mode of production, the economic system is said to
be dominant. Also, in every mode of production there is a 'determinant'
instance (Poulantzas, 1973, p. 14) and this, it is claimed, is invariably the
economic system. Thus in the capitalist mode of production, the
economic system is both dominant and determinant.
Each of the elements (or instances) of any mode of production (that is,
the economic, the political and the ideological) is said to form a structure
which, in tum, is made up of a number of elements and a number of
relations. According to Balibar, every economic system is a structure
which is made up of three elements and two relations and it is the specific
combination of these elements and relations which differentiate one
economic structure from another (see Althusser and Balibar, 1970,
pp.212ft).
The three elements which make up any economic structure are
labour, means of production, and non-labour, and the relations which
combine these elements are the relations of property (or ownership) and
real appropriation (or control) (see Althusser and Balibar, 1970,
pp. 212-14; Poulantzas, 1973, pp.28-30). In the capitalist economic
system, the relations of property and real appropriation are said to be
such that the capitalist (that is, non-labour) owns labour power and
means of production (relation of property) and also controls the
production process by which these two elements are combined (relation
of real appropriation). Thus, in the capitalist economic system the
relations of property and real appropriation are said to 'overlap and
Marxist Traditions of Castells and Lojkine 83
coincide' (Althusser and Balibar, 1970, p.215) and are described by

Castells as 'homologous' (Castells, 1976a, p. 180).2
At any particular point in history, a society will have a combination of
different modes of production, one of which will be dominant. This
'complex unity' (Poulantzas, 1973, p. 15), in which a number of different
modes of production coexist, one of them being dominant, is called a
'social formation' or a 'conjuncture'.
The analysis of any mode of production thus involves, according to
the Althusserian School, an analysis of the three 'regional' (Poulantzas,
1973, p. 12) structures which comprise it (that is, the economic structure,
the political structure and the ideological structure). In the case of the
capitalist mode of production, only the economic structure has yet been
systematically analysed - by Marx himself. Thus Althusserians see the
analysis of the political and ideological structures of capitalism as a
major priority. They argue that such an analysis must, following the
example of Marx's pioneering economic analysis, be in terms of
structures, that is, in terms of elements and relations. 3 Poulantzas
attempts to develop this type of structural analysis4 in relation to the
political structure of the capitalism. Thus the Althusserian school
simultaneously outlines the various 'regional theories' that need to be
developed as part of the analysis of the capitalism, as well as stipulating
the methodological procedures, gleaned from Marx's economic
analysis, by which these theoretical advances may be achieved (see
Glucksman 1974, p.293).
One additional feature ofthe Althusserian framework may be noted.
This feature concerns the distinction between 'structures' and 'prac-
tices', and is fundamental to the whole Althusserian problematic.
Althusser explains the two concepts in the following way: 'All the levels
of social existence are the sites of distinct practices: economic practice,
political practice, ideological practice, technical practice and scientific
(or theoretical) practice. We think the content of these different practices
by thinking their peculiar structure' (Althusser and Balibar, 1970,
p.58).5
The basic idea here is that there are two levels in reality: the superficial
level (of practices) and the fundamental level (of structures). The
relationship between these two levels is that the underlying reality of
structures determines the more superficial reality of practices (but not
entirely, since practices can vary within given structural limits). This is
what Poulantzas means when he describes a structure as defining 'the
limits of variation' (Poulantzas, 1973, p. 95) of practices. The same idea
is implicit in the following quotation: 'The determination of the
practices by the structure, consists in the production by the structure of

the limits of variation of the class struggle: it is these limits which are the
effect of the structure' (Poulantzas, 1973, p. 95).6
These are some of the basic concepts and terms which make up the
Althusserian framework. Before completing this account, however, it is
necessary to outline briefly Poulantzas's theory of the capitalist state,
since this too is relevant to understanding Castells's work. According to
Poulantzas, the function of the state is to act as 'the factor of cohesion
between the levels of a social formation' (Poulantzas, 1973, p.44). It
does this in two ways: through repression and through ideology. On this
basis, Poulantzas, following Althusser, classifies the institutions of the
state (which are renamed state apparatuses) according to whether they
contribute to cohesion through repression or through ideology. These
two types of state apparatus are referred to as the repressive state
apparatus and the ideological state apparatus respectively (Poulantzas,
1972, p.25; see also Althusser, 1971, pp. 131-40).
The term repressive state apparatus is used to designate those
institutions of the state which, within capitalism, serve the function of
repressing any serious threat to the system of capitalist exploitation and
accumulation. For Poulantzas and Althusser, the repressive state
apparatus contains the following elements: the head of state, the
Government, the administration, the army, the police, the courts and the
prisons (Althusser, 1971, p. 131; Poulantzas, 1972, p. 151). Each of these
institutions is classified as 'repressive' because they can have recourse, if
need be, to the use of violence, in the course of their functioning. For
Marx and Lenin, the state was identified exclusively with the repressive
state apparatus, but Althusser and Poulantzas propose to extend the
definition of the state to include other institutions which also playa role
in the maintenance and cohesion of the capitalist system. These
institutions are referred to as the ideological state apparatus.
The term ideological state apparatus covers a wide gamut of
institutions which include, according to Poulantzas, the church, the
schools, the mass media, the arts, the political parties, the unions ('with
the exception of course, of the revolutionary part of trade union
organizations' (Poulantzas, 1972, p.251)) and the family ('from a
certain point of view' (ibid.)). Despite the plurality, diversity and
'relative autonomy' of these various institutions, both Althusser and
Poulantzas feel justified in classifying them together as part of the state
apparatus since anything which constitutes a 'factor of cohesion'
between the levels of a social formation must, they argue, be part of the
state (Althusser, 1971, p.139; Poulantzas, 1972, pp.251-2).7
The significant point about the apparatuses of the state within

capitalist society is that their 'structure' is such that they serve the
interests of the dominant capitalist class while at the same time being
'relatively autonomous' from this class. Poulantzas argues that the state
serves the interests of the dominant class through a 'power bloc'
(Poulantzas, 1973, p. 230) which unites the various fractions of the
dominant class and ensures that the state is subordinated to their
interests. However, within the power bloc there is one class fraction
which is dominant and this is called the 'hegemonic class or fraction'
(Poulantzas, 1973, p.237). Within advanced capitalist societies, this
hegemonic class fraction is monopoly capital and it unites the various
fractions of the capitalist class and ensures that the state operates in the
interests of the entire capitalist class.
These are the main features of Poulantzas's theory of the state. This
account and the account of the Althusserian framework which preceded
it are designed to provide the conceptual and terminological back-
ground necessary for an understanding of CastelIs's work. This account
has not tried to assess the Althusserian framework since this will be
undertaken, in so far as it is directly relevant, within the specific context
of Castells's work. It is now necessary to examine an alternative Marxist
tradition - the theory of state monopoly capitalism - since this is the
conceptual background to the work of Jean Lojkine.
5.2 THE THEORY OF STATE MONOPOLY CAPITALISM
Lojkine's work is situated within the Marxist tradition known as state

monopoly capitalism (henceforth, SMC).8 This tradition has been the
official doctrine of the European Communist parties for many years,
including the French Communist Party of which Lojkine is a member
(see Pickvance, 1977a, p. 220; see also Lebas, 1983, pp. 14ft). Thus the
theory of SMC considerably predates the more recent Althusserian
approach. Indeed these two approaches to Marxism are in conflict on a
number of important issues as this section reveals.
Four elements of the theory of SMC are highlighted in Lojkine's
work, and together these constitute the main core of the theory. Each of
these four elements will now be briefly outlined.
The first element of the theory is that there are three stages in the
history of capitalist development. The three stages are: classical
competitive capitalism, simple monopoly capitalism and state monopo-
ly capitalism (Lojkine, 1977a, p. 122; see also 1977c, p. 21). In Lojkine's
view the capitalist countries of Western Europe and America have

passed through the first two stages and are now in the third and final
phase of capitalist development. In the case of France the transition
from competitive to monopoly capitalism occurred at the end of the
nineteenth and the beginning of the twentieth century (Lojkine, 1977d,
p.259). The methodological implication of this is that a correct
understanding of any capitalist society must begin from a correct
'periodization' (ibid., p. 120) of the history of that society. Indeed the
failure to respect this methodological point is one of the criticisms which
Lojkine directs at Poulantzas's analysis of the capitalist state which
(implicitly) assumes that 'political domination is identical in capitalist
states. . . whether these states are at the stage of competitive capitalism
... or at the monopoly stage' (1977a, p. 277).
The second element of the theory is the claim that monopolistic firms 9
entirely dominate the economy and the state in the period of SMC. The
dominance of monopolies in this period, according to Lojkine, is
reflected in the fact that they receive a higher rate of profit than non-
monopolistic firms (Lojkine, 1977a, p. 112). In addition, their interests
are given special protection by the state since 'the whole of the
organization of the economy and of the society is under the thumb of the
large monopolies' (ibid., p. 118). The implication ofthis, Lojkine argues,
is that the primary class conflict in SMC is between monopoly groups
and non-monopoly groups (the latter consisting of medium and small
capitalists and landowners and workers). The central Marxian conflict
between capitalists and workers still remains, but it is secondary rather
than primary (see Lojkine, 1972b, p. 143; 1977a, p. 113 and Chapter 5,
pp. 275-311; 1977b, pp. 143-4; 1977c, p. 22; 1977d, pp. 257,260,261;
1978, pp.46, 50). The party political implications of this, as Jessop
points out, is that SMC is 'often associated with the political programme
of an anti-monopoly front' which 'would embrace small and medium
capital as well as the petit bourgeoisie and wage-earning class, and
would attempt to capture the state apparatus for its own use' (Jessop,
1977, pp. 360-1). This element of the theory presents yet another source
of disagreement between Lojkine and Poulantzas. Poulantzas, as has
just been seen (section 5.1), argues that the capitalist class are united in a
'power bloc' while Lojkine argues that they are divided into two opposed
groups: monopolies and non-monopolies (see Lojkine, 1977a,
pp.275-311).
The third element of the theory is that the rate of profit will tend to fall
in the course of capitalist development and this becomes most acute in
the period of SMC. This arises because of the tendency for constant
capital to increase relative to variable capital thereby giving rise to what

Lojkine terms an 'overaccumulation' of constant capital and its
subsequent 'devalorization' (Lojkine, 1977a, pp.96-101, 122). This
characteristic of the theory ofSMC is based directly upon Marx's law of
the falling rate of profit (see section 2.1). It is the basis of the claim that
the chronic crisis of capitalism is deepest in the period of SMC, where
non-monopoly groups are particularly adversely affected (Lojkine,
1977e, pp.116-18).
The fourth element of the theory is that state intervention in the
economy is 'a growing necessity' (Lojkine, 1976a, p. 140). The ultimate
purpose or function of this intervention is 'to maintain the cohesion of
the social formation as a whole' (ibid., p. 139) and it does this by
attempting, as Lojkine variously puts it, 'to regulate' (1977a, p. 178), 'to
attenuate' (ibid.), 'to suppress' (ibid., p. 179) and 'to resolve' (ibid.) the
profitability crisis ofthe capitalist economy. The effect of this interven-
tion, it is claimed, is to regulate the problem in the short term while
exacerbating it in the longer term (Lojkine, 1977a, pp. 180, 353; 1977e,
pp. 116-17). It exacerbates the problem in the long term because it
increases the organic composition of capital (as a result of state
investment in physical and social infrastructure - see section 7.3.2)
which leads, ceteris paribus, to further falls in the rate of profit, and to yet
further interventions by the state. Thus state intervention becomes a
permanent feature of the monopolistic stage of capitalist development;
hence the term state monopoly capitalism.
It is the question of the state and state intervention which represents
the source of greatest disagreement between Lojkine and the Althus-
serian school, particularly Poulantzas (see Pickvance, 1977a,
pp. 219-27).10 Lojkine rejects the Althusserian theory of instances and
hence rejects the notion that the state has 'relative autonomy' from the
economic base (Lojkine, 1975, p.28 fn23; 1977a, pp. 78-9, 275-311;
1977b,pp.141, 142, 147; 1977e,p.113).Inturnherejectsthenotionthat
the state acts in the interests of the entire capitalist class. In his view the
state is 'determined' by the economic base (1977a, p.257) and is an
'active reflection' of it (1977b, pp. 141ft) and hence is an active reflection
of the class struggle. Since the monopolies dominate the other class
fractions in the class struggle, it follows, on Lojkine's premises, that they
also dominate the state to which they are united by a 'unique
mechanism' (1972b; 1977a, p. 116; 1977e, p. 119; Lojkine and Pretecei-
lIe, 1970, p. 80). Thus the state tends to reflect and protect the interests of
the monopolies in the current stage of capitalism.
These are the four characteristics which constitute the core of the
theory ofSMC as seen by Lojkine. This theory plays an important role in

his analysis of collective consumption and urban development (which
are discussed in sections 7.2 and 7.3) and it is in that context that an
assessment of the theory of SMC will be made.
5.3 CONCLUSION
This chapter has briefly outlined two different traditions of Marxism

namely Althusserianism and State Monopoly Capitalism. These two
traditions are the contexts within which the respective works of Castells
and Lojkine are located. No attempt was made to assess these two
traditions, however. This will be done, where relevant, in the course of
assessing the work of Castells and Lojkine. The aim of this chapter was
simply to provide the essential conceptual and terminological back-
ground necessary for a proper understanding of their work. It is now
necessary to undertake a systematic examination of that work.
6 The Marxist Urban
Sociology of Manuel
Castells
6.0 INTRODUCTION
The purpose of this chapter is to assess, systematically, the writings of

Manuel Castells. These writings date from 1967 and, as the bibliography
reveals, they are extremely copious. They represent one of the first
attempts within urban sociology to develop a Marxist analysis of urban
areas. Castells's writings can be classified into three different categories:
those concerned with a critique of conventional urban sociology; those
concerned with formulating an alternative Marxian approach for urban
analysis; and those concerned with empirical research. While this
classification is not watertight it is broadly accurate and will be adopted
in this chapter. Thus, Castells's critique of conventional urban sociology
will be discussed in section 6.1; his formulation of a Marxist approach to
urban analysis in sections 6.2 to 6.6; and his empirical research will be
examined in section 6.7.
This threefold classification of Castells's work corresponds broadly to
the chronological sequence in which it was published. The first stage
represents his early writings in the late 1960s in which he advanced his
critique of conventional approaches to urban sociology. The second
stage involved the formulation of a Marxist approach to the analysis of
urban areas, which was quite a new approach within urban sociology
when it first appeared in the late 1960s and early 1970s. The third stage
represents the application of this new perspective to empirical research
which led to the publication of a large number of research reports
authored and coauthored by Castells throughout the 1970s (see section
6.7.1).
It is necessary to assess the value of Castells's intellectual endeavours,
not only because of their comparative novelty, but also because they
have become influential within the social sciences generally (as the
number of citations to his work reveals - see General Introduction, note
1). The assessment presented in this chapter will focus primarily, though
not exclusively, upon issues in Marxian political economy essentially
89
because this is the ultimate foundation upon which Marxist analysis

rests. Thus the analysis presented in Part I (Chapters 1 to 4) is central to
the arguments presented in this and the next chapter. The crucial
question, therefore, which will be addressed to Castells's work, is
whether it is based upon a valid Marxian political economy and whether
the latter, in turn, is conceptually capable of handling the spatial
dimension necessitated by an urban analysis.
6.1 THE CRITIQUE OF URBAN SOCIOLOGY
6.1.1 Introduction
Castells's critique of urban sociology was first published in French in

1968 (Castells, 1968 - republished in English in Castells, 1976a). It has
subsequently been repeated, extended and translated in a number of
different publications. It was repeated and extended in French in 1969
(Castells, 1969c - republished in French in Castells, 1971 b - republished
in English in Castells, 1976b) again in French in 1972 (Castells, 1972c-
republished in English in Castells, I 977a), and in English in 1975
(Castells, 1975c - republished in English in Castells, 1976d). This
critique attempts to assess the 'scientific relevance' (Castells, 1976a,
p. 33) of urban sociology with a view to identifying those areas of it
which, in Castells's view, have 'produced knowledge' (Castells, 1976b,
p. 84). Castells's critique, which acknowledges that many of his criti-
cisms were made by previous writers, was enthusiastically received by
many commentators and has been variously described as 'brilliant'
(Pickvance, 1978, p. 174), 'rigorous and comprehensive' (Harloe, 1977,
p. 2), 'convincing' (Reynaud, 1974, p. 617), 'systematic' (Garnier, 1974,
p. 124) and 'radical' (Pickvance, 1974a, p.205).
Castells concentrates his critique on the twin topics of urbanism and
urbanisation and these are outlined and discussed below in subsections
6.1.3 and 6.1.4. Prior to that, however, it is necessary to clarify the terms
'science' and 'ideology' which Castells uses in his critique before
considering the more substantial issues involved. This will now be done.
6.1.2 'Science' and 'Ideology' in Urban Sociology
Castells's critique (with the exception of Castells, 1977a, Parts I and II)
begins from the proposition that any science must possess one of two
criteria if it is to be truly scientific. It must have either a 'specific
Marxist Urban Sociology of Castells 91
theoretical [or scientific] object' or a 'specific real object' (Castells,

1976b, p. 60). If it does not possess one of these two criteria then it 'does
not exist as a science' and hence is 'ideological' (ibid.).
On the basis of this syllogistic formulation of the relations between
science and ideology Castells argues that urban sociology is unscientific
- first, because it lacks a 'specific real object' due to 'the absence of any
clear delimitations' of the term 'urban' (ibid., p. 73), and secondly,
because it lacks a 'specific theoretical [or scientific] object', since the
'themes around which urban sociology has historically attempted to
constitute itself as a science' (that is, urbanism, urbanisation and the
ecological system) 'do not possess the characteristics of theoretical
distinctiveness' (ibid.). Castells concludes, therefore, that 'urban
sociology is an ideology' (ibid., p. 60).
He argues further that urban sociology is unscientific because, while
possessing a 'theoretical [or scientific] object', it is both 'non-specific'
and 'non-explicit' (Castells, 1976a, p. 38) and hence, according to his
definition, is unscientific. Castells claims that a close examination of the
main works of the Chicago school, which foreshadowed 'all the
subsequent developments of the discipline' (Castells, 1976b, p.61),
reveals their concern with 'the processes of social disorganization and
individual maladjustment, the persistence of autonomous sub-cultures,
deviant or otherwise, and their resistance to integration' (Castells,
1976a, p. 37). For Castells, these studies in the 'sociology of integration'
(ibid., p.39) are 'perfectly legitimate' (ibid.) though limited from the
point of view of 'understanding urban social life' (ibid.). He concludes
by rejecting urban sociology as unscientific and ideological.
Castells's definition of science in terms of a specific theoretical object
and a specific real object originates in the writings of Louis Althusser
(Althusser and Balibar, 1970; see also Castells, 197Oc; Castells and
de Ipola, 1976 - republished in 1979). Two distinctive features should be
noted about this definition. The first is that the distinction between a
theoretical (or scientific) object and a real object is used by Althusser and
Castells to emphasise the crucial role played by theory in the production
of scientific knowledge: it is the theoretical object which provides the
conceptual map for understanding the real object. This view, according
to Althusser and Castells, contrasts with the empiricist view which
characterises scientific knowledge as a process of abstraction from the
real object.
The second feature of this definition of science is the claim that
Marxism is scientific because Marx's writings constitute a radical
epistemological break with Hegel and the classicals, whose theoretical
object is essentially ideological. For both Althusser and Castells, as

indeed for Marx, the distinction between science and ideology corre-
sponds to the division between Marxist and non-Marxist thought
respectively. It is this particular (and somewhat controversial)) usage
which Castells employs in his critique of conventional urban sociology;
hence his claim that the latter is ideological should be understood in this
peculiarly Marxist sense (see Pickvance, 1974b, pp.203-5). It is now
necessary to consider the more substantive issues involved in Castells's
critique, namely urbanism and urbanisation.
6.1.3 Urbanism
Castells devotes considerable attention to criticising the theory of

urbanism of Louis Wirth (1951), since this theory provides, he believes,
'the essential theoretical basis of urban sociology' (Castells, 1976b, p. 65;
see also Castells, 1976a, pp.36-40; 1977a, Chapter 5, pp.75-85).
Wirth's theory is that the spatial characteristics of urban areas (as
defined by size, density and heterogeneity) determine, to a large degree,
the cultural characteristics of urban dwellers (as defined by secondary
relations, role segmentation, anonymity, isolation, instrumental rela-
tions, the absence of direct social control, the diversity and transcience
of social commitments, the loosening of family ties and individualistic
competition). Unlike the traditional criticisms of Wirth (whch have
focused upon the fact that his description of urban culture was
ethnocentric and temperocentric) Castells's criticism is that the real
determinant of urban culture is not the spatial form of urban areas but
the mode of production in which it is situated. In the case of urbanism, it
is capitalism rather than spatial form which is its ultimate cause.
Castells accuses Wirth of making a 'spurious correlation' (Castells,
1976b, p. 69) between the spatial characteristics of urban areas and the
culture of urban residents since both, in fact, are simultaneously
determined by capitalism. He suggests further that Wirth's treatment of
urban areas, as entities which exist independently of the mode of
production, represents 'too impoverished a vision of sociological theory
to be seriously defended' (Castells, 1976b, p. 68).
Castells's own position, however, contains an ambiguity. On the one
hand he claims that the mode of production is the determinant of the
spatial and cultural characteristics of urban areas (see Castells, 1976a,
pp. 38, 54, 55; 1977a, pp. 81- 2). On the other hand he claims that
industrialisation is also the determinant of these spatial and cultural
characteristics. This ambiguity was first pointed out by Pickvance
(1974b, pp. 212-16) and is illustrated in Figure 6.1 (which is a slight
Marxist Urhan Sociology of Castells 93
Spatial characteristics of
urban areas
(e.g. size, density,
heterogeneity)
Mode of production
or (C)
industrialisation
Cultural characteristics
of urban areas
(e.g. urbanism)
Figure 6.1 The detenninants of urban culture according to Castells and Wirth
modification of Figure 1 in Pickvance, 1974b, p. 213). In terms of Figure

6.1, where the arrows indicate the direction of causality, Wirth's position
corresponds to arrow (C) while Castells's position vacillates between
arrows (A) and (B). (Pickvance, 1974b, pp. 213 -14, in a more generous,
if exegetically more questionable interpretation, characterises Castells's
position in terms of arrows (A) and (C).)
Apart from the ambiguity in Castells's position, his critique of Wirth
is cogent and useful. This is particularly so from the Marxist point of
view since it opens up the possibility of developing a Marxist approach
to the analysis of urban areas.
Castells extends his critique of urbanism to include the sociological
literature on 'suburbanism' as well as the urban analysis of Henri
Lefebvre (1968, 1972, 1973). He rejects all of these as 'urban ideology'
because they treat the city as causally autonomous and as the 'most
important' (Castells, 1977a, p. 87) reality. He criticises the thesis that a
certain suburban culture is associated with a certain (suburban)
'ecological unit' (ibid., p. 101) since this involves the same assumptions
and errors as Wirth's thesis. These assumptions, he states, imply an
'implicit link between an ecological context and a cultural context' (ibid.,
p.104).
The work of Henri Lefebvre is of a different tradition to urbanism and
suburbanism but is nevertheless 'very close to Wirth's thesis concerning
the way social relations are produced' (ibid., p. 90). Although Lefebvre is
of the Marxist tradition, Castells finds his urban analysis objectionable
because it is conducted in terms of the category 'urban' rather than in
terms of modes of production. Thus Lefebvre is criticised for referring to
'urban contradictions' (ibid., p. 86) rather than to contradictions
stemming from the dominant mode of production. The 'whole problem'
(ibid., p. 89) with Lefebvre's analysis, according to Castells, is summar-
ised and crystallised in the term 'urban' which is treated as a

'transhistorical form' (ibid., p. 90), to the complete neglect of the social
structure. As a result, Lefebvre replaces the notion of 'class struggle'
(ibid., p. 91) by 'urban struggle' (ibid.) as the motive force of history (see,
however, Castells, 1983a, p. 15 for a more favourable view of Lefebvre's
work).
The essential point of Castells's critique of urbanism, suburbanism
and Lefebvre is that urban areas are not autonomous entities which exist
independently from the rest of society. They are part of the society in
which they are situated and, from a Marxist perspective, they can only be
properly understood in terms of the dominant mode of production
prevailing in that society. In this sense Castells's critique prepares the
way for the explicit application of a Marxist perspective to the analysis
of urban areas.
6.1.4 Urbanisation
Urbanisation refers, inter alia, to the process by which urban areas grow
and to the resulting spatial structure of those areas. Castells is
particularly critical of two themes that have been part of the conven-
tional approach to the study of urbanisation: the concentric zone model
of Burgess and the concept of 'overurbanization' (see Castells, 1970a;
1970b; 1976a, pp. 40-41; 1977a, Chapter 3, pp. 39-63). His criticism of
both these themes is that they treat urbanisation as if it were a uniform,
unilinear and universal process.
The assumption that urbanisation follows an inevitable and universal
pattern is typified, according to Castells, by the work of Burgess who
depicts all urban growth by a series of concentric zones and explains
them as the outcome of ecological processes: competition, concentra-
tion, centralization, segregation, invasion and succession. Castells's
criticism of Burgess and this 'school' is that it treats 'as a universal
feature a social process which is found only under particular conditions'
(Castells, 1976a, p.40). Castells is also critical of the concept of
'overurbanization' which has been developed for the purpose of
understanding the process of urbanisation in 'dependent' societies (see
Castells, 1970a, pp. 1168-90; 1971a; 1977a, pp. 43-63). His criticism,
which is not original (see Sovani, 1964), is that if urbanisation and
industrialisation have occurred simultaneously in advanced capitalist
society, this does not imply that the pattern must be either universal or
inevitable. Indeed an historical investigation of dependent societies
would reveal why the process of urbanisation is likely to be systematical-
ly different from that which occurred in advanced capitalist societies.

As in his critique of urbanism, Castells's critique of urbanisation
emphasises the necessity of relating urbanisation to the mode of
production in which it is occurring. It is a valid critique and a useful
antidote to those studies which tend to treat urban areas as autonomous
entities unrelated to the society in which they are located (Castells,
1970a, p. 1156).
6.1.5 Conclusion
Castells's critique of urban sociology is significant, not because of its

originality, but because it is tied to the promise of a new Marxian
approach to the analysis of urban areas. Since Marxism has never before
been applied to the analysis of urban questions (apart, possibly, from
Engels's (1958) precedential study of Manchester), this represents a new
departure for both Marxism and urban sociology. It is for this reason
that the value of Castells's work must be judged, not by the perspicacity
of his critique, but by the Marxist alternative which he proposes. It is this
alternative which is examined in the subsequent sections.
6.2 COLLECTIVE CONSUMPTION
6.2.1 Introduction
In 1972 Castells's book, La Question Urbaine, was published in French

(Castells, 1972c) and in 1977 it was translated and published in English
(Castells, 1977a). This is probably Castells's best-known work and has
been described by one reviewer as 'one of the most significant works in
urban sociology ever written' (Pickvance, 1978, p. 173).2 In this book
Castells restates his critique of conventional urban sociology (Castells,
1977a, Chapters I and 2, pp. 7-112) and replaces it by what he terms 'an
adaptation of Marxist concepts to the urban sphere, using a particular
reading of Marx given by the French philosopher Louis Althusser'
(Castells, 1977a, p. ix; see also p. 125). It is this 'adaptation' which will be
examined in this and subsequent sections.
Castells's first step in this 'adaptation' is to select the theme of
'collective consumption' - which he rescues from 'the murk of urban
sociology' (Castells, 1976b, p. 75) - and treat it as the major defining
characteristic of urban areas within capitalism. There are two important
aspects to this part of Castells's analysis which will be discussed in this
section. The first concerns the way in which he arrives at this definition
of an urban area (discussed in subsection 6.2.2) while the second
concerns his analysis of the role and function of collective consumption
in capitalist society (discussed in subsection 6.2.3). Before proceeding to
an examination of these topics, however, it may be useful to first clarify
the meaning of the term collective consumption as it is used by Castells.
The term 'collective consumption' - which is frequently used inter-
changeably with 'collective means of consumption' (Castells, 1978a,
p. 42) and 'means of collective consumption' (Castells, 1977c, p. 64) - is
defined by Castells as follows: 'We class as "collective means of
consumption" expenditure on education and culture, social welfare,
transport, housing and urbanization' (Castells, 1978a, p.42). On the
basis of this definition and from the way in which he uses the term in
various other places (Castells and Godard, 1974, pp.228-9; 1976b,
p.295; 1977a,pp.451,460; 1977b,p.64; 1978a,pp.41-2; 1981b,pp.8,
9; 1983b, p.9) it is reasonable to infer that, for Castells, the term
collective consumption refers to the state provision of the 'indirect
salary' (Castells, 1975a, p. 175). In other words, collective consumption
refers to that part of the real wage which is paid, indirectly (by the state)
to the worker in contrast to that part which is paid directly (by the
employer). Thus the workers' wage contains at least two elements: the
element paid by the employer and the element paid by the state. There
may also be a third element if workers receive goods and services from
voluntary welfare organisations (though Castells does not consider this
possibility). This definition of collective consumption is quite straight-
forward and can be easily handled within the conventional Marxian
framework (as will be seen in subsection 6.2.3).
On the basis of these clarifications, it is now possible to proceed to an
examination of the way in which the concept of collective consumption
is used in Castells's urban sociology.
6.2.2 Defining an Urban Area as an Area of Collective Consumption
Castells claims that an urban area - 'the urban', 'urban space' and
'urban unit' (Castells, 1977a, pp.234-5) being synonyms of same -
should be defined as an area of collective consumption since this
definition would allow the entire urban question to be investigated in a
'theoretical way' (ibid., p. 236). The procedure by which he arrives at this
definition can be summarised in terms of four separate propositions, as
follows:
(i) An urban area should be defined in terms of one of the instances of

the social structure (Castells, 1977a, p.235). Given that Castells
follows the Althusserian definition of social structure (see section
5.1) this translates to the proposition that an urban area should be
defined as either an economic, a political or an ideological entity.
(ii) An urban area, at least within capitalist societies, should be defined
in terms of the economic instance since this is the dominant
instance in the capitalist mode of production (ibid., p.235).
(iii) The economic instance consists of two elements, namely, means of
production and labour power, and an urban area should be defined
in terms ofthe latter, not the former. The reason for this is that 'the
urban seems t:p me to connote directly the process relating to labour
power' (ibid., p. 236) and is an area 'delimited by ajob market and
by the relative unity of its daily life' (ibid.) as measured by 'the map
of commuter flows' (ibid.). Means of production, by contrast, are
said to correspond to the domain of regional problems (ibid.).
(iv) The reproduction oflabour power consists oftwo elements namely,
individual consumption, and collective consumption, and the city
should be defined in terms of the latter (ibid., p. 445). The reason
for this is that collective consumption contains 'most of the realities
connoted by the term urban' (ibid.; see also Castells, 1976h,
pp.180ff; Castells and Godard, 1974, pp. 174, 192).
Castells's arguments are extremely cursory and unconvincing (as indeed

various commentators have already noted - Rarloe, 1979, p. 22; Pahl,
1977, p. 314; 1978; Pickvance, 1978, p. 175). The main difficulty is that
he places too much importance on collective consumption as the
distinctive and defining characteristic of urban areas to the neglect of
other 'urban' processes (for example, production, property develop-
ment, and so on). Moreover, the boundary of an urban area cannot
consistently be defined, as he suggests, in terms of the job market since
the areal size (and hence the boundary) of the job market varies
considerably for workers of different skills (see section 8.7). It would
seem that a more cogent and comprehensive definition of an urban area
would treat it as a built-up area, that is, a spatial concentration of
physical structures consisting of buildings (for example, houses, fac-
tories, offices, shops, and so on) and infrastructure (for example, roads,
water sewerage, electricity, parks, and so on) (see Chapter 8). In turn,
three basic processes are associated with every urban area. The first is the
process by which the physical structures of the urban area (that is, the
buildings and infrastructure) are produced and reproduced. This

process includes, inter alia, the production of collective consumption.
The second is the process by which these physical structures are used for
the production of other commodities. The third is the process by which
these physical structures are used for final consumption. It is these three
processes taken together which constitute an urban area. Collective
consumption is only one item in these processes. Thus the 'urban
question' is much broader in scope than the issue of collective
consumption as Castells's claims.
6.2.3 The Effects of Collective Consumption
It is possible, on the basis of various obiter dicta in Castells's writings, to

identify three different theories by which he attempts to explain the
'strategic role' and 'decisive place' (Castells, 1977a, pp.457, 458) of
collective consumption within the capitalist economy. His first theory is
that collective consumption plays an important role in the 'constant
battle against the tendency toward a lower rate of profit (Castells, 1975a,
p. 176; 1977a, p.461) by 'counteracting' (Castells, 1978a, p. 19) this
tendency. Castells's second theory is that collective consumption serves
the function of alleviating the problem of deficiency of aggregate
demand because 'the economy of advanced capitalist societies rests
more and more on the process of consumption, i.e. the key problems are
located at the level of the realisation of surplus value' (Castells, 1976d,
p.294; 1976h, p. 182; 1977a, p. 63; 1981b, p. 8). The third theory is that
collective consumption is a response to the 'development of the class
struggle and the growing power of the workers' movement that extends
its bargaining power to all areas of societal life' (Castells, 1975a, p. 176;
1976h, p. 184).
These three explanations of the role and function of collective
consumption correspond to the three types of Marxian crisis which were
analysed above (see section 2.2). In other words, Castells links collective
consumption and, by implication, the state, to the major crisis points of
capitalist society. Unfortunately, however, he does not elaborate in
sufficient detail any of these three explanations. It may therefore be
useful to consider some of the major issues which arise in each case.
The first explanation, as has been seen, claims that collective
consumption performs the role and function of 'counteracting' the
tendency for the rate of profit to fall (Castells, 1978a, p. 18; see also
1980a, pp. 51-2,69-70). The tendency for the rate of profit to fall refers
to Marx's famous 'law' which he outlined in the third volume of Capital
(Marx, 1974c, Part III, Chapter 13-15, pp.211-60), and which he

regarded as the major source of crisis in capitalist societies. This 'law' or
tendency was thoroughly examined above (see section 2.1) and it was
seen that it could only occur if the organic composition of capital
increased more rapidly than the rate of surplus value. (It was also seen,
that there is nothing in the nature of capitalism to ensure that this
tendency will necessarily and inevitably occur, thereby refuting the idea
that this is a 'law' as distinct from a possibility.)
It follows from this that if collective consumption is to counteract the
tendency for the rate of profit to fall, it must do so either by reducing the
organic composition of capital or by increasing the rate of surplus value
or by a combination of both. Castells argues that it has the effect of
reducing the organic composition of capital. His argument (which can
only be reconstructed from obiter dicta, since it is nowhere clearly
developed) is that collective consumption counteracts the falling rate of
profit by removing 'unprofitable' (Castells, 1978a, pp.41-2; 1976h,
pp. 184- 5) sectors from capitalist production. Since these are unprofita-
ble because they have a higher than average organic composition of
capitaV it follows that their removal from the capitalist sector has the
net effect of reducing the organic composition of capital in that sector
and hence, ceteris paribus, counteracting the falling rate of profit.
Castells explains the effect of state expenditure and investment as
follows: 'By investing at a loss the general rate of profit of the private
sector holds steady or increases' (Castells, 1978a, p. 19) and concludes:
'social expenditures of the state not only thus favour big capital, but they
are also indispensable to the survival of the system' (ibid.) and are 'one of
the principal cogs of monopoly capitalism' (Castells, 1977a, p. 161).
Castells presents no empirical evidence to support any of these claims,
however. Moreover, the relevant evidence could only be produced after
extensive and complex econometric analyses. His claims are thus based
upon conjecture rather than on fact. 4
Castells's second explanation seems to involve the claim that collec-
tive consumption performs the function of solving the realisation crises
which periodically occur within capitalism. This is possibly the most
conventional view held about the nature of collective consumption in the
sense that it refers (implicitly, in this case) to the basic Keynesian notion
that expenditure by the state can have the effect of increasing the level of
aggregate demand, thereby allowing firms to increase output and
produce to full capacity. It is now part of the conventional wisdom that
Keynesian policies can only be effective in a capitalist economy if that
economy is undergoing what Marx termed 'a realisation crisis' (see
section 2.2). Such a crisis will express itself in the existence of unsold
stocks of finished commodities (beyond what is normal to ensure a
steady flow of same) as well as the underutilisation of the economy's
capacity (that is, every firm will not be using its productive potential to
the full). In such a situation state expenditure can have the effect of
increasing the level of aggregate demand, thereby allowing firms to
realise the profits which are crystallised in their existing stocks of unsold
commodities and to increase their output up to the level of full capacity.
If, however, a particular capitalist economy is already at full capacity
(that is, every firm is producing the maximum possible output), then an
increase in state expenditure on (for example) collective consumption
would simply lead, at least in the short term, to inflation.
This seems to be the unstated argument behind Castells's claim about
the relationship between collective consumption and realisation crises.
In the light of these remarks, therefore, it is possible to outline more
formally the precise empirical conditions which would have to be
satisfied if Castells's claim were to be tested and proven correct. There
are three such conditions:
(a) that there is a realisation crisis in at least one capitalist country
which expresses itself in the existence of unsold stocks of com-
modities and the underutilisation of productive capacity;
(b) that this realisation crisis preceded in time, the introduction of, or a
major increase in, collective consumption;
(c) that the realisation crisis in the country in question was solved by the
introduction of or the increase in, collective consumption.
It is possible that the rapid growth in collective consumption in the
1930s and 1940s operated in this way, though whether this continues to
be the case is an interesting empirical question which Castells does not
discuss.
The third explanation claims that the growth of collective consump-
tion is a response to 'the growing power of the workers' movement'
(Castells, 1975a, p. 176). At the theoretical level, this claim emphasises
the respective bargaining power of workers vis-a-vis capitalists and the
state in the determination of wages. In this sense it effectively and
correctly involves a rejection of Marx's theory of subsistence wages (see
section 2.3). At the empirical level, Castells presents no evidence to
support his claim. The relevant evidence in this case would have to show
that the genesis of collective consumption was the outcome of the
growing bargaining power of workers. It would also have to show that
variations over time and place in the quality and quantity of collective
Marxist Urban Sociology of Castells WI
consumption were a reflection of variations in that bargaining power. In

addition it would also involve establishing if in fact collective consump-
tion increases or decreases real wages, and for which group of workers
(and this would be done by subtracting the amount which each worker
pays in taxes from the amount of collective consumption which each
receives).
These then are the three nascent theories about collective consump-
tion which are alluded to in Castells's work. The brief analysis
undertaken here reveals that there are complex theoretical and empirical
issues beneath each of these theories. It also reveals how unwise it is to
make intemperate claims, such as Castells does, about the role and
function of collective consumption without first considering these
issues.
6.2.4 Conclusion
The theme of collective consumption is central to Castells's work.

However, he attributes an inordinate importance to it by claiming that
urban areas derive their specificity from the organisation of the
reproduction oflabour power and are accordingly structured round the
process of collective consumption. Castells also claims that collective
consumption plays a key role in solving the major crises of capitalism
although this claim is a matter of belief rather than a matter of fact. This
aspect of Castells's analysis is lacking in a solid grounding in Marxist
political economy and its empirical implications. In general, therefore,
this section of Castelis's analysis is not particularly useful from the point
of view of developing a Marxist urban sociology. It is now necessary to
examine his concept of the urban system.
6.3 THE URBAN SYSTEM
The term 'urban system' refers to the set of concepts produced by

Castells for understanding urban areas (see Castells, 1969d, p. 92; 1976b,
p.78; 1976c, p.153; 1977a, pp.237ff; 1981a, pp.385ff; Castells and
Godard, 1974, pp. 192ft). These concepts are the result of his efforts to
'extend in the field of the analysis of space the fundamental concepts of
historical materialism' (1977a, p. 125).
Castells extends the conceptual framework of Marxism, not by
extending the basic concepts of Marxian political economy (as outlined
in Part I) but by elaborating the Althusserian notion of a mode of
production. According to the Althusserian view (see section 5.1), a mode

of production consists of three interrelated elements or systems: the
economic, the political and the ideological. Castells argues that these
three systems need a more detailed 'specification' (1976c, p. 153) if they
are to be of any use in understanding urban areas. He begins this
'specification' with the economic system by arguing that this can be
decomposed into three elements: production (P); consumption (C) and
exchange (E). The specification of the political system involves redefin-
ing it as Administration (A) because it refers to the regulation and
control of the relations between P, C and E (for example, local
government, urban planning agencies, and so on). Finally, the
ideological system becomes the Symbolic element (S).
In summary then, Castells's urban system contains five elements:
Production (P), Consumption (C), Exchange (E), Administration (A)
and Symbolic (S).5
Castells then argues that these five elements of the urban system must
be further broken down into sub-elements, since the former are 'much
too general to be translated into explanatory propositions' (Castells,
1976c, p. 157). The procedure by which this is done is rather obscure:
each sub-element is 'defined by the refraction on it of other elements
(including itself) and/or other instances of the social structure (Castells,
1977a, p. 238). The outcome is that the five elements of the urban system
are broken down into sub-elements as illustrated in Table 6.1.
The taxonomy in Table 6.1 is further elaborated 'by specifying, within
each sub-element, levels and roles' (ibid., p.240). Castells shows, in
Table 6.2, how this might be done in the case of two sub-elements (C 1
and P 3).
This is Castells's concept of the urban system. It is, he insists, 'only a
concept and, as such, has no other use than that of elucidating social
practices and concrete historical situations in order both to understand
them and to discover their laws' (ibid., p. 241). In terms of this criterion
Castells's concept does not seem to achieve the purpose for which it was
intended since it is an elaborate taxonomy apparently without any
heuristic use. The reason for this is that the processes by which the
elements, sub-elements, levels and roles are 'internally' related is as
unclear as is their 'external' relation to the decisions and actions of the
main actors in a capitalist society (that is, capitalists, workers, landow-
ners, and the state). As a consequence Castells's classification of such
phenomena as raw materials, green spaces and historic buildings as,
respectively, P 2, C2 and E6 appears rather pointless and uninformative.
The concept of the urban system is also unsatisfactory because it does
Table 6.1 Castells's concept of the urban system
Elements Sub-elements Example
Production (P) Instruments of work PI Factories

Objects of work P2 Raw materials
Objects of work P3 Industrial environment
Objects of work P4 Administration, Informa-
tion
Consumption ( C) Simple reproduction of Housing and minimal
labour power C] material amenities (drains,
lighting, roads, etc.)
Extended reproduction Green spaces, pollution,
of labour power C2 noise, etc. (environment)
Extended reproduction
of labour power C3 School amenities
Extended reproduction
of labour power C4 Socio-culturai amenities
Exchange (E) Production-+ Consumption E] Commerce and distribution
Consumption -+ Production E2 Commuting (urban trans-
port)
Production -+ Production E3 Goods transport
Consumption-+Consumption E4 Circulation (residential
mobility)
Consumption-+ Ideological E5 Emission of information,
shows, etc.
Production-+ Ideological £6 Historic buildings
Consumption -+ Political £7 Decision-making centres
Production -+ Political E8 Business centres
Administration (A) Global/local
Specific/general
Symbolic (S) Failure to recognise/
recognition/
communication
Effect of legitimation
Source: Castells, 1977a, pp.238-40.
not achieve the other purpose for which Castells intended it, namely an
extension of the Marxian framework to incorporate the analysis of
urban areas. The basis of the Marxian framework resides ultimately in
Marx's political economy, particularly as elaborated in the three
volumes of Capital (Marx, 1974a, 1974b, 1974c) and any elaboration of
the Marxian framework should probably start from, or be directly
Table 6.2 The sub-elements in Castells's urban system
Sub-element Levels Roles

C1 (Housing) Luxury housing Lodger
Social housing, ( +, -) Tenant
Slums, etc. Co-owner
Owner
P3 (Industrial zone) Well-equipped Articulation of industry with the:
Badly-equipped natural environment (water,
space)
Communications (network of
transportation)
Social milieu
technology (industrial interdepen-
dencies)
Source: Castells, 1977a, pp.240-41.
related to, this basis. The Althusserian concept of a mode of production

bears only a tenuous relation to Marxian political economy and
Castells's elaboration of it clearly illustrates the ultimate aridity to which
it can lead. In this sense the failure of the concept of the urban system is
due to the fact that it is not anchored in Marx's political economy.
6.4 URBAN POLITICS
Castells, following the Althusserian school (see section 5.1), makes a

distinction between structures and practices. The relevance of this
distinction in the present context is that the concept of the urban system,
which has just been outlined, is a concept for understanding the
structure, but not the practices, of urban areas. In order to understand
the practices which occur in urban areas it is necessary to analyse the
class struggle and the relationship ofthis struggle to the state. It requires,
in Castells's terminology, an analysis of 'urban politics' (Castells, 1977a,
pp. 243-5,261-75) and this, he argues is 'the heart of the sociological
analysis of the urban question' (ibid., p. 244) .
. Urban politics consists of two different fields: urban planning and
urban social movements. These two topics are the main focus of
Castells's analysis of urban areas, particularly his empirical analysis (see
section 6.7). His interest in these two topics reflects his preoccupation
with the mechanisms by which urban areas in a capitalist society are
maintained (through urban planning) or transformed (through urban

social movements). In this way Castells establishes the link, however
tenuous, between the preceding analysis of the urban system and the
subsequent analysis of urban planning and urban social movements. It is
now necessary to examine Castells's conceptual treatment of these two
topics.
6.5 URBAN PLANNING
6.5.1 Introduction
Castells has written relatively little, at a conceptual level, on the nature

of urban planning, even though it is an important theme throughout his
work and is particularly important in his empirical research (see Castells
et al., 1970, Godard, Castells et al., 1973 and Castells and Godard, 1974
which are discussed in subsections 6.7.3 and 6.7.4 respectively). The
basic concepts of his original article in 1969 (Castells, 1969a -
republished in English in Castells, 1981a) have been repeated on a
number of occasions without any major alteration (see notably Castells,
1976c and 1977a, Chapters 10 and 12 respectively).
Urban planning, according to Castells, refers to state intervention in
an urban area which has been necessitated by the contradictions of the
capitalist mode of production. It is a political response to what he
variously terms 'a contradiction' (1977a, p. 260), a 'dislocation' (ibid.), a
'displacement' (1969a, p. 426) and an 'irregularity' (1977a, p. 427) in the
social structure. Its effects are described in equally varied terms: to
'regulate' (1977a, p. 269), to 'counterbalance' (ibid., p. 425) and to 'go
beyond' (ibid., p.427) the contradictions of the capitalist system.
Castells's definition of urban planning clearly refers to any form of
state intervention in an urban area. Thus it does not differentiate
between, for example, state land use planning and state property
development, even though both are different types of intervention in
terms of their purposes and effects (see sections 8.4 to 8.6). The adoption
of such a broad definition, while not excluding the possibility of finer
distinctions being made between different types of state intervention,
has the disadvantage of not focusing upon the particular features of any.
Castells's conceptual analysis of urban planning can be divided into
two parts: the first concerns the definition of urban actors and the
second concerns the causes of and constraints upon urban planning.
Both of these will now be examined.
Table 6.3 Castells's typology of urban actors
Theoretical structural
combination Example of a 'concrete' actor
O-G-P Large international finns

O-G-C Trade unions
O-L-P Chambers of Commerce and Industry
O-L-C Neighbourhood associations
A-G-P Organs of planning
A-G-C
A-L-P Committees of concerted action or regional expansion
A-L-C Municipal institutions
Source: Castells, 1969a, p.424.

o = organisation; G = global; P = centred on production; C = centred on consump-
tion; L = local; A = authority.
6.5.2 The Typology of Urban Actors

Urban actors, according to Castells, are defined in relation to 'the
different elements and sub-elements of the urban system' (Castells,
I 969a, p. 423). In addition to the elements of the urban system outlined
in section 6.3 - that is, P, C, E, A, and S - Castells adds four other 'sub-
elements' namely, Authority (A), Organisation (0), Local (L) and
Global (G). Without any explanation Castells proceeds to construct a
typology of urban actors. His typology is contained in Table 6.3.
Castells does not explore further the uses to which this taxonomy of
urban actors might be put, stating that they are only 'illustrations'
(Castells, 1969a, p.442, fn. 39) though precisely what they illustrate is
left unclear. In its existing form Castells's taxonomy seems rather useless
and pointless. 6 A simpler approach, and one more directly related to the
methodology of Marx's political economy would define urban actors in
terms of their economic and political interests and would reveal that the
most important urban actors include private property developers, state
property developers, landowners, state land-use planning authorities,
and so on. This approach is adopted in Chapter 8, and seems to be more
effective in revealing some of the processes which take place in urban
areas.
6.5.3 The Causes of and Constraints upon Urban Planning
Urban planning arises, as Castells frequently insists, because there is one

central contradiction within the capitalist system, namely, the contradic-
tion between 'the private control of labour power and means of
production on the one hand, and the collective character of the

reproduction of these two elements on the other' (Castells, 1976c, p. 166;
1977a, p.269; 1981a, p.388). At the urban and regional level, this
contradiction is expressed as a contradiction between 'the increased
profit a firm can obtain by selecting a location in an established
industrial zone within a previously built-up area and the dysfunctional
consequences of generalized behaviour of this kind' (Castells 1976c,
p. 166). Elsewhere Castells restates the latter point: 'each individual firm
profits by pushing towards greater concentration whereas organized
decentralization would bring the greatest overall technical advantages'
(Castells, 1976c, p. 152). In this situation the state must intervene in
order to solve the problems which ensue.
This, in essence, is Castells's explanation of the cause of urban
planning. It is an extremely summary and mechanistic explanation. The
implicit assumption seems to be that the existence of this contradiction is
itself sufficient to trigger off interventions by the state. In reality the
process is much more complex and a fuller conceptualisation of the
process would require an analysis of how this contradiction affects the
interests of different groups of capitalists and workers and how this, in
turn, is transmitted to the state in such a way that it is impelled to
intervene on behalf of one group or another or both. Viewed in this light
there is nothing as automatic about state intervention as Castells's
presentation seems to imply.
However, Castells's analysis is deficient, not just because it fails to
establish the connection between urban planning and (as he sees it) the
central contradiction of capitalism, but also because his analysis of the
latter is erroneous in a number of ways. Three errors may be noted in his
formulation of this contradiction. The first is that it is not necessarily the
case, contrary to Castells's claim, that every firm will obtain increased
profit by locating in an established industrial zone in a previously built-
up area. Indeed Castells's research on industrial location in Paris
showed that there was a tendency for large manufacturing firms to
decentralise from established industrial zones (Castells, 1967; Touraine
et al., 1967, 1968; Castells, 1969b; 1975d, Chapter 2; 1977a, pp. 137-45;
see subsection 6.7.1 for a critical assessment of this research).
Secondly, although there may be dysfunctional consequences from
the location of a firm in a particular area (for example, congestion,
pollution, and so on), there may equally be highly positive consequences
(for example, backward and forward linkages with other firms, and so
on). Since both consequences are equally possible, it is clearly an error to
cite only one of them, as Castells's does, and treat it as the only possible
consequence.
Thirdly, organised decentralisation would not necessarily bring the

greatest over-all technical advantages relative to highly-centralised
locations, as Castells claims. It all depends upon such factors as the level
and efficiency of infrastructural facilities, the type of industrial activity,
the distribution of the population in the various urban areas, and so on.
It is not possible, without considering these factors, to make broad
generalisations about the 'technical' merits of centralisation versus
decentralisation.
These three errors reveal the deficiency of Castells's economic analysis
which is a feature of his entire work. In other words his work is not firmly
grounded in any political economy, Marxist or otherwise.
Castells also examines the constraints under which the state operates
in a capitalist society. In this context he introduces the important
concept of 'limits' (Castells, 1976c, p. 166; 1977a, p. 269; 1981a, p. 389)
to convey the notion that the state (like every other interest in society)
operates within a set of constraints. He cites two limits which constrain
the state. The first is that the state cannot alter the ownership relation,
and the second is that there can be no direct intervention by the state in
production, although interventions in such matters as zoning, financial
incentives and collective consumption are possible (Castells, 1981a).
Castells seems to be rather confused about the nature of these limits
since zoning, for example, involves reduction in the private property
rights of landowners (see section 8.8), which effectively proves that the
ownership relation is not inviolable within capitalism. Moreover, the
ownership relation and production itself are constantly being altered by
the state through regulations relating to the quality, the safety, the sale,
the price, the advertising and the labelling of commodities. These
regulations affect the production process itself, and are effective
reductions in the private property rights of the producers of those
commodities (that is, capitalists). It would thus seem that the limits to
state intervention in a capitalist society are not those identified by
Castells. Nevertheless, there are limits to the actions of the state and, as
will be shown (see section 8.5) they can be listed specifically.
6.5.4 Conclusion
This section has shown that Castells's concept of urban planning does
not provide a useful framework for the analysis of this type of state
intervention. There are three reasons for this. The first is that Castells
treats urban planning as synonymous with all forms of state intervention
so that its particular features (such as zoning development control, state
property development, and so on) are lost from view. Secondly, the
concept of urban actors (as defined by Castells) seems to be useless from
the point of view of understanding urban planning. Thirdly, the analysis
of the causes of, and constraints upon, urban planning was seen to
contain a number of errors at the level of political economy. Despite
these defects, however, Castells's subsequent analysis of urban planning
in Dunkirk (Castells and Godard, 1974) contains a number of very
useful insights on the political and ideological effects of urban planning,
largely because it does not rely upon the type of analysis developed in
this section (see subsection 6.7.6).
6.6 URBAN SOCIAL MOVEMENTS
6.6.1 Introduction
Castells's first and fullest conceptual treatment of urban social

movements was presented in 1970 to the Seventh World Congress of
Sociology at Varna in Bulgaria (Castells, 1976c)7 and the main elements
of his approach have been repeated, and added to, in a number of
subsequent publications (Castells, 1972b, republished in Castells, 1972d,
pp. 6-17; 1977a, Chapter 12; Castells, et al., 1974, republished in 1978).
His concern in these writings is with producing a Marxist conceptual
framework for understanding urban social movements. The analysis in
this section will be primarily concerned with outlining and evaluating
that framework (subsection 6.6.3). However, it is necessary first to
clarify precisely the meaning of the term urban social movement as it is
used by Castells (subsection 6.6.2).
6.6.2 The Definition of an Urban Social Movement
An urban social movement, as Castells uses the term, has two

characteristics. The first is that it must produce a major change in the
capitalist system, and the second is that it must be related to other
movements in the society (Castells, 1976c, p. 155; 1977a, p.263).
As regards the first characteristic, Castells distinguishes an urban
social movement from other types of practices. To this end he constructs
a typology of practices and their corresponding effects. These are
summarised in Table 6.4.
In terms of this typology, urban social movements are the cause of
major change in a capitalist society. Indeed Castells insists that they are
Table 6.4 Castells's typology of practices and their effects
Practices Effect
Regulation Reproduction of the urban system

Reform Change in one element of the system
Maintenance of order Reproduction of another structural instance
through the urban system
Urban social movement Transformation of the structural law of the
urban system
Social movement, with an Challenge to the political system
urban base
Demogogic movement No effect, apart from the practice itself
Source: Castells, 1976c, p. 163; 1977a, p.268.
the 'true source of change and innovation in the city' (l972d, p.14)
because they produce 'a qualitatively new effect on the social structure'
(1976c, p.151). This qualitatively new effect must involve radical
changes in either the capitalist relations of property and real appropria-
tion, or in the system of authority (ibid.).
As regards the second characteristic of an urban social movement
Castells emphasises that 'one of my central hypotheses ... is that there
is no qualitative transformation of the urban structure that is not
produced by an articulation of the urban movements with other
movements, in particular (in our societies) with the working class
movement and with the political class struggle' (l977a, p. 453). Thus, for
Castells, the linking and fusion of different movements is the crucial
catalyst for producing radical change in a capitalist society.
Two issues arise in relation to this definition. The first is that it may be
difficult to apply empirically since the criteria for identifying 'true' and
'radical' change are not made clear. Castells states that radical change
must involve, inter alia, a change in the property relation in a capitalist
society. However, the rights of private property (that is, the property
relation) in a capitalist society can be, and are, changed in a wide variety
of ways through state regulations relating to the use of land, the
pollution of the air, the construction of buildings, and so on, as well as
regulations relating to the price, quality, safety, marketing, and so on, of
commodities. The point is that all of these regulations involve changes in
the property relation within capitalism (because they reduce the rights of
private property) though they may not necessarily be regarded by
Castells or others as 'radical'. In other words, 'true', 'radical' change is
not a clear-cut phenomenon, not even within the Marxist perspective.

