This book is a stimulating and original introduction to the economics of industrial society. It is intended for use as a student text, but will also be of interest to all those - whether students or teachers - looking for new ways of understanding the economic problems of industrialised countries. It provides an effective critique of current economic theories, and develops an original model of the economics (whether neo-classical, Marxist, or Keynesian) of modern industrial society. Throughout the book the analysis is oriented towards the solution of problems in the real world, and towards explaining the operation of economic institutions in different countries. The work looks at the way individual markets operate, the determination of foreign exchange rates, the problem of unemployment, and the fiscal and monetary policies needed to tackle unemployment. The standard text book approach does not allow for the different types of market that exist. For the main agricultural and mining products there are commodity exchanges, whereas no such markets exist for the products of manufacturing industry. The way in which prices are determined will vary considerably depending on whether or not a commodity exchange exists. This means that the price mechanism will not be uniform for every single commodity. This book analyses the case of an industrial country entirely dependent on other countries for its raw materials (agricultural and mining products). The prices of these raw materials are determined on the commodity markets of the exporting countries; and it is only after passing through the filter of the exchange rate that they will exert an influence on the prices of manufactured products. For that reason foreign exchange rates play a significant role in the theory of production. Furthermore, this book follows Marx and Keynes in refuting Say's Law, which states that a state of general overproduction is impossible. (In this sense the book is anti-neoclassical.) Investment is not determined by savings, but is determined independently of savings, either autonomously or according to certain investment functions, and is dependent on the operation of financial markets. This book therefore attempts an integrated analysis of the real economy and the money economy. In an economy for which Say's Law does not hold true, there will inevitably be no automatic full employment. For that reason this book discusses thefiscaland monetary policies available to help reduce unemployment. It also considers stagflation and the Wicksellian cumulative process, making clear their mechanisms. Throughout, the mathematical analysis is kept simple, and wherever possible diagrams are used so that the argument will be intelligible to those new to the subject.
The economics of industrial society
The economics of industrial society MICHIO MORISHIMA Translated by DOUGLAS ANTHONY, JOHN CLARK, and JANET HUNTER
The right of the University of Cambridge to print and sell all manner of books was granted by Henry VIII in 1534. The University has printed and published continuously since 1584.
C A M B R I D G E UNIVERSITY PRESS Cambridge London New York New Rochelle Melbourne Sydney
Published by the Press Syndicate of the University of Cambridge The Pitt Building, Trumpington Street, Cambridge CB21RP 32 East 57th Street, New York, NY 10022, USA 10 Stamford Road, Oakleigh, Melbourne 3166, Australia © Cambridge University Press 1984 First published 1984 Reprinted 1985 Library of Congress catalogue card number: 84-7684
British Library Cataloguing in Publication Data Morishima, Michio The economics of industrial society. 1. Economics I. Title 330 HB171 ISBN 0 52126700 5 hard covers ISBN 0 52131823 8 paperback
Transferred to digital printing 2002
BS
Contents
Preface
page
Introduction 1 Which economics? 2 The outline of this book
vii 1 5
PART ONE THE FORMATION OF PRICES 1 Markets and the price mechanism 1 How are prices determined? 2 The structure of auctions 3 Formulations of economists 4 The existence of equilibrium 5 Determining production prices 6 Two types of market economy 2 The function of exchanges 1 The multilayered structure of market economies 2 Interrelationship of demands and supplies within a single exchange 3 Price repercussions amongst substitutive goods 4 Price repercussions between different exchanges 5 Futures markets 3 Fixing product prices 1 More on the full-cost principle 2 The period of production and hedging 3 Equilibrium production prices and their movement 4 The traditional theories: the marginal productivity theory and the labour theory of value 5 How should we treat the traditional theories of the firm?
13 15 21 23 25 32 38 39 43 53 61 68 72 78 83 95
vi
Contents
4 Determination of foreign exchange rates 1 The century of internationalism 2 The effects of fluctuations in foreign exchange rates 3 How exchange rates are determined (I) 4 How exchange rates are determined (II) 5 Changes in exchange rates 6 The futures market
99 105 112 118 122 128
PART TWO THE WORKING OF THE NATIONAL ECONOMY 5
The modern industrial society 1 An outline of the model 2 The structure of the model (I) 3 The structure of the model (II) 4 Equilibrium in real commodity markets - the principle of effective demand 5 The structure of the financial sector 6 Equilibrium on financial markets - the determining of the interest rates 6 Is full employment possible? 1 The labour market 2 Labour unions 3 Unemployment (I): Marx 4 Unemployment (II): Keynes and the classical school 5 Wages and unemployment 6 Labour and machinery 7 Fiscal policy 1 The investment multiplier 2 The effect of fiscal expenditures 3 The effects of a reduction in taxation 4 The multiplier effect of balanced budgeting 8 Monetary policy 1 Credit and business stimulation 2 Deflation and inflation 3 The roles of the central bank and the exchange stabilization fund 4 The Wicksellian cumulative process
135 140 147 152 160 167 177 183 188 197 204 212 221 231 237 243 251 257 262 271
Additional notes
281
Exercises
293
Index
299
Preface
This book has been written as an introductory economics textbook for first or second year students, but it is also intended to be used on occasions as a textbook for courses on the principles of economics for third or fourth year students, and as a supplementary reader for private study by MSc students. Furthermore, at the risk of setting my hopes too high, I hope it will also be of interest to fellow economists. The content of this book does, in fact, reproduce the content of the lectures which I have been giving to first year students at the LSE for several years now. Many of these students come to us having already studied 'A' Level economics in the sixth form, and the proportion of our students who have done this has recently been showing a marked increase. It is therefore desirable to present to them lectures with a higher-level content than 'A' level economics - often P. A. Samuelson's Economics and R. G. Lipsey's An Introduction to Positive Economics, for example - or at least lectures which do not just repeat the content of such works. This book marks my supply in response to this sort of demand. The approach of this book differs to a considerable degree from standard economics. If one really wishes to study economics at the present time one has to pay assiduous attention to consumer theory and the theory of the firm, dependent on the concepts of the marginal rate of substitution and marginal productivity respectively. That students should acquire 'the habit of rigorous thinking' through this kind of learning process is obviously an extremely good thing, but the rigour emphasized by these theories is no more than a geometric (or mathematical) rigour. All historical exactness or fidelity to the facts is either made light of, or totally disregarded. We should, of course, be pleased that economics has increasingly 'progressed' towards an exact logical composition, but it is regrettable that this has resulted in at least one of the most important divisions of economics ending up as a grandiose philosophy seeking to find out the logical implications of the pursuit of utility maximization on the part of vii
viii
Preface
the individual and profit maximization on the part of the firm - what we might call a philosophy of freedom. In this book such theories as these are not discussed at all. The theorem of the realization of a Pareto optimum in a situation of competitive equilibrium has entirely been ignored. The same is true of the equation for consumer behaviour - income effect and substitution effect - and of the marginal productivity theory; they have either been disregarded or accorded only the barest mention. Instead this book attempts to analyse the price mechanism in accordance with reality and at the same time to introduce students directly to the major problems of economics - i.e. an analysis of the way in which the real economy operates and the best way to bring about a change in direction in this operation. For an analysis of such problems as these a high-level consumer theory is quite unnecessary, as this book makes abundantly clear. What is more important is a knowledge of historical experience and the observation and formation of the way in which actual institutions work. As is stated in the introduction, the operation of the economy in 'middle-ranking' industrial nations such as Britain, Japan, Germany and Italy, differs from its operation in large-sized' industrial nations like the United States. Medium-sized nations have to be dependent on other countries for their raw materials, therefore have to export in order to raise this capital. Hence foreign exchange problems occupy a pivotal place in production theory, and friction - or, in the very worst cases open conflict - arises between financiers and industrialists regarding whether the exchange rate should be high or low. Such problems as these are not matters of life and death for the large industrial nations which can be more or less self-sufficient, and where imports and exports are very small in proportion to GNP, but for the medium-ranking industrial countries they are of primary importance. I therefore believe that we should have for these medium-sized nations economic theories (and textbooks) different from those we use in the case of the United States, and this book is my own first step in the direction of the quest for this kind of theory. For that reason I would like to recommend it not merely to university students but also to those who have already graduated and who now find themselves surrounded by the operation of the 'real economy'. I suspect, however, that one part of the second section of the introduction to this book will be difficult to understand, not merely for students but also for those who have gone through economics courses. Should this be the case, I hope that although the reader may have encountered in the introduction passages which are difficult to under-
Preface
ix
stand, he or she will not stick at that point but read on to the end of the book and then afterwards reread the introduction. It is likely to be a good deal easier the second time around. I have explained further in the Additional Note about those topics which are dealt with in normal textbooks but which here are either disregarded or given scant attention. Although based on my lectures at LSE, this work was first published by Iwanami Shoten for Japanese readers, which accounts for the large number of examples taken from the case of Japan. For the English translation I am indebted to the efforts of D. Anthony, J. Clark and J. Hunter and would like to express my gratitude to them. March 1984
MICHIO MORISHIMA
Introduction
1 Which economics?
The variety of economics There are, fortunately or unfortunately, a large number of schools within economics. Each school possesses its own particular theories, hence there is a great diversity of economic theories. They can be roughly classified into such schools as, for example, classical economics, neoclassical economics, Marxian economics, and Keynesian economics, but these can then be further classified into sub-schools such as the neo-Ricardian, neo-Marxist, neo-Keynesian and even neo-Austrian. On top of that we have institutional economics, the historical school as well as power school of economics. There are, in effect, a galaxy of plausible theories, but they all aim to analyse and explain the capitalist economy, so very few economists indeed even think that these theories can be used to shed light on socialist economies. What is needed for this is a quite separate theory. Similarly, when considering a capitalist economy which incorporates a relatively high degree of planning the theory of competition used for the free enterprise system cannot possibly be applied without substantial modifications. It is taken as a matter of course that for analysing different systems different theories must be used. This approach is regarded as the common sense one by economists of socialist countries. No one is likely to believe, for example, that the theory used in analysis of the Soviet Union will, just as it stands, be appropriate to Poland, Yugoslavia or China. Among economists in capitalist countries, however, you are unlikely to find many opportunists who make assertions along the lines of 'Classical theory is the most valid for country A, Marxian theory for country B, and Keynesian for country C ; if an economist is a Marxist then that economist is likely to claim without any flexibility that Marxian economics is the correct theory for all capitalist countries: the United States, Britain, Japan and so forth. And in that respect neoclassical economists and Keynesians are no 1
2
Introduction
different from their Marxist counterparts. They remain loyal to the theory in which they themselves believe and are convinced that the real capitalist economy must in each and every respect operate in the manner prescribed by their theory.1 When looked at closely socialist economies differ tremendously from one country to another and possess considerable individuality. Free enterprise economies likewise are not completely uniform. The real economy is not managed by an abstract ethereal being known as Homo economicus. Countries may share a capitalist economy, but their historical experience and cultural traditions differ, and the lives, beliefs and modes of behaviour of their people are certainly not the same. Not only that, but the methods and attitudes of organizations such as companies, banks and labour unions are quite different. They encounter, as a result, very different problems. The problems of wage differentials between large and small enterprises and of unpaid family employees, for example, which have been more or less solved in Britain, remain very severe difficulties in Japan. Furthermore each of these countries can also react in a different manner to the same stimulus. Where the difference in reaction is not more than a matter of degree it is possible to handle these economies with models of the same type using 1
Lionel Robbins' famous definitions of 'economic' was probably connected with the establishment of this sort of conviction. According to Robbins economics is the branch of learning which is concerned with the allocation of scarce resources among the various aims competing with each other for them. Those who define economics in this way as the study of the management of scarce resources are likely to consider that the most efficient way of administering these resources will be the same in any country. They therefore usually tend to regard neoclassical theory as the most appropriate scholarship to teach this kind of means of management, and apply it to all countries, on occasions even to the Soviet Union. Hence the conviction that neoclassical theory, and neoclassical theory alone, is the true economics. (This is the sort of reason why many Western economists have ended up by subscribing to Kantorovich's theory of production planning as one element of neoclassical theory.) My own tacitly adopted definition of economics is different: It is that economics is the branch of learning which concerns itself with elucidating how the elements of the material life of the people of a country compete with each other, and how they are mutually interdependent. Therefore whenever people are directly confronted by some difficulty relating to the management of their material lives, economists are under an obligation to look into how that particular problem can be removed. In view of the fact that the way in which the material lives of the people are related to each other differs from system to system, and the difficulties which crop up also differ according to system, the content of economic theory may also be expected to differ accordingly. In considering matters of policy, there are frequent cases where the conclusion is unavoidable that a difficulty cannot be dealt with without changing or amending the system, so it is essential that whatever their country economists should be guaranteed a wide measure of freedom to criticize the system. Robbins, L., An Essay on Nature and Significance of Economic Science, 2nd edn., London, Macmillan, 1949; Kantorovich, L. V., 'Mathematical Methods in the Organization and Planning of Production', (1939), Management Science, 6, pp. 366-422.
Which economics?
3
different numerical values for the coefficients (parameters) constituting the framework, but where the reactions are qualitatively different they have to be analysed using quite separate models. As the peculiarities of each economy become more and more well-defined we become more and more dissatisfied with these ready-made, conventional models - be they neoclassical or whatever - and the development of more appropriate models becomes a matter of primary concern to economists. Large, medium-sized and small industrial countries In this book we shall leave aside completely the differences in the characters of the people and the disparities in their cultural traditions, and instead classify economies quite simply according to scale, dividing them into three groups: large, medium-sized and small.2 Large capitalist countries will be deemed to mean countries such as the United States which are rich in natural resources, which are more or less selfsufficient, having virtually no dependence on any other country, and whose economies can develop any industrial sector that should be necessary. Countries whose land area is small and which are perforce dependent on other countries for many of their industrial raw materials and fuel are designated 'medium-sized industrial nations'. These countries, however, are by no means small, and are strong enough to be able to develop domestically all sectors of industry. The ratio of the volume of raw material imports to the level of production of manufacturing industry was in the case of the US A in 1977, 2.5%. The same figures for 1978 for Britain, Italy and Japan were 12%, 10% and 9% respectively, so these three countries are classed medium-sized industrial nations in our sense of the term.3 The same year thefiguresfor West Germany and France were 5.6% and 6.6% respectively, so these countries, too, can be said to demonstrate a marked 'medium-sized' tendency in comparison with the United States. Moreover in our 'medium-sized nations' the agriculture and mining sectors are weak, with the amount of production of these sectors comprising only a very small percentage of total GNP. In the case of both Britain and Japan these percentages in 2
3
The ethos of a people frequently governs their destiny. For discussion of this problem, see my Why has Japan 'Succeeded'?, Cambridge University Press, 1982. To obtain these figures imported fuels should, strictly speaking, be included. However, since many imported fuels are consumed outside the industrial manufacturing sector very detailed statistics are needed for any calculation of the proportion of the price of the products of manufacturing industry accounted for by imported raw materials and imported fuels. If we here for the time being regard all imported fuel stuffs as being used by manufacturing industry the figure comes to 16% for the USA, 26% for Japan, 30% for the UK, 16% for Germany, 24% for France and 31% for Italy, providing us with almost the same conclusion as the text. (Figures relate to 1977 for the USA, 1978 for the other countries.)
4
Introduction
1978 were 4%; for West Germany, France and Italy the figures were 7%, 9% and 12% respectively, so that both Britain and Japan can be said in this respect to be archetypal 'medium-sized' nations. Of the other three, however, both France and Italy have fairly sizeable agricultural sectors and in this sense cannot be regarded as typical of 'medium-sized industrial nations'. Small industrial countries are unable to produce all industrial commodities. Some of them may be unable to produce capital goods, and have to purchase machines from abroad, while others may be unable to produce certain kinds of consumer goods and have no choice but to import these consumer goods from other countries to satisfy demand from the people. In as far as its population is small, a small-sized country will not possess the labour force adequate for the domestic production of all kinds of industrial products, so such small countries are likely to be forced into the position of 'small industrial countries'. However, there also exist countries, which, while they may be sizeable in terms of population, cannot, for reasons such as the low level of education, develop those sectors of industry which require particular skills. Some industrial countries in the process of development become 'small industrial countries' for this reason, and many of these countries may easily before long become 'medium-sized industrial countries' if the opportunity offers. Some may even progress to become 'large countries'. Model for countries lacking natural resources In this book I intend to try to analyse those countries we have defined as 'medium-sized industrial countries', and in an effort to reach greater clarification of the concept of a 'medium-sized country' I shall venture to make a somewhat bold abstraction along the following lines. Let us consider the case of a country which possesses no natural resources whatsoever of its own. There exist, therefore, no agricultural sector and no mining sector. This country's industry, however, is capable of producing sufficient goods - both in the consumer goods sector and in the capital goods sector - to satisfy the demands of its people. It therefore imports no finished industrial goods from any other country. However, these industrial sectors require raw material and fuel. A country with no natural resources of its own has no choice but to import such raw materials and fuel from elsewhere. Given that imports of this kind are unavoidable this country is going to have to export part of its industrial production to pay for its imports. The problems of earning through exports such-and-such an amount of foreign currency, and of how much foreign currency will be necessary to purchase imports, i.e.
The outline of this book
5
the problem of trade receipts and disbursements, is a significant matter, literally one of life and death, for the country. Should the foreign exchange rate change the cost of imported raw materials and fuel will also change, thus influencing the costs of production of industrial goods and hence their price. In this way, in a country lacking in natural resources, industrial production (which is the same as that country's total national production) is directly linked to the exchange rate. Since there is no agriculture, foodstuffs have to be imported from other countries, but in our no-resource country there will be no importing of agricultural products for immediate food use. Despite this there is unlikely to be any shortage of food, because according to our model one section of the consumer goods production sector, that manufacturing tinned goods, has as its raw materials imported agricultural and marine products and supplies the people with these tinned goods. Clearly this kind of model is a distorted one, one which represents in an utterly deformed way the realities of a medium-sized industrial country. Nevertheless this kind of crude abstraction is absolutely imperative if we are to reveal the essentials of the economy which we have selected as the object of our analysis. Just as the twisted faces drawn by Picasso expressed very well the individuality of their subjects so in science does one gain a more precise grasp of the real fundamentals by exaggerating one side of the reality through abstraction and eliminating completely the other side. In this work we will not be concerned with the other sizes of industrial country, but if, while studying the theory of 'medium-sized industrial countries' we bear in mind the question of how to reach a model for the large and small industrial countries, then this will develop our ability to conceptualize. 2 The outline of this book Part one: The theory of prices
This work is divided into two main parts. Part One discusses the way in which prices are determined. Ordinary textbooks tend first of all to state that a price is determined at the point where the demand curve and the supply curve intersect, and then move on to an explanation of each of these curves (expounding the utility or indifference-curve theory of consumer demand and the theory of firms based on the principles of profit maximization; these theories yield the demand curve and the supply curve, respectively). However, this means of determining prices fits only one group of goods: agricultural, forestry and marine products, and some minerals. The prices of many industrial products are not
6
Introduction
regulated by the market so as to equate demand and supply; prices have already been decided at the factory at the time of shipment, and the amount of shipment is regulated according to the volume of demand. Of course even when the price is determined in this way by the supplier (the factory) there is frequently price competition between suppliers, competition which is in some cases so fierce that it can be more appropriately termed a 'price-war', and there is no question of such price competition being a revision of prices to regulate supply and demand. It is aimed at bringing down competitors. Any equating of the supply and demand of these goods is achieved not by price regulation but by regulating the quantity available (the volume of shipments). On top of this the foreign exchange rate is determined by the dealers of the various banks according to particular formulae. Furthermore in the real economy wages are not determined by the congruence of supply and demand, being bid up and down in the manner described by the textbooks. In short, in the real economy the price of each good is decided according to a number of formulae, and the first half of this book will attempt to explain the manner of determining the prices of those goods which have the greatest bearing on our 'country without natural resources'. Discussion will include the means by which the prices of agricultural, forestry, marine and mining products which constitute the raw materials of our industrial sector, are determined on the (foreign) commodity markets; the nature of the full-cost principle which supplies the formula by which each factory decides on the price of its products; and how the exchange rate is determined on the international interbank market. Discussion of the labour market and the financial markets will be reserved for Part Two of the book. Part two: Say's Law and the principle of effective demand In Part Two we will discuss the circulation of both goods and currency. Since our country without natural resources has no agriculture no landlord class need be considered. The people make their living as workers, entrepreneurs or capitalists, or if they are unable to become any of these they become unemployed. The entrepreneurs are distinct from the capitalists. It is the capitalists who supply the capital for the enterprise, and the entrepreneurs who are responsible for its direction and management. Cases where capitalists are at the same time entrepreneurs are, of course, quite frequent, but in view of the fact that there are countless capitalists who despite contributing capital have no role in planning or management, and because there are also large numbers of entrepreneurs who possess no capital at all - or hardly any - and carry on their business with loans of capital from elsewhere, entrepreneurs
The outline of this book
7
and capitalists must be kept distinct from each other. Enterprises normally raise funds through the issue of stocks, but we shall assume here that funds are raised through the issue of (fixed-interest) debentures. Hence the capitalists who finance the enterprise are rentiers. There are in economics two fundamentally conflicting methods of thought. The first of these takes the view that overproduction is impossible when the economy is taken as a whole, that is to say if the total volume of production pxXx + p2X2 is decided then a total demand in accordance with this level will always be generated.4 The second is the view that it is total demand which determines the total volume of output, and not vice versa. The former is called Say's Law, and the latter the principle of effective demand. Ricardo, who dominated the field of economics up to the first half of the nineteenth century, recognized Say's Law, so this law was called by Keynes the postulate of the classical school.5 Since such a postulate is necessary for general equilibrium to be achieved, theorists who believed in general equilibrium, including Walras, subscribed to Say's Law, but from the midnineteenth century, scholars denying the validity of Say's Law began to appear, for example Marx and his disciples. The ideas of these opponents of Say's Law eventually bore fruit in the form of Keynes' principle of effective demand.6 The basic theme of the history of economics over the century preceding the publication of Keynes' General Theory can be regarded as the attempt to overturn the world of Say's Law (Ricardian economics) and to construct a system not governed by this law (Keynesian economics). This book is written from the standpoint of the principle of effective demand, but before coming down conclusively on the side of this principle I shall explain briefly what I regard as the deficiencies in Say's Law. Any decision on the level of output involves a decision on the wages to be received by workers and the income of entrepreneurs, and these individuals all consume part of their respective incomes, and save the remainder.7 If it can be always guaranteed that investment will be 4 5 6
7
Xx is the volume of production of consumer goods, X2 the volume of production of capital goods. px and p2 are their respective prices. Keynes, J. M., The General Theory of Employment, Interest and Money, London, Macmillan, 1936, p. 26. Walras' position is an ambiguous one. In the text of his Elements of Pure Economics (translated by W. Jaffe, Richard D. Irwin, Inc., Homewood, Illinois, 1954), he clearly refutes Say's Law, while affirming it in his mathematical model. Any denial of Say's Law in his mathematical model would have amounted to a recognition that such a thing as a situation of general equilibrium, far from coming into being, did not generally exist. (M. Morishima, Walras' Economics, Cambridge University Press, 1977, pp. 70-122.) Concerning Marx see his Theories of Surplus Value, part II, London, Lawrence and Wishart, 1969. See Additional Note a.