The second issue is that Castells is primarily interested in urban social
movements because of his optimistic belief in the possibility of radical
change through the fusion of urban protests with other struggles.
However, it would seem that many of the protest movements which
typically take place in urban areas are unlikely to achieve the status of an
urban social movement in Castells's sense because many of them tend to
be concerned exclusively with purely local issues (for example, the
provision of schools, transport, public lighting, police protection, and so
on). Castells's own research actually confirms this (see subsections 6.7.4,
6.7.5 and 6.7.8) which suggests that his emphasis on urban social
movements, as distinct from protests generally, is somewhat misplaced.
These considerations suggest that Castells's preoccupation with urban
social movements is excessively narrow and is therefore not very useful.
Indeed, as will be seen below (subsections 6.7.4, 6.7.5 and 6.7.8), his
actual researches into the empirical reality of urban protest clearly reveal
that the typical urban protest does not develop into an urban social
movement.
6.6.3 A Framework for the Analysis of Urban Social Movements
Castells's framework for the analysis of urban social movements

consists of the concepts of the urban system and urban actors as well as a
set of methodological rules to guide empirical research on this topic.
Since the former have been thoroughly examined already (see section 6.3
and subsection 6.5.2), the analysis in this section will be concerned above
all with Castells's methodological rules.
It is possible to construct, from Castells's writings (1972d, pp. 42-5;
1976c, pp. 163-5; 1977a, pp. 268-9), six different methodological rules
which should guide empirical research on urban social movements.
Although these rules are not listed as such in Castells's work, each of
them is nevertheless present in his writing, albeit in a less systematic form
that that presented here.
The first methodological rule is to identify what Castells variously
terms the 'problems', 'stakes' or 'issues' involved (1972d, p.42; 1976c,
p. 173). The issues, he argues, must be identified and coded in structural
terms. This would seem to mean, given Castells's concept of structure,
that if a particular issue concerns, for example, a claim for better
housing, then this will be classified 'structurally' as C, (see section 6.3).
This methodological rule is cryptically crystallised in Castells's advice
to: 'Give it a name' (1976c, p. 163).
The second methodological rule is to identify the characteristics of the

population affected by the issue, that is, 'the social base' (Castells,
1972d, p. 42). The latter must, in turn, be distinguished from that
fraction of the population which mobilises round the issue, that is, 'the
social force' (ibid.).
The third methodological rule is to identify 'the adversary' (ibid.),
which the social force must confront.
The fourth methodological rule is to identify the politico-ideological
orientation of the organisation through which the social base is
mobilised into a social force (ibid.). Castells places considerable
emphasis on the difference between those organisations which are solely
concerned with specific but limited issues (for example, tenants'
organisations) and those which attempt to link specific issues to broader
political issues (for example, a radical political organisation). In
Castells's view, it is only organisations of the latter type which can,
strictly speaking, give rise to an urban social movement.
At a more general level Castells insists that organisations are
methodologically secondary in the analysis of urban social movements.
He states - with the implicit purpose of differentiating his approach
from what Pickvance terms 'the Anglo-Saxon tradition of "participa-
tion" studies' (Pickvance, 1976, p. 198; Castells, 1977a, p. 324) - that the
usual starting point for the study of urban social movements, to wit,
'organizations' is the 'wrong one' (Castells, 1976c, p.171). This is
because 'the genesis of an organization does not form part of the analysis
of social movements, for only its effects are important' (Castells, 1976c,
pp. 169-70). The correct approach, he argues, 'must start by identifying
the contradictions (,problems') or drawing attention to the mobiliza-
tions specific to those problems' (ibid., p.171).
The fifth methodological rule is to identify the strategy or 'type of
action' (Castells, 1972d, p.43) by which the organisation pursues its
claim. These strategies can be classified into legal actions (for example,
petitions, demonstrations, delegations, and so on) and illegal actions.
Any analysis should examine the relative success of each of these means
in attaining their end.
The sixth and final methodological rule is that the 'effects' of the
movement must be analysed (ibid.). Castells distinguishes three different
situations: those where there are no effects; those where there are 'urban
effects' (that is, the effect is limited to one specific issue); and those where
there are 'political effects' (that is, the effect alters the relationship of the
successful group vis-a-vis other groups and the state). For Castells, a
truly urban social movement will produce both urban effects as well as
political effects.
This is the essence of Castells's approach to the study of urban social
movements. Apart from the taxonomic character of the concepts of the
urban system and urban actors which are unlikely to be of any use in the
understanding of urban social movements, Castells's methodological
rules for empirical research are, by contrast, likely to prove useful. The
above outline of Castells's framework, in the form of methodological
rules, has the advantage of showing how that framework might be
applied in empirical research.
There is, however, one important element missing from Castells's
conceptual analysis, namely, the spatial dimension. Many protest
movements in urban areas are locality-based and arise because the
residents in an area are all equally affected by a particular issue (for
example, inadequate public transport, crime, redevelopment, and so
on). In other words the spatial sphere of influence of many issues is likely
to be limited to specific geographical areas so that only the inhabitants of
that area are directly and immediately affected by it. This fact (which is
conceptualised more formally in section 8.8) is a crucial element in
explaining the size of the social base affected by a particular issue and,
depending upon the level of mobilisation, the resulting social force. In
addition it may explain why many protest movements in urban areas are
issue-specific and are unconcerned with broader political issues. It
would thus seem that a more rigorous analysis of the spatial dimension
would considerably strengthen and deepen Castells's analysis. In
addition, it would offer the possibility of developing a more explicitly
spatial dimension to Marxian political economy, a step which has not
yet been undertaken.
6.6.4 Conclusion
The analysis in this section has shown that while Castells's definition of
an urban social movement is seriously deficient, his methodological
rules for the study of the latter have some merit. Castells's analysis
could, however, be improved by introducing a spatial dimension and by
incorporating this within the framework of Marxian political economy.
As is demonstrated in Chapter 8, this offers a real possibility for
developing a Marxist analysis of urban protest. It is now necessary to
examine how Castells applies this framework in empirical research.
6.7 EMPIRICAL RESEARCH
6.7.1 Introduction
Castells has undertaken at least 18 different pieces of research, usually in

collaboration with other researchers and research assistants. These are
listed below in approximate chronological order, according to their date
of original publication:
1. Industrial location in the Paris region (Castells, 1967; Touraine, et
al., 1967, 1968; Castells, 1969b; 1975d, Chapter 2; 1977a,
pp. 137 -45).
2. New towns in Britain (Castells, 1969a, pp.429-32; 1977a,
pp.277-83).
3. Urban renewal and other urban problems in the United States
(Castells, 1969a, pp.433-6; 1970b; 1976e; 1976f; 1977a,
pp.379-420).
4. Urban renewal in Paris (Castells, et al., 1970; Castells, 1972a;
Godard, et al., 1973; Castells, 1977a, pp.304-22; 1978a,
pp.93-125).
5. Retirement and old age in France (Castells and Guillemard, 1971).
6. Protests against urban renewal in Paris (Castells, 1972c,
pp.406-30; 1972d, pp. 19-45; 1977a, pp. 326-48);
7. Protests in Montreal (Castells, 1972c, pp.431-44; 1972d,
pp.49-67; 1977a, pp. 348-60);
8. Shanty town dwellers in Santiago, Chile (Castells, 1972c,
pp.444-71; 1972d, pp.91-116; 1977a, p.360-75).
9. Urban development in Dunkirk (Castells and Godard, 1974;
Castells, 1975d, Chapter 3, pp. 141-97; 1975e; 1977c, pp.66-78;
1978a, Chapter 4, pp.62-91).
10. Urban protests in Paris (Castells, et al., 1974, 1978; Castells, 1983a,
Part 2, pp. 73-96).
11. Immigrant workers in Western Europe (Castells, 1975b).
12. Urban protests in Paris (Castells, 1977b, 1978a, Chapter 6,
pp.126-51).
13. The service sector in the US (Castells, 1976g).

14. Shanty town dwellers in Monterrey, Mexico (Castells, 1977d).
15. Neighbourhood groups in Madrid, Spain (Castells, 1978b, 1983a,
Part 5, Chs 21-7, pp.213-88).
16. The economic crisis in the US (Castells, 1980a).
17. Latin and gay communities in San Francisco (Castells, 1983a,
Section III).
18. Squatter movements in Peru, Mexico and Chile (Castells, 1983a,
Section IV).
These research publications vary, not only in terms of their theme and
its location, but also in terms of their size and quality. Some are major
pieces of research based upon a systematic collection and analysis of
data (for example, those listed as 6, 7, 8, 12, 14 and 15). Some are reviews
of concepts and data which already exist in the literature (for example,
those listed as 2, 3, and 13), while others are on topics quite unrelated to
urban questions (for example,S). Faced with this diversity it has been
necessary, for the purpose of this book, to select and assess those
research reports which simultaneously represent Castells's major re-
search endeavours, as well as the range of urban themes that are covered
in his work. On this basis seven different pieces of research have been
selected (those listed as 1,4,6,7,9,11 and 12) and will be assessed below.
Each will be discussed in their approximate order of original publica-
tion. The strategy will be to assess the concepts, data, methods and
results of each piece of research in order to discern if it contributes to a
Marxist analysis of urban issues.
6.7.2 Research on Industrial Location in the Paris region
This research was carried out by four researchers (including Castells)

and the results were published in preliminary form in 1967 (Touraine, et
al., 1967). They were also submitted for Castells's doctoral thesis
(Doctorat de 3 erne Cycle en Sociologie: 1967). They were published in
final form in 1968 (Touraine, et al., 1968) and have been published
subsequently, at varying lengths, on three occasions (Castells, 1969b;
1975d, Chapter 2, pp.49-139; 1977a, pp.137-45).
The research involved an examination of 940 industrial establish-
ments which were located in the Paris region between January 1962 and
July 1963 (Touraine, et aI., 1967, p. 375; Castells, 1977a, p. 137). The
basic purpose of this research was simple: to show that different types of
industrial establishment were located in different parts of the Paris
region. The methodological procedure adopted for this purpose in-
volved classifying industrial establishments into nine different types,
and classifying the communes of Paris into three different types of area
and examining the frequency distribution of each type of industrial
establishment in each type of area. (Touraine, et al. (1967, p. 376) refer
to these industrial establishments as the independent variable, and the
areas of location as the dependent variable, although this is evidently a
misuse of terms since there is no causal connection between these two
variables.)
The criterion according to which industrial establishments are
classified into different types is autonomy: 'technical' autonomy and
'economic' autonomy (Touraine, et al., 1967, pp. 373ft). According to
Castells there are three types of 'technical' autonomy:
Type A refers to industrial establishments which have low technical

autonomy since they are small family firms producing a specialised
product and are 'entirely subordinated, from the technological point of
view, to other industrial sectors; for example general engineering
companies' (Castells, 1977a, p. 138).
Type B refers to industrial establishments which are highly mechanised
and which mass-produce a highly standardised product (for example,
motor cars).
Type C refers to industrial establishments which are highly autonomous
and are concerned with the creation of new products (for example,
electronics).
Similarly there are three types of 'economic' autonomy:
Type I refers to industrial establishments which have low economic

autonomy because they are dependent upon a specific market.
Type 2 refers to industrial establishments whose means of production
have a fixed location.
Type 3 refers to industrial establishments which have high economic
autonomy since they are 'without spatial constraint from the point of
view of their functioning' (Castells, 1977a, p. 138).
Castells proceeds from this to construct a typology of nine different

types of industrial establishments according to their autonomy. These
are illustrated in Table 6.5 where each of the symbols in the table have
already been defined.
Table 6.5 Castells's typology of industrial

establishments
Al A2 A3
BI B2 B3
CI C2 C3
Source: Castells, 1977a, p. 138.
This typology of industrial establishments is rather unsatisfactory

since the concept of 'autonomy' is ill-defined, as is the distinction
between 'technical' and 'economic' autonomy. As a result the precise
meaning of each type remains unclear.
The next stage of the research was to classify all the communes in the
Paris region into different types (Touraine, et al., 1967, p. 386). These
types are:
Type cr. which refers to areas of high density, close to the centre of the
Paris agglomeration.
Type J3 which refers to areas where there is a strong concentration of
transportation networks.
Type 'Y which refers to areas of high social status.
(Touraine, et aI., 1967, pp. 384-5; Castells, 1977a; pp. 139-40).
The basic hypothesis of the study was that industrial establishments AI,
B3 and C3 would tend to locate in areas cr., J3 and'Y respectively. A number
of other subsidiary hypotheses were also put forward for each of the
other types of industrial establishments (see Touraine, et al., 1967,
pp. 384-5; Castells, 1977a, pp. 140-41). The basic rationale for these
hypotheses was that firms which were highly dependent upon a
centralised market (AI) would tend to locate in the centre (cr.), while
firms which were concerned with access to locationally fixed inputs
(B2) would tend to locate in areas with a strong concentration of
transportation networks (/3) and, finally, firms concerned with research
and development (C3) would tend to locate in prestigious areas (y) (for
example, the southern and western suburbs). The date showing the
frequency distributions of each type of industrial establishments in each
type of area is broadly consistent with these hypotheses (see Touraine et
al., 1967, pp. 394-402; Castells, 1977a, p.43). The same result is also
revealed more simply and straightforwardly by the maps showing the
location of the different types of industrial establishments (Touraine et
al., 1967, pp. 367-83, 388-93).
This was Castells's first piece of published research and predates his
subsequent Marxist writings. In the light of the research that has since
been carried out on industrial location in the 1970s and early 1980s (see
section 8.7), Castells's failure to distinguish between industrial establish-
ments which are branch plants of multiplant firms and those which are
uniplant firms now appears as a striking and serious limitation. The
research is now of little theoretical or practical interest essentially
because the relationship between different types of firms and their
corresponding location behaviour is poorly formulated (possibly be-
cause the form of the research was more influenced by data availability
than by theoretical considerations).
6.7.3 Research on Urban Renewal in Paris
A preliminary report of this research was published in 1970 (Castells, et

ai., 1970) and the final report was published in 1973 (Godard, et al.,
1973). In addition, a summary of the results has been published in three
different places (Castells, 1972a; I 977a, pp.304-322; 1978a,
p. 93 -125). The results have thus been exhaustively reported.
The research was carried out in 1970 by a team of six researchers who
were, according to Castells, 'under our direction' (1972a, p.95). The
research, which took six months (Castells, 1974, p. 241), attempted to
assess the effects of urban renewal in Paris from 1954 to 1970. The basic
research procedure was to establish why certain areas in Paris had been
selected for renewal projects and what were the social consequences of
those projects. Thus the research involved an examination of the pre-
renewal and the post-renewal characteristics of the areas in which
renewal projects were located.
Castells attempted to measure the pre-renewal characteristics of the
renewal areas in three ways: first, he computed the 'index of differentia-
tion' for the areas in which the renewal projects were located; secondly,
he correlated a number of variables with the size of the renewal projects
in each area; and thirdly, he compared the areas within which renewal
occurred with Paris as a whole on a number of selected variables. Each of

these measures will now be examined.
Castells devised an 'index of differentiation for urban renewal
(IDUR)' in order to establish the degree of association between the
renewal areas and a number of variables. The formula for the index is:
Value of the variable (%) in
the renewal areas
IDUR -1
Value of the variable (%) in
the whole of the city of Paris
(see Castells, et al., Castells, 1972a, p. 111; Godard, et al., 1973, p. 25;
Castells, 1977a, p. 308).
This index was calculated for 23 of the 30 renewal operations and the
results are presented in Table 6.6.
Table 6.6 The index of differentiation for 23 renewal areas

in Paris on selected variables
Variable Index
Proportion of Algerians in the population + 1.20
Proportion of upper management and liberal professions - 0.57
Dwellings without water +0.36
Proportion of semi-skilled and unskilled workers +0.50
Overcrowding in the dwellings + 0.47
Dwellings without WC +0.31
Proportion of foreigners -0.30
Proportion of persons over 65 - 0.19
Rate of women's activity -0.13
Proportion of craftsmen and tradesmen +0.04
Proportion of persons under 19 +0.02
Rate of activity of the population +0.06
Sources: Godard, et al., 1973, p. 26, Table 6; Castells, 1977a, p. 308, Table 54. (It
is noteworthy that there is considerable discrepancy in the indexes contained
in each of these tables even though both are based upon the same research. It is
assumed that the first table (Godard, et al., 1973, p.26, Table 6) is the correct
one.)
Castells interprets this table as follows: 'They [that is, the renewal
areas] are sectors in which housing has deteriorated, of course, but they
are also sectors strongly marked by the presence of immigrant and
unskilled, and by the absence of the upper strata of the population'

(Castells, 1977a, p. 308). The difficulty with this interpretation is that it
is impossible to assess if it is an accurate account of the various renewal
areas. The reason for this lies in the index of differentiation itself which
could have the same value in two entirely different situations for
example, 80%/40% - 1 = 1; 8%/4% - 1=1). This fact raises serious
doubts about the usefulness of this index as a way of isolating the
distinctive characteristics of the renewal areas. In these circumstances a
more accurate characterisation of the various renewal areas could
probably be gleaned by simply inspecting the actual percentage values of
the various variables in each area and in Paris as a whole. However, this
information, although used by Castells to calculate the index of
differentiation, does not appear in any published account of this
research.
Castells then correlates each of the above variables with the areal size
of each renewal operation in order to establish more precisely the
distinguishing characteristics of the renewal areas. The results of this
Spearman rank correlation test are tabulated in Table 6.7.
Table 6.7 Correlations between the areal size of renewal operations

and selected variables in Paris
Selected variable Correlation
Proportion of persons over 65 +0.28

Proportion of semi-skilled and unskilled workers +0.18
Proportion of foreigners plus active women +0.14
Proportion of Algerians in the population +0.20
Proportion of craftsmen and tradesmen +0.09
Proportion of upper management and liberal professions +0.04
Overcrowding in the dwellings -0.09
Rate of activity of the population -0.05
Dwellings without water -0.27
Dwellings without WC -0.06
Proportion of persons under 19 -0.17
Source: Godard, et ai., 1973, p. 27, Table 7; Castells, 1977a, p. 309, Table 54.
(Once again there are considerable discrepancies between these two tables. It is
assumed that the first table is the correct one.)
These results, according to Castells, indicate that the renewal

operation was 'all the more important in that it concerned old people,
semi-skilled and unskilled workers, foreigners, working women and
Algerians. And, on the other hand, there was a low but inverse relation
with the variables concerning the deterioration of the housing' (Castells,

1977a, p. 309). This interpretation is broadly consistent with the figures,
although two points should be noted: first, all of the correlations are
extremely low, thus indicating a weak association; secondly, the
correlations for some of the variables (notably proportion of persons
over 65, overcrowding in dwellings and dwellings without WC) are the
opposite of that suggested by the index of differentiation. Both of these
points reveal that the evidence is inconclusive.
Castells also examined other characteristics of the renewal areas and
found that the latter tended to be areas of industrial decline and areas
where the majority of the residents tend to be politically left-wing
(Godard, et al., 1973, pp. 28-34).
The research also examined the post-renewal characteristics of the
renewal areas. Because many operations were still in progress while the
research was being carried out, this aspect of the research is incomplete.
However, two findings are of considerable importance. The first relates
to housing, where it was found that renewal frequently resulted in
'eviction of most of the former residents and the occupation of the new
space by social categories of higher status' (Castells, 1977a, p. 313). The
second finding was that 62 per cent of all land under renewal was being
converted to office use (Godard, Castells, et at., 1973, p. 52) thereby
accentuating the industrial decline in those areas. These results have in
fact been confirmed by more recent investigations of urban renewal in
Paris (see notably Zetter, 1975; Bentham and Moseley, 1980).
The main feature of this research is that it is concerned with the
characteristics of certain areas in Paris, and with the way in which the
characteristics of those areas have changed as a result of urban renewal.
This methodological procedure was adopted in order to reveal what
Castells terms 'the social logic at work in the operation' (1977a, p. 306).
While this methodological procedure is useful it is limited. It is limited
because it does not investigate the actions of each of the actors involved
in the renewal process (developers, planners, landowners, residents, and
so on, since it is ultimately their actions, each pursuing their respective
interests, which determine the outcome of the renewal process. Castells's
analysis would need to proceed to this level to ensure that the 'social
logic' of the renewal process is in fact revealed.
6.7.4 Research on Protests against Urban Renewal in Paris
This research was carried out in 1970 in Paris with the assistance of two
other researchers (F. Lentin and J. Olives) (Castells, 1972d, p. 24, fn 2).
The results were published twice in 1972 (Castells, 1972c, pp. 406-30-
translated to English and republished in 1977a, pp. 326-48, and 1972d,

pp.19-45).
This study is closely linked to Castells's study of urban planning in
Paris (see subsection 6.7.3 above), since the urban protests under
investigation here were a direct response to the urban renewal as
analysed in that study.
One of the areas studied was given the fictional name of 'Cite du
Peuple' and its population was predominantly working class with a high
proportion of immigrants. The area was scheduled for redevelopment
and, as a result of this, a protest action was organised in 1965 by the
national tenants association to protect residents from the threat of
eviction (Castells, 1972d, p. 26). The demand was made, supported by
petitions, meetings, speeches and demonstrations, that a larger propor-
tion ofHLMs (the French equivalent oflocal authority housing) be built
in the redeveloped area. This demand was acceded to in 1966, although
the new HLMs were subsequently allocated to residents from outside the
area and this led to the disintegration of the movement (Castells, 1972d,
p.29; 1977a, pp. 326-30).
Subsequently, a group of architecture students (,strangers to the
district', Castells, 1977a) attempted to mobilise the residents into
opposing evictions by denouncing the redevelopment plan and calling
for revolutionary struggle. This, however, had no effect upon the
residents and the students left (1972d, pp. 29-30).
In another part of the 'Cite du Peuple' - 'a slum largely inhabited by
unskilled workers, immigrants or North African Jews' (Castells, 1972d,
p. 32) - an organisation was formed by young workers and working-
class students from the area, with the aim of linking the demand for
'rehousing in the same district with moderate rents' (ibid., p. 33), with a
general political struggle (that is, the denunciation of the bourgeoisie
(ibid., p.34). Gradually, however, the movement dwindled as fewer
people attended meetings and residents left the area.
In another part of Paris - the 'Midi' district - there was no
mobilisation against the evictions even though this district had a strong
left-wing tradition (Castells, 1972d, p.36). A revolutionary group of
young workers and students set up a local committee, distributed a news
bulletin and held a rent strike. However, the movement finally collapsed
because the revolutionary group had to leave to carry out revolutionary
work elsewhere (ibid., p. 37).
Finally, in the 'Cite d'Aliarte' district a movement, supported by
revolutionary students and a left-wing party, managed to halt, albeit
temporarily, the evictions necessitated by redevelopment. A number of
immigrant workers who were threatened with eviction occupied seven

houses, but they were ultimately dislodged by the police (ibid., p. 39).
This is a brief summary of Castells's research on the resistance to
redevelopment in Paris (as reported in Castells, 1972d, pp. 19-45). The
research is interesting, though highly impressionistic and lacking in
detail on crucial questions such as the chronology of events as they
occurred in the district, the size of the social base affected in each area,
and the size of the social force which emerged from this base. In addition
to these deficiencies, the duration of the research and the research
procedure itself remains obscure. In the introduction Castells states that
his results are based upon 'precise research, personal experience and
sometimes by sharing in the struggle' (ibid., p.16) although the
connection between the former and the latter are not clearly brought
out. In this sense Castells's research would seem to conform neither to
the standard procedures of sociological research, nor indeed to his own
methodological procedures as outlined in subsection 6.6.3. In short, it is
an interesting, but very inadequately executed, piece of research.
6.7.5 Research on Urban Protests in Montreal
This research, carried out between 1969 and 1971, examined what
Castells regards as 'one of the most important urban struggles of recent
years' (l972d, p. 49). The results of the research were published twice in
1972 (1972d, pp. 49-67 and 1972c, pp. 431-44 - translated to English
and republished in 1977a, pp. 348-60). The information upon which the
research is based comes, Castells states, 'from several sources' (l977a,
p. 348, fn 17). These sources are listed as follows:
personal experience, especially in the Mouvement d'Action Politique
Municipale during my stay in Montreal in 1969; the work of Quebec
students in my urban politics seminars at Montreal University in 1969
and at the University of Paris in 1970 and 1971; long discussions with
Evelyne Dumas, author of a series of articles on this theme in the
Montreal Star; continual exchange of information and ideas with
Ginette Treusdell, a Quebec sociologist who has written a study on
this theme; bulletins, tracts, publications and internal reports of
FRAP and Montreal citizens' committees. (Ibid.; see also Castells,
1972d, pp. 66-7).
The movements which Castells examined originated in 1963 in
'citizens' committees' which were set up in various parts of Montreal for
the purpose of solving the problems of unemployment and the lack of
adequate housing and communal facilities. In 1970 these committees

were united under an umbrella organisation, FRAP (Front d'Action
Politique), and were renamed CAPs (Comite d'Action Politique). Castells
examined the viscissitudes of four of these CAPs in four different
districts of Montreal.
The first district was Sainte-Anne, a working-class slum area with 20
per cent unemployment. Castells found that the attempt to establish an
urban protest here was a complete failure because 'a discrepancy soon
became apparent between the "locals" solely concerned with problems
of facilities, and the militant "outsiders" ... 'who wanted to bring
politics into every action' (Castells, 1977a, p. 355). Castells explains this
failure as follows:
on the one hand the social base, lower proletarian, was a favourite
area for charitable organizations and had a very long way to go before
political awareness could emerge from daily oppression; on the other
hand it was partly a consequence of the characteristics ofthat district;
the issues were seen in the context of social work and the balance of
power was never questioned (Castells, 1977a, p. 356).
The second district examined by Castells was Saint-Jacques, which
had similar social characteristics to Sainte-Anne. In this district the CAP
was initially concerned with social problems and formed a co-operative
and clinic for which they gained 'the respect of a large section of the
population' (Castells, 1977a, p. 356). However, the CAP was also, and
increasingly, committed to 'a rigorous Marxist line' (ibid.) which they
attempted to propagate, through militants, by taking 'advantage of the
local base obtained through urban claim actions' (ibid.). As in the
district of Sainte-Anne, there seems to have been no fusion of urban and
political issues in this district, and hence no urban protest developed.
The third district of Montreal which Castells examined was Cotes-
des-Neiges. This district was inhabited largely by students and profes-
sors and was 'an extremely petty-bourgeois district' (ibid., p. 357). The
aim of CAP was to develop an urban protest (through the fusion of
urban and political struggles) by remaining 'close to the local level'
(ibid.) while maintaining a long-term revolutionary view. This approach
resulted in failure because 'the CAP found itself out ofline with its local
base, an upwardly mobile petty-bourgeoisie' (ibid., p. 358).
The fourth and final district investigated by Castells was Hochenlaga-
Maisonneuve. In this district, according to Castells 'there was a
combination of all the elements necessary for the desired unification of
struggle' (ibid. - these elements being the presence of a majority of
unionised workers and a citizens' committee. In 1971 an urban protest

was initiated in order to oppose the realisation of a proposed motorway
plan which threatened to demolish 'thousands of homes' (ibid.). An
organisation was formed which combined 'unions and working-class
organizations' (ibid.). Castells described the campaign which ensued in
the following way: 'Information sheets were circulated, meetings held, a
petition taken from door to door and signed by thousands of people;
mass demonstrations were organized, including one where the demon-
strators drove along the proposed motorway route' (ibid.). The result of
the campaign was nevertheless a 'total failure' (ibid., p. 359) because,
according to Castells, the interests in favour ofthe motorway plan were
'so powerful' (ibid.) and support for the campaign 'dwindled and the
local population lost interest' (ibid., pp. 358-9).
Castells states that this research is 'necessarily summary and
schematic' (ibid., p. 348), presumably because much of it is based upon
second-hand accounts, rather than upon systematic observation, of the
events which he is analysing. In this sense, the value of the research is
necessarily limited by the limitations of the data. Nevertheless, its results
may prove useful, if only in showing that urban social movements, in
Castells's strict sense (see subsection 6.6.2) are rare occurrences indeed.
6.7.6 Research on Urban Development in Dunkirk
Castells's research into the urban development of Dunkirk was under-

taken in collaboration with Francis Godard and Vivian Balanowski.
The results of the research were published in full in 1974 in Monopolvi/le
(Castells and Godard, 1974) and have been subsequently published, in
more abbreviated form on four occasions (Castells, 1975d, Chapter 3,
pp. 141-97; 1975e; 1977c, pp.66-78; 1978a, Chapter 4, pp. 62-91).
The data used in the research was gleaned from a wide range of
sources:
documents (census, research reports, planning documents etc);

the daily press and periodicals;
letters from firms, associations and the administration;
120 interviews;
contacts with key informants;
participant observation;
a forty-minute film;
photographs.
(see Castells and Godard, 1974, p.418).
The authors present no discussion whatever of the methodological issues

which arose in the gathering and coding of this varied collection of data
on Dunkirk. This is a serious omission in a book which regards itself as a
contribution to 'scientific research' (Castells and Godard, 1974, p. 441)
(Lojkine's comment on this book is thus strange when he writes: 'in this
study methodology is the object of a very precise consideration'
(Lojkine, 1981 b, p. 92). In the introductory section entitled 'Theoretical
Problematic and Methodology of the Research' (Castells and Godard,
1974, pp.7-20) Castells and Godard emphasised that the ideas and
concepts of La Question Urbaine must be 'put to the test of observation'
(ibid., p. 14) and that, in turn, these 'observations must be coded and
reinterpreted with the aid of concepts' (ibid., p. 13). Yet, as will be seen,
Castells is often quite careless and cavalier in his handling of the data.
Monopolville is divided into five large chapters. The first two chapters
present the theoretical orientation of the book as well as a considerable
amount of information on Dunkirk and the Nord region. The third
chapter presents an outline of the class structure of the town while the
fourth discusses some of the pressing urban problems which it is
currently experiencing. The fifth and final chapter discusses, inter alia,
the nature of urban planning and urban protest in Dunkirk. Each of
these chapters will now be briefly reviewed and assessed.
The first two chapters (pp. 2l-lO4) present a considerable amount of
information on Dunkirk, as well as a synopsis of the theoretical
orientation of the book. The entire book is stamped by Castells's
theoretical approach which has been systematically analysed here in the
preceding sections (6.2 to 6.6). However, in the first chapter Castells and
Godard add a new dimension to that approach by referring to the state
monopoly capitalism thesis (see section 5.2 for an outline of the main
tenets of this thesis). This is the first time in Castells's work where the
state monopoly capitalism thesis is explicitly mentioned. However, the
Althusserian approach continues to predominate and, for Castells, the
state monopoly capitalism thesis seems to involve no more than
acknowledging the necessity of 'periodization' within capitalism (Cas-
tells and Godard, 1974, pp. 22- 3) and the fact that the current period of
capitalism is dominated by a combination of monopolistic firms and
state intervention. (Castells, like Lojkine, seems to use the term
monopoly as a vague indicator of size rather than as an indicator of
market share. For convenience Castells's and Lojkine's usage will be
retained.)
The essence of Castells's theoretical approach is based upon what he
regards as one of the 'basic contradictions of capitalism' (Castells and
Godard, 1974, p. 23) which is the law of the falling rate of profit. The
consequences of this law are twofold: on the one hand it leads to a
growing concentration of capital in huge monopolistic firms in order to
counteract this fall; on the other it leads to increased intervention and
expenditure by the state in order to make necessary but unprofitable
investments (for example, infrastructure). These two consequences of
the falling rate of profit are said to be clearly visible in Dunkirk which
has recently undergone a rapid transformation, as a result of the joint
actions of the monopolies and the state; hence the sub-title of
Monopolville: L'entreprise, L'etat, l'urbain) (The Firm, the State, the
Urban).
Although M onopolville is a serious attempt to show how the town of
Dunkirk is being moulded and transformed by the combined actions of
capitalist firms and the state, it is not, as its theoretical orientation might
suggest, an attempt to show how the actions of capitalist firms and the
state respond to the law of the falling rate of profit. Indeed the actual
rate of profit enjoyed by the various firms in Dunkirk, although possibly
very difficult to obtain in the case of the branch plants of multi-plant
firms, is one piece of information that is conspicuously missing from
Castells's and Godard's book, as is the effect of state expenditure on
these different rates of profit. This fact confirms again (see subsection
6.2.3) that for Castells, the law of the falling rate of profit is a dogma
rather than an explanation and, moreover a dogma which is erroneous
(see section 2.1). The same remark, as will be seen, applies with equal
force to the work of Lojkine (see notably subsection 7.3).
Dunkirk was chosen as one of the 'new poles of industrial develop-
ment' (Castells and Godard, 1974, p.8) under the aegis of the Fifth
(1966- 70) and Sixth (1971- 5) French economic plans. These plans, like
the economic plans of other EEC countries in this period, placed great
emphasis on the necessity of restructuring and revitalising key manufac-
turing industries in order to make them nationally and internationally
competitive. Dunkirk was selected as a coastal growth pole with this aim
in mind for the propulsive iron, steel and petrochemical industry.
The iron and steel industry is the main activity in Dunkirk and is
owned and controlled by Usinor (a French firm first founded in 1948)
who located there in 1960 (Castells and Godard, 1974, p. 10). The other
French steel giant, Wendel-Sidelor, has moved most of its operations to
the coastal site of Fos-sur-Mer and this site has become a growth pole
analogous in many ways to Dunkirk (see Castells and Godard, 1974,
p. 66).) The petrochemical industry in Dunkirk is not as developed as the
iron and steel industry, but the signs are that the French petroleum
Table 6.8 Some characteristics of Dunkirk and the Nord Region

according to Castells
Characteristic Score
Occupational structure Above average percentage of manual

workers
Education and training 50% of young people (14-17) receive no
skilled training
Salaries Below average
Quality of housing Below average
Female participation in labour Below average
force
Source: Castells and Godard, 1974, pp. 62-3.
company, Total, is on the verge of a major expansion there (see Castells

and Godard, 1974, p. 81). As a result, the economic growth of Dunkirk
is coming to depend increasingly on monopolistic enterprises.
The selection of Dunkirk as a growth pole, according to Castells and
Godard, is due not only to its advantages as a coastal site suitable for
industries of primary transformation (such as iron and steel and
petrochemicals) where imported raw materials can be readily processed
andre-exported (see Castells and Godard, 1974, pp. 43-55, 61). It is also
due to the decline of many traditional industries (notably coal and
textiles) in the populous hinterland of Dunkirk and the Nord region (see
Castells and Godard, 1974, pp. 55-69) and to the fact that Dunkirk and
the Nord region are relatively disadvantaged in terms of a number of
characteristics (see Table 6.8).
One consequence of this pattern of disadvantage is that today
Dunkirk suffers from a shortage of skilled labour which is being met by a
combination of immigration and long-distance commuting (Castells
and Godard, pp. 101-2) with the result that skilled workers are more
than ordinarily privileged in comparison with unskilled workers.
During the past few decades the population of Dunkirk has grown
rapidly, though not as rapidly as was earlier predicted (see Coing, 1977,
p. 184). Its growth rate, throughout the 1960s and 1970s, has been
exceptional by comparison with the growth rates for the total French
population and the total urban population in France (see House, 1978,
pp. 125-40).
In Chapter 3 (and in the latter part of Chapter 2), Castells and Godard
identify the various class fractions and their interests in Dunkirk. This
is one of the most interesting and commendable parts of the entire book,
essentially because it is central to any Marxist analysis. Tables 6.9 and
6.10 give a schematic summary of the main fractions of the capitalist and
working classes and their corresponding interests.
According to Castells, it is the non-monopoly class fractions (notably
those numbered (4a), (5) and (6)) which are most in crisis and he cites as
evidence (albeit without citing the source) the 'fact' that 1000 small and
medium-sized firms disappeared between 1964 and 1968 (see Castells
and Godard, 1974, p.90).
This analysis is one of the most promising parts of Castells's and
Godard's book, though it is not developed as systematically, nor as
exhaustively, as the topic requires. One of the interesting findings to
emerge from this analysis, in view of Castells's references to state
monopoly capitalism, and in view of Lojkine's adherence to this thesis
(see notably section 5.2), is that the central class conflict in Dunkirk does
not seem to be between the monopolies on the one hand and all other
non-monopoly groups (that is, medium and small capital and all
workers) on the other (see Tables 6.9, 6.10).
Three pieces of evidence in particular seems to indicate this:
(a) Certain groups of non-monopoly capitalists (notably subcontrac-
tors, numbered (4a) in Table 6.9) benefit considerably from the
presence of monopolies.
(b) Certain groups of workers (notably skilled workers, numbered (1) in
Table 6.10) also benefit from the presence of monopolies.
(c) All workers benefit from the expansion of monopolistic supermark-
ets through lower prices (see Castells and Godard, 1974, p. 124).
These three cases seem to reveal that an anti-monopoly alliance between
all non-monopoly groups is not an inevitable - nor even a highly
probable - feature of state monopoly capitalism, at least as found in
Dunkirk. (This point, however, is not made by Castells and Godard.)
In addition to identifying the different class fractions, their interests,
and the conflicts between them, Castells and Godard also investigate
the ideology of these different class fractions (Chapter 3, pp. 150- 72).
This analysis is based upon 'various observations' (Castells and Godard,
1974, p. 161) of the daily life of these class fractions and on the 'general
information' (ibid.) which Castells and his collaborators possess of the
Dunkirk agglomeration. On the basis of this evidence Castells identifies
four different ideologies:
(1) The ideology of consumption and the nuclear family.
(2) The ideology of class solidarity and conflict.
Table 6.9 Castells's description ofthe fractions of the capitalist class and
their interests in Dunkirk
Fractions of the capitalist class Interests
(1) The integrated monopoly sector (l) Smooth and efficient system of
for iron and steel. The production production;
of steel, for example, is integrated (b) Reserving space for expansion
with the production of ships, and storage, preferably as close as
boilers, pipes, etc. (see Castells possible to the quays;
and Godard, 1974, pp. 74, 76-9, (c) a rail network within the
85-6, 114-15). industrial-port complex;
(d) no competition for its skilled
labour force
(2) The non-integrated monopoly (a) a deep-water port capable of
sector. The refining of petrol, for taking oil-tankers of 400000 +
example, is currently not tons;
integrated with any other activity (b) storage space close to the quays;
(see Castells and Godard, 1974, (c) Efficient road and rail network
pp. 74, 79-82, 87, 115). for transporting the petrol
throughout France and Europe
(3) Large (monopoly) companies (a) Increase in port traffic;
specialising in port activities - e.g. (b) Increase in the capacity of the
port authority, ship owners, port;
freight owners (see Castells and (c) Efficient road and rail network
Godard, 1974, pp. 75, 82-4, 87) for linking Dunkirk with the rest
of France and Europe
(4) Small and medium enterprises (a) continued development ofthe
specialising in port activities port and the monopolies
(a) subcontractors with strong (b) their existence is threatened by
financial backing (e.g. building the monopolies
companies)
(b) autonomous family
enterprises (Castells and
Godard, 1974, pp. 75, 82-4,
87-8,115)
(5) Agriculture: (a) small farmer Urban expansion is absorbing
(b) large farmer agricultural land and removing a
(c) market gardener number of farmers and market
gardeners off the land. The small
farmer suffers most from the sale of
his land while the larger farmer is
likely to set up an alternative business
either in the building or transport
industry.
(6) Shopkeepers and artisans Their interest is to stay in business but
(particularly in the case of small
shopkeepers) their existence is
threatened by the growth of large
supermarkets.
Source: Castells and Godard (1974).
Table 6.10 Castells's description of the fractions ofthe working class and
their interests in Dunkirk
Fractions o/the working class Interests
(1) Skilled workers in the monopoly In view of the scarcity of skilled

sector (see Castells and Godard, labour these workers are relatively
1974,p.117) privileged and their interests lie in
maintaining these privileges
(2) Workers in small and medium- These workers are in a weak market
sized enterprises (see Castells and position and their interest is in
Godard, 1974,p. 118) preserving their jobs even at the
expense of lower wages
(3) Building workers These tend to be foreign workers
whose interests tend to be quite
different to those of French workers
in Dunkirk
(4) Unskilled workers (see Castells They are in a weak market position
and Godard, 1974, p. 117)
(5) Dockers Their jobs are being threatened by
technological advances, notably
containerisation.
Source: Castells and Godard, 1974.
(3) The ideology of individual artisan autonomy.

(4) The ideology of traditional rural community and extended family
(Castells and Godard, 1974, pp. 162-8).
Each class fraction is said to possess its own particular combination of
these different ideologies and it is this combination which constitutes its
ideology. (Dockworkers, for example, combine ideology (2) with
ideology (4).)
The major difficulty with this aspect of the research is that the reader
has no way of assessing the empirical status and accuracy of this
classification of ideologies and their distribution among the different
class fractions in Dunkirk. This is a difficulty because the empirical basis
of the analysis - that is, the data - is not (contrary to normal scientific
practice) made accessible to the reader. There is, in other words, no
indication of the set of empirical observations that were made which are
so neatly summarised by these four ideologies and which, in turn, are
neatly distributed among the different class fractions. Without such data
there is no way of assessing the validity of this analysis; it must therefore
be regarded with some scepticism.
In Chapter 4 (pp. 173 - 292) Castells and Godard discuss some of the
problems in the 'urban system' which arise in relation to the reproduc-
tion of labour. In this chapter the housing problem is given particular
attention. At the time of writing Castells found a serious housing
shortage in Dunkirk (1974a, pp. 198-212) although the more recent
evidence (see notably Coing, 1977, p. 184) seems to suggest that this has
been reversed and that there is now an 'overproduction' (ibid.) of
houses, to the extent that the 1975 census recorded 6100 houses vacant in
Dunkirk.
Castells and Godard also discuss (pp. 235-59) the transport prob-
lem in Dunkirk and the transport requirements of the various class
fractions and the way in which they are inadequately dealt with in
Dunkirk. This discussion focuses on the demands for improved road
and rail networks and on the inadequacy of the existing, privately-run,
bus service.
The problems of housing and transport are referred to as problems
associated with, respectively, the consumption element and the ex-
change element in the urban system. This type of terminology, derived
from Castells's concept of the urban system, pervades and vitiates
Chapter 4 of M onopolville and is of no real heuristic value. Castells's and
Godard's treatment of the data in this chapter is also rather careless
since the sources of the 'facts' related to housing and transport are rarely
cited.
The third and fourth parts of Chapter 4 (pp. 364-412 and pp. 412-39
respectively) contain an analysis of urban planning and urban protests.
The analysis of urban planning begins by defining the latter as all forms
of state intervention (Castells and Godard, 1974, p.365) though the
subsequent discussion is largely confined to the institutional structures
in France and Dunkirk for urban and regional land use planning.
Castells goes through the history of the production of various planning
documents and it is on the basis of this analysis that he reaches his
conclusion that urban planning serves the function of reconciling,
integrating and regulating the diverse interests in the city within 'the
limits drawn by the structural laws of the dominant mode of production'
(Castells and Godard, 1974, p. 411). Castells and Godard are particularly
emphatic that urban planning never challenges the dominant interests of
the capitalist class (for example, it never challenges, but always
facilitates such monopolies as Usinor).
The analysis of urban planning also highlights, particularly in his later
accounts of this research (1978a, pp. 82-8), the political and ideological
effects of planning. The planning authority, according to Castells, is at
the centre of various conflicting class interests and its role is to produce
plans which demonstrate that the various types of development

envisaged by the plan reflect the 'common interest' rather than the
particular interest of one particular class or class fraction. The visibility
of the contradictions and conflicts between classes is reduced by
presenting plans as statements of 'harmony and agreement of contradic-
tory demands, rubbing out these contradictions by colouring them
differently on a map, or more seriously by separating them spatially
(zoning), or temporally (phased development), thus enabling divergent
or even contradictory social demands to be represented in the same
scheme' (ibid., p. 84).
Viewed in this light, the planner is both a negotiator and a technician.
Castells continues: 'he [sic] uses the prestige of neutrality and technical
competence to present himself as a professional and "scientist" above
conflict, and as such is in a comfortable position to arbitrate between
various social partners' (ibid., p. 85).
On the basis of this analysis, Castells concludes that the major effects
of urban planning are to be found, not in the particular types of property
development which it envisages and approves but in its ideological and
political effects. The ideological effect consists in the 'rationalization-
legitimation' of conflicting interests, while the political effect consists in
'negotiation and mediation' between these conflicting interests (ibid.,
p.86).
This is probably the most convincing part of Castells's analysis of
urban planning in Dunkirk and provides a number of insights which are
likely to prove relevant to understanding urban planning in other
capitalist countries. Indeed, some of the empirical research on urban
planning in Britain is largely confirmatory of Castells's analysis (see
notably Lambert, et al., 1978; Flynn, 1981).
The fifth and final chapter is concerned with 'urban politics' (in the
broad sense) in Dunkirk. This chapter contains five parts. The first two
parts contain a summary of the various political parties in France and
Dunkirk and the various class fractions which these parties tend to
represent, as well as a briefresume of the results of the 1971 municipal
elections where the Socialist Party gained control of the Dunkirk urban
community (Castells and Godard, 1974, pp. 293-363).
Castells and Godard finally discuss urban protests even though
these were particularly scarce in Dunkirk (ibid., p. 414). His analysis is in
fact concerned with explaining the absence of urban protests. Two
reasons are given for this absence. The first is that the urban problems in
Dunkirk (such as the housing problem) inhibit the emergence of urban
protests. For example, those living in HLMs which are of poor quality,
or have excessively high rents, are discouraged from engaging in protest
action since many families are obliged to live in even worse conditions
such as caravans or disused barracks (ibid., p. 429). The second reason is
that the trade unions and the political parties have not offered their
support and organisational capacities to the development of urban
protests over housing, transport, and so on (ibid., pp. 429-30). In fact,
Castells and Godard found that, in Dunkirk, conflicts in the work place
were never related to conflicts over urban problems (ibid., p.423).
The preceding synopsis has highlighted the main features of the study
of Dunkirk as well as its salient merits and defects. It is now necessary to
briefly summarise.
The principal merit of CastelIs's book is that it attempts to analyse the
urban growth and development of Dunkirk by focusing on the different
class fractions participating in that development and the different (and
often competing) interests which each has in relation to that develop-
ment. It is this approach which is at the heart of any Marxist analysis of
the city. Unfortunately, however, Castells and Godard do not develop
this line of analysis to its fullest extent thereby neglecting some of the
more interesting and complex empirical issues (such as, for example, the
way in which the income of the various class fractions is affected, directly
and indirectly, by the central and local state).
The main defect of this book is its carelessness with respect to the
methodological details of doing research. This was seen to be the case
both with respect to the use of 'official' statistics in the book, the exact
source of which was rarely cited, but more particularly with respect to
the data generated by the authors in the course of the research, most of
which, on the basis of the information contained in this book must be
regarded with some scepticism.
6.7.7 Research on Immigrant Workers in Western Europe
This research is an examination of some official statistics on immigra-

tion in Western Europe and was published in 1975 (Castells, 1975b).
Most of the statistics were produced and published by the OECD, the
EEC and the Governments of France, Germany and Britain, and were
collected for Castells with the aid of a research assistant (ibid., p. 33). No
date is given as to when the research was undertaken.
Castells's starting point is that immigration is high in certain West
European countries, as measured by the percentage of foreigners in the
population in 1971: Germany (5.55 per cent), France (6.87 per cent),
Belgium (7.39 per cent), Luxembourg (10.81 per cent), Switzerland
(15.71 per cent), Sweden (5.0 per cent), United Kingdom (3.4 per cent)
(ibid., p. 37). 'We ... want to argue', writes Castells, 'that immigration
is not a conjunctural phenomenon linked to the manpower needs of

expanding economies but a structural tendency characteristic of the
current phase of monopoly capitalism' (ibid., p.44).
The main structural tendency to which Castells refers is the law of the
tendency of the rate of profit to fall (ibid., pp. 44ft), and immigration is
seen as an important device in counteracting this tendency. The
argument is that the fall in the rate of profit which results from a rising
organic composition of capital is counteracted by increasing the
exploitation of immigrant workers. This is done by increasing the rate of
both absolute surplus value (that is, by reducing wages) and relative
surplus value (that is, by increasing the length and intensity of the
working day). This, in turn, has the effect, via competition, of reducing
the wages of indigenous workers. From this Castells concludes that
immigration causes 'a relative lowering of wages, thus contributing to
the structural counter-tendency which helps delay the fall in the rate of
profit' (1975b, p. 49). It should be noted that the only relevant empirical
evidence which Castells presents in favour of this analysis is one table
which shows that, in 1968, in France, the wages of foreign workers were
less than the wages of French workers (ibid., p.47).
The importance of immigrant workers in counteracting the tendency
for the rate of profit to fall arises, Castells claims, because of the fact that
thay are not unionised and united with the indigenous trade union
movement (ibid., p. 53). This, in turn, is due to their 'legal-political
status' (ibid.) as well as 'racism and xenophobia' (ibid.). The absence of
unionisation mkes it possible for capitalists to exploit immigrant
workers at a high rate of exploitation while also making it possible, via
competition, to increase the rate of exploitation of indigenous workers.
The only solution to this, according to Castells is 'the common discovery
by immigrant and indigenous workers of their basic identity of interests'
(ibid., p. 60).
This is the essence of Castells's analysis. It is essentially an attempt to
explain the effects of immigration on wages and profits using Marx's law
of the falling rate of profit (see section 2.1 for a critique of this law; see
also subsection 6.2). However, none of the evidence presented by
Castells is relevant to the proof or disproof of that explanation.
Castells's carelessness with respect to empirical evidence is also
evident in this interpretation of the tables of data which are presented in
the article. His interpretation of nearly half these tables (Tables I, 2, II,
12, 13, 14 and 15) is, strictly speaking, incorrect. Two examples may be
used to illustrate this. Citing Tables 13, 14, 15 and 16, Castells states that
'32% of immigrants in Germany live in temporary dwellings ... while
98% of shanty town dwellers in France are immigrants. (See Tables 13,
14, 15, 16)' (ibid., p.49). In fact, Tables 13, 14 and 15 refer to the
socioeconomic status of employees in Germany, Switzerland and
France, respectively, while Table 16 is non-existent. (In addition, four
diagrams are cited as evidence but are not published in the article.) An
equally striking disparity occurs in the interpretation of Tables 11 and
12, which claims to show that 'the majority [of immigrants] are
unmarried or 'forced bachelors (See Tables 11 and 12)' (Castells, 1975c,
p. 47). In fact, only Table 11 contains any evidence on the marital status
of immigrants, and this shows that the majority of immigrants in
Germany in 1970 were married, while the majority in France in 1968
were single.
In addition to these two defects there is also an internal inconsistency
in Castells's position. One of the fundamental assumptions made by
Marx (see section 2.3), and seemingly adopted by Castells, is that there is
perfect competition in the labour market. On the basis of this
assumption Castells is able to argue that the wages of immigrant
workers have an effect upon the wages of indigenous workers (l975b,
p. 49). However, if there is not perfect competition in the labour market
because immigrant and indigenous workers operate in 'segmented
labour markets' (see Cain, 1976), then the wages of the former would
have no effect upon the wages of the latter. This could occur if immigrant
workers had different types of jobs from indigenous workers (as Castells
believes: 1975b, p.43); if immigrant workers had lower wages than
indigenous workers (as Castells believes: ibid., pp.49, 50, 54); and if
immigrant workers were non-unionised while indigenous workers were
(as Castells believes: ibid., pp. 53, 59, 60). Under these circumstances,
there would not be a competitive labour market, and the wages of
immigrant workers could be expected to have no effect upon the wages
of indigenous workers. While this does not establish empirically that the
labour markets of Western Europe are competitive or segmented, it does
indicate that Castells's position is inconsistent and confused.
This research therefore is of extremely limited value because Castells
does not provide empirical evidence to support his empirical claims and
even when empirical evidence is provided it is frequently interpreted in
an incorrect way.
6.7.8 Research on Urban Protests in Paris
This research is based upon a pilot study of nearly 180 different urban
protests in Paris between 1968 and 1973. The results were first published
in 1977 (Castells, 1977b - translated and published in Castells, 1978a,
Chapter 6, pp. 126-51). The period between 1968 and 1973 was chosen
Table 6.11 CasteIls's typology of urban protests in Paris, 1968-1973
Type ofprotest Example of stake Social base Mobilisation
Economic Housing and urban Working class High