8
Introduction
equal to this volume of savings, then the total level of production will always be equal to the sum of the total amount of consumption plus the volume of investment, i.e. to the total level of demand. Hence Say's Law will operate.8 Overall surplus production is not possible; if there is overproduction in industries producing consumer goods there is bound to be underproduction in capital goods industries.9 In this way in order for Say's Law to take effect the amount of investment generated must be equal to the total amount of savings, regardless of the prescribed level of the total volume of output. This, however, is quite simply impossible. The decision on whether to invest rests with the entrepreneur, and neither the workers nor the rentiers have any part in it. When total production is at a high level the total volume of savings is accordingly high, so savings can without any problem exceed investment. Conversely when the volume of output is small total savings will be less than total investment. This being the case there is no question of total savings equalling investment at every level of total output. It will be no more than a case of savings equalling investment by chance at a certain level of total output. This stipulated value of output is known as the equilibrium output value, and since any volume of production differing from the equilibrium value generates savings which are either greater or less than investment, then overall overproduction and underproduction are very possible.10 It is this that serves to negate Say's Law. The entrepreneurs' investment decisions are devoid of any mechanism which might enable them to revise their investment plans to invest at exactly the same level as that reached by savings at any time, whatever the volume of output may be; they also lack any flexibility which might produce this kind of revision. Conversely, in the real economy the amount of production accords with the amount of investment decided on by entrepreneurs. In this manner production is carried out on a scale which will produce the I2 is the volume of demand for investment, p2l2tne amount of investment funds. Should this amount of investment always be equal to savings, the result will be e (*) PxXx + p2X2 = Pl(D? + D 1 + D[ + El + Gx) + p2(I2 + E2 + G2) according to the (t) formula given in the Additional Note a. That is to say the total amount of production equals total demand. y Formula (*) in note (8) can alternatively be written px[Xx - (D? + D\ + D[+El + Gx)] + p2[X2- (I2 + £ 2 + G2)] = 0 (**) If the part within the first square brackets has a plus value (i.e. surplus production of consumer goods), then the formula within the second square brackets has a minus value (i.e. underproduction of capital goods). 10 If S>I then the left side of the formula (**) in note (9) has a positive value, so the sections within both the first and second square brackets are together likely to assume a positive value, hence giving rise to general overproduction and contradicting Say's Law. 8
The outline of this book
9
equilibrium value of output, but production of this sort of scale is not necessarily such as will result in full employment among workers. At times when entrepreneurs are not particularly keen to invest the equilibrium amount of production will be small, and unemployment is, as a result, unavoidable. Under Say's Law there are no obstacles to full employment. As long as production is carried on at the level necessary to realize full employment, then investment will adapt itself so as to equal the appropriate level of savings, and supply and demand are balanced at the full employment level of output; neither overproduction nor underproduction will exist. In an economy where Say's Law does not operate insufficient investment becomes an obstacle to full employment. And in an economy where the inclination to invest is insufficient the government must either positively generate demand to remove this insufficiency, or in some way encourage financial agencies into stimulating the desire to invest. Hence fiscal and monetary policy will constitute the main topic of Part Two of this volume, and there will also be some discussion of problems related to dealing with the ill effects of these policies (such as deficit financing, inflation, and stagflation). The approach adopted by this book This book is both theoretical and analytical, so mathematics is frequently utilized. The mathematics is, however, not difficult, and the level of knowledge attained with O-level mathematics is adequate for an understanding. My own belief is that economics is not a single pure science, but a grand integrated body of knowledge. Therefore in order to gain an understanding of economic theory it is not enough merely to be conversant with the mathematical framework of the theory. There must also be some considerable knowledge of the social, institutional and historical foundations of that theory. Any attempt to disregard this social, institutional and historical background, and to consider and examine economic theory as no more than formal logic and chains of mathematics, cannot possibly be the right way to study economic theory. At appropriate junctures in this book, therefore, pages are devoted to the explanation of economic systems and attempts made to consider matters from an economic history standpoint. I am under no illusions that such discussions are adequate, but I have adopted this sort of approach here in the hope of persuading the readers of this book that economics is an integrated science. It expresses my own antipathy towards the way in which theoretical economics has become no more than a mathematical skeleton.
PART ONE
The formation of prices
Markets and the price mechanism
1 How are prices determined?
Views of the academics Let us first consider the views of two representative scholars, Marx and Walras. The former, Marx, considers that the prices of goods are determined 'by competition between buyers and sellers, by the relation of demand to supply, and of desire to offer'.1 He says that The same commodity is offered by various sellers. With goods of the same quality, the one who sells most cheaply is certain of driving the others out of the field and securing the greatest sales for himself. Thus the sellers mutually contend among themselves for sales, for the market. Each of them desires to sell, to sell as much as possible and, if possible, to sell alone, to the exclusion of other sellers. Hence, one sells cheaper than another. Consequently, competition takes place among the sellers, which depresses the price of the commodities offered by them. But competition also takes place among the buyers, which in its turn causes the commodities offered to rise in price. Finally, competition occurs between buyers and sellers; the former desire to buy as cheaply as possible, the latter to sell as dearly as possible. The result of this competition between buyers and sellers will depend upon how the above-mentioned sides of the competition are related, that is, whether the competition is stronger in the army of buyers or in the army of sellers.2
Of course, since in the real world friction exists, actual prices do not necessarily coincide with equilibrium prices. Nevertheless Marx thought that 'the study of such frictions, while important to any special work on wages, may be dispensed with as incidental and irrelevant in a general analysis of capitalist production'.3 That is to say, as a rule Marx took prices to be equilibrium prices. Orthodox economists are completely at one with Marx from this standpoint. Walras for example, writes as follows 1 2 3
Marx, K., Wage, Labour and Capital in Karl Marx and Frederick Engels, Selected Works, London, Lawrence and Wishhart Ltd., 1978, p. 76. Ibid, p. 76. Marx, Capital (Volume III), Moscow, Progress Publishers, 1966, pp. 142-3.
13
14
Markets and the price mechanism
As buyers, traders make their demands by outbidding each other. As sellers, traders make their offers by underbidding each other. The coming together of buyers and sellers then results in giving commodities certain values in exchange, sometimes rising, sometimes falling, sometimes stationary. The more perfectly competition functions, the more rigorous is the manner of arriving at value in exchange. The markets which are best organized from the competitive standpoint are those in which purchases and sales are made by auction, through the instrumentality of stockbrokers, commercial brokers or criers acting as agents who centralize transactions in such a way that the terms of every exchange are openly announced and an opportunity is given to sellers to lower their prices and to buyers to raise their bids. This is the way business is done in the stock exchange, commodity markets, grain markets, fish markets, etc. Besides these markets, there are others, such as the fruit, vegetable and poultry markets, where competition, though not so well organized, functions fairly effectively and satisfactorily. City streets with their stores and shops of all kinds - baker's, butcher's, grocer's, tailor's, shoemaker's etc. - are markets where competition, though poorly organized, nevertheless operates quite adequately. Unquestionably competition is also the primary force in setting the value of the doctor's and lawyer's consultations, of the musician's and singer's recitals, etc. In fact, the whole world may be looked upon as a vast general market made up of diverse special markets where social wealth is bought and sold. Our task then is to discover the laws to which these purchases and sales tend to conform automatically. To this end, we shall suppose that the market is perfectly competitive, just as in pure mechanics we suppose, to start with, that machines are perfectly frictionless.4 The market So then, what is a market? In economics a market is both the place where transactions (exchanges) are carried out, and the arrangements for exchange, or the organization through which buying and selling take place. As Walras points out, whilst there may be loose arrangements, there may also be elaborate organizations regulated by law which are completely systematized (such as the various types of exchange, etc.). Since goods and goods (or goods and money) are exchanged in the market, an exchange ratio, that is a price, will be fixed there. For example, if 20 yards of linen are exchanged for 2 coats, the price of a coat in terms of linen will be their exchange ratio, that is one coat: 10 yards of linen. In the everyday economy, goods are not normally exchanged for other goods but for money, and so the price of goods will be their exchange ratio with money. Thus the price expressed in terms of pounds will be 1 coat for a certain number of pounds. In broad terms three markets may be distinguished: commodity 4
Walras, Elements of Pure Economics (translated by W. Jaffe), Homewood, Illinois, Richard D. Irwin Inc., 1954, pp. 83-4.
The structure of auctions
15
markets, markets for factors of production, and security markets. Commodity markets may be further divided into consumer goods and producer goods markets (machinery, buildings, raw materials, etc.), while the market for factors of production may be broken down into the labour market and the market for land. Security markets are markets for shares, for bonds, and so on. Markets for factors of production must be handled cautiously because of the important role of human, social, and historical factors, and therefore we shall treat them separately below, and continue our explanation mainly in terms of commodity and security markets. 2 The structure of auctions Cross-trading
There are three kinds of transaction method; cross-trading or negotiated transactions, trading by bid or tender, and competitive trading. In cross-trading there is a transaction as long as the seller (supplier) and the buyer (demander) agree on a price. Suppose that there is a clothes shop A which would like to sell a coat at £16.00, and an individual B who would like to buy it at £16.00, with C who would buy it at £14.00, and D who would buy it at £12.00. Because agreement is soon reached between A and B on price, in cross-trading, a coat will quickly be sold from A to B. With the next coat it seems more likely that agreement will be possible with C rather than D, and so A may well first approach C. In this case too, so long as either A or C, or both, are not prepared to make some concession no agreement will be reached; but if A is intent on selling and C is similarly intent on buying, a second cross-trade will probably be achieved at £15.00 between A and C. Thus, with cross-trading similar coats may be exchanged at different prices on the same day. However, B may complain at this point. Having just bought from A at a price of £16.00, when B hears that A later sold to C at £15.00 he is likely to return to shop A and stubbornly negotiate to get the price of the coat he bought down to £15.00. In order to avoid such disturbances not only must there be agreement achieved on price between individual sellers and buyers who cross-trade, but amongst all the sellers and buyers. The tender system and competitive trading are arrangements for achieving high levels of agreement and for establishing a single price for each good. That is, the law of indifference is realized. Trading by tender With trading by tender, there are both tenders for buying goods and tenders for selling, and it is the latter in particular which is called an
16
Markets and the price mechanism
auction and is carried out periodically at a specified place. Examples of the former are as follows. (1) Where the State and regional public bodies procure construction work, the contents of the work are publicly announced. The construction companies having estimated the costs, the government office as a rule entrusts the company which presented the cheapest estimate with the work. (2) In ordering the construction of new ships, too, contracts are made according to the same formula. And (3) where there are several different methods of manufacturing the same product the 'tender formula' will be employed to determine which method of production will be adopted and which rejected. That is, the cost of the respective methods of production will be estimated, and the least-cost method will be adopted. Auctions often take place in the selling of antiques and luxury residences. But there is no set method for these, and there are various types, such as the British type and the Dutch type. In the British type the buyer bids up gradually from a low to a higher price, whilst in the Dutch type he bids down from a higher to a lower price. Let us now sell two coats at auction. Let A be the seller and B, C and D the three buyers. As in the previous example, suppose that B will buy 1 coat up to a price of £16.00, and C and D respectively will buy 1 coat each at £14.00 and £12.00. Firstly, in the British type, the auctioneer may well propose a low price, such as £11.00 for example. At this price B, C and D all signal their intention to buy, so with a demand of 3 but a supply of 2 the price will be bid up. When the auctioneer raises the price in units of £0.10 and it reaches £12.10, D will drop out, the demand will be 2 and will be equal to the supply. That is, in the British type of auction £12.10 is the equilibrium price. Since both B and C will have bought at the same price neither should have any complaint, and D too will have no reason to complain since he has given up trading. Since B was happy to buy up to a price of £16.00 he will congratulate himself that he has bought cheaply; and C too, who would have bought up to £14.00, may well think that it was not a bad bargain. (The total amount by which the purchasers have profited is called the consumer's surplus, and is, (£16.00 - £12.10) + (£14.00 - £12.10) = £5.80.) The Dutch type of auction goes as follows. The auctioneer suggests a high price, for example £17.00. There is no buyer and the price is bid down. At £16.00 B will begin to demand it, but there will still be an excess supply, and the price will be bid down further. When it eventually reaches £14.00 the demand will be 2 and will equal the supply. The equilibrium price in the Dutch type of auction is £14.00 and of course the consumer's surplus is £2.00. Thus even with auctions, when the type differs the equilibrium value
The structure of auctions
17
attained will not necessarily be the same. However, it differs from cross-trading in that in an auction all the buyers are treated equally and all transactions trading between B and A, and between C and A are carried out at a uniform price. Auctions are transactions at a uniform contract price and establish the rule of one price for one good. There may, however, be some readers who think that because in an auction seller A will supply 2 coats regardless of the price proposed, the inclinations of A are absolutely ignored. However, A will not sell a coat at any price simply at the behest of the buyers. For example, if the price were above £15.00 A may certainly want to sell, but beneath that price he may not want to; and at a price of exactly £15.00 he may be indifferent between selling or not. But in order to have the inclinations of the seller reflected in the bargain, the rules of the auction given above should be slightly amended. That is, A entrusts the sale of 2 coats to the auctioneer and when the auction starts sits among the buyers. Whilst the price is not one at which A wishes to sell, A should be permitted to indicate his intention to buy back the 2 coats. Therefore when the price is less than £12.00, the total demand will be the 3 coats B, C and D wish to buy and the 2 coats A bids for, making 5 coats in all. With this demand the price will be bid up. Where the price is £12.10-£14.00 total demand will be 4 coats, where it is £14.10-£14.90 demand will be 3 coats, falling to 1 coat at £15.10-£16.00. Since the supply is 2 coats the price will be bid up to £15.00. At £15.00 one coat will be sold to B, and the remaining coat will be left unsold. This means that if the price is lowered to £14.90 A will buy up the total supply at that price and so B's demand will be an excess demand. Therefore price will be bid up. The price of £15.00 is thus equilibrium price, and one coat remains unsold at equilibrium. But because it does not matter to A whether he sells or not at that price he will be content to withdraw. Exchanges The competitive trading which is carried out at commodity and stock exchanges has expanded and reinforced the bid or tender system or auctions at auction houses, so as to enable many suppliers and demanders to participate at the same time. Exchanges are not public places in the sense that anyone can transact directly there. Transaction must be through a broker approved by the exchange; but once having commissioned a broker anyone can engage in transactions, and so the organization is a semi-public one. Thus, whilst only a limited number of brokers have access to the exchange, an unlimited volume of business can be handled.
18
Markets and the price mechanism
There are various kinds of competitive trading. For example, prices were determined as follows in the settlement-trading market of the pre-war Tokyo Rice Commodity Exchange. In Table 1, let us assume that A to H are the eight brokers and that they came to the exchange with orders to sell or buy as shown in the Table. Carte blanche here means that they do not specify the price; an order to sell 300 bushels of standard rice (or to buy 400 bushels) at the market price, whatever it was, would be a carte blanche order. The opposite of this is a stop or limit order which is to sell or buy at a specified price. For example, B has orders to sell 300 bushels if the price is £10.00 or above, and F has orders to buy 100 bushels if the price is £13.00 or below. Table 1. Sample rice exchange orders Seller
Price
Quantity
Buyer
Price
Quantity
A
Carte blanche £10.00 £11.00 £12.00
300 bushels
E
400 bushels
300 bushels 200 bushels 100 bushels
F G H
Carte blanche £13.00 £12.00 £10.00
B C D
100 bushels 300 bushels 300 bushels
In the competitive trading process the price will change, but because it is impossible for it to be made to change continuously in infinitely small units, let us assume that it changes in discrete, minimum units of £1.00 (one pound). If finer adjustment is necessary, it might be adjusted in 50-pence or 10-pence units. The minimum unit is decided by the statutes of the exchange. Simulation of competitive trading Let us now handle the demand and supply of Table 1 according to the practice of the Tokyo Rice Exchange. When competitive trading begins, first carte blanche orders to buy and sell are transacted. In our example, the carte blanche order to buy is 100 bushels in excess of demand, and the convention is that these 100 bushels will be transacted at the lowest price (at £10.00, 100 bushels will be supplied from the 300 bushels B intends to supply). Where there is an excess supply on carte blanche ordering, this excess is transacted at the highest demand price. Buyers and sellers who have struck a bargain slap each other's hand to indicate the fact, and the exchange's book-keeper will record who sold or bought how many bushels. However, he will not record the price at which the bargain was struck.
The structure of auctions
19
In the market, however, B still wants to sell his remaining 200 bushels at £10.00. When we look over to the demand side, F, G and H still want to buy 100, 300 and 300 bushels respectively at £10.00, and thus there is an overall demand of 700 bushels in total. Of this, H's demand for 300 bushels will be withdrawn if the price goes above £10.00, G's demand for 300 bushels will disappear if it goes above £12.00; but as long as no seller is found F will stay in the market until his price of £13.00 is reached. Accordingly, the course of competitive trading will be different thereafter according to which of F, G or H, B transacts with. Competitive trading consists of transactions through a single agreed price, and transactions made in the course of the auction are all settled at the final price (equilibrium price). Even if H were to transact with B at the price of £10.00, should the price go above £10.00 H must pay B the higher figure. That is, even though it was H's intention only to buy rice at £10.00 per bushel he will have bought rice at a price above £10.00. In order to avoid such a mistake, H must be permitted to resell (without cost) the rice he has bought from B. If B transacts with H and the price then goes above £10.00, H will immediately resell; but when B transacts with G or F they will not resell what they have bought until the price rises to £12.00 and £13.00 respectively. Since the way in which resale orders appear differs, the course of competition trading thereafter depends on whom B transacts with now. Let us suppose that B has struck a bargain with H. Since only 200 bushels remain of B's supply, even after buying from B, H will probably continue waving his hand shouting 'buying 100 bushels at £10.00'. Since the total supply at £10.00 has already come on the market there will be no supply to satisfy the demands of H, F and G who remain on the floor of the exchange. However, if the price were to rise to £11.00 C will want to sell 200 bushels. Yet at £11.00, H's demand (100 bushels) would disappear from the floor; but because F and G still want to buy it is possible these bargains will be struck between C and F or G. Whilst the total demand is for 400 bushels and the new supply is only the 200 bushels of C, there is also the resale from H. That is, H has already bought 200 bushels at a price of £10.00, but since he has no intention of buying a single bushel at a price of £11.00, as soon as it reaches £11.00 he has to cancel the order he made earlier. As explained, in order to do this H must resell the 200 bushels he has already bought in order to offset past purchases. Total supply is now 400 bushels, including the supply from this resale. This is equal to total demand and bargains will be made between C, H and F, G
20
Markets and the price mechanism
at the price of £11.00. This fact will then be recorded by the exchange's book-keeper. On the floor of the exchange, D waves his hand shouting 'selling 100 bushels at £12.00', and H shouts 'buying 300 bushels at £10.00', but there is no longer anyone in the market who will buy at £12.00 or sell at £10.00. This is because B who wished to sell at £10.00 and F and G who were prepared to buy at £12.00 and £13.00 respectively, have already completed their selling and buying under the preferred conditions of a price of £11.00. If the price is £11.00, D and H cannot complain at not selling or not being able to buy. Thus £11.00 is the price (equilibrium price) which none of the traders then present in the market have any objection to. When he judges an equilibrium price has been achieved, the watchman hits a clapper and declares that competitive trading is over. The equilibrium price governs all transactions and has established the rule of one price for one good. Prohibition of disorderly bidding The above simulation of competitive trading in the Tokyo Rice Exchange followed the rules and normal practice of Exchange. That is, as long as there are outstanding orders to buy at a given price, it is forbidden to put in a selling bid at a lower price. Conversely, whilst selling orders are not completely exhausted at a given price it is not permitted to put in a buying bid at a higher price. To allow these practices would be to condone conduct inimical to the orderly working of the market. When there are outstanding orders on the market to buy at a given price /?, and sellers had in any case been prepared to sell at a price lower than p, it is only commonsense for them to sell at p. Not to do so, and to try to sell at a price lower than p would be disturbing. Similarly, when there are outstanding orders on the market to sell at a given price/?, and buyers had in any case been prepared to buy at a price higher than /?, it is commonsense to buy at p. Not to do so, and to try and buy at a price higher than/7 would be equally disturbing. Bidding in contravention of these normally accepted conventions with regard to transactions is known as 'disorderly bidding' and is prohibited. Therefore, the only bids that can be lodged when outstanding bids to buy remain on the market (an excess demand) are bids to sell at a higher price. Similarly, the only bids permitted when outstanding offers to sell remain on the market (an excess supply) are bids to buy at a lower price. In fact, the reason why, in the above simulation of competitive trading, the price rose from £10.00 to £11.00 was because of the rule of the Exchange whereby offers to sell at £11.00 come into consideration because there are outstanding bids to buy at £10.00. That is, under this
Formulations of economists
21
rule, if there is an excess demand (supply), the price will rise (fall). The competitive trading mechanism is merely a kind of 'social computer' which eventually discovers equilibrium prices by adjusting prices according to this formula. 3 Formulations of economists
Demand and supply curves How then do economists analyse equilibrium prices? Economists say that prices are determined where the supply and demand curves intersect, but in our example of an exchange, how can we depict the demand and supply curves? Let us begin with the supply curve. Because the only supply will be from carte blanche sales when the price is less than £10.00, total supply is 300 bushels. At a price of £10.00, B will try to sell 300 bushels, which together with carte blanche sales will make a total supply of 600 bushels. At £11.00 B will, of course, continue to supply, and when we include the 200 bushels from C who now joins in the selling, total supply will be 800 bushels. At and above £12.00, supply will be 900 bushels. This is shown graphically in the series of black dots of Fig. 1, which places price on the vertical axis and quantity on the horizontal axis. Because price changes at discrete intervals in £1 units, even if we join the black dots in a line there will be no meaning to the spaces between neighbouring black dots. In our case the series of sporadic black dots makes up the supply curve, but in economics we usually consider that price changes continuously, and we show the supply curve by means of a continuous curve. The demand curve can also be derived from Table 1 by a similar procedure. At prices above £13.00 there is only carte blanche demand and therefore total demand will be 400 bushels; but at the price of £13.00 the 100 bushels of F demands are added (total demand equals 500 bushels). At £12.00 there will be G's demand as well (total demand 800 bushels). Since there is no new demand at £11.00, total demand will be 800 bushels as before, but at £10.00 H's 300 bushels will be added and total demand will be 1100 bushels. Thereafter there will be no change in total demand even if the price were to fall further. The series of hollow circles represents the demand curve. As is clear from Fig. 1, because a black dot and a hollow circle coincide at the price of £11.00, this is the price which equates demand and supply, and is therefore the equilibrium price. In order to arrive at such a price competitive trading began at £10.00, but an excess demand arose at that price. In order to satisfy this excess demand we should mobilize extra supply; for that purpose we should inquire into how the supply will be forthcoming at prices above
22
Markets and the price mechanism £'s 15 14 13 12 11 10
300
600
900
1200
Bushels
Figure 1
£10.00. When bargains were struck at a price higher by one unit according to our 'market conventions', there was an increase in supply of 200 bushels (from C), and the elimination of 300 bushels of demand (from H). Thus 500 bushels of excess demand were eliminated from the market. The price adjustment function So economists explain the mechanism of price determination, via competitive trading as follows. Let the total demand for rice be £>, the total supply be S and price be/?. Because D and 5 depend on/?, we have D = F(p) and 5 = G(p). The changes in D in response to changes in p depict the demand curve, and the changes in 5 give the supply curve. The equilibrium price p° will be determined by the intersection of the two curves. That is F(p°) = G(p°)
(1)
When the price is/?, if there is an excess demand (F(p) > G(p)) the price will rise (p > 0); and if there is excess supply (F(p) < G(p)) the price will fall (p < 0). Here p shows the extent to which price p changes at a given moment in the course of competitive trading. (To put this in more detail, when we take t as the parameter which shows the progress of competitive trading p shows the rate of change of p with respect to t,
The existence of equilibrium
23
dp/dt.) When demand and supply are equal (F(p) = G(p)), price will cease to change (p = 0). Following Samuelson, economists write the relationship between the degree to which price changes (whether positive or negative) and the volume of excess demand as follows p = H(E(p)) where E(p) = F(p) - G(p) and //(0) = 0
(2)
This is called the price adjustment function, and it is 'market conventions' which sustain this type of price adjustment mechanism. 4 The existence of equilibrium
No trading In our simulation the seller D did not participate in any phase of competitive trading. For this reason there may be those who consider perhaps that D has no relationship with the equilibrium achieved; but D is in fact one of those who uphold this equilibrium. This can be shown as follows. We assumed in the previous example that D would sell 100 bushels of rice at £12.00, but let us now suppose that D changes his mind and tries to sell at £11.00. On this assumption competitive trading would have proceeded as follows. Firstly, carte blanche selling and buying will take place and 300 bushels will be sold by A to E. E's remaining demand for 100 bushels will be met by agreeing a sale with B at £10.00. The further 200 bushels which B wishes to sell will be sold to meet the demand at the price of £10.00 (to H, for example). At £10.00 there will be excess demand because there will also be F's demand (100 bushels), G's demand (300 bushels), and H's demand (100 bushels). According to the market conventions sales at £11.00 will be sought after. (Up to this point the situation is exactly the same as in the previous simulation.) At £11.00, not only will C and D supply 200 bushels and 100 bushels respectively, but there will be an addition to supply of 200 bushels because H will cancel the purchase of the 200 bushels he made earlier and resell them. Therefore the total supply will be 500 bushels. On the other hand, at a price of £11.00, demand will total 400 bushels with F wanting 100 bushels and G 300 bushels. Suppose that the 200 bushels supplied by C and the 200 bushels re-sold by H are now sold to F and G, and D's supply remains unsold on the market. According to 'market conventions' price will be forced down to £10.00. If the price goes down to £10.00, D's sales will disappear from the market because he does not want to sell below £11.00. However, C, who does not want to sell at £10.00, has already disposed of his supply when
24
Markets and the price mechanism
the price was £11.00, and therefore he must buy back the 200 bushels he sold, now that the price has fallen again to £10.00. Further, amongst the buyers H has once bought but then resold what he bought. Thus he is in the same position as if he had not bought at all. H will accordingly appear as a buyer of 300 bushels and hands will be waved and bids made on the floor of the exchange for a total demand of 500 bushels. On the other hand, the supply will be zero. The price will be pushed up once more to £11.00. Thus when D sells 100 bushels at £11.00 the price will permanently oscillate between £10.00 and £11.00 and an equilibrium price will not be established. The market's watchman declares 'No Trading', and all deals made in the process of competitive trading up to then are cancelled. It is clear from this that the reason why equilibrium was established at £11.00 in the previous simulation was because D did not sell at £11.00, but attempted to sell at £12.00. Thus, because of this kind of supply schedule being assumed for D, he did not take any part in competitive trading, and it was this very lack of his activity which produced the equilibrium. That is to say, equilibrium is the product of the collective operations of all the buyers and sellers, who have gathered in the market, irrespective of whether they are active in the market or not. Multiple equilibria Thus D plays an important role in that he allows an equilibrium to exist and causes it to be established. Moreover, depending upon his supply schedule, not just one but several (say two) equilibria are possible. In order to see this, let us assume that in Table 1 D will only supply 100 bushels when the price is £13.00 and not £12.00. In this case the black dot corresponding to the price of £12.00 will move leftwards by 100 bushels and coincide with the circle at that height. That is, both £11.00 and £12.00 will be the equilibrium prices with D's new supply schedule, and which equilibrium price is established will depend upon what price competitive trading starts from. The following series of deals will bring about a new equilibrium price of £12.00. Hitherto carte blanche selling and buying first took place, and according to the common practice of the Tokyo Exchange the 100 bushels of excess demand left over was met from B's supply at the lowest price (£10.00). Let us now assume that the rules have been changed so that this left-over demand has to be met out of the supply which is forthcoming at the maximum price of £13.00. At this price B, C and D are all ready to supply (total supply is 600 bushels), and the 100 bushels of carte blanche demand together with F's demand for 100
Determining production prices
25
bushels at £13.00 will be met from amongst the supply forthcoming from B, C and D (say C and D, for example). Since 400 bushels of supply will not be taken up and will remain in the market, price will be lowered by 1 unit in accordance with 'market conventions', and demand at £12.00 will be forthcoming. G will demand 300 bushels at £12.00, and in order to cancel the transaction in which he sold when the price was £13.00, D will try to buy back 100 bushels. For this reason total demand is 400 bushels, whilst on the other hand supply consists of B's 300 bushels and C's 100 bushels, given a total of 400 bushels (C has already sold 100 bushels when the price was £13.00). At £12.00 supply and demand balance, and both are cleared off the market. That is, equilibrium has been achieved. Thus, where several equilibrium prices exist, competitive trading will converge on a low equilibrium price where it began at a low price, and on a high equilibrium price where it began at a high price. There are cases where the price from which trading begins is determined by the rules and conventions of the exchange; but there are also exchanges which leave this opening price to the discretion of the market's auctioneer. In either case, exchanges are the most powerful kind of organization for discovering and disseminating equilibrium prices. However, situations may exist where, even with such an organization, trading must be abandoned, with the cry of 'equilibrium price not achieved', because of the state of demand or supply; also it may be unavoidable that because of the rules of the exchange low (or high) equilibrium prices are consistently established. Despite this, however, exchanges produce, in normal circumstances, agreements which satisfy all those who have put in orders to buy and orders to sell, irrespective of whether or not they all are actually able to buy or sell on the market. In this sense the exchange must be seen as an extremely important mechanism which contributes greatly to the democratic management of the economy. 5 Determining production prices Commodities without an exchange However, it is impossible to provide this kind of exchange for every kind of commodity. Shares, public bonds and corporate debentures apart, exchanges are restricted to some of the products of agriculture and forestry and their processed products (such as grains, beans, potatoes, cotton, coffee, cocoa, wool, eggs, meat, rubber, bean wastes, sugar, wood, plywood etc.). They also exist for textiles (cotton yarn, silk thread, rayon yarn, rayon staple etc.), and for mineral products (gold,
26
Markets and the price mechanism
platinum, silver, copper, tin etc.). No exchanges where prices are determined exist for most manufactured products (automobiles, electrical products, furniture etc.) or for services (railways, hotels, films etc.). Even with commodities for which there is an exchange competitive trading does not necessarily take place, for there are commodities where only cross-trading is carried out even in exchange. Competitive trading where many suppliers compete with each other could not take place unless they all supply a commodity of very similar quality. Where quality is difficult to standardize (as with commodities such as silk cocoons and wool) each lot must be graded individually and its own price fixed. Consequently, only trading by bid or tender will be carried on, and not competitive trading. Furthermore, even for commodities which have an Exchange, not all transactions take place at the exchange. Besides cross-trading which goes on over the counters of shops, wholesalers form associations and gather periodically at a pre-determined place and sell these commodities by bid or tender. This sort of market can be called a 'quasi-exchange'. At a fully-fledged exchange sale or purchase can only be made through a recognized dealer; but in a quasi-exchange there are no restrictions on buyers, whether they be organized traders or individuals acting on their own behalf, who can come and go freely. This is despite the fact that suppliers are limited to the very wholesalers who organized the 'quasiexchange'. In any case, strictly supervised competitive trading only handles a part of the demand for an extremely small number of commodities, and the effectiveness of the competitive trading price mechanism is correspondingly partial and local. Of course, where an Exchange has been established for a particular commodity, transactions outside the exchange will rarely be carried out at prices very different from the equilibrium price determined in the Exchange. Price fixed within the Exchange will affect transactions outside to a greater or lesser extent. When the demand and supply in the exchange is an unbiased sample of the demand and supply of that commodity in the country in general, there can be little objection to treating the equilibrium price established in the Exchange as closely resembling the equilibrium price in the country at large. However, for commodities for which there is no exchange this method of approximation cannot be established. In considering how the prices of commodities with no exchanges are decided, it would be hasty, superficial, and dangerous to consider - without examining what actually happens and by analogy with transactions within exchanges - that prices will be established for those commodities which satisfy all the parties to the transactions.
Determining production prices
27
The full-cost principle5 How are prices decided in the case of manufactured products? Let us suppose for the sake of simplicity that factories produce only one kind of product. The factory computes a standard price by adding fair profits at a fixed rate to production costs. Production costs per unit of product are made up of (1) raw material costs, (2) fuel costs, (3) power costs, (4) the cost of chemicals, (5) the cost of tools, (6) water costs, (7) wages, (8) depreciation costs, (9) management costs, (10) various miscellaneous costs, and so on. If volume of production is very small, costs per unit product will be high; conversely, if the volume of production is extremely large, and the factory must mass-produce beyond its capacity, costs per unit will also be comparatively high. Therefore the production costs c per unit product will be a function of the factory's output x; where x is small c will be large, and with increases in JC, c will at first decline but will then increase (when it exceeds the capacity of the factory). That is, c will describe a U-shaped curve as x increases (we call this the average cost curve). Costs c increased by a fixed rate will be the price of the good (this fixed rate m is known as the net mark-up rate), and the part marked up will form the profit per unit of output. Price determination by this method is known as the full-cost principle.6 Thus in order to determine price p, c has to be determined, and for c to be fixed x must be determined. Since it takes a certain length of time to produce a product, those who run the factory must either produce only that amount for which they have received orders, or they must produce in anticipation of a volume of demand gauged through market research. Where output is small, production will be to order (for example, in the case of ships and special machinery); but with mass production it is impossible to take orders individually, and production must be by estimate. 5
6
On the full-cost principle see for example: Wilson, T. and Andrews, P. W. S., Oxford Studies in the Price Mechanism (Oxford: The Clarendon Press, 1951), which includes inter alia, Hall, R. L. and Hitch, C. J., Trice Theory and Business Behaviour'. Going back further Kalecki, M., Essays in the Theory of Economic Fluctuations (London: George Allen and Unwin, 1939) may be considered a pioneering study on the full-cost principle. See also Wiles, P. J. D., Price, Cost and Output (New York: Frederick A. Praeger, 1963). Among the ten items which make up the costs of production, the sum of (1) to (7) are known as prime costs c', and the sum of (8) to (10) are the indirect costs c". Because it is difficult to assess c" it is usually taken as a proportion of prime costs c' at a prescribed rate m", i.e., c" = m"c'. Thus c = c' + c" = (l + m")cf. Because price is determined according to p = (1 4- m)c, we can write p = (1 -I- m) (1 + m")c' = (1 + m')c', where m' is usually called the mark-up rate. Obviously there is a relationship between it and the net mark-up rate m: 1 + m = (1 + m')/(l + m").
28
Markets and the price mechanism
Where production is estimated in this way, output which is actually sold (xa) will not necessarily be equal to the estimated sales (xe)J When they are equal the factory will make its forecasted profits; but where some production estimated in this way is left unsold (that is, xe>xa), profit will be less than was forecast, and in the worst case profits may even be negative. That is, the factory will make a loss. If the factory is just about to go bankrupt it might hold a bargain sale of its products.at auction, as in the example of the sale of coats at the beginning of this chapter. Let us assume that a demand for 200 coats has been forecast, and that the price of producing a coat is £16.00 (cost + mark-up). Since demand depends on price it will be necessary to revise the initial forecast if £16.00 is too high. However, we will assume that the clothing firm judges it can sell 200 coats if the price is £16.00, and that 200 coats have been produced. However, in reality it transpires that there are orders for only 150 coats at £16.00 and 50 coats remain unsold. The company now has two choices. The first is to lower the price and sell off the 50 coats, and the second is to leave the price as it is and dispose of the remainder in the future. Where it adopts the first method people who have already bought the coats at £16.00 will complain; or else they may say nothing but lose their faith in the company. Thus even if the first method is beneficial to the company in the short term it will not be adopted other than in an emergency situation such as when it 'shuts up shop'. Instead it is likely to adopt the second method, which is to maintain its price, add the unsold goods to the future supply, and at the same time reduce its future output. Even in the converse case where demand is higher than the output it produces, price may not be raised as it would be under competitive trading. The 200 coats produced will steadily be sold off at £16.00 to the waiting buyers, but when the firm has sold about 150 coats it may well come to realize that 200 coats will be insufficient to satisfy total demand. Yet in this case too the company may not try to discourage demand by increasing the price. Instead it may either mobilize stock from the warehouse, or if it has no stock will simply declare 'sold out' and apologize to its customers. It is clearly advantageous to the company to raise its price and adjust overall demand to 200 coats rather than discard demand, but in real life managers and merchants do not behave 'rationally' as depicted in economic textbooks. They are more mechanical and bureaucratic than one would expect. Also, by sticking to the 7
We will discuss in Chapter 2 the way in which the estimated output xe is determined. Here, however, we shall treat it as if it has already been determined.
Determining production prices
29
price they first decided upon they feel they are keeping faith with their customers. Thus in the market for mass-produced, manufactured products demand and supply are regulated by the volume of output let onto the market rather than by price. Where production exceeds demand the volume of stocks increases, and accumulated stocks are discharged onto the market where there is an excess demand. Through adjusting stocks in this manner demand will usually be satisfied, apart from when there is an exceptional shortage of goods. Despite the fact that most manufactured products are sold by cross-trading rather than competitive trading, the principle of one price for one commodity is established for them, because each supplier will continue to sell them at an unchanged price as long as he is able to do so by adjusting stocks of his products.
Price competition under the full-cost principle If there is excess demand (supply) the price will rise (fall), so that buyers who cannot go to high prices will be eliminated (as will suppliers who cannot accept low prices). This form of competition via competitive trading is the basic conceptual formula of economists, and it has been widely disseminated and become accepted as common knowledge. On the other hand, under the full-cost principle - in which excess demand (supply) is eliminated by adjusting the quantity supplied and exerts no pressure on price - we may ask how are buyers and sellers induced to compete, and in what way are they eliminated from the market? To begin with the production costs of firms are not all the same. Even with the same company, each factory is located in a different place, uses different machines, and there are differences in the skill of the workers. Consequently, if the mark-up rate is given, differences in costs will produce differences in price. If there is no difference in the quality of companies' products buyers will, whenever possible, try to buy at the cheapest prices. Therefore high-priced products can survive only for the short period it takes for the information to circulate amongst buyers that their price is high. Even under the full-cost principle competition requires that there be one price for any given commodity. Therefore, factories and companies with high costs have to be content with smaller net mark-up rates, and rates will differ to an extent which reflects the differences in costs. If we take the profit per unit of production as TT, the costs of production as c, and the prices as /?, it follows from the formula for the full-cost principle p = c+ 77=(l + 7r/c) c = (l + m)c
(3)
30
Markets and the price mechanism
that those companies and factories with a high c must have a low net mark-up rate m (that is profit per unit of cost incurred TT/C will be low). Otherwise they will lose out as a result of price competition. Moreover, even if they can afford to tolerate a low mark-up rate for a while, they will soon fall into financial difficulties because funds must be used in the most effective manner; obviously, no lender will be interested in companies with a poor mark-up rate. The result is that such companies will have to reduce their output, and will be caught in a vicious circle where they lose the benefits of mass production and their costs of production rise more and more. In the end factories and companies whose efficiency is low will be closed down and their output will be eliminated from the market. Thus during the second stage of competition there will be a single, uniform cost of production for any given commodity. With a single price and a single cost of production per commodity, there will also be one mark-up rate per commodity. Moreover, if the mark-up rate differs amongst commodities it is unlikely that funds will flow in the direction of products with a low mark-up rate. Thus factories which produce these products will eventually get into financial difficulties. Thus in the end the mark-up rate must be the same for all products. In this way the principle of one mark-up rate for one commodity, and the tendency to uniformity in mark-up rates amongst all commodities, are both the outcome of funds flowing in the direction of efficient firms as a result of the calculation of economic self-interest. Competition under the full-cost principle does not stop at that point. Buyers do not necessarily buy commodities directly from the factory for there may be shops and stores of various sizes between them and the factory. There are corner shops, department stores, supermarkets, and consumer cooperatives, and small shops will buy in goods through small regional wholesalers, and the latter will buy in through large national wholesalers. The more stages there are in the distribution process, the higher will final retail prices be as the operating costs and the profits of each shop and store are added on, and the power to compete will be successively weakened. The necessity to compete may well eliminate an excessively complicated and roundabout distribution process. 8 Let us now compare the following as three of the simplest routes for getting consumer goods to their purchasers: (i) from factory direct to consumer; (ii) from factory to consumer via small retail shops, (iii) from factory to consumer via supermarket, chain store, or consumer coopera8
On the revolution in distribution due to the advent of supermarkets, see for example Zimmerman, M., The Super Market: A Revolution in Distribution, McGraw-Hill, 1955.
Determining production prices
31
tive. Since sales costs are a constituent part of total costs of production and depend more on how much is bought at one time rather than on the volume of output, the actual sales costs which are incurred and charged when goods are sold to large purchasers will be noticeably lower than the estimated average sales costs which are included in the price according to the full-cost formula. It is therefore usual for factories to sell to large purchasers such as supermarkets, chain stores, consumer cooperatives etc. at far lower prices than to individuals and small retail outlets. Thus if supermarkets and the like buy in quantity and economize on the costs of acquiring their goods, they will be able to sell at a much lower price than the corner shop. In some cases, they can even sell at retail prices which are much cheaper than those recommended by the factory. With rationalized and simplified block-purchasing and sales methods as their main weapon, supermarkets and the like have begun to meet the challenge of the very small retail outlets in the price war and ultimately defeated many of them. Most of those who work in small shops are either the owners themselves or members of their families, and the proportion of ordinary employees is extremely small. Consequently, their sales costs depend largely on how the owner costs his own labour and that of his family, and thus to some extent they are arbitrary and flexible. Small shopkeepers and their families have fought back by contenting themselves with very low incomes but there are limits to these tactics. Most of them will ultimately go under. That is, there are price wars even under the full-cost pricing principle, and as a result only the most economic distribution channels will survive and the rest will go to the wall. However, it must be remembered that price-cutting did not occur because of an excess supply of the commodity. Even if the lines they were interested in were selling smoothly and without hitch, the supermarkets would still devise some sales method which undercut their competitors and allow them to increase their share of total sales.9
9
There are, in addition, commodities whose prices are determined neither by the full-cost principle nor in a competitive trading market. There were for example taxis and rickshaws in the late 1920s in Japan. Because the price was decided by negotiation between the driver or rickshaw man and the client, it was a cross-transaction, and the price depended a good deal on who was more skilful at negotiating. Therefore the principle of one price for one good did not hold true. In that case, however, if the driver proposes a high price the client will summon another taxi, and if the client persists in offering an extremely low fare the taxi will seek other clients. Where there is excess supply, that is many empty cars, the position of the clients will strengthen and the price will fall; but where there is an excess demand, that is where there is a long line of clients waiting for taxis, the drivers will resolutely press for higher fares.
32
Markets and the price mechanism
6 Two types of market economy Neoclassical economists and Keynes Thus almost all manufactured industrial product prices are determined according to the full-cost principle, while the main agricultural, forestry and fishery products, together with some mining products, have exchanges where competitive trading takes place. The full-cost principle and competitive trading are very different price fixing mechanisms, and the character, physiology and dynamics of the whole of a national economy will differ according to whichever of the two mechanisms predominates. Since neoclassical economics (or so-called Walrasian economics) assumes that all commodities are exchanged via competitive trading, it will be effective in analysing economies where the competitive trading mechanism predominates. On the other hand the Keynesian-type of economics in the form developed by Kalecki10 can be seen as taking prices to be fixed by the full-cost principle. Therefore this type of economics is appropriate for analysing economies where manufacturing industry predominates. In what follows we shall refer to the pure model where the prices of all commodities are decided by competitive trading as flexprice economies or neoclassical-type economies. We shall call economies where the price of all commodities is decided by the full-cost principle fixprice economies or Keynesian-type economies (strictly speaking, Kalecki-Keynes type economies), since prices are fixed independently of the excess demand for the commodities once costs are determined. Real economies fall between these two extremes, being mixed forms of flexprice and fixprice economies. We can calculate the degree to which real world economies are mixed by measuring the ratio of the value of output produced in the flexprice sector to the value of output produced in the full-cost sector; or roughly by the ratio of the value of the output produced in agriculture, forestry, fisheries and mining to the value of the output produced in manufacturing. If we compute this ratio for 1960 and 1973 by country as in Table 2, we see Britain and West Germany are fixprice economies nearest to the pure type. Japan and Italy in 1960 were still provided with fairly large flexprice sectors, but rapidly became fixprice types between 1960 and 1970. 10
See for example the previously quoted work by Kalecki and also Kalecki, M., Theory of Economic Dynamics: An Essay on Cyclical and Long-Run Changes in Capitalist Economy, London: George Allen and Unwin, 1954. We are below interpreting Keynes' theory along Kalecki's lines in spite of the fact that Keynes' own theory of the firm and his theory of prices are both neoclassical.
Two types of market economy
33
Table 2. The ratio of the value of output of agriculture, forestry, fisheries and mining to the value of output of manufacturing (unit: %)* Year
Japan
America
Britain
West Germany
France
Italy
1960 1973
57 25
23 24
19 13
20 16
31 24
58 37
* Computed from World Bank, World Tables 1976.
Table 3. Ratio of the value of agricultural output to manufactured output (unit:%)* Year 1789/1815 1801 1825/1835 1841 1860/1869 1872/1882 1896/1900 1901 1907 1913 1919 1929 1939 1949 1959 1969
Britain
Germany c
France
Italy
250 139 200 65 133
275'* 140
46 15 18
214* 95*
51 16° 13 13* 15 10 8
33 23« 15 9
30<* 23 16
214 181 138 103 91 59 35
a
value for 1920, bvalue for 1938, c values after 1945 for W. Germany only, lvalue for 1950, *value for 1908/10, 'value for 1861/65. 8 indicates that Transportation is included. * Taken from Carlo M. Cipolla (ed.), The Fontana Economic History of Europe, vol. 4(2), 1973, p. 811 and vol. 6(2), 1976, pp. 751, 752, 754.
It was clearly in the past that Britain became a fixprice economy (perhaps about 1890. See Table 3). Other countries at the time (not only Japan and Italy, but also America, Germany, France etc.) still had a proportionally large flexprice sector. If we take this into consideration,
34
Markets and the price mechanism
Table 4. Ratio of the population engaged in primary industry to that engaged in secondary industry (unit: %) Year
Japan
America Britain
1866-1875 1876-1885 1886-1895 1896-1905 1906-1915 1916-1925 1926-1935 1936-1945 1946-1955 1956-1965 1966-1975
1700 866 554 344 250 274 188 94 36
200 200 150 123 100 79 71 55 34 21 13
_ 26 22 19 17 14 13 13 11 6
Germany France
Italy
_ 116 92 85 71 71 66 52 25 14
153 121 158 139 157 124 121 132 68 40
113 79 81 59 60 54 30
The figures are for a given year within the period. They were computed from pages 374 and 375 of the Bank of Japan, Statistics Bureau, 'Main Economic Statistics of Japan from the Meiji Period Onwards', 1966.
it was certainly no accident that the 'Keynesian Revolution' in economics occurred in Britain. Just as economics first flourished in Britain which was the first country in the world to develop a modern capitalist economy, so it was also Britain which was first to become a fixprice economy and to devise the economic science for it - Keynesian economics. Considered in this way, the historical trend of academic theory from the neoclassical economics to the Keynesian was not due simply to autonomous developments in pure theory, but should be seen as the response of theory to substantive changes in the economy itself. Historical background to the making of fixprice economies All countries, even those which are now advanced industrial countries, were once primary product producing countries with the emphasis on agriculture, forestry and fisheries. The industrial revolution occurred in Britain during the period 1760-1830, and in Japan during the period from about 1890 to 1910. But even after their industrial revolutions in most industrial countries a large part of the total population worked in the primary sector. What the fact revealed in Table 4 shows - namely that in most industrial countries until quite recently the ratio of the working population in primary industry (agriculture, forestry and fisheries) to that in secondary industry (broadly defined to include mining and construction as well as manufacturing) was over 50% - is
Two types of market economy
35
that these countries still retained quite a large primary sector for a long period of time even after they had undergone their industrial revolutions. Britain alone is the exception. In Britain by the beginning of the 1880s the population employed in agriculture, forestry andfishingfell to about one quarter of that employed in the secondary, industrial sector; thereafter it gradually declined, and now the ratio of the former to the latter is less than 10%. Britain has relegated primary industries to the status of an appendage to the economy, and whilst the same kind of phenomenon has been late in coming it is appearing in other industrial countries too. By the mid-1950s in America and by the late 1950s in West Germany, primary industry had contracted to the relative size of British primary industry in the 1880-1980 period, and the same phenomenon of a flight from agriculture was readily identifiable throughout the 1970s in Japan, France and Italy. If we call this change the second industrial revolution, that is one which almost completely eliminates primary industry and establishes secondary industry as the mainstay of the economy, we can say that the second industrial revolution took place in Britain in the 20-year period from 1876 to 1895, and in America and Germany from 1950 to 1965. In Japan it perhaps occurred in the period after 1965. Why then did Britain experience her second industrial revolution much earlier than other countries? Britain, which had taken the world lead in achieving the first industrial revolution with its accompanying economic domination and military strength, completed an Empire that spanned the world by about 1850 and controlled the seven seas. Comparing the colonies with the mother country, it hardly needs stating that the latter was relatively advantageously placed as regards industry (had a comparative advantage in industry), while the former were comparatively well placed for the production of agricultural goods. Since it was realized that the loyalty of the colonies would be difficult to retain under the mercantilist, imperialist policy followed hitherto whereby the mother country exploited the colonies, liberal political thought and free trade theory became popular from about 1820.n When the principles of comparative advantage of the free trade school were applied to the British Empire, then obviously the mother country would concentrate on industry and agriculture would be transferred to the 11
American independence was achieved in 1776, Canada rebelled in 1837, and Australia in 1854. In order not to repeat the failure in America, negotiations were necessary on liberalization of trade with the colonies, particularly the white dominions. A policy of free trade with the colonies was wholeheartedly supported. But since free trade was later found to hinder the development of industry in the colonies, they eventually became unsatisfied with it and went as far as to demand a protectionist policy.
36
Markets and the price mechanism
colonies.12 The mother country received supplies of raw materials (agricultural products) from the colonies, and the colonies purchased consumption and capital goods (finished manufactured products) from Britain. It was indeed this kind of co-prosperity based on free trade between mother country and colony which provided the material basis for getting the colonies to pledge their loyalty to Britain over such a long period. The philosophy of free competition was also promoted and disseminated. As a result of the second enclosure movement which reached a peak between 1800 and 1820, landlords monopolized vast acreages and many farmers lost their land. British agriculture survived in the form of large-scale capitalist agriculture where agricultural capitalists rented huge areas of land from the great landowners and employed agricultural labourers to work there. On the other hand, farmers who had lost their land and the poorer townspeople who had fallen on hard times sought a new world and emigrated to the colonies. Every year after 1830 hundreds of thousands of people emigrated to Canada, Australia, New Zealand and South Africa. Although British agriculture rationalized itself and managed to survive, agricultural production hardly increased in money terms in the 60 years from 1841 to 1901, whereas industrial production increased 4.5-fold during that period. As a result the ratio of the value of agricultural production to that of industrial production declined from 65% to 15%, and the ratio of the agricultural population to the industrial population had already declined to 26% by 1881. We can conclude that the abnormally small British agriculture of the time was the outcome of the thorough application of the principle of comparative costs. It was also the achievement of the British liberal colonial policy of the times. It is therefore hardly surprising that other industrial countries lacking the background against which Britain operated, failed for a long time to reduce the size of their agricultural sectors. Japan, for example, constructed her Manchurian Empire (1932-1945) on the British model and tried to rid herself of her surplus agricultural population by doing so; but her policy was not in essence liberal, and was more the old-style imperialist colonial exploitation. Nothing was more important for the establishment of free trade between Japan and Manchuria and the achievement of co-prosperity than that the newly-born Manchurian Empire should not be drawn into war. But the military authorities and politicians of the time in Japan lacked such insight and brought on war after war. Japan's ambition to build a 'Japanese Commonwealth' in East 12
See Additional Note b.