policy
Political Transport Multiclass Medium
Ideological Environment Student Low
Sources: 1977b, p. 50; 1978b, pp. 131-2.
for study because 1968 was 'a significant historical watershed' (Castells,
1978b, p. 129) while 1973 was 'the date of the start of the research'
(ibid.).
Castells explains that the data used in the study was 'obtained by a
systematic analysis of newspapers and weeklies, as well as the pam-
phlets, documents and dossiers of unions and political organizations. In
many cases field work supplemented our information' (ibid.).
The analysis begins with a typology of urban struggle. Castells
distinguishes between economic struggles, political struggles and
ideological struggles in terms of their stake (or issue) and their level of
mobilisation. This typology can be summarised as shown in Table 6.11.
This typology is a rather loose device for differentiating between
different types of protest action, since they are largely interchangeable.
It is possible that a typology which focused upon the spatial dimension
of different struggles (that is, the size of the area and population ·affected
by the issue) might be a more effective device in this case. All struggles
are limited, depending on the issue, to specific geographical areas (for
example, a neighbourhood, an urban area or an entire society) (see
section 8.8). It is quite possible that the types of issues cited by Castells
may vary along this dimension.
There would seem to be three principal results which emerge from
Castells's research.
The first is that a protest is likely to produce a successful economic (or
'urban') effect if it is not related to any political organisation and if it is
not supported by single persons, immigrants or students (Castells,
1978b, p. 137). Castells emphasises the political aspect of this result: 'the
politicization of a struggle does not increase its efficiency at the urban
level' (ibid.). He cites the example of the squatting movement whose
excessive politicisation led to its isolation and ultimate defeat (ibid.,
p.146).
The second result is that a protest is likely to have a successful effect if
it aims to be overtly political and if its social base and social force is
predominantly French and does not include students or immigrants
(ibid., pp. 135-9).
The third result is that a protest is likely to have a successful
ideological effect (that is, challenge the dominant ideology) if it is
concentrated on the housing issue and is supported by a student (rather
than by a family or immigrant) social base (ibid., pp. 139-42). On the
role of students Castells states: 'despite their rather negative effect on the
successful outcome of protest action, [they] reveal themselves to be
ideological agents par excellence' (ibid., p. 140).
Castells emphasises the political implications of these results, par-
ticularly the first one. He states that: 'a protest struggle must be kept
independent of a political struggle in order to obtain positive urban
effects on the one hand and positive political effects on the other' (ibid.,
p. 144). However, while the two types of protest must be kept separate,
they should, he argues, be developed simultaneously. These are the main
themes and results of Castells's research. It is virtually impossible,
however, to evaluate the validity of this research since Castells provides
no information on how the various variables are measured (for example,
the size and characteristics of the social base, the level of mobilisation of
this social base, the effects produced by the resulting social force, and so
on) or on how these variables are intercorrelated. There is no quan-
titative evidence presented to support any of the claims made by Castells
in this research report. Moreover, since nearly 180 different protest
actions were studied, many of which had ceased before the research had
started, it is possible (indeed likely) that accurate and comprehensive
information was not available in every case. Castells provides no
discussion of these issues. For these reasons his research is of extremely
limited value.
6.7.9 Conclusion
Castells's research is an earnest attempt to apply his 'Marxist' urban

theory to actual empirical situations. This research is moderately
successful in some cases (notably in the study of Dunkirk: subsection
6.7) but in virtually every case it is blemished by serious methodological
defects in the treatment of data. Castells's treatment of data is extremely
careless: in some cases because his interpretation of the data is wrong (as
in his study of urban renewal in Paris: subsection 6.7.3), and in his study
of immigrant workers in Western Europe: subsection 6.7.7; in others
because the production of the data itself, particularly in those cases
where the data is based upon 'personal experience' and 'newspaper

reports' is never discussed or explained (as in the study of protests in
Paris: subsections 6.7.4 and 6.7.8; and in Montreal: subsection 6.7.5; in
yet other cases because factual claims are made without the data source
being cited (as in the study of Dunkirk: subsection 6.7.6), and finally
because the full empirical implications and data requirements of his
claims are not fully appreciated (as in the study of Dunkirk: subsection
6.7.6 and the study of immigrant workers in Western Europe: subsection
6.7.7).
These considerations seriously limit the validity of Castelis's empirical
research. In conjunction with the defects identified in his theoretical
analysis, they seriously limit the value of his entire work.
6.8 CONCLUSION
Castells's writing, as has been seen, can be divided into three parts: his
critique of conventional urban sociology; his formulation of a 'Marxist'
urban sociology; and his empirical research. It is now necessary to
summarise the main conclusions to emerge from the assessment of his
work.
Castells's critique of conventional urban sociology was seen to be
incisive, cogent and largely correct. His empirical research, by contrast,
was seen to be blemished by a frequently haphazard use of empirical
evidence. As regards his conceptual framework for the analysis of urban
areas, two major sets of problems were identified. The first is the
inadequacy of his definition of urban areas. This definition, which
isolated collective consumption as the distinguishing and 'structuring'
characteristic of urban areas, was seen to be inadequate since it placed
undue emphasis on collective consumption at the expense of other urban
processes. As is argued more fully in Chapter 8, an urban area is,
fundamentally, a spatial concentration of physical structures, that is,
buildings (for example, houses, factories, offices, shops, and so on) and
infrastructure (for example, roads, water, and sewerage services,
schools, hospitals, parks, and so on). Evidently collective consumption
is only one part of this spatial concentration. Moreover, Castells's
attempt to 'explain' collective consumption in terms of Marxian crisis
theory was seen to be wholly inadequate both conceptually and
empirically.
The second set of problems relates to Castells's analysis of urban
planning and urban social movements. The difficulty with this analysis
lies in its reliance upon seemingly useless taxonomies such as the urban
system and urban actor. Moreover, it would seem that the processes of
urban planning and urban protests are in fact part of more general
processes constituting urban areas. As is argued in Chapter 8, urban
areas consist of three basic processes: the process by which the urban
area is produced; the process by which the urban area is used for further
production; and the process by which the urban area is used for final
consumption. From this point of view, urban planning is simply part of
the process by which urban areas are produced while urban protests (or
most of them) are part of the more general process by which urban areas
are used for final consumption (and the conflicts inherent in that
process).
The more general conclusion of this chapter is that Castells's analysis
is not a significant contribution to the Marxian analysis of urban areas
because it fails in its self-appointed goal 'to extend in the field of the
analysis of space the fundamental concepts of historical materialism'
(Castells, 1977a, p. 125). The ultimate reason for this failure seems to lie
in Castells's adherence to, and elaboration of, the Althusserian fram-
ework. The consequence of this, in turn, is that Castells's analysis is
highly functionalist, in the literal sense that many of his explanations
take the form of identifying the (alleged) functions which particular
actions or activities fulfil (for example, collective consumption or urban
planning), rather than identifying the actual effects of those actions. It is
only by knowing the effects of particular actions that it is possible to
infer their 'functions'. The consequence of Castells's functionalism, in
turn, is that he explicitly (see notably Castells, 1976b, pp. 77-8; 1974b,
p.239; 1981a, pp. 379,380,386) avoids any analysis which would treat
urban processes (such as urban planning or urban social movements) as
the outcome of the conscious and calculated decisions and actions of the
actors in a capitalist society (that is, capitalists, landowners, workers, the
state, and so on). Paradoxically, it is precisely Marx's focus on
individual decisions, each made within a particular set of constraints and
structures, and the collective outcome of those decisions, which is one of
the important methodological features of his political economy.
The need for an analysis of urban areas along Marxist lines thus
remains. An attempt will be made in Chapter 8 to meet that need by
developing an urban sociology which avoids the major errors of
Castells's approach, but which nevertheless retains the Marxian
emphasis upon the role of interest and conflict in social interaction.
However, before proceeding to that task, an assessment will now be
made of Lojkine's attempt to develop a Marxist analysis of urban areas.
7 The Marxist Urban
Sociology of Jean Lojkine
7.0 INTRODUCTION
The purpose of this chapter is to evaluate the writings of Jean Lojkine.

Lojkine's writing, like that of Castells, attempts to apply a Marxist
perspective to the analysis of urban areas. The corpus of his work, which
is much smaller than that of CastelIs (see Bibliography) can be classified
into two categories: his theoretical, Marxist, analysis of urban areas, and
his empirical research. Unlike Castells, whose analysis develops out of a
critique of conventional urban sociology, Lojkine's starting point is
Marxism itself (or that variant of Marxism known as state monopoly
capitalism - see Chapter 5). More specifically, Lojkine's urban analysis
emerges from his analysis of the state and the urban policies of the state.
It is in this sense that he regards the Marxist analysis of the state,
particularly the contemporary French state, as 'the guiding thread' in all
his writing (Lojkine, 1977d; also 1977a, p. 123).
There is no clear chronological sequence between the two parts of
Lojkine's writing (that is, the theoretical analysis and the empirical
research). In fact, his first major piece oftheoretical analysis appeared in
1972 (Lojkine, 1972c - translated and published in English in 1976a), the
same year in which his first major piece of empirical research was also
published (Lojkine, 1972a). Nevertheless, there is a logical separation
between these two parts of his work and this will be maintained in the
presentation below. Thus, his conceptual analysis will be examined in
sections 7.1 to 7.5, followed by an examination of his empirical research
in section 7.6.
The central focus guiding this examination of Lojkine's writing is
Marxian political economy as outlined in Part I. This perspective, while
appropriate to any piece of Marxist writing, is particularly so in the case
of Lojkine whose analysis is very explicitly based upon the assumptions
and propositions of Marx's political economy. As a result most of
Lojkine's work can only be adequately assessed from this perspective.
Thus the crucial issue to be investigated is whether his use of Marxist
political economy is rigorous and cogent, and whether it is adapted to
cope with the spatial dimension which is an essential part of any urban
analysis.
The analysis begins with Lojkine's definition of an urban area.
141
7.1 THE DEFINITION OF AN URBAN AREA
Lojkine defines an urban area as the spatial concentration of the general

conditions of production (Lojkine, 1976a, pp. 120ff; 1977a, pp. 125ft).
He defines the general conditions of production, in turn, as consisting of
the following elements:
the means of production and reproduction;
the collective means of consumption (that is, the totality of medical,
sports, educational, cultural and transport facilities);
the means of social circulation (that is, the commercial and banking
institutions);
the means of material circulation (that is, the means of communication
and transport).
(Lojkine, 1976a, pp. 120-21; 1977a, pp.125-6).
Thus, within every urban area, 'the totality of general conditions of
production, including collective means of consumption, have become
concentrated within a limited area' (Lojkine, 1976a, p. 123).1 Urbanisa-
tion occurs within capitalism, according to Lojkine, because of 'the
tendency for capital to increase the productivity oflabour by socializing
the general conditions of production' (ibid., p. 124). 'The city thus
appeared', he continues, 'as the direct effect of the need to reduce
indirect costs of production, and the costs of circulation and consump-
tion in order to speed up the rotation of capital and thus increase the
period during which capital was used productively' (ibid., p. 127).
These quotations reveal the centrality ofthe concept of production in
Lojkine's definition of urban areas and in his explanation of urbanisa-
tion. While this orientation is consistent with the basic premises of
Marxian theory, the process by which urban areas are used in the
production of commodities is only one of the processes constitutive of
urban/built-up areas. As will be argued more fully in Chapter 8, there
are, in addition, two other processes which are also an integral part of
every urban area. These are the process by which the urban area is itself
produced through property development, as well as the process by which
it enters into final consumption. These two processes are neglected in
Lojkine's work.
In addition to his definition of urban areas, Lojkine also argues that
urban areas within capitalism have two distinctive features. He writes:
'in our opinion what does characterize the capitalist city is on the one
hand the growing concentration of the collective means of consumption
. . .; and on the other hand the particular mode of concentration of the
totality of means of reproduction (of capital and of labour power)'
Marxist Urban Sociology of Lojkine 143
(ibid., pp. 120-21). Of these two characteristics, it is the collective means

of consumption which receives the most attention in Lojkine's work
since the latter, he believes, 'bring out the novelty of the urban mode of
life' (ibid., p. 123) and 'gradually create a style oflife, new social needs,
hence the phrase urban civilization' (ibid., p. 120). It is thus necessary to
examine the concept of the collective means of consumption as it is used
in Lojkine's work. This will be done in the next section.
7.2 THE COLLECTIVE MEANS OF CONSUMPTION
7.2.1 Introduction
Lojkine analyses what he regards as the four most important character-

istics of the collective means of consumption. For him the collective
means of consumption are, first, collective; second, unproductive; third,
unprofitable; and fourth, durable, immobile and indivisible. This
section will analyse the cogency and rigour with which Lojkine analyses
these four characteristics (subsections 7.2 to 7.2.5).
7.2.2 The 'Collective' Nature of the Collective Means of Consumption
Lojkine defines the collective means of consumption as 'the totality of

material supports of the activities devoted to the extended reproduction
of social labour power' (Lojkine, 1976a, p. 121). These are exemplified
by reference to 'the totality of medical, sports, educational, cultural and
public transport facilities' (ibid.). Lojkine refers to these facilities as
'collective' though his use of the latter term is neither clear nor
consistent. In fact it is possible to identify four different usages of the
term in his work.
In his first usage, Lojkine states that the collective means of
consumption are 'collective' because they are provided by the state. This
usage is implicit in his assertion that the 'great importance' (ibid., p. 121)
of the collective means of consumption is evidenced by the proportion of
'expenditure in the budgets of the major capitalist countries' (ibid.)
devoted to the provision of these facilities. This usage is identical to that
of Castells (subsection 6.2.1) and is quite straightforward from a
Marxist perspective.
In his second usage, Lojkine claims that the collective means of
consumption are collective because 'the mode of consumption is
collective and is thus by its nature opposed by individual, private
appropriation' (ibid., p. 122). This is exemplified by stating that 'parks
and lessons cannot be consumed individually - at least not in their

current increasingly socialised form' (ibid.). In this case Lojkine uses the
term collective to refer to those use-values which can be consumed
simultaneously by a number of individuals. The problem with this usage,
however, is that many use-values designated by the term 'collective
means of consumption' are not in fact 'collective' in this sense (for
example, medical services, such as operations and kidney machines,
cannot be consumed simultaneously as can parks and lessons) while
many use-values designated as individual consumption would in fact be
'collective' (for example, all spectator sports and entertainments). Thus
the second usage of the term 'collective' in contrast to the first, fails to
specify what is the quintessential characteristic of the collective means of
consumption. It should therefore be rejected (see Pahl, 1978; Saunders,
1980a, p. 122; Theret, 1982, pp. 349-50).
In his third usage, Lojkine claims that the collective means of
consumption are 'collective' because they satisfy needs which 'can only
be satisfied collectively' (Lojkine, 1977a, p. 136). These needs refer, inter
alia, to the need for health, education and transport. However, this claim
would seem to be false since these needs can be, have been, and in some
cases still are, satisfied individually. There would thus seem to be little
justification for defining the collective means of consumption according
to the particular need which it satisfies.
In this fourth usage Lojkine claims that the collective means of
consumption are 'collective' because this represents the 'best' (ibid.,
p. 136) way of satisfying these social needs. While this may be the case
(though it is debatable) this definition would seem to provide a
justification for 'collective' consumption rather than a criterion for
distinguishing between individual and collective consumption. As a
result it does not grasp the specific features of the collective means of
consumption.
The result of this analysis suggests that Lojkine's first definition of the
collective means of consumption is the most straightforward and clear
cut because it identifies the term 'collective' with state provision.
Lojkine's other usages seem to be imprecise (in the case of the second),
factually incorrect (in the case of the third), or irrelevant (in the case of
the fourth).
7.2.3 The 'Unproductive' Nature of the Collective Means of

Consumption
One of Lojkine's most important claims about the collective means of

consumption is that they are 'unproductive' (Lojkine, 1977a, p. 134). He
claims that the collective means of consumption are essential to the

reproduction oflabour and are a part of variable capital. However, they
have the 'peculiarity' (ibid., p. 132) that they are 'not consumed directly
by individual labour power' and hence are not directly transformed
'from variable capital to wages' (ibid.).
Lojkine distinguishes between two types or fractions of variable
capital: individual consumption (which he identifies with wages), and
collective consumption. He claims that individual consumption is
directly consumed while the collective means of consumption have the
indispensable function of transforming 'the remainder of variable
capital into wages' (ibid.).
Lojkine elaborates by making a distinction between 'objects of
consumption' and 'means of consumption' (ibid., p. 134). The difference
is that the former are use-values which enter final consumption whereas
the latter are a set of activities which make final consumption possible
(ibid.). Thus, in Lojkine's usage, individual consumption comprises
material objects which are consumed individually, while the collective
means of consumption are activities or services which are not consumed
but instead 'make consumption possible' (ibid.).
On this basis Lojkine claims that there is a 'parallelism' (ibid., p. 133)
between the collective means of consumption and the means of social
circulation, that is, 'commercial and banking institutions' (Lojkine,
1976a, p. 120). He claims that the collective means of consumption play
an 'identical social function' (ibid.) in the capitalist system of production
to the means of social circulation. This is because the means of social
circulation perform the function of transforming 'value from the money
form to the commodity form' (ibid.) while the collective means of
consumption perform the function of transforming 'variable capital into
wages and then into means for the reproduction oflabour power' (ibid.).
Lojkine infers that since the means of social circulation are unproduc-
tive, as Marx claimed (Marx, 1974b, Part I), the collective means of
consumption must also be unproductive. Both of these processes,
according to Lojkine, are 'necessary from the social point of view but
totally unproductive' (Lojkine, 1977a, p. 134), The collective means of
consumption are an 'expenditure oflost funds, in so far as they create no
value but, on the contrary, effect a reduction on the value already
created' (ibid., p. 135). Even though they 'increase the productivity of
labour' (ibid., p.137) they nevertheless remain unproductive and
'superfluous' (ibid., p. 135) from the point of view of the reproduction of
capital.
Lojkine's argument is seriously flawed in a number of ways. Five
problems in particular may be identified.
The first problem is his claim that the collective means of consump-
tion are not consumed. This proposition seems to contradict the obvious
fact that the services designated by the term collective means of
consumption (such as health, education, transport, and so on) are
actually consumed. It is thus difficult to see the justification for this
claim.
The second problem lies in the claim that the collective means of
consumption perform the function of 'transforming' a fraction of
variable capital into wages. This appears to be misconceived since
variable capital is simply wages by another name. Correspondingly, the
alleged parallelism between the collective means of consumption and the
means of social circulation does not exist since variable capital does not
have to be transformed into wages unlike commodities which have to be
transformed, through the means of social circulation, into money and
vice versa.
The third problem with Lojkine's argument concerns his central
proposition that the collective means of consumption are 'unproduc-
tive'. This proposition is difficult to assess straightforwardly since, as
has been seen in Chapter 4, Marx used the terms productive and
unproductive labour in two different and inconsistent senses (which, for
convenience, were labelled 'Definition I' and 'Definition II'). In
addition, Marx's Definition II was found to be untenable (see notably
section 4.2). Lojkine does not avert to this ambiguity in Marx's usage
though he does appear to use the term in the sense of 'Definition II' so
that it is open to the difficulties identified above. Nevertheless, it is
possible that the collective means of consumption may be unproductive
according to Marx's Definition I. This, however, is an empirical
question which cannot be established a priori but can only be established
by measuring the net effect of the collective means of consumption upon
the rate of profit and hence upon the rate of capital accumulation. Thus
it remains an open question as to whether the collective means of
consumption are in fact 'unproductive'.
The fourth problem concerns the proposition that the collective
means for consumption 'increase the productivity of labour' (Lojkine,
1977a, p. 137). Two issues arise in this context. The first is whether it is
empirically true that the collective means of consumption increase the
productivity of labour. Lojkine provides no evidence to prove that it
does. The second issue is that if the collective means of consumption
increase the productivity of labour then its effect (depending upon the
costs, and hence the taxes, needed to produce it) may be to increase the
output of the capitalist sector and hence, ceteris paribus, the rate of
profit. If this possibility were shown to exist empirically then the

collective means of consumption would in fact be 'unproductive'
(according to Marx's Definition I). These considerations reveal the
complex issues which are left unexplored in this part of Lojkine's
analysis.
The fifth problem relates to Lojkine's assertion that the collective
means of consumption are 'superfluous from the point of view of the
reproduction of capital' (Lojkine, 1977a, p. 135). The difficulty with this
claim is that the profitability of capitalism is dependent upon, inter alia,
a labour force which has an adequate level of nutrition, health,
education, skill, mobility, and so on. Since one of the effects of the
collective means of consumption is to produce such a labour force, they
would seem to be essential rather than superfluous to the reproduction
of capital. The fact that capitalists attempt to minimise the costs of the
individual and collective reproduction of labour power, as Lojkine
points out (ibid., p. 140), only proves that these are necessary costs; it
does not prove that they are superfluous.
These are the main problems with Lojkine's claim that the collective
means of consumption are unproductive. As has just been seen, it is a
claim which he fails to justify because his understanding of Marx's
political economy seems to be seriously deficient.
7.2.4 The Unprofitable Nature of the Collective Means of Consumption
Lojkine states that the production of the collective means of consump-

tion is unprofitable from the point of view of the individual capitalist
(Lojkine, 1977a, p. 138). He explains this by referring to 'the slowness of
the rotation of nonproductive capital in the consumption sector' (ibid.).
This is exemplified by stating that houses, schools and hospitals 'last for
many years' (ibid.). Lojkine's argument, in short, is that the collective
means of consumption are inherently unprofitable because their produc-
tion requires a capital investment lasting many years.
The main difficulty with this argument is that the production of many
commodities (and not just the collective means of consumption) require
that capital be invested for long periods. Examples of this are the
investments in plant and machinery for the generation of electricity, the
assembly of cars, the refining of oil, the mining of coal, the production of
chemicals, and so on. The production of these commodities is not,
however, normally unprofitable. Thus the period of rotation of capital
in the production of the collective means of consumption does not seem
to be the correct explanation for their lack of profitability.
It would seem that the reason why many of the goods designated by
the term 'collective means of consumption' are unprofitable is that the
level of effective demand for them (that is, those able and willing to pay
for them) is too low. This is why education (and to a lesser extent medical
care) can only be produced profitably if they are confined to a relatively
small number of affluent consumers (and, even then, direct and indirect
state subventions are frequently required to make them profitable).
Thus it is only when these goods and services are produced for
consumers who do not, and cannot, constitute an 'effective demand',
that their production becomes unprofitable. This is typically the reason
why the state in most capitalist countries, is unable to produce the
collective means of consumption for a profit: the latter are normally
distributed, not according to ability to pay but according to the state's
definition of need. Thus the collective means of consumption are
unprofitable, not because of some inherent features involved in their
production but because there is a deficiency in the aggregate demand for
them.
7.2.5 The 'Durable', 'Immobile' and 'Indivisible' Nature of the Collective

Means of Consumption
Lojkine describes the collective means of consumption as use-values

which are simultaneously durable, immobile and indivisible (Lojkine,
1977a, p, 138). Each of these three characteristics is said to be derived
from the fact that these use-values are immaterial services which cannot
(unlike material products) be separated from or exist independently of
their act of production (ibid.). The purpose of the present analysis will be
to clarify the meaning of the terms 'durable', 'immobile' and 'indivisible'
and establish whether they are in fact genuine characteristics of the
collective means of consumption.
The term 'durable', as used by Lojkine, indicates that the consump-
tion of the collective means of consumption has a 'useful effect'
(Lojkine, 1976a, p. 122) which is 'durable' (ibid.). This is elaborated by
affirming that 'their consumption does not involve destruction as in the
case of the consumption of food' (ibid.). Lojkine does not explain the
matter further. The difficulty with his argument in its existing form is
that there is no way of comparing the useful effects of consuming
education with the useful effects of consuming food. The reason for this,
in turn, is that in order to establish a comparison between two things (or
in this case, two effects) it is necessary for both to be reduced to the same
common denominator. But since the effects of education and food
cannot be reduced to a common denominator, any more than can a

quantity of education and a quantity of food, it follows that it is
impossible to establish whether the effects of one are more durable than
the effects of the other. For this reason Lojkine's claim that the collective
means of consumption are more 'durable' than the individual means of
consumption would seem to be impossible to prove.
The term 'immobile' is not explained by Lojkine, though it seems
plausible to infer from the text (Lojkine, 1977a, pp. 138 -9) that it refers
to the fact that the services of tHe collective means of consumption can
only be consumed at the time and place of their actual production. This
is a feature of all services (which contrasts with the 'mobility' of material
products with respect to the time and place of their production) and
hence is not the exclusive peculiarity of the collective means of
consumption.
The term 'indivisible' is not explained by Lojkine either, though it
seems reasonable to assume that it is being used in the normal economic
sense to designate the fact that some commodities have a minimum
economic size (see Bannock, Baxter and Rees, 1972, p. 135). The usual
example of this is certain types of plant and machinery which, for either
engineering or cost reasons, have a minimum economic size. In
designating the use-values of the collective means of consumption as
highly indivisible, Lojkine is thus presumably claiming that such services
as education, health and transport cannot be produced in different sizes
or amounts according to individual requirements. This claim is basically
correct since the equipment needed to produce the collective means of
consumption (that is, buildings, roads, technical equipment, trained
staff, and so on) cannot be produced in discrete units, and hence are
indivisible.
The results of this analysis reveal that the collective means of
consumption are immobile and indivisible, as Lojkine claims. It is
however impossible to prove that they are more durable than other
consumption goods, as he also claims.
7.2.6 Conclusion
The analysis in this section has examined what Lojkine regards as the
four main characteristics of the collective means of consumption. Four
results have emerged. The first is that the term 'collective' (as used in the
expression 'collective means of consumption') is most accurately and
unambiguously identified with 'state provision'. Lojkine's other usages
of the term 'collective' were found to be either imprecise, factually
incorrect, or irrelevant. The second result is that the collective means of
consumption are not 'unproductive' in the sense claimed by Lojkine.

The third result is that the collective means of consumption are
unprofitable because there is a deficiency in the aggregate demand for
these commodities and not, as Lojkine claims, because of the slow
rotation of capital involved in their production. The fourth result is that
the collective means of consumption are immobile and indivisible as
Lojkine claims, but it is impossible to prove that they are more durable
than other consumption goods.
7.3 THE THREE PROBLEMS ASSOCIATED WITH

CAPITALIST URBAN DEVELOPMENT
7.3 Introduction
There are, according to Lojkine, three major 'crisis points' (Lojkine,

1976a, p. 143) or 'contradictions' (l977d, p. 264) associated with urban
development in capitalist societies. These are: the financing of urban
expenditure; the locational strategy of firms; and urban land rent.
Lojkine's analysis of these problems essentially involves applying to
them some traditional concepts in Marx's political economy. The details
of his analysis will now be examined.
7.3.2 The Problem of Financing Urban Expenditure
The first problem associated with urban development in capitalist

societies is, according to Lojkine, the limit to the financing of 'urban
expenditure', that is, the means of communication and the collective
means of consumption (Lojkine, 1976a, p.128; 1977a, p. 160; 1977d,
p. 264). The limit to urban expenditure arises, Lojkine argues, because
the latter increases the organic composition of capital which causes the
rate of profit to fall. This limits the further expansion of urban
expenditure presumably because (though Lojkine does not make this
explicit) the decline in the rate of profit causes a decline in investment
and employment which causes a decline in the amount of revenue
received by the state and this, in turn, leads to a decline in the amount of
the state's urban expenditure.
Lojkine's statement of this argument may be reformulated in terms of
three basic stages. In the first stage he classifies the collective means of
consumption as a type of capital ('expenses capital', Lojkine, 1976a,
p. 130) analogous to constant capital. The second stage of the argument
affirms that both types of capital (that is, constant capital and expenses
capital) have the same effect of increasing the organic composition of
capital. The third stage is then to deduce that, since urban expenditure
has the effect of increasing the organic composition of capital, it also has
the effect of reducing the rate of profit upon which its growth depends.
As a result, the financing of urban expenditure faces an inexorable limit
as a result of the tendency of the rate of profit to fall. Each of these stages
of Lojkine's argument will now be critically examined.
The first step of Lojkine's argument attempts to establish that one
element of urban expenditure (that is, the collective means of consump-
tion) is like 'expenses capital' (that is, the means of social circulation) in
that they produce 'zero extra value' (Lojkine, 1976a, p. 131), while the
other element of urban expenditure (that is, the means of material
circulation) are assumed to be 'highly devalorized' (that is, produce little
extra value). He claims that the collective means of consumption are like
the means of social circulation in being 'a deduction from surplus value
already produced' (ibid., p. 130). His argument is that both perform an
'identical social function' (Lojkine, 1977a, p. 133) within capitalism, in
that the means of social circulation perform the function of transform-
ing 'value from the money form to the commodity form' (ibid.), while
the collective means of consumption, as part of variable capital,
perform the functions of transforming 'variable capital into wages and
then into means for the reproduction of labour power' (ibid.). In other
words, the collective means of consumption and the means of social
circulation are designated as 'expenses capital' since they are 'necessary
to the productive utilization of capital ... but do not transmit or add
any value' (Lojkine, 1976a, p. 130).
The first step of Lojkine's argument is misconceived for at least two
reasons: first because it is based upon a false analogy and second, be-
cause it is inconsistent with the value analysis which Lojkine adopts.
Lojkine's analogy is misconceived because, as has already been seen (sub
section 7.2.3), the collective means of consumption do not transform
variable capital into wages, unlike the means of social circulation which
transform commodities into money and vice versa. Variable capital and
wages are simply different expressions for the same thing.
The second difficulty with Lojkine's argument is that the treatment of
the collective means of consumption as expenses capital is inconsistent
with Marx's value analysis which Lojkine adopts. This is because the
collective means of consumption are a fraction of variable capital which
means precisely that they are not a 'deduction from surplus value
already produced' but an outlay of capital for the purpose of producing
surplus value. In other words, if the collective means of consumption are

a fraction of variable capital, they cannot, analytically speaking, be seen
as a deduction from surplus value already produced since no surplus
value can ever be produced without the prior outlay of variable capital.
A third difficulty is that Lojkine provides no empirical evidence
whatever to show that the collective means of consumption are 'totally
devalorized capital' (Lojkine, 1976a, p. 13) (that is, produce no extra
value) and that the means of material circulation are 'very highly
devalorized capital' (that is, produce little extra value). To establish the
former would require empirical evidence to show that for each
expenditure on the collective means of consumption there is an equal
and corresponding reduction in the mass of profit/surplus value (as well
as in the rate of profit), while to establish the latter would require
evidence to show that for each expenditure on the means of material
circulation there is a correspondingly smaller increase in the mass of
profit/surplus value (through a decline in the rate of profit). None of this
requisite evidence is provided by Lojkine.
The second step ofLojkine's argument affirms that urban expenditure
acts upon the organic composition of capital in the same way as does
constant capital: an increase or decrease in either expenses capital or
constant capital leads, respectively, to an increase or decrease in the
organic composition of capital. Lojkine's explanation is that urban
expenditure 'acts on the organic composition of capital in the same way
as constant capital: it raises the organic composition by increasing the
mass of accumulated social capital without itself being productively
utilized' (Lojkine, 1977a, p. 163).
There are two flaws in this stage of Lojkine's argument. The first
which is identical to that found in Marx (see sub section 2.1 above), is
that an increase in the 'mass' of aggregate (that is, social) capital does not
necessarily imply a corresponding increase in the 'value' of aggregate
capital. This does not follow because the concepts of constant capital
and the organic composition of capital refer to value quantitities and not
physical quantities (that is, masses), so that any increase or decrease in
the mass of the latter (however that might measured) does not imply any
corresponding increase or decrease in their value.
The second flaw is that no evidence is provided to show that the means
of material circulation are 'highly devalorized'. To show this it would be
necessary to demonstrate that the total value of the means of material
circulation causes the organic composition of capital to increase more
rapidly than the rate of exploitation. Without this evidence Lojkine's
argument is seriously weakened.
The third stage of Lojkine's argument affirms that urban expenditure

'strengthens the tendency for the rate of profit to fall' (Lojkine, 1976a,
pp. 131-2), because of its effect upon the organic composition of capital.
The argument, following Marx, is that since all profit arises from the
exploitation oflabour power (that is, variable capital plus surplus value)
the decline of variable capital relative to constant capital implies,
assuming a constant rate of exploitation, a fall in the rate of profit. The
fall in the rate of profit is said to cause, inter alia, an 'over-accumulation'
of constant capital because it is capital which cannot be used 'to exploit
labour at a given degree of exploitation' (Marx, 1974c, p. 255). In other
words an increasingly larger amount of constant capital will be used to
produce the same quantity of profit (see section 2.1). Lojkine refers to
this as the 'devalorization' of constant capital (Lojkine, 1976a, p. 131;
1977a, p. 163ft); hence his 'over-accumulation - devalorization hypoth-
esis' (Lojkine, 1976a, p. 131).
Lojkine's argument is that of the two components comprising urban
expenditure (that is, the collective means of consumption and the means
of material circulation) the former are 'totally devalorized capital
producing no extra capital (expenses capital)' (ibid., p. 131) while the
latter are 'very highly devalorized capital' (ibid.). Hence his conclusion
that: 'the financing of the collective means of consumption, like the
financing of means of material and social circulation, increase the mass
of capital which is devalorized in relation to that which is productively
utilised at or above the average rate of profit' (ibid.).
The main problem with this argument is the lack of any empirical
evidence of where and by how much the rate of profit is falling and, if so,
whether this fall is due to the effect of urban expenditure on the organic
composition of capital. Lojkine assumes, independently of any
empirical evidence, that urban expenditure strengthens the tendency for
the rate of profit to fall. He further assumes, following Marx, that there
is a tendency for the rate of profit to fall. As has been shown above
(section 2.1), that assumption is unwarranted.
In this section Lojkine's claim was examined: that there is an inherent
limit within capitalism to the financing of urban expenditure because the
latter causes the organic composition of capital to rise and the rate of
profit to fall. It was seen that this claim is without justification since
Lojkine provides no evidence to show either that urban expenditure
raises the organic composition of capital and even if it did, that this
would cause the rate of profit to fall. In the absence of such evidence
Lojkine's claim is purely an assertion.
7.3.3 The Problem of the Locational Strategy of Capitalist Firms
The second problem associated with urban development in capitalist

societies stems, according to Lojkine, from the locationa1 strategy of
firms (Lojkine, 1976a, pp. l33-5; 1977a, pp.170-71; 1977d,
pp. 264- 5). The problem, he argues, derives from the fact that firms tend
to locate in those areas which offer the best 'situational advantage'
(Lojkine, 1976a, p. l34). The consequence of this, in turn, is 'the
increasing underdevelopment of the regions least well-equipped in
urban infrastructure' (ibid.) on the one hand, and 'the growing
congestion of the gigantic 'megalopolises' (ibid.) on the other. Lojkine
concludes from this that the locational criteria and strategy of large
capitalist firms is in 'contradiction with the technological and social
necessities of any real territorial planning, i.e. of developed cooperation
at the level of the nation as a whole' (ibid.).
Lojkine does not elaborate his analysis further. His basic and
recurrent point is that the locational strategy of capitalist firms is such
that they will tend to locate in developed rather than underdeveloped
regiO'ns. The problem with this generalisation is that it does not accord
with the facts since there is growing evidence that manufacturing firms in
advanced capitalist countries are now tending to locate in smaller urban
and even rural areas and in areas that were formerly regarded as
underdeveloped (see, for example, Goddard, 1978, p. 311; Massey and
Meegan, 1978, p.287; Massey, 1979, p.237; Perrons, 1981; Keeble,
Owens and Thompson, 1983).
Lojkine does not consider that the locational strategy of a firm is likely
to depend upon its characteristics such as size (for example, uniplant or
multiplant) and product (for example, extraction, manufacturing,
services) as well as upon the availability and cost of inputs in various
regions. His analysis is therefore of rather limited value.
7.3.4 The Problem of Urban Land Rent
The third limit to any rational planning of urban development,

according to Lojkine, derives from the existence of urban land rent.
However, as will be seen presently, much of Lojkine's analysis is
concerned, not with substantiating this claim, but with elaborating the
concept of urban land rent (Lojkine, 1971; 1976a, pp. l35-8; 1977a,
pp.I72-7).
Lojkine's analysis of urban land rent or, to use his term 'urban
property rent' (Lojkine, 1971, p. 89) is essentially an attempt to show
that the conceptual categories developed by Marx for understanding
agricultural land rent and land prices are equally applicable in the
context of urban land rent and land prices. On the basis of this analysis,
he makes a number of claims about the effect of urban land rent on the
rate of profit, capital accumulation and urban planning which will be
examined below.
Lojkine begins his analysis by affirming that: 'In a given economic
sector, two conditions determine the possibility of property rent being
formed' (ibid., p. 89). These two conditions are:
1. the organic composition of capital in this sector must be below the
social average organic composition of capital;
2. there must be an obstacle, within this sector, to the free circulation of
capital so that the surplus profit generated in this sector (as a
consequence of the low organic composition of capital) will not be
redistributed to other sectors but appropriated by the landowners
within this sector (ibid.).
According to Lojkine: 'Marx proved in Capital that these two conditions
are to be found in the sector of agricultural production' (ibid., p. 89).
On the basis of his definition of rent, Lojkine then considers a number
of urban situations where these two conditions might exist, with a view
to establishing the existence of urban property rent. A number of
different situations are referred to, though seldom is a systematic
analysis developed. Before critically examining one of these situations,
which Lojkine does attempt to analyse, it may be useful to provide a
brief diagrammatic summary of the different situations to which he
alludes. This is done in Table 7.1.
Lojkine analyses one of the situations identified in Table 7.1, namely,
the case of a property developer who leases offices to capitalists for
commercial, financial or administrative use. His basic strategy is to
establish whether his two conditions for the existence of rent are present.
In this light, he attempts to prove that his first condition is present by
arguing that the capital invested in this sector, although unproductive,
nevertheless allows its owner to 'appropriate a more or less large amount
of surplus value (and hence profit)' (Lojkine, 1971, p.91). Lojkine
explains that this is because:
the organic composition of capital is still relatively low in this sector

given the massive employment of underpaid female labour (commer-
cial and office employees) and the still limited introduction of
mechanization (the productivity of the banking and commercial
sector is far lower, despite recent progress, than that existing in the
mining or industrial sector) (ibid., p.91).
Table 7.1 Lojkine's analysis of different types of rent in different urban

situations
Condition (J) for

the emergence of Condition (2) for
rent: the emergence of Types of rent
a lower than rent: arising:
average organic an obstacle to Differential rent I
Situations giving composition of the free mobility Differential rent Il
rise to rent capital (C/ U) of capital Absolute rent
Owner of land A lower than The landowner Differential rent I

leasing to a capital- average organic acts as a barrier to Differential rent II
ist farmer composition of investment in Absolute rent
capital agriculture
Owner of residen- Not specified Not specified Differential rent I

tial property leas- Absolute rent
ing to tenants No differential
Rent I
Own~rs of a fac- A higher than Not specified Not specified

tory leasing to a average organic
capitalist composition of
capital
Owners of com- A lower than The property de- Differential rent I

mercial, financial average organic ve10per acts as an Differential rent II
or administrative composition of obstacle to the free Absolute rent
offices leasing to capital mobility of capital
capitalists
Owners of collec- A higher than Not specified Differential rent

tive facilities average organic Absolute rent
(roads and rail- composition of No differential
ways) leasing to capital Rent II
capitalists
Source: Lojkine, 1971.
Lojkine argues that the second condition for the existence of urban
property rent is also present because of the existence of an agent who
hinders the free mobility of capital and the equalisation of the rate of
profit. This agent is the 'property builder-developer' (ibid., p. 91) who,
according to Lojkine, 'purchases, develops and equips the urban land
and leases the commercial (or office) premises to capitalist agents'
(ibid.).
According to Lojkine, this analysis is 'amply validated by contem-

porary facts: the leasing of commercial or office premises by private
property developers' (ibid.). As further evidence he refers to property
development at Parly II, La Defense and Maine Montparnasse.
Lojkine then elaborates his analysis by claiming that 'the three types
of rent distinguished by Marx' (ibid.) namely, differential rent I,
differential rent II, and absolute rent, coexist in the case of commercial,
financial or administrative offices. In explanation, he claims that:
(1) differential rent I arises because of the situational advantages of a
particular plot of urban land. According to Lojkine, 'the means of
communication playa key role here in differentiating land' (ibid.)
and cites the example of the proximity of the A 13 motorway to Parly
II;
(b) differential rent II arises because of differences in productivity
arising from differences in capital investment in commercial centres
and office buildings;
(c) absolute rent arises because of the ownership of land and the
obstacle which this raises to the free circulation of capitals and the
formation of an average rate of profit (ibid.).
There are three serious difficulties with this analysis. The first is that
Lojkine's definition of rent corresponds to only one of the types
identified by Marx, namely, monopoly rent I (or, as Marx termed it,
absolute rent) (section 3.3). Lojkine implies that the conditions for the
existence of this type of rent apply to all types which is not the case (see
Chapter 3 above). Moreover, the first of these conditions (that is, a low
organic composition of capital) is not even necessary for monopoly rent
I.
The second difficulty is that Lojkine fails, in any case, to show that his
two criteria for the existence of rent actually exist empirically. As regards
the first criteria he simply affirms its existence by referring to the large
numbers employed in the commercial and financial sector. But this
evidence, however suggestive to some, does not establish anything about
the organic composition of capital. Lojkine also fails to establish the
second criterion because he fails to show in what way the developer and
owner of an office block acts as an obstacle to the free mobility of capital
and the equalisation of capital. This proposition is affirmed and never
proven.
The third difficulty arises with respect to Lojkine's discussion of the
three types of rent. Two specific difficulties may be noted here. In the
first place, there are four types of rent present in Marx's work and not
three as Lojkine claims (Chapter 3). Secondly, there is an inconsistency
between Lojkine's initial definition of rent (in terms of two conditions)

and his definition of differential rent I and II. This is because he
acknowledges that the two conditions which are necessary for rent to
exist are not in fact necessary for the existence of differential rent I or II.
These are the most serious problems associated with Lojkine's
analysis of office property and they reveal a fundamental conceptual
confusion in this part of his work.
Lojkine then makes four claims about the nature and consequences of
rent in general. These four claims are:
(1) urban rent acts as 'an obstacle' to 'the development of social
productive forces' (Lojkine, 1976a, p.135), to 'monopolistic
accumulation' (ibid., p. 136), and to 'economic growth' (ibid.,
p.135);
(b) urban rent strengthens the tendency for the rate of profit to fall
(Lojkine, 1976a, p. 137; 1977a, p. 174);
(c) urban rent is a 'structural obstacle to all real urban planning'
(Lojkine, 1976a, p. 137; 1977a, p. 175);
(d) urban rent gives rise to 'economic and social segregation' (Lojkine,
1977a, p. 175).
Lojkine makes no serious attempt to substantiate or elaborate any of
these four claims. As regards the first two, it is clear that they are closely
related since the rate of profit determines the rate of capital accumula-
tion and the rate of economic growth. In other words, both these claims
are identical. The question therefore, is whether and, if so, under what
conditions does the existence of rent lead to a fall in the rate of profit
and, in turn, to a reduction in the rate of economic growth. 2 As was seen
in Chapter 3, the answer to this question depends upon the type of
commodity being produced (that is, a basic or a non-basic) and on the
type of rent (that is, differential rent I and II, monopoly rent I and II).
Moreover, the issue of whether rent is an 'obstacle' to capital accumula-
tion, as Lojkine claims, depends essentially upon whether it arises
because of the existence of the private ownership ofland or independent-
ly of it. It would seem that only those types of rent which exist because of
the private ownership ofland (that is, monopoly rent I) can properly be
regarded as 'obstacles' to capital accumulation. Rent which exists
independently of the private ownership ofland (that is, differential rent I
and II, and monopoly rent II) is no more an 'obstacle' to capital
accumulation than other costs which must be borne in the course of
capital accumulation.
Lojkine's third claim is that rent is a 'structural obstacle' to all 'real'

and 'rational' urban planning. It is, however, impossible to evaluate the
validity of this claim without first clarifying the meaning of the terms
'real' and 'rational' planning. Unfortunately, Lojkine does not clarify
these terms so that no evaluation of this claim is possible.
Lojkine's fourth claim is that urban land rent gives rise, via land
prices, to 'economic and social segregation' (Lojkine, 1977a, p. 175). He
distinguishes between, three different types of 'urban segregation' (ibid.,
p.176):
1. an opposition between the centre where land prices are high (because
of 'agglomeration effects' and the periphery where land prices are
low;
2. a separation between desirable and undesirable residential areas;
3. a separation between office, industrial and residential zones.
Lojkine's main argument (ibid., pp. 176-7) is that land prices in the city
centre are now so high that the centre is 'increasingly reserved for the
headquarters of international firms' (ibid., p. 176). Identifying inter-
national firms with monopolistic firms, Lojkine concludes that both
wage-earners (of all sizes) and non-monopoly capitalists are excluded
from the city centre by monopoly capital.
Lojkine's claim is basically an empirical claim and, although no
evidence is presented to substantiate it at this point, his empirical
research in Paris (Lojkine, 1972a) and Lyon (Lojkine, 1974a) does
provide some confirmatory evidence on this point (see subsections 7.6.1
and 7.6.2).
The analysis in this section has revealed a number of weaknesses in
Lojkine's treatment of the Marxian concept of rent. This was seen, not
only in his exegesis of Marx's concept, but also in his application of it to
the urban context. In addition it was seen that some of Lojkine's claims
about the effects of rent were quite unfounded.
7.3.5 Conclusion
Lojkine's analysis of the three problems associated with urban develop-

ment in capitalist societies is seriously flawed essentially because it is
based upon an inadequate understanding of the nature and limitations
of Marxian political economy. It is not, therefore, a significant
contribution to the Marxist analysis of urban areas.
It is now necessary to examine Lojkine's analysis of urban planning.
7.4 URBAN PLANNING
Lojkine's analysis of urban planning is an extension of his analysis of the

three problems which, in his view, are inextricably associated with
capitalist urban development. These problems, which have been
analysed in the preceding section, are the financing of the collective
means of consumption, the locational strategy of firms and urban land
rent. In Lojkine's view, the 'capitalist relations of production' (Lojkine,
1976a, p. l28) which generate these problems simultaneously act as an
'obstacle' (ibid.,p. 137; 1977a, p.l72), a 'limit' (Lojkine, 1976a, p.128;
1977a, p. l60) and a 'hindrance' (Lojkine, 1977a, p. l72) to any 'real'
(Lojkine, 1976a, p.137) 'rational' (ibid.,p.128; 1977a, p.159),
'socialized' (l977a, p. 159) urban policy.
Lojkine's analysis of urban planning is also an extension of his views
on the nature and consequence of state intervention within the period of
SMC (see section 5.2.). This view is essentially that all state intervention
is an attempt to regulate the economic and social contradictions
generated in this stage of capitalist development. However, since this
intervention is subordinated to 'the logic of private capital accumula-
tion' (Lojkine, 1977d, p. 258) it tends to exacerbate rather than regulate
the problems which it attempts to solve. Thus, for Lojkine, 'all state
policy acts less as an instrument of regulation than as an indicator of a
society torn by the conflict of antagonistic classes' (ibid., p. 258). In
short, it is an 'active reflection' (ibid., pp. 141ft) of the class struggle.
Lojkine differentiates his view of urban planning from two other
views which he regards and rejects as functionalist (ibid., pp. 141, 147).
These views treat urban planning as either 'a mechanism of state
regulation designed to resolve the technical problems of urban develop-
ment' (ibid., p. 14l) or 'a form of ... regulating social contradictions'
(ibid.). The error in these views, according to Lojkine, is the assumption
that the state can regulate the problems generated by capitalism within
the framework of capitalism. This, in Lojkine's view, is impossible.
Lojkine develops the concept of urban planning by isolating what he
regards as its three different dimensions (Lojkine, 1977a, p. 191). These
dimensions are:
1. a planning dimension, which refers to the written plans, that is, the
documents which reveal the planning goals being adopted by the
planning authority;
2. an operational dimension, which refers to the financial and judicial
interventions of the state in relation to those plans;
3. an urban dimension, which refers to the effects of the two preceding

dimensions
(ibid., see also 1977d, p.259 where urban planning is defined only in
terms of the first two dimensions).
Lojkine emphasises two points in particular about this concept of

urban planning. The first is that there is a 'complete opposition'
(Lojkine, 1977a, p. 192) between the goals of planning as elaborated in
the planning texts and documents (that is, the first dimension) and the
actual implementation of those goals (that is, the second dimension).
This opposition takes the form of delays or cancellations in the
implementation of those aspects of the plan which are unprofitable or
financially expensive (for example housing, green spaces, schools, other
collective facilities; ibid., pp. 192, 195). There is thus 'a constant
distortion between plans and practices' (ibid., p. 186) which tends to
increase as the financial requirements of plans increase (ibid.,
pp. 197 -8). For this reason Lojkine quotes approvingly from Castells to
the effect that urban plans are, above all, 'ideological texts which in no
way deprives them of their social efficacity but characterizes the
intervention of the political as bearing not on the urban system but on
the general ideological instance' (Lojkine, 1977a, p. 186; Castells and
Godard, 1974, p. 276). (At the same time Lojkine criticises Castells for
his tendency to actually identify urban planning with urban plans (ibid.).
The second point which Lojkine emphasises about his concept of
urban planning is that the true 'logic' (Lojkine, 1977a, p.203) and
'sociological content' (ibid., p.254) of urban planning can only be
identified from its effects, that is, the third dimension. On this basis he
reviews the evidence on the effects of the state's urban policy in a number
of French cities; Paris (Lojkine, 1972a; 1974c; Castells, 1973a; Freys-
senet, Regazzola and Retel, 1971), Dunkirk (Castells and Godard, 1974,
Lyon (Lojkine, 1974a) and Rennes (Huet et ai., 1973). All of this
evidence, according to Lojkine, confirms that the state's urban policy in
contemporary France is in favour of the monopolies and against the
interests of all non-monopoly groups (Lojkine, 1977a, p.273; 1977b,
pp. 149ff; 1977c, pp. 21-2; 1977d, p. 259; Lojkine and Preteceille, 1970,
p. 80). In other words, its effect is to 'exacerbate the principal
contradiction, at the current stage of capitalism, between the mon-
opolies and the non-monopoly social strata' (Lojkine, 1977a, p.273).
Lojkine bases this conclusion on the claim that the bulk of the state's
urban expenditures are concentrated in the centre of urban areas (for
example, Paris and Lyon) or in large industrial complexes (for example,
Dunkirk) which are increasingly occupied by the monopolies, to the

neglect of expenditures on the reproduction of labour. In short, the
effect of the state's urban policy is the 'segregative distribution of the
general conditions of production' (Lojkine, 1977d, p.258).
These are the main features of Lojkine's analysis of urban planning.
Although this analysis contains the useful distinction between the three
dimensions of urban planning it nevertheless remains underdeveloped
and excludes some important aspects of urban planning. Three aspects
in particular should be noted.
The first is that urban plans have a double purpose: they not only
provide the framework for state expenditure in particular areas (as
Lojkine correctly points out), but they also provide a set of guidelines
through which the planning authority regulates private property
development. These guidelines are then enforced by the planning
authority through the process of development control. Lojkine's
analysis contains no reference to development control and hence
excludes an important area of interaction between the state and the
private sector and hence, in turn, excludes the possibility of revealing the
constraints which private property developers as well as landowners
impose upon the state and vice versa.
The second aspect of urban planning which is absent in Lojkine's
analysis is the process by which plans are actually produced. Plans are
produced by officers and politicians of the state (whose interests may
frequently be in conflict) and it is through these individuals that the
constraints and conflicts engendered in a capitalist society must be
interpreted and reflected in the plan. Lojkine's analysis contains no
reference to the way in which plans are made and hence to the process by
which class conflicts are reflected in the state.
The third aspect of Lojkine's analysis which remains weak is his
treatment of the effects of the state's urban policy. Throughout this part
of his analysis Lojkine nowhere takes into account the elementary fact
that all state expenditure is ultimately financed from taxes on wages,
profits and rents, so that a completely accurate account of the effects of
the state's urban policy in a particular locality must be based upon a
comparison of the distribution of taxes between the different classes,
with the distribution of the subventions which they receive back from the
state through urban planning. Moreover, if such a comparison were
made it may well reveal that not all non-monopoly groups (that is,
medium and small firms as well as professional, artisanal and manual
workers) are equally affected by the state's urban policy as Lojkine in
fact maintains.
These are some of the main deficiencies of Lojkine's analysis of urban

planning. An attempt will be made below (in Chapter 8) to bring these
considerations together into a more coherent and plausible framework
for the analysis of the state's urban policy.
It is now necessary to consider Lojkine's analysis of urban social
movements.
7.5 SOCIAL MOVEMENTS
Lojkine's analysis of social movements (see notably Lojkine, 1977a,

pp. 312--41) is essentially concerned with clarifying the main elements
which make up a social movement, and with exploring the possibility for
the emergence of a 'new type of social movement' (Lojkine, 1977b,
p. 153; 1977a, pp. 334ft). He defines a social movement as a mobilisation
of the dominated class for the purpose of making a number of demands
on the dominant class, thereby challenging their power and, possibly,
overthrowing it. This definition is implicit in a number of statements
made by Lojkine:
'the social movement only has significance by its capacity to oppose the
dominant class and all of its hegemonic system' (Lojkine, 1977a, p. 313).
'the real historical influence of a social movement can only be defined by
the analysis of its relationship to political power' (ibid., p. 320).
'the social movement is defined, in the last resort, by the capacity to
change the socio-economic system in which it is born' (ibid.).
These quotations indicate that, for Lojkine, a social movement is

more than just a protest or a reform movement. It is a movement with
radical intent which may 'open the road to socialism' (ibid., p. 316) and
may ultimately bring about a socialist revolution. As an example of a
successful social movement Lojkine cites the case of the Russian
Revolution of April and October 1917 (ibid., pp. 316ft). As examples of
unsuccessful social movements he cites the case of 'the French social
movements of 1848, 1871 and 1968' (ibid., p. 321).
Lojkine then proceeds to analyse in greater detail the basic elements of
a social movement. He argues that a social movement is 'the combina-
tion of two social processes' (ibid., p.318). These two processes or
'dimensions' (ibid., p. 319) are:
I. class mobilisation, and

2. the political stake (ibid., pp. 318-19).
Lojkine's discussion of each of these processes will now be examined.