Two types of market economy
37
Asia - the Great East Asia Co-Prosperity Sphere - collapsed without bearing fruit. The reason that industrial countries other than Britain succeeded or are about to succeed in achieving a second industrial revolution after the Second World War is because peace continued after the war and the international economy got back into working order, so that an international division of labour based on the principle of comparative advantage was able to be established without political ties between the mother country and the colonies. For establishing trade along the lines laid down by the principles of comparative advantage it may have been necessary in the nineteenth century not only that these political relationships should have existed, but that in addition many people should have emigrated from the mother country and produced the conditions for trade between like and like (between those remaining in the mother country and their colonial cousins).13 At the time only Britain was provided with such conditions and other countries had to wait for their time to come. 13
Even in the period when economic policy was liberal and laissez-faire, Britain operated a strict discriminatory policy against the non-white peoples of India.
The function of exchanges
1 The multilayered structure of market economies
The case of industrial manufacturing countries There are on the one hand goods whose prices are decided by competitive trading (agricultural and mining products etc.), and on the other hand goods whose prices are determined by the full-cost principle (mainly manufactured industrial products). The prices of commodities in the national economy are not regulated by a single principle; commodities may be put into two separate compartments in which quite different principles predominate. That is, the price mechanism of the national economy is grounded on two different principles, but this dual price mechanism is not simply a parallel mix of the sectors where the full-cost principle operates and those which accord with the competitive formula. It is more like a multilayered structure with one founded on the other, as we shall see below. Such a multilayered structure will certainly not crumble away even in the case of a pure industrial manufacturing country whose agricultural and mining sectors have completely disappeared, so that all the prices of commodities produced in that country will be decided by the full-cost principle. For such a country to produce, raw materials (agricultural and mining products) which must be imported from abroad are necessary, but in the markets from which they are shipped their prices will be decided by competitive trading. In deciding the costs of manufactured industrial products, the prices of raw materials must be converted into units of the country's currency on the relevant foreign exchange rates, and wage, processing and other costs must be added to the raw materials cost so calculated. That is, the full-cost principle is constructed on the theory of competitive trading which predominates in the areas producing raw materials and on the exchange rate theory which controls the international economy; and in the final analysis of costs we must go right back to clarifying the structure of competitive trading in the areas producing raw materials. 38
Interrelationship of demand and supply within a single exchange
39
The case of countries producing raw materials Similarly, the price fixing mechanism in a country producing raw materials is not a simple structure of the competitive trading type alone, but is dependent upon the full-cost principle of areas importing raw materials. Raw materials' prices fixed by competitive trading depend on the demand and supply of the raw materials, but this demand depends on the demand for the industrial product whose raw material it is and thus on the prices of industrial products. If the price of raw materials rises the cost, and therefore the price, of industrial products will rise accordingly, and as a result demand for industrial products will decrease. Demand for these raw materials will also decrease, but because the chain which links raw material prices and the demand for them is the full-cost principle, the competitive trading mechanism in purely raw material producing countries will also rest on the full-cost principle in overseas countries. Consequently it rests on the pricing in foreign exchange markets and the full-cost principle. 2 Interrelationship of demands and supplies within a single exchange Parallel competitive trading
From these considerations we now find that analysis of the competitive trading mechanism is of great importance to economics - not only for agricultural countries and those large capitalist countries whose own agricultural sectors supply materials to their industries, but also for the 'medium-sized' countries which lack natural resources and can, therefore, accommodate only manufacturing industries so that they are entirely reduced to fixprice economies. The proper analysis of exchanges is still indispensable in this age of fixprice economies, just as it was when all the sectors of the economy were very much of the flexprice type.1 Despite changes in the nature of national economies, analysis of the way price changes spread from one commodity to another within an exchange, and furthermore analysis of the circumstances in which price changes are mutually disseminated amongst exchanges, are important even today. In this sense the essence of the analysis contained in neoclassical price theory has certainly not been superseded. In markets (exchanges) which have been organized so that competitive trading is carried out it is usual that only those goods which belong 1
What is meant in this volume by the 'fixprice economy' is not necessarily an economy where all prices are kept constant. It means only that the prices of products are determined by the full-cost principle and, therefore, prices are not flexibly adjusted according to excess demand for (or excess supply of) those commodities.
40
The function of exchanges
to a more or less same category are exchanged. It is not possible to exchange cotton fibre or share in the sugar exchange. However, sugar is classified according to kind (using a representative characteristic - for example, colour), into white, yellow, brown, etc., and the price of each kind of sugar is set in its market. Of course a particular market (the sugar market) is influenced by events outside the market and by other markets (the cotton fibre and share markets). However, in what follows we shall separate a particular market from all other markets and consider it in isolation. Such abstraction distorts reality but it is the most effective simplification for making clear how the price mechanism operates within a particular market. Let us now suppose that only white and yellow sugar are traded at the sugar exchange, and let the demand for those commodities respectively be Dx and D2, the supply Sx and 52 , and prices (money prices) bep1 and p2. Contrary to the example of competitive trading of a single kind of rice on the rice exchange given in the previous chapter, in this market two kinds of commodity are traded in. The demands for white and yellow sugar are in competition one with the other, and if the latter increases the former may well decrease. Since demand for yellow sugar will depend on its price, demand for white sugar depends not only on its own price but also on that for yellow sugar. In analysing the price formation process of markets where competitive trading takes place in parallel for two (or more than two) kinds of goods we must explain the supply and demand for them, and in order to do so we must know how the demand (or supply) of these goods is mutually interlinked. However, in actual exchanges competitive trading for white and yellow sugars does not necessarily take place in parallel. The method fairly widely adopted in most exchanges is a sequential one, the next good is dealt with after competitive trading in the first good is completed.2 But, as is natural where there are many kinds of commodity which must be handled there is not enough time to take bids for each commodity in sequence and therefore the commodities are divided into 2
In Walras' theory of tatonnement it is assumed that the prices of commodities are adjusted in a fixed order (Walras, ibid, pp. 169-72). Against this Arrow and Hahn feel it clear that Walras had not literally thought prices were adjusted in afixedorder, and that he had assumed a price adjustment mechanism with a sequential system wholly because of its explanatory convenience. See Arrow, K. J. and Hahn, F. H., General Competitive Analysis, Edinburgh, Oliver and Boyd, 1971, p. 5. Nevertheless, it is clearly a fact that in actual exchanges competitive trading for commodities does take place in a fixed sequence. However, since it is clear that Walras did not regard his 'theory of tatonnement' as simply a theory of price determination within a single exchange, we should consider him as having posited his theory as one which was perhaps based on the pure model of a single exchange but which did not differ greatly from the economy as a whole.
Interrelationship of demand and supply within a single exchange
41
several groups (group 1, group 2, etc.) and competitively traded for within the group in a fixed sequence, whilst the sessions for each group take place at the same time and in parallel. Analysis of cases where competitive trading at a particular exchange takes place at a fixed time for one commodity only was given in the previous chapter, so we will analyse below the case where bids take place at the same time for two commodities (white sugar and yellow sugar). Substitutive goods and complementary goods Related commodities are classified into substitutive and complementary goods. Like white and yellow sugar, substitutes have almost the same use, and the term denotes goods one of which can be substituted for the other. Complementary goods, like coffee and sugar, are those where if one is used so is the other, and the term denotes goods where if one is bought so is the other. Independent goods in contrast, are those which are neither substitutes nor complements, that is, goods where demand for one is quite without influence on the other.3 Because more or less the same kind of goods are usually transacted for at the exchange, there only substitutive relations within the same exchange and complementary relations would be an absolutely abnormal phenomenon. Complementary relations appear between exchanges (between the sugar and coffee exchanges),4 and because information does not flow as quickly and accurately between exchanges as within them, demand and supply within an exchange frequently ignore events which have occurred at other exchanges or only respond slowly. In the extreme case where telephone links between exchanges are cut off and each exchange becomes an isolated and closed compartment, in market terms coffee and sugar will become independent goods even though they are complementary in their uses. (Likewise a good which is a substitute on the basis of use - for example, brown sugar - will turn into an independent good when it is transacted for at separate exchanges, and, in particular, where communications are cut off between exchanges. Thus substitutability and complementarity on the basis of use are merely necessary conditions for substitutability and complementarity in the 3
Definitions of substitutes and complements have been presented in various forms. The above definition belongs to the most unsophisticated and uncontrived amongst them, but is close to what Mosak calls 'gross substitutes'. See Mosak, J. L., GeneralEquilibrium Theory in International Trade, Bloomington, Indiana, The Principia Press
4
Inc., 1944, pp. 42-51. There are amongst them those like the New York Coffee and Sugar Exchange which are composite exchanges that handle two groups of commodities, within each of which there prevails substitutability, while there is complementarity between them.
42
The function of exchanges
market, and for both use and market qualities to coincide the further condition of 'smooth and speedy flow of information' must be satisfied.) So how is the demand for the two kinds of sugar, yellow and white, interlinked? To begin with, (i) if the price for white sugar px increases, the demand Dx will decrease, and the demand for yellow sugar D2 will increase. (That is, there will be substitution from white to yellow.) (ii) If on the contrary the price of yellow sugar p2 rises, it will produce substitution from yellow to white sugar, D2 will decrease and Dx increase. However, that is not all. Where px and/? 2 increase in identical proportions there will be no change in relative prices between white and yellow sugar, and since neither will be comparatively more expensive than the other, substitution between them will not occur. Nevertheless, because white sugar will become more expensive compared to other goods besides yellow sugar, people will probably economize on white sugar, and the demand for white sugar will resultingly decline. That is, (iii) a proportional price increase in px and p2 will reduce demand Dx. Similarly, it will also cause a fall in D2. Supply will change as follows. Sugar suppliers will supply part of the quantity Xx and X2 which is available, and will keep the remainder Xx and X2 for future business or for household use. How much they keep for future supply and how much they supply now will be influenced by forecasts of future sugar price and production levels, but we shall suppose here that there is no expected change (or at least that there is no change whilst competitive trading is taking place). In this case supply is likely to change as follows. (Note that St = Xt-Xhi = l,2.) (i') If the price for white sugar px increases the attraction of selling now rather than in the future will increase, and because people will rush to sell now what they had put aside to sell in future, Sx will probably increase. Moreover because sugar suppliers will try to compensate for a part of the decline in the quantity Xx of white sugar they retain by increasing the quantity X2 of yellow sugar which they retain, the quantity S2 of yellow sugar supplied will decline by that amount. In the same way, (ii') when the price p2 of yellow sugar increases, S2 will increase and Sx decline. Further, (iii') where/?! andp 2 increase in equal proportion, Sx will be indirectly affected by p2 and S2 by pu but these cross-effects will be too weak to be cancelled out by the direct effects of the change in their own prices, px and p2 respectively, so that the proportional increase in the prices will cause Sx and S2 to increase. If we call E the quantity of excess demand which is obtained by subtracting the quantity supplied from the quantity demanded, then EX = DX — Sx, and E2 = D2 — S2. From (i)-(iii) and (i')-(iii') above we can see that the volume of excess demand fluctuates as follows, (a) If px
Price repercussions amongst substitutive goods
43
Table 5. Excess demand functions where two commodities are substitutive A rise in Pi
Pi
Effect on E2
increases, Dx will decrease and Sx increase, therefore Ex will decrease. With regard to E2 on the other hand, because D2 increases and S2 decreases, E2 will increase. Similarly, (b) if p2 increases, E2 will decrease and Ex increase, (c) If px andp 2 increase in proportion, Dx and D2 will decrease and Sx and S2 will increase, and therefore Ex and E2 will both decline. Propositions (a)-(c) can be shown as in Table 5. However, a is a parameter showing the proportional changes inpx and/?2. That is, if a increases both prices will increase in proportion, and if a decreases both will decrease proportionally. 3 Price repercussions amongst substitutive goods
General equilibrium within an exchange At what equilibrium prices does competitive trading cease when excess demands for two commodities change as in Table 5? Further, how do equilibrium prices change where demands (say, carte blanche orders) have been added during the course of competitive trading? We shall deal with these problems in this section, but because they are meaningless when no equilibrium prices exist with a given excess demand table, we shall assume below that prices change continuously during competitive trading in order to guarantee the existence of equilibrium prices, and that accordingly excess demands will also change continuously. As we saw in the previous chapter, in actual competitive trading prices change discretely in fixed unit slices, and thus so do excess demands. As a result it is possible for excess demand to arise at a given price, and for excess supply to arise at the next highest price unit. If the price is raised since an excess demand has arisen in the market, an excess supply will appear, and when the price is lowered again an excess demand will arise and the price will have to be raised again. A so-called 'unquotable' condition will develop where the price just goes up and down and an equilibrium will never be attained. In order to exclude the possibility of
44
The function of exchanges
the non-existence of an equilibrium because of discontinuous price changes, we shall assume below that prices change continuously and that both the excess demand functions Ex = Ex(px, p2) and E2 = E2(px, p2) are continuous functions of prices. Firstly, in order to confirm the existence of equilibrium let us find the points where demand and supply are equal in the plane with the two axes representing prices/?!, p2 (i.e. the points at which excess demand is 0). Even if the demand and supply for white sugar are equal this may not necessarily be so for yellow sugar, and therefore the points where Ex(pXy p2) = 0 are merely partial equilibria (the points at which equilibrium is achieved for white sugar) are not necessarily a general equilibrium point where demand and supply for both white and yellow sugar are in equilibrium. It goes without saying that the general equilibrium point is that point where 'the set of partial equilibrium points for white sugar' and 'the set of partial equilibrium points for yellow sugar' intersect; that is, the point which belongs to both sets. As we shall see below, since we can express graphically the set of partial equilibrium points for white sugar as a curve of partial equilibrium for white sugar and similarly for yellow sugar, the intersection of the partial equilibrium curves for each kind of sugar gives the general equilibrium point. 5 Partial equilibrium curves These curves may be derived as follows. To begin with, if we fix the price of yellow sugar at an arbitrary fixed value (for example p*2), the excess demand for white sugar Ex(px, p*2) will solely depend on its price px. If/?! is 0, i.e. white sugar is free, there will be an excess demand (that is Ex will be positive). However, since Ex will decrease when px increases, when px becomes large, for example when it reaches /?*, Ex may well be 0. Furthermore, if/?! gets larger, Ex will become negative. Now at p[ let Ex be negative; Ex(p[, p:2)<0. Fixing px at p[ and adjusting p2 so that Ex once again equals 0 means that we must raise p2. The reason for this is that, as assumed in Table 5, when p2 increases Ex will increase. Thus Ex begins to increase from a negative value, and at/?2 it once more becomes 0; Ex(p[, p2) = 0. Clearly the point a' = (/?{, p2) will be above and to the right of the point a* = (p*,p2). Since at both points a* and a'', Ex is 0, these points are both partial equilibrium points for white sugar. The curve Ex = 0 in Fig. 2 is the locus of such points. As the figure shows the curve rises to the right, and on the left of the curve the excess demand for white sugar is positive and to the right it is negative. 5
The analysis below follows Hicks' Value and Capital, Oxford, Clarendon Press, 1939, pp. 66-77.
45
Price repercussions amongst substitutive goods
Pi
£,=0
Pi
Pt
Pi
Figure 2
The partial equilibrium curve is that obtained by joining partial equilibrium points. Let us join a point p on the curve with the origin, and let us call the straight line pO thus obtaining the price line. Its slope indicates the price ratio of yellow to white sugar (that is, the relative price). The partial equilibrium curve of white sugar rises to the right, but even with a curve sloping upward to the right, there are two cases. One, as in Fig. 3a where the curve cuts the price line from above. However, as a proportional increase in both prices is assumed to give rise to a decrease in the excess demand for white sugar, the case of Fig. 3a cannot occur, and the partial equilibrium curve will necessarily take the form it does in Fig. 3b. We can easily prove this. In Fig. 3a, let the price line pO extend for example to 1.3 times its length to p"0. Thus we get at point p, Ex(pu p2) = 0 at pointp"yEx{p'ip'i)>Q
(1) (2)
Expression (2) is obtained because in Fig. 3a the point p" lies to the
46
The function of exchanges P2
Pi Figure 3 a
Pi Figure 3b
47
Price repercussions amongst substitutive goods
P2
(a)
(c)
Pi
Figure 4
left of the El = 0 curve where Ex is always positive. However, (2) contradicts Table 5. Since the last column of the table states that Ex must decrease when prices increase proportionally, when prices change from p to p", Ex will decrease from 0 to a negative value, and at point p" we should get £i
(3)
Clearly (3) is in contradiction with (2) or Fig. 3a, and we know that Fig. 3b is the correct graph. Existence and stability of general equilibrium The partial equilibrium curve for yellow sugar may be drawn in the same way. This also rises to the right, and cuts the price line from below when seen from the p2 axis (seen from the px axis, it cuts the price line from above). Figure 4 may be obtained if we draw the partial equilibrium curves for white and yellow sugar together on the same plane. At the
48
The function of exchanges
point where two curves intersect, /?°, we obtain a general equilibrium because there both Ex and E2 are simultaneously 0. These two curves segment the plane into four zones. For either curve, on the side nearer to the origin the excess demand for its corresponding good is positive, and on the side furthest from the origin its excess demand is negative. Therefore in Fig. 4, £ i > 0 , E2<0 in zone (a), Ex > 0, E2 > 0 in zone (b), Ex < 0, E2 > 0 in zone (c) and Ex < 0, E2 < 0 in zone (d). In well-organized markets prices are regulated by the relative magnitude of demand against supply. That is, competitive trading takes place in accordance with the rule that the price of a good rises where demand exceeds supply (excess demand is positive), and falls in the reverse case. Accordingly, px will rise and p2 fall in (a). In this zone therefore, the arrow which shows the direction of the price change will point down to the right. Similarly, the arrow will be upward pointing to the right in zone (b), upward pointing to the left in zone (c), and downward pointing to the left in zone (d). p2 will be unchanging at the boundary of (a) and (b) but px will rise there and the arrows will in consequence point horizontally to the right. For the same reason the arrows point horizontally to the left at the boundary of (c) and (d), straight upwards at the boundary of (b) and (c), and straight downwards at the boundary of (a) and (d). In this way we may obtain the graphical arrangement of the arrows shown in Fig. 5.6 We may compare this figure with a maritime chart which shows the flow of ocean tides. At whichever point on the ocean a piece of driftwood floats it will ultimately float to the general equilibrium point/?0, given that such ocean currents exist. That is, competitive trading will reach p° and cease. Hicks' Law of prices Lastly let us consider how the general equilibrium point is affected when there have been additional carte blanche orders for a given good, for example white sugar. Since a fixed quantity of white sugar will be bought - for example 100 units - whatever the price, the original points where Ex = - 100 will become new partial equilibrium points (that is, the points where Ex +100 = 0). Since the new partial equilibrium curve for white sugar must be located within the zones (c), (d) where excess demand was originally negative, the partial equilibrium curve will move right as in Fig. 6. AEX in this figure indicates the volume of additional carte blanche orders for white sugar. In Fig. 6, p° is the old general equilibrium point and/?1 the new one. 6
See Additional Note c.
49
Price repercussions amongst substitutive goods Pi
X \
\1 I/ \
0
\
x P\
Figure 5
£•,=0
0 Figure 6
P\
50
The function of exchanges
We may obtain the following three propositions if we compare the two: (i) the price of white sugar will rise (that is, p\>P\)', (ii) the price of yellow sugar will rise too (that is, p\ >pi)\ and (iii) the ratio of the price increase for yellow sugar will be lower than that for white (that is, Pi/p^Pi/Pi)Laws (i) and (ii) are self-evident from Fig. 6, but to obtain (iii) we should join p° and p1 respectively to the origin O and compare the two price lines. The respective slopes of these lines show Pi/Pi andpl/pl* a n d the figure indicates that/^/p? >P2/p\- Therefore as (iii) states it must be that/?}//?? >pl//?§. The above was premised on the fact that there are two kinds of goods, but this assumption is not indispensable to our deriving these conclusions. However many kinds of goods there are, that is, where there is an exchange divided into many sectors where competitive trading takes place for a single good in each, and in the exchange as a whole competitive trading for many goods takes place simultaneously and in parallel, we may obtain the same conclusion notwithstanding the number of goods as long as there exists the sort of dependency of excess demand on prices as is shown in Table 5.7 Firstly, the general equilibrium is necessarily stable (that is, the flow of prices created by excess demand in the end carries the market to a state of general equilibrium). Secondly, where excess demand for one good increases (i) its price will rise, (ii) the prices of all other goods will rise too, and (iii) their rates of increase will not be as great as that for the good whose demand has increased. These three laws of price changes are called Hicks' Three Laws after their discoverer, and are the most fundamental laws of prices obtainable with regard to the competitive trading mechanism. From the previous session to the next Let us suppose that after the session for white and yellow sugars has ended in our exchange, competitive trading now takes place for medium white and brown sugars. Let medium white sugar be the third good and brown sugar the fourth, and let their respective demand, supply and 7
Where for example there are three goods, Table 5 may be expanded as follows; however, a is the parameter which shows the proportional changes in pl9 p2, Py Rise in Pi
Effect on E2 £3
Pi
P3
Price repercussions amongst substitutive goods
51
price be D3, S3, p3 and D4, S4, /?4. Because medium white and brown sugar are substitutes, there is no objection to our considering what we have said before about white and yellow sugar as being perfectly true as it stands for them too. A general equilibrium which equates both demand and supply for medium white and brown sugars is possible, and moreover it is a stable equilibrium. That is, from whatever price set competitive trading commences, a general equilibrium price set will be discovered sooner or later and competitive trading will end at that point in time. This will be so as long as the exchange teller adjusts prices according to the rules (that is, as long as he adjusts prices according to a formula whereby he raises prices if there is an excess demand for the respective commodities and lowers them if excess supply arises). Hicks' Laws will also hold. That is, if the demand for either medium white or brown sugar increases not only will the prices of both rise, but the rate of price increase for the sugar whose demand has increased (for example, medium white sugar) will be larger than that for the other (brown sugar). However, we must not forget that apart from this the following price repercussions exist. The increase in the demand for white sugar (or yellow sugar) will affect the prices of medium white and brown sugars. Not only are these two kinds of sugar now in session mutually substitutable, they are also substitutable for the goods (white and yellow sugars) whose trading session has already ended. If the demand for white sugar increases the prices of both white and yellow sugars will rise (according to Hicks' Laws), and therefore due to the substitutability between these sugars and medium white or brown sugar demand will switch over from white and yellow to medium white and brown sugars, and the demand for the latter two will increase. The result will be that the prices of medium white and brown sugar rise. (In Fig. 7, E3 and E4 are the partial equilibrium curves for medium white and brown sugars when there is no increase in the demand for white sugar, and £3 and E\ are their partial equilibrium curves after an increase in demand for white sugar. The new general equilibrium point is located to the above right of the old one, and therefore the prices of medium white and brown sugars will both increase.) Thus effects of the increase in demand for white sugar will continue to be present in the exchange after trading in white sugar has ended. That is, the increase in the demand for a good will not only force the prices of that good and all others traded in simultaneous sessions to rise, it will also force the prices of all goods traded in that exchange in sessions thereafter to rise. What then is the relationship between the rate of increase in the price of medium white or of brown sugar and that for white sugar which was
52
The function of exchanges PA
E*
Figure 7
the epicentre of these price changes? Let us assume that excess demand for medium white and for brown sugar changes as in Table 5', in order to make possible a comparison of the rates of price increase. In this table, /3 is a parameter which shows that proportional changes have taken place not only in /?3, /?4, but also in pu p2. That is, if p increases, then/?!, /?2, /?3, /?4 will all have simultaneously increased proportionally, and a decrease in j8 shows a proportional fall in each of them. When these four prices change in proportion no substitution will occur amongst the four kinds of sugar because the price of any one of them will not be comparatively cheaper or more expensive compared with the others. But because these sugars will be comparatively cheaper or more expensive compared with other goods, the excess demand for medium white and brown sugars will probably be affected. The last column of Table 5' shows that where the four prices have increased in proportion both medium white and brown sugars will be more expensive when compared with goods other than white, yellow, medium white and brown sugars, and as a result the excess demand for them E3, EA will decrease. Given these interrelationships of excess demand, we can prove that
Price repercussions between different exchanges
53
Table 5'
Rise in P3
PA
Effect on EA
Hicks' Third Law of Price Repercussions does hold for medium white and brown sugar: An increase in the demand for white sugar gives rise to an increase in the prices of both medium white and brown sugar at rates lower than the rate of increase in the price of white sugar itself. (We are not herewith involved in the proof of this proposition, which is left to the reader.) It can also be shown that an increase in the demand for white sugar raises the prices of both medium white and brown sugars less if trading of these kinds of sugar takes place after the session for white and yellow sugars has ended than they will if the four are traded simultaneously. 4 Price repercussions between different exchanges
Arbitrage dealings and equalization of prices As we have seen above, an increase in the demand for white sugar not only pushes up the price for yellow sugar which is competitively traded for at the same time as white sugar; after completion of competitive trading in white sugar it also operates to push up the prices of other commodities traded on that exchange (medium white sugar, brown sugar, etc.). Thus, throughout the goods traded in the same exchange there would prevail a 'co-variational' relationship among their prices. That is, all prices determined in a single exchange tend to fluctuate in the same direction; either they all rise together or fall together. What then are the price repercussions which are transmitted between different exchanges? In the actual economy there are many markets (sugar, coffee, cocoa, silk yarn, cotton yarn, wool and other exchanges etc.), and there may not be just one market in each economy for each commodity: several can exist (like the Tokyo Sugar Exchange, the Osaka Sugar Exchange, etc.). In the above analysis of price effects within the same exchange we tacitly suppose that the exchange was cut off from the outside and turned into a secret chamber; that is, there was noflowof information between exchanges and events which occurred in
54
The function of exchanges
one exchange or were in the process of occurring were never made known to other exchanges. But in reality the communications between exchanges are very close, and since dealers have information sent to them all the time by telephone from all parts, price repercussions are not confined within a single exchange. Where transmission of information is immediate there is unlikely to be any difference in competitive trading simultaneously taking place whether it is happening within the same exchange or in a different exchange. We shall assume in what follows that the transmission of information is perfect (information is transmitted costlessly, immediately and accurately), and will analyse what kind of price repercussions occur between various goods in competitive trading which simultaneously takes place in different exchanges. Let us first consider the case where competitive trading takes place for the same goods at the same time on two exchanges (for example, the sort of situation where competitive trading goes on at the same time for white and yellow sugar in both Tokyo and Osaka). As we have already seen, if the demand for white sugar increases in the Tokyo Exchange the prices of white and yellow sugar will increase in Tokyo. But there will be no change in the Osaka prices if this price information is not transmitted there. Yet if the exchange of information between the two exchanges is perfect and the Tokyo prices are communicated minute by minute to Osaka and vice versa, a large flow of demand and supply will occur between Tokyo and Osaka. That is to say, if it is known that the Tokyo price for white sugar is higher than the Osaka price, suppliers in Osaka will not sell there but will try to sell in Tokyo, and buyers in Tokyo will probably stop buying in Tokyo and try to buy in Osaka. (We shall assume the cost of transporting white sugar between Tokyo and Osaka can be ignored.) Nor is this all. Arbitrage dealing will occur - that is transactions for profit which result from the margin between buying cheap in Osaka and selling dear in Tokyo.8 Thus when Osaka supply flows to Tokyo and Tokyo demand flows to Osaka, excess supply will be created in Tokyo and an excess demand in Osaka, the Tokyo price will fall, and the Osaka price will rise. If, as a result of such price fluctuations, the price of white sugar becomes higher in Osaka than Tokyo, the flow of supply and demand will reverse direction. Supply will increase in Osaka and demand in Tokyo; the price in Osaka will fall and that in Tokyo rise. Such price 8
Earlier works concerning the problem of arbitrage includes Cournot, A., Research into the Mathematical Principles of the Theory of Wealth, New York: Macmillan, 1887, Chapter 3.