As regards the first dimension, he states that class mobilisation
consists of two elements:
(a) the social base, and

(b) the organisation.
Together these form the 'social force' (ibid., p. 318). It follows from
this that the extent to which a social force can be mobilised is crucially
dependent on the nature of both the social base and the organisation.
Lojkine points out that a social base can vary according to its size and
according to its heterogeneity. It can be either large or small in size and
homogeneous or heterogeneous in class composition. The social
movement which led to the Russian Revolution of 1917 was, according
to Lojkine, clearly a case where the social base was large in size and
heterogeneous in class composition (though this is not the only possible
combination for mobilising a social force). In addition, a social base can
be spread over areas of different size (national, regional, urban or local)
and this too can affect the mobilisation of a social force.
Lojkine makes two important points about organisations. The first is
that a social movement does not occur 'spontaneously' (ibid., p. 321). It
requires an organisation. The second point is that organisations can
vary both with respect to their 'degree of openness' (ibid., p. 318) and
according to whether they exist at a national, regional, urban or local
level.
Lojkine attributes particular significance to the spatial aspect (that is,
national, regional, urban, local) in the mobilisation of a social force. He
argues that a national, as distinct from a regional, urban or local social
movement, implies 'a different relationship to the power of the state
insofar as the latter relies, in the last instance, on the capacity of the
dominant class to maintain the 'national' cohesion of the whole of the
social formation' (ibid., p. 319). Lojkine goes further and argues that
only a national social movement is capable of attacking the political
power of the dominant class (ibid.).
As regards the second dimension (that is, the political stake) Lojkine
states that a social movement is crucially dependent on its 'political
stake' (ibid., p. 319). The precise nature of the political stake involved in
any social movement can only be ascertained through an analysis of the

ideological and political content of both its claims and its actions. It is,
according to Lojkine, only after the precise nature of the political stake
has been established, that it is possible to assess the capacity of a social
movement to actually challenge the political power of the dominant
class (ibid., p. 320).
As with the first dimension, the spatial aspect is also important here.
According to Lojkine, only a political stake which is national in its
import has the possibility of being truly revolutionary, that is, of posing
a challenge to the central, national, political power of the dominant
class. Moreover, even if, in the case of a given social movement, the
mobilisation of the social force is limited to a regional, urban or local
level, but it has a national political stake, it could still be a revolutionary
movement.
Both Lojkine and Castells share the belief that a social movement
must be radical in intent and pose a challenge to the dominant class in
the society. For Castells an urban protest can only be successful if it is
fused with other movements (see section 6.6.2) whereas, for Lojkine, a
social movement can only have a chance of success if it is not limited
solely to an 'urban' issue. Thus Lojkine's position is virtually identical to
that of Castells (despite Lojkine's protests to the contrary; ibid.,
pp. 322-5). Neither regard a local struggle over an item of collective
consumption as having any potential for radical or revolutionary
change, and both regard them, therefore, as politically and analytically
insignificant.
Lojkine, finally and briefly, refers to 'a new type of social movement'
(Lojkine, 1977b, p. 153; 1977a, pp. 334 ft) which is likely to emerge in
France in the near future. He writes: 'French urban social movements
have until recently been characterized by a total separation between the
economic demands fought for by the working class within the work
place (higher wages and the slowing down of speeds) and the 'social'
demands linked to the reproduction of labour power (housing, trans-
port, health, cultural facilities, and so on). The new feature of the 1970s
is, on the contrary, a strongly expressed desire by the labour movement
to make a close link between demands within the workplace and those
related to the 'environment' (Lojkine, 1977b, pp. 153-5; 1977e, p. 23).
The distinctive feature of this new type of social movement, according
to Lojkine, is that a number of different demands are brought together
and made part of the same movement. As an example of this he cites the
case of a strike which occurred in Turin in July 1969 (Lojkine, 1977a,
pp.335-6). This strike was organised by the trade unions and by

political parties of the left to protest against the increase in rents and in
the cost of living. In other words, a number of different demands were
linked together in this movement. Social movements of this type have
not yet occurred in France, according to Lojkine, with the possible
exception of a strike which took place at the Rateau factory in La
Courneuve, Paris, between February and April 1974.
These are the main features of Lojkine's analysis of social movements.
It is similar, in many respects, to Castells's analysis (see section 6.6) so
that many of the comments made in that context apply equally here. In
particular, both are primarily interested in those protest actions which
are radical and revolutionary in both intent and effect. However, as has
been argued above in the case of Castells, this would seem to be unwise
since many of the protest actions which occur in urban areas are
undertaken by purely local groups (such as residents associations,
tenants associations, environmental lobbies, and so on) and are
concerned with purely local issues (such as schools, public transport,
crime, noise, and so on). Many of these protest actions are not radical
either in intent or effect yet, as is argued in Chapter 8, they are an
important feature of the consumption process in urban areas. Thus
Lojkine's exclusion of such protest actions, whatever its political
justification, would seem to be an error.
The most important and insightful part of Lojkine's analysis is the
spatial dimension. He recognises that a social base, an organisation, and
an issue, have a spatial dimension in the sense that they can vary spatially
from a national to a local level and this fact is of central importance in
understanding not only the level of mobilisation of a social force but also
the potential impact of the movement. In this respect, Lojkine's analysis
is superior to that of Castells.
Lojkine's analysis concludes in the sanguine belief that a new type of
social movement is emerging in the 1970s in France and elsewhere, and is
likely to become even more important in the future. While this is clearly a
possibility, it may not be as inevitable as Lojkine believes, essentially
because neighbourhood and locality-based interests (that is, those
concerned with the issues of collective consumption) do not always nor
necessarily coincide with work-based interests (that is, those concerned
with issues of pay, conditions of work, and so on). Thus Lojkine's new
type of social movement is only likely to emerge when, ceteris paribus,
there is a coincidence between these two sets of interests.
It is now necessary to examine how Lojkine applies the conceptual
framework outlined in the preceding sections to his empirical research.
7.6 EMPIRICAL RESEARCH
7.6.1 Introduction
Lojkine has undertaken two major pieces of research: one on urban

development in Paris (Lojkine, 1972a and 1974c; 1976b) and one on
urban development in Lyon (Lojkine 1974a). More recently he has been
undertaking research on the political aspects of urban development in
Lille, the preliminary results of which are now becoming available
(Lojkine, 1980, 1981b). The present section will be confined to an
examination of his research in Paris and Lyon since these constitute the
main corpus of his research work.
7.6.2 Research on Urban Policy in Paris in the Period 1945-72
Lojkine's study of Paris was first published in 1972 and is an attempt to

reveal 'the real sociological content' (Lojkine, 1972a, p.12) of the
changes which have been occurring in the Paris region over the past few
decades. These changes effectively involve the 'deindustrialization'
(ibid., p. 15) of the city and its transformation into a major 'tertiary
centre' (ibid., p. 12); they also involve, in addition, a growing separation
and 'segregation' (ibid., p. 41) of places of work from places of residence,
with consequent problems for commuters, given the inadequate trans-
port system. Lojkine's book is essentially an attempt to describe, in
broad outline, the processes which underlie these changes.
The book is divided into three chapters. The first chapter (pp. 15-81)
provides a general account of the major economic changes which have
been occurring in Paris since 1945. These changes are described, first, in
terms of the way in which they affect the 'dominant class' and second, in
terms of the way in which they affect the 'dominated class'. The second
chapter (pp. 83-186) attempts to discover the role which the planning
authorities have played in relation to these economic changes, with
particular reference to the spatial dimension. Finally, in the third
chapter (pp. 187 - 261), Lojkine examines a number of specific examples
of urban planning which, he argues, are particularly revealing of the
nature of state activity. Each of these chapters will now be briefly
summarised and assessed.
The first chapter of Lojkine's book begins by looking at the
relationship of the dominant class to the deindustrialisation of Paris.
Lojkine notes that deindustrialisation began in the 1930s and increased
rapidly after the end of the Second World War (though the actual
evidence for this is only sourced and not presented (Lojkine, 1972a,
p. 15, fn I). This process effectively involved the removal of many
industrial installations from Paris to either the suburbs or the provinces
and their replacement by offices which tended to locate to the west of the
traditional centre of Paris (for example, La Defense). These offices are
usually the headquarters of large firms or centres for research and
development and these, in turn, have attracted luxury shops, hotels and
houses.
The main thrust of Lojkine's argument is that the deindustrialisation
of Paris cannot be explained as the outcome of either technological
change (ibid., p. 16) or urban planning (ibid., pp. 18-19). The explana-
tion lies rather in the way in which different firms have responded to the
changing conditions of profitability of the centre of Paris vis-a-vis its
suburbs and the provinces. According to Lojkine, large industrial firms
(such as Renault and Simca) have transferred their various plants from
the centre of Paris and concentrated them in the suburbs or the provinces
because this strategy offers them a number of advantages:
cheaper suburban sites with proximity to the Paris labour and consumer
markets;
avoidance of the high costs of land and congestion in Paris;
a cheaper and more docile labour force;
a reduction in the costs associated with dispersal by concentrating
different production units on one suburban or provincial site (ibid.,
p.17).
In contrast to these large industrial enterprises, for whom relocation
and spatial concentration outside Paris are rational and beneficial
courses of action, medium and small firms have largely been unable to
follow and relocate outside Paris because:
unlike larger industrial firms (which frequently do not require a highly
skilled labour force), medium and small firms frequently depend on a
highly skilled labour force which is not obtainable outside the centre of
Paris;
the costs of moving are often prohibitive;
the grants system for relocating favours the larger rather than the
smaller firms (since the grant is proportional to the number of jobs
created);
they frequently do not have access to new finance (ibid., p. 20).
These factors, together with the fact that many of these medium and
small firms are unable to afford the currently expensive sites in the centre
of Paris, has had the consequence that many such firms are disappearing
through voluntary liquidaton or bankruptcy (ibid., p. 20).
Thus the relocation and expansion of large firms coupled with the
disappearance of many medium and small firms suggests the more
general conclusion that there is, as Lojkine puts it, 'a close correlation
between the movement of economic concentration and industrial
decentralization' (ibid., p.20).
Turning to the 'dominated classes' (who include all wage-earners,
shop-keepers, artisans, and so on), the main consequence of the twin
processes of deindustrialisation and tertiarisation is that there is now,
within the Paris region, 'a growing spatial disequilibrium between place
of work and place of residence' (ibid., p. 41). This disequilibrium, which
Lojkine usually refers to as 'urban segregation' (ibid.), expresses itself in
the fact that the centre and neighbouring suburbs of Paris are
increasingly becoming places of employment while residential areas are
being pushed further towards the urban periphery. The major con-
sequence of this is that many workers are required to undergo long,
twice-daily commuting journeys to work. Lojkine cites evidence (ibid.,
pp. 46- 7) which confirms that the daily movement of commuters
between Paris and its suburbs is continuing to grow apace as is inter-
suburban commuting. The consequence of this, in combination with the
inadequate financing of public transport generally, is that the time and
the discomfort of travelling to work have increased considerably in
recent years (ibid., pp. 48-9).
In the second chapter of his book, Lojkine examines various urban
planning policies (particularly those relating to housing and public
transport) which have been pursued between 1945 and 1971, of which,
he claims, the main effect has been urban social segregation. Lojkine
provides a particularly long discussion of public transport policies (ibid.,
pp. 100-135; see also 1974b) the essence of which is that the public
transport system 'reinforces' (Lojkine, 1972a, p. 121) and 'accentuates'
(ibid., p. 92) segregation: it reinforces segregation because the rail
network is 'radiocentric' (ibid., p. 103) in design and, as such, facilitates
long commuting journeys between the centre and the periphery of the
Paris region at the expense of inter-suburban journeys; it accentuates
segregation because the transport system is underfinanced (as a result of
attempting to run it according to the principles of the free market) so
that commuting journeys take longer and are more uncomfortable than
they might otherwise be.
In Chapter 2 Lojkine also discusses three aspects of urban policies
over which there has been some controversy within the state itself (ibid.,
pp. 135-86). The first controversial issue was whether the growth of
Paris should be curtailed or whether it should be allowed to expand.
Lojkine traces the various 'oscillations' (ibid., p. 141) between these two
planning objectives, from the Third Republic (1870-1940) through
various planning documents notably PADOG (1960),3 SDAU (1965)
(1969t and the preparatory documents for the Sixth National Plan
(1971- 5). However Lojkine concludes that this controversy is 'above all
ideological' (Lojkine 1972a, p. 146), essentially because the actual
course of events in such matters as housing, employment, the location of
new offices, the relocation of manufacturing industries, indicates that
the growth of the Paris region (defined according to ZEAT), 5 far from
being curtailed, has in fact grown more rapidly than France as a whole
over the past 20 years.
The second controversial issue was whether the redevelopment and
expansion of the business and commercial centre of Paris should
proceed westwards, or whether it should be concentrated in the centre
and east of Paris. The former approach was crystallised in PADOG
(1960) in its proposal to continue the development of La Defense (to the
north-west of Paris) as the locus of new office expansion (Lojkine,
1972c, p. 149); the latter approach was advocated by the Commissariat
au Plan6 and by SDAU (1965) who advocated the redevelopment of the
centre and east of Paris and a cessation of the development at La Defense
(ibid., pp. 149- 50). The actual course of events, as measured by building
and planning approvals for offices (ibid., pp. 153-8), reveals that the
development at La Defense, and in the western sector generally,
continued apace, which effectively proves, according to Lojkine, that the
state endorses the dominant tendencies of the Paris property and office
market and thereby acts 'as a means for systematizing segregation rather
than as a means of suppressing it' (ibid., p. 166).
The third and final area of debate has been the mode of financing
urban public transport. The main debate has been between those
(notably the Ministry of Finance) who argue that it should be financed
by levying the appropriate and commercially-determined fares on the
users/workers (ibid., p. 183), against those (virtually all of the planning
agencies) who argue that it should be financed, in part at least, by a tax
on employers and property developers. In the event, it is the former
option, which Lojkine describes as 'Malthusian and liberal' (ibid.,
p. 181), which has largely prevailed 'since 1945 even 1930' (ibid., p. 183).
The third chapter of Lojkine's book is an attempt to show how the
actions of the state, particularly as expressed through various planning
agencies and policies, can be understood in one of three ways: as the
organisation of the hegemony of the dominant class; as the active

reflection of the class struggle; as the maintenance of the cohesion of a
social formation. An example of each of these three aspects of state
activity will now be examined to reveal the tenor and content of this
chapter.
Lojkine cites the example of the planning and development of the
office complex at La Defense (of 1.6 million 2 of office space) as an
example of how the state organises the hegemony of the dominant class.
The idea for the development of an office complex at La Defense was
effectively initiated by the state with the setting up, in 1958, of The Public
Institution for the Development of La Defense (l'Etablissement Public
Charge de l'Amenagement de la Defense). At the time, this appeared
'perfectly strange' (ibid., p. 214) to the majority oflarge firms who have
since located there, essentially because it was not until the late 1960s that
the demand for office space actually began to grow rapidly (ibid.,
p. 219). In the light of subsequent events which have seen the transfor-
mation of La Defense into a vibrant centre for the headquarters of many
large multinational firms, the actions of the state in this regard can be
seen as a perfect example of 'anticipation' (ibid., p. 218) of the future
needs of the dominant class and, in this sense, Lojkine regards the state
as the organiser of the interests and hegemony of the dominant class.
Lojkine then cites a number of different examples to show how urban
planning is also a reflection (an 'active reflection'; ibid., p. 247) of the
class struggle. These examples (such as the extension of certain metro
lines, the postponement of certain fare increases on public transport, the
decentralisation of some office employment) are designed to show that
the actions of the state can also be a response to pressure from the
dominated class. Indeed Lojkine affirms that all of these measures
'directly reflect the pressure of those social classes who are victims of
urban segregation' (ibid., p. 239). Lojkine does not, in fact, substantiate
this claim with any evidence: there is no evidence of any protest or social
movement in relation to any of these issues, nor indeed is there any
indication that the state's actions were an anticipated response to such
protest movements. There is, nevertheless, a more general and valid
point which is implicit in Lojkine's analysis, which is that the state can
(and does) respond to demands and pressures from the working/wage-
earning class, without significantly altering the capitalist system of
production, as long as these concessions do not seriously undermine
productivity and economic growth. Finally, Lojkine gives an example of
how the state, through its planning agencies, can take measures which
seem to be governed above all by a concern with maintaining the
cohesion of capitalist society. Lojkine's example is the project for the

development of a 'Cite Financiere' (ibid., pp. 199-211) in the centre of
Paris (Opera-Saint-Lazare-Chateaudun) which would rival Wall Street
in New York and 'The City' in London. This project was strongly
advocated by the banking and insurance industry but was strongly
opposed by many of the relevant planning agencies, notably DATAR, 7
who regarded it as a measure which would further concentrate
population and employment in Paris at the expense of the provinces.
Writing in 1972, Lojkine argued that, in this case, the logic of the
planning agencies was directly opposed, and in contradiction, to the
logic of the banking and insurance industry; and on this basis he
concluded that it was 'impossible to explain' (ibid., p.209) the state's
activity in relation to this project without reference to the cohesive role
and function of the state. In a subsequent article, Lojkine (l974c) has
examined the project more thoroughly in the knowledge that the project
has gone ahead albeit with a decentralisation of a wide-range of banking
and insurance activities. His interpretation in this article is that the
contradiction between the banking and insurance industry and the
planning agencies is in fact only apparent. This is because the revolution
in information technology (notably through computers) has made it
possible to have a high degree of decentralisation of 'technico-adminis-
trative functions' (ibid., p. 128) in the provinces headed by one large
centre in the 'Cite Financiere' in Paris for the purpose of making and
implementing investment decisions in the various equity, gilt, and
property markets. Lojkine's later analysis (l974c) is a considerable
improvement on his earlier analysis (1972a), both in terms of its
approach (that is, comparing the optimal strategy open to the banking
and insurance industry, in the light of recent technical innovations, with
the planning strategies of the various state agencies) and in terms of the
richness of the data. As such, his 1974 interpretation of the events
surrounding the 'Cite Financiere' project would seem to be more
accurate than that suggested in 1972.
Lojkine's book, as the preceding synopsis reveals, contains a very
considerable amount of information and analysis on urban planning
and development in the Paris region. It is one of the first research
monographs which has attempted to examine urban planning and
development from a Marxist perspective. According to one French
reviewer, it is also the 'first study' (Lautman, 1974, p.274) of the
development of Paris as a whole in the post-war era. The research is
based largely on an analysis of planning documents, official reports, and
on a number of interviews, though Lojkine gives no information about
the number or nature of those interviews (see Lojkine, 1972a, pp.37,

207, 208). The research is unavoidably incomplete in the sense that a
number of important projects (notably the one million square metres of
office space at Bercy in the east of Paris) were unfinished at the time of
writing.
Because of the wide-ranging nature of Lojkine's project, it is
inevitable that some issues are dealt with in insufficient detail or
precision. This is particularly the case with Lojkine's concept of class,
which divides the entire class system into two groups: the dominant and
the dominated, reflecting Lojkine's abiding belief that the central
cleavage in the contemporary phase of capitalist development is between
monopoly and non-monopoly groups (see section 5.2). While Lojkine
does not make some attempt in Chapter 1 to differentiate between the
different fractions of the dominant class (largely in terms of size: large,
medium, small), he makes no attempt to differentiate between the
different fractions of the dominated class which includes all wage-
earners, artisans, shopkeepers, and so on. While the evidence from
Lojkine's research does reveal that all of the members of this class have
suffered the segregative effects of urban planning and development in
Paris, it is unlikely that they have all been affected uniformly. The fact
that there does not appear to have been any concerted action by them in
the form of an urban social movement might be taken as indicative of
this.
In terms of the over-all corpus of Lojkine's work, this book reveals
that a basic division exists between much of Lojkine's theoretical work
and his actual research. Lojkine's book, while providing a broad
Marxist account of 25 years of planning and development in Paris, does
not deal with those issues which are actually central to his theoretical
writings and which are so much in need of empirical research. This is
particularly the case with his analysis of collective consumption and his
analysis of rent. As has been seen (subsection 7.3.3), Lojkine claims that
state expenditure on collective consumption has the effect of reinforcing
the fall in the rate of profit. This claim is elaborated at some length, and
was first published in the same year as the present study (1972), yet no
attempt is made to subject it to the test of empirical evidence. Similarly,
with the concept of rent (subsection 7.3.5), which was first published in
1971, but which receives no attention in this study despite its centrality to
the change in land use which has been occurring in Paris over the past 25
years. This division between theoretical and empirical research (which is
also to be found in Castells - see notably section 6.7) may be the cause of
many ofthe errors which were identified in Lojkine's theoretical analysis
(sections 7.1 to 7.5) and is one of the basic weaknesses, not only ofthis
research, 8 but of his entire work.
7.6.3 Research on Urban Policy in Lyon in the Period 1945-72
Lojkine's study of Lyon was carried out between 1970 and 1972 with the
collaboration of two other researchers (Fran~oise Orlic and Catherine
Skoda) and was published in 1974 (Lojkine, 1974a). Lyon, with a
population of just over 1 million in 1976 (see House, 1978, p. 142), is the
second largest city in France (though still nine times smaller than the
largest, Paris), and Lojkine attempts to 'preserve the parallelism'
(Lojkine, 1974a, p. 31) between this study and his previous study of Paris
(see subsection 7.6.2) by confining both studies to the same historical
period (1945 - 72), and also by attempting to analyse urban planning and
development in both cities in terms of its consequences for different
classes and class fractions.
Like his previous study of Paris, Lojkine's study of Lyon is based
upon documentary research. This research draws upon a variety of
materials: planning documents, research reports, reports of parliamen-
tary debates and a variety of books, newspapers and magazines.
Lojkine, however, does not discuss any of the methodological issues
which arose from the use of these data sources.
Lojkine's book is divided into three chapters. The first chapter
(27 pages) is a brief analysis of the expenditure by the various agencies of
the state on the city of Lyon. The second chapter (75 pages) examines
commercial capital, banking capital and industrial capital respectively,
and the way in which each of these class fractions have been changing in
recent years. The third chapter (48 pages) is concerned with the working
class and with the way in which it has been affected by recent urban plans
and policies. Each of these chapters will now be briefly summarised and
assessed.
In the first chapter (ibid., pp. 29-56) Lojkine identifies four different
state bodies, each of which makes a contribution to the total expenditure
of the state in Lyon. These state bodies (listed according to their sphere
of power and influence) are:
The City of Lyon (Ville de Lyon);
The General Council of the Rhone (Conseil General du Rhone);
The Urban Community of Lyon (La Communaute Urbaine de Lyon);
The National Government.
Lojkine discusses the expenditure of each of these four state bodies

with particular reference to their expenditure on public transport and
housing.
Beginning with the City of Lyon, Lojkine highlights three sets of facts
about the pattern of expenditure of this body. The first is that
expenditure (and the borrowing necessary to finance it) rose particularly
rapidly after 1960 (ibid., p. 32). The second fact is that expenditure on
public transport is inadequate. This is evidenced, according to Lojkine,
by the fact that there has been a deficit on the running of the public
transport service since 1963 (ibid., p. 35) and also by the fact that, in
1967, the public transport network only covered an area of 150km2 while
the size of the Lyon agglomeration is 2000km2 (ibid., p. 37). Unfortun-
ately, Lojkine does not indicate the actual amount or the proportion of
the total budget which was devoted to public transport except to state
that, since 1961, it was 'even less' (ibid., p. 34) than that spent on HLM
housing - which, on the basis of data subsequently presented on page 43,
means that it was below 6 per cent of the total expenditure.
The third set of facts relate to expenditure by the City of Lyon on
housing. The basic data are that, in the period 1953-61, annual
expenditure on housing averaged 10 per cent, but was considerably
below this in the preceding and succeeding periods and in 1970 it was
only 1 per cent of total expenditure (though there does appear to be a
discrepancy between the two different figures which Lojkine gives for
the total expenditure in 1970; see Lojkine, 1974a, pp. 32, 43). Lojkine
makes no comment on the underlying causes of this fluctuating pattern
of expenditure.
Turning to the expenditure of the General Council of the Rhone,
Lojkine notes that 65 per cent of this expenditure was devoted to 'social
aid' (ibid., p.45), (that is, education, training, health, social action,
sporting, cultural and administrative facilities), while most of the
remainder was devoted to roads and airports. As a result, very little of its
expenditure was devoted to public transport or housing.
Lojkine then discusses the expenditure pattern of the Urban Com-
munity of Lyon and states that this gives priority to roads and education
but not to public transport or housing (ibid., p.50). Unfortunately,
Lojkine provides no data giving an itemised breakdown of this
expenditure.
Finally, in Chapter 1, Lojkine briefly examines national state expen-
diture in Lyon and concludes that this expenditure is primarily directed
at infrastructural development. In other words it is primarily directed at
the means of communication (roads, motorways, airports) rather than

the means of collective consumption (ibid., p. 56).
Lojkine's analysis in Chapter I leaves little doubt that public trans-
port and housing receive a small share of total state expenditure. While
the data is not always complete, nor always presented in the most
systematic manner, this central point still forcefully emerges. However,
it should be emphasised that this is the only solid fact to emerge from the
first chapter. This chapter is certainly not a comprehensive analysis of
state expenditure in Lyon (as its title suggests), nor indeed a comprehen-
sive analysis of state expenditure on public transport and housing in
Lyon.
In Chapter 2 (ibid., pp. 61-137) Lojkine discusses the position of the
different fractions of capital in Lyon (commercial capital, banking
capital, industrial capital) and the way in which they have been changing
in recent years.
The first class fraction to be discussed is commercial capital. In this
section (ibid., pp.61-82) Lojkine puts forward the view that, in the
Lyon region, the commercial class is undergoing 'the classical process of
centralization and concentration of capital' (ibid., p. 71). This, in effect,
means that small shopkeepers are being replaced by larger more
monopolistic commercial enterprises. The evidence presented in this
study (Lojkine, 1974a, pp. 67 -71) and elsewhere (Lojkine, 1972b,
pp. 159-61) strongly supports this interpretation.
Lojkine then examines banking capital and the relationship between
'urban policy and banking capital in the Lyon region' (ibid., p. 83). This
section (ibid., pp. 83-97) describes the efforts of the central, regional
and urban state authorities to create, in the Part-Dieu area of Lyon, a
commercial and financial centre of major national and international
importance. Lojkine's main argument is that at least one aspect of this
policy is doomed to failure in the sense that Lyon can never become a
major national or international centre for finance. This is so for three
reasons. The first is that, in Lyon, most of the insurance companies
(notably the Union des Assurances de Paris) and the banking groups
(notably Credit Lyonnais and Banque Nationale de Paris) are sub-
sidiaries of companies which are controlled from Paris. The second
reason is that most of these groups and companies are currently
reorganising to take advantage of the revolution in information
technology associated with the computer and the microchip. This
reorganisation has taken the form of separating purely routine adminis-
trative activities (such as compiling and issuing bank statements,
processing claims, and so on) from purely commercial activities (which
involve dealing directly with clients) and, in tum, the separation of these
two functions from those of control and decision-making. This reorgan-
isation has resulted in a highly-decentralised commercial and adminis-
trative operation while at the same time it is highly centralised and
controlled from Paris. The third reason is that Paris is likely to remain
the major control and decision-making centre in the French financial
world because 'the major means of communication and exchange are to
be found in Paris and in Paris only' (ibid., p.97). For these reasons,
according to Lojkine, Lyon is unlikely to supplant the traditional
supremacy of Paris in the financial world.
The final section of Chapter 2 is concerned with industrial capital. The
major fact which Lojkine highlights here, is that large monopolistic
firms are either supplanting small and medium-sized firms (particularly
in the civil engineering field - see Lojkine, 1974a, pp.126-7) or
absorbing them through mergers (as with the merger between Berliet
and Michelin-Citroen - see ibid., pp. 134-6) or through employing them
as sub-contractors (ibid., pp. 102, 132). This growing dominance oflarge
monopolistic firms is, in tum, reflected in the urban and regional plans
of the state which is currently (that is, since 1970 - see ibid., p. 118)
making huge investments in order to provide and service large industrial
estates (over 1000 hectares in size) to attract multinational monopolistic
firms. Lojkine describes this 'subordination of regional political power
to large industrial capital' (ibid., p. 122) as 'a new phenomenon in
France' (ibid.) and is to be found not only in Lyon (where those in
political power are described as 'overtly of the right or camouflaged
under centrist labels' (ibid., p. 1l3) but throughout the entire area of the
'Grand Delta' which has become 'a pole of interest for monopolistic
groups' (ibid., p. 119).
Despite the radical differences in the economic interests of small and
medium industrial capital on the one hand, and monopoly capital on the
other, the Lyon Chamber of Industry and Commerce manages to
maintain a 'political alliance' (ibid., p. l33) between them. This is
exemplified, according to Lojkine, by Satolas airport which was
supported by the Chamber even though it appears to be in the interests
of small and medium, rather than of monopoly, capital (ibid.,
pp. 130-33).
In Chapter 3 (ibid., pp. l37-84) Lojkine looks at the relationship of
urban planning and policy to the working class. His discussion centres
around housing, education and public transport.
As regards housing, Lojkine points out that the strongest spatial
concentration of the working class is in the East and North of Lyon since
this is where there is the highest concentration of HLMs and grands

ensembles. These grands ensembles are located at the urban periphery
and many of them were built since 1960 to rehouse those (that is, the
traditional working class, but also artisans, shopkeepers and small
industrialists) who were 'deported' (ibid., p. 161) from the centre of
Lyon as a result of urban 'renewal' (ibid.) there.
One of the main reasons why working-class housing tends to be
located at the periphery rather than the centre is that the City of Lyon
leases its city centre land (which would be suitable for HLMs) to
developers who build offices and commercial premises on it. Lojkine
adds that the 70-year leasing arrangements granted by the City of Lyon
allow the capitalist tenant to appropriate differential rent I and II and
the City of Lyon to appropriate absolute rent (ibid, p. 142). Lojkine
states that this is only a 'hypothesis' (ibid.) which could nevertheless be
'demonstrated' (ibid.) by comparing the value of the absolute rent with
the value of differential rent I and II.
Turning to education, Lojkine points out that there is a very unequal
spatial distribution of educational facilities in Lyon since they are much
more heavily concentrated in the centre than in the suburbs (ibid.,
p. 164). However, he does not show that this unequal spatial distribution
of educational facilities has an independent effect on access to, and
success in, education so that this aspect of his analysis remains
incomplete.
As regards public transport, Lojkine argues that the basis of the
problem in Lyon is that public transport is governed and operated by
what he variously terms 'the same old logic of capitalist profit' (ibid.,
p. 169), 'the narrowly commercial optic' (ibid., p. 167), and 'the good old
marginalist logic' (ibid., p. 169). This policy, according to Lojkine, is
erroneous, because public transport is inherently unprofitable (though
the precise reasons for its unprofitability are not given). The adverse
consequences of the public transport policies in Lyon are that buses are
overcrowded; travel to the centre of Lyon from certain suburban areas
can require two buses; in certain suburban areas, bus stops can only be
reached after a walk of 15 minutes; there is, in certain cases, a lack of
synchronisation between some buses and trains (ibid., pp. 166-7).
Lojkine also discusses and disputes the 1966 Waldman report on the
metro system in Lyon (ibid., pp. 171ft). However, it is impossible to
assess this section of Lojkine's analysis since the various pieces of data
which are crucial to the debate do not appear to be strictly comparable
and, moreover, there is no detailed map showing the location of the
various places cited in the discussion.
These are the main themes of Lojkine's study of Lyon. It is a
descriptive and informative study of some of the more recent develop-

ments in that city, particularly those which have occurred since the
1960s. Although the title of the book states that it covers the period
1945 - 72, this is somewhat misleading since most of the data postdates
1960. However, even in this reduced time-span Lojkine's book covers a
wide gamut of topics - from state expenditure on housing and education
to developments in the world of banking and insurance. It is this breadth
of scope which is probably the main weakness of the book since it
contains a relatively small amount of data upon each of the various
topics. In fact, it is less rich in empirical detail than the companion
volume on Paris (see subsection 7.6.2). For this reason Lojkine's study of
Lyon is more a preliminary investigation than a definitive study.
Another feature of this study is that most of the major issues which are
raised in Lojkine's conceptual analysis (see sections 7.0 to 7.5) do not
appear here. The one exception to this is the concept of rent which is
briefly alluded to, although Lojkine treats it in a very confused manner.
He claims that the tenants ofland in the centre of Lyon (that is, the banks
and insurance companies) are able to appropriate differential rent I and
II while the landowner (that is, the municipal authority of Lyon) is able
to appropriate absolute rent. This claim is confused for three reasons.
First, tenants cannot appropriate differential rent I; they can only
appropriate differential rent II (see section 3.2). Secondly, the possibility
of the City of Lyon appropriating absolute rent would require that it is
the sole owner of all city centre land or that it acts in collusion with other
landowners to raise land prices (see section 3.3). Thirdly, the amount
and type of rent appropriated by the landowner and tenant in this case
could only be established according to the method outlined above (see
section 3.5), and not by comparing the size of the different rental
components, as Lojkine suggests.
In summary, Lojkine's book is a useful and informative investigation
of various developments in the City of Lyon since the 1960s. However, it
is very preliminary in nature, not only because of the paucity of data on
each topic, but also because many of the issues in Lojkine's conceptual
analysis are not treated here or are treated in a rather confused manner.
In short, too little data is spread over too wide a geographical, historical
and conceptual area.
7.6.4 Conclusion
The central theme running through Lojkine's two major pieces of

research is that the effect of urban development and redevelopment in
Paris and Lyon is to advance the interests of the monopolies against all
non-monopoly groups. The state presides over and facilitates this

process, thus 'confirming' the thesis of state monopoly capitalism that
the state protects and advances the dominance of the monopolies in the
current phase of capitalism. As has been seen, Lojkine presents some
evidence to support this thesis, though it is considerably less univocal
than his conclusions suggest. This suggests a more general difficulty with
Lojkine's research: that it is too broad in scope to allow detailed
empirical analysis of one particular issue.
In his more recent researches on urban development in Lille, Lojkine
has shifted his focus decisively to local politics, and is retrospectively
critical of his failure, particularly in the study of Lyon, to identify the
links between the economic effects of urban development on different
class fractions and the political consequences of those effects (see
Lojkine, 1980; 1981b, pp. 93-4). This new self-criticism may mean that
Lojkine's subsequent research will be free from some of the blemishes
that have been identified in this section.
7.7 CONCLUSION
It is now necessary to draw together the various findings which have

emerged from the analysis in this chapter. These findings relate to both
the conceptual and the empirical parts of Lojkine's work. The empirical
part of Lojkine's work has just been analysed (in section 7.6) and it was
found that his studies of Paris and Lyon are useful preliminary studies
which, however, suffer from a lack of detail on many of the important
topics. Moreover, there seemed to be a divorce between the themes
covered in this research and t'.e major issues in his conceptual analysis.
For example, the effect of the collective means of consumption on the
rate of profit, which is a central element in his conceptual analysis, is not
even considered in his empirical research.
Lojkine's conceptual analysis not only defines the field of urban
sociology but also outlines and analyses the main themes which, in his
view, fall within that field: collective consumption and its financing; the
locational behaviour of firms; urban land rent; urban planning and
urban social movements. It is now necessary to summarise briefly the
usefulness of Lojkine's analysis of these topics.
Lojkine defines an urban area as a spatial concentration of the general
conditions of production (section 7.1). He then selects one of these
conditions - the collective means of consumption - and claims that it is
one of the most distinctive characteristics of urban areas within
capitalism. In this way the collective means of consumption become

one of the central themes of Lojkine's urban sociology (section 7.2).
Lojkine's definition of the field of urban sociology is somewhat
narrow because it does not include all of the processes which take place
within urban areas. As was argued earlier (section 7.1), urban areas are
the locus of three major social processes: the process by which the urban
area is produced through property development; the process by which it
is used in the production of other commodities and the process by which
it is used for final consumption. This definition of the field of urban
sociology, which will be elaborated at length in the next part of this book
(Part III, Chapter 8) shows that the field of urban sociology can be
broader in scope, and richer in content, than that envisaged by Lojkine's
definition.
The collective means of consumption are a central theme in Lojkine's
writing although his analysis of them (section 7.2) was seen to contain a
number of errors. Apart from the ambiguities surrounding his use of the
word 'collective' (which was used in four different senses) his explana-
tion of the 'unproductive' and 'unprofitable' nature of the collective
means of consumption was shown to be faulty. In turn his analysis of the
problems relating to the financing of urban expenditure is also
untenable. The reason for this is that it relies upon Marx's law of the
tendency for the rate of profit to fall which has been shown to be invalid
(see section 2.1).
Lojkine's analysis of the locational behaviour of firms (subsection
7.3.3), is extremely brief and cursory and seemed to be uninformed by
the literature in their field. Similarly, his analysis of rent (subsection 7.3)
was based upon an inadequate and confused understanding of Marx's
concept of rent and of the limitations of that concept.
In his analysis of urban planning (section 7.4) Lojkine is particularly
anxious to show that the major effect of state intervention in this field is
to advance the interest of the monopolies against the interest of all other
non-monopoly groups. While considerable evidence is adduced to
support this claim, Lojkine's apparent preoccupation with it led him to
totally neglect a number of other important issues about urban planning
such as the actual process by which plans are produced and the effect of
the implementation of that plan, through development control, on
private property development.
Lojkine has not yet carried out any empirical research on social
movements although his conceptual analysis of this topic is both cogent
and informative (section 7.5). In particular, his focus upon the spatial
dimension of social movements highlights an important feature of these
movements which is neglected in other work, such as Castells's.

These are the main findings to emerge from an examination of
Lojkine's work. Most of the errors in that work can be traced directly to
Lojkine's lack of a clear and rigorous understanding of Marxian
political economy, and the latter's limitations, and to his lack of detailed
knowledge of the processes underlying urban planning and develop-
ment. For this reason his writing does not appear to be a significant
contribution to Marxist urban sociology. As a result, the value of
Lojkine's work, like that of Castells, lies essentially in its role as a
catalyst for other work in this field leading, possibly, to the development
of a more cogent analysis of urban areas. It is in this spirit that an
attempt is made to develop a new approach to urban sociology in the
next part of the book (Part III, Chapter 8).
Part III
An Alternative Approach to
Marxist Urban Sociology
8 From Marxist to Post-
Marxist Urban Sociology
8.0 INTRODUCTION
The purpose of this chapter is to make some innovations in the field of

Marxist urban sociology. These innovations will attempt to avoid some
of the difficulties which have been identified in the work of Manuel
Castells and Jean Lojkine, while at the same time addressing many of the
issues which are central to it.
The analysis begins with a synthesis of the results of the preceding
chapters in order to provide a summary assessment of the work of
Castells and Lojkine (section 8.1). In response to some of the problems
identified in their work an alternative model will be outlined (section 8.2)
and subsequently applied to the analysis of urban processes and conflicts
(section 8.4 to 8.8).
8.1 THE MARXIST URBAN SOCIOLOGY OF MANUEL

CASTELLS AND JEAN LOJKINE
8.1.1 Introduction
The writings of Castells and Lojkine represent a pioneering attempt to

develop a Marxist urban sociology. It is now necessary to bring together
the various strands of their work which have been examined in the two
preceding chapters in order to provide a more summary and definitive
assessment of it. This is essential before proceeding to examine the future
developments in this field. To this end the central themes in the writings
of Castells and Lojkine will be briefly summarised and assessed. These
themes are: the definition of an urban area (subsection 8.1.2); collective
consumption (subsection 8.1.3); urban planning (subsection 8.1.4) and
urban movements (subsection 8.1.5).
8.1.2 The Definition of an Urban Area Contained in the Writings of

Castells and Lojkine
Castells and Lojkine have two quite different definitions of an urban

area. For Castells an urban area is an area of collective consumption (see
185
186 An Alternative Approach
subsection 6.2.2), whereas for Lojkine it is an area where there is a spatial

concentration of the general conditions of production (see section 7.1).
Castells's definition is based upon his assumption that the urban area is
distinguished as a unit by its 'everydayness', that is, the reproduction of
labour power. Since reproduction, and hence consumption is, he argues,
increasingly organised collectively, he infers that an urban area is a place
of collective consumption - 'collective' being used as a synonym for
'state provision' (see Castells and Godard, 1974, pp. 228-9; Castells,
1976b, p.295; I 977a, pp.451, 460; 1977b, p.64, 1978a, pp.41-2);
1981b, pp. 8-9: 1983b, p. 9). In addition he claims that the boundary of
an urban area is coterminous with the boundary of the job market and is
fixed by commuting patterns (see Castells, 1977a, pp.234-6; 1976h,
pp. 180ff; Castells and Godard, 1974, pp. 174-92).
The main difficulty with Castells's definition is that it limits unneces-
sarily the range of issues which are considered to be 'urban' (see Harloe,
1977, p. 22; Pah1, 1977, p. 314; 1978; Pickvance, 1978, p. 175). Indeed the
issue of consumption itself is considerably richer, both conceptually and
empirically, than Castells's analysis suggests, as will be seen more fully in
section 8.8. Moreover, the way in which the urban area is itself produced
through property development, and the way in which it is used in the
production of other commodities, are processes which are also neglected
in Castells's framework (see section 8.4 to 8.7) although they would seem
to be essential concerns to Marxist urban sociology.
Castells's definition is open to the further objection that the boundary
of an urban area is not normally coterminous with the boundary of the
job or labour market: labour market areas can be equal to, greater than,
or less than, the boundary of an urban area, depending upon such
factors as the skill level of the labour in each market (see subsection 8.7.2
below). Thus Castells's definition of an urban area is difficult to sustain.
Lojkine's definition of an urban area is much more comprehensive
and convincing than that offered by Castells since it includes the main
physical elements which are to be found in any urban area (that is, those
things which are necessary for production and final consumption (see
Lojkine, 1976a, pp.120ff; I 977a, pp.125ft). However, like Castells,
Lojkine only considers those elements of an urban area which are either
owned privately (for example, factories, offices, and so on) or by the
state (for example, schools, hospitals, parks, and so on); neither of them
consider the category of communal goods such as air, the visual
environment, smoke, noise, congestion, and so on, which are also
important elements in an urban area and which may be important
sources of conflict within those areas (see subsection 8.8.6). In addition
From Marxist to Post-Marxist Urban Sociology 187
Lojkine's analysis focuses upon the processes by which urban areas are
used in capitalist production and in final consumption but neglects to
consider the process by which the urban area is itself produced through
property development (see section 7.1). Thus neither Castells nor
Lojkine offers entirely satisfactory definitions of an urban area or of the
field of urban sociology generally.
8.1.3 The Analysis of Collective Consumption in the Writings of Castells

and Lojkine
Castells and Lojkine attempt to explain the role and function of

collective consumption by using traditional Marxian concepts, most
notably Marx's theory of crisis (see Chapter 2) and productive and
unproductive labour (see Chapter 4). Although there are different
emphases in both writers, the law of the falling rate of profit is central to
the explanations of both. Both writers assume that the role and function
of collective consumption must be related to the law of the falling rate of
profit and, on this basis, Castells reaches the firm conclusion that the
function of collective consumption is to 'counteract' the tendency for the
rate of profit to fall (Castells, 1978a, p. 18; 1980a, pp. 51-2), 69-70; see
also subsection 6.2.3) whereas Lojkine reaches an equally firm but
opposite conclusion that the function of collective consumption is to
'strengthen the tendency for the rate of profit to fall' (Lojkine, 1976a,
pp. 131- 2; see also subsection 7.3.2). Since there is no firm evidence for
either conclusion, because of the difficulty of measuring and testing
Marx's law, neither claim is particularly useful. In both cases, however,
there appears to be an implicit assumption that the effects of collective
consumption in any capitalist country can be inferred from Marx's law
of the falling rate of profit. This particular example illustrates in striking
fashion how Castells and Lojkine tend to use Marxian concepts in a
dogmatic and functionalist manner. Their usage tends to be dogmatic
because they do not actually examine Marxian concepts to ensure that
they are internally consistent or empirically accurate. (The analysis in
section 2.1 reveals that this is particularly necessary in the case of Marx's
law of the falling rate of profit.) Their usage also tends to be functionalist
because they infer the effects of collective consumption by deduction
from the functions which it is supposed to play within a capitalist
society, given certain Marxian premises. The error in this procedure is
that the 'function' of collective consumption can only be inferred from
its effects and the latter in turn can only be established by empirical
inquiry which, in this case, would be extremely difficult to undertake.
Castells also offers two additional explanations of the role and

'function' of collective consumption. He claims, on the one hand, that
collective consumption alleviates the problem of deficiency of aggregate
demand because 'the economy of advanced capitalist societies rests
more and more on the process of consumption, i.e., the key problems are
located at the level of the realization of surplus value' (Castells, 1976d,
p.294; 1976h, p. 182; 1977c, p. 63; 1981b, p. 8.). He also claims, on the
other hand, that collective consumption is a response to the 'develop-
ment of the class struggle and the growing power of the workers'
movement that extends its bargaining power to all areas of societal life'
(Castells, 1975a, p. 176; 1976h, p. 184). These are major and important,
though unsubstantiated, claims about the nature of collective consump-
tion, and the corroboration of either would involve a major research
undertaking. In this, as in other contexts, Castells seems to be more
concerned with defining the perspective, and setting the agenda for
future research, rather than actually undertaking the research itself.
Lojkine too applies some additional Marxian concepts to clarify the
nature of collective consumption. He points out that collective consum-
ption is unprofitable (see subsection 7.2.4) as well as unproductive (see
subsection 7.2.3). He explains that collective consumption is unprofita-
ble to produce because investment in such things as houses, schools,
universities, and hospitals 'lasts for many years' (Lojkine, 1977a, p. 138).
As was seen above, the difficulty with this argument is that many
profitable investments in the private sector, such as equipment for the
extraction and refining of oil, the assembly of cars, the production of
chemicals, and so on, also 'last for many years'. In addition, some forms
of collective consumption are in fact produced profitably in the private
sector in many capitalist countries (for example, private medicine and
private education). These considerations would seem to indicate that,
whatever the real reason for the unprofitability of collective consump-
tion, it probably does not lie, like Lojkine claims, in the difficulties
associated with the supply of these goods in a capitalist system.
Lojkine also claims that collective consumption is 'unproductive'.
However, he does not seem to recognise that there are two quite
different, and inconsistent, usages of the terms productive and un-
productive in Marx's work (see Chapter 4). As a result, his claim is
inherently ambiguous. In this, as in other cases (most notably in his
discussion of rent - see subsection 7.3.4), Lojkine's analysis, like that of
Castells, tends to be highly conjectural and speculative. Although both
draw heavily upon concepts from Marx's political economy, neither
gives serious consideration to the methodological problems involved in
testing those concepts. As a result, the value of their analysis of collective

consumption lies more in suggesting possible applications of Marxian
political economy that in its rigorous application.
8.1.4 The Analysis of Urban Planning in the Writings of Castells and

Lojkine
Castells and Lojkine, despite their similarities, represent quite different

traditions within the Marxist paradigm (see Chapter 5) and this
difference becomes particularly apparent in their discussions of urban
planning. Castells's writing is part of the tradition known as Althus-
serianism (see section 5.1) while Lojkine's is in the tradition known as
state monopoly capitalism (see section 5.2).
For Castells, urban planning is a form of state intervention whose
effect is to regulate the contradictions generated by capitalist urban
development (see subsection 6.5.1). In Althusserian terminology, it is an
intervention of the political system on the economic system. Lojkine, by
contrast, argues that urban planning reflects rather than regulates the
contradictions of capitalist urban development and that, in the current
stage of capitalism (that is, state monopoly capitalism), its effects are to
promote the interests of monopolistic groups against those of non-
monopoly groups (see section 7.4).
Castells and Lojkine differ, not only in terms of their definition of
urban planning, but also in terms of the problems which they identify as
necessitating it. For Castells, urban planning is necessitated by the
contradiction between the private nature of production and the
collective nature of reproduction (see subsection 6.5.3), whereas for
Lojkine it is necessitated by three contradictions associated with urban
development in capitalist societies, namely, the financing of urban
expenditure, the locational strategy of firms, and urban land rent (see
sections 7.3 and 7.4). The analysis of these various problems leads
Castells and Lojkine to address some of the central and most important
issues in Marxian political economy, yet, as was seen above (subsection
6.5.3 and sections 7.3 and 7.4) this is one of the weakest aspects of their
work, essentially because they do not consider either the internal validity
of these concepts or the problem of operationalising them empirically.
Despite the different emphases in their work, both Castells and
Lojkine treat urban planning as part of the more general process of state
intervention. While this approach has the advantage of focusing upon
those features which urban planning shares with other forms of state
intervention, it has the disadvantage of concealing important internal
differences within the category of urban planning. As a result, their

analysis tends to exclude any explicit consideration of such issues as
zoning, development control, compensation, the internal and external
constraints upon planners in making and implementing plans, the effect
of land-use planning upon private property rights and property
development, and so on. None of these issues are investigated in their
writings, yet together they constitute one type of state intervention,
namely, land-use planning, which has an important effect upon the
urban environment and, moreover, an effect which is probably different
from other types of state intervention. In this sense it would appear that
there are some disadvantages with Castells's and Lojkine's definition of
urban planning.
8.1.5 The Analysis of Urban Movements in the Writings of Castells and

Lojkine
Castells and Lojkine are totally opposed to the capitalist system in all its
forms and share a common commitment to some form of socialism. It is
this fact, possibly more than any other, which accounts for the style and
content of their writing. It accounts, in particular, for the approach
which they adopt towards urban movements.
Castells and Lojkine both define an urban social movement as being
concerned with 'true' and 'radical' change which will 'open the road to
socialism' (see subsection 6.6.2 and section 7.5). They regard neighbour-
hood and community protests (such as those concerned with the
provision of public transport, schools or housing) as unimportant, both
analytically and politically, as long as they remain within the horizon of
that particular neighbourhood or community. Thus both Castells and
Lojkine are, in principle, interested mainly in movements that have,
actually or potentially, a national and not just a purely local dimension.
This definition was criticised (see subsection 6.6.2 and section 7.5) for
being too narrow, because it excluded from consideration many of the
protest actions which typically occur in urban areas (for example,
protest actions over such issues as rents, repairs to local authority
houses, planning proposals, crime, and so on). Although many of these
movements could scarcely be described as 'radical', they are nevertheless
a vital part of the political interaction that takes place between the local
or central state and the residents in that jurisdiction. Any definition
which wishes to exclude these neighbourhood and community-based
movements from the general category of urban movement is likely to
miss an important part of the politics that take place within urban areas.
In his research Castells was unable to find any 'real' urban social
movements (as he would define them) and was obliged, by default, to
focus upon protests over such issues as housing and redevelopment
whose horizons tended to be local rather than national (see notably
subsections 6.7.4, 6.7.5 and 6.7.8).
Given their definition of an urban movement, Castells and Lojkine
place considerable emphasis upon the fusion of struggles over collective
consumption with struggles in the workplace. Indeed, Lojkine has
expressed the belief that movements involving this fusion are likely to be
of growing importance in the coming decades (Lojkine, 1977a, pp. 334ff;
1977b, p. 153). While this may turn out to be the case, it is likely to occur
in spite of, rather than because of, a growth in protests over collective
consumption. This is because the latter tend to be limited to the spatial
confines of a particular externality field (see subsections 8.8.3, 8.8.6 and
8.8.9). In Britain the evidence would seem to disprove Lojkine's
expectation since community movements, according to Saunders, 'are
typically fragmented, localized, limited to a narrow range of concerns,
and politically isolated from broader radical movements' (Saunders,
1980b, p. 551).
Although the concept of urban social movement contained in the
work of CastelIs and Lojkine is excessively narrow, they nevertheless put
forward a useful framework of concepts for the description and analysis
of social movements. They argue that a social movement can best be
understood by identifying: the issue or stake of the movement (that is,
the pro blem affecting, or goal adopted by, a particular group), the social
base (that is, the population affected by the particular problem or issue),
the organisation (that is, the vehicle through which the social base
achieves its goal), the social force (that is, that part of the social base
which is mobilised into action through the organisation and the effect
(that is, the outcome of the action pursued by the social force). These
concepts and the methodological procedure underlying them are likely
to prove useful in characterising and comparing different protest
movements.
8.1.6 Conclusion
This review of the work of Castells and Lojkine suggests that their
attempt to develop a Marxist analysis of urban areas must be regarded,
despite its occasional insights, as having rather limited value. The
evidence supporting this claim was presented systematically and exhaus-
tively in Chapters 6 and 7 and in the preceding subsections of this
chapter (8.1.1 to 8.1.5). This evidence suggests that, despite the centrality
of Marxian political economy in their work, they rarely use its concepts
in any methodologically rigorous or creative way. Both tend to use
Marxian concepts in a rather dogmatic and functionalist way. In
consequence, they fail to make some of the modifications and innova-
tions to the Marxian framework which, as will be seen presently (in
section 8.2), seem to be necessary in order to adapt it to the spatial and
often highly localised nature of class and class conflict at the urban level.
Nevertheless, their work is significant because of its role in generating
interest and in helping to define the research agenda for the growing field
of Marxist urban sociology. It is in this context that the subsequent
sections of this chapter (8.2 to 8.9) will attempt to develop a more
empirically-grounded, if less strictly Marxist, analysis of some of the
central issues in this new field of urban sociology.
8.2 A CONFLICT APPROACH TO URBAN SOCIOLOGY: THE

CONSTRAINED-ACTOR MODEL
Marxist political economy, apart from certain obiter dicta about the
separation of town and country (Marx, 1947c, p. 333) is virtually non-
spatial and this immediately creates a problem when applying it to the
urban context. In addition, there is the problem that Marx's analysis of
class and class conflict tends to be at a high level of abstraction,
particularly in Capital, whereas class conflict in urban areas tends to be
more locality-based. Thus, notwithstanding the lack of methodological
rigour with which Castells and Lojkine use Marxist political economy,
the latter itself seems to require some innovations in order to make it
more amenable to the analysis of urban issues (for example, issues such
as the conflicts associated with negative externalities in residential areas
or the conflicts between developers and planners over property develop-
ment).
The analysis developed here derives its basic thrust from the Marxian
emphasis upon the role of interests, both material and ideal, in human
action, and upon the conflicts of interest which usually ensue from the
pursuit of those interests. However, unlike the standard Marxian
approach, where classes are treated as the main actors or agents in
society, the concept of actor will be broadened here to include
individuals, groups and organisations which have a separate and
identifiable interest and which act purposively to pursue that interest. 1
This broader concept of interest reflects the hierarchical levels of
abstraction at which the concept of interest can be treated so that the
approach developed here represents a change of levels rather than a

change in the basic perspective. Nevertheless, the focus upon actors,
whether they are individuals, groups or organisations, is not strictly
Marxist, and is probably closer to Weber. For this reason the term
'constrained-actor' will be used as a label of convenience for the model
developed and applied here.
The constrained-actor model focuses upon instrumentally rational,
self-interested action to the exclusion of other types of human action,
notably, altruistic, affective and traditional action. The reason for this is
that it is assumed that instrumentally rational self-interested action is the
most salient type of action in the various urban contexts discussed below
(while acknowledging that other types of motivations are likely to
operate in other urban contexts). This view of society which is
complementary to, though less deterministic than, the Marxian view,
can be stated more systematically in terms of four basic propositions.
The first proposition is that an actor is defined by the particular
interest which it pursues. Those interests may be material (for example
profits, wages, subventions from the state, neighbourhood facilities, and
so on) or ideal (for example, political and ideological influence and
legitimacy). In turn an actor may be a single individual (for example, a
capitalist, worker, landowner, resident, and so on) or an organisation
(for example, a union of employers or workers, the central or local state,
a resident's association, and so on). It follows from this that a
sociological analysis of any social process must begin by identifying the
actors involved, and the interests which they are pursuing. This typically
was Marx's procedure, for whom the three main actors were capitalists,
workers and landowners, each defined by their respective interest,
namely, profits, wages and rents.
The second proposition is that the material interests of actors are
determined by their legally defined rights over resources (for example,
capital, labour, land, subventions from the state, houses, and so on) and
the pursuit of their self-interest typically involves maximising the gains
from those resources within the limits set by the law and upheld by the
state (see Saunders, 1980a, pp.45-8). This, it will be recalled, is the
fundamental basis of Marx's theory of class: capitalists have property
rights over capital, workers have property rights over labour and
landowners have property rights over land. However, the constrained-
actor model, unlike the strict Marxian model, allows for the fact that the
interests of an individual organisation are also determined by their
legally defined property rights over items of final consumption (for
example, dwellings).
The interests of the state (that is, its policies and goals) are more
difficult to define than those of other actors, and can vary within and
between capitalist countries, depending upon such factors as its
organisational structure, the values and attitudes of its incumbents and
the relative power of the various classes and interest groups in that
society. These factors, in turn, are reflected in the law which defines the
scope of state power as well as the manner in which the power can be
exercised.
The third proposition is that actors in advanced capitalist societies
tend to act in an instrumentally rational way and hence to calculate the
efficiency of the means used to attain their self-interest. This, as Weber
observed, is part of the process of 'rationalisation' of western societies
(Weber, 1974, pp. 13-31), an assumption that is also implicit in Marx's
Capital (Marx, 1974a, 1974b, 1974c).
The fourth and final proposition is that the pursuit of self-interest
takes place through interaction with other actors, equally in pursuit of
their self-interest, so that the interests of one actor are typically
constrained and limited by the interests of another. In this way conflicts
of interest arise which are a characteristic feature of capitalist societies.
The relationship between workers and capitalists, so exhaustively
analysed by Marx, is a perfect example of the operation of these
constraints: workers' interest in higher wages (for example) are contin-
ually constrained by capitalists' interest in higher profits and vice versa.
There are, however, many other types of conflict, notably conflicts in the
sphere of final consumption, which were not investigated by Marx. As
will be seen below, these can be adequately analysed within the
framework of the constrained-actor model outlined here.
The analysis below focuses primarily upon the micro-level conflicts
which take place in the urban context (for example, conflicts associated
with property development, or over negative externalities in residential
areas) although it is not intended to exclude consideration of the more
macro-level issues which are central to Marxist analysis generally. In
other words, the analysis here leaves open the question of the connection
between classes and various interest groups such as residents' associa-
tion, environmental lobbies, local planning authorities, landowners,
developers, and so on. Thus, although the approach developed here does
not solve all of the problems identified in the writings of Castells and
Lojkine, it does have the advantage of using concepts that can be
operationalised more easily while aware of the need to subsequently
relate those concepts to more abstract levels of analysis (see Pickvance,
1984, especially pp. 43-7; see also Castells, 1983a and 1983b where the
latter seems to be moving in this direction).
Before applying the constrained-actor model to urban analysis

proper, however, it is necessary to define as precisely as possible the
nature of urban areas, and the urban processes which typically take
place within them. This will be done in the next section.
8.3 URBAN AREAS AND URBAN PROCESSES: THE FIELD OF

URBAN SOCIOLOGY
An urban area is a built-up area consisting of physical structures, that is,

buildings (for example, houses, offices, factories, schools, hospitals,
churches, theatres, and so on) and infrastructures (for example, roads,
railways, water and sewerage systems, electricity, telecommunications,
parks, streets, and so on). It is, in other words, a spatial concentration of
land-using, humanly-created property.2 (A similar stress upon the built
environment is also to be found in some ofthe writings of Harvey - 1976,
republished in 1978a; 1978b, republished in 1981.)
Urban areas are essentially dynamic in the sense that the size and
composition of their stock of property typically changes over time. Some
of the existing stock of property becomes obsolete and is replaced while
additional new buildings and infrastructures are produced at the edges
of the built-up area. In addition, the use to which property is put can
itself change over time (for example, from residential to office) as can the
social class of the users (for example, from lower to middle income group
through gen trifica tion).
Urban areas are typically produced by private developers using
private and state capital. In turn, they are used for both producing other
commodities and for final consumption. Urban sociology, as defined
here, is essentially concerned with understanding how the property
concentrated in an urban area is produced and used. To be more precise,
it is concerned with understanding three processes which together
constitute an urban area. The first is the process by which an urban area
is produced through private and state property development. The
second is the process by which an urban area is used to produce goods
and services as a result of the location decisions of firms and agencies
and the migration and commuting decisions of workers. The third is the
process by which an urban area is used for final consumption through
the use of houses, schools, parks, theatres, museums, and so on.
This formulation of the field of urban sociology is broader and more
comprehensive than that found in the writings of either Castells or
Lojkine. As will be seen in sections 8.4 to 8.8, this formulation not only
incorporates the main issues which are to be found in their work (that is,
collective consumption, industrial location, urban planning, urban
social movements) but also includes a number of other issues (notably
property development and the spatial nature of final consumption)
which are not found in their writings but which are nevertheless an
integral part of the processes taking place within urban areas.
The three processes constitutive of an urban area will now be analysed
using the constrained-actor model. The methodological procedure in
each case will be to identify the various actors involved in each process,
the interests which they are pursuing, and the constraints which they
encounter in trying to achieve them. The analysis begins with an
investigation ofthe process of property development (sections 8.4 to 8.6)
followed by the process by which urban areas are used in production
(section 8.7) and in final consumption (section 8.8).
8.4 THE PRODUCTION OF URBAN AREAS: THE PROCESS

OF PROPERTY DEVELOPMENT
The majority of buildings and infrastructure which constitute the

physical fabric of an urban area in advanced capitalist societies is
produced by capitalists and the state through the process of property
development. Property development can arise in a number of ways:
sometimes it is a direct response to the demand for buildings and
infrastructures; at other times it is an indirect response, in anticipation of
future demand; at yet other times, it is a response to a demand which is
itself created by property developers. Thus property development is a
major force affecting the quantity, type and location of buildings and
infrastructures in an urban area.
A number of different actors or agents are involved in the process of
property development. These include the private property developer
whose role typically involves organising the finance, assembling the sites
and contracting the construction company; the various state agencies
which hire development companies to carry out property development;
the financial institutions which provide the finance capital; landowners,
who provide the sites; the state land-use planning authority whose
permission is necessary for most development to occur; various
professional agents such as architects, engineers, surveyors, planners,
and so on; and finally the construction companies who actually built the
physical structures. These various actors, each of which operates within
a particular set of constraints or 'structures', represent operationally
distinct parts of the development process.
Finance capital
State interacts Landowners

property ~
developer with
to Buildings
Professional agents . , and
produce infrastructure
Private interacts State planning authority

property .,
developer with
Construction companies
and allied trades
Figure 8.1 The process of property development
The relationship of these different actors to the over-all process of

property development is outlined, in highly simplified form, in Figure 8.1
(see Lorimer, 1972, p.19 and Ambrose, 1977, p.42, for a slightly
different diagram of the same process).
It is now necessary to analyse separately the process of state property
development (section 8.5) and the process of private property develop-
ment (section 8.6). This analysis is very similar to that of Ambrose and
Colenutt whose book, The Property Machine, provides a detailed
empirical analysis of the activities of the various actors involved in
property development in Britain and its resulting costs and benefits
(1975; see also Ambrose, 1977). Lorimer (1972) has done a similar
analysis for Canada. The analysis of Lamarche (1976), by contrast, is
primarily concerned with the nature of the income derived from
property development rather than with the process itself and is thus less
relevant in this context.
8.5 STATE PROPERTY DEVELOPMENT
The state is probably the largest property developer in most advanced

capitalist countries. Its property developments include the production of
facilities (that is, buildings and infrastructure) for transport (roads,
railways, airports, and so on), energy (electricity, gas, and so on), waste
disposal (sewerage system, water, dumps, and so on), communications

(telephone, satellites, and so on), defence (missile sites, defence bases,
and so on), education (schools, universities, museums, galleries, and so
on), health (hospitals, clinics, asylums, and so on), shelter (houses, flats,
hostels) and recreation (parks, swimming pools, and so on).
These facilities are essential parts of production and final consump-
tion in most advanced capitalist countries. Most of them are used
simultaneously for both production and final consumption and hence
are both wage and non-wage goods (see Offe and Ronge, 1975, p. 144). It
is thus impossible to quantify precisely the proportions of state
expenditure which are devoted to further production and to final
consumption.
State property development, at least in Britain, occurs through a
variety of institutional arrangements. Some developments (for example,
housing) are produced by the local state; some (for example, motorways)
by the central state; some (for example, water and sewerage treatment)
by regional state bodies; and some (for example, railways) by national-
ised industries. Thus the over-all process is characterised by considera-
ble organisational complexity.
In order to understand the nature of state property development, it is
necessary, following the methodological principles of the constrained-
actor model, to identify the interests (that is, the goals and policies)
which the state is pursuing in undertaking property development, and
the constraints which it encounters in attempting to achieve them.
However, unlike capitalists and workers, whose interests can be
reasonably approximated by reference to their perceived material
interests (that is, maximisation of profits and wages), the interests of the
state in a capitalist society can vary considerably, depending upon its
organisational structure, the values and goals of its incumbents and their
electoral support, as well as the relations of power and influence
connecting it with the different classes and interest groups in that society
(see Saunders, I 980a, pp.201-8ff; Dunleavy, 1980, Chapters 4-5;
Clark and Dear, 1981; Cox, 1983). It follows from this that the precise
interests and goals which the state pursues in a capitalist society cannot
be affirmed a priori; they can only be known through empirical in-
vestigation. Nevertheless, following the lead of such writers as Offe and
Ronge (1975), it is possible, by making some simplifying assumptions, to
identify some of the constraints which typically limit the state's freedom
of action in a capitalist society, and thereby limit its freedom of action in
relation to property development.
The logic of the analysis of Offe and Ronge is that the state, in a
society of capitalist commodity production, will tend to be constrained

to protect the interests of capitalists (or certain sectors of them) above all
other classes, since it is their investment decisions which result in profits
(and hence satisfy capitalists' interests), employment (and hence satisfy,
to some extent, workers' interests) and state revenue (and hence satisfy, to
some extent, state interests). Thus the protecting of capitalists' interests
tends to be the logical outcome of the state's position vis-ii-vis the different
interests in a capitalist society. In practical terms, this protection
frequently takes the form of legislation, grants, trade protection, tax
reliefs, and so on, as well as the provision of physical infrastructure
(roads, telecommunications, energy, water, and so on) which are
favourable to the interests of capitalists. It is in this sense that the state in
a capitalist society is a capitalist state (see Offe and Ronge, 1975,
pp. 139ft). However, in most advanced capitalist countries, the state has
also found it necessary, in pursuance of its interest, to increasingly
protect the interests of workers (employed and unemployed as well as
the old, sick and the young) through social welfare payments and other
facilities (schools, hospitals, houses, parks, and so on).
It follows from this that certain state activities involve the expenditure
of relatively large amounts of state revenue. Since the latter normally
depends upon taxation (or, more rarely, on the profitability of state
enterprises), state expenditure may be constrained fiscally because of
resistance to increased taxes or because the state - congruent with, and
probably influenced by, certain class interests - acts to maintain low tax
rates. In such situations, fiscal constraints may result in certain types of
state expenditure being constrained more than others, depending upon
the configuration of class forces involved. Thus, for example, Saunders
found that in Croydon, when the local state was faced with limited
revenue in 1973, it reduced expenditure on health, education and social
services while increasing expenditure on roads and carparks (see
Saunders, 1980a, p. 303). In this example, those aspects of state property
development which directly benefited business interests were privileged
vis-ii-vis those which would have benefited working-class interests (see
also O'Connor, 1973, and Habermas, 1976, for a more general
discussion of the fiscal constraints on state expenditure).
These considerations suggest that state property development should
be understood, not only by reference to the values and relations which
are internal to the state but also in terms of its external relations to the
various classes and interests in the society. While this methodological
principle is no substitute for concrete, empirical analysis, it does provide
a framework for understanding the 'interests' of the state in relation to
property development and the constraints which it encounters III

attempting to achieve them.
8.6 PRIVATE PROPERTY DEVELOPMENT
8.6.1 Introduction
The three principal actors involved in the private property development

process (see Figure 8.1) are the developer, the landowner and the state
planning authority and, for simplicity, the analysis in this section will
focus primarily upon these three actors. The interest of the private
property developer is in making a profit by producing either buildings or
infrastructures, while the interest of the landowner is in selling or leasing
land for the highest obtainable price. The interests ofthe state planning
authority, by contrast, can vary considerably, due to the forces and
factors identified in the preceding section (8.5). Thus the state planning
authority in a capitalist society can have a variety of interests or goals,
such as: attracting jobs to their area, maintaining low property tax rates,
promoting civic pride and tourism, conserving its architectural heritage,
preserving green spaces, and so on. This variety of goals was exemplified
in Croydon where Saunders found that the interest of the planning
authority, and of the local government generally, involved maintaining
a low property tax rate, preserving the green belt and ensuring low
housing densities (all of which were congruent with the interests of
middle-class owner occupier), as well as in promoting civic pride and free
enterprise (which were congruent with city centre business interests) see
Saunders 1980a, pp. 244, 307). Despite this variety of goals and interests,
local authorities usually face constraints from other actors in the process
of private property development and, using examples from Britain, these
will be discussed in subsection 8.6.4.
Other actors, too, notably state agencies, environmental groups and
final consumers, can also be sources of constraint and conflict in the
development process. These various constraints are summarised in
matrix form in Table 8.1. The constraints operating upon private
developers, landowners and the state planning authority will now be
analysed respectively.
8.6.2 Constraints upon the Interests of Private Property Developers
Table 8.1 reveals that the interests of private property developers in a

capitalist society can be constrained in six different ways. First of all,
they can be constrained by competition from other property developers,

which ensures that their rate of profit is maintained at a competitive
level. This constraint may, however, be reduced in the case of larger
developers where the number of actual and potential competitors is
smaller and, correspondingly, the possibility of monopolistic or
oligopolistic profits is greater.
Secondly, private property developers can be constrained by landow-
ners, both private and state. This can occur in a number of ways. It can
occur if private landowners refuse to sell their land except at an
extortionate price, that is, above the price set in a competitive market.
This possibility could arise if the developer is in the process of
assembling a number of different sites, each of which is essential to a
development project, but is hindered by one landowner who demands a
price which reflects the redeveloped rather than the existing value of the
site. This is known as 'price gouging' (Davis and Whinston, 1961, p. 111;
see also Elkin, 1974, pp. 71-2).
Private property developers can also be constrained by state landow-
ners who may refuse to sell suitable development land. This occurred in
the City of Nottingham betwen 1979 and 1980 where 'more than two-
thirds of the potential residential sites in the inner city were found to be
in the ownership of the Nottingham City Council' (Nicholls, et at., 1982,
p. 332) but the 'Labour Council adopted a policy of no disposal ofland
for private housebuilding' (ibid., p. 339). Alternatively, private property
developers may be constrained by the collective actions of landowners
who are opposed to a particular type of development within their area
(for example, owner-occupier residents opposed to an office or factory
development in their area). Such collective action could cause a delay or
even a cancellation of the proposed development and hence a thwarting
of the developer's interests.
Thirdly, the developer is constrained by the state planning authority.
This occurs because the planning authority, through its powers of
zoning and development control, effectively controls the supply of
development land; in Britain this has occasionally led to scarcities in the
supply of 'suitable' development land (see Drewett, 1973, p. 177;
Goodchild, 1978, pp. 13 -14). 3 In addition, the developer is also con-
strained by the costs of making a planning application (for example,
ascertaining the requirements likely to prove acceptable), the delay in
obtaining a decision (see Drewett, 1973, pp. 177-8; Neilson, 1976,
p. 161; Goodchild, 1978, pp.17-l8) and, in the event of permission
being granted, the conditions which may be attached to it. Alternatively,
the developer may be refused planning permission as occurred in 14 per
cent of all planning applications in England and Wales in 1978/79
N
o
N
Table 8.1 The constraints and conflicts generated in the process of private property development. Sources of constraint on actors
in the development process
Private Environmental
property State planning Other state and other
developer Landowner authority agencies groups Consumers
Private (1) Compe- (I) Refusal to (1) Restriction on (I) Price (I) Cause The level of
property tition keeps sell land supply ofland controls delay or effective
developer profits in (2) Collective (2) Cost of (2) Taxes on refusal of demand
check action by application profits planning
adjacent (3) Delays permission
landowners(4) Costly (2) Cause
and users conditions amenities
(5) Possible to be
refusal of preserved
permission or
improved
Actors in the Landowners (1) Develop- (I) Land (1) Refusal of (1) Taxation of(I) Prevail The leve1 of
development mentmay values may planning land upon effective
process reduce the be reduced permission to planning demand
value of by adjacent use land in the authority
adjacent uses ofland most to refuse
land and profitable way permission
property (2) 'Planning for the
blight' most
profitable
use
The state (I) Ifprivate (I) Compen- (l) Exclusionary (l) A lack of (I) Resistance (I) Actual and
planning land is sation to zoning and co-ordin- to planning anticipated
authority zoned for landowners fiscal policies ation decisions political
unprofit- (2) Collective can adversely between the which behaviour
able uses, action by affect other planning could cause of the
developers landowners jurisdictions authority harm to the electorate
are unlikely (2) Competition and other envlron- (e.g. voting,
to develop in attracting state ment protests,
it jobs agencies etc.)
IV
ow
(Department of the Environment, 1980, Table 1). In addition, the

developer of offices which are above a certain size, and in certain areas,
must obtain an Office Development Permit (ODP) from the Department
of the Environment, which may be refused (although as Pickvance,
1981, pp. 274ff, has shown, the rate of refusal ofODP's has been quite at
variance with its stated policy goals). The net effect of these constraints is
to increase the costs incurred by the developer, whose development, in
tum, imposes costs and benefits on others (see Ambrose and Colenutt,
1975, pp.60-67, 159-62; Balchin and Kieve, 1982, pp.1l8-19).
The fourth constraint is that posed by other state agencies such as the
central government. This can arise through the taxation of developers'
profits. It may also arise ifthere are price controls upon the sale of new
buildings such as houses or offices or rent restrictions on leasing.
Fifthly, the developer may be constrained by environmental groups
whose actions to protect amenities may cause the developer's planning
application to be delayed or even refused. In addition, they may cause
the developer, via the planning authority, to incur extra costs in
preserving or improving the amenities affected by their development.
Finally, the developer is constrained by the level of effective demand.
In the case of housing demand, this constraint is affected by the level of
wages and employment, the availability ofloans and the rate of interest
as well as subventions from the state to house purchasers. In the case of
factories, offices, shops, and so on, effective demand is affected by the
possibility of using these properties profitably, and this depends upon
such factors as wage levels, interest rates, competition, government
regulations, and so on. In this sense, the level of effective demand is the
ultimate constraint upon the interests of the private property developer.
8.6.3 Constraints upon the Interests of Landowners
Table 8.1 also reveals how the interests of landowners may be

constrained in the development process. This can also occur, as in the
case of the private property developer, in six different ways. First, the
interest oflandowners may be constrained by a property developer if the
latter's development has the effect of reducing the value of land and
property in its immediate vicinity. A possible example of this would be
the development of a factory, an airport, a motorway or a dump close to
a residential area. Conversely, a development may increase land values
in its vicinity in which case the interest of the landowner would be
advanced simultaneously with that of the developer.
The second constraint comes from other owners and users ofland and
derives from the fact that the value of urban land and property depends,
not only upon the quality and maintenance of adjacent properties, but
also upon the use to which those adjacent properties are put, as well as
the status of their users. Thus, if there is prejudice against a particular
status group (for example, an ethnic minority), the presence of the latter
in a particular area can, ceteris paribus, have the effect of reducing the
value of adjacent land and property.
The third constraint comes from the state planning authority which
has the right and the power to determine the use to which a landowner
may put his land in an urban area. This effectively means that the state
may inhibit a landowner from using his land in the most profitable way
so that the latter suffers a loss as a result. Additionally the planning
authority may have long-term plans for an area (for example, road
proposals) which may cause land and property values in an area to fall as
soon as the plans become public knowledge. Moreover, they may
continue to fall even if the plans are subsequently withdrawn, as long as
any suspicion remains that the plans may be implemented. This
phenomenon is known as 'planning blight'.
Fourthly, the interest oflandowners may also be constrained by other
agencies of the state. This may take the form ofa central government tax
on land whose value appreciates as a result of planning or development.
In Britain the development land tax operated in this way until its
abolition in 1985 (although in a much weaker form than was originally
envisaged) (see Balchin and Kieve, 1982, pp. 135-8; Cullingworth,
1982, pp. 88-90,97-101).
Fifthly, environmental groups may constrain the interests oflandow-
ners by prevailing upon the planning authority to refuse permission to
landowners to develop their land in a profitable but environmentally
damaging way (for example, housing developments on scenic mountain-
sides).
Finally, the interests of landowners are affected by the level of
effective demand from consumers (that is, consumers of offices,
factories, houses, and so on) and hence by the same factors which affect
the private property developer.
8.6.4 Constraints upon the Interests of the State Planning Authority
Table 8.1 also reveals the different constraints which affect the interests
of the state planning authority. The first of these comes from the private
property developer and derives from the fact that developers are unlikely
to undertake a property development in an area if it is zoned for uses
which they regard as unprofitable. This constraint is likely to vary

considerably between countries and to be greatest in those capitalist
countries where virtually all development is undertaken by the private
sector. In these circumstances, planners must zone land in a way which
broadly accords with the expectations of developers or face the prospect
of having no development at all. Planning, in such situations, according
to Pickvance (1977c), becomes 'trend planning' (ibid., p. 42) as distinct
from 'interventive planning' (ibid.), because 'the development plan
merely reflects market trends in the allocation ofland' (ibid.). This, he
claims, is the predominant form of planning in Britain since 1951 largely
due to the fact that 'private sector developers carried out the majority of
developments' (ibid., p.43).
The second constraint is that posed by landowners and varies from
one country to another according to the laws of compensation. The
planning authority is likely to be highly constrained in the type of zoning
which it can enforce if, for example, the laws of compensation ensure
that landowners are compensated for any zoning which reduces the
value of their land to below the maximum market value. In this situation
the planning authority will tend, depending upon its financial resources,
to zone land according to the laws of the market, in order to avoid such
compensation claims. Conversely, if the laws of compensation do not
favour landowners, then the planning authority will be less constrained
by the threat of compensation. In Britain, compensation is governed by
the provisions incorporated in the Land Compensation Act (1973), and
allows compensation to be paid in such cases as planning blight,
compulsory purchase, depreciation caused by public works and revoca-
tion of an existing planning permission (see Cullingworth, 1982,
pp. 88-90,97-101).
Landowners (for example, owner-occupier residents in an area) may
also constrain the planning authority by engaging in collective action to
have a planning decision reversed. As a result, the planning authority is
likely to take into account the potential and actual opposition which its
decisions are likely to meet in a particular area and to treat that
opposition as a limit to its actions.
The third constraint is that posed by other planning authorities. This
constraint derives from the fact that the consequences, both intended
and unintentional, of the actions of planning authorities, cannot always
be confined within their jurisdictional boundaries. Thus, for example, a
local authority where housing and land is scarce, is likely to be
constrained in solving its housing problems if the major policies of
adjacent planning authorities are to preserve green belts, maintain open
spaces and permit only low density residential development (see

Williams and Norman, 1971; Saunders, 1980a, pp.247-8). Similarly,
the fiscal problems of a local authority facing high expenditure demands
from its residents (for example, low income residents with large families)
can be exacerbated if neighbouring authorities strictly adhere to a policy
of low property taxes. Since one of the ways this policy is maintained is
by restricting entry of those who are most likely to be a fiscal burden (for
example, by restricting the amount of housing available to low income
families), its effect is to ensure that the fiscal burden is, ceteris paribus,
unequally distributed between local governments (see Baumol, 1967;
Albin, 1971; Aronson, 1974). In the same way, the efforts of a local
authority to attract jobs to an area may be constrained by other local
authorities who are able to offer more attractive packages to lure firms
and state enterprises (see Molotch, 1976; Logan, 1978). In all of these
ways, planning authorities can be constrained by the actions of other
planning authorities.
The fourth constraint is that posed by other state agencies or policies
where there is a lack of co-ordination and co-operation between the two
(see for example, Marriott, 1969, p.200).
The final constraint is that posed by the consumers of urban property
(that is, offices, factories, houses, schools, parks, hospitals, and so on).
In this case it is likely that the local authority will be subject to pressures
from the various class interests within its jurisdiction while at the same
time being constrained by the anticipated and actual reactions of the
electorate.
8.6.5 Summary
This analysis of the three main actors in the development process reveals
the mutually constraining effects which each has upon the other, as well
as the constraints imposed by other actors. Although it is not strictly a
Marxist analysis in the sense that the main actors are not classes, nor are
the actors related to each other as deterministically as in the Marxian
model, the analysis does emphasise the crucial role of conflict in the
property development process and the power of the various actors
involved in that process. In short, the analysis here provides a set of
concepts which can be readily operationalised, thereby overcoming the
disadvantages of some of the more abstract Marxian concepts.
It is now necessary to examine the second process which takes place in
urban areas, namely the process by which the urban area is used in the
production of commodities.
8.7 PRODUCTION IN AN URBAN AREA
8.7.1 Introduction
A large proportion of the wealth in capitalist societies (that is,

'commodities' which are both use values and exchange values, as well as
'products' which are only use values: see Marx, 1974a, Chapter 1) is
produced by capitalists and the state within urban areas. However, an
urban area can only be used to produce wealth if capitalists and/or the
state decide to locate their production operations there and if workers
live (or migrate to live) within commuting distance of those production
operations. Thus the decisions to locate, migrate and commute within
urban areas are an important part of the process by which the urban area
is used for production. Indeed, it is these decisions which determine
whether an urban area grows, declines or remains stable. In turn, they
determine the quantity and type of production of goods and services in a
particular urban area and, by implication, the quantity and type of
employment and unemployment in that area.
The methodological procedure which shall be used in this analysis is
the same as that outlined in section 8.2, and applied in the sections (8.4 to
8.6). This procedure which, for simplicity, will be applied to production
by capitalist firms (that is, ignoring the public sector), involves
investigating the interests which capitalists and workers pursue in
making their respective decisions to locate (in the case of the former) and
migrate and commute (in the case of the latter), as well as the constraints
which they encounter in pursuing that interest.
The decisions and actions of capitalists and workers in relation to
location, migration and commuting can be conveniently summarised in
terms of the concept of a labour market: capitalists' location decisions
constitute the demand side of the market while workers' migration and
commuting decisions constitute the supply side. The area covered by this
labour market is a labour market area and the precise relationship
between an urban area and a labour market area will be clarified in the
course of the analysis below. The analysis begins with an investigation of
the factors affecting the supply of labour in an urban area (subsection
8.7.2) followed by an investigation of the factors affecting demand
(subsection 8.7.3).
8.7.2 The Supply of Labour in an Urban Area
The supply oflabour in an urban area (or in a sub-area of an urban area)

consists of those living within commuting distance of work in that area.
This area constitutes a labour market area and, over time, the number of
workers in that area can be increased or decreased by migration. 4
It is well established that the size of the labour market area tends to
vary between workers according to their skill, largely because of
differences in their migration behaviour, but also because of differences
in their commuting behaviour. In Britain, for example, the size of the
labour market area tends to be much smaller for unskilled manual
workers than for higher professional, administrative and managerial
workers. This is because the former have lower migration rates than the
latter (see Johnson, Salt and Wood, 1974, pp. 106-7; Evans and Russell,
1980, pp. 212-13; Salt, 1980, p.258). In addition, they also commute
over shorter distances (see Berthoud, 1980, pp.244-8). In short, the
labour market area for unskilled manual workers tends to be less than,
or equal to, the size of the urban area within which they live. By contrast,
the labour market area for professional, managerial and administrative
workers tends to be coterminous with the entire national (and, in some
cases, international) urban system.
It is necessary to explore in more detail the reasons for this difference
in the size of labour market areas by exploring how the interests of
different types of workers vary in relation to migration. 5 For brevity and
simplicity the analysis will be confined to two groups of workers at
opposite ends of the skill spectrum: the unskilled manual worker and the
professional administrative and managerial worker.
The low migration rates of unskilled (and some skilled) manual
workers arises essentially because their material interests are not likely
to be significantly improved by migration. This is so for a number of
reasons. First, because their job is unlikely to have a career structure so
that migration from one area to another would not materially improve
their position. Secondly, competition for jobs is probably greatest in the
unskilled category in all areas, so that migration from one area to
another would not confer any decisive advantage. Alternatively, in the
case of some skilled workers (for example, coal and steel workers) their
skills may be specific to a particular area. Thirdly, the worker may be
living in a local authority house and is likely to encounter great difficulty
in getting a replacement in another area (see Johnson, Salt and Wood,
1974, pp. 1-2; Berthoud, 1980, pp. 248-51; Hughes and McCormick,
1981) - although as Gleave and Palmer (1978) have shown, local
authority tenants display a higher rate of movement within local
authority areas than owner-occupiers, but a lower migration rate
between regions. These are some of the reasons why it may not be in the
material interests of unskilled manual workers in advanced capitalist
societies to migrate. In historical terms, this represents a reversal of the
high migration rates of this group in the early stages of the industrial
revolution (Fielding, 1982, p. 28). The net effect of these factors is that
the labour market area of unskilled manual workers tends to be less than
or equal to the size of the urban area within which they live.
Professional, administrative and managerial workers generally have
relatively high migration rates in Britain and Western Europe. However,
as Fielding (1982, p. 28) points out, this was not always the case nor is it
currently the case in the United States (see Johnson, Salt and Wood,
1974, p. 106). The reason for this essentially is that their material
interests can be significantly improved by migration. This is because
migration may be associated with career advancement and salary
increases and the transaction costs of moving house from one area to
another are likely to be small relative to total earnings, particularly when
they are borne by the employer rather than the employee (see Salt, 1980,
pp. 258-60; Fielding, 1982, pp. 26-7). The net effect of these factors is
that professional, administrative and managerial workers have relative-
ly high migration rates with the consequence that their labour market
area tends to be coterminous with the national (and possibly also, the
international) urban system.6
It is now possible, on the basis of this review of the literature, to clarify
the nature of the relationship between an urban area and a labour
market area. This relationship can be conceptualised as a continuum. At
one end of the continuum, the labour market area is coterminous with a
part of the urban area or, in the case of smaller urban areas, with the
entire urban area so that all commuting occurs within that particular
area. This end of the continuum corresponds to the labour market area
of unskilled manual workers. At the other end of the continuum, the
labour market may be spread over a number of urban areas so that
commuting and migration occurs in various directions between each of
the urban areas. This end of the continuum corresponds to the labour
market area of professional, administrative and managerial workers. It
follows that the labour market areas of intermediate skilled workers will
fall between these two extremes, with the size of the labour market area
increasing directly with skill.
This conceptualisation of the relationship between urban areas and
labour market areas can also be reinforced, indirectly, by evidence on
the spatial distribution of unemployment in different skill groups in
Britain in the 1970s (see notably Cheshire, 1979, p.32; Evans and
Russell, 1980, pp. 212 -13; Corkingdale, 1980, p. 185; Richardson, 1980,
pp.214-17; Evans and Richardson, 1981, p.114; Elias and Keogh,
1982, p.11). The evidence shows that the rate of unemployment in
different urban areas tends to be uniform for professional, adminis-

trative and managerial workers but to vary for unskilled workers. The
most plausible explanation of this, according to Evans and Russell
(1980, p. 213), is that 'at the lowest skill levels, the spatial extent of the
labour market is likely to be coincident with a travel-to-work area, but
that the labour market tends to become a national market as the skill
level increases'. Within urban areas, however, rates of unemployment
tend to vary much more widely for all skill groups. In the Greater
London area, for example, Evans and Russell (1980, p. 210) found that
'. . . when the influence of skill level has as far as possible been
eliminated, ... there is still a persistent difference between the
unemployment rates prevailing in inner London and those prevailing in
outer London'.
The main reason for this seems to be that the unemployed (like the
employed) live in segregated neighbourhoods depending upon such
factors as their age, stage in the life cycle, skill and income. In other
words, the labour market interacts with the housing market to produce a
variegated pattern of intra-urban unemployment. The other factor
affecting unemployment is local variations in the demand for labour,
which will be considered in the next subsection.
The analysis in this subsection has revealed that urban areas and
labour market areas are conceptually distinct. They should not,
therefore, as frequently occurs (see, for example, Broadbent, 1977,
pp.94ff; Castells, 1977a, p.236 and subsection 6.2.2), be treated as
identical. 7
8.7.3 The Demand for Labour in an Urban Area
The demand for labour by capitalists and the state in an urban area is the
outcome of their past and present decisions to locate there and to employ
the labour living there. In this subsection, the analysis will be confined to
the decisions of firms in order to show how they affect the demand for
labour in different areas.
In broad terms, the decisions of firms as they affect the demand for
labour in an area can be classified in the following way:
1. The decisions of existing firms in an area:
(a) to maintain existing levels of employment;
(b) to expand existing levels of employment through:
(b ' ) the creation of new jobs,
(b") the relocation of jobs from another area;
(c) reduce existing levels of employment through:

(c') layoffs,
(c") relocation of jobs to another area;
(d) cease employment in an area through:
(d') demise of the plant or firm,
(d") relocation of plant/firm to another area;
2. The decisions of new firms to the economy:
(a) to locate in an area;
(b) not to locate in an area.
The significance of this taxonomy of decision-making is that it allows
the change in employment in any urban area to be broken down into two
components:
1. the absolute job loss or gain (to the area and the economy) which
would be the outcome of decisions Ib', lc' and Id'; and
2. the locational job loss or gain (to the area but not to the economy)
which would be the outcome of decisions Ib', lc', Id', 2a and 2b.
The value of this approach has been clearly illustrated by the work of
Massey and Meegan (1978). In their study of the decline in manufactur-
ing employment (in the electrical, electronics and aerospace equipment
sectors) between 1966 and 1972 in the cities of London, Manchester,
Merseyside and Birmingham, they found that the majority of job losses
were absolute rather than locational and were due either to closures
(decision Id') or layoffs (decision lc'). This fact, coupled with the failure
of new manufacturing firms to locate in the inner areas of Britain's larger
conurbations (decision 2b), is one ofthe main reasons for the relatively
higher levels of unemployment among manual workers in the 1970s and
1980s in these areas. This result has been replicated by many other
studies of job losses in the manufacturing sector in Britain's major cities
in the period from the mid-1960s to 1980 (see Elias and Keogh, 1982,
pp. 2-7 for a review of the evidence).
The significance of this result is that the private sector demand for
labour in an area is the outcome of a number of different types of
decisions made by firms, each pursuing their interest. However, the
demand for labour in an area can be further understood by investigating
the characteristics of the firms making those decisions. Increasingly the
demand for all types oflabour in the private sector (both manufacturing
and services) is constituted by multiplant, multiproduct and, frequently
multinational firms. According to Healey (1982): 'in 1972 approximate-
ly three quarters of employees in the private sector of manufacturing

industry in the United Kingdom worked for multiplant enterprises'
(ibid., p. 38).
The implication of this is that the type and quantity of labour
demanded in an area is increasingly being determined by the location
decisions of these mUltiplant firms. Numerous studies of their location
patterns in Britain, the United States and Sweden show that their
headquarters (and hence the associated functions of financial control,
research and development and top management) 'tend to be concen-
trated into a small number of large metropolitan complexes' (Dicken,
1976, p. 402). This applies to both national and international multiplant
firms whose headquarters tend to be located at the top, or in the top tier,
of the national and international urban hierarchy respectively (see
Westaway, 1974; Goddard, 1978; Goddard and Smith, 1978; Massey,
1979). Conversely, there is also some evidence that multiplant firms are
locating their branch plants (and their associated low-skilled assembly
function) in smaller urban centres (see Goddard, 1977, 1978; Massey
and Meegan, 1978, p.287; Massey, 1979, p.237) or even rural sites
(Keeble, Owens and Thompson, 1983). The outcome ofthese locational
decisions is that the demand for highly-skilled professional labour will
tend to be concentrated in the largest urban centres (or in their
outskirts), while the demand for lower-skilled manual labour is shifting
down and away from the urban hierarchy. In short, the spatial
distribution of the demand for different types of labour in advanced
capitalist societies is changing as a result of, inter alia, the location
decisions of multiplant firms. In other words, the way in which urban
areas are being used in the production of commodities is increasingly
dependent upon their size and hence upon their position in the national
and international urban hierarchy.
8.7.4 Summary
This section has attempted to analyse the way in which urban areas are
used in the production of commodities. This was done through the
analysis of the location decisions of firms and the migration and
commuting decisions of workers. These two sets of decisions were
analysed in terms of the concepts of a labour market and a labour
market area.
These decisions (together with those of the state, which were not
examined here) determine not only the amount and type of production
that takes place within a particular urban area; they are also the major
determinants of the process of urbanisation itself, and hence of the

forces causing the size of an urban area to increase, decrease or remain
stable. In short, the entire shape of the urban system in any capitalist
country is the outcome of a myriad of different decisions by firms, the
state and workers, each pursuing their interest in the context of the
constraints and opportunities posed by each other.
It is now nc;cessary, finally, to analyse the process by which urban
areas are used in final consumption.
8.8 CONSUMPTION IN AN URBAN AREA
8.8.1 Introduction
The act of final consumption refers to the process by which labour is

reproduced through the consumption of wage goods. The conventional
Marxian treatment of final consumption is the following:
Workers are paid money wages by capitalists in exchange for their
labour and this enables them to buy wage goods from capitalists.
These wage goods, usually along with unpaid domestic labour, are
then consumed by workers who, again, exchange their labour with
capitalists for another round of money wages; and so on, the cycle of
production and reproduction proceeds. (see subsections 1.1.2 and
2.3.2).
One of the important developments in capitalism since Marx's time
is the growing involvement of the state in the production of wage (or
final consumption) goods. These include housing, schools, hospitals,
public transport, sport and recreational facilities, landscaped open
spaces as well as water, sewerage and transportation networks. The
state produces these goods by spending the revenue which it receives
from taxes (on wages, profits and rents) and from its trading
operations (when they make a surplus) and distributes them accord-
ing to a variety of market and non-market mechanisms. The precise
distributional effect of such state expenditure will depend, in the case
of each individual citizen, on the difference between the taxes paid and
the consumption goods received.
The real wages of the worker in advanced capitalist societies can
thus be divided into two parts: the direct wage (that is, the money
wage paid directly by the employer - that is, the firm or the state - in
exchange for labour) and the indirect wage (that is, the 'wage' received
indirectly from the state). In the terminology of Castells and Lojkine

(subsections 6.2.1 and 7.2.2 respectively) the wage can be divided into
individual consumption and collective consumption.
These considerations are part ofthe conventional Marxian treatment
of consumption. However, there is an important spatial dimension to
consumption which has not been analysed by Marx or by subsequent
Marxists. This dimension derives from the fact that many consumption
goods are fixed to particular locations (for example, houses, schools,
hospitals, roads, parks, open spaces, and so on) and can only be
consumed at those locations. The analysis below will attempt to
conceptualise this spatial dimension within the framework of the
constrained-actor model.
The analysis begins with a typology of consumption goods (subsec-
tion 8.8.2) followed by a description of the spatial dimension of
consumption goods (subsection 8.8.3). The dwelling unit is then selected
as the central element in the matrix of consumption goods on the ground
that the location of one's dwelling is a major factor determining the
quantity and quality of other goods which an individual consumes
(subsection 8.8.4). The analysis then shows how changes in the
environment of the dwelling unit affects different tenure groups
differently, with significant distributional effects (subsection 8.8.5).
Finally, the typical reactions of residents to changes which affect
consumption goods within the area of their dwellings will also be
examined (subsections 8.8.6 to 8.8.9). In this way it is hoped to provide
the basis for a spatial analysis of consumption.
8.8.2 A Typology of Consumption Goods
Consumption goods - in fact all goods - can be classified as either

private, state or communal according to their ownership (Macpherson,
1973, 1978). A private consumption good (for example, a house, a car,
an overcoat, and so on) is defined by the fact that its owner has three
related property rights over that good. These rights are:
I. The right: (a) to exclude others from it; and

(b) to determine the use to which it is put.
2. The right to receive an income from it, if possible.
3. The right to transfer it (through sale, lease or inheritance).
These rights confer complete or pure private ownership over a good (see
Alchian, 1965, p.819; Cheung, 1978, p.51; see also Demsetz, 1967;
Table 8.2 A typology of consumption goods
Ownership of consumption good

Private State Communal
Example: Example: Example:
Food Schools Fresh Air
Clothing Hospitals Visually pleasant
Positive Furniture Public transport environment
Household A residential area
appliances, etc. free from crime
Effect of
consumption
good
Example: Example: Example:
Noxious gases, Polluted air
liquids, or solids, Congested roads
Negative Domestic refuse which are owned Noise
by the state
Furubotn and Pejovich, 1972; Alchian and Demsetz, 1973; Pejovich,

1982).8 Typically, a private consumption good (like a private capital
good),9 is distributed through the market where access is based upon
ability to pay, although access through non-market mechanisms such as
inheritance or marriage also occurs.
A state consumption good (for example, roads, streets, footpaths, as
well as public schools, hospitals, parks, houses, and so on) involves the
same bundle of property rights as a private consumption good except
that the way in which the state exercises its right may be very different
from that of the typical capitalist or worker (since, as has been seen in
sections 8.5 and 8.6, their interests are very different). Access to state
consumption goods typically occurs in three different ways: through the
market mechanism, that is, based upon ability to pay; through the
bureaucratic mechanism, that is based upon the state's definition of
need; and through the free-for-all mechanism, that is, based upon the
first-come-first-served principle. It is the latter form of access which
effectively - that is, de facto if not de jure - transforms a state
consumption good into a communal consumption good.
A communal consumption good (for example, air, common land,
weather, outer space, the visual environment, the social status of an area,
and so on) is defined by the fact that everyone has the right to use it and
hence to benefit from or be harmed by it. In short, no one can be
excluded from its consumption (see Dales, 1975, p.491; see also Sax,
1971).10 The most distinctive feature of communal consumption goods is
that they are distributed free of charge to those living within accessible
distance from them. It is for this reason that certain types of state
consumption goods (roads, footpaths, streets, parks, and so on) are, in
fact, if not in law, communal consumption goods.
Consumption goods can be either positive or negative according to
whether their consumption has a positive or negative effect upon the
consumer's well-being. Leaving aside those borderline cases where
consumption goods have both positive and negative effects (for
example, alcohol, tobacco, cars, and so on), it is possible to construct a
typology of consumption goods. This is done in Table 8.2.
It is now necessary to examine the spatial dimension of these various
consumption goods.
8.8.3 The Spatial Dimension of Consumption
Many consumption goods are fixed to particular locations (for example,

houses, schools, hospitals, theatres, pubs, scenic areas, and so on) and
create a positive or negative effect - an externality 11 - around the point
where they are located according to whether they are positive or negative
consumption goods. It is this externality effect which gives consumption
goods their spatial dimension (see Pahl, 1975, Chapters 7 and 9).
The effect created by the location of a consumption good in an area
can be understood in terms of the concepts of externality field (see
Harvey, 1973, p.60; Smith, 1977, p.91; Bale, 1978, p.334) and
externality gradient (see Bale, ibid.). An externality field refers to the
area over which the consumption good has an effect, either positive or
negative. Within the externality field the intensity with which the effects
of a consumption good are felt are likely to vary, that is, increase or
decrease, with distance from its location, and this can be depicted by the
externality gradient.
The concepts of externality gradient and externality field are illus-
trated diagramatically in Figures 8.2 and 8.3, using the example of a
negative communal consumption good (for example, smoke emissions
from a factory chimney which are consumed by those breathing the
affected air).
Figure 8.2 shows that the negative effect of the consumption good
(measured on the oy axis) decreases with distance (measured on the ox
axis) from its location (0). The externality field in Figure 8.3 is derived
from the externality gradient and is based upon the assumption that the
Effect (i.e.,
cost or
benefit) of
the
consumption
good
miles
Distance from the location
of the consumption good
Figure 8.2 The externality gradient of a negative communal consumption good
Externality field
distance from the

location of the cosumption
good
-----,1-. _ _ area affected by the

consumption good
Figure 8.3 The externality field of a negative communal consumption good
effect of the consumption good is uniformly diffused in aU directions

from its location. Thus, since ox' is the distance over which the
consumption good has an effect, the rotation of ox' through 360' from
the source (0) defines the externality field, that is, the area affected by the
y
Effect (i.e.,
cost or benefit)
of the consumption
good
o x x" miles
Distance from the location of the
consumption good
Figure 8.4 The externality gradient of a consumption good with both positive
and negative effects
area where consumption good

has a negative effect
_--f----area where consumption good

has a positive effect
Figure 8.5 The externality field of a consumption good with both positive and
negative effects
consumption good (1t X (ox'i). It follows from this that if the popula-
tion is evenly distributed in the area, then every unit increase in
the distance over which the consumption good has an effect will result in
a trebling of the area affected and hence a trebling of the number of
people affected by the externality.
Not all consumption goods are unequivocally positive or negative,
however; many have both positive and negative aspects (for example, an
airport, a motorway, a pub, and so on). This is illustrated in Figures 8.4
and 8.5.
Figures 8.4 and 8.5 show the case (for example, an airport) where the
negative effects of the consumption good are greatest for those living
closest to its point oflocation (0) and thereafter decline to point x' after
which the effect becomes positive until point x", beyond which there is
no effect whatever.
It is now necessary to show, using this analysis, how the various
elements in the matrix of consumption goods are related and consumed
in the urban context.
8.8.4 The Dwelling Unit
The dwelling unit l2 is probably the most important consumption good

for any individual, not only because it is the place within which many
other consumption activities occur (eating, sleeping, recreation,
procreation, and so on) but also because it is the point of origin from
which journeys to work as well as to other sources of consumption
(schools, shops, clinics, sports centres, parks, theatres, and so on) are
undertaken. In other words, the location of a dwelling determines its
proximity to other private consumption goods (for example, shopping
centres, golf courses, pubs, theatres, and so on) state consumption goods
(for example, schools, hospitals, transportation networks, landscaped
open spaces, local taxes, and so on) and communal consumption goods
(for example, clean air, quietness, the social status of the area, the level of
crime, and so on). It is for this reason that the consumption of a dwelling
unit is spatially and inextricably connected to other locationally fixed
consumption goods (see Downs, 1981, pp. 16-19). In other words, the
dwelling determines the spatial and in some cases the social constraints
on access to other consumption goods (see Pahl, 1975, pp. 201, 248-64;
see also Hirsch, 1978, for a broader discussion on the problems of access
to consumption goods). Thus, as Bourne (1981, p.3) points out
'obtaining the key to a house, flat or apartment also brings with it the
right to a given level of public services (police, schools, parks, etc); it
determines one's local tax rate and in turn it delivers a set of benefits
(security, etc.) or disbenefits (crime, vandalism, etc.) to both the
individual household and the larger community' .13
The location of consumption goods, in turn, normally has a
simultaneous effect upon a number of contiguous sites within a given
area. Using the concepts of externality gradient and externality field,
these locationally fixed consumption goods (for example, shopping
centres, public parks, noise, pollution, and so on) can be seen as creating
a field of effects, which normally decline with distance. In this way, the
unequal distribution oflocationally-fixed consumption goods within an

urban area gives rise to different but internally homogenous sub-areas.
Numerous studies have attempted to quantify the effect which the
location of various types of consumption goods can have upon the price
of dwellings within a particular area. These studies have focused upon
such consumption goods as pollution (Ridker and Henning, 1967),
expenditure on education (Oates, 1969) and status (Bailey, 1966), and
have shown how the proximity of dwellings to these goods can affect the
price of those dwellings. This confirms that an intimate connection
exists, and is perceived to exist (see Agnew, 1978), between the dwelling
and other locationally-fixed consumption goods (see also Saunders,
1978, p.247; Nicholls, et al., 1982, p.338). In a capitalist market
economy, the price of a dwelling, which is determined not only by its
location vis-a-vis other consumption goods but also by such factors as
the scarcity of sites, the cost of building materials, the availability of
mortgage finance, and so on, acts as the mechanism governing access to
that dwelling, and hence access to different parts of the urban area. One
of the outcomes of this process is that the inequalities in income within
and between classes, as a result of their position in the sphere of
production, are spatially reflected by their location in the different parts
of the urban area. Urban sociologists and urban geographers have
devoted considerable attention to documenting, through the use of
social area analysis and factorial ecology, the existence of these various
sub-areas, though less attention had been given, at least up until the late
I 960s, to the processes by which they are created (Ambrose, 1977, p. 39).
The significance of these considerations from the point of view of the
constrained-actor model is that it is possible to identify at least three
different actors (or tenure groups) whose interests in relation to housing
are different and who are therefore affected in different ways by changes
in the value of their dwellings. These three tenure groups, that is, owner-
occupiers, local authority tenants and private sector tenants are defined
by their property rights over dwellings and may be referred to as
'housing classes' because they constitute different sets of interests. 14
Although there are significant differences within each of these tenure
groups in terms of income and the quality of their dwellings and its
environment, each nevertheless shares a common material interest in
relating to housing, not only because they tend to be similarly affected by
changing property values, but also because state subventions to housing
are distributed primarily on the basis of tenure (although the size of the
subvention is affected by income and other factors). Other tenure groups
- such as housing association tenants - may also be identified if it can be
shown that their material interest in relation to housing is different from

that of other tenure groups.
8.8.5 Tenure Groups and Changes in the Site Value of Dwelling Units
In Britain there are three main types of housing tenure: owner-

occupation; rental from a private landlord; and rental from the local
state. (Housing associations and co-operatives are yet another form of
tenure but, unlike some European countries, this sector is very small in
Britain.) The precise mechanism by which access is gained to these
different tenurial positions is determined ultimately by income and stage
in the life cycle and proximately by the 'gatekeepers' or 'managers' of the
housing supply (see Pahl, 1975, Chapter 13; see also Williams, 1978).
In the private sector, where dwelling units are purchased or leased
from private landlords, these managers are estate agents, mortgage
managers, and landlords, who ensure that dwellings are distributed
according to market criteria (that is, ability to pay) so that those with the
highest (earned or unearned) income will obtain the best dwelling units
and vice versa. In the state sector, where dwelling units are rented from
the state, the manager is the local state authority who ensures that the
dwellings are distributed according to the state's definition of need.
Each tenure group is affected differently by changes in the price of
their dwelling unit. These changes are caused by spatial factors (that is
changes in the location and quality of private, state and communal
consumption goods) and by non-spatial factors (for example, inflation,
interest rates, state subventions, and so on). The analysis here will focus
only upon those changes caused by spatial (or environmental) factors
for two reasons: first, in order to highlight the spatial interdependencies
involved in the consumption of urban dwellings, and secondly, in order
to demonstrate that the differences in material interests between
different tenure groups or housing classes are not solely 'contingent'
upon such non-spatial factors as high inflation, low interest rates and
state subventions as Dunleavy (1979, p.420), Saunders (1980a,
pp. 97 -8) and Pratt (1982, pp.485-8, 495) have claimed. These
different interests are a direct result of the operation of different sets of
property rights within a built-up area.
The price of a dwelling, as has been seen, can be affected spatially by
changes in the location of private consumption goods (for example, the
change may relate to a shop, a golf club, or any other privately-owned
amenity), state consumption goods (for example, the change may relate
to the siting of a motorway, an airport, a dump, a park, and so on), or
communal consumption goods (for example, the change may relate to

the pollution of the air by smoke, noise, or smells, or to the increase in
crime in the area, and so on). Some of these changes may increase the
site-value of a dwelling unit while others may reduce it. If they increase it
then owner-occupiers receive, inter alia, an increase in their imputed
income from home ownership while tenants in the private sector will
have an increase in the actual rent which they pay to the landlord, as long
as competitive market conditions prevail (see Downs, 1981, Chapter 6).
The effect in the case of local authority tenants cannot be determined a
priori since it depends upon the rent policy of the local authority and the
constraints imposed upon it, in turn, by central government. Converse-
ly, if the price of the dwelling is reduced then owner-occupiers suffer a
reduction in their imputed rent while private sector tenants get a
reduction in their actual rent, assuming normal market conditions. The
effect, in the case oflocal authority tenants depends again upon the rent
policy of the local housing authority (see Neutze, 1976, pp. 85-6 for a
useful discussion of the way in which increases in land and house prices
affect different tenure groups.)
The property rights of an individual over a dwelling are thus an
important consideration in assessing the distributional consequences of
changes in the positive and negative consumption goods affecting that
individual's area. The implication of this is that a group of individuals
(for example, manual and non-manual workers) may have the same
material interests in relation to production but quite different material
interests in relation to housing if their tenure positions are different.
These different class positions may lead, depending upon the way in
which they are seen by each individual, to different types of individual
and collective action at both the national and neighbourhood level. The
evidence in Britain suggests that, at both these levels, there is a
significant difference between owner-occupiers and tenants in their
political allegiances (see Dunleavy, 1979) and also in terms of the goals
of local community action (see Saunders, 1978, 1980a and b; Silburn,
1975).
It is now necessary to analyse more fully the individual and collective
reactions of different tenure groups to changes which affect the site value
of their dwellings, and hence affect their material interests.
8.8.6 Individual and Collective Reaction to Changes in Consumption
Residents of an area, irrespective of their tenurial position, have an

obvious interest in increasing the positive consumption goods in their
area and reducing the negative ones. As a result, the location of

consumption goods in an area is frequently an issue around which
individual and collective action arises. The way in which this occurs can
be seen in the case of an area facing the threat of an increase in some
negative consumption good (for example, an increase in local taxes 15 an
increase in the level of crime, an increase in air and noise pollution, a
deterioration in the visual environment or in the social status of an area,
and so on). Although the analysis below will be illustrated by examples
of negative consumption goods the same processes of individual and
collective action could be applied to the location of positive consump-
tion goods (for example, parks, creches, schools and other desirable
amenities).
The state (both central and local) is usually one of the main targets
towards which individual and collective action is directed in an urban
area, since it plays a crucial role in producing, distributing and
regulating many of the consumption goods (both positive and negative)
which are located there and which are the source of what Cox terms
'localized welfare impacts' (Cox, 1978, p. 99). It is for this reason that
action over consumption in an urban area is frequently political action,
as the growing body of literature on urban conflict and urban politics
clearly shows (see Cox, 1973, 1978; Agnew, 1978; Dear and Long, 1978;
Reynolds and Honey, 1978; Ley and Mercer, 1980; Cox, 1981; Cox and
Johnston, 1982; Cox and McCarthy, 1982; Robson, 1982).
Residents in an area affected by a negative consumption good (for
example, air pollution) can react in a number of different ways. They
may leave or stay, protest or remain silent. The full range of options,
based upon a modification of Hirschman's model is outlined in Table 8.3
(see Hirschman, 1970; see also Orbell and Uno, 1972; Dear and Long,
1978).
Table 8.3 A typology of the reactions of residents to a

negative consumption good in their area
Silence Protest
Stay (1) Resignation (3) Collective action