Price repercussions between different exchanges
55
fluctuations will continue until the price in Tokyo and Osaka are balanced. That is, where Tokyo information can be obtained without cost in Osaka (or the converse) and where transportation costs for sugar between Tokyo and Osaka are zero, prices in Tokyo and Osaka will be absolutely equal. Common national prices will thus be formed, and such prices will also, to a certain extent, be effective outside the exchanges. It is indeed true that the exchange of information between exchanges and the outside will not be as close and rapid as it was between exchanges. However, those who transact in rather large quantities even though outside the exchange will be bound to take account of the price within it. Thus if the price on the exchange is cheaper than that outside it, buyers outside the exchange may well try to buy sugar on the exchange rather than buy outside. Conversely if the price on the exchange is higher than that outside, suppliers outside the exchange will try to sell within the exchange. Therefore, the price formed at the exchange will not only be binding upon trading within the exchange but also regulate large-scale trading outside it. The formation of prices which have such broad controlling power is a result of the cheapening of transportation costs and of the costs of obtaining information. Before the nineteenth century, the exchange of information and means of large-scale transportation were undeveloped, and markets in separate areas were little interested in each other because the costs of obtaining information were so high. Even if traders paid a lot to obtain information and were aware of price differentials, transportation costs were too high to realize a profit from such price differences. The result was that regional markets were isolated. However, by the twentieth century, especially after the Second World War, both communications and transportation technology made great advances. Telegrams, telephones, radio, television and computers were developed, reliable information became obtainable at low cost, and along with this automobile highways were built, container transportation spread, and the efficiency of truck transportation greatly improved. The area within which goods circulated expanded to a national and to a world-wide scale, and in response prices are equalized not only nationally but also internationally. Effects on exchanges for substitutive goods Let us next consider the effects of an increase in demand which has developed on an exchange for a certain commodity (say coffee) upon another exchange where its substitutes (say cocoa) are dealt with. If transmission of information between the exchanges is as rapid and accurate as it is within an exchange, even if the coffee exchange and
56
The function of exchanges
cocoa exchange are independent organizations, both commodities will be transacted just as if they were on one fictional, composite 'coffee and cocoa exchange'.9 All kinds of coffee and cocoa will be transacted for at this 'composite exchange', and all these goods will be mutually substitutable. Thus the analyses in section 3 of this chapter are applicable exactly as they are to this composite exchange too. That is, if the demand for Brazilian coffee increases its price will rise (Hicks' First Law). At the same time not only will the prices of other coffees rise, the price of cocoa will rise too (Second Law). Furthermore, the extent of the increase of these prices will not be as great as that for Brazilian coffee (Third Law). Effects on exchanges for complementary goods Where information is transmitted perfectly between two exchanges which deal in complementary goods (for example, coffee and sugar), the two exchanges can be regarded just as if they were two sectors of one composite exchange, as was the case with the substitutes, coffee and cocoa. Whilst these two exchanges can be said to be organizationally independent they are functionally integrated. In normal cases, transactions for similar commodities take place on the same exchange and so all of the commodities handled are substitutive goods, but exchanges also exist which handle complementary goods, such as the New York Coffee and Sugar Exchange. Therefore, though we proceed with our analysis by means of the idea of a fictional, composite coffee and sugar exchange, we cannot be criticized for divorcing ourselves altogether from reality. Let coffee now be commodity 1 and sugar commodity 2, their excess demands be Eu E2, and their prices be pu p2. \ip\ rises, demand for coffee will decline (supply will increase) and because the demand for its complement sugar will also probably decrease in response, it can be assumed that the rise in/?! will bring about a decrease in Ex and E2. With px on the horizontal axis and p2 on the vertical axis, we can draw the partial equilibrium curves for coffee and sugar on the plane with these two axes. The curves will decline to the right as shown in Fig. 8. If we now move horizontally to the right from point a on the partial equilibrium curve for coffee and reach point ft, for example, the volume of excess demand for coffee will be negative because the price of coffee is rising. To return it to zero we must move perpendicularly down from point b to point c, for example. This is because the excess demand for 9
As an outstanding concrete example of a composite exchange we have the New York Commodity Exchange Center. This incorporates the New York Mercantile Exchange, the New York Coffee and Sugar Exchange, the New York Cotton Exchange and COMEX (Commodity Exchange Inc.).
Price repercussions between different exchanges
0
57
Pi
Figure 8
P7
0
Pi
Figure 9
Pi
Figure 10
coffee will increase when the price of sugar falls. (By the same reasoning, the partial equilibrium curve for sugar will also fall downwards to the right.) When we draw the two partial equilibrium curves on the same figure
58
The function of exchanges
we obtain Fig. 9, but the partial equilibrium curve for coffee is not necessarily the more steeply inclined as in the figure, so it is also possible for the partial equilibrium curve for sugar to be steeper as in Fig. 10. Each partial equilibrium curve divides the price plane into two. On the side nearer to the origin excess demand for the good is positive, whilst on the side furthest from the origin it is negative. By the rules of the market if excess demand for a commodity is positive its price will rise, and if it is negative the price will fall. Therefore we can draw small arrows which show the direction of price changes on the price plane once we are given partial equilibrium curves. Thus when we join up these arrows we may obtain a flow diagram of price movements. As we soon understand if we take a little care to confirm it, where the slope of the partial equilibrium curve for coffee (commodity 1) is steeper than that for sugar (commodity 2) (as in Fig. 11), the flow will be towards the intersection of the curves (the general equilibrium price set p°) and p° will be stable, but where the slopes of the curves are in the converse relationship (as in Fig. 12), clearly prices will ultimately be pushed further away from/? 0, as long as they do not drift along on one of the special currents (a or a1). That is to say, p° is unstable (more precisely, p° is a saddle point). Where general equilibrium is unstable it will not be possible to establish general equilibrium prices by competitive trading unless by great good fortune the prices are initially set at a point on a or a'. Prices will never reach the general equilibrium point and continue to fluctuate limitlessly, the teller will eventually have to declare 'no trading' and the session must halt. Therefore, whilst we shall assume below that equilibrium is always stable, we must recall that, as in Fig. 12, complementary goods can be a cause of instability. If demand for coffee increases its partial equilibrium curve will move to the right. Since we are assuming that equilibrium is stable, the slope of the partial equilibrium curve for coffee will be steeper than that for sugar. Therefore, as shown in Fig. 13, the new general equilibrium price p', after an increase in demand, will be located to the right and below p° which was the price set before the increase. That is, the increase in the demand for coffee will increase the price of coffee and push down the price of its complementary commodity, sugar. Substitutes for complements are complements Such results will be obtained however many kinds of commodities are transacted at the two exchanges. If for example, Arabica coffee and Robusta coffee are traded at the coffee exchange, and white, yellow and brown sugars at the sugar exchange, an increase in the demand for Arabica coffee will force up the prices of both Arabica and Robusta
59
Price repercussions between different exchanges
Pi
0
Pi
Figure 11
P2
P\ Figure 12
60
The function of exchanges
0
Pi
Figure 13
coffee, and force down the prices of white, yellow and brown sugars (given that the general equilibrium for coffee and sugar is stable). This case can be further extended to where there are more than two exchanges. Let us now suppose that there are four exchanges for coffee, cocoa, sugar and honey and that information is rapidly and accurately transmitted amongst them. With such perfect information there is no need to consider the exchanges as functionally independent organizations, and there is also no obstacle to proceeding with the analysis as if there were a composite coffee and cocoa exchange and a composite sugar and honey exchange. Only substitutes are transacted at the respective composite exchanges and relations between the two exchanges are complementary. That is, the relation between one commodity on the integrated coffee and cocoa exchange (say, Arabica coffee) and one commodity on the integrated sugar and honey exchange (say, white sugar) is one of complementarity; but in addition the relationship between any other good in the coffee and cocoa exchange which is a substitute for Arabica coffee (say, cocoa) and white sugar is also one of complementarity. That is, a substitute for Arabica coffee is a
Futures markets
61
complement of white sugar, which is itself a complement of Arabica coffee. Similarly a complement (Arabica coffee) of white sugar, which latter is a substitute for the No. 1 honey which is traded in the sugar and honey exchange, is a complement of No. 1 honey. The mesh of transactions which knit together the composite coffee and cocoa and the composite sugar and honey exchanges in this way satisfies the relationships that substitutes of complements are complements, and the complements of substitutes are complements.10 As long as such relations are satisfied, there is no difference in relations between the two integrated exchanges are compared with relations between the single coffee exchange and the single sugar exchange. That is, an increase in the demand for the Arabica variety of coffee forces up the prices of all kinds of coffee and cocoa, and forces down those of sugar and honey. The law that the prices of substitute goods move in the same direction and those of complementary goods in opposite directions is thus not only correct within exchanges but also between them, given perfect circulation of information and the stability of the general equilibrium of coffee, cocoa, sugar and honey. 5 Futures markets Sham buying and speculation We implicitly assumed above that transactions are made for actual goods. That is to say, within a short period after the completion of competitive trading, for example within five days, the contracted transaction must actually be implemented. Consequently, the seller must actually be in possession of the goods and the buyer must also have sufficient money to pay for them. Or else the seller must make preparations to hand over the goods within a few days, and the buyer must dash about raising the money. Then if they fail in this the transaction will go by default, and the defaulter will be forced to incur painful sanctions. In actual exchanges, however, transactions take place in futures goods in addition to actual goods, and the former predominate in the 10
This kind of system satisfies the following relationships too: (i) 'substitutes of substitutes are substitutes'; (ii) 'complements of complements are substitutes'. For example, cocoa, which is a substitute for Robusta coffee whose substitute is Arabica coffee, is itself a substitute for Arabica coffee; and honey, which is a complement of cocoa whose complement is white sugar, is a substitute for white sugar. On the stability of such a system and the laws of price changes under it see, for example, Morishima, M., 'On the Laws of Change of the Price-System in an Economy which Contains Complementary Commodities', Osaka Economic Papers, Vol. 1,1952, and Morishima, M., 'A Generalization of the Gross Substitute System', The Review of Economic Studies, Vol. XXXVII (2), April 1970.
62
The function of exchanges
break-down of total transactions. Futures transactions are those where it is unnecessary to deliver the goods immediately on contracting a deal (or to do so within a few days). Instead they may be delivered and accounts settled on a fixed day after quite a margin of time (at the end of the month or of next month, or six months later). The time limit by which futures must be delivered and the deal implemented is known as the 'terminal month' or the 'delivery month'. Normally at the exchange futures transactions take place in parallel for terminal months of different lengths, but there are limits to the length of period for transactions in futures (these differ according to country, exchange and commodity, short terms normally being of three months and long terms being up to 24 months), and the number of sessions for transactions in futures in a given day is determined in relation to the length of term. For purposes of simplifying the explanation of trading in futures let us assume the longest terminal month is three months. We call transactions where delivery must be made at the end of the current month 'current month delivery', or 'month end delivery (options)' and those which must be accepted at the end of the next month, or at the end of the month after 'intermediate delivery' and 'future delivery' respectively. Sessions for transactions in the three types of futures take place for the commodities (for example, wheat) on the exchange. Since there is a gap in time between contract and delivery there is no need for the seller to hold the actual goods (actuals), nor for the buyer to have the cash for them. Where futures transactions have a terminal month which is a long way ahead (for example, transactions in 12-months' wheat), no sellers are likely to have the actual goods. Therefore much sham takes place in futures without the actual goods to back the transactions. 11 However, where there is false selling (let us suppose an individual A sham-sells commodity M on the intermediate delivery market), A does not have the actual goods (actuals or physicals) to deliver to the buyer of M at the next month end. Therefore, there would be default on the contract so that A must in the period from the day the contract was made until the end of the next month, repurchase M, at least by the same volume as the sham sale either this month on the intermediate delivery market or next month on the current month delivery market. However, a buyer like A does not purchase and take deliveries of the commodity M with a view to using or consuming it. A will re-sell M by the terminal month. (In fact, in the present example A has already re-sold before purchasing.) This is the way 'sham' transactions occur in the futures market, but 11
See Additional Note d.
Futures markets
63
since 'sham' transactions will all be cancelled by the corresponding repurchase or resale on the day when accounts are settled what remains after cancellation is the buying and selling of actuals (physicals), and therefore the delivery of actual goods will take place without let or hindrance. People who engaged in 'sham' trading on the futures market will have no need to pay the whole of the purchase price, needing only to pay the difference between the buying and selling price. That is to say, transactions in futures are dealings in margins or differences. Those who bought high and sold low will have made a loss, conversely profits may be obtained by 'sham' trading. Consequently, able people who are perfectly capable of figuring out future price changes are certain to make a profit in the futures market, so long as prices continue to fluctuate. Moreover almost no capital is required to make such a profit. Therefore those confident that they can forecast the movement of futures prices gather in the market for futures which is the site for their speculative activities. They do not come to the market with a view to commodities but with a view to prices. However, because they need to know about the situation as regards production, consumption, the environment (movements in domestic and international politics, the weather, etc.), the activities of large speculators and so on in order to formulate accurate forecasts, all kinds of information gather in futures markets, and the exchange becomes a kind of information centre. Those who deduce correct forecasts from this information make a profit, and those who err in forecasting make a loss. Hedging Depending on the exchange and the commodity, sham trading in futures takes place every day in many times more than the actual quantity of goods delivered, sometimes several hundred times more. Yet those who activate the futures market include not only the speculators prepared to take the risk and who are dreaming of a quick killing. In addition to them, producers, wholesalers, importers, etc. appear in the market as sellers or buyers and trade for the purpose of 'hedging' to reduce risk even though it is impossible to avoid it completely. Let us consider the example of wheat. Wheat (winter wheat) is sown in late autumn or early winter and harvested in early summer of the following year. It goes without saying that the size of the harvest will depend on the weather. There will be a good crop if it is favoured by the weather, and bad weather will produce a poor crop. Thus with a good crop wheat will be cheaper, and when there is a poor crop the price will rise. Farmers will thus be haunted by violent fluctuations in income according to their luck, but up to a point they can stabilize their income
64
The function of exchanges
by using the futures market. For example, if they sell about half the average annual crop on the futures market some months before harvest they have no need to fear a collapse in price below the present selling price for the part of the harvest already sold. If the harvest is good they will only have to sell off cheaply what remains from the harvest. That is, farmers will have been able to halve the damage from the crisis caused by a good harvest by using the futures market. Similarly, if wheat importers sell part or all of what they import on the futures market, they will be able partially or wholly to avoid the damage of a fall in prices which might arise from an excess supply when imports actually arrive on the market. Conversely, however, where prices rise, farmers' incomes will be less because they will not be able to enjoy the benefit from the price increase in respect of that part of the harvest they have already sold. Thus if they sell off on the futures market they lose tfie chance of a large income, but in return they reduce the risk of having to be satisfied with a very small income. That is, futures markets serve the purpose of stabilizing farmers' incomes. Role of futures markets In the exchanges, for each commodity handled, there are, as a rule, a spot market and futures markets with different terminal months. Because today's wheat is substitutable for wheat at the end of the month or the end of the next month or the month after that, there is a relation between the futures prices and the spot price, and the former cannot diverge too far from the latter. Thus the spot price will rise when the demand for spot goods now increases, and the futures price will probably rise along with it. It is moreover unlikely that the futures prices will be boosted out of balance. Let the current delivery price of a certain commodity at a certain point of time t be pu the intermediate delivery price be /?2? a t a given time t' later in the same month let them be respectively p[ and/?2. If it is expected that the intermediate delivery price in terms of the current delivery price at t is higher than the corresponding price at t', i.e., if the inequality
P\
Pi
is expected to hold, then people will sell s2 of the high priced intermediate delivery and buy dx of the cheap current delivery; then later when a cheaper relative price is established they buy back s2 of the
Futures markets
65
intermediate delivery (d2 = s2) and re-sell dx from the current delivery (s[ = dx). It is obvious that from such transactions a surplus of m = [p[s[ ~pidx] + \p2s2-p'2d'2] will be obtained. If we take du s2 such that pxdx = p2s2, then m=p[s[-p'2d'2=P-±pxs[-P-±p2d'2 P\ Pi But since dx = s[, and s2 = d2, we have pxs[ =p2d2 so that
1
Pi
Therefore when the above inequality for relative prices is established the surplus m is positive. That is, sham trading on the current and intermediate delivery markets, together with the re-selling and buying back at a later date which cancels it out, will produce a profit. Such combinations of transactions are called arbitrage over time, and since intermediate delivery goods will be sold due to arbitrage, p2 will fall; px will rise because current delivery goods will be bought due to arbitrage. As a result the relative price ratio of futures will rapidly equalize, and soon proper relative prices will appear. Not only will this be so but changes in the futures prices px, p2, are also likely to be appropriate ones. As we have already said, there are many speculators active in the futures market. If they foresee that in future px or p2 will suddenly rise, they will buy in before px, p2, actually begin to rise. Then, since they will re-sell what they have already bought in order to profit from the difference in the prices, when the price at last does begin to rise violently, the price will in fact not increase as much as had been forecast. Similarly, where a violent fall in prices is forecast, the speculators will soon begin to sell out. But because they will buy back to profit from the price margin when the price at last does begin to fall violently, the price fall will be moderated through their support. With commodities whose supply fluctuates seasonally like agricultural products, prices will fall in the harvest period and rise out of season; but when a futures market is established the price equalization function of speculative dealings and hedging will start to work and very much reduce the extent of rises and falls in the prices of agricultural products. Futures markets have the internal capacity automatically to normalize prices in this fashion, but besides this exchanges additionally set an
66
The function of exchanges
Pc -
Pa
upper and lower limit to fluctuations in pu p2, and thus to prevent the overheating or overcooling of the markets.12 Lastly, futures markets have the capacity to put into circulation a larger volume of actual goods (actuals) than spot markets. Let us assume that the demand and supply curves for actual goods are represented by DD' and 55' respectively, as in Fig. 14. Where speculation and hedging are not allowed and the demand and supply for the actuals directly balance each other on the market, price will take on a value (for example p°) between the two end points, S and D of the curves, and both demand and supply will be zero, as the figure shows. However, in the futures market where speculation and hedging may occur, the demand for actual goods qa and what speculators sell may correspond at price pa. If that is the case, speculators must later buy back what they sham-sold and therefore they will once more appear in the futures market and look for sellers. New sellers will also appear later on than this in the market in order to repurchase. They will buy back 12
In futures markets it is not permitted to fix prices during that day which exceed the upper or the lower limit.
Futures markets
67
that part qb at price pb from suppliers of actual goods, buy back the remainder qa - qb at price pc from other suppliers of actual goods. At the end of the terminal month, qa of actual goods will be delivered from the suppliers with actual goods to the buyers of actual goods. The two speculators who have intervened in the course of this process will buy back all that they sham-sold and must pay the difference. That is, through the intervention of the speculators and by their loss, a quantity qa of actual wheat has come into circulation. That is, in the futures market demand can meet supply smoothly because of speculators' intervention, which results in the circulation of a large volume of real goods.13 15
In this case speculators will suffer a loss, but in other cases they may make a profit. Even if exchange does not take place in a real-goods economy as a consequence of the fact that buyers of goods and sellers of goods confront each other who are neither of them prepared to make a loss, the circulation of goods will begin if there are amongst them speculators who are prepared sometimes to make a loss and sometimes a profit.
Fixing product prices
1 More on the full-cost principle The average cost curve As indicated earlier, the prices of most industrial products are not fixed in the market (exchange) so as to equalize supply and demand; they are set by the enterprise itself. In order to obtain a profit the enterprise will mark up average production costs to get the product price per unit, but the mark-up ratio has to be a reasonable one. We will discuss later what constitutes a reasonable rate; we will assume for the present that the entrepreneur already has a generally accepted idea of what it should be. Thus the problem of price-fixing rests entirely on how to estimate average costs. Costs are classified into fixed costs which do not increase even though the volume of output increases, and variable costs which increase and decrease with the volume of production. Variable costs may be further subdivided into proportional costs which vary in proportion to the volume of production and non-proportional costs which do not. The latter are of two kinds; those which vary less than proportionally, and those which vary more than proportionally. Fixed costs are interest and depreciation costs on fixed capital, rents and remuneration to employees not engaged directly in production directors, executives, technicians, watchmen, porters etc. However, most raw material costs and the wages of labour are proportional costs. Power costs, fuel costs and maintenance costs of buildings, machines and tools etc. are all non-proportional costs, and they vary less than proportionately while the volume of production is small; but should production reach such a high level that machines are inevitably driven to their limits then it is likely that they will vary more than proportionately with output. Various operating costs such as sales and purchasing costs will be unchanged while the volume of production is low, and will vary less than proportionally when it is high. Taxes and rights' fees for patents, copyright and such-like will depend on the tax system and on 68
More on the full-cost principle
69
rights' transfer contracts, or else may be invariable in some cases and variable in others. In the normal event wages, which are also proportional costs, may have to have an overtime allowance added to them where production is the result of labour working overtime. Thus as the proportion of output produced in overtime rises, wages rise more than in proportion with output. In any case, it goes without saying that fixed costs per unit product will decline with an increase in production, and that proportional costs per unit product will not change. As against this non-proportional costs per unit product will tend to decrease while the volume of production is low, but will begin to increase per unit if the volume of production is high and exceeds the capacity of the factory. Therefore, having added up all these different kinds of costs we find that average costs will decrease initially with an increase in output, but ultimately will increase. Even after average costs have started to rise, only a part of nonproportional costs will initially increase more than proportionally, but thereafter most non-proportional costs will gradually start to increase more than proportionally. If we measure output along the horizontal axis and average costs along the vertical axis, we will obtain a U-shaped curve. Fixed costs per unit product will decrease rapidly while the volume of production is low but rising; but when the product is produced in large quantities the variation in fixed costs will gradually get smaller. Therefore, in the case of products which are produced in large quantities and fixed costs are not large compared with proportional costs, changes in fixed costs per unit product due to changes in the volume of production can practically be ignored. On the other hand, average costs will begin to rise markedly when production is pushed beyond the capacity of the factory; and as long as production is not carried on at such an exceptionally high level average costs will be more or less unchanged with respect to changes in output. For example, where total fixed costs are £8000 and proportional costs per unit-product are £3.90, average costs will be £4.00 with an output of 80000 units. However, if output were 40000, average costs would be £4.10, and if output were 160000 units they would be £3.95, and so average costs will be practically unchanged. That is, the average cost curve for this factory will describe a U-shape with an almost horizontal bottom over a broad range of output from 40000 to 160000 units; and as long as the volume of production remains within this range average production costs may be easily determined, and thus the price of product marked up at the mark-up rate may also be easily fixed. They will be practically unchanged over a broad range whatever happens to the volume of output.