Leave (2) Migration (4) Protest in exile
In the present context only reactions (l), (2) and (3) will be considered
since option (4) is more appropriate to cases such as political refugees.
The first reaction is resignation and acceptance of the negative
consumption good. The second is for some (or all) of the residents to
migrate to another area in order to avoid the negative consumption
good. The third is for residents in the affected area to engage in collective
action to have the negative consumption good reduced or relocated.
Each of these three reactions will now be discussed respectively
(subsections 8.8.7 to 8.8.9).
8.8.7 Resignation
Resignation is possibly the most frequent response of residents to the

location of a negative consumption good in their area. There are at least
three reasons why residents may adopt this strategy. First, the effects of
the negative consumption good may be insignificant and thus insuf-
ficient to generate either migration or collective action. Secondly,
residents may feel powerless to effect a relocation or even a reduction in
the negative consumption good, thus reflecting their alienation from the
political process (see Mullins, 1977, p. 33). Thirdly, the residents may
not have the money, time or energy to either migrate from the area, or to
engage in collective political or legal action. In consequence, they may be
resigned to what they see as the inevitable.
8.8.8 Migration
The choice to migrate or 'exit' (Hirschman, 1970) is determined, in this

context, by the resident's perception of the size of the effect created by
the negative consumption good, the possible success of collective action
and the constraints of migration facing each resident. If the effect of the
negative consumption good is large (for example, a rising level of crime
or taxation, the location of an airport or motorway, the movement of a
low status group into a high status area,16 and so on) and if the
possibility of successful collective action is low then, ceteris paribus,
migration is likely to be considered as a rational course of action.
However, as has been seen (subsection 8.7.2) those on lower incomes
face major constraints to migration by comparison with higher income
owner occupiers because of the costs involved in migration. Moreover,
migration is likely to be a satisfactory solution for only those owner-
occupiers who move before the full consequences of the negative
consumption good are reflected in the price of their dwellings. The
possibility of migration for local authority tenants depends ultimately
upon the policy of the local authority while, in the case of private sector
tenants, it depends upon such factors as the location and price of private
rented accommodation in other areas (see Cox, 1978, p.99; Cox and
Johnston, 1982, pp. 200-1).
Migration, however, does not have the effect of reducing Or relocating
a negative externality so that the remaining residents as well as the new
immigrant residents will be obliged to bear its full costs. This may still
happen even if they engage in collective action (see Dear and Long, 1978,
pp. 117 -18).
8.8.9 Collective Action
Collective action, as the term implies, is action by a group to obtain a

goal or interest shared by the group. In the present context, the exclusion
of a negative consumption good from a given residential area would, by
definition, be a shared material interest. Collective action requires an
organisation through which the group interest can be expressed and
advanced and, in the present context, this organisation is typically a
residents' committee or association although it may also be the local
branch of a political party or an environmental organisation.
The emergence of an organisation to advance the collective interest of
a group is neither straightforward nor inevitable. Indeed, as Olson
(1971, Chapter 1) has shown, if individuals are economically self-
interested, they will be unable, because of the 'free-rider problem', to
form an organisation to protect or advance their collective interest -
unless they resort to special incentives or coercion to ensure organ-
isational membership. If, by contrast, individuals act from, moral
motives (for example, from a sense of justice or compassion) then an
organisation to further the collective interest is likely to form more easily
through the altruism of its members.
In the case of residents' associations, the furtherance of collective
interests may also encounter the obstacle that the interests of all the
residents may not be entirely identical. This could happen for a number
of reasons. First, there may be variations in the age, stage in the life cycle
or tenure status of the residents. As has been seen (subsection 8.4.5)
negative (and positive) consumption goods affect each tenure group
differently; owner-occupiers are more likely than other tenure groups to
bear the full costs (and benefits) of these goods and, other things being
equal, they may be more likely to engage in collective action than other
tenure groups.17 Secondly, certain consumption goods (for example, a
motorway, a railway line, an airport) may be of the type whose effects
are predominantly negative for those living closest to them, and
predominantly positive for those living at greater distances (subsection
8.4.4). This creates a divergence (and possible a conflict) of interest

between two different groups of residents. Thirdly, some negative
consumption goods (for example, those relating to the social status of an
area) are crucially dependent upon the belief or attitude that certain
racial, religious, ethnic or income groups are of inferior status and this
attitude may not be universally shared by all the residents of an area.
Fourthly, the negative consumption good may affect a large area and
hence a large number of residents and, as a general rule, large groups are
much more difficult to organise than small groups. These four
considerations reveal the types of impediment which are likely to
confront the emergence of a common interest in a given urban area.
The fact that an organisation emerges in a residential area, in spite of
these impediments, does not guarantee, however, that it will be effective
in furthering the interests of residents. The effectiveness of an organisa-
tion in mobilising opposition depends upon a number of factors of
which the following are the principal ones: the amount of money at its
disposal; the quality of leadership; the internal structure of the
organisation; its linkages (both vertical and horizontal) with other
organisations, for example, political parties, trade unions, the media,
other residential associations, and so on (see Pickvance, 1975,
pp. 40-41); its contacts with influential individuals (for example, senior
personnel in central or local government); its choice and number of
tactics and the type and cost of the demands being made. Anyone of
these factors could mean the difference between the success or failure of
an organisation in attaining its goals.
An investigation of the way in which residents collectively attempt to
exclude negative consumption goods from their area reveals that a wide
range of exclusionary strategies have been used. These strategies vary
according to the source and type of the consumption good and from one
country to another. The following are some of the main exclusionary
strategies (apart from the market, which is also an exclusionary strategy,
but does not require collective action).
1. Controlling the sale of dwelling units to new residents. This strategy,

according to Cox (1973, p. 25), is used in the US by white residents in
'neighbourhoods threatened by racial transition' (ibid.). In such
cases, 'residents planning to sell property are frequently asked to
place it in the hands of an agency expressly established by a
neighbourhood association for the purpose oflimiting the business
carried on by block busters' (ibid.; see also Wolf and Lebeaux, 1969,
Chapter 3). This technique facilitates the racial residential segrega-
tion of neighbourhoods as well as ensuring that the supply of

housing is more restricted for non-whites than for whites so that,
ceteris paribus, the former pay more for the purchase or lease of
housing than the latter (see Ridker and Henning, 1967, pp. 250-51,
255-6; Kain and Quigley, 1975, p. 540; see however, Bailey, 1966,
pp.218-20).
2. Exclusionary land-use zoning. This strategy, which is widely used in
the US, involves zoning land according to certain criteria which
effectively exclude low income groups. These criteria include the
requirement that the size of the site and the building must be large,
the prohibition of mobile homes and multiple family dwellings,
restriction on the number of bedrooms, and so on (see Williams and
Norman, 1971). Alternatively the area may be zoned for open space
(see White, 1981).
3. The creation of physical barriers. Physical barriers in the form of
roads, rivers, walls, railway lines, and so on, act as boundaries
between different residential areas. These barriers may be important
in 'protecting' a higher status neighbourhood from what it regards
as too close proximity to a lower status neighbourhood. Where such
barriers are not present, they may be erected. The most famous
example of this occurred in Oxford, England, in the 1930s, when a
middle-class housing estate constructed a wall - the Cutteslowe
Wall - to reduce their proximity to an adjacent local authority
housing estate (see Collison, 1963).
4. The use of vigilante groups. The residents in an area may adopt this
type of strategy to combat crime and vandalism. However because it
is illegal it is likely to be covert and of short duration.
5. The use of political pressure. This strategy is most likely to be used
when it is within the power of the (central or local) state to reduce or
relocate a negative consumption good (for example, the flight path
of aircraft over a particular residential area). A large number of
tactics may be used in the furtherance of this stragegy. Pickvance
(1975, p. 37) has classified these into 'institutional', that is, those
within the framework of the law (for example, voting, petitions,
deputations, demonstrations, creation of media interest, and so on)
and 'non-institutional', that is, those outside the framework of the
law (for example, violent or illegal demonstrations, sit-downs and
sit-ins, rent strikes, and so on). The effectiveness of any particular
tactic is likely to be highly dependent upon the circumstances of the
case, though Saunders' study of the actions of residents' associa-

tions in the London Borough of Croydon reveals the overriding
effectiveness of institutional tactics (see Saunders, 1980a, p. 121).
8.8.10 Summary
The analysis in this section has attempted to develop a spatial dimension

to the concept of final consumption by attempting an account of the fact
that certain elements of the real wage (for example, houses, hospitals,
schools, and so on) are fixed to particular locations and can only be
consumed at those locations. The analysis focused upon the dwelling
unit as the central element in the matrix of consumption goods. It was
shown how changes in the value of the dwelling unit, particularly those
arising from changes in its location vis-a-vis other consumption goods
can affect different tenure groups differently, and can lead to various
forms of individual and collective action. Thus the analysis of the spatial
dimension of final consumption in this section has opened up new issues
and problems about the nature of final consumption within urban areas.
8.9 CONCLUSION: NEW DIRECTIONS FOR MARXIST

URBAN SOCIOLOGY
This chapter has outlined some new directions for Marxist urban
sociology. These directions were based upon the constrained-actor
model which, although not strictly Marxist, nevertheless retains the
Marxian emphasis upon the role of interest andiconflict inhumaminter-
action. The result is a more micro-level analysis than that contained in
the works of Castells and Lojkine, and one which overcomes some of the
difficulties found in their work.
The novelty of the contribution presented in this chapter lies in the
application of the constrained-actor model to the analysis of the three
processes constitutive of an urban area. The first of these is the process
by which urban areas are produced in a capitalist society and the
analysis of this involved a conceptual investigation of the logic of state
and private property development (sections 8.4 to 8.6). This investiga-
tion provided a broader framework of analysis than that found in
Castells and Lojkine where there is no explicit treatment of the process
of private property development and where the treatment of state
property development, via the concept of collective consumption, is far
from satisfactory (sections 6.2 and 7.2). Thus the analysis in those
sections provided a framework within which more detailed empirical

studies of private and state property development in different capitalist
countries can be undertaken.
The second process is the process by which urban areas are used in the
production of commodities in a capitalist society. Since an urban area
can only be used to produce commodities if capitalists and/or the state
decide to locate their production operations there, and if workers live (or
migrate to live) within commuting distance of those production
operations, the analysis of this process focused upon the location
decisions of firms and the migration and commuting decisions of
workers. This analysis was carried out (in Section 8.7) using the concept
oflabour market areas and it provided the basis for a clarification of the
precise relationship between urban areas and labour market areas. The
significance of this analysis is that it explicitly treats the issues of
industrial location, labour migration and commuting patterns which
have an important bearing upon the use of urban areas and which have
not hitherto been adequately incorporated within urban sociology
generally or Marxist urban sociology in particular.
The third and final process constitutive of urban areas is the process of
final consumption. The novelty of this part of the analysis (section 8.8)
lies in its typology of consumption goods and in demonstrating that
consumption goods have a spatial dimension. Many consumption
goods, whether they are private, state or communal and whether they
are positive or negative, are normally located and fixed at specific places
and can only be consumed at those places. These considerations were
conceptualised by treating the dwelling unit as the central element in the
matrix of consumption goods whose value is determined, inter alia, by
the externality effects generated by its location vis-a-vis other consump-
tion goods. This analysis contrasts with the traditional Marxian view
where consumption goods are treated non-spatially as bundles of wage
goods. The above analysis also examined the significance of housing
tenure as well as the various types of actions which residents typically
undertake in order to maintain or improve the amount of consumption
goods in their area.
The main conclusion to emerge from this chapter and from the book
generally is that the field of Marxist urban sociology, as found in the
writings of Manuel Castells and Jean Lojkine, could be considerably
advanced by developing a more micro-level analysis of urban processes
and conflicts. This analysis would complement rather than replace the
traditional Marxist approach since both macro- and micro-level explan-
ations are essential to all sociological analysis, including Marxist urban
sociology.
Notes
General Introduction
1. One of the simplest measures of the volume of interest generated by their

work is the number of citations of it which are listed in the Annual Social
Science Citation Index (see Institute for Scientific Information, 1969-1984).
Table Nllists the annual number of citations of their work since their names
first appeared in the index.
Table NI Number of citations per year of the works of Manuel Castells

and Jean Lojkine and the average number of citations of each cited author
in the Social Science Citation Index (1969-84)
Year Castells Lojkine Average
1969 4 1 3.34
1970 3 I 3.27
1971 7 I 3.42
1972 5 0 3.39
1973 5 0 3.41
1974 23 0 3.40
1975 42 I 3.61
1976 22 3 3.68
1977 31 8 3.66
1978 61 4 3.72
1979 80 II 3.90
1980 103 12 4.00
1981 97 12 4.28
1982 92 15 4.42
1983 110 28 3.94
1984 104 14 4.03
Source: Compiled from the Annual Social Science Citation Index for each year
between 1969 and 1984 (see Institute for Scientific Information, 1969-1984).
Although Table NI reveals the volume of interest generated by these two

authors it does not indicate whether that interest is favourable or unfavourable.
However, three facts can be clearly discerned from the table. The first is that
Castells's work has generated much greater interest than that of Lojkine. This
is due not only to the slightly larger volume of Castells's work (although
Castells frequently publishes the same material in slightly modified form in a
231
232 Notes
number of different places) but also because Castells's work has been widely
translated into English, Spanish, Portugese, Italian and Danish (see UNESCO
1974-1977). Lojkine's work has not been as widely translated. The second fact
is that there has been a fairly steady increase in interest in both authors,
particulary since the mid-1970s. The third fact is that the number of citations of
Castells's work, with the exception of 1970, has been consistently above the
average of other authors cited in the Social Science Citation Index, while the
number of citations of Lojkine's work has also been above average since 1977.
The writings of both authors are listed in the Bibliography below.
1 Marx's Labour Theory of Value
1. Walker comments, appositely: 'By socially necessary Marx means, of course,

commercially necessary. In other words, an industrialist cannot, because of
competition, use a productive method less mechanized than that prevalent in
his industry generally, without eventually losing money' (Walker, 1978,
p. 162).
2. The reason for Marx's preference must be seen in the context of his
distinction between the use value and the exchange value oflabour (which is
discussed later in this section). To anticipate: when Marx speaks of the use
value of labour power he refers to the value of the output which that labour
power creates and describes it by the term 'the value of labour' (see Marx,
1974a, Chapter 7, pp. 173-92; 1975, section 7, pp.43-6). By contrast, when
he speaks of the exchange value of labour power he refers to the value of the
wages which are necessary to reproduce that labour power'and describes it by
the term 'the value of labour power' (ibid.).
3. There is in tum an important distinction, within the concept of constant
capital, between the stock of constant capital (that is, fixed constant capital)
and the flow of constant capital (that is, circulating constant capital). Fixed
constant capital (k) refers to the value of fixed equipment, stocks of raw
materials and half-finished goods net of depreciation which last for more
than one period of production. Circulating constant capital (c) refers to the
value of the depreciation of fixed capital equipment plus the value of raw
materials, fuel, etc. which is used up in one production period. It is usually
assumed by Marxists (though perhaps not always justifiably: see Purdy, 1973,
p. 18; and Hodgson, 1974a, pp. 57-8) that all fixed constant capital is used
up during one production period (that is, k = 0), so that constant capital is
equal to circulating constant capital (that is, C = c) (see Howard and King,
1975, p. 199). In the present analysis, total capital is assumed to be equal to
circulating constant capital. This is an assumption of convenience, which is
appropriate for the present elementary analysis.
4. Morishima (1973, Chapter 1) distinguishes between two different definitions
of the value of a commodity which are present in Marx. The first definition,
which is the one just outlined, defines the value of a commodity as equal to the
value of those constituent elements which become embodied in it, while the
second defines the value of a commodity as equal to the socially necessary
Notes 233
labour which is required to produce one unit of the net output of that
commodity. Morishima remarks that: 'At first sight these may look like
identical definitions, and Marx actually regarded them as synonymous. They
are, however, found to be different views of value, their equivalence being
established only if rigorous proof is provided' (ibid., p. 11). The distinction
between these two definitions will not be pursued here: it will be assumed,
following the convention in these matters, that the first definition is the
correct (if not the only) definition.
5. Marx's assumption about money is, evidently, no longer correct. Money is no
longer a produced commodity (like gold) whose value/price is determined by
its value/price of production. Money now takes the form of pieces of paper
and plastic whose worth is fixed by convention. Marx's assumption that all
money is gold adds an additional complexity to his analysis in the sense that
the money value or money price of a commodity can change, not only as a
result of changes in the production of that commodity but also as a result of
changes in the production of gold.
6. It is sometimes argued that the dualism between values and surplus value on
the one hand, and prices and profit on the other, corresponds not only to two
levels of reality but also to two different historical eras (see, for example,
Meek, 1956, 1976b;Sherman, 1970,p. 278; Marx, 1974c,p.177;Engels, 1974,
pp. 895ft). The argument here is that in the preceding historical era of simple
commodity production commodities were exchanged at their values, whereas
in the contemporary era of capitalist commodity production, commodities
exchange, not in terms of their values, but in terms of prices. Thus, the
argument goes, values are not only theoretically and onto logically prior to
prices, but also historically prior.
Morishima and Catephores (1975; 1976) provide a simple refutation of this
argument by showing that the historical era of simple commodity production
could never have existed. They point out that simple commodity production
involves a society of independent producers who are independent in three
senses: first, there is no ex-ante co-ordination; second, each has individual
ownership over his/her own means of production and third, there is mobility
of producers among jobs. By applying this definition to the historical
evidence they conclude that: 'Simple commodity production has never been
realised in history in its full or pure form, or even in a tolerably approximate
form, because of the lack of mobility of producers among jobs in the pre-
capitalist age' (ibid., p. 315).
7. There is some disagreement over the question of Marx's mathematical
ability. Engels's (1959) view was that Marx was 'well versed in mathematics'
(ibid., p. 19) while Bortkiewicz (1952) refers to the 'scantiness' (ibid., p. 55) of
Marx's knowledge of mathematics. Morishima's (1974a) judgement is
perhaps the most balanced in pointing out that while Capital is not explicitly
mathematical, many passages can be translated into rigorous mathematical
language and that, 'mathematical problems, even new mathematical
problems are concealed in his economics' (ibid., p.6l2).
8. Those who attempt to defend Marx's labour theory of value by separating the
issue of prices from the issue of profit have acquired the label 'fundamental-
234 Notes
ist' in contrast to the label 'neoRicardian' which has been reserved for those
who reject the labour theory of value and adopt the Sraffa approach (see Fine
and Harris, 1976, p. 144). However, the neoRicardian label is misleading
since Sraffa's work demolishes the labour theory of value, held by both Marx
and Ricardo, which claims that prices and profit are regulated exclusively by
the quantities oflabour embodied in commodities (see Hodgson, 1976, p. 20;
1977b, pp.90-92; Roncaglia, 1978, pp. xvi-xix, 131-47; Bhaduri and
Robinson, 1980, p. 104)
9. It should be noted that not all of the arguments relevant to this debate have
been considered here. Possibly the most important argument relates to the
effect of joint production (that is, pure joint production and fixed capital)
upon the transformation of values into prices and surplus value. Steedman
has shown that, when there is joint production, it is possible to have a
capitalist economy where there is, simultaneously, positive profits and
negative surplus value and vice versa (see Hodgson and Steedman, 1975;
Steedman, 1975b and 1977, Chapter 10; however, see also Morishima, 1976
and Steedman, 1976). This is a highly anomalous result and proves that the
labour theory of value breaks down in the case of joint production. Similarly
the existence of a choice of techniques seriously undermines the primacy of
the labour theory of value. This is because the value of a commodity can only
be calculated once the technique of production is known. However, when
there is a choice of techniques, the technique that is actually used will be
determined by capitalists based upon current prices and profit rates. In other
words, prices and the rate of profit must be known before values can be
calculated. In this case, therefore, prices and profit are prior to values rather
than the reverse (see Steedman, 1977, pp. 64-5; Elster, 1985, p. 136).
2 Marxian Crisis Theory
1. Glyn and Sutcliffe (1972) measured profits (including interest but less stock
appreciation and depreciation) arising from production in the company
sector in the UK. They found that the before-tax share of profits in net
output fell from 25.2 per cent in 1950 to 12.1 per cent in 1970 (ibid., Table
3.2, p. 58), while the before-tax rate of profit fell from 16.5 per cent in 1950
to 9.7 percent in 1970 (ibid., Table 3.3, p. 66). The after-tax rate of profit fell
from 6.7 per cent in 1950 to 4.1 per cent in 1970 (ibid.), Glyn and Sutcliffe
offer the following explanation: 'the decline in the share of profits and the
rate of profit in the U.K. has resulted from the combination of international
competition and wage pressure' (ibid., p. 70).
2. Boddy and Crotty (1975) based their research on a study of the St Louis
Federal Reserve Bank which showed that, in the various expansionary
periods between 1947 and 1972, wages managed to squeeze profits in the
non-financial corporate sector resulting in 'a pronounced decline in the ratio
of profits to wages' (ibid., p. 5). During this period, the percentage ratio of
before-tax net profits (that is, profits less depreciation and net interest) to
gross product fell from 19.4 per cent to 10.6 per cent (ibid.).
Notes 235
3. Weisskopf's (1979) research was based upon data collected by the US

National Income and Product Accounts and related to the non-financial
corporate business sector of the economy (which accounts for 60 per cent of
the gross national product). Weisskopfs research shows that the rate of
profit in this sector (defined as the ratio of 'before-tax net capital income
including corporate profits plus net interest' to 'the net stock of capital
including both fixed capital and inventories' (ibid., p. 349)) fell from an
average of 11.5 per cent in 1949 to 6.5 per cent in 1975 (ibid., p.350)
amounting to an average annual decline of 1.2 per cent over the period. 'This
decline' he argues 'was attributable almost entirely to a decline in profit
share' (ibid., p. 352). Weisskopf's conclusion is that: 'The long-term decline
in the rate of profit from 1949 to 1975 was almost entirely attributable to a
rise in the true share of wages, which indicates a rise in the strength of
labour' (ibid., p. 370).
4. Heap (1980) shows, on the basis of OEeD data (Hill, 1979), that pre-tax
profits in the manufacturing sector and in the transport and industry sector
fell precipitately in all of the major industrialised countries in the period
1955-76 as a result of ' a squeeze on the share of output going to profits and
a rise in the capital-output share' (Heap, 1980, p. 66).
5. The meaning of the terms 'long-run' and 'short-run' in the present context is
slightly different from the conventional meaning. In the conventional usage
(see Lipsey, 1971, p. 212; Bannock, Baxter and Rees, 1972, pp. 260, 273), the
term long-run refers, at the microeconomic level, to the period of time which
a firm requires to adjust output (and, by implication, the inputs necessary to
produce that output) to a change in demand. The length of the long-run will
be determined by that input which takes the longest time to adjust and will
be equal to the latter. At the macroeconomic level, the term long-run is used
more generally to refer to the period of time required for underlying
'structural' changes (for example, in employment, output, investment, and
so on) to occur in the economy. By contrast, the term short-run refers, at the
microeconomic level, to the period of time during which it would be
impossible for a firm to adjust its output (and, by implication, the inputs
necessary to produce that output) to a change in demand. The length of the
short-run will be determined by that input which takes the longest time to
adjust and will be less than the latter. At the macroeconomic level, the term
short-run is used more generally to refer to the period during which
underlying 'structural' change could not possibly occur.
In the present context, the problem arises with the term long-run since it
cannot be assumed to end endogenously. As a result, its length remains
indefinite. Thus it is necessary to specify that the long-run can only be ended
exogenously (for example, by state intervention) and it is this which
determines its length. (Weisskopf, 1978, p. 242, uses a similar definition.) It
is only because of this additional specification that it is possible to describe
Marx's law of the falling rate of profit as a long-run phenomenon.
6. It should be emphasised that the issue here is not whether there exists a
tendency for the rate of profit to fall and a counter-tendency for it to rise
(since Marx clearly acknowledged that both tendencies exist) but whether
236 Notes
one tendency is actually dominant. If neither is dominant then it makes no

sense to speak ofa 'law' (see Steedman, 1975a, pp. 79-80; Hodgson, 1974a,
pp.97-8).
7. It may be of some interest to recall one of the few studies which have
attempted to test Marx's theory against the empirical evidence of a particular
capitalist economy. Gillman (1957) made one such attempt by measuring
the organic composition of capital and the rate of profit (in terms of prices
rather than values) for the manufacturing sector in the US for the period
1849-1952. He found, in brief, that neither the organic composition of
capital nor the rate of profit consistently changed in the direction which
Marx's law would have predicted. However, even if Gillman's evidence had
shown a pattern identical to that predicted by Marx, there would still be the
question of the appropriateness of using price data as a measure of Marx's
'law', particularly since Marx always defined the rate of profit and its
tendency to fall in value terms (see Hodgson, 1974a, pp. 70-75; Desai, 1979,
pp.193-8).
8. Another cause of realisation crisis, not discussed here for reasons of space,
arises from the fact that different parts of constant fixed capital 'have
different durabilities' (Marx, 1974b, p. 457; see also pp. 453-73). This case
is discussed extensively by Kenway (1980). The reason for this type of
realisation crisis is that capital goods depreciate at different rates so that the
demand for the output of the sector producing those goods is likely to be
discontinuous and fluctuating, thereby giving rise to periods of overproduc-
tion (when demand is low) and underproduction (when demand is high).
This possibility arises even in the case of simple reproduction, contrary to
the interpretation of Bleaney (1976, p. 106) and Desai (1979, p.41).
9. Marx's models of reproduction, like most models in economics, abstracts
from what Robinson calls 'historical time' (Robinson, 1980, p. 223; see also
Bhaduri and Robinson, 1980, p. 105) in the sense that it does not consider
the historical process by which the economy has reached its current state of
equilibrium. It is assumed, for analytical purposes, that the present (as
distinct from the future) equilibrium has always existed and is identical in
each preceding period, stretching back indefinitely through what Robinson
calls 'logical time' (Robinson, 1980, p. 220; see also Bhaduri and Robinson,
1980, p. 105).
10. The same effect, it should be noted, would arise if the commodity - assuming
it is indispensable and non-substitutable - entered final consumption as a
wage good.
11. Malthus's Essay on Population was published in 1798 and put forward the
theory that population size would always be limited by the scarcity of
resources. His argument was that population would tend to increase at a
compound rate (like a geometric progression) whereas resources would tend
to increase at a simple rate (like an arithmetic progression) (see Blaug, 1962,
Chapter 3). As a consequence, if population were not voluntarily controlled
it would ultimately be controlled by starvation. This point can be illustrated
graphically by plotting the curve of an arithmetic progression (correspond-
Notes 237
12 c
11
10
9
B
8 Population
is
>:
e 7
....0""
6
~
~ 5
Resources
4
2 3 4 5
Years
Figure Nl The growth of population and resources according to Malthus
ing to the growth of resources) and the curve of a geometric progression

(corresponding to the growth of population). From Figure Nl it can be seen
that the line AB describes the growth path of resources over time while the
line AC describes the growth path of population over the same time. It
follows from that that X is the point where the growth in population and
resources are in equilibrium. The area to the left of X represents periods of
plenty, while the area to the right of X represents periods of scarcity (that is,
surplus population). The level of wages is the mechanism which determines
the balance (or imbalance) between population and resources. Marx's main
criticism of Malthus was that a surplus population (in relation to the
available resources) was not necessarily due to natural increase alone; it
could also be due to capital-using/technical innovation within capitalist
society itself.
12. Morishima and Catephores (1978) distinguish between a short-term and a

long-term theory of the supply oflabour in Marx's work (ibid., p. 121). They
claim that the industrial reserve army is the key element in Marx's short-
term theory while mechanisation is the key element in his long-term theory.
However this view is highly dubious since it is mechanisation which gives
rise, ceteris paribus, to the industrial reserve army. In other words
238 Notes
mechanisation and the industrial reserve army cannot be distinguished in

this way: they are both phenomena of the long-term.
13. Marx's theory of the falling rate of profit (see section 2.1) is perfectly
consistent with his theory of wages as long as it specified that the quantity of
constant capital must increase at a faster rate than the quantity of surplus
value (Walker, 1978, p.171). It should perhaps be added that the
consistency of these two theories is no proof that they are empirically
correct.
14. As note 1.
15. As note 2.
16. As note 3.
17. As note 4.
3 Marx's Theory of Rent
1. It is for this reason that Sraffa classifies land as a 'non-basic' (Sraffa, 1960,
p. 74). His justification is that since different plots of land are 'employed in
production, but not themselves produced they are the converse of com-
modities which, although produced, are not used in production. They are, in
fact, already included under the wider definition of non-basics' (ibid.). This
implies that land does not affect the rate of profit which seems to be in
contradiction with the fact that if production of a basic commodity is
extended to land of inferior quality then, ceteris paribus, it will cause a
decline in the rate of profit (see section 3.1). There is, however, no
contradiction since Sraffa's distinction between basic and non-basic
commodities applies to commodities being produced at a given point in time
and under given technical and social conditions of production (land being
one such given condition). Once these given conditions change (as occurs,
for example, when production is extended to inferior land), then the
relationships between wages, prices and profits, postulated by Sraffa, also
change (see Roncaglia, 1978, Chapter 3, especially p. 61).
2. It is possible to construct a typology of land according to its quantity and
quality from which it is possible to deduce the type of situations where rent
will arise if that land is owned by landowners and leased to capitalists for the
production of commodities (see Table N2).
Table N2 A typology of land
Quantity of land
Scarce Plentiful
Quality of Variable (1) Rent (2) Rent
land Invariable (3) Rent (4) No rent
Notes 239
It can be seen from this typology that rent will arise in all cases irrespective of
the variability in quality «(I) and (3)) and will also arise, even when land is
plentiful (2), as long as there are not unlimited quantities of one particular
type ofland. The only case where no rent will arise is where land is invariable
in quality and plentiful in quantity (4). The typical situation in capitalist
societies is where land is scarce and variable in quality (I), or plentiful but
variable in quality with the supply of each type ofland being limited (2). The
correctness of this typology and its implications will become apparent as the
analysis of rent unfolds in the subsequent sections (3.1 to 3.4).
3. The terms input, output, net input and net output refer throughout to a
quantity of goods 'calculated at the ruling levels of the rate of profits, wages
and prices' (Sraffa, 1960, p. 75).
4. It should be noted that the fertility of any plot of land in a capitalist society
can only be known after it has been used in the production of commodities
and, consequently, its profitability has been established. Thus fertility, far
from being an invariable and inherent property of the land, as Ricardo and
Marx believed, is crucially dependent upon the prevailing profit (and wage)
rates in each capitalist economy (see Sraffa, 1960, p.75; Montani, 1975,
pp. 71-9; Kurz, 1978, pp. 20- 23).
5. In a capitalist system, the scarcity ofland, and hence of corn, could arise in a
situation of simple reproduction (that is, a zero rate of economic growth) if
workers gained an increase in real wages (assuming corn to be the only wage
good) or in a situation of extended reproduction (that is, a positive rate of
economic growth) if further expansion was curtailed because of the scarcity
of corn. Thus, a basic precondition of differential rent I (and, indeed of all
rent) is that land is scarce (see Sraffa, 1960, p. 76) or, more precisely, that
land is perfectly inelastic in supply (see Keiper et al., 1961, pp. 108-10).
6. If the first option is adopted the consequences which follow will depend
upon whether the efficiency of the new technology (as measured by the
output per unit of input) is greater than, less than, or the same as the old
technology. If the new technology is more efficient than the old and if wages
are constant, then the consequence will be a high rate of profit. The effect of
this, in turn, will be that competition between capitalists will reduce the rate
of profit to its level before the technological innovation and the owner ofthe
land will receive a rent. If the new technology is of the same efficiency, then
there will be no change. In other words, the rate of profit will remain the
same and no rent will be paid. If the new technology is less efficient then it
will not be adopted.
7. There are two alternatives to this assumption. The first is that the
commodity being produced is a basic but the wage is variable (both upwards
and downwards) depending on workers' power. If, for example, it is
assumed that workers' are powerless, then the effect of extending produc-
tion from type (i) to type (ii) land will be an increase in the unit price of
production of the commodity leading to a corresponding reduction in real
wages (since the commodity is a basic). This will lead, in turn, to a higher
rate of profit for the capitalist(s) producing on type (i) rather than on type
240 Notes
(ii) land. This will lead, in tum, through competition, to the emergence of
differential rent I on type (i) land since the capitalist(s) producing on type (ii)
land would be willing to pay a rent for the least of type (i) land up to the
point where the rate of profit was the same on both plots. In this case
differential rent I would arise without any reduction in the overall rate of
profit.
The second possible assumption is that the commodity is a non-basic and,
in this case, the effect will be to increase the price of production of the
commodity (because of the higher unit costs of producing on type (ii) land),
thereby giving rise to a higher rate of profit for those capitalists producing
on type (i) land. However, this higher rate of profit will be reduced to that
attainable on type (ii) land since the capitalists producing on type (ii) land
would be willing to pay a rent for the lease of type (i) land up to the point
where the rate of profit on both types of land was the same. In this way
differential rent I emerges on type (i) land, although the rate of profit
throughout the economy will not be affected.
8. The fact that differential rent I arises independently of the existence of the
actual ownership of land (whether the owners be the state, landowners or
capitalists) means that it will exist in any system of production where land is
scarce and where there is more than one technique of production (due either
to the variable quality of the land or to the different combinations of inputs
invested in it). The fundamental difference between differential rent I in a
capitalist economy, and in a centrally planned economy, stems from the way
in which the prices of commodities are determined in these two economies: a
centrally planned economy would not include profit in the capitalist sense
and this would evidently affect prices as well as the size and calculation of
differential rent I. A centrally planned economy could not abolish
differential rent I without also abolishing a price system where each
commodity has a single and uniform price throughout the economy.
One implication of this is that differential rent I is perfectly consistent
with the capitalist economy in the sense that it does not hinder the process of
capital accumulation. Ball argues otherwise, claiming that if capitalists
(notably farmers) owned land rather than leasing it from landowners, that
this would lead to more investment in land and hence to more capital
accumulation. He uses two arguments to support this contention. The first
is that if farmers owned their own land then 'the drive for accumulation by
capitalist farmers would result in additional capital being invested in each
soil type until prices of production had been equalized across all lands under
cultivation' (Ball, 1977, p. 390). This argument implies that those capitalist
farmers who appropriate differential rent I (that is, surplus profits) will
continue to invest until this surplus disappears. No rational calculating
capitalist would undertake such a course of action and there are no
competitive forces obliging him/iher to do so.
Bail's second argument is that when a tenant farmer (as distinct from the
capitalist farmer) pays a differential rent I to a landowner he/she is hindered
from making any additional investment since 'there is no reason why the
landlord should acquiesce to a loss of rent which must ensue if the farmer
were to increase output by additional investment of capital at the expense of
rent' (ibid., p.391). This assertion is dubious for three reasons. First,
Notes 241
because any investment which increases output will actually result in an

increase in rent, leaving the rate of profit unchanged (assuming that the
commodity being produced is a basic and that real wages are constant).
Second, although tenant farmers cannot possibly increase the rate of profit
(because of the combination of constant wages and competition) they may
still be obliged to invest and innovate if only to maintain the existing rate of
profit. Third, the period of the lease is a crucial factor here and this, as will be
seen in section 3.2, could allow the tenant farmer appropriate differential
rent II for the duration of the lease which may be a sufficient incentive to
invest.
9. This is probably the reason why Ricardo expected the stationary state to
emerge. Ricardo's prognosis of the stationary state was based on the
premise that a growing economy would give rise to a growing population
which, in tum, would give rise to a growing demand for food (typically,
com). As a result of this, the production of com would have to be extended
to land of inferior quality, thereby giving rise to diminishing returns. This in
tum would cause com prices to rise which, given a constant real wage,
would cause profits to fall and rents to rise. This declining profit could only
be arrested, Ricardo argued by a fall in the price of com made possible by
either (a) a fall in real wages or (b) improvements in agriculture or (c) the
discovery of new markets. Ricardo was particularly concerned with (c),
which formed the basis of his argument against the Com Laws. He regarded
wages as 'more or less permanent' (Ricardo, 1966, p. 22) and improvements
as unable to 'overcome the natural impediments' (ibid., p. 32) of poor lands.
The result was that, given the persistence of the Com Laws, and given that
the rate of profit in manufacture moves in the same direction as that in
agriculture (ibid., p. 12) the rate of profit in the economy as a whole will fall
continually, leading inevitably to the 'stationary state'. The fact that the
increase in the price of agricultural commodities may not be inevitable, as
Marx pointed out, thus severely undermines Ricardo's case that the
stationary state is inevitable.
10. There is one difficulty with Marx's formulation of this point. This difficulty
arises because Marx tends to regard two different technologies as involving
two different amounts of capital: the more efficient the technology the larger
the amount of capital and vice versa. This is Marx's characteristic view of
technology which led him, as has been seen (section 2.1), to the erroneous
view that technical change within capitalism would inevitably lead to a
rising organic composition of capital (and, ceteris paribus, to a falling rate of
profit). The relevant point here is that a technical change does not
necessarily imply a larger amount of capital; it could equally well involve a
smaller amount or even the same amount. It is thus important to avoid
describing different techniques in terms of their amounts of capital; the
economically relevant description is in terms of their efficiency.
II. The term 'monopoly rent I' is used here, in preference to Marx's term
'absolute rent', because it brings out clearly the fact that this type of rent
arises from monopolistic practices, that is, collusion, rather than competi-
tion, between landowners. As with differential rent, there are two types of
monopoly rent.
242 Notes
12. Marx also believed that this inhibited investment in agriculture although as
Gibson and Esfahani (1983) point out, this depends upon the savings and
investment behaviour of landlords. They write: 'if landlords have the same
savings behaviour as capitalists, it is not at all clear how total investment is
supposed to diminish. If, on the other hand, landlords consume more of
their total income, which historically does seem to have been the case,
investment will fall' (ibid., p.95).
4 Marx's Theory of Productive and Unproductive Labour
1. The use of the terms 'Definition I' and 'Definition II' to distinguish between
the two different ways in which Marx used this distinction, does not imply
any chronological sequence in his usage nor does it imply that he was even
aware of these different usages.
2. Ibid.
3. Marx uses the terms productive and unproductive in the 'capitalist sense'
(Marx, 1969, p.153). In other words, they are categorisations of labour
from the point of view of capitalist production. He distinguishes this from
the 'absolute sense' (ibid.; ibid., p. 393) of the terms which treats labour as
productive (or unproductive) if it produces (or fails to produce) a use-value.
4. The distinctive feature of the capitalist sector (in contrast to the non-
capitalist sector) is that surplus labour is accumulated as capital (rather than
in some other form). This is clearly revealed by Marx's example of the self-
employed tailor working for a client. He writes
It may be that the quantity of labour performed by the jobbing tailor is
greater than that contained in the price he gets from me . . . This,
however, is all the same as far as I am concerned ... it is in no way a
means to any enrichment, any more than any other way for spending
money for personal consumption is a means of enrichment to me (Marx,
1969, p.402).
In other words, for labour to be productive, it must not only produce
surplus value, but in addition this surplus value must be accumulated by the
capitalist. Marx's argument, it will be apparent, assumes that the self-
employed tailor does not accumulate his own surplus labour as capital. In
this case his labour is unproductive according to 'Definition 1'.
5. By definition, wages can never function as capital since they are wholly
consumed to meet the subsistence requirements of the workers.
6. Marx points out that there are two 'contradictory definitions' of productive
labour in Adam Smith (Marx, 1969, p. 156; see also Dobb, 1973, p. 61). On
the other hand there is the 'correct' definition where Smith defines as
productive all labour which 'produces surplus value for the capitalist'
(Marx, 1969, p. 152). Marx regarded this definition of productive labour as
one of Smith's 'greatest scientific merits' (ibid., p. 157). On the other hand,
there is Smith's 'wrong conception of productive labour' (ibid., p.155)
Notes 243
where the latter is defined as that which produces a 'vendible commodity' in

contrast to 'services which generally perish in the very instant of their
performance' (ibid., p. 163).
Marx's first definition of productive labour corresponds to Smith's
'correct' definition. However, there is some evidence, as Bullock (1973,
p. 88) and Hunt (1979, p. 314) have pointed out, that Marx adopted Smith's
second criterion of the material nature of the product in his later work as a
'second, different and subsidiary definition' (Marx, 1969, p. 155). Hunt
concludes from this that 'the issue of whether productive workers create a
physical commodity is treated in a very confusing, if not contradictory,
manner by Marx (ibid., p. 314). Bullock, by contrast, defends Marx on this
usage with the highly questionable assertion that 'immaterial products
cannot be accumulated' even though 'individual capitalists may accumulate
money capital through their control of the production and sale of such
commodities' (Bullock, 1973, p. 89). Subsequently, however, Bullock (1974)
abandons this position in favour of that advanced here.
7. Marx's failure to consider the effect of the non-capitalist sector upon the
rate of accumulation in the capitalist sector can also be seen in his discussion
of petty commodity producers (that is, handicraftsmen and peasants).
These, according to Marx, 'belong neither to the category of productive nor
unproductive labourers' (Marx, 1969, p. 407). However, this is true only ifit
is assumed that there is no trade between petty commodity producers and
the capitalist sector. The existence of trade between the two would clearly
have an effect on capital accumulation and hence, in this case, would make
petty commodity producers, either productive or unproductive (according
to 'Definition 1'). Harrison makes a similar mistake to that of Marx by
labelling as 'non-productive ... all wage labour not performed under the
capitalist mode of production but financed out of revenue generated in the
capitalist sector' (Harrison, 1973, p. 74).
8. Marx did not classify all storage costs as circulation costs and hence as
unproductive according to 'Definition II'. He classified as unproductive
those storage costs which are 'due merely to the time required for the
conversion of existing values from the commodity form into the money
form' (Marx, 1974b, p. 142). By contrast, those costs of storage which,
strictly speaking, form part of the production process (such as distilling) he
classified as productive. By the same argument he also classified transport
costs as productive since they too are the 'continuation of a process of
production within the process of circulation' (ibid., p. 155).
9. Gough (1972, p.66) acknowledges that these two different definitions of
productive and unproductive labour coexist in Marx but he fails to note or
tease out their inconsistencies while O'Connor (1975, p. 305) is apparently
unaware of any inconsistency.
10. Lebowitz, in an attempt to explore the contemporary relevance of Marx's

tretament or circulation costs refers to 'the tension between the desire to
reduce circulation costs to a minimum and the requirements of realization'
(Lebowitz, 1972, p. 337) which results in 'the secular growth in wasteful,
244 Notes
unproductive activity' (ibid.). Lebowitz accepts unquestionably the Marx-

ian assumption that circulation costs are unproductive and, as a result, fails
to address the crucial problem with Marx's second definition, that is, the
problem of distinguishing between those costs which are technically
indispensable to the production of a commodity in any mode of production
and those which are necessary only in a capitalist mode of production. As a
result, Lebowitz's characterisation of circulation costs as 'wasteful' is open
to question.
5 The Different Marxist Traditions of CasteUs and Lojkine
I. Two qualifications to this characterisation of Marx's historical materialism

are necessary. The first is that Marx himself never fully or rigorously
developed the theory of historical materialism. Indeed the term 'historical
materialism' was never used by Marx, and only first appears in the writings
of Engels (see Giddens, 1971, p. 4). The second qualification is that Marx's
empirical use of historical materialism, particularly in The Eighteenth
Brumaire (Marx, 1979) is a great deal more nuanced and sophisticated than
his theoretical sketch of it in 'Preface to a Contribution to the Critique of
Political Economy' (Marx, 1976).
Both of these qualifications suggest that Marx's nascent theory of
historical materialism does not necessarily imply a simple (or simplistic)
determinism of the superstructure of the base. Althusser's work is devoted
precisely to clarifying the nature of the determinism between the base and
the superstructure.
2. Although this homology exists in capitalism, it is not present in earlier
modes of production and it is this, according to Balibar, which 'explains'
(Althusser and Balibar, 1970, p. 215) why Marx constantly confounds them
in a single concept' (ibid.). In feudal society, according to Balibar, the
peasants generally retained control of the means of production (that is, the
relation of real appropriation) while the feudal ruling class maintained
economic ownership (that is, the relation of property). In capitalism, by
contrast, the capitalist performs the 'double function' (Althusser and
Balibar, 1970, p. 214) of being the owner of the means of production and
hence the ultimate exploiter oflabour (that is, the relation of property) and
the direct organiser and controller of production (that is, the relation of real
appropriation).
3. Both Balibar (Althusser and Balibar, 1970, p. 215) and Poulantzas (1973,
pp. 24- 5) once argued that the various instances of any mode of production
could be analysed in terms of a set of 'invariant' elements which were
combined by a 'variable' set of relations. Following criticisms of formalism
(see, inter alia, Laclau, 1975, pp. 103ff; Glucksman, 1974, p. 298) this claim
has been weakened to the assertion that every instance is composed of
various elements and relations which may vary from one mode of
production to another (Poulantzas, 1976, pp.78-80).
Notes 245
4. In view of this it is perhaps curious that the Althusserian school should

consistently and vehemently reject that their position is structuralist
(Althusser and Balibar, 1970, pp. 7, 226; Poulantzas, 1973, p.26; Castells
and de Ipola, 1976, pp. 128-31). The reason for their rejection of the
structuralist label is that they associate it exclusively with the work of Levi-
Strauss (1972). The term is evidently not confined to the latter and it does
appear to be an appropriate label to describe the Althusserian position.
5. Thus, for example, in Althusserian terminology, the state consists of a
political structure and a set of political practices which, in French, are
translated as Ie politique and la politique respectively.
6. It should be noted here that there is a basic inconsistency between the
Althusserian distinction between structures and practices on the one hand,
and their rejection of historicism and humanism on the other. This
inconsistency arises in the following way: since structures, according to the
Althusserians, impose limits on the variability of practices, the question
arises as to the cause (and hence explanation) of this variability. The only
plausible answer to this question is that this variability is due to historical
and subjective factors. However, such factors are labelled by the Althus-
serian school as historicism and humanism respectively, and are rejected as
having no explanatory role in relation to practices.
7. One of the major problems with this definition is that it fails to grasp what
Poulantzas himself terms 'the very specificity of the political' (Poulantzas,
1973, p.38). This is because practically everything can be construed as a
factor of cohesion (and hence part of the state); and in fact practically
everything is included as part of the state, according to Poulantzas's
definition. The result is that the political is not treated as a structure (in the
strict Althusserian sense but rather as, to quote Laclau, 'a quality which
pervades all the levels of a social formation' (Laclau, 1975, p. 100; see also
Miliband, 1972, p. 262; Jessop, 1977, p. 355).
8. Kirk (1982, p. 142) is clearly misinformed in writing that 'Lojkine ... like
Castells ... based his work on a structuralist reading of Marx following
that of Althusser'. Lojkine's work is based upon the theory of state
monopoly capitalism.
9. Lojkine use the terms monopoly firm, multinational firm and large firm as if
they were interchangeable, which they are not, since a monopoly firm is not
necessarily a multinational firm, and a multinational firm is not necessarily
a monopoly firm, and a large firm is not necessarily either a multinational or
a monopoly firm.
10. However, much of the disagreement is more apparent than real. This was
also the conclusion reached by Pickvance (1977a, pp. 219-27) in his review
of the debate between Lojkine and Poulantzas. He writes: 'What emerges
from our examination of the Poulantzas-Lojkine debate is that there remain
important differences of emphasis between the two writers, but that to a
striking extent the debate goes round in circles, with each accusing the other
of positions they deny' (ibid., p. 226).
246 Notes
6 The Marxist Urban Sociology of Manuel Castells
1. It is controversial on two grounds. The first is that the precise nature and
extent of the continuities and discontinuities between Marx, Hegel and the
classicals, upon which Althusser bases his proof that Marxism is scientific, is
a matter of some controversy between Marxists. In other words, Althusser's
'proof' that Marxism is scientific rests upon a particular interpretation of
Marx which is not shared by all Marxists. The second controversy is that a
theory which serves certain ideological functions may not necessarily be
unscientific, contrary to the AIthusserian view. The distinction between
scientific and non-scientific theories depends ultimately upon the decisions of
scientists, rather than upon the ideological effects of those theories.
2. Others, however, have been less impressed. Elliot (1980, p. 153) has described
it as 'a pretentious and ... a profoundly irritating book', while Glass (1977,
p. 669) has described it as 'a load of humbug'.
3. It should perhaps be emphasised that Castells does not explicitly state that
collective consumption is unprofitable because it requires a high (that is,
higher than average) organic composition of capital. However, this would
seem to be implicit in his argument.
4. All of this assumes that facts matter, yet this is an assumption which may not
be applicable to Castells's discussion of the law ofthe falling rate of profit. He
writes: 'That we are not able to find a long-term secular trend for the organic
composition to increase and the rate of profit to decrease does not contradict
the theory; an intrinsic part of the theory is that reality is the result of
conflicting forces' (Castells, 1980a, p.40).
5. In his original characterisation of the urban system (Castells, 1976b, p. 79;
1976c, p. 154), Castells refers to four elements (P, C, E, A) only. The element
(S) is only included in a later publication (Castells, 1977a, p. 238).
6. In a footnote to his elaboration of the concept of 'urban actors' Castells
claims to reject 'the formalism of universal taxonomies' (Castells, 1969a,
p. 442, fn43) and emphasises the need for 'theoretical relevant variables'
(ibid.) Such evident inconsistency between Castells's claims and his actual
achievements suggests the contemporary relevance of Locke's observation
which seems to apply aptly to all of Castelis's work: 'Vague and insignificant
forms of speech, and abuse oflanguage, have so long passed for mysteries of
science; and hard and misapplied words, with little or no meaning, have, by
prescription, such a right to be mistaken for deep learning and height of
speculation, that it will not be easy to persuade either those who speak or
those who hear them that they are but the covers of ignorance, and
hinderance to true knowledge' (Locke, 1., 1964, p.58).
7. The title of this article is: 'Theoretical Propositions for an Experimental
Study of Urban Social Movements' (Castells, 1976c). Castells's justification
for using the term 'experimental' is that he assumes 'as constant all elements
not included in a particular analysis' (ibid., p. 172). However, an analysis can
only be experimental if certain variables are controlled and they can only be
Notes 247
controlled by manipulation not by assumption. As Schnore has pointed out:

'To ignore a factor is not to control it' (Schnore, 1965, p. 388). Thus Castells's
use of the term 'experimental' in this context is unjustified.
7 The Marxist Urban Sociology of Jean Lojkine
l. Lojkine is here implicitly (Lojkine, 1977a, p. 124) and later explicitly (ibid.,
pp. 14S-6) critical of the way in which Castells defines an urban area in terms
of collective consumption, thus limiting it to only one of the general
conditions of production (see section 3.2.2). Lojkine rejects this definition as
'one of the dominant themes of bourgeois ideology' (Lojkine, 1977a, p. 124).
2. Lojkine actually contradicts this claim in one place when he describes the
growth in investment in property (particularly urban property) by monopol-
istic companies as 'an important instrument for warding off, the fall in the
rate of profit in the industrial sectors' (Lojkine, 1977a, p. 287; see also p. 282).
This statement clearly contradicts claim (b) above.
3. The acronym PADOG is an abbreviation of Plan d'amenagement et
d'organisation generale. It was, in 1960, the plan for the organisation and
development of the Paris region whose notable feature was a policy
(ultimately unsuccessful) to contain the growth of Paris.
4. The acronym SDAU is an abbreviation of Schema directeur d'amenagement
et d'urbanisme. It became, in 1965, the successor to PADOG (1960) and was
itself, in tum, revised in 1969 in the light of the 1968 census, as well as in the
light of the notable political events of that year. The most characteristic
feature of SDAU (l96S) was the creation of eight new towns in the Paris
region.
S. The acronym ZEAT is an abbreviation of Zone d'erudes et d'amenagement du
territoire. There are eight such zones in France which were devised under the
Sth National Plan (l971-S) by the Commissariat au Plan for planning
purposes in the context of both France and the EEC.
6. The Commissariat au Plan is a technical body whose function it is to prepare
long and medium-term plans for the central government. It has, to date,
prepared seven national plans.
7. The acronym DAT AR is an abbreviation of Delegation a {'amenagement du
territoire et a {'action regionale. It was set up in 1963 for the purpose of co-
ordinating area and regional plans. Although it is attached to the Ministry of
the Interior, it has no direct powers.
8. This is not the only weakness which has been identified in Lojkine's study of
Paris. Pickvance (l977a, pp.238-4S) has shown, in a painstaking and
thorough critique, that Lojkine's study is also rather weak in its analysis of
the precise relationship between social classes and state policies. The critique
presented here, by contrast, is primarily interested in identifying the extent to
which the theoretical themes developed in Lojkine's urban sociology are also
pursued in his empirical research.
248 Notes
8 From Marxist to Post-Marxist Urban Sociology
I. The term actor is applied to groups and organisations in the Weberian sense
that the latter are 'the resultants of the particular acts of individual persons'
(Weber, 1968, p. 13). Weber adds: 'When reference is made in a sociological
context to a state, a nation, a corporation, a family, or an army corps, or to
similar collectivities what is meant is ... a certain kind of development of
actual or possible social actions of individual persons' (ibid., p. 14).
2. Buildings and infrastructures are also found in rural areas so that the
essential difference between rural and urban areas seems to lie in the degree
of concentration of property (and hence in the level of spatial interdepen-
dence) rather than in the kind of property per se.
3. The other effect of restricting the supply of suitable land is that it increases
the price ofland and property and hence harms the interests of those wishing
to buy property while benefiting those who already own it. In Marxian
terms, this means that owners of land can appropriate monopoly rent I (in
addition to differential rent I) as a result of zoning restrictions (see sections
3.3 and 3.5).
4. This apparently straightforward definition of a labour market area is
extremely difficult to apply empirically because, as will be illustrated in
more detail below, there are a number of different labour market areas for
each of the different skill types oflabour and the boundaries of each of these
labour market areas tend to be imprecise and overlapping. British research
(see notably Hall et at., 1973; Johnson, Salt and Wood, 1974) has tried to
overcome these difficulties by simply defining labour market areas in terms
of the major commuting patterns within an urban area and have avoided the
problem of imprecise and overlapping boundaries by adopting arbitrary (if
reasonable) statistical conventions.
5. It is perhaps worth emphasising that household movement does not
necessarily involve a change of job (that is, labour migration). In other
words, labour migration is only a subset - and in Britain, possibly only a
relatively small subset (see Johnson, Salt and Wood, 1974, pp.5-6, 69,
113-14) - of total movement. Clark and Onaka (1983, p. 50) in a review of
18 different studies of migration in the US, UK, Canada, and New Zealand,
put forward the following classification of reasons for migration:
6. The correctness of this analysis can also be confirmed, somewhat more
casually, by a perusal of the job advertisements in different newspapers. Job
advertisements for highly skilled, non-manual workers such as professionals
and managers tend to be advertised in national newspapers, while
advertisements for unskilled manual workers are normally only to be found
(if at all) in local newspapers (see Johnson, Salt and Wood, 1974, p.34;
Evans and Russell, 1980, p. 213; Evans and Richardson, 1981, p. 113). This
confirms, from the demand side, that the size of the area in which employers
look for labour varies considerably with the level of skill of the labour being
demanded.
Notes 249
MOVE
/~
Forced Voluntary
Housing
~
ADJUSTMENT
"'- INDUCED
Employment
Space Job change
Quality/design Retirement
Cost
Tenure change
Neighbourhood Life cycle

Quality Household formation
Physical environment Change in marital
Social composition status
Pu blie services Change in household
size
Accessibility
Workplace
Shopping/school
Family/friends
Figure N2 Reasons for household relocation
A useful survey of recent research on migration (that is, research since 1975)
is contained in Clark (1982).
7. One of the implications of distinguishing between an urban area and a

labour market area is that the source of unemployment should be sought
within the labour market for each particular type of labour, rather than in
the area where the unemployed live. The significance of this in the British
context, according to Cheshire (1979), is that the high rates of unem-
ployment among unskilled workers in the inner areas of Britain's major
conurbations is due more to the lack of demand for that type oflabour vis-a-
vis the supply, rather than to the fact that this group live or are 'trapped'
(ibid., pp. 36ft) in the inner city (see also Metcalf, 1975; Corkingdale, 1980,
pp. 184, 190; Richardson, 1980, pp. 214-17).
8. Private ownership does not always exist in this pure form, however. For
example, a proprietor may lease a house or a plot ofland to a tenant, thereby
temporarily ceding rights (1) and (2) to the latter (in exchange for rent).
Similarly, the zoning of land by the state effectively abolishes the right (1 b)
oflandowners over 'their' land while price and rent controls are, in effect, a
reduction of right (2). Thus private ownership of property refers to a
number of different empirical situations and is not an undifferentiated
entity.
250 Notes
9. It should be noted that some goods can be simultaneously consumption

goods and capital goods (for example, roads, airports, telephones, and so
on). This, however, does not destroy the conceptual distinction between
capital goods and consumption goods since their role in the production
process is quite different.
10. The more usual term for a communal good is a public good. Both, however,
are identical since the criterion for defining them is non-excludability.
Private goods, by contrast, are excludable. Public and private goods can be,
in turn, either pure or impure according to whether they are, respectively,
non-rival (that is, the good is not affected by the number of individuals using
it) or rival (that is, the good is affected by the number of individuals using it).
Table N3 gives a complete typology of pure and impure public and private
goods.
Table N3 A typology of private and public goods
Private goods Public goods
Excludable Non-excludable
Pure private good Impure public good

Examples: Examples:
Rival food and drink roads, streets, etc.
household furniture libraries, museums, etc.
schools, buses,
amenity areas
Impure private good Pure public good

Examples: Examples:
Non-rival television transmissions national defence
regulated by licences
lighthouses
airport noise
air pollution
II. There is a large number of synonyms used for the term 'externality', such as:
spillover effect, external effect, uncontracted effect, neighbourhood effect
and third party effect. It is usually defined in the following way: 'The essence
of externalities, whether in production or consumption, is that their costs or
benefits are not reflected in market prices, and so the decision of the
consumer or firm creating the externalities ... generally does not take its
effect into account' (Bannock, Baxter and Rees, 1972, p. 159). The concept
of externality does not appear in Marx's work. It first appeared in 1890
(seven years after Marx's death) in Alfred Marshall's Principles of
Notes 251
Economics (1961) and it was Marshall's pupil at Cambridge, Arthur Pigou,

who brought the concept to prominence in 1920 in his Economics of Welfare
(1960).
12. On the term 'dwelling unit', the remark of Donnison and Ungerson (1982,
p.42) is apposite: 'At a conference called in 1966 by the United Nations
Economic Commission for Europe, the definition of "dwelling" was more
or less settled for the whole of Europe ... "Dwellings" are structurally
separate units built for people to live in, with an entrance opening on to the
street or a space within the building to which the public has access'. The term
dwelling unit will be used in conformity with this definition.
13. Thus, apart from one's dwelling, the neighbourhood is the most basic
spatial or territorial entity in which one lives. However, neighbourhoods, in
turn, exist within cities, cities within regions, regions within countries, and
so on. These different spatial levels or places coexist in a nested fashion and,
at each level, they can have an effect, independently of one's class or status
position, upon the quality of one's life and life chances. It is for this reason
that place, like class, can be a basis for shared interests and hence for
collective action (see Molotch, 1976; Logan, 1978; see also Smith, 1977 for
an insightful analysis of the independent effects of place on human welfare).
The analysis in this section will be focused upon the neighbourhood level
where, through housing, there can be an overlap between class-based and
place-based interests.
14. There has been considerable debate in the literature over the concept of
housing classes since it was first used by Rex and Moore in 1967 (see Rex
and Moore, 1967). The concept has been criticised by Marxists who, using
the term 'class' to refer to positions in relation to production, regard
'housing classes' as interest groups in relation to the distribution (not the
production) of housing; Weberians, too, have questioned whether the
differences of interest between housing classes are contingent rather than
enduring (see Saunders, 1980a, Chapter 2, for a clear synopsis ofthis debate;
see also Thorns, 1981; Pratt, 1982). As will be seen in the next subsection
(8.8.5), there are real differences of interest between tenure groups in
relation to housing, so that it is quite justified to refer to such groups as
housing classes (in the Weberian sense of having different amounts of
market power). Saunders has more recently argued that the term 'class'
should be confined to production-based divisions while the term 'sector'
should be used for consumption-based divisions such as divisions over
housing, health, education, and so on (Saunders, 1984, pp.206-7; see
Harioe, 1984, for a critique).
15. A tax can be regarded as a negative consumption good if it reduces the real
income of the consumer (that is, the taxpayer). In such cases its effect is the
same as that of any other negative (state or communal) consumption good,
and residents will tend to react to it in the same way. Thus there is some
evidence in the US that residents (particularly affluent residents) will tend to
migrate from high to low tax areas (see Cox, 1973, pp. 27ft). Aronson (1974)
found similar evidence of fiscal migration in a study of the Leeds area of
252 Notes
England between 1965 and 1972. The Tiebout model (Tiebout, 1956) was
the first attempt at explaining this type of migration within a neoclassical
economic framework.
16. The usual example of this is the movement of non-whites into a white
neighbourhood where the latter are prejudiced against the former. This
frequently leads to a temporary fall in house prices, during the period of
racial transition, followed by a rapid increase in price due to the
concentration of non-white demand in those particular areas. It is this, inter
alia, which probably explains why 'most researchers have concluded
that blacks do pay more than whites for housing of comparable size and
quality .. .' (Kain and Quigley, 1975, pp. 63-5ff; see also Wilkinson and
Gulliver, 1971, p.33).
17. A recent survey in Norman, Oklahoma, on attitudes to the location of
mental health facilities found that owner-occupiers were more opposed than
tenants to the location of such facilities in their area because of'the potential
impact of the facility on real estate values and resale possibilities' (Smith and
Hanham, 1981, p. 153). This difference in attitude would, ceteris paribus,
make owner-occupiers more likely than tenants to engage in collective
action.
Bibliography
ABRAHAM-FROIS, G. and BERREBI, E. (1979) Theory of Value, Prices and
Accumulation: A Mathematical Integration of Marx, von Neumann and Sraffa
(London: Cambridge University Press) (first published in French in 1976).
AGNEW, J. A. (1978) 'Market Relations and Locational Conflict in Cross-
National Perspective', in Cox, K.R. (ed.) Urbanization and Conflict in Market
Societies (London: Methuen) Chapter 6: 128-43.
ALBIN, P. S. (1971) 'Unbalanced Growth and Intensification of the Urban
Crisis', Urban Studies, June: 139-46.
ALCALY, R. E. (1975) 'The Relevance of Marxian Crisis Theory', in
Mermelstein, D. (ed.) The Economic Crisis Reader (New York: Vintage
Books): 132-8.
ALCALY, R. E. (1978) 'An Introduction to Marxian Crisis Theory', in The
Union for Radical Political Economics (eds) U.S. Capitalism in Crisis (New
York: Union for Radical Political Economics): 15-21.
ALCHIAN, A. A. (1965) 'Some Economics of Property Rights', II Politico, 30:
816-29.
ALCHIAN, A. A. and DEMSETZ, H. (1973) 'The Property Right Paradigm',
The Journal of Economic History, 33 (I): 16-27.
ALTHUSSER, L. (1969) For Marx (London: Allen Lane, The Penguin Press)
(first published in French in 1965).
ALTHUSSER, L. (1971) Lenin and Philosophy and Other Essays (London: New
Left Books, translated by Ben Brewster).
ALTHUSSER, L. and BALIBAR, E. (1970) Reading Capital (London: New
Left Books) (first published in French in 1968).
AMBROSE, P. (1977) 'The Determinants of Urban Land Use Change', in
Massey, D. and Ambrose, P. Values, Relevance and Policy, Section III, Units
25-6 (Milton Keynes: The Open University Press): 35-79.
AMBROSE, P. and COLENUTT, B. (1975) The Property Machine (Harmonds-
worth: Penguin Books).
ARMSTRONG, P., GLYN, A. and HARRISON, J. (1978) 'In Defence of
Value: A Reply to Ian Steedman: An Extended Review of Marx after Sraffa,
by Ian Steedman, London: New Left Books, 1977; Capital and Class, 5: 1- 31.
ARONSON, 1. R. (1974) 'Voting with Your Feet' New Society, 29 (621): 545-7.
BAILEY, M. 1. (1966) 'Effects of Race and of Other Demographic Factors on
the Values of Single-Family Homes', Land Economics, 42: 215-20.
BALCHIN, P. N. and KlEVE, J. L. (1982) Urban Land Economics (London:
The Macmillan Press).
BALE, J. (1978) 'Externality Gradients', Area, JO (5): 334-6.
BALL, M. (1977) 'Differential Rent and the Role of Landed Property',
International Journal of Urban and Regional Research, I (3): 380-403.
BALL, M. (1980) 'On Marx's Theory of Agricultural Rent: A Reply To Ben
Fine', Economy and Society, 9 (3): 304-39.
BANDYOPADHYAY, P. (1982) 'Marxist Urban Analysis and the Economic
Theory of Rent', Science and Society, 46: 162-96.
253
254 Bibliography
BANNOCK, G., BAXTER, R. E. and REES, R. (1972) The Penguin Dictionary

of Economics (Harmondsworth: Penguin Books).
BARAN, P. (1957) The Political Economy of Growth (New York: Monthly
Review Press).
BAUMOL, W. J. (1967) 'Macroeconomics of Unbalanced Growth: The
Anatomy of Urban Crisis', The American Economic Review, 57 (3): 415-26.
BAUMOL, W. (1974) The Transformation of Values: What Marx "Really"
Meant (An Interpretation)" Journal of Economic Literature, 12: 51-62.
BENETTI, C. (1976) Valeur et Repartition (Paris: Presses Universitaires de
Grenoble et Maspero).
BENTHAM, G. and MOSELEY, M. (1980) 'Socio-economic Change and
Disparities within the Paris Agglomeration: Does Paris have an "Inner-City
Problem"?, Regional Studies, 14: 55-70.
BERTHOUD, R. (1980) 'Employment in a Changing Labour Market', in
Evans, A. and Eversley, D. (eds) The Inner City: Employment and Industry
(London: Heinemann): 232-52.
BHADURI, A. (1969) 'On the Significance of Recent Controversies on Capital
Theory: A Marxian View', The Economic Journal, 79: 532-9.
BHADURI, A. and ROBINSON, J. (1980) 'Accumulation and Exploitation: an
Analysis in the Tradition of Marx, Sraffa and Ka1ecki', Cambridge Journal of
Economics, 4: 103-15.
BLAUG, M. (1962) Economic Theory in Retrospect (London: Heinemann
Educational Books).
BLEANEY, M. (1976) Underconsumption Theories: A History and Critical
Analysis (London: Lawrence Wishart).
BODDY, R. and CROTTY, J. (1975) 'Class Conflict and Macro-Policy: The
Political Business Cycle', The Review of Radical Political Economics, 7 (I):
1-19.
BOHM-BAWERK, E. VON (1949) 'Karl Marx and the Close of his System', in
Sweezy, P. (ed.) Karl Marx and the Close of his System by Eugen von Bohm-
Sawerk and Bohm-Sawerk's Criticism of Marx by Rudolf Hi/herding (Clifton,
New Jersey: Augustus M. Kelly): 9-118 (first published in German in 1896,
and in English in 1898).
BORTKIEWITC, L. VON (1949) 'On the Correction of Marx's Fundamental
Theoretical Construction in the Third Volume of Capital', in Sweezy, P. M.
(ed.) Karl Marx and the Close of his System by Eugene von Bohm-Bawerk and
Bohm-Bawerk's Criticism of Marx by Rudolf Hilferding: Appendix (Clifton,
New Jersey: Augustus M. Kelly): 197-21 (first published in German in July
1907 and in English in 1954).
BORTKIEWICZ, L. VON (1952) 'Value and Price in the Marxian System',
International Economic Papers, no. 2: 5-60 (first published in German in July
1907 and in English in 1949).
BOSE, A. (1980) Marx on Exploitation and Inequality: An Essay in Marxian
Analytical Economics (London: Oxford University Press).
BOURNE, L. S. (1981) The Geography of Housing (New York: John Wiley).
BREUGEL, I. (1975) 'The Marxist Theory of Rent and the Contemporary City:
A Critique of Harvey', in Political Economy of Housing Workshop (ed.)
Political Economy and the Housing Question (A first volume of papers
Bibliography 255
presented to the Housing Workshop of the Conference of Socialist Econ-

omists) London, pp. 34-46.
BROADBENT, T. A. (1975) 'An Attempt to Apply Marx's Theory of Ground
Rent to the Modern Urban Economy', Centre for Environmental Studies:
Research Paper No. 17, London.
BROADBENT, T. A. (1977) Planning and Profit in the Urban Economy
(London: Methuen).
BULLOCK, P. (1973) 'Categories of Labour Power for Capital', Bulletin of the
Conference of Socialist Economists, 2 (6): 82-99.
BULLOCK, P. (1974) 'Defining Productive Labour for Capitai', Bulletin of the
Conference of Socialist Economists: 1-15.
BYRNE, D. and BEIRNE, P. (1975) 'Towards a Political Economy of Housing
Rent', in Political Economy of Housing Workshop (eds) Political Economy
and the Housing Question (A first volume of papers presented to the Housing
Workshop of the Conference of Socialist Economists) London: 47-67.
CAIN, G. C. (1976) 'The Challenge of Segmented Labour Market Theories to
Orthodox Theory: A Survey', Journal of Economic Literature, 14 (3-4):
1215-57.
CASTELLS, M. (1967) Les Politiques d'Implantation des Entreprises Industrie-
lles dans la Region de Paris: Etude Sociologique (These pour Ie Doctorat de
3eme Cycle en Sociologie, Faculte des Lettres et Sciences Humaines,
Universite de Paris, Nanterre; Ecole Pratique des Hautes Etudes, Vleme
Section: Sciences Economiques et Sociales).
CASTELLS, M. (1968) 'Y a-t-il une sociologie urbaineT, Sociologie du Travail:
72-90.
CASTELLS, M. (1969a) 'Vers une Theorie sociologique de la planification
urbaine', Sociologie du Travail: 413-43.
CASTELLS, M. (l969b) 'Enterprise industrielle et developpement urbaine',
SynopSiS, Sept.-Oct.: 67-76.
CASTELLS, M. (1969c) 'Theorie et ideologie en sociologie urbaine', Sociologie
et Societes, I (2): 171-91.
CASTELLS, M. (l969d) 'Le Centre Urbain: Projet de Recherche Sociologique',
Cahiers Internationaux de Sociologie, May: 83-106.
CASTELLS, M. (1970a) 'Social Structures and Processes of Urbanization:
Intersocietal Comparative Analysis', Annales, August: 1155-99.
CASTELLS, M. (1970b) 'La renovation urbaine aux etats-unis: synthese et
interpretation des donnees actuelles', Espaces et Societes, I: 107 - 35.
CASTELLS, M. (l970c) 'Les nouvelles frontieres de la methodologie
sociologique', Social Sciences Information, 9 (6): 79-108.
CASTELLS, M. (197Ia) 'L'urbanisation dependante en amerique latine',
Espaces et Societes, 3: 5-23.
CASTELLS, M. (1971 b) 'La Sociologie et la question Urbaine', L 'Architecture
d'Aujourd'hui, Sept.: 92-7 (first published, in modified form, in 1969c).
CASTELLS, M. (1972a) 'Urban Renewal and Social Conflict in Paris', Social
Sciences Information, II (2): 93 -1l24.
CASTELLS, M. (1972b) 'Lutte de classes et contradictions urbaines: I'emer-
gence des mouvements sociaux urbains dans Ie capitalisme avance', Espaces et
Societes, 6-7: 3-8 (also published in 1972d, pp. 6-17).
256 Bibliography
CASTELLS, M. (l972c) La Question Urbaine (Paris: Maspero).

CASTELLS, M. (I 972d) Luttes Urbaines et Pouvoir Politique (Paris: Maspero).
CASTELLS, M. (1974) 'Controverse surle pouvoirlocal: remarques surl'article
de Pierre Birnbaum', Revue Franc:aise de Sociologie, 15 (3): 237-42.
CASTELLS, M. (1975a) 'Advanced Capitalism, Collective Consumption and
Urban Contradictions: New Sources of Inequality and New Models for
Change', in Lindberg, Leon N. (ed.) Stress and Contradiction in Modern
Capitalism (Lexington, Mass.: Heath): 175-97.
CASTELLS, M. (l975b) 'Immigrant Workers and Class Struggles in Advanced
Capitalism: the Western European Experience', Politics and Society, 5 (I):
33-66.
CASTELLS, M. (l975c) 'Urban Sociology and Urban Politics: From a Critique
to New Trends of Research', Comparative Urban Research, 3 (I): 7-13.
CASTELLS, M. (1975d) Sociologie de l'espace industriel, Editions Anthropos,
Paris.
CASTELLS, M. (l975e) 'La Condition Sociale de la Planification Urbaine: Le
Cas de la Region de Dunkerque', Recherches Sociologiques, November:
401-26.
CASTELLS, M. (197 6a) 'Is there an Urban Sociology?, in Pickvance, C. G. (ed.)
Urban Sociology: Critical Essays (London: Tavistock Publications): 33-59
(first published in French in 1968a).
CASTELLS, M. (1976b) 'Theory and Ideology in Urban Sociology', in
Pickvance, C. G. (ed.) Urban Sociology: Critical Essays (London: Tavistock
Publications: 60-84 (first published in 1969c).
CASTELLS, M. (l976c) 'Theoretical Propositions for an Experimental Study of
Urban Social Movements', in Pickvance, C. G. (ed.) Urban Sociology: Critical
Essays' (London: Tavistock Publications): 147-73.
CASTELLS, M. (1976d) 'Urban Sociology and Urban Politics: From a Critique
to New Trends of Research', in Walton, J. and Masotti, L. (eds) The City in
Comparative Perspective (London: John Wiley): 291-300 (first published in
1975c).
CASTELLS, M. (1976e) 'La Crise Urbaine aux Etats-Unis: vers la Barbarie',
Les Temps Modernes, February: 1177-240.
CASTELLS, M. (1976f) 'The Wild City', Kapitalistate, 4-5, 1-30.
CASTELLS, M. (1976g) 'The Service Economy and Postindustrial Society: A
Sociological Critique', International Journal of Health Services, 6 (4):
596-607.
CASTELLS, M. (l976h) 'Crise de I'Etat, Consommation Collective et Con-
tradictions Urbaines', in Poulantzas, N. (ed.) La Crise de I'Etat (Paris: Presses
Universitaires de France): 179-208.
CASTELLS, M. (l977a) The Urban Question: A Marxist Approach (London:
Edward Arnold) (first published in French in I 972c).
CASTELLS, M. (l977b) 'Les conditions sociale d'emergence des movements
sociaux urbains (a partir d'une enquete exploratoire sur les luttes urbaine dans
la region parisienne, 1968-73)', in International Journal of Urban and
Regional Research, I (I): 45-75 (translated and published in English in
1978a).
CASTELLS, M. (1977c) 'Towards a Political Urban Sociology', in Harioe, M.
Bibliography 257
(ed.) Captive Cities: Studies in the Political Economy of Cities and Regions
(London: John Wiley): pp. 61- 78.
CASTELLS, M. (l977d) 'Marginalitt~ urbaine et mouvements sociaux au
Mexique: Ie mouvement des "posesionarios" dans la ville de Monterry',
International Journal of Urban and Regional Research, 1 (1): 145-50.
CASTELLS, M. (1978a) City, Class and Power (London: Macmillan Press).
CASTELLS, M. (l978b) 'Urban Social Movements and the Struggle for
Democracy: The Citizens' Movement in Madrid', International Journal of
Urban and Regional Research, 2 (1): 133-46.
CASTELLS, M. (1980a) The Economic Crisis and American Society (New
Jersey, USA: Princeton University Press).
CASTELLS, M. (1980b) 'Cities and Regions beyond the Crisis: Invitation to a
Debate', International Journal of Urban and Regional Research, 4 (1): 127-9.
CASTELLS, M. (1981a) 'Toward a Sociological Theory of City Planning', in
Lemert, C. (ed.) French Sociology: Rupture and Renewal since 1968 (New
York: Columbia University Press): 374-95 (first published in French in
1969a).
CASTELLS, M. (1981b) 'Local Government, Urban Crisis and Political
Change' in Zeitlin, M. (ed.) Political Power and Social Theory: A Research
Annual, 2 (Greenwich, Conn.: Jai Press) pp. 1-19.
CASTELLS, M. (1983a) The City and the Grassroots: A Cross-Cultural Theory
of Urban Social Movements (London: Edward Arnold).
CASTELLS, M. (1983b) 'Crisis, Planning and the Quality of Life: Managing the
New Historical Relationships between Space and Society', Environment and
Planning D: Society and Space, I: 3-21.
CASTELLS, M., CHERKI, E., GODARD, F. and MEHL, D. (1974)
Sociologie des Mouvements Sociaux Urbains: Enquete sur la Region Par-
isienne: Volume I: Problematique Theorique-Methodologie Analyse des Ten-
dences Generales: Volume 2: Crise du Logement et Mouvements Sociaux (paris:
Ecole Des Hautes Etudes En Sciences Sociales, Centre d'Etude des
Mouvements Sociaux).
CASTELLS, M., CHERKI, E., GODARD, F. and MEHL, D. (1978) Crise du
logement et mouvements sociaux urbains: Enquete sur la region parisienne
(Paris: Mouton) (first published in 1974).
CASTELLS, M. AND de IPOLA, E. (1976) 'Epistemological Practice and the
Social Sciences', Economy and Society, 5: 111-44 (republished in 1979).
CASTELLS, M. AND de HIPOLA, E. (1979) 'Epistemological Practice and the
Social Sciences', in Freiberg, 1. W. (ed.) Critical Sociology European
Perspectives (New York: Irvington Publishers) 217-53 (first published in
1976).
CASTELLS, M., DELAYRE, H., DESSANE, C., GODARD, F.,
O'CALLAGHAN, c., PUlG, G. and SKODA, C. (1970) 'Paris 1970:
Reconquete Urbaine et Renovation-Deportation', Soci%gie du Travail:
485-513.
CASTELLS, M. and GODARD, F. (1974) (in collaboration with Balanowski,
V.) Monopo/ville: L'entreprise, L'Etat, L'urbain: Analyse des rapports entre
l'entreprise, l'Etat et l'urbain apartir d'une enquete sur fa croissance industrielle
et urbaine de fa region de Dunkerque (Paris: Mouton).
258 Bibliography
CASTELLS, M. and GUILLEMARD, A. (1871) 'La Determination des

Pratiques Sociales en Situation de Retraite; Sociologie du Travail: 282-307.
CHESHIRE, P. C. (1979) 'Inner Areas as Spatial Labour Markets: A Critique
of the Inner Area Studies', Urban Studies, 16: 29-43.
CHEUNG, S. (1978) 'The Myth of Social Cost', Hobart Paper No. 82 (London:
The Institute of Economic Affairs).
CLARK, G. and DEAR, M. (1981) 'The State in Capitalism and the Capitalist
State', in Dear, M. and Scott, H. J. (eds) Urbanization and Urban Planning in
Capitalist Society (London and New York: Methuen): 45-61.
CLARK, W. A. V. (1982) 'Recent Research on Migration and Mobility: A
Review and Interpretation', Progress in Planning, 18 (I): I-56.
CLARK, W. A. V. and ONAKA, J. L. (1983) 'Life Cycle and Housing
Adjustment as Explanations of Residential Mobility', Urban Studies, 20 (I):
45-57.
CLARKE, S. and GINSBURG, N. (1975) 'The Political Economy of Housing',
in Political Economy of Housing Workshop (eds) Political Economy and the
Housing Question (A first volume of papers presented to the Housing
Workshop of the Conference of Socialist Economists) London: 3-33.
COCKBURN, C. (1977) The Local State: Management of Cities and People
(London: Pluto Press).
COING, H. (1977) 'Monopolville: I'enterprise et la ville', International Journal
of Urban and Regional Research, I (I): 180-85.
COLLISON, P. (1963) The Cutteslowe Walls: A Study in Social Class (London:
Faber & Faber).
CORKINGDALE, J. T. (1980) 'Employment Trends in the Conurbations', in
COX, A. (1983) 'On the Role of the State in Urban POlicy-Making: The Case of
Inner-City and Dispersal Policies in Britain', in Pons, V. and Francis, R. (eds)
Urban Social Research: Problems and Prospects, Sociological Review Mon-
ograph 30 (London: Routledge & Kegan Paul): 31-45.
COX, K. R. (1973) Conflict, Power and Politics in the City: A Geographic View
(New York: McGraw-Hili).
COX, K. R. (1978) 'Local Interests and Urban Political Processes in Market
Societies', in Cox, K. R. (ed.) Urbanization and Conflict in Market Societies
(London: Methuen) Ch.4: 94-108.
COX, K. R. (1981) 'Capitalism and Conflict around the Communal Living
Space', in Dear, M. and Scott, A. J. (eds) Urbanization and Planning in
Capitalist Societies (London: Methuen) Ch.16: 431-55.
COX, K. R. and JOHNSTON, R. J. (1982) 'Conflict Politics and the Urban
Scene: A Conceptual Framework', in Cox, K. R. and Johnston, R. J. (eds)
Conflict, Politics and the Urban Scene (London: Longman) Ch. I: 1-27.
COX K. R. AND MCCARTHY, J. J. (1982) 'Neighbourhood Activism as a
Politics of Turf: A Critical Analysis', in Cox, K. R. and Johnston, R. J. (eds)
Conflict, Politics and the Urban Scene (London: Longman) Ch. 10: 196-219.
CULLINGWORTH, J. R. (1982) Town and Country Planning in Britain, 8th edn
(London: Allen & Unwin).
DALES, J. H., (1975) 'Beyond the Marketplace', Canadian Journal of Econ-
omics, 8: 484-503.
Bibliography 259
DAVIS, O. A. and WHINSTON, A. B. (1961) 'The Economics of Urban

Renewal', Law and Contemporary Problems, 26: 105-17.
DEAR, M. J. and LONG, J. (1978) 'Community Strategies in Locational
Conflict' in Cox, K. R. (ed.) Urbanization and Conflict in Market Societies
(London: Methuen) Ch. 5: 113-27.
DEMSETZ, H. (1967) 'Toward a Theory of Property Rights', American
Economic Review, 57: 347-59.
DEPARTMENT OF THE ENVIRONMENT (1980) Development Control
Statistics 1978-79 (London: Department of the Environment).
DESAI, M. (1979) Marxian Economics (Oxford: Blackwell).
DICKEN, P. (1976) 'The Multiplant Business Enterprise and Geographical
Space: Some Issues in the Study of External Control and Regional Develop-
ment', Regional Studies, 10: 401-12.
DICKINSON, H. D. (1957) 'The Falling Rate of Profit in Marxian Economics',
Review of Economic Studies, 24: 120- 30.
DMITRIEV, V. K. (1974) Economic Essays on Value, Competition and Utility,
translated by D. Fry and edited with an introduction by D. M. Nuti. (London:
Cambridge University Press) (Consists of three essays, first published
separately, in Russian, in 1892 and 1902, and first published in book form, in
Russian, in 1904 and in English in 1974).
DOBB, M. (1972) 'The Sraffa System and Critique of Neo-Classical theory
of Distribution', in Hunt, E. K. and Schwartz, J. G. (eds) A Critique of
Economic Theory: Selected Readings (Harmondsworth: Penguin Books):
205-21.
DOBB, M. (1973) Theories of Value and Distribution since Adam Smith: Ideology
and Economic Theory (Cambridge: Cambridge University Press).
DONNISON, D. and UNGERSON, C. (1982) Housing Policy (Harmonds-
worth: Penguin Books).
DOWNS, A. (1981) Neighbourhoods and Urban Development (Washington
D.C.: The Brookings Institution).
DREWETT, R. (1973) 'The Developers: Decision Processes', in Hall, P.,
Gracey, H., Drewett, R. and Thomas, R., The Containment of Urban England:
Volume 2: The Planning System: Objectives, Operations, Impacts (London:
Allen & Unwin) 163-93.
DUNLEAVY, P. (1979) 'The Urban Basis of Political Alignment: Social Class,
Domestic Property Ownership, and State Intervention in Consumption',
British Journal of Political Science, 9: 409-43.
DUNLEAVY, P. (1980) Urban Political Analysis: The Politics of Collective
Consumption (London: The Macmillan Press).
EDEL, M. (1976) 'Marx's Theory of Rent: Urban Applications', in Political
Economy of Housing Workshop (ed.) Housing and Class in Britain (A second
volume of papers presented to the Political Economy of Housing Workshop
of the Conference of Socialist Economists) London 7-23 (also published in
Kapitalistate, 4 (1976): 100-124).
ELIAS, P. and KEOGH, G. (1982) 'Industrial Decline and Unemployment in
the Inner City Areas of Great Britain: A Review of the Evidence', Urban
Studies, 19: 1-15.
ELKIN, S. L. (1974) Politics and Land Use Planning: The London Experience
(London: Cambridge University Press)
260 Bibliography
ELLIOTT, B. (1980) 'Manuel Castells and the New Urban Sociology', British
Journal of Sociology, 31 (2): 151-8.
ELSTER, J. 1985 Making Sense ofMarx (London: Cambridge University Press;
Paris: Editions de la Maison des Sciences de I'Homme).
ENGELS, F. (1958) The Condition of the Working Class in England, translated
from the German and edited by Henderson, W. O. and Chaloner, W. H.
(Oxford: Blackwell) (first written and published in German in 1845; first
published in English in 1892).
ENGELS, F. (1959) Anti-Duhring (Moscow: Foreign Languages Publishing
House) (first published in German as a series of articles in Vorwart in 1877 - 8
and in book form in 1878; first published in English in 1959).
ENGELS, F. (1974) 'Supplement to Capital, Volume Three', in Marx, K.
Capital: A Critique ofPolitical Economy. Volume III: The Process ofCapitalist
Production as a Whole (London: Lawrence & Wishart) (first published in
German in 1894; first published in English in 1909).
EVANS, A. W. and RICHARDSON, R. (1981) 'Urban Unemployment:
Interpretation and Additional Evidence', Scottish Journal of Political Econ-
omy, 28 (2): 107-24.
EVANS, A. and RUSSELL, L. (1980) 'A Portrait of the London Labour
Market', in Evans, A. and Eversley, D. (eds) The Inner City: Employment and
Industry (London: Heinemann): 204-31.
FIELDING, A. J. (1982) 'Counterurbanization in Western Europe' , Progress in
Planning, 17 (1): 1- 52.
FINE, B. (1979) 'On Marx's Theory of Agricultural Rent', Economy and
Society, 8 (3): 241-78.
FINE, B. (1980) 'On Marx's Theory of Agricultural Rent: A Rejoiner', Economy
and Society, 9 (3): 327-39.
FINE, B. and HARRIS, L. (1976) 'Controversial Issues in Marxist Economic
Theory', in Miliband, R, and Saville, J. (eds) Socialist Register (London:
Merlin Press): 141-78.
FLYNN, B. (1981) 'Managing Consensus: Strategies and Rationales in Policy-
Making' in Harloe, M., (ed.), New Perspectives in Urban Change and Conflict,
(London: Heinemann Educational Books): 50-62.
FREYSSENET, M., REGAZZOLA, T. and RETEL, J. (1971) Segregation
spatiale et deplacements sociaux dans l'agglomeration parisienne (Paris: Centre
de Sociologie Urbaine).
FURUBOTN, E. G. and PEJOVICH, S. (1972) 'Property Rights and Economic
Theory: A Survey of Recent Literature', The Journal of Economic Literature,
10: 1137-62.
GAREGNANI, P. (1972) 'Heterogeneous Capital, the Production Function
and the Theory of Distribution', in Hunt, E. K. and Schwartz, J.G. (eds) A
Critique of Economic Theory: Selected Readings (Harmondsworth: Penguin
Books): 245-91.
GARNIER, J. P. (1974) 'A propos de "la question urbaine", Espaces et
Societes, 8-9: 123-9.
GIBSON, B. and ESFAHANI, H. (1983) 'Nonproduced Means of Production:
Neo-Ricardians vs. Fundamentalists', Review of Radical Political Economics,
15 (2): 83-105.
Bibliography 261
GIBSON, B. and MCLEOD, D. (1983) 'Non-Produced Means of Production in

Sraffa's System: Basics, Non-Basics and Quasi-Basics', Cambridge Journal of
GIDDENS, A. (1971) Capitalism and Modern Social Theory: An Analysis of the
writings of Marx, Durkheim and Max Weber (London: Cambridge University
Press).
GILLMAN, J. (1957) The Falling Rate of Profit: Marx's Law and its
Significance to Twentieth Century Capitalism (London: Dobson)
GLASS, R. (1977) 'Verbal Pollution: Review of The Urban Question by M.
Castells', New Society, 41 (782): 667-9.
GLEAVE, D. and PALMER, D. (1978) 'Mobility of Labour: Are Council
Tenants Really Handicapped?', Centre for Environmental Studies Review, 3:
74-77.
GLUCKSMAN, M. (1974) Structuralist Analysis in Contemporary Social
Thought: A Comparison of the Theories of Claude Levi-Strauss and Louis
Althusser (London: Routledge & Kegan Paul).
GLYN, A. and SUTCLIFFE, B. (1972) British Capitalism, Workers and the
Profits Squeeze (Harmondsworth: Penguin Books).
GODARD, F. CASTELLS, M., DELAYRE, H., DES SANE, C. and
O'CALLAGHAN, C. (1973) La Renovation Urbaine a Paris: Structure
urbaine et logique de classe (Paris: Mouton).
GODDARD, J. B. (1977) 'Urban Geography: City and Regional System',
Progress in Human Geography, 1: 296-303.
GODDARD, J. B. (1978) 'Urban and Regional Systems', Progress in Geogra-
phy, 2: 309-17.
GOODCHILD, R. (1978) 'The Operation of the Private Land Market', in
Pearce, B. J., Curry, N. R. and Goodchild, R. N., Land, Planning and the
Market, Department of Land Economy, University of Cambridge,
Occasional Paper no. 9: 11-47.
GOUGH, I. (1972) 'Marx's Theory of Productive and Unproductive Labour',
New Left Review, 76: 47-72.
HABERMAS, J. (1976) 'Problems of Legitimation in Late Capitalism', in
Connerton, P. (ed.) Critical Sociology: Selected Readings (Harmondsworth:
Penguin Books): 363-87.
HALL, P., THOMAS, R., GRACEY, H. DREWETT, R. (1973) The Contain-
ment of Urban England, vol. I and II (London: Allen & Unwin).
HALLETT, G. (1979) Urban Land Economics: Principles and Policy (London:
Macmillan Press).
HARLOE, M. (1977) 'New Trends in Urban Sociology: Weberian and Marxist
Approaches', in Harioe, M. (ed.) Captive Cities: Studies in the Political
Economy of Cities and Regions (London: John Wiley): 1-47.
HARLOE, M. (1979) 'New Perspectives in Urban and Regional Research:
Progress and Problems', in Harioe, M. (ed.) New Perspectives in Urban
Change and Conflict (London: Heinemann Educational Books): 1-26.
HARLOE, M. (1984) 'Sector and Class: A Critical Comment', International
Journal of Urban and Regional Research, 8 (2): 228-37.
HARRIS, D. J. (1983) 'Accumulation of Capital and the Rate of Profit in
Marxian Theory', Cambridge Journal of Economics, 7: 311-30.
262 Bibliography
HARRISON, J. (1973) 'Productive and Unproductive Labour in Marx's

Political Economy', Bulletin of the Conference of Socialist Economists, 2 (6):
70-82.
HARVEY, D. (1973) Social Justice and the City (London: Edward Arnold).
HARVEY, D. (1976) 'Labour, Capital and Class Struggle around the Built
Environment in Advanced Capitalist Societies', Politics and Society, 6 (3):
265-96.
HARVEY, D. (1978a) 'Labour, Capital, and Class Struggle Around the Built
Environment in Advanced Capitalist Societies', in Cox, K. (ed) Urbanization
and Coriflict in Market Societies (London: Methuen) I: pp.9-37.
HARVEY, D. (I 978b) 'The Urban Process under Capitalism: A Framework for
Analysis', International Journal of Urban and Regional Research, 2 (I):
101-31.
HARVEY, D. (1981) 'The Urban Process under Capitalism: A Framework for
Analysis', in Dear, M. and Scott, A. J. (eds) Urbanization and Urban Planning
in a Capitalist Society (London: Methuen) chapter 5: 91-121.
HARVEY, D. (1982) The Limits to Capital (Chicago: The University of Chicago
Press).
HARVEY, D. and CHATTERJEE, L. (1974) 'Absolute Rent and the
Structuring of Space by Governmental and Financial Institutions', Antipode:
A Radical Journal of Geography, 6 (I): 22-36.
HEALEY, M. J. (1982) 'Plant Closures in Multi-plant Enterprises: The Case of
a Declining Industrial Sector', Regional Studies, 16 (I): 37-51.
HEAP, S.H. (1980) 'World Profitability Crisis in the 1970s: Some Empirical
Evidence', Capital and Class, 12: 66-84.
HILL, T. P. (1979) Profits and Rates of Return (Paris: OECD).
HIRSCH, F. (1978) Social Limits to Growth (London: Routledge & Kegan
Paul).
HIRSCHMAN, A. O. (1970) Exit. Voice and Loyalty: Responses to Decline in
Firms. Organizations. and States (Cambridge, Mass.: Harvard University
Press).
HODGSON, G. (1973) 'Marxist Epistemology and the Transformation
Problem', Bulletin of the Conference of Socialist Economists, 2 (6): 47-64.
HODGSON, G. (1974a) 'The Theory of the Falling Rate of Profit', New Left
Review, 84: 55-82.
HODGSON, G. (1974b) 'Marxian Epistemology and the Transformation
Problem', Economy and Society, 4: 357-93.
HODGSON, G. (1975) Trotsky and Fatalistic Marxism (Nottingham: Spokes-
man Books).
HODGSON, G. (1976) 'Exploitation and Embodied Labour-time', Bulletin of
the Conference of Socialist Economists, 4: 1-25.
HODGSON, G. (l977a) 'Sraffa, Value and Distribution: An Expository Essay
on the Capital Controversy', British Review of Economic Issues, 1: 44-55.
HODGSON, G. (1977b) 'Papering over the Cracks' in Miliband, R. and Saville,
J. (eds) Socialist Register (London: Merlin Press): 88-105.
HODGSON, G. (1982) Capitalism. Value and Exploitation: A Radical Theory
(Oxford: Martin Robertson).
HODGSON, G. and STEEDMAN, I. (1975) 'Fixed Capital and Value
Analysis', Bulletin of the Conference of Socialist Economists, 4 (3) 1- 7.
Bibliography 263
HOUSE, J. W. (1978) France: An Applied Geography (London: Methuen).

HOWARD, M. C. and KING, J. E. (1976) The Political Economy of Marx
(London: Longmans).
HOWARD, M. C. and KING, J. E. (eds) (1976) The Economics of Marx:
Selected Readings of Exposition and Criticism (Harmondsworth: Penguin
Books).
HUET, A., et al. (1973) Role et portee economique, politique et ideologique de la
Participation a l'amenagement urbain, L'Office social et culture! de Rennes,
France.
HUGHES, G. and MC CORMICK, B. (1981) 'Do Council Housing Policies
Reduce Migration Between Regions'? The Economic Journal, 91: 919-37.
HUNT, E. K. (1979) 'The Categories of Productive and Unproductive Labour
in Marxist Economic Theory', Science and Society, 43: 303-25.
INSTITUTE FOR SCIENTIFIC INFORMATION (1969-1984) Annual
Social Science Citation Index 1969-1984 (Philadelphia, Penn.: Institute for
Scientific Information).
JESSOP, B. (1977) 'Recent Theories of the Capitalist State', Cambridge Journal
of Economics, 1: 353-73.
JOHNSON,J. H., SALT,J. and WOOD,P. (1974) Housing and the Migration of
Labour in England and Wales (Lexington, Mass. Saxon House; Lexington
Books).
KAIN, J. F. and QUIGLEY, J. M. (1975) Housing Markets and Racial
Discrimination: A Microeconomic Analysis (New York: National Bureau of
Economic Research).
KEEBLE, D., OWENS, P. L. and THOMPSON, C. (1983) 'The Urban-Rural
Manufacturing Shift in the European Community', Urban Studies, 20 (4):
405-18.
KEIPER, J. S., KURNOW, E., CLARK, C. D. and SEGAL, H. H. (1961)
Theory and Measurement of Rent (Philadelphia, Penn.: Chilton Company).
KENWAY, P. (1980) 'Marx, Keynes and the Possibility of Crisis', Cambridge
Journal of Economics, 4: 23-35.
KIRK, G. (1982) 'Theoretical Approaches to Urban Planning', in Blowers, A.,
Brook, c., Dunleavy, P. and McDowell, L. (eds) Urban Change and Conflict:
An Interdisciplinary Reader (London: Harper & Row) pp. 135-47 (first
published in 1980).
KURZ, H. D. (1978) 'Rent Theory in a Multisectoral Model', Oxford Economic
Papers, 30 (1): 16-37.
LACLAU, E. (1975) 'The Specificity of the Political: The Poulantzas- Miliband
debate', Economy and Society, 5 (1): 87 -110.
LAMARCHE, F. (1976) 'Property Development and the Economic Founda-
tions of the Urban Question', in Pickvance, C. G. (ed.) Urban Sociology:
Critical Essays (London: Methuen) ch.4, pp.85-118 (first published in
French in 1972).
LAMBERT, J., PARIS, C. and BLACKABY, B. (1978) Housing Policy and the
State: Allocation, Access and Control (London: The Macmillan Press).
LANGE, O. (1968) 'Marxian Economics and Modern Economic Theory', in
Horowitz, D. (ed.) (1968) Marx and Modern Economics (New York: Monthly
Review Press): 68 -87 (first published in The Review ofEconomic Studies, June
1935.)
264 Bibliography
LAURIA, M. (1984) 'The Implications of Marxian Rent for Community -

Controlled Redevelopment Strategies', Journal of Planning Education and
Research, 4 (I): 16-24.
LAUTMAN, J. (1974) 'Review of Lojkine, J., 1972, La Politique Urbaine dans la
Region Parisienne 1945-1971, Revue Francaise de Sociologie, 15 (2): 27-7.
LEBAS, E. (1981) 'The New School of Urban and Regional Research: Into the
Second Decade', in Harloe, M. and Lebas, E. (eds) City, Class and Capital:
New Developments in the Political Economy of Cities and Regions (London:
Edward Arnold): ix - xxxiii.
LEBAS, E. (1982) 'Urban and Regional Sociology in Advanced Industrial
Societies: A Decade of Marxist and Critical Perspectives', Current Sociology,
30(1): 1-271.
LEBAS, E. (1983) 'The State in British and French Urban Research, or the
Crisis of the Urban Question', in Pons, V. and Francis, R. (eds) Urban Social
Research: Problems and Prospects, Sociological Review Monograph 30
(London: Routledge & Kegan Paul): 9-30.
LEBOWITZ, M. A. (1972) 'The Increasing Cost of Circulation and the Marxian
Competitive Model', Science and Society, 36: 331-8.
LEFEBVRE, H. (1968) Le Droit a la Ville (Paris: Anthropos).
LEFEBVRE, H. (1972) La Pensee Marxiste et la Ville (Paris: Caterman).
LEFEBVRE, H. (1973) Espace et Politique: Le Droit a la Ville II (paris:
Anthropos).
LEVI-STRAUSS, C. (1972) Structural Anthropology (London: Harmonds-
worth). (First published in French in 1958.)
LEY, D. aud MERCER, J. (1980) 'Locational Conflict and the Politics of
Consumption', Economic Geography, 56: 89-109.
LIPSEY, R. G. (1971) An Introduction to Positive Economics, 3rd edn (London:
Weidenfeld & Nicolson) (first published 1963).
LOCKE, J. (1964) An Essay Concerning Human Understanding edited by A. D.
Woozley (London: Collins) (first published in 1690).
LOGAN, J. R. (1978) 'Growth, Politics and the Stratification of Places',
American Journal of Sociology, 84 (2): 404-16.
LOJKINE, J. (1969a) 'Pour une analyse marxiste du changement social',
Sociologie du Travail, July-Sept.: 259-73.
LOJKINE, J. (1969b) Contribution aune tMorie marxiste des ideologies (Paris:
Cahiers du Centre d'etude et de recherche marxiste).
LOJKINE, J. (1971) 'Y a-t-il une rente fonciere urbaineT, Espaces et Societes, 2:
89-94.
LOJKINE, J. (1972a) La politique urbaine dans la region parisienne 1935-1972
(Paris: Mouton).
LOJKINE, J. (1972b) 'Pouvoir Politique et lutte des classes a I'epoque du
capitalisme monopoliste d'etat', La Pensee, 166: 142-68.
LOJKINE, J. (1972c) 'Contribution a une theorie marxiste de l'urbanisation
capitaliste', Cahiers Internationaux de Sociologie, 52: 123 -46.
LOJKINE, J. (1974a) La politique urbaine dans la region lyonnaise 1945-1972
(Paris: Mouton).
LOJKINE, J. (1974b) 'La Politique des Transports', in Les Cahiers du Centre
d'Etudes et de Recherches Marxistes (CERM), Urbanisme Monopoliste,
Urbanisme Democratique, CERM: 191-7.
Bibliography 265
LOJKINE, J. (1974c) 'Le projet de cite financiere a Paris', Espaces et Societes,

13-14: 111-34.
LOJKINE, J. (1975) 'Strategies des grandes entreprises, politiques urbaines et
mouvements sociaux urbains', Sociologie du Travail: 18-40.
LOJKINE, J. (l976a) 'Contribution to a Marxist theory of Capitalist
Urbanization', in Pickvance, C. G. (ed.) Urban Sociology: Critical Essays
(London: Tavistock Publications): 119-46 (first published in French in
I 972c).
LOJKINE, J. (1976b) Strategies des grandes entreprises et politique urbaine: Ie
cas des banques et des assurances (Paris: Centre Nationale de la Recherche
Scientifique: Centre d'etude des Mouvements Sociaux).
LOJKINE, J. (1977a) Le Marxisme, L'hat et la Question Urbaine (Paris: Presses
Universitaires de France).
LOJKINE, J. (1977b) 'Big Firms' Strategies, Urban Policy and Urban Social
Movements', in Harloe, M. (ed.) Captive Cities: Studies in the Political
Economy of Cities and Regions (London: Wiley): 141-56.
LOJKINE, J. (1977c) 'L'analyse marxiste de l'etat', International Journal of
Urban and Regional Research, I: 19-23.
LOJKINE, J. (1977d) 'L'etat et L'Urbain: Contribution Ii une analyse
materialiste des politiques urbaines dans les pays capitalistes developpes',
International Journal of Urban and Regional Research, I (2): 256-71.
LOJKINE, J. (l977e) 'Crise de I'Etat et Crise du Capitalisme Monopoliste
d'Etat', La Pensee, 193: 113-26.
LOJKINE, J. (1978) 'Sur l'usage du concept de contradiction dans une analyse
materialiste d'etat', La Pensee, 197.
LOJKINE, J. (1980) 'Politique urbaine et Pouvoir local', Revue de Sociologie
Fran(aise, 21 (4): 633-51.
LOJKINE, J. (1981a) 'For a Marxist Analysis of Social Change', in Lemert, C.
C. (ed.) French Sociology: Rupture and Renewal Since 1968 (New York:
Columbia University Press) pp. 355-69 (first published in French in 1969a).
LOJKINE, J. (198Ib) 'Urban Policy and Local Power: Some Aspects of Recent
Research in Lille', in Harloe, M. and Lebas, E. (eds) City, Class and Capital:
New Developments in the Political Economy of Cities and Regions (London:
Edward Arnold): 89-104.
LOJKINE, J. and PRETECEILLE, E. (1970) 'Politique urbaine et stratl!gie de
c1asse', Espaces et Socihes, I: 79-84.
LORIMER, J. (1972) A Citizen's Guide to City Politics (Toronto: James, Lewis
& Samuel).
MACPHERSON, C. B. (1973) 'A Political Theory of Property' ,in Macpherson,
C. B. Democratic Theory: Essays in Retrieval (Oxford: Clarendon Press):
120-40.
MACPHERSON, C. B. (1978) 'The Meaning of Property', in Macpherson, C.
B. (ed.) Property: Mainstream and Critical Positions (Oxford: Blackwell):
1-13.
MANDEL, E. (1968) Marxist Economic Theory (London: Merlin Press).
MARKUSEN, A. R. (1978) 'Class, Rent and Sectoral Conflict: Uneven
Development in Western U.S. Boomtowns' Review of Radical Political
MARRIOTT, O. (1969) The Property Boom (London: Pan Books).
266 Bibliography
MARSHALL, A. (1961) Principles of Economics (London: The Macmillan

Press) (first published in 1890).
MARX, K. (1969) Theories ofSurplus-value (Vol. 3 of Capital) Part II, Moscow:
Progress Publishers (first written in Gennan between 1861-1863; first
published in Gennan in 1905; first published in English in 1963).
MARX, K. (1973) Grundrisse: Foundations of the Critique of Political Economy
(London: Penguin Books, in association with New Left Books) (first written
in Gennan between October 1857 and March 1858; first published in Gennan,
in two volumes, in 1939 and 1941, and again in 1952; first translated and
MARX, K. (1974a) Capital: A Critical Analysis of Capitalist Production, vol. I
(London: Lawrence & Wishart) (first written in Gennan between 1865 and
1867; first published in Gennan in 1869; first published in English in 1887).
MARX, K. (1974b) Capital: A Critique ofPolitical Economy, vol. II: The process
of circulation (London: Lawrence & Wishart) (first written in Gennan
between 1865 and 1867; first published in Gennan in 1885; first published in
English in 1919).
MARX, K. (1974c) Capital: A Critique of Political Economy, vol. III: The
Process of Capitalist Production as a whole (London: Lawrence & Wishart)
(first written in Gennan between 1863 and 1865; first published in Gennan
1894; first published in English 1909).
MARX, K. (1975) Wages, Price and Profit (Alternative title: Value Price and
Profit) (Peking: Foreign Language Press) (first written in English and
delivered as a speech, in June 1865, to the General Council of the First
International; first published in 1898).
MARX, K. (1976) Preface and Introduction to A Contribution to the critique of
Political Economy (Peking: Foreign Language Press) (first published in
Gennan in 1859 and in English in 1951).
MARX, K. (1977) 'Economic and Philosophical Manuscripts', in Marx, K.
Early Writings (London: Penguin Books and New Left Review): 279-400
(first written in Gennan between April and August 1844; first published in
Gennan in 1932; first complete translation, by T. B. Bottomore, and
publication in English in 1963).
MARX, K. (1979) 'The Eighteenth Brumaire of Louis Bonaparte', in Marx, K.
and Engels, F. Collected Works, Vol. II, 1851-1853 (London: Lawrence &
Wishart): 99-197 (first written in Gennan in December 1851 to March 1852;
first published in Gennan in 1852; first published in English in 1979).
MARX, K. and ENGELS, F. (1955) 'Letter from Marx to Engels in
Manchester, August 2, 1862, London', in Selected Correspondence of Marx
and Engels (Moscow: Progress Publishers) pp. 128-33.
MASSEY, D. (1979) 'In What Sense a Regional Problem?' Regional Studies, 13:
233-43.
MASSEY, D. and CATALANO, A. (1978) Capital and Land: Landownership by
Capital in Great Britain (London: Edward Arnold).
MASSEY, D. B. and MEEGAN, R. A. (1978) 'Industrial Restructuring versus
the Cities', Urban Studies, 15: 273-88.
MEDlO, A. (1972) 'Profits and Surplus-Value: Appearance and Reality in
Capitalist Production', in Hunt, E. K. and Schwartz, J. G. (eds) A Critique of
Economic Theory: Selected Readings (Hannondsworth: Penguin Books).
Bibliography 267
MEDIO, A. (1977) 'Neoclassicals, Neo-Ricardians and Marx', in Schwartz, J.