70
Fixing product prices
The estimated demand curve What happens where the volume of production is estimated to be extremely small? Let us consider the above arithmetical example as referring to the production of books. Paper and binding charges are proportional costs, editorial costs and typesetting charges are fixed costs. Printing costs will be non-proportional costs which increase less than proportionally, but we shall ignore them in our arithmetical example along with certain other costs such as advertising and sales costs. Now where the book has a very general appeal and estimated sales are more than 40000, average costs will be about £4.00, and mark-up at a rate of 25% may be fixed at a retail price of £5.00. But where the book is specialized and estimated sales are only 1000, average costs will be £8000/1000+ £3.90 = £11.90, and the retail price may be £14.88. The price of the book will thus depend first of all on whether it is of general appeal with a large expected estimated demand, or whether it is a specialist book with only a small expected demand. In the former case there will be almost no difference in the price even if expected demand changes, but in the latter case there will be a wide margin of difference in the price according to how demand is estimated. For example, where expected demand is only 500 books, average costs will be £19.90 and the retail price will be £24.88 (given that the mark-up rate in each case is 25%). That is to say, the price of commodities with a small output which does not reach the long flat bottom of their respective U-shaped average cost curves will depend a great deal on the size of expected demand. There is another situation where the size of the expected demand greatly influences the determination of the price. This is where fixed costs are very large compared to proportional costs. In the previous example, fixed costs were £8000 in total, and fixed costs per unit product were £0.04 where the factory produced to the limit of its productive capacity, which is 200000 units. This figure is extremely small in comparison with proportional costs of £3.90 per unit product, with the result that average production costs were virtually unchanged over a broad range of output from 40000 to 200000 units. However, if in the arithmetical example the fixed costs were £800000 (given that the productive capacity of the factory was still 200000 units as before), the fixed costs per unit product would be £4.00 at the capacity level of production and would thus be comparatively high compared with proportional costs of £3.90. So we would no longer be able to ignore changes in fixed costs per unit product which accompany changes in the volume of output, and we would only be able to regard average costs as fixed with regard to a volume of output in the range of from 185 000 to
More on the full-cost principle
71
200000 units. That is, if fixed costs are large compared with proportional costs the flat part of the U-shaped average cost curve will be shorter. Therefore, there is a high probability that expected demand will not fall within that range; and if it does not, average cost and so price will be elastic with respect to expected demand. How then will prices be fixed in these cases? That is, where products have a very short horizontal section to their average cost curves; or, as with specialized academic books, there is a long horizontal section to their average cost curve but an extremely limited demand. In these cases the firm will estimate demand and read off its average costs from the cost curve which corresponds to a volume of output that equals the estimated demand. It will next add a mark-up at the mark-up rate and thus set the price of the product. But demand depends upon price, and so the price which is calculated on the basis of the expected demand must be completely consistent with the price the firm had in mind when it made its estimate of likely demand. We can easily confirm whether the price the firm sets is consistent or not by drawing in the expected demand curve on the diagram of the average cost curve.1 Figure 15 shows a part of the U-shaped curve which includes a mark-up m on average costs c(x) (that is (1 + m)c(x)), and an expected demand curve DD' which slopes downwards to the right. That is, when the price is px the firm can expect a demand of xu and if it produces only xx it can sell its product at price p2. But if the price is p2, the enterprise can expect a demand of x2, and can sell output x2 at a price lower than p2. At both the points xx and x2 the price the firm had in mind when it estimated the demand for its product and the price which it sets on the basis of that expected demand are not the same. Therefore the prices the firm sets at these points are not consistent prices. Only at the point where the estimated demand curve intersects the curve of supply price (1 4- m)c(x) - points x and x' in the figure - will the prices set by the firm be consistent. As the figure shows two consistent prices normally exist, but where the curve of expected demand is twisted more than three can exist, and indeed an unlimited number can exist in extreme cases. In these cases 1
The expected demand XA of firm A will depend not only on the price p of the product, but also on the expected demand XB for the competitor B. That is, XA = fA(p, XB). Similarly, XB =fB(p, XA), and the curve of expected or estimated demand in Fig. 15 is obtained from the two. Thus, or p = hA(XA) Not only isfA a subjective estimate made by A , / f l is also A's subjective estimate of what B will probably estimate. At the same time B too may well derive p = hB(XB) from his own subjectively estimated curves. A similar situation arises when there are more than two competitors.
72
Fixing product prices P
Pi P2
0
x
xx
x2
x
D
x
Figure 15
which price is selected will depend completely on the policy the firm follows and one probable criterion for the selection is the size of the profit anticipated. Since it can be shown that the larger is the volume of output the larger will be the profit,2 and if the points x, xf, x", etc. are points where consistent prices can be set, then profit will be at a maximum at the point where the x is largest from among these JC'S. In the case of Fig. 15, the profit at x is larger than that at x'. The enterprise will set a price p for its product and anticipate a demand of x? 2 The period of production and hedging
Re-assessment of costs of production Let us divide goods into three groups. The first group consists of goods whose prices are set according to the full-cost principle; the second 2 3
Profits will be mc(x)x when the volume of output is x. Since total costs c(x)x will increase when x increases, the larger JC is, the greater the profits will be. Where the actual volume of sales does not accord with the expected one an enterprise can go without altering the price of its products, by either foregoing the surplus demand or carrying over the excess supply to the future. However an enterprise must as far as possible avoid this kind of situation. An enterprise must therefore set the price of its products so as to have it as consistent as possible. Furthermore, as maintained by the neoclassical school, even where an enterprise has fixed its margin so as to maximize profit, if other enterprises are selling their products at a lower margin, that enterprise will have no choice but to change to prices with a lower mark-up rate. In this way the mark-up rate is pushed down to where prices are at the minimum. See the discussion later (p. 88 ff.) on price competition.
The period of production and hedging
73
consists of goods where prices are set by competitive trading, and the third group consists of labour and land. The determination of prices of goods in Group II was considered in detail in Chapters 1 and 2. We will explain later how prices for goods in Group III, wages, are set, while land problems and therefore, rents, are ignored throughout the book. In what follows here we shall go into some detail concerning the way prices of goods in Group I have to be determined, assuming the prices of goods in Groups II and III are given. The prices of goods in Group I will be set according to their respective average costs of production, but they are often themselves used as producer goods or in the production of other Group I goods, as for example with paint and electric cables etc. Consequently, a part of the production of goods in Group I will depend on the prices of some other goods in the same group. That is,firmstake the current prices of Group I goods as given, evaluate their respective average costs, mark up the average costs so decided by a mark-up rate, and thusfixthe prices of the various goods in Group I. However, prices set in this way are not necessarily equal to the prices of Group I goods which were used at the time when average costs were assessed. Therefore, the assessment of average costs must be corrected using the new prices. Thus as a result of this re-assessment firms will have to revise their production prices. Revision of prices brings about a further re-assessment of costs, and this latter produces a further revision of prices. Such 'iteration' may well continue until prices calculated from marked-up average costs are equal to the prices used for the assessment of average costs - that is, until a consistent price system is obtained. Once such a state is reached 'iteration' will cease; further revisions of prices will not take place and therefore the consistent production prices arrived at will be equilibrium production prices. The speed with which equilibrium prices are established rests on the degree of sensitivity with which firms react. As long as the firm is rational it will always assess its costs on the basis of current prices, and must always revise its production prices in response to changes in these; but since it requires extraordinary effort constantly to carry out these 'iterative calculation operations', firms often tend to be negligent or remiss in doing them.4 Thus current production prices are not necessarily 'consistent prices', and will merely be 'interim prices' which have 4
Price revision tends to be slow in socialist societies in particular. They are afflicted by the long sequence of time-consuming bureaucratic procedures for price revisions and by the bureaucratic way of thinking which considers that to revise prices for any reason whatever is to confirm that an error has been made by the authorities. The planning bureaucrats thus prevent achievement of equilibrium prices (that is, the establishment of appropriate and fair prices).
74
Fixing product prices
stopped short somewhere along the endless chain of repeated iterative calculation needed to reach equilibrium prices. Yet if we were to assume that firms are enthusiastic enough in re-evaluating their costs, then we should have current prices which are quite close to equilibrium production prices, even though they are only 'interim' prices. Cost assessment and the current prices assumption The above explanation is implicitly premised on the fact that each cost item is assessed at its price prevailing at the time when firms fix their product prices. However, since in reality time is required to produce things, the input of producer goods into the production process must be made before products are completed and put on sale. The prices required to obtain producer goods of this sort are not the same as the prices of producer goods at the time when product prices are determined. Therefore, costs which are assessed on the basis of current prices will differ from actual costs incurred in the past for the production of the relevant goods. Such a way of assessing costs would seem to be irrational at first sight but it can be rationalized if we reason as follows. Firms have stocks of producer goods, and at time t0 they will withdraw them from stock and inject them into the production process. However, until the final product is completed we can regard these producer goods as remaining in the factory, even though they have changed their form. Only at time ti when the finished product is completed and sold will these producer goods leave the factory. At that point, that portion of the stock of producer goods which leaves the factory must alone be replenished. Therefore, we can think of the unit price p(tx) of the producer goods needed to replenish stocks as the unit price of producer goods which leave the factory. That is, producer goods which were used up in producing final products which are sold at time tx are evaluated at current prices p{t^) prevailing at time tx. Even upon such a basis, there is no change in the fact that the average cost c(p(^)) calculated in terms of the current cost approach will not be the same as the average cost c(p(t0)) which was evaluated at prices p(t0) which were actually paid to acquire the producer goods used up in producing the final product. Ifpih) is larger than/?(f0), the current price approach overestimates costs, and in the converse case it underestimates them. However, these over-or-under estimations can to some extent be avoided, as we will explain below, if these producer goods are covered (hedge-sold) in the futures market at the time when they were acquired. Let us assume that a units of producer good 1 are required to produce
The period of production and hedging
75
one unit of the final product. According to the present or current price approach average costs will be
and according to the historical costs approach average costs will be
Here px(t) is the price of producer goods 1 at time t, and c(l) and c(0) are average costs evaluated according to the current price and historical cost approaches in relation to cost items other than the input of producer good 1. If, at the same time as injecting a into the production process at time t0, the firm false-sells an amount a of producer good 1 in the futures market with time tx as the term date, at price qi(t0), and buys back a at time tx, the enterprise will obtain precisely
as trading income from such hedging. Average costs determined by the historical approach are true costs when hedging of this sort does not take place; but where producer goods have been hedge-sold in the futures market, true costs are those where the trading income from hedging has been subtracted from average costs determined according to the historical approach. That is Pl(t0)a
+ c(0) - [qx(h)a - px{h)a\
(1)
Therefore, if we assume that /?i(f0) which is the price of good 1 at time t0 and qi(t0) which is the price of futures are equal, then true costs will be Pi(h) a + £(0). That is, with regard to producer good 1 where hedging took place it is the current price approach which gives the true cost. Therefore the average cost c(p(ti)) determined according to the current price approach is the true average cost where hedging has taken place for all producer goods. Hedging and risk avoidance The reason the firm hedge-sells in this way in the futures market at the same time as it purchases its producer goods, is in order to avoid the danger of fluctuations (falls) in futures prices. Let us now assume there are two firms A and B producing the same good, and that B hedge-sold the producer good 1 it acquired simultaneously with their purchase, while firm A did not. We assume that after the purchase of producer good 1 its price begins to fall, so that at the time where the product is completed, the price of good 1 is lower than it was at the time when the producer good 1 was bought in, i.e. P\{t\)
76
Fixing product prices
hedge-sold the producer goods it has a trading income from this, and so it can offset this against that part of costs which have risen because it bought its producer goods at a high price at time t0. Therefore it is able to estimate lower costs by the current price approach. However, enterprise A which has not hedge-sold must claim historical costs which are high, and therefore enterprise A will be at a competitive disadvantage compared with enterprise B. In the converse case, that is where the price of producer good 1 which was hedge-sold rises between time t0 and time tu firm B will bear a trading loss due to its hedging. Firm B will therefore assess its costs so as to cover this loss and will be at a competitive disadvantage compared with firm A which is not faced with this necessity. However, in this case it will be impossible for firm B to stand idly by and watch the price of producer good 1 rising continuously between times t0 and tx without repurchasing the producer goods 1 it has hedge-sold. If it buys back at an early stage in the price rise its trading loss due to hedging will not be a large one and therefore firm B will not be at too great a competitive disadvantage compared with firm A. A firm which hedge-sells its producer goods at the time it purchases them will clearly be in an advantageous position when their prices fall and will not stand at too great a disadvantage when they rise. Thus a rational and sensible firm is always likely to hedge-sell in the futures market the producer goods it has on hand. Therefore, when the prices of producer goods fall, firms will not calculate costs according to the high purchase prices of producer goods, but at their lower current prices. Where prices rise firms will probably buy back the producer goods they hedge-sold at a point in time (for example t') before tu true costs (1) which includes the trading loss from hedging will be Pl(t0)a
+ c(0) - [qi(t0)a -Pl{t')a\
=Pl(t')a + c(0)
(1')
and costs will not rise as much as they would have under the pure current value approach. 5 That is, true costs will be elastic when prices fall but not all that elastic when the prices rise. Despite this asymmetry when the current price p\(ti) is higher than pi(t0), P\{t') will also be higher t h a n p ^ o ) ; thus/?!(/') will probably move in the same direction as P\{tx) moves. In this sense we can state that costs will ultimately be regulated by current prices even in the case where the prices of producer goods rise. Note that in (1 ; ) we assume px{tQ) = qx(Q for the sake of simplicity.
The period of production and hedging
77
Where there is no futures market Yet there are many producer goods which have no futures market. To carry through the current value method of costing for such producer goods a department must be set up within the enterprise to accumulate the losses and profits which accompany falls and increases in the prices of producer goods, and that department must be accounted for separately from the production department. Let us call this department the 'Department in charge of hedge-selling (or covering department)', and where there is no futures market for producer goods 1, let them be hedge-sold to the 'covering department' within the firm. If we imagine that the 'covering department' buys up producer goods 1 purchased by the production department at the purchase price Pi(^o), and sells them back to the production department at the current value P\(tx), the costs of the production department taking into consideration the trading profit (or loss) between it and the 'covering department' will be Pl(t0)a
+ c(0) - \px(t»)a - Pl{tx)a\
(1")
The trading income of the 'covering department' on the other hand will be
If the production department thus hedge-sells producer goods to the 'covering department', costs in relation to those produced goods will come to be assessed entirely according to the current value approach. As the reader will quickly realize* this is because (1") above is equal to px(tx)a + c(0).
On the other hand, however, the 'covering department' will have to accumulate trading losses and profits. When prices fall (when P\(tx)
78
Fixing product prices
3 Equilibrium production prices and their movement
Adjustment of estimated volume of output As we stated earlier, firms which calculate average costs by the current value method require 'iterative calculation' in order to be able to obtain a consistent production price system where they each use each others' products as producer goods. Not only is this so, but since average costs also depend on how they estimate the volume of output, and since this estimate depends on the volume of demand and hence the price of the firms' products, not only price but also the estimated volume of output will be adjusted in the process of 'iterative calculation'. Let us imagine that average costs have increased from c(x) to c(x) because of a rise in wages or else the price of producer goods. The curve of average costs plus the mark-up rate moves upward as shown in Fig. 16. As long as the estimated volume of output remains at Jc, the price
Figure 16
fixed by the full cost principle must be p*. However, such a price will most likely be incompatible with the conjectured, estimated production volume Jc. At p* which is higher than/?, a demand of Jc cannot any longer be anticipated, and the firm must estimate for a smaller volume of demand and hence, of output. As long as output does not exceed the firm's production capacity (that is, production takes place on the descending part of the U-shaped average cost curve), the contraction in the estimated volume of output will bring about an increase in average costs and result in a further rise in the product price. The increase from
Equilibrium production prices and their movement
79
p to p* in the figure is the rise in the product price due to increases in the price of producer goods, and the additional rise from/?* to/?' shows the effect produced by adjusting the estimated output. Thus, a rise in the price of producer goods will induce an increase in the product price, and at the same time it will produce a decrease in the firm's estimated output. As a result a higher price will be fixed for the firm's products because the fixed cost burden per unit product will increase. Profitability ofproduction But where these products are used as producer goods (inputs) by other firms, price changes will not be confined to those described above. Because the price of these products has increased, the average cost curve of other enterprises rises, and their product prices rise whilst their estimated output will contract. As a result of the passing on of price movements from one firm to another, the outcome is that the price of producer goods required by the first firm will again rise, and prices may well be more than / / , with the estimated output less than x'. Such additional price changes will also affect other firms. A spiralling upward movement in product prices and a spiralling downward movement in estimated outputs will thus be produced, but such a vicious cycle must cease at some point. This eventual state of affairs is in fact one where a 'consistent system of production prices' or 'equilibrium production prices' are established in the manner described earlier. The repercussions of product price changes can be analysed in exactly the same way as the repercussions of competitive trading prices. For the sake of simplicity we shall consider a situation where changes in the estimated volume of output have barely any effect on product prices; that is, the case where estimated output lies on the horizontal part of the U-shaped average cost curve. In order to simplify matters further, let us assume that there are only two kinds of firm producing products 1 and 2, and that firms of the same kind all have identical average cost functions. Let us also assume that these products are used by all firms as producer goods. Now, let the average cost functions of the firms which produce product 1 and 2 be respectively pxan + p2a21 + Vx\ Pia12+p2a22 + V2
where Vx, V2, show the fixed costs and wage costs etc. per unit product. The coefficients a^s are the production coefficient, so that a2l for example indicates the amount of product 2 required to produce one unit of product 1. In the same way a12 is the amount of product 1 required to produce one unit of product 2. The equations for equilibrium produc-
80
Fixing product prices
tion prices may be shown by Pi = (1 + m) [pxan +p2a2i + Vx] p2 = (1 + m) \pxan + p2a22 + V2] If we solve these equations algebraically for p u p2 given that Vu V2 have a positive value, at/s are positive or 0 and m is positive, we shall obtain equilibrium production prices. However, the solution so obtained will not necessarily have positive values as equilibrium prices are required to have. In order to obtain that kind of solution then the coefficients ai}- and m in the above equations do not merely have to be positive or 0; they must also satisfy more restrictive conditions which will be described below. Were those conditions to be satisfied the production coefficients atj could be said to be 'profitable' at the mark-up rate m. That is, this would be the case where a special combination of positive prices (/?*, /?*) would exist which would establish the inequalities p \ > (1 + m) \p\an + p*2a2l] p \ > (1 + m) [p\al2 + p\a22] This is the definition of the profitability of atfs at m. Were this p* = (p*, p2) to exist and were/?* and/?*> to be made proportionally large enough, the differences between the right-hand and the left-hand side of the inequalities would increase, and therefore for sufficiently large/?**, /?** the following inequalities would be established: Pr
> (1 + m) \pran
**(l
+
+ /?I*02i + Vx]
) \ * * +
**+V]
(4)
Here p \ * = kp\, p \ * = kp*2, and A: is a sufficiently large proportionality factor. If prices /?** are established, firms producing products 1 and 2 can all meet their costs and furthermore are able to receive a mark-up rate which is greater than m. Therefore firms will be 'profitable' at the rate m. Existence of equilibrium prices In Fig. 17, pi is taken on the horizontal axis and/? 2 on the vertical axis. (3) is established at point/?*, and (4) is established at point/?**. If we fix /?** i n the first equation of (4), and decrease /?**, a2i + Vl]
(5)
will be established at p[. p[ is the partial equilibrium price of product 1
Equilibrium production prices and their movement
81
r
0
b
px
Pi
Figure 17
when/?2 = p\*• This is greater than the partial equilibrium price/?" with p2 = 0,6 that is p'[ = (1 + m) [p'lciii + VJ
(6)
Let these points where partial equilibrium prices obtain be a = (p{, /?**) and b = (p'[, 0). As shown in Fig. 17, b must be located below and to the left of a. In the same way we may also obtain for product 2 the partial equilibrium points c = (p**, p2) when px =/?**, and d = (0, p2) when Px = 0. Since it can be shown that p2 >p2 > 0, d will be below and to the left of c. Since the partial equilibrium price curve of product 1 joins a and b, and that of product 2 joins c and d, both curves must intersect within the positive quadrant of the price plane, and their intersection will be at the point p° which is the point of general equilibrium of production prices. Thus if the coefficients at/s and m fulfil the conditions for profitability, positive equilibrium production prices will exist, and conversely whatever prices are taken, if (3) is not established then equilibrium prices will not exist. The production coefficients will be set by the state of technology but the mark-up rate is set by economic considerations. If firms demand very high rates they will be 'unprofitable' whatever prices 6
Since p*, p*2 > 0 , we may obtain from (3)
and therefore 1 > (1 + m)an. If we solve (5) and (6) with respect to p{, p'[, and consider the condition just obtained, we can immediately prove that p[>p'{>0.
82
Fixing product prices
are, and therefore, equilibrium production prices which are 'consistent' will never be established. This fact puts an upper limit on the value of m. The repercussions of changes in production prices By means of thoroughgoing application of the current value approach to the calculation of costs, firms repeatedly re-calculate their costs at the latest prices and revise their product prices. As a result of these 'iterative calculations' the prices firms fix for their products will approach equilibrium product prices. Of course, the convergence will not be complete and therefore actual prices will differ somewhat from their equilibrium values. But if we analyse how equilibrium prices vary we should be able to grasp the main trends in the variation of current prices. Now, if Vx changes, how might the equilibrium prices /??, p\ change? In Vx are included costs which are incurred in acquiring goods from Group II (producer goods whose prices are determined by competitive trading), goods from Group III (primary factors of production such as labour etc.), and fixed costs. For example, if the prices of Group HI goods which are necessary for the production of product 1 but are not used in the production of product 2 rise, or if only the wages of firms producing product 1 rise, V1 alone will increase whilst V2 will remain unchanged. Since such an increase in Vx will cause an increase in the solutionsp\ and/?'! to (5) and (6), points a and b in Fig. 18 will move to the right to a' and b'. Consequently the new partial equilibrium curve of product 1 will be given by the dotted line which passes through a', b'. The new equilibrium point will be at the intersection p1 of this dotted line and the partial equilibrium curve cd of product 2. If we express the difference px -p° between the new and old equilibrium points as Ap, we can immediately read off from Fig. 18 that Hicks' following three laws are established: (i) The price of good 1 whose cost Vx has increased will rise. Apx > 0. (ii) The price of good 2 whose costs do not increase directly will also rise. Ap2>0. (iii) The rate of price increase for good 1 will be larger than that for other goods. Ap1/p1 > Ap2/p2The above analysis is premised on the fact that the goods of Group I whose prices are set by the full-cost principle are of two kinds, but we can now carry out a similar analysis however many kinds of goods there are, and show that (ii) and (iii) above will be established in relation to all other goods, no matter how many kinds there are. Hicks explained that there are repercussions of price changes according to his three laws
The traditional theories
83
Pi
b'
Pi
Figure 18
amongst goods of Group II, but as the above (i), (ii), (iii) show, analogous laws of price changes are established amongst goods of Group I. 4 The traditional theories: the marginal productivity theory and the labour theory of value
The two orthodox theories In economics the theory that production prices are determined according to the full-cost principle does not yet form the main stream of contemporary thinking. The present situation is one where the marginalist theories of the neoclassical school are regarded as orthodox on one side of the world of economics, while on the other the theories of the classical economists (Ricardo, Marx or Sraffa) based on the labour theory of value are regarded as orthodox. How then is the full-cost principle different from these theories?