(ed.) The Subtle Anatomy of Capitalism (Santa Monica, Cal.: Goodyear
Publishing Company): 381-411.
MEEK, R. (1956) 'Some Notes on the "Transformation Problem''', The
Economic Journal, 66: 96-107.
MEEK, R. (1976a) 'The Falling Rate of Profit', in Howard, M. C. and King, J.
E. (eds) (1976) The Economics of Marx: Selected Readings of Exposition and
Criticism (Harmondsworth: Penguin Books): 203-18.
MEEK, R. (1976b) 'Is there an "Historical Transformation Problem"?: A
Comment', The Economic Journal, 86: 342-7.
METCALF, D. (1975) 'Urban Unemployment in England', The Economic
Journal, 85: 578-89.
MILIBAND, R. (1972) 'Reply to Nicos Poulantzas', in Blackburn, R. (ed.)
Ideology in Social Science: Readings in Critical Social Theory (London:
Fontana/Collins): 253-60.
MOLOTCH, H. (1976) 'The City as a Growth Machine: Toward a Political
Economy of Place', American Journal of Sociology, 82 (2): 309-32.
MONTANI, G. (1975) 'Scarce Natural Resources and Income Distribution',
Metroeconomica, 27: 68-101.
MORISHIMA, M. (1973) Marx's Economics: A Dual Theory of Value and
Growth (Cambridge: Cambridge University Press).
MORISHIMA, M. (1974) 'Marx in the Light of Modem Economic Theory',
Econometrica, 42 (4): 611- 32.
MORISHIMA, M. (1976) 'Positive Profits with Negative Surplus Value - A
Comment', The Economic Journal, 86; 599-603.
MORISHlMA, M. and CATEPHORES, G. (1975) 'Is There an "Historical
Transformation Problem"?', The Economic Journal, 85: 309-28.
MORISHIMA, M. and CATEPHORES, G. (1976) 'The "Historical Transfor-
mation Problem": A Reply', The Economic Journal, 86: 348-52.
MORISHlMA, M. and CATEPHORES, G. (1978) Value Exploitation and
Growth: Marx in the Light of Modern Economic Theory (London: McGraw-
Hill).
MULLINS, P. (1977) 'The Social Base, Stake, and Urban Effects ofa Brisbane
Urban Social Movement', Australian and New Zealand Journal of Sociology,
13 (1): 29-35.
MURRAY, R. (1977) 'Value and Theory of Rent: Part One', Capital and Class,
3: 100-22.
MURRAY, R. (1978) 'Value and Theory of Rent: Part Two', Capital and Class,
4: 11-33.
NAQVI, K.A. (1960) 'Schematic Presentation of Accumulation in Marx', Indian
Economic Review, 5 (1): 13-22.
NEILSON, L. (1976) 'Developers as Conservative Decision-Makers', in
McMaster, J. C. and Webb, G. R. (eds) Australian Urban Economics: A
Reader (Sydney: Australia and New Zealand Book Company): 151-73.
NELL, E. (1972) 'The Revival of Political Economy', Australian Economic
Papers, 11 (8): 19-31.
NEUTZE, G. M. (1976) 'Policy Instruments in the Urban Land Market', in
McMaster, J. C. and Webb, G. R. (eds) Australian Urban Economics: A
Reader (Sydney: Australia and New Zealand Book Company): 78-128.
268 Bibliography
NICHOLLS, D. C., TURNER, D. M., KIRBY-SMITH, R. and CULLEN, J.

D. (1982) 'The Risk Business: Developers' Perceptions and Prospects for
Housebuilding in the Inner City', Urban Studies, 19: 331-41.
NUn, D. M. (1972) '''Vulgar Economy" in the Theory of Income Distribu-
tion', in Hunt, E. K. and Schwartz, J. G. (eds) A Critique ofEconomic Theory:
Selected Readings (Harmondsworth: Penguin Books): 222-32.
OATES, W. E. (1969) 'The Effects of Property Taxes and Local Public Spending
on Property Values: An Empirical Study of Tax Capitalization and the
Tiebout Hypothesis', Journal of Political Economy, 77: 957-71.
O'CONNOR, 1. (1973) The Fiscal Crisis of the State (New York: St. Martin's
Press).
O'CONNOR, 1. (1975) 'Productive and Unproductive Labour', Politics and
Society, 5 (3): 297 - 336.
OFFE, C. and RONGE, V. (1975) 'Theses on the Theory of the State', New
German Critique, 6: 137-47.
OKISHIO, N. (1961) 'Technical Change and the Rate of Profit', Kobe
University Economic Review, 7: 85-99.
OLSON, M. (1971) The Logic of Collective Action: Public Goods and the Theory
of Groups (Cambridge, Mass.: Harvard University Press) (first published in
1965).
ORB ELL, 1. M. and UNO, T. (1972) 'A Theory of Neighbourhood Problem
Solving: Political Action versus Residential Mobility', American Political
Science Review, 61: 471-89.
PAHL, R. E. (1975) Whose City? And Further Essays on Urban Society
(Harmondsworth: Penguin Books).
PAHL, R. E. (1977) '''Collective Consumption" and the State in Capitalist and
State Socialist Societies', in Scase, R. (ed.), Industrial Society: Class, Cleavage
and Control (London: Allen & Unwin): 153-71.
PAHL, R. E. (1978) 'Castells and Collective Consumption', Sociology, 12:
309-15 ..
PEJOVICH, S. (1982) 'Karl Marx, Property Rights School and the Process of
Social Change', Kyklos, 35: 383-97.
PERRONS, D. C. (1981) 'The Role ofIreland in the New International Division
of Labour: A Proposed Framework for Regional Analysis', Regional Studies,
15 (2): 81-100.
PICKVANCE, C. G. (1974a) 'Review Essay: Toward a Reconstruction of
Urban Sociology', (Review of Castells, 1972c), American Journal of
Sociology, 80 (4): 1003-8.
PICKVANCE, C. G. (1974b) 'On a Materialist Critique of Urban Sociology',
The Sociological Review, 22 (2): 203-20.
PICKVANCE, C. G. (1975) 'On the Study of Urban Social Movements', The
Sociological Review, 23 (1): 29-44.
PICKVANCE, C. G. (1976) 'On the Study of Urban Social Movements' in
Pickvance, C. G. (ed.) Urban Sociology: Critical Essays (London: Tavistock
Publications): 198-218.
PICKVANCE, C. G. (1977a) 'Marxist Approaches to the Study of Urban
Politics: Divergencies Among some Recent French Studies', International
Journal of Urban and Regional Research, 1 (2): 219-55.
PICKVANCE, C. G. (1977b) 'Physical Planning and Market Forces in Urban
Bibliography 269
Development', National Westminster Bank Quarterly Review, August: 41-50.

PICKVANCE, C. G. (1978) 'Review of: The Urban Question: A Marxist
Approach by Manuel Castells, Edward Arnold, London, 1977', The
Sociological Review, 26 (1): 173-6.
PICKVANCE, C. G. (1981) 'Policies as Chameleons: An Interpretation of
Regional Policy and Office Policy in Britain', in Dear, M. and Scott, A. J.
(eds) Urbanization and Urban Planning in Capitalist Society (London and New
York: Methuen): 231-65.
PICKVANCE, C. G. (1984), 'The Structuralist Critique of Urban Studies', in
Smith, M. P. (ed.) Cities in Transformation: Class, Capital and the State,
Urban Affairs Annual Reviews, vol. 26 (Beverly Hills and London: Sage
Publications): 31- 50.
PIGOU, A. (1960) The Economics of Welfare (London: Macmillan Press) (first
published in 1920).
POULANTZAS, N. (1972) 'The Problem of the Capitalist State', in Blackburn,
R. (ed.) Ideology in Social Science: Readings in Critical Social Theory
(London: Fontana/Collins): 238-53 (first published in 1969).
POULANTZAS, N. (1973) Political Power and Social Classes (London: New
POULANTZAS, N. (1975) Classes in Contemporary Capitalism (London: New
POULANTZAS, N. (1976) 'Problems actuels de la recherche marxiste sur
I'etat', Dialectiques, 13: 30-43.
PRATT, G. (1982). 'Class Analysis and Urban Domestic Property: A Critical
Examination', International Journal of Urban and Regional Research, 6 (4):
481-502.
PURDY, D. (1973). 'The Theory ofthe Permanent Arms Economy', Bulletin of
the Conference of Socialist Economists, 6: 12-32.
QUADRIO-CURZIO, A. (1980) 'Rent, Income Distribution and Orders of
Efficiency and Rentability', in Pasinetti, L. L. (ed.) Essays on the Theory of
Joint Production (London: The Macmillan Press): 218-40 (first published in
Italian in 1977).
REX, J. and MOORE, R. (1967) Race, Community and Conflict: A Study of
Sparkbrook (Oxford: Oxford University Press).
REYNAUD, E. (1974) 'Review of La Question Urbaine by M. Castells', in
Revue Franc;aise de Sociologie, 15 (4): 617-26.
REYNOLDS, D. R. and HONEY, R. (1978) 'Conflict in the Location of
Salutary Public Facilities', in Cox, K. R. (ed.) Urbanization and Conflict in
Market Societies (London: Methuen) Ch. 7: 144-60.
RICARDO, D. (1966) 'An Essay on The Influence of the low Price of Com on
the Profits of Stock showing the Inexpediency of Restrictions on importation,
with Remarks on Mr. Malthus' Two last publications: "An Inquiry into the
Nature and Progress of Rent" and "The Grounds of an Opinion, on the
Policy of restricting the Importation of Foreign Com" , (Abbreviated Title:
'Essay on Profits') in Sraffa, P. (ed.) The Works and Correspondence of David
Ricardo, Volume IV; Pamphlets and Papers 1815-1823 (London: Cambridge
University Press): 10-41 (first published in Feburary, 1915).
RICARDO, D. (1970) On the Principles of Political Economy and Taxation, P.
Sraffa (ed.) (London: Cambridge University Press) (first published in 1817).
270 Bibliography
RICHARDSON, R. (1980) 'Unemployment and the Labour Market', in

Cameron, G. C. (ed.) The Future of the British Conurbations: Policies and
Prescriptions for Change (London: Longman): 208-23.
RIDKER, R. G. and HENNING, J. A (1967) 'The Determinants of Residential
Property Values with Special Reference to Air Pollution', Review of
Economics and Statistics, 49: 246-57.
ROBINSON, J. (1965) 'Piero Sraffa and the Rate of Exploitation', New Left
Review, 31: 28-34.
ROBINSON, J. (1967) An Essay on Marxian Economics (London: The
Macmillan Press).
ROBINSON, J. (1973) 'The Theory of Value Reconsidered', in Collected
Economic Papers, vol. 4 (Oxford: Blackwell) pp. 59-66 (first published in
Australian Economic Papers, 1979, 8 (12): 12-19).
ROBINSON, J. (1980) 'Time in Economic Theory', Kyklos, 33: 219-29.
ROBSON, B. T. (1982) 'The Bodley Barricade: Social Space and Social
Conflict', in Cox, K. R. and Johnston, R. J. (eds) Conflict. Politics and the
Urban Scene (London: Longman) Ch. 3: 45-61.
ROEMER, J. E. (1977) Technical Change and the "Tendency of the Rate of
Profit to Fall" " Journal of Economic Theory, 16: 403-24.
ROEMER, J. E. (1978a) 'Marxian Models of Reproduction and Accumula-
tion', Cambridge Journal of Economics, 2: 37-53.
ROEMER, J. E. (1978b) 'The Effect ofTechnological Change on the Real Wage
and Marx's Falling Rate of Profit', Australian Economic Papers: 152-66.
ROEMER, J. E. (1979) 'Continuing Controversy on the Falling Rate of Profit:
Fixed Capital and Other Issues', Cambridge Journal of Economics, 3: 379-98.
ROEMER, J. E. (1981) Analytical Foundations of Marxian Economic Theory
(London: Cambridge University Press).
ROEMER, J. E. (1982) A General Theory ofExploitation and Class (Cambridge,
Mass.: Harvard University Press).
RONCAGLIA, A. (1978) Sraffa and the Theory of Prices (London: John Wiley)
(first published in Italian in 1975).
SALT, J. (1980) 'Labour Migration, Housing and the Labour Market', in
SAMUELSON, P.A. (1971) 'Understanding the Marxian notion of Exploita-
tion: A Summary ofthe So-called Transformation Problem between Marxian
Values and Competitive Prices', Journal of Economic Literature, 9: 399-431.
SAUNDERS, P. (1978) 'Domestic Property and Social Class', International
Journal of Urban and Regional Research, 2: 233-51.
SAUNDERS, P. (1980a) Urban Politics: A Sociological Interpretation (Har-
mondsworth: Penguin Books).
SAUNDERS, P. (1980b) 'Local Government and the State', New Society, 51
(910): 550-51.
SAUNDERS, p. (1984) 'Beyond Housing Classes: The Sociological Significance
of Private Property Rights in Means of Consumption', International Journal
of Urban and Regional Research, 8 (2): 202-27.
SAX, J. L. (1971) 'Takings, Private Property and Public Rights', The Yale Law
Journal, 81 (2): 149-86.
Bibliography 271
SCHNORE, L. F. (1965) 'The Spatial Structure of Cities in the Two Americas',

in Hauser, P. M. and Schnore, L. M. (eds) The Study of Urbanization (London
and New York: John Wiley: 347-98.
SCOTT, A. J. (I 976a) 'Land and Rent: An Interpretative Review of the French
Literature', Progress in Geography: International Reviews of Current Re-
search, 9: 102-45.
SCOTT, A. J. (l976b) 'Land Use and Commodity Production', Regional
Science and Urban Economics, 6: 147-60.
SCOTT, A. J. (1979) 'Commodity Production and the Dynamics of Land-use
Differentiation', Urban Studies, 16: 95-104.
SCOTT, A.J. (1980) The Urban Land Nexus and the State (London: Pion).
SETON, F. (1976) 'The Transformation Problem', in Howard, M. C. and King,
J. E. (eds) The Economics of Marx: Selected Readings of Exposition and
Criticism (London: Penguin Books): 162-76.
SHAIKH, A. (1978) 'An Introduction to the History of Crisis Theories', in The
SHERMAN, H. J. (1970) 'The Marxist Theory of Value Revisited', Science and
Society, 34: 257-92.
SILBURN, R. (1975) 'The Potential and Limitations of Community Action', in
Lambert, C. and Weir, D. (eds) Cities in Modern Britain: An Introductory
Reader (Glasgow: Fontana/Collins): 391-7.
SMITH, A. (1976) The Wealth of Nations (Harmondsworth: Penguin Books)
(first published in March, 1776).
SMITH, C. J. and HANHAM, R. W. (1981) 'Proximity and the Formation of
Public Attitudes towards Mental Illness', Environment and Planning A, 13:
147-65.
SMITH, D. M. (1977) Human Geography: A Welfare Approach (London:
Edward Arnold).
SOVANI, N. V. (1964) 'The Analysis of "Over-Urbanization"', Economic
Development and Cultural Change, 12 (2): 113-22.
SOWELL, T. (1960) 'Marx's "Increasing Misery" Doctrine', The American
Economic Review, 50: 111-20.
SOWELL, T. (1976) 'Marx's Capital after One Hundred Years', in Howard, M.
C. and King, J. E. (eds) The Economics of Marx: Selected Readings of
Exposition and Criticism (Harmondsworth: Penguin Books) pp. 49-76 (first
published in Canadian Journal of Economics, 33 (1967): 50-74).
SRAFFA, P. (1960) Production of Commodities by Means by Commodities.
Prelude to a Critique of Economic Theory (London: Cambridge University
Press).
STEEDMAN, I. (1972) 'Marx on the Rate of Profit', Bulletin of the Conference
of Socialist Economics, 4: 104-9.
STEEDMAN, I. (1973) 'The Transformation Problem Again', Bulletin of the
Conference of Socialist Economists, 2 (6): 37-42.
STEEDMAN, I. (1975a) 'Value, Price and Profit', New Left Review, 90: 71-80.
STEEDMAN, I. (1975b) 'Positive Profits with Negative Surplus Value', The
Economic Journal, 85: 14-23.
STEEDMAN, I. (1976) 'Positive Profits with Negative Surplus Value: A Reply',
272 Bibliography
The Economic Journal, 86: 604-8.

STEEDMAN, I. (1977) Marx after Sraffa (London: New Left Books).
STEEDMAN, I. et al. (1981) The Value Controversy (London: Verso Editions
and New Left Books).
SWEEZY, P. (1942) The Theory of Capitalist Development: Principles of
Marxian Political Economy (New York and London: Monthly Review Press.
SWEEZY, P. M. (1973) 'Some Problems in the Theory of Capital Accumula-
tion', Bulletin of the Conference of Socialist Economists, 2 (6): 25-36.
THERET, B. (1982) 'Collective Means of Consumption, Capital Accumulation
and the Urban Question: Conceptual Problems Raised by Lojkine's Work',
International Journal of Urban and Regional Research, 6 (3), 345-71.
THORNS, D. C. (1981) 'The Implications of Differential Rates of Capital Gain
from Owner Occupation for the Formation and Development of Housing
Classes', International Journal of Urban and Regional Research, 5: 205-17.
TIEBOUT, C. B. (1956). 'A Pure Theory of Local Expenditures', Journal of
Political Economy, 64: 416-24.
TOURAINE, A., AHTIK, V., OSTROWETSKY-ZYGEL, S. and CAS-
TELLS, M. (1967) 'Mobilite des Entreprises et Structures Urbain', Sociologie
du Travail: 369-405.
TOURAINE, A., AHTIK, V., OSTROWETSKY-ZYGEL, S. and CAS-
TELLS, M. (1968) Mobilite desEntreprises Industrielles en Region Parisenne
(Paris: Cahiers de l'Institut d'Amenagement et d'Urbanisme de la Region
Parisienne) Volume II. .
UNESCO (1974-1977) Index Translations. 1974-1977. Vols 27-30, UNESCO,
7 Place de Fontenoy, 75700, Paris.
VAN PARIJS, P. (1980) 'The Falling-Rate-of-Profit Theory of Crisis: A
Rational Reconstruction by way of Obituary' , The Review of Radical Political
Economics, 12 (1): 1-16.
WALKER, A. (1978) Marx: His Theory and its Context (London: Longman).
WALKER, R. (1974) 'Urban Ground Rent: Building a New Conceptual
Framework', Antipode: A Radical Journal of Geography, 6 (1): 51-9.
WALKER, R. A. (1975) 'Contentious Issues in Marxian Value and Rent
Theory: A Second and Longer Look', Antipode: A Radical Journal of
Geography, 7 (1): 31- 53.
WEBER, M. (1968) Economy and Society: An Outline of Interpretive Society,
vols I and 2 (translated by Roth, G. and Wittich, C.) (Berkeley: University of
California Press) (first written in German in 1910-14; first published in
German in 1922; first published completely in English, 1968).
WEBER, M. (1974) The Protestant Ethic and the Spirit of Capitalism (London:
Unwin University Books) (first published in German in 1904-5; first
WEISSKOPF, T. E. (1978) 'Marxist Perspectives on Cyclical Crisis', in The
WEISSKOPF, T. E. (1979) 'Marxian Crisis Theory and the Rate of Profit in the
Post-war U.S. Economy', Cambridge Journal of Economics, 3: 341-78.
WESTAWAY, J. (1974) 'The Spatial Hierarchy of Business Organizations and
its Implications for the British Urban System', Regional Studies, 8: 145-155.
Bibliography 273
WHITE, M. J. (1981) 'Self-Interest in the Suburbs: The Trend Toward No-

Growth Zoning', in Tropman, J. E., Dluhy, M. J. and Lind, R. M. (eds) New
Strategic Perspectives on Social Policy (New York: Pergamon Press): 434-49.
WILKINSON, R. K. and GULLIVER, S. (1971) 'The Impact of Non-Whites
on House Prices', Race, 13 (1): 21-36.
WILLIAMS, N. and NORMAN, T. (1971) 'Exclusionary Land Use Controls:
The Case of North-Eastern New Jersey', Syracuse Law Review, 22: 475-507.
WILLIAMS, P. (1978) 'Urban Managerialism: A Concept of Relevance?' Area,
10 (3): 236-40.
WINTERNITZ, J. (1948) Values and Prices: A Solution to the So-called
Transformation Problem', The Economic Journal, 58: 276-80.
WIRTH, L. (1951) 'Urbanism as a Way of Life', in Hatt, P. K. and Reiss, A. J.
(eds) Cities and Society: The Revised Reader in Urban Sociology (New York:
The Free Press of Glencoe) pp.46-63 (first published in The American
Journal of Sociology, 1938, July, Vol. 44).
WOLF, E. P. and LEBEAUX, C. N. (1969) Change and Renewal in an Urban
Community: Five Case Studies of Detroit (New York: Praeger Publishers).
WRIGHT, E. O. (1978) Class, Crisis and the State (London: New Left Books).
ZETTER, R. (1975) 'Les Hailes: A Case Study of Large Scale Redevelopment in
Central Paris', Town Planning Review, 46 (3) 267 -94.
Author Index
Abraham-Frois, G. 18,56 Catelano, L. 56
Agnew, J. A. 221,224 Chatterjee, L. 56
Albin, P. S. 207 Cheshire, P. C. 210, 249n
Alcaly, R. E. 24,35,36 Cheung, S. 215
Alchian, A. A. 216 Clark, G. 198
Althusser, L. 81-4,91-2,95,97, Clark, W. A. V. 248n
244-5n,246n Clarke, S. 56
Ambrose, P. 197,204,221 Coing, H. 132
Armstrong, P. 21 Colenutt, B. 204
Aronson, J. R. 207, 251n Collison, P. 228
Corkingdale, J. T. 210, 249n
Bailey, M. J. 221 Cox,A. 198
Balanowski, V. 125 Cox, K. R. 224,226,227, 251n
Balchin, P. N. 204 Crotty, R. 24, 53, 234n
Bale, J. 217,228
Balibar, E. 81-3, 244-5n Dales,J.H. 217
Ball, M. 56, 240-1n Davis, O. A. 201
Bandyopadhyay,P. 56 Dear, M. J. 198,224,226
Bannock, G. 149, 235n, 250n Demsetz, H. 215,216
Baran, P. 74 Desai, M. 12, 236n
Baumol, W. J. 10,207 Dicken, P. 213
Baxter, R. E. 149, 235n, 250n Dickinson, H. D. 30
Beirne, P. 56 Dmitriev, V. K. 14
Benetti, C. 60 Dobb, M. 20, 242n
Bentham, G. 121 Donnison, D. 251n
Berrebi, E. 18, 56 Downs, A. 220, 223
Berthoud, R. 209 Drewett, R. 201
Bhaduri, A. 20, 234n, 236n Dunleavy, P. 198,22,223
Blaug, M. 8,47, 236n
Bleaney, M. 236n Edel, M. 56
Boddy, R. 24, 53, 234n Elias, P. 210,212
Bohm-Bawerk, E. Von 8,13-18,22, Elkin, S. L. 201
33, 35, 49, 233n Elliot, B. 246n
Bose, A. 18, 56 Elster, J. 18,24,36, 249n
Bourne, L. S. 209 Engels, F. 95, 233n
Breugel, L. 56 Esfahani, H. 56, 242n
Broadbent, T. A. 56,211 Evans, A. 209,210, 248n
Bullock, P. 243n Evans, A. W. 209,210, 248n
Burgess, 94
Byrne, D. 56 Fielding, A. J. 215
Fine, B. 56
Cain, G. C. 136 Flynn, B. 133
Cataphores, G. 13, 17, 18,36,46, Freyssenet, M. 161
13, 17, 18, 36, 46, 233n, 237n Furubotn, E. G. 216
274
Author Index 275
Garegnani, P. 20 Keeble, D. 154,213

Garnier, J. P. 90 Keiper, J. S. 68,239n
Gibson, B. 56, 242n Kenway, P. 236n
Giddens, A. 244n Keogh, G. 210,212
Gillman, J. 74,236n Kieve, J. L. 204
Ginsberg, N. 56 King, J. E. 15, 24, 28, 31, 33, 35, 36,
Glass, R. 246n 47,52,232n
Gleave, D. 209 Kirk, G. 245n
Glucksman, A. 83, 244n Kurz, H. D. 56
Glyn, A. 21,24,53, 234n
Godard, F. 81,96,97, 101, 105, 109, Laclau, E. 244- 5n
114,119-21,125-34,161,186 Lamarche, F. 197
Goddard, J. B. 154,213 Lambert, J. 133
Goodchild, R. 20 I Lange, 0. 50
Gough, I. 74,243n Lauria, M. 56
Guillemard, A. 114 Lautman, J. 172
Gulliver, S. 252n Lebas, E. xvii, 85
Lebeaux, C. N. 227
Habermas, J. 55, 199 Lebowitz, M. A. 243-4n
Hall, P. 248n Lefebvre, H. 93-4
Hanham, R. W. 252n Ley, D. 114
Harloe, M. xvii, 90, 97,186, 251n Levi-Strauss, C. 245n
Harris, D. J. 35 Lipsey, R. G. 235n
Harrison, J. 21,74, 243n Locke, J. 246n
Harvey, D. 56, 195,217 Logan,J.R. 207,251n
Healey, M. J. 212 Long, J. 224,226
Heap, S. H. 24, 53, 235n Lorimer, J. 197
Hegel, G. 91,246n McCarthy, J. J. 24
Henning, J. A. 221, 223 McCleod, D. 56
Hill, T. P. 53,235n McCormick, B. 209
Hirsch, F. 220 Malthus, T. 50,60, 236-7n
Hirschmann, A. 0. 204,225 Markusen, A. R. 56
Hodgson, G. 15, 16, 18,22,35, Marriot, O. 207
232n, 234n, 235n, 236n Marshall, A. 13,250-ln
Honey, R. 224 Massey, D. 56, 154,21,213
House,J.W. 174 Medio,A. 21
Howard, M. C. 15,24,28,31,33, Meegan, R. A. 154,212,213
35,36,47, 52,232n Meek, R. 30, 233n
Huet, A. 161 Mercer, J. 224
Hughes, G. 209 Metcalf, D. 249n
Hunt, E. K. 72, 74, 243n Miliband, R. 245n
Molotch, H. 207,251n
Ipola, E. de 245n Montani, G. 239n
Moore, R. 251n
Jessop, B. 86, 245n Moseley, M. 121
Johnson, J. H. 204,210, 248n Moroshima, M. 13, 17, 18,36,46,
Johnston, R. J. 224,226 23-23n,234n,237n
Mullins, P. 225
Kain, J. F. 228,252n Murray, R. 56
276 Author Index
Navqi, K. A. 38 Salt, J. 209,210, 248n

Neilson, L. 201 Samuelson, P. 22,33
Nell, E. 20 Saunders, P. 144, 191, 193, 198, 199,
Neutze, G. M. 223 200, 297, 221, 222,223, 229, 251n
Nicholls, D. C. 201,221 Sax, J. L. 217
Norman, T. 69,228 Schnore, L. F. 247n
Nuti, D. M. 20 Seton, F. 16
Shaikh, A. 24,35,47
Oates, W. E. 221 Sherman H. J. 233n
O'Connor, J. 55, 199, 243n Silbum, R. 223
Offe, C. 198, 199 Smith, A. 67,71, 242-3n
Okishio, N. 33 Smith, C. J. 252n
Onaka, J. L. 248n Smith, D. M. 217, 251n
Olson, M. 226 Sovani, N. V. 94
Orbell, J. M. 224 Sowell, T. 24,36,52
Owens, P. L. 154 Sraffa, P. 17-23,56,72, 234n,
238-9n
Pahl,R. 97,144,186,217,220,222, Steedman, 1. 12, 18,22,30,35, 234n,
223 236n
Palmer, D. 209 Sutcliffe, B. 24, 53, 234n
Pejovich, S. 216 Sweezy, P. 8,15,16,24,31,35,36,
Perrons, D. C. 154 42,50
Pickvance, C. 85,87,90,92-3,95,
97,112,186,194,204,206,227, Theret, B. 144
228, 245n, 247n Thompson, C. 154, 213
Pigou, A. 251n Thoms, D. C. 251n
Poulantzas, N. 81-4,244- 5n Tiebout, C. B. 252n
Pratt, G. 222, 251n Tugan-Baranowsky, M. 42
Preteceille, E. 87, 161 Touraine, A. 107, 114, 116-18
Purdy, D. 32 Ungerson, C. 251n
Uno, T. 224
Quadrio-Curzio, A. 56
Quigley, J. M. 228, 252n Van Parijs, P. 35
Walker, A. 47,52, 232n, 238n
Rees, R. 149, 235n, 250n
Walker, R. 56
Rete!, T. 161
Weber, M. 194,248n
Reynaud, E. 90
Weisskopf, T. E. 24,35,53, 235n
Reynolds, D. R. 224
Whinston, A. B. 201
Rex, J. 251n
White, M. J. 228
Ricardo, D. 13,56,57,60, 234n,
Wilkinson, R. K. 252n
241n
Williams, N. 69,207,228
Richardson, R. 210, 248n, 249n
Ridker, R. G. 221,228 Williams, P. 222
Robinson, J. 8,20,52, 234n, 236n Wintemitz, J. 16
Wirth, L. 92
Robson, B. T. 224
Roemer, J. E. 18,35,54 Wolf, E. P. 227
Wood,P. 209,210,248n
Roncaglia, A. 234n, 238n
Wright, E. o. 31
Ronge, V. 198,199
Russell, L. 209,210, 211, 248n Zetter, R. 121
Subject Index
Althusser 224n: scientific functionalism 140, 186;
knowledge 91-2,246 ideology 90-2; immigrant
Althusserianism xviii, 81- 5, 88, 189: workers 134-6; index of
Castells and 90-2,95,97, differentiation 119-21;
101-3,104,126,140, 180, 245n industrialisation 92; interest in
work of 231- 2n; law of falling
Balibar: characterisation of economic rate of profit 35,98-9,127,
systems 82 135,186; location of industry
Basic commodities, 17-18,57,59, 107 -8, 115-18; Marxian crisis
61,66,65,73,158,238-41n; theory 25,55,98-101,139,
non-basic commodities 17 -18, 187; Marxist political economy
64-6,73,158,238-41n 89-90,101-2,103,104,106,
Bortkiewicz: critique of Marx's 108-9,113,188-9,192;
solution to Transformation methodology 116,118-20,123,
problem 13; solution to 125-6,129,131,132,134,135,
Transformation problem 136, 137, 138-9, 186, 188, 189,
14-18; and Bortkiewicz 192; property relations 108,
Corollary 17-18 110; real appropriation 110;
realisation crises 47,99-100;
Capital: constant capital 5-6, 13, research on Dunkirk 109,
14,27,28,30-1,38-9,86-7, 125-34; science 90-2; state
145,151,152,153, 236n (Marx's monopoly capitalism 126, 129;
theory of 5-6, 13, 14,27, trade union power 100-1, 188;
28-31, 232n, 238; price transport 132, 134; urban
coefficient of 14; value of 9, actors 106, 11, 121, 140,246n;
14,21,27; over-accumulation urban areas 89,92-3,94,101,
of 27,86-87,153); variable 104,105,137,139,247n
capital 5-6,27,28,30,38-9, (definition of 95-7, 139,
70,87,145,151,152,153 (Marx's 185-6); urban culture 92;
theory of 5-6,28,30,70; price urban planning 81,104,105-9,
coefficients of 14; value of 13, 122,126,132-3,139-40,
14,21,27) 189-90 (definition of 105);
Capitalist production: Marx's urban politics 81, 104-5, 133;
characterisation of 4; Sraff's urban protests 121-5,126,132,
characterisation of 18 136-8,140; urban renewal
Castells: Althusserianism 90-92, 118-21; urban social
95,97,101-3,104,126,140,180, movements 81,104,109-13,
245; Chicago school 91; class 139, 190-1 (social base of
fractions 128-9,132-3,134; 112-13,123,138,191;social
collective consumption 25, forceof 112-13,138,191);
95-101, 139-40, 165, 186, urban sociology 89-140;
187-8,215,229, 246n (definition critique of 89,90-5,139,141,
of 96; effects of 98 -10 1); 185,191 (formulation of Marxist
housing 121, 132, 138; urban sociology 95-113,139,
277
278 Subject Index
CastelIs: Althusserianism - contd. 216-17, 220, 222, 230, 250n;

195); urban systems 81, 101-5, reactions to changes in
106, 111, 132, 140, 246n) consumption of 223-29; state,
urbanism 90-1,92-94; 216,220
urbanisation 90,91,94-5,96 Crisis: Marx's theory of xviii, 24-6;
Chicago School 91 Castells's and Marxist crisis
Circulation, costs of 73-5,243-4n theory 25,55,98-101,139,
Class fractions 85, 161, 163, 167, 187; Lojkine and Marxist crisis
173, 174, 176-7: Castells's theory 25,55,187
theoryof 128-9,132-3,134;
Lojkine's theory of 161, 163, Dwelling unit 215,220-9,230,
167,174,176-7,178,180,247n 251n,239n
Collective action 226-30, 251n
Collective consumption 25, 187 -9, Exploitation oflabour 71: Marx's
196,229: Castells's theory of theory of 4,6; rate of 8,28-9,
25,95-101, 139-40, 165, 186, 30,32-3,46,152-3
187-8,215,229, 246n (definition Extended reproduction 44-6, 239n,
of 96; effects of 98 -10 1); 251n: Marx's model of 40-2
dwelling unit 215,220-9,230; Externalities 217,230, 250-251n:
falling rate of profit 98-9; fields 191,217-19,220;
Lojkine's theory of 25, 88, gradients 217-19,220;
142-50,151-2,186,187-8,215, negative 192,194,217-20
229, 247n; Marxian crisis Fiscal crisis 55
theory 25, 139, 187; realisation
crisis 99-100; trade union Housing 133-4, 165, 170, 175-6,
power 100-1 171-8, 190: and Castells 121,
Commodities 67: Marx's definition 132, 138; and Lojkine 173,
of 3; value of and price of 8, 175-6,177-8
12-13,16,18,49 Housing classes (see Tenure groups)
Commuting 97,128,167,186,195,
208-11,230 Ideological state apparatuses 84-5
Competition 10, 16, 17,38,49,59, Industrial reserve army 48,50,53,
61-62,64,68,135,136,201, 54: Marx's theory of 48, 50
241n, 248n: Marx's theory of Interests 106, 107, 108, 121, 125,
10,16,17,49,59,61-2,64,68, 129,132,134,140,166,171,
136 192-4, 196, 198-207,208-11,
Composition of capital: organic 212, 221,223,226-7,229,251n
6-7,8,10,27,28-33,35,50,66,
87,99,135,150-3,155,157, Labour markets 97, 136, 168, 186,
236n, 246n; Marx's theory of 8, 208-11,230, 248n
10,27,28-32,50,66; technical Labour power 27, 82, 97, 108: as a
27,31-2; value 27,31 commodity 48-50,54; Marx's
Constrained-actor model xvii, 192- 5, definition of 4-6; reproduction
196,198-207,215-29 of 97,101,107,132,143,145,
Consumption goods: Communal 146, 147, 165, 186, 214; shortage
186,216-17,220, 222,230, 250n; of 36,46-7, 128
negative 216-20,224-5, Labour theory of value: Marx's
226-7,228,230, 251n; positive theory of 3-7, 18,21-22,56,
216,220,224,226,230;private 66, 234n
Subject Index 279
Land 56-69: ownership of 57,64, state monopoly capitalism 81,

240n; price of 57, 248n; rent 85-8,129,141,160-1,177,179,
of 57-69 181,189, 245n; transport 149,
Landowners: constraints on 204-5 167,170,171,175-176,178;
Land-use 173: planning of 105, urban areas 141, 180, 247n
110, 132, 190; zoning 69, 108, (definition of 142-3,186-7,
190,201,206,228,248n, 249n 19l);urbandeve1opment 150,
Law offalling rate of profit 24- 35, 167-80,189; urban planning
85,127,150-1,153,158,181, 160-3,167-8,169-72,174,180,
186: Castells's interpretation of 181, 182, 185, 189-90; urban
35,98-9, 127, 135,186; politics 180; urban protests
Lojkine's interpretation 35, 85, 190, 191; urban segregation
150,151-3,173,180; Marx's 159, 167, 169, 170, 171, 173;
theory of 24-35 urban social movements
Lefebvre, theory of suburbanism 163-6,173,180,181-2,185,
93-4 190-1 (social base of 164-5,
Legitimation crisis 55 166,191; social force of 164-5,
Location of industry 107- 8, 117, 166,191); urban sociology
150,154,180-1,189,195,196, 141-82,182,185,191,195;
211-13,230: Castells's research urbanisation 142;
on 107-8,115-18;Lojkine's
research on 150,154,180,181, Macroeconomic planning: absence
186 of 25, 36, 40-2
Lojkine: class fractions 161, 163, Marx: barter 37; constant capital
167, 174, 176- 7, 178, 180, 247n; 5-6,13,14,27,28-31, 232n, 238
collective consumption 25,77, (value of 9, 14,21,27; over-
88, 165, 170, 180-1, 187 -8,215, accumulation of 27); capitalist
229, 247n; collective means of production, characterisation of
consumption 142-50,151-2, 4; commodities (definition of 3;
153,176,180-81;housing 173, value of and price of 8,12-13,
175-6, 177 -8; functionalism 16, 18,49, 232-233n);
186; interest in work of competition 10,16,17,49,59,
231- 2n; law offalling rate of 61- 2, 64, 68, 136; composition of
profit 35, 85, 150, 151- 3, 173, capital (organic 8, 10, 27,
180; location of industry 150, 28-32,50,66; technical 27,
154, 180, 186; Marxian crisis 31-2; value 27, 31) crisis,
theory 25,55, 187; Marxist theory of xviii, 24- 6;
political economy 141,147, exploitation oflabour 71
151,159,182,188-9,192; (theory of 4, 6; rate of 8,
Marx's theory of productive 28-30,46); extended
labour 70,144-7,187-8; reproduction 44-5 (model of
means of material circulation 40-2); industrial reserve army
142,151,152,153; means of 48,50, 53, 54, 237 - 238; labour
social circulation 142, 145, 146, power 4-6,27 (as a
151,153; methodology 152-3, commodity 48-50,54;
157,159,168,171,172-3,174, shortage of 46- 7); labour
175,176,178-9,180,186,188, theory of value 3-7, 18,21-2,
189,192;rent 56,69,150, 56,66, 233-4n; law offalling
154-9,173,178,179-80,181; rate of profit 24-35,86-7,
280 Subject Index
Marx: barter - contd. Poulantzas, theory of capitalist state

98-9,181, 235-6n, 238n, 241n; 84-5
productive labour, theory of Price gouging 201
xviii, 69, 70-77, 146, 188, Prices 3,7,9, 12,21, 65,68,234n
242-4n; profit 4-5,7,9, 16, Productive labour: Marx's theory of
17,18,27,28-29,51,66,68,71, xviii, 69, 70-7, 146, 188,
193,233n;241n(rateof 8,9,10, 242-4n; Lojkine's theory of 70,
11, 12, 16, 15, 17,22,27,28-9, 144-7,187-8
31,41,43,45,47,49,54; actual Price coefficients 14
rate of 27,30; average rate 9, Profit 4-5,7,9,16,17,18,27,
10,49); maximum rate of 27, 28-9,31,51,65,66,68,71,100,
29,30,52; price rate of 12-13, 152,155,162, 234-5n, 239n: rate
14; value rate of 12-13, 14); of 8,9,10,11,12,16,15,17,22,
realisation crisis 24-5,35-47, 24,26,27,28-9,32-5,41,43,
236n; rent 56-69,154-5,157, 45,46-7,54,59,60,62,63,68,
181,193 (differential rent I 57, 73,127,146,152,155,156,201,
58-61,157; differentialII 57, 234n, 236n, 238-241n (actual
61-4, 157; monopoly rent I 57, rate 27,30; average rate 9, 10,
64-6,157, 241n; monopoly rent 49,157; maximum rate 27,
II 57,66-67,157; theory of 29-30; price rate of 12-13,14,
xviii, 55, 56-69, 238-42n); rising 236n; Sraffa's theory of 20-1;
real wages 24-5,26,47-54; value rate of 12-13, 14, 236n)
simple reproduction 44-6 Property development 132, 142,
(model of 38-9); surplus 181,186,187,190,192,194,
value 9, 10, 11, 13, 14, 16, 18, 196-207,229: private 106,168,
21,22,28,39,66,233n,238n, 181,195,197,200-7,229-30;
242n (absolute surplus value 6, state 105,106, 108-9, 195,
30; relative surplus value 7,30; 197-200,229-30
rate of 6, 28, 30; theory of Property developers 196:
4-7); technical innovation 27, constraints on 200-4
29-31,50; transformation
problem (formulation of 7-9; Realisation crisis 35-47,99-100:
solution to 9-12; critique of and Castells's theory of 47,
solution 12-14); use value 99-100; and Marx's theory of
3-5, 232n, 242n; value, theory 24-25,35-47,236n
of xviii, 151; variable capital Relations of property 82,108, 244n,
5-6,28,30,70 (value of 13,14, 249n: Castells's interpretation
21,27); wages 4,20-2,26,46, of 108,110
70-1, 193, 232n (Marx's theory Relations of real appropriation 82,
of 50-1, 238n) 110, 244n: and Castells 110
Relative autonomy 82, 87
Okishio's theorem 33
Rent 56-69, 150, 162, 238-42n:
Over-urbanisation 94
differential rent I 57,58-61,
Planning authorities 106, 132, 162, 68-9,157-8,178,179,239-40n,
167,170-2,194,196,200,201, 248n; differential rent II 57,
205: constraints on 205- 7; 61-4,68-9,157-8,178,179;
interests of 200 Lojkine's theory of 6,69,150,
Population growth: lack of 46; 154-9, 173, 178, 179-80, 181;
Malthus's theory of 236- 7 Malthus's theory 60; Marx's
Subject Index 281
Rent - contd. Transformation problem 3,7-9,

theory of xviii, 55, 56-69, 233-234n; Bortkiewicz's critique
154-5,157,181,193; monopoly of Marx's solution 13;
rent I 57,64-6,68-9,157-8, Bortkiewicz's solution 14-18;
178, 179, 248n; monopoly rent invariance postulates 16;
II 57,66-7,68-9,157-8,178, Marx's formulation of 7-9;
179; Ricardo's theory of 58, Marx's solution 9-12;
60-1, 239n, 241n Meek's solution 16;
Repressive state apparatuses 84-5 Steedman's critque of Marx's
Rising real wages 24- 5,26,47 - 54 solution 12;
Winternitz's solution 16
Sector I, II, III 14-15, 17,38-41 Transport 132,134, 149, 165, 167,
Simple reproduction 44- 5, 236n, 169,173,175-6,178,190:and
239n: Marx's model of 38-9 Castells 132, 134; and Lojkine
Simultaneous equations 13, 15 149,167,170,171,175-6,178
Socially necessary abstract labour 3,
48,232-3n Urban actors 106,111,121,140,
Sraffa: basic and non-basic 194: definition of 196,248n;
commodities 17-18,19-20, Castells's theory of 106,
238n; capitalist production, 121,140,246n
characterisation of 18; Marx's Urban areas 89,92-3,94,101,104,
labour theory of value 20-1, 113, 137, 166, 185-7, 190,
233-4n; profits and prices 195-6,214,230: Castells's
18-22, 238n, 239n; and zero theory of 89,92-3,94, 101,
wages 29-30 104,105,137,139,247n
State expenditure 72,76-7,99-100, (definition of 95-97, 139,
127, 150, 161-2, 174-6, 198, 185 -6); final consumption in
199,214 214-29; Lojkine's theory of
State monopoly capitalism xviii, 81, 141,180, 247n (definition of
85-8,126, 189,245n;and 142-3,186-7,191); production
Castells 126, 129; and Lojkine of 208 - 214, 229; property
81,85-8,129,141,160-1,177, development in 196-207, 229
179,181,189,245n Urban development 150,167-180,
Surplus value 9,10,11,13,14,16, 189: control of 162,201;
18,21,22,28,66,151,152,153, Lojkine's research on 150,
155, 188: absolute surplus value 167-80,189
6,30, 135; Marx's theory of Urban planning 81,104,155,185,
4-7; price coefficients of 14; 189-90, 196: Castells's theory
rate of 6,28,30,35, 135; of 81, 104, 105-9, 122, 126,
relative surplus value 7,30,135 132-3,139-40,189-90
(definition of 105); Lojkine's
Technical conditions of production theory of 160-3,167-8,
21 169-72,174,180,181,182,185,
Technical innovation 24-5,26,27, 189-90
29-35,36,42-4,49,50,58,61 Urban politics 81,104: Castells's
Tenure groups 215,221-3,225, theory of 81, 104- 5, 133;
226-7,229,230, 239n, 251n Lojkine's theory of 180
Trade union power 25,26,48,53-4, Urban protests 111,133-4,190-1:
100-1: and Castells 100-1, 188 Castells'stheoryof 121-5,126,
282 Subject Index
Urban protests - contd. Lojkine 141-82, 182, 185, 191,

132, 132, 136-8, 140; Lojkine's 195
theory of 190-1 Urban systems 81: Castells's theory
Urban renewal, Castells's research of 81,101-5,106,111,132,
on 118-21 140, 246n
Urban social movements 81,104, Urbanisation 94-5,214: and
185,196: Castells's theory of Castells's theory of 90,91,
81,104,109-13,139,190-1 94-5,96
(social base of 112-13, 123, Urbanism 90-1,92-94: Wirth's
138,191; socialforce of theory of 92-3; Castells's
112-13, 138, 191); Lojkine's theoryof 90-1,92-4;
theory of 163 - 6, 173, 180, Lojkine's theory of 142
181-2,185,190-1 (social base
of 164-5, 166, 191; social force Value 32: exchange 3-5,75,208;
of 164-5,166,191) use 3-5,75,144,145,148,149,
Urban sociology: definition of 208; Marx's theory of xviii, 151
xvii-xviii; 142, 181, 186, 195;
Castells 89 -140 (Castells's Wages 21,32-4,46,48-54,58-9,
critique of 89,90-5, 139, 141, 60,63,65,68,70-1,72,100-1,
185,191; formulation of Marxist 135, 136, 162,214-15, 234-5n:
urban sociology 95-113, 139, Marx's theory of 50-1; zero
195); growth of Marxist urban wages 29-30,32
sociology xviii, 185, 192; Weather conditions 26, 36,44-6
Wirth, theory of urbanism 92-3

Kieran McKeown Auth. Marxist Political Economy and Marxist Urban Sociology A Review and Elaboration of Recent Developments

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Kieran McKeown Auth. Marxist Political Economy and Marxist Urban Sociology A Review and Elaboration of Recent Developments

Uploaded by

Copyright:

Available Formats

MARXIST POLITICAL ECONOMY AND MARXIST

A Review and Elaboration of Recent Developments

o Kieran McKeown, 1987

List a/Tables and Figures Xlll

PART I MARXIAN POLITICAL ECONOMY

1 Marx's Labour Theory of Value 3

2 Marxian Crisis Theory 24

2.2 Realisation Crises 35

3 Marx's Theory of Rent 56

4 Marx's Theory of Productive and Unproductive Labour 70

PART n FRENCH MARXIST URBAN SOCIOLOGY: THE

5 The Different Marxist Traditions of CasteUs and Lojkine:

6 The Marxist Urban Sociology of Manuel Castells 89

6.7.2 Research on Industrial Location in the Paris

7 The Marxist Urban Sociology of Jean Lojkine 141

7.6.3 Research on Urban Policy in Lyon in

8 From Marxist to Post-Marxist Urban Sociology 185

8.8 Consumption in an Urban Area 214

Nl Number of citations per year of the works of Manuel Castells

1.1 The relationships between physical value and price quantities 22

NI The growth of population and resources according to Malthus 237

economy which are suggested, implicitly or explicitly, by their work.

Marxian Political Economy

The purpose of this chapter is to provide an over-all assessment of

1.1 MARX'S LABOUR THEORY OF VALUE

The purpose of Marx's labour theory of value is to explain how the

of sociallyl necessary abstract labour embodied in them. Every com-

value (S) where:

~ = the organic composition of capital in sector i where i is any

where i and j refer to any sector between 1 and n.

~ = the rate of surplus value or exploitation in sector i, where i

is, by increasing the level of productivity in the wage-goods sector:

where i and j refer to any sectors between I and n.

1.2 THE TRANSFORMA nON PROBLEM

This problem is the pons asinorum of Marx's economic analysis

where i refers to any sector or commodity between I and n (see Marx,

Marx acknowleged this problem by pointing out that prices could

1.3 MARX'S SOLUTION TO THE TRANSFORMATION

There are essentially two components involved in the transformation of

Marx put forward the following:

This definition of the general rate of profit in terms of the arithmetic

it! Si = [r it! (Ci+ V) ]

Marx seems to have conceptualised the relationship between surplus

economy which regulates the distribution of surplus value irrespective of

Marx described this as a 'generallaw' (Marx 1974c, p. 161) which acts

1.4 THE ERRORS IN MARX'S TRANSFORMATION

Any solution to the transformation problem effectively involves trans-

since it involves circular reasoning. It is circular because the only reason

their respective total prices (see Bortkeiwicz, 1952, p. 10). It is therefore

Bortkiewicz, 1952, p.24; Hodgson, 1973, p.363; Morishima 1974a,

1.5 THE CORRECT SOLUTION TO THE TRANSFORMATION

The identification of the errors in Marx's solution to the transformation

A number of subsequent solutions have been put forward along the

(c\x + C2X + C3X) - (C\X + V\Y)

(V\Y + V2Y + V3Y) - (C2X + V2Y)

(S\Z + S2Z + S3Z) - (C3X + V3Y)

have an essential part in the determination of prices and the rate of

1.6 THE SRAFFIAN APPROACH TO PROFIT AND PRICES

Sraffa's book is concerned with the relationships between prices, profits

corresponding to the inputs and output of one of the n commodities in

basics' (ibid.). Two important implications follow from this distinction.

implication for Marx's labour theory of value is that the explanation of

theory of value. The first is that Sraffa's approach is more fundamental