84
Fixing product prices
As we shall make clear later, our theory is positioned between the two opposing orthodox theories. In one aspect it resembles neoclassical theory, and in another it resembles the classical school. Because of its position in relationship to the orthodox theories, our theory may perhaps suffer the fate of being abhorred by all those who call themselves orthodox as near relatives are sometimes hated. Yet the minute difference between us and our close relatives is very important. To ignore this would mean a return to our neoclassical or classical ancestors, and we would succeed to the errors they committed. Neoclassical marginalism Pealing firstly with the neoclassical economists, their basic conception is the same as the full cost principle, encompassing a 'U-shaped average cost curve' c{x) and an 'estimated demand curve' x{p). The latter describes how estimated demand for the firm's own product will change when its price changes. But in the neoclassical theory of the firm an inverse demand curve p(x) is usually employed which takes the volume of estimated demand as the independent variable, and price as the dependent variable, thus reversing the relationship. This price shows at what price we must sell, in order to sell-off a given output of x. Because the price has to be lower to sell a larger volume of output, the differential coefficient of p(x), p' = dp/dx, will be negative. Total sales R are p(x)x, total cost C is c(x)x, and profit II is R - C. (i) 77= R - C, (ii) R = p(x)x, (iii) C = c(x)x If we differentiate these with respect to JC, we obtain (i') IT = R'- C , (ii') Rf =p'x+p,
(iii') C = c'x + c
but so long as the marginal profit IT (that is, the increase in profit obtained for a unit increase in output x) is positive, profit is increasing; if it is 0, there will be for the time being no change; and if it is negative profit is decreasing. If profits which were hitherto increasing temporarily cease to change and decrease thereafter, profits at the point of temporary halt will be at a maximum. We therefore obtain II' = 0 as the condition for maximum profits, that is, /?'= C\ Namely, at the point of maximum profits marginal revenue (i.e., the increase in revenue obtained for a unit increase in output) will be equal to marginal costs (i.e., the increase in total costs required for a unit increase in output). Because p' < 0, the marginal revenue R' from the above equation (ii') will be less than the average revenue p. That is, the marginal revenue curve which shows how R'{x) changes with changes in x, will be beneath
85
The traditional theories
C
Figure 19
p{x), the average revenue curve (or the inverse estimated demand curve). Similarly, the marginal cost curve will be beneath the average cost curve where the average costs are falling (c' <0) from (iii'), and conversely where average costs are increasing (cr > 0), the marginal cost curve will be above the average cost curve. Thus at the point where the average cost curve ceases to fall and starts to rise and average costs are at a minimum (c' = 0), the marginal cost curve and the average cost curve will intersect ( C = c). Considering the above in graphic form, we can obtain Fig. 19. As we have already stated the volume of output x° which corresponds to the intersection a of the marginal revenue curve and the marginal cost curve is the output which maximizes profits. According to the neoclassical school, the firm will produce x° - where the average cost of production will be c° = c(x°) - and will sell x° at the price p°=p(x°) to obtain a profit, p° — c°, per unit of output. If we take the ratio of profits to average costs as A:0, then
86
Fixing product prices
Variations The above is the basic form of the neoclassical theory of the firm, but two important variations of it also exist, dealing with the cases where the marginal revenue curve is horizontal and where the marginal revenue curve is kinked. In the case of neoclassical perfect competition, the output of each firm can be only an extremely small part of the society's overall demand, and the firm can estimate for a limitlessly large demand where the price is fixed. Consequently, as far as the firm is concerned, this is the same as if there were no restriction on the demand for its output, and the only condition facing it is that it must sell each unit of output at the same fixed price. Given this condition marginal revenue is a constant, and therefore, the marginal revenue curve is horizontal, so that the above diagram will change to that in Fig. 20. The firm is able to realize maximum profits at the point where marginal costs and the fixed price p° are equal. Extremely pessimistic firms will have a kinked marginal revenue curve. In Fig. 19, we drew a downward sloping demand function on the assumption that demand likely to be lost when the price is raised above p° by one unit is the same as the extra demand acquired when the price is lowered by one unit below/?0. However, the pessimistic firm may well assume a great deal of demand will be forfeited when the price rises and
Figure 20
87
The traditional theories
0
x°
R'
x
Figure 21
that only a very small increment in demand will be gained if the price is lowered. The marginal revenue curve will, in this kind of case, bend or kink at the stage where output x° is reached, as shown in Fig. 21. Needless to say, profits will be maximized at the intersection of this kinked marginal revenue curve and the marginal cost curve.7 In the above neoclassical theory all that is explained is how the volume of output and how price is determined where the firm's production plan is formed so as to maximize profits. As regards competition between firms it can say nothing even in the case of perfect competition. Therefore, we can accept their theory as more or less correct only when there is a monopoly and no competing firms exist; or where the firm produced a product for a perfectly organized market which means the firm accepts prices set by the exchange and certainly 7
In the case of perfect competition dealt with earlier, we assumed that firms would lose all their customers if they raised their price but would attract unlimited numbers of customers if they lowered it. In other words, looked at in terms of the case we are now dealing with, we can regardfirmsas being very pessimistic with regard to price increases and exceedingly optimistic with regard to decreases in price under perfect competition. A firm which is pessimistic with respect to price cut will never declare price war as will be discussed later.
88
Fixing product prices
does not engage actively in a price competition. In fact the prototype of the neoclassical theory shown in Fig. 19 was first presented by Cournot as a theory of monopoly, and there is no element of price competition present in it at all.8 On the other hand, in the case of Fig. 20 too, the horizontal marginal revenue curve which is presented as a Theory of Perfect Competition' also has no element of price competition, and each enterprise accepts the price which was set at the exchange, and merely responds to it. Thus the neoclassical theory is a theory without price wars. But where competitors exist, even though each firm aims at profit maximization and thus the volume of output is determined according to the neoclassical formula, immediately hereafter 'price wars' will break out between firms and the following course of events will develop. Price wars
For any firm there is a limit to its mark-up rate below which it cannot fall if the firm is to continue to exist. Let that limit value be ra. The firm's mark-up rate must at the least be ra or greater than ra. When it is less than ra and it is judged there is no prospect of the mark-up rate recovering for a long period in future, the enterprise may well dissolve and dispose of its assets. Explanation of how the value of ra is determined will be left to page 174, and we will here consider what sort of situation will arise in the case of ra being assigned a value which is different from k°, the mark-up rate at the neoclassical profit maximization point. There is, firstly, no necessity for k° to be equal to ra. There are cases where k° is smaller than ra, and there are also cases where it is larger than ra. To begin with, if £°
0. Next, what kind of situation might develop if k°>m? Let us assume that two enterprises of the same type - that is, enterprises A and B which have the same cost curves and the same estimated demand curves - operate in accordance with neoclassical principles. They will produce an output x° which will maximize their profits, and will sell this at a price p° = p(x°), attaining a mark-up rate k° such that p° = (1 + /c°)c°, where However, they will only be able to behave according to the neoclassiCournot, A., Recherches sur les Pricipes mathematiques de la Theorie des Richesses, Paris, 1838, Chapter 5.
The traditional theories
89
cal principles of profit maximization up to this point, and beyond it a price war is at once bound to break out. That is to say, what makes it possible for both A and B respectively to sell output x° at p° is the fact that both firms are selling their product at the same price. If the products of A are cheaper than those of B, the latter's customers will all move over to A, and A will be able to sell at least 2x° of its product. That is, setting k°> kA>m,we suppose A sells its own outputx° at the price PA
= {\ + kAy
(7)
In doing so the profits from x° will momentarily decline, but since B's customers will throng to A because it is selling cheaply, then if A expands its output from x° to 2x°, A will be able to get larger profits by dint of larger sales and a lower profit margin. However, the increasing output from x° to 2x° may cause a marked increase in the average costs of production. It will certainly not then be good policy to increase output to 2x°. It should confine its increase in output within the limits where there occurs no significant increase in average costs. That is, it should produce an output nx° such that c(x°) — c(nx°) does not have a significantly large negative value (where 1 < n ^ 2). Since in that case profits are
nn=pAnx°-c(nx°)nx0 provided that the product is sold at pricep A , they will be larger than the profit 77° = k°c°x° which is earned when only x° was produced and sold before the fall in price. For if we consider (7), the difference between the two levels of profit will be n»-n°=
[nkA - k°]c°x° + [c° - c(nx°)]nx°
(8)
Since the part within the second set of square brackets on the right-hand side of this equation will only attain a negative value such as can merely be ignored, whether or not Un is larger than 77° will be determined by whether the part within the first set of square brackets on the right-hand side is positive or not. If kA has a value sufficiently close to kQ this part will necessarily be positive so that we obtain TIn> 77°.9 That is, if firm A 9
Let pA be taken such that it is larger than the marginal cost at x() but smaller than p°. The TIn is an increasing function of n at n = 1. This can be shown in the following way. Differentiating IJn with respect to n, we obtain dnn/dn = {pA- {c'(nx°)nx° + c(nx°)}]x° On the right-hand side of this equation the part in the braces represents the marginal cost at x°. Since pA is taken to be larger than this, dTIn/dn is positive at n = 1; hence TIn > 77° for some n > 1 if pA is taken sufficiently near to p°.
90
Fixing product prices
is content with a profit margin kA which is lower than k°, usurps customers from B and increases its sales to nx°, it can attain a much larger profit Un than the maximum profit W it would have attained had there been no price war. It is therefore worth A's while to declare war. Once challenged by A, B must also go to war. As long as B goes on selling at/?0 it will lose demand (n - l)x° to A and profits will decrease to &°c° (2 - n)x0.10 In order to maintain demand at its former level B must imitate A and lower the price pB of its own product to the level of pA. If it does so A will earn profits kAc°x°, because the clients usurped by A will be lost back to B. Therefore, in order once again to steal back the customers it has lost, A may well further lower its mark-up rate from kA to k'A. Only by so doing will A be able to raise its profits above kAc°x°. B will also fight back by making k'B — k'A. Thus once a price war breaks to try to achieve large sales through low profit margins, the mark-up rate will be continually lowered whilst kA and kB are greater than m. When finally they approach m, the price war will gradually come to an end because a greater reduction in the mark-up rate will make it impossible for the firms to survive. It goes without saying that for both firms the situation after the war will be clearly worse than the situation before commencement of hostilities when profits were maximized.11 However, such a price war will be unavoidable so long as both are independent, competitive firms, and there is no other way for them to avoid it but to form a cartel and jointly maintain prices.12 As long as industry is competitive, the mark-up rate of competing enterprises should just equal the rate m which makes their continuing existence possible. If we take k to be the firm's actual mark-up rate, it will be possible for k > m only where the firm is either a monopoly or forms an effective cartel with other firms in the same industry. We may be able to calculate the degree of monopoly exercised by a firm according to the ratio (k — m)/(k° — m). According to this index, when k = A:0, that is when the current mark-up rate is equal to the maximum profit mark-up rate, the degree of monopoly is 1; and when k = m, that is when the firm has to be content with a mark-up rate which just enables it to survive, the degree of monopoly is equal to zero. Most 10 11
12
Since 1 < n < 2, this profit will clearly be smaller than the maximum profit kc°x°. At each of the stages at which kA is lowered, A will always have done so because it anticipated an increase in profits due to larger sales at a lower profit margin. But because B responds in kind all of A's moves will be rendered negatory, and A's situation will ultimately be worse that it was at first. If A had been aware of this fact he would probably not have declared war. The same applies to B. If neither initiates the hostilities no price war takes place, and in this case A and B can be regarded as constituting a cartel tacitly. See Additional Note e.
The traditional theories
91
firms which have a monopoly or which actually participate in cartels, notwithstanding their position, have to be content with a mark-up rate much smaller than the k° of the neoclassical school due to the pressure from consumers, the mass-media and from legal restrictions. Ricardo Classical scholars such as Smith, Ricardo, and Marx believed prices were equal to 'production costs'; that is, production costs inflated by the addition of average profits, i.e. production prices. The full-cost principle may be considered as a version of resuscitation of the classical production cost theory. Let us consider an economy formed from agriculture, a consumption goods industry, and a capital goods industry. We shall call these industries 1, 2, 3. The capital goods industry (industry 3) produces all-purpose machines, which serve as tractors in agriculture, as spinning machines in the consumption goods industry, and as the robots which make all-purpose machines in the capital goods industry. Let alh a2h a3h lt be the production coefficients for the i-th industry; alh a2i represent the quantity of agricultural products and the quantity of consumption goods required to produce one unit of the i-th product; and a3h lt represent the number of machines and the number of workers also required to produce one unit of the i-th product. If 1000 units of the i-th product are produced by one machine the number of machines per unit product a3l will be 1 machine/1000 units. We can write cost of production equations (full-cost equations) for the three industries respectively as follows: Pi = (1 + m) \pxan + p2a2l + p3a3X + wl\] p2 = {l + m) \pxan + p2a22 + p3a32 + wl2]
(9)
p3 = (1 + m) \pxax3 + p2a23 + p3a33 + wl3]
For the classical economists, the task of their price theory is to explain by what the rates of exchange between products of different industries are determined (that is relative prices - for example, pjp2 etc.), and at what level these ratios settle. With the mark-up rate (or the classical economists' 'rate of profit') m and the wage rate w given, we can solve the above equations with respect to prices/?!, p2, p3 and divide them by each other to find the exchange rate pjp2. However, the classical economists were not satisfied with these mathematical solutions. Ricardo says The exchangeable value of the commodities produced would be in proportion to the labour bestowed on their production; not on their immediate production
92
Fixing product prices
only, but on all those implements or machines required to give effect to the particular labour to which they were applied.13 That is, if we let the amount of labour required to produce a unit of good i be A,-, the relative price pijp] will be equal to the relative amount of labour required A,/Ay, which is Ricardo's labour theory of value. How large then are the quantities of labour A1? A2, A3, required to produce a unit of each product? We shall start our explanation with A^ To produce a unit of agricultural net output not only will we need to produce consumption goods and machines, we shall need to produce more than one unit of agricultural output (due to the multiplier effect). The reason for this is that we need agricultural products, consumption goods, and machines to produce agricultural products, and must therefore also produce these consumption goods and machines. In order to do so additional agricultural products, consumption goods and machines will be required. Thus expansion of production by one unit in the agricultural sector will effect the consumption goods and machinery industries, and so these affects will further rebound on agriculture. Let the output of each industry required to produce a unit of agricultural net output be Xu X2 and X3. The volume of agricultural output needed to produce these outputs will be anXu auX2, and aX3X3, and so the output from agriculture Xx must cover these requirements and furthermore create one extra unit (the net product). In the case of consumption goods there is no need to create a net product, and it will suffice if X2 just covers the volume of consumption goods a2XXx, a22X2, a23X3, required to produce Xu X2, X3. The same will hold for machines. Therefore, Xx = anXx + aX2X2 + aX3X3 + 1 X2 = a2XXx + a22X2 + a23X3 + 0
(10)
X3 = a3XXx + a32X2 + « 3 A + 0 Furthermore, since the quantities 1XXX, 12X2, lyX3 of labour are needed to produce Xu X2, X3 respectively, the quantity of labour Xx required to produce a unit of agricultural net output will be given by (11) To obtain the quantity of labour necessary to produce a unit of the consumption good we replace 1 by 0 in the first equation of (10) and 0 by 1 in the second, keeping 0 in the third; we then solve the new set of 13
Ricardo, D . , On the Principle of Political Economy and Taxation, in The Works and Correspondence of David Ricardo, ed. P. Sraffa, vol. 1, Cambridge University Press, 1953, p. 24.
The traditional theories
93
equations with respect to A^-'s and substitute them into equation (11) to obtain A2. The labour value of the machine may be calculated in a similar way. It can be shown that the labour values A1? A2, A3 obtained in this way are equal to the solutions to the following equations: 14 A2 = A^n + A2#2i + ^3fl3i + h A2 = X\an + A2a22 + \3a32 + l2
(12)
A3 = \xal3 + A2fl23 + \3a33 + l3
If we write the equations for determining labour value in this form, what Ricardo is saying will become apparent. He contends that the ratios of prices which are obtained by solving (9) are equal to the corresponding values obtained by solving (12); but it is clear that in some cases this proposition is correct, and in other cases it is incorrect.15 Ricardo himself was fully conscious of this fact, but he considered, in the most part of his Principles, 'all the great variations which take place in the relative exchange value of commodities to be produced by the greater or less quantity of labour which may be required from time to time'.16 He confined himself to pointing out examples where his propositions would no longer be true. Marx It was Marx who elaborated the conditions under which Ricardo's labour theory of value would stand. Apart from the capitalist economies he considered societies which were still at the stage where workers and capitalists did not form separate groups, and he called such societies simple commodity production societies. Here profits and wages were 14
Solving (10) and substituting into (11), we obtain kx = L(l — A)~le{, where L is the row vector (/1? /2, /3), A the matrix {atj) and ex the column vector whose first component is 1, all others being 0. Similarly the labour values of the consumption goods and the machine may be calculated by the formulae, A2 = L{\ - A)~le2 and A3 = L(I — A)~le3, respectively. These three formulae can collectively be written in the form A = L(l-A)-1(el,e2,e3) where A is the row vector (A1? A2, A3). In view of the fact that (ex, e2, e3) is the 3 x 3 identity matrix, we obtain from the above equation A = AA+L
15
16
which is equation (12). The proposition holds true when all a^s are 0. However, where in agriculture an, a2l, a31 are all 0 and in the consumption goods and capital goods industries at least one of a \v a2n a3i *s positive, the labour theory of value will not be valid. That is, pjp^kj\ (i = 2,3). Ricardo, op. cit., pp. 36-7.
94
Fixing product prices
not differentiated, and the surplus was received in its entirety by the producers (that is, the workers). If in such a society there were to be differences in the rates of income earned per unit of labour in different industries, producers would probably move from industries with a low rate of income to those with a high one. Producers would cease to move, when the rate of income became the same throughout all industries. That is, the equations obtained by putting m = 0 in (9) would constitute the price equations for a static, simple commodity producing society. (Note that in these equations w would no longer represent the wage rate but the rate of income received per unit of the producers' labour.) By comparing these price equations with the value determination equations (12) we immediately realize that p / /w = A1, and therefore that PilPj ~ Kl^r That is to say, Ricardo's labour theory of value is completely correct for a static, simple commodity production society. This kind of society is, however, not a capitalist society. In a capitalist society m should be positive, and for the labour theory of value to be valid when m > 0, the production coefficients atj must satisfy a special condition. Marx called the proportion of indirect labour included in product i to its direct labour, that is ( A ^ + A2tf2/ + >M3/)A> the value composition of that product, and demonstrated that the labour theory of value was correct where the value compositions of all products were equal to each other, and also that the labour theory of value was correct in that case alone.17 Since the condition that the value composition of all products be equal is a very strict one, so the labour theory of value will only be valid in a capitalist society in very exceptional cases. As opposed to Ricardo who considered that the labour theory of value was valid except in exceptional cases, Marx considered that it would not normally be valid. Thus Marx abandoned Ricardo's theory of labour value but carefully preserved Ricardo's concept of value A,. He explained, by using the concept of value, how, in a capitalist society, workers were exploited by capitalists, and in so doing revealed how 'exploitation is the sole source of profit'. I have called this the fundamental Marxian theorem but do not go into its mathematical proof in this volume.18 Marx has thus used the concept of labour value to prove that profits are created when and only when there is exploitation. For Marx, the labour theory of value is no longer, as it was for Ricardo, a theory of prices. It is indispensable for the theory of exploitation, on the basis of which Marx rigorously establishes his own theory of profits; moreover, 17 18
For more detail see: Morishima, M., Marx's Economic, Cambridge University Press, 1973, pp. 82-3. See ibid., pp. 63-8.
How should we treat the traditional theories of the firm?
95
he obviously recognizes that relative prices generally deviate from relative values. In this sense it may be said that Marx is clearly superior to Ricardo, but in other aspects his theory has the same weaknesses as Ricardo's. They both assume that prices are determined by production costs - that is by the full-cost principle. Yet in the real economy the prices of one group of goods are set by competitive trading for which exchanges exist. The real economy is an eclectic mixed economy founded on both the full-cost principle and competitive trading. The theoretically pure Ricardo and Marx both ignored this fact, and constructed a 'unitary' model sticking consistently to a single principle.19
5 How should we treat the traditional theories of the firm?
Monopolists, price-takers and small proprietors I would like to outline my thoughts on how I regard the theories of the firm as they have stood in the past. I divide products into two groups; the first of which consists mainly of products from manufacturing industry which have no perfectly organized market (exchange), and the second of which consists of products which have exchanges (agricultural, forestry and fishery products, mineral products). With Group I products, where output is monopolized by a single firm the production activity of the firm may be explained by the classical monopoly theory. But when the mark-up rate decided by the monopoly firm (the difference between the monopoly price and average costs divided by average costs) is extremely large, the firm will not be able to accumulate profits without let or hindrance even though it has a monopoly. Due to the pressure of public opinion, moral pressure or political pressure from the government, the mark-up rate cannot exceed a given upper limit; subject to these constraints, the firm, with given demand and cost functions, will draft its production plan so as to maximize profits. The theory of the full-cost principle will pertain to all other firms producing Group I goods. In that even the number of competitors does not present any problem, but of course where there are few competitors it may well be easy for them to act in concert. Should they do so by forming an explicit or implicit cartel, they will behave in most respects 19
One might say that in the classical and Marxian economics prices are determined by competition and, hence, are not set by any particular individual or organization, while in the full-cost principle theory prices are set by the manufacturers independently of competition. This is a wrong interpretation of what I said in the text. What I said there is that for agricultural products competition is made mainly in exchanges, whereas for manufacturing products it is made, in terms of the mark-up rates, within the framework of the full-cost principle. Thus our prices are also competitive prices.
96
Fixing product prices
as monopolists, but to the extent that they remain competitors the competition will be intense (even with two competitors). The price war which will take place between competitors may well more or less standardize their mark-up rates at the minimum value. The prices of goods in Group II will not be set by individual firms but in the exchanges. Such prices will then be declared as fair price for that time and will possess authority outside the exchange. Whether or not firms sell through exchanges, they will adjust the quantities they produce so as to maximize their profits with the official price determined on the exchange as a datum. In accordance with the equilibrium conditions of neoclassical 'perfect competition' whereby marginal costs equal product prices, firms will increase (decrease) their output when product prices rise (fall). Therefore, supply of Group II goods will be an increasing function of their price. In addition, small proprietors also exist in modern societies. Whilst themselves owning the land, tools and machines required for production, they are themselves workers. For such small proprietors income from their assets (capital) and income from labour are undifferentiated, and their income is what is left when each and every cost incurred in production has been subtracted from the price of their products. Their income equation will therefore be w/,- = Pi ~ Piau - p2a2i
(where w indicates the undifferentiated, combined rate of income received for one unit of labour by small proprietors), which, if rearranged, tallies with Marx's value equation (9) for a simple commodity production society. Where such small proprietors together produce between them the goods which they themselves consume in the production process and form a self-sufficient society of simple commodity production, the labour theory of value will hold within that society. However, in actual economies small proprietors all transact with capitalist firms (which produce goods in Group I or Group II), and therefore small proprietors do not make up a closed society. Thus the labour theory of value as a theory of price will not be valid for them. Price-adjustment and quantity-adjustment sectors If we treat small proprietors as separate, we can see at a glance the price-fixing mechanisms for goods of Group I and those of Group II are quite different. With goods of Group II product prices will be regulated by competitive trading in the market. If the demand which accumulates at the exchange is d and supply s, then p = F(d(p) — s(p)). Since firms will produce so that marginal costs are equal to the market
How should we treat the traditional theories of the firm?
97
price p° thus determined, we can add up the output of all firms and obtain the total output X(p°) for that good. Now if for the sake of simplicity we ignore the gestation period for production, all of the goods produced are likely to be supplied at the price p°. On the other hand, total demand in the society will be D(p°). If we assume that an unbiased sample of society's total demand and supply accumulates at the exchange, then d(p°) = nD(p°), s(p°) = nX(p°), where n is a stable fraction. When market equilibrium is established at the exchange, we must have d(p°) = s(p°), and therefore D(p°) = X(p°). In the case of goods of Group I, output is estimated when prices are fixed, but such an output is never anything more than an estimate. In actuality such estimates are often mistaken, and the effective demand which appears will sometimes exceed and sometimes be less than that estimate. Where production is inadequate it will rapidly increase, and where it is excessive it has to be sold off over a period. Yet even in this case the firm will not re-calculate its average costs of production in response to the amount it actually produces, in order to revise its price. Since it will continue to sell at the price it has already fixed, when it increases its output it will earn profits which are markedly greater than those anticipated, because the estimated fixed costs per unit of output overestimate the actual fixed costs per output. In the converse case where the actual volume of sales does not reach the size which was anticipated, profits fall rapidly below those anticipated and the firm often incurs great losses because it under-estimated the fixed costs per output. Firms which produce goods in Group I must absorb unsold goods into stocks when there is excess supply (D<X). They may well contract production in order to avoid swollen stocks. That is X< 0. Since stocks will decrease when there is an excess demand (D>X), firms will carry out an increase in production (X> 0). X here shows the total output of the good, and production will show a tendency to increase or decrease according to whether X is positive or negative. Therefore we can obtain the output adjustment function for goods in Group I:20 X=F(D-X) As long as there is nothing untoward in this adjustment process, prices will not be revised. Prices will be revised where costs have risen by 20
Put in more detail, producingfirmsdo not directly encounter thefinalbuyer. What they encounter is demand from wholesalers, supermarkets, consumers' associations and trading companies. If we let this demand be D', and the demand of the final buyer be D, then we may express the adjustment mechanism in the process of distributing Group II goods as X= F(D' -X),D' = G(D - D').
98
Fixing product prices
reason of an increase in wages or raw material costs, but they will not usually be revised by reason of an inadequate or an excess supply. In other cases too, where to lower prices may appear to improve the operating situation of the firm, firms will not alter prices unless they are driven into a corner. This is partly because the losses they continually incur are compensated for by large profits made when demand was buoyant; but it is also partly because to lower prices would be to announce publicly that they had been driven into a corner, which from the point of view of the management of the firm would be a bad thing. 21 Thus since the excess demand for Group I goods will be regulated not by altering prices but by changing the volume of output, we may say these industries constitute the fixprice or quantity adjustment sector. In contrast, those sectors which produce goods in Group II constitute the flexprice or price adjustment sector. As we have already stated, at the end of the nineteenth century industrial countries with the exception of Britain still had a large flexprice sector, but at present the size of the fixprice sector in most industrial countries is much larger than before, and the flexprice sector is gradually assuming the status of an appendage. 22 21
22
Even when they are monopolists, where firms which produce Group I goods find that the quantity of demand they actually attract differs from their estimated output, they may well leave prices unchanged and adjust quantities. Whilst this is true, since America, France and Italy still have a quite large flexprice sector, we may say that these countries' economies are mixed economies made up of flexprice and fixprice sectors. For a mathematical model of such mixed economies see: Morishima, M., Walras' Economics, Cambridge University Press, 1977, Chapter 7.
Determination of foreign exchange rates
1 The century of internationalism
Onset of protectionism It is no exaggeration to say that the period from the latter half of the nineteenth century and right through the twentieth century was 'the era of internationalism'. Britain was the pioneer in this. The period from 1840 to 1870 was one in which Britain strenuously advocated liberalism in economic ideas, and during this time the internationalization of the economy was pushed forward to the extent that the value of net imports and exports which had formed only 13% and 9% respectively of the Gross National Product in 1841 made up 24% and 20% in 1871.1 At the same time industrialization proceeded in Britain and she came to import food and raw materials from abroad and to sell industrial products to foreign countries. Britain thus became the 'workshop of the world' and other countries become 'Britain's granary'. Such an international division of labour only became possible in the context of Britain's huge colonial empire. Britain's industrial pre-eminence also made her militarily strong. Thus with the establishment of this international division of labour which enabled Britain to concentrate on her industries, the British Empire became stronger and larger. All the cogs in the machine fitted smoothly. Stimulated by Britain's success France and Germany both opened the doors of their economies, and then eventually even Japan began to march out into the markets of the world. We may infer that the period 1850-1880 in France corresponds to the period when Britain's economy was opened up (1840-1880), and in Japan it is probably 1880-1910. In France the ratios of imports and exports to GNP were on average 5% and 6% in the 10-year period 1845-1854, but reached an average of 16% and 13% in the ten-year period 1875-1884. In Japan they were just under 4% and 5% in 1880 but reached 17% and 18% respectively by 1
Some imports are re-exported abroad. Net imports is the figure we obtain by subtracting re-exports from imports. 99
100
Determination of foreign exchange rates
1910. Then, as did Britain, these countries built a large sphere of influence on the basis of the fruits of industrialization and with the regions under their influence as sources of supply of food and raw materials they attempted to reach an even more advanced stage of industrialization. Thus, directly connected with the earlier period of liberalism, the period of aggressive imperialist competition appeared. These countries, including Britain, sought sources of supply of food and raw materials and markets for their industrial products; competition to grab colonies was rife and their struggle eventually exploded in the form of the First World War. Attracted by the profits accruing to a monopolist, outsiders will attempt to break into an industry; monopoly therefore gives rise to competition. Competition ends when the victorious competitor becomes a monopolist. It is certainly not difficult to find historical examples of chains of causality where when X produces its opposite, the opposite will produce X. In the area of trade philosophy and policy, the philosophy of free trade gave birth to protectionism, and conversely the latter eventually provoked the reappearance of the former. In fact the British period of free trade was the successor to the period of parliamentary mercantilism. In that era, all the wealth of the colonies - both in gold and silver and in other resources - was exploited for the prosperity of the mother country. They were subordinated to the mother country. The relationship between the mother country and its colonies was that of head to limbs, aims to means. They together formed a self-sufficient commercial hierarchy. However, the opposition of the American colonists who wanted the abolition of the Maritime, Stamp, Sugar and Tea Acts, and subsequently, American Independence, rang the death-knell on this colonial system. Whilst free trade had its positive aspects in that it secured benefits for the mother country, it was also a policy designed to pacify the colonies and the product of compromise with them. Having been shown the success of the British Empire, other countries (Germany, France, America etc.) tried to emulate Britain. However, late in industrialization, they had to tread a quite different path from Britain. Unlike Britain, they did not put their faith in the philosophy of free trade and specialize in industries which, by international comparison, seemed advantageous to them. The industries in these countries which were relatively advantageous at that time in comparison with Britain were not in manufacturing, but were agriculture or mining. Therefore we should have to conclude that what they ought to have done according to the theory of comparative costs would have been to export the products of agriculture and mining to Britain and receive in return British industrial products. But to have done so would have
The century of internationalism
101
closed off to them the road to industrialization. If they were to become industrial countries in their own right and compete with Britain as such, they must needs abandon the theory of comparative costs upon which the philosophy of free trade was based and establish a new trade philosophy. It was the nationalist and protectionist trade theory of Friedrich List which met this need, and it suggested a variety of protectionist trade policies for the nurture and protection of her infant industries in order to overcome Germany's late entry onto the stage of world history. Putting up international fences When industry had been developed up to a certain point, colonies had to be maintained as the source of supply of raw materials and a potential sales outlet for manufactured goods. These colonies, like those of Britain in the mercantilist era, were subordinated to and had to serve their mother countries. Not only was there ferocious economic competition between the early and the late developing capitalist countries, such competition taking the form of improvements in the quality of products, the development of new production techniques, the lowering of production costs and of prices, the exclusion of the goods of other nations, by means of tariff barriers, and so on; wars often occurred to secure new colonies or to redistribute existing ones. Powerful countries would fight over a less-powerful one; or else it would be left to one of them to invade it. Thus the antagonism produced in the 'defence' of the British Empire against the imperial advance of late developing capitalist countries (chiefly Germany) eventually exploded in the form of the First World War. However, even when the result of the Great War was settled, the struggle for colonies did not end. Japan, which had fished in the troubled waters of the Great War, sent troops into the Shantung area of China in 1927 and caused the Manchurian Incident of 1931, the Shanghai Incident of 1932 and the Sino-Japanese War of 1937 (the China Incident). As well as making Manchuria independent in 1932 and enclosing the North-Eastern region of China in its own sphere of influence, Japan nibbled away at other regions of China during the Sino-Japanese War. With Hitler seizing power in Germany in 1933, Germany began to undermine the Versailles system. In October of the same year Germany left the League of Nations, and in 1935 declared it would expand its armaments. Japan, too, had left the League of Nations in 1933 before Germany, and had announced to America in 1934 its abandonment of the Treaty of Washington. Thus the League of Nations became little more than a name, and the era of international co-
102
Determination of foreign exchange rates
operation after the Great War, whose basic philosophy had been liberalism, became a thing of the past. In fact the London arms reduction conference of 1936 also broke down. The rising countries of Japan, Germany and Soviet Russia, particularly Japan and Germany, desired to build up their own systems of self-sufficiency by establishing spheres of influence which would enable them to rank with the British Empire. In 1935, Italy invaded Ethiopia. Within this historical context, the internal structure of the British Empire was compelled to change. The autonomous Dominions within the Empire (Canada, Australia, New Zealand, South Africa) were not inclined to sacrifice their own young industries for the sake of industrial development in the mother country, and problems of free trade versus protectionism not only existed between the British Empire and other capitalist countries, but were also a bone of contention between Britain and the Dominions within the Empire which possessed the potential to become industrial nations. In order to protect and develop their infant industries the Dominions were allowed to raise trade barriers. Moreover, in order that the Dominions should continue to provide export markets for Britain, the Empire had no other recourse but to raise high tariff barriers against foreign goods and grant preferential treatment in Britain's domestic market for goods from the Dominions in return for similar treatment from them with regard to British goods (Imperial Preference). That is, the British Empire intended to foster the infant industries of the Dominions at the cost of goods from other countries, but were it to carry out such a policy of exclusion it would incur the counter attack of other countries, and lose export markets in countries outside the Empire. Despite this, Britain went so far as to adopt a protectionist trade policy to make such a sacrifice because of the rapid advance of the late-developing capitalist countries and the necessity of defending the British Empire from within so as not to be ruined by these countries. A further reason was the need to protect British agriculture from the danger of collapse in the face of the world slump, even though that agriculture had already contracted to its minimum scale. Formation of economic blocs By 1932 the policy of Imperial Preference was adopted in its entirety, and Britain turned from a policy of free trade to one of protectionism. The period of free trade which had begun with the abolition of the Corn Laws in 1846 had run its course after 86 years. It will be clear from the following figures that the tariff policy was a great success. Canada, Australia, New Zealand, India and the West Indies were Britain's main trading partners within the Empire; if we compare the total of imports
The century of internationalism
103
from these countries to Britain and Britain's total exports to them with the total imports from Germany, France, Holland and Belgium and Britain's exports to those European countries, we find the following. In 1929 imports from the former were 1.1 times those from the latter, but had dramatically increased to 2.8 times in 1938. As for exports, those to the former were 1.2 times greater than those to the latter in 1929, and had increased to 1.6 times greater in 1938.2 Thus the result of the Imperial Preference policy was to shut out the countries of Europe from trade with Britain, and trading ties between Britain and the Dominions and Colonies within the Empire were markedly strengthened. The British Empire became a single, solid economic bloc, but because Germany and Japan also tried to establish their own spheres of influence, even by force of arms, the world economy entered an era of large economic regions encompassing many countries (Grossraum-Wirtschaft). These economic blocs were, at that time, each independent fighting units, and so the danger of war between them was always present. If we consider the problem in this way, we must conclude that the Second World War was the inevitable outcome of protectionist trade policies. With the various economic blocs as their bases, currency areas were formed, such as the Sterling, the Dollar area, and the Yen area. The countries which belonged to the Sterling area were not necessarily Dominions or Colonies of the British Empire (at first countries outside the British Empire such as Scandinavian countries, Iceland, Portugal, Argentina, Egypt, Sudan, Iraq, etc. also belonged to it; but when the Second World War broke out Scandinavia, Argentina, Portugal, etc. dropped out). Moreover not all Dominions and Dependent Territories of the British Empire belonged to the Sterling area: Canada and Hong Kong are outside the area. All currencies within the area were linked to the pound and their value in pounds kept fixed. Dollars earned in transactions by countries within the area with those outside it were gathered at the Bank of England which established an exchange equalization account (or an exchange stabilization fund) which it operated to keep the external price of the pound fixed. Multicentric economic blocs The Second World War struck a blow which shook the foundations of this system of arrangements. While it was natural that the economic blocs centring on Germany and Japan, who were the defeated countries, should have been dissolved, it was impossible even for the economic 2
Thesefiguresare calculated from Mitchell, B. R., and Dean, Phyllis, Abstract of British Historical Statistics, Cambridge University Press, 1962, pp. 320-6.
104
Determination of foreign exchange rates
bloc formed by the British Empire which was the victor to survive unaltered. The war taught nationalism to the populations of Colonies and Protected Territories in Asia and Africa with the result that after the war many became independent. It became impossible for Britain to maintain the imperial order. The Empire was broken up and in its place the British Commonwealth - an autonomous group of societies which had joined together freely and equally and whose components were Britain and the white Dominions - was expanded to include the non-white countries which had become independent after the war. Other than that there should continue to exist a union in the form of a Commonwealth of the many races contained in the old Empire, there was no real, substantive policy for the continuation of the Empire. After such major surgery the economic bloc formed by the Empire still had some life in it after the war, but the bloc itself no longer had its past power to lead or its solidarity, nor were its ties as economically attractive as formerly. However, such tendencies certainly did not mean that the end of the era of economic blocs was approaching. Leaving aside the economic and political bloc built by the communist countries of Eastern Europe which was a product of the Second World War, the great success of the EEC in which the major industrial countries of Europe have participated indicates that, in the domain of the international economy, still there is more than a little benefit to be got from collusion amongst several powerful countries. When Britain was forced to choose between the Commonwealth bloc or the EEC, it first countered by forming the European Free Trade Area (EFTA) together with Sweden, Norway, Denmark, Switzerland, Austria and Portugal - many of whom had at one time been members of the Sterling area - but to no avail. Britain aside, just as no industrial country which was strong enough to be compared with Germany, France and Italy, existed in the Commonwealth bloc, neither was there such a country in EFTA. In the latter half of the twentieth century when the diversification and subdivision of industrial products has become very marked, it is no longer possible for an economic bloc centring on a single industrial country of the scale such as Britain's, to produce a comprehensive range of industrial products efficiently. However closely the countries within such a bloc collude, the bloc's power to seal off and exclude others cannot be strong. Modern international trade relations which began with the hierarchical division of labour between the mother country and its colonies, had, after a long history, to develop towards monopolistic competition where the advantage lay with collusive alliances of powerful and diverse industrial nations (the so-called multi-centric economic blocs). That is, we have
The effects of fluctuations in foreign exchange rates
105
left an era of alliances between countries which were heterogeneous and complementary, and entered an era of collusion between countries which are essentially the same but in competition one with another. 2 The effects of fluctuations in foreign exchange rates Fixed exchange rate systems
Thus there was the period of free trade, the period of economic blocs where countries were heterogeneous with complementary economies, and the period of blocs of competitive countries, and yet it is very clear that the period for 100 years or more from the second half of the nineteenth century was remarkably internationalist. Of course, when heterogeneous and complementary countries formed blocs (for example, the British Commonwealth or the Greater East Asia CoProsperity Sphere) it was possible for the bloc as a whole to cut itself off and sever intercourse with areas outside the bloc. However, blocs formed from homogeneous and competitive countries will together become bankrupt if they trade only amongst themselves and do not seek raw materials and markets outside their bloc. Even though they are bloc economies they are, as a rule, open; and even in the extreme case of a closed bloc, the bloc economy itself is already highly internationalized when seen from the point of view of any given country within it. The main problem for the international economy is how to convert the various currencies of the countries from which it is made up. In the age of the gold standard when each currency was backed by gold, exchange rates were more or less equivalent to legally fixed parities. If the currency laws of Japan and Britain specified respectively that 1 yen comprised 11.57422 grains of pure gold, and that 1 pound would include 113.0016 grains of pure gold by volume, the amount of gold contained in 9.6632 yen and 1 pound would be equivalent, and the legally fixed parity would be 1 pound = 9.6632 yen. If the transport costs (including security costs and insurance) of sending 113.0016 grains of gold from Japan to Britain (or the reverse) are 1 yen, the actual exchange rate would neither go below 1 pound to 8.6632 yen, nor above to 10.6632 yen. That is, the value when transport costs are subtracted from the legally fixed parity will mark the lower limit of the exchange rate, and the value when transport costs are added to the legally fixed parity will mark the upper limit. (If for some reason the exchange rate falls below 8.6632 yen, a person wishing to exchange yen for pounds will first buy gold with sterling in Britain, and will spend 1 yen to send the gold to Japan where he will change his gold into yen. In doing so he will obtain 9.6632 yen and thus the net yen he acquires by this method will be
106
Determination of foreign exchange rates
8.6632 yen. Accordingly, should the exchange rate fall below the lower limit, a person changing pounds into yen will take steps to send gold and the exchange rate will not go below the lower limit. Similarly, should it rise above the upper limit a person changing yen into pounds will send gold and so the actual exchange rate will hit the upper limit and not exceed it.) Thus actual exchange rates will be kept within a fixed range around the legally fixed parity under the gold standard, and because transport costs, security costs and insurance fees will get smaller with progress in transport technology and the reduced fear of attack by pirates, the range will contract and the exchange rate will come to vary only slightly around the legally fixed parity. We may thus conclude that the gold standard is virtually a system of fixed exchange rates. From thefixrate to theflexrate regime The gold standard was temporarily shelved during the First World War, but after it the major countries returned to the gold standard. In the period of the great depression in the 1930s, however, these countries were compelled once more to prohibit the convertibility of gold as well as the export of gold.3 During the depression, all countries devalued their currencies in order to increase exports, but because they all tried to increase exports by the same method, devaluation followed by retaliation repeatedly occurred. At the time of the rebuilding of the international monetary system after the Second World War, there was established what may in broad terms be regarded as a kind offixedexchange rate system (the system centring on the International Monetary Fund), although it was not the same as the pre-war gold standard. The intention behind this move was to avoid a repeat of the chaos and confusion which attended fluctuating exchange rates before the war. To begin with, (1) the IMF fixed the exchange rate between the US dollar and pure gold at 1-ounce troy of pure gold = $35. Next (2) the countries belonging to the IMF fixed their own exchange rates either in terms of the US dollar or gold (e.g. £1 = $2.8, $1 = ¥360) and registered these rates. Then to preserve these rates, (3) America undertook with foreign Monetary Authorities to buy or sell gold without limit on the amount at a rate of $35 = 1 ounce troy. (4) Foreign Monetary Authorities took on the obligation through their exchange banks to trade in the exchange market in dollars with no limits imposed when the exchange rate of their 3
In September 1931 Britain abandoned the gold standard, and in December of the same year Japan once more forbade the export of gold and announced an Order to End Gold Convertibility. America left the gold standard in 1933, and France finally abandoned it in 1936.
The effects offluctuationsin foreign exchange rates
107
own currency fell 1% below or rose 1% above the rate registered with the IMF.4 This was based on the assumptions that America held enough gold to fulfil its obligations, and that each member country had an adequate exchange equalization account or exchange stabilization fund. Therefore, when America ceased to convert gold in August 1971 due to an insufficiency in its supply, the system - the IMF principle of fixed exchange rates - collapsed. Therefore, there is at present no upper or lower limit to fluctuations in exchange rates. For most of the period after the founding of the IMF £1 equalled ¥1008 and $1 equalled ¥360, but by January 1982 £1 equalled about ¥420 and $1 about ¥220. This indicates that the yen was cheap relative to sterling and the dollar throughout the 1950s and 1960s, which would seem to have contributed to the growth in Japanese exports. Moreover, this state of affairs lasted for a long time because of the IMF's fixed exchange rate system, and during this time Japan built up a powerful exporting industry sector. By the time of the change to a fluctuating exchange rate system Japan's balance of trade had already improved remarkably and, as a result of the change, the yen suddenly emerged as a strong currency. Exchange rates and product prices
Similarly, a weak pound should bring about an increase in British exports. But according to what logic will this result come about? In what follows we shall assume that Britain's exporting industries are manufacturing industries (therefore industries whose prices are determined by the full-cost principle), and explain first of all how changes in the exchange rate exert an influence on prices in these industries. Next we shall consider how such price changes affect export and imports. If we take a single manufacturing industry, some of its products will be sold on the domestic market and others will be sold as exports to foreign countries. The prices of the products will be expressed in pounds sterling in the home market and in dollars abroad. Similarly, when the industry uses imported raw materials, it must convert the dollar prices of 4
Looked at from the British point of view the exchange rate £1 = $2.8 is the rate in terms of foreign currency (the rate per unit of one's own currency in terms of an amount of foreign currency). However, the rate of $1 = ¥360 is, in Japanese eyes, the rate in terms of their domestic currency (the rate per unit of the foreign currency in terms of an amount of domestic currency). In Japan exchange rates are usually expressed in the domestic currency, but in Britain usually in the foreign currency. The monetary authorities in Britain which deal in foreign currency have an obligation to the IMF to buy dollars without limit at the rate of £1 = $2.8 x 1.01, and to sell dollars without limit at the rate £1 = $2.8 x 0.99. The monetary authorities in Japan who deal in the domestic currency must sell dollars without imposing any limit if the rate goes to $1 = ¥360 x 1.01, and buy dollars if it goes to $1 = ¥360 x 0.99.
108
Determination of foreign exchange rates
raw materials in their countries of origin into pounds when calculating costs. Exchange rates will be used when translating sterling prices into dollar prices, and as stated in footnote 4 two kinds of method exist for doing this. The first is to express the value of the unit of the domestic currency (pounds sterling) in terms of foreign currency, e.g. £1 equals $2; the second is to give the domestic currency equivalent of a unit of the foreign currency, e.g. $1 = £0.5. Let us show the former by q, and the latter by r, with their dimensions as $/£ and £/$ respectively. Then q = 1/V. When q changes from £1 = $2 to £1 = $1.8, the pound weakens, because the pound which hitherto had a value of $2 will now have a value of only $1.8. If we show the same situation for r, $1 = £0.5 will become $1 = £0.55, and so if the pound weakens q decreases and r increases. Now let product prices be p£ and imported raw material prices be 77$. The reason we go out of our way to add the subscripts £ or $ to these symbols is to make it clear that these prices are expressed respectively in sterling and dollars. In terms of the full-cost principle formula, p£ = (1 + m)Q, we have to deal with average costs C£ in terms of sterling in order to calculate/^, but to do this we must translate the raw material costs 77$ expressed in dollars into raw material costs 77£ expressed in pounds. For this purpose the exchange rate expressed in terms of the domestic currency r will be used which has the dimension £/$. That is % = r7r%- Furthermore, products will be sold on the domestic market at p£ but will be sold abroad with their prices expressed in dollars. In translating p£ into dollars the exchange rate expressed in terms of the foreign currency q will be used; it has the dimension $/£. That is, the prices in foreign markets p% will be equal to qp£. If the pound weakens, r will increase as explained earlier. Thus the sterling price of imported raw materials which forms a part of costs will rise, but the other elements in costs will not change, or they will only increase at a rate below that of the rate of increase of r. Consequently the rate of increase in total average cost Q will be less than that for r, and therefore the rate of increase of the product price p£ will also be less than that for r.5 Since we have qr= 1, q will fall at the same rate as r 5
This may be proved as follows. The mark-up ratio is given by m, and average costs may be divided into three parts. The first part of costs C{ are British industrial products which are used again as materials etc. The second part of costs is C2 which are imports from abroad used as raw materials or fuels. The third part of costs C3 are other costs, namely fixed costs, wages etc. C3 is unchanged when r rises, and C2 will only increase by the same or less than the rate at which r increases, the former where it is impossible to substitute domestic products for imported raw materials and the latter where domestic raw materials are substituted for imported ones which have become relatively more expensive. (contd.)
The effects of fluctuations in foreign exchange rates
109
increases. Therefore, the rate of fall of q will be greater than the rate of increase of /?£, and so the prices p$ = qp£ of those products on the foreign market will fall. That is to say, if the pound weakens the domestic price Pi of the products of British manufacturing industry will rise (so creating inflation), but these products may be sold more cheaply abroad. The effect on exports and imports of a devaluation of the pound Let the domestic demand for industrial products be DD, and foreign demand be DF. If we make DM the demand for similar products imported from abroad, of the total domestic demand DD, DM will be made up of foreign goods, and the remainder DD — DM will be made up of domestically produced goods. The sterling value of exports from these industries will be p£DF, and the value of imported manufactured goods will be rp'%DM and rir%k [(DD — DM) + DF] will be the value of imported producer goods. Here /?$ and TT$ are the foreign prices of imported manufactured products and imported producer goods, respectively, and k is the volume of imported producer goods needed to produce one unit of product.6 Where imported producer goods cannot be substituted for by other producer goods which can be domestically supplied, they are known as non-competitive imports, and in such cases k will not change. As against this, where the sterling price rrr% of imported goods has risen because of an increase in r, and, as a result, there has been a switch to domestically supplied producer goods away from imported goods, these imported producer goods are known as competitive imports. In the case of competitive imports k declines when r increases. The price of all British industrial products will be affected by an increase in r. However, let us consider those most strongly affected; that is, those industries where the rate of increase of product prices is highest. Let the rate of increase be h, and let us examine whether it is possible for h to be equal to or higher than the rate of increase r. In the equation the left-hand side will increase by the rate h, but on the right-hand side C2 will only increase by the same rate as r increases or at a lower rate. C3 will be fixed. Therefore in order that h be equal to or higher than the rate of increase of r, Cx must increase by a rate greater than h. But the rate of increase of Cx will depend on the rates of increase of British industrial product prices, and it will not be possible for Cx to rise by a rate higher than its maximum rate h. Were we to suppose a 'rate of increase where h^f a contradiction would arise, so the rate of increase will be h
110
Determination of foreign exchange rates
The balance of payments for exports and imports, that is the balance of trade, will be given for manufacturing industry by PiDF
- rp'%DM - rir%k[(DD - DM) + DF]
As we saw before, if r rises the domestic pricep£ will rise at a slower rate than the rate of increase in r; exports prices qp£ will fall and import prices rp'% for these products will rise at the same rate as r. Thus the total domestic demand for products DD will decrease (since the price of both domestic products and imports will rise), and the demand for imports DM will also decrease (since the price of imports will rise by a rate higher than that for domestic products). But the demand for exports DF will rise (since the export price will fall). We can conclude from this that the first term in the above equation (the value of exports) will increase. However, it is not possible to judge whether the second term (the value of imported finished products) and the third term (the value of imported raw materials and other producer goods) will increase or decrease. It is therefore difficult to say whether the balance of trade will improve or worsen. However, if we re-write the above equation as follows [Pi ~ rir%k] DF - r\p'% - 7rsk]DM -
nrskDD
we can infer that the first term will probably increase and the second and third terms probably decrease. To begin with the first term, as we stated in note 5, letting Q be that part of costs which results from using producer goods which are produced by British industry, C2 be that part of costs resulting from the use of imported producer goods, and C3 costs other than these, such as wages etc., then we can write the full cost formula as/?£ = (1 4- m) (Q + C2 + C3); but when we consider that r7r$fc = C2, the inside of the square brackets of the first term in the above formula for the trade balance is Cx + m(Cx + C2) + (1 + m)C3. C3 did not change when r increased, but Cx and m{Cx + C2) may well increase.7 Thus what is inside the square brackets will probably increase, and DF too will increase, and therefore we may regard the whole of the first term as likely to increase. Next, the inside of the square brackets of the 7
Let Q + C2 before the increase in r be C^ + C§, and let it be C\ + C\ after the increase in r. Let the mix of imports and domestically made producer goods be fixed at the combination they attain after the rise in r, and let C\ + C\ be this combination evaluated at the prices and the exchange rate ruling before the rise in r. Because the increase in r increases domestic product prices, C} + C\>C\ + C\. Therefore if C\+C\&C\ + C\, we obtain C\+C\>C\ + C\, and the combination of imports and domestically produced producer goods before the increase in r will not have been the most efficient combination. Therefore, C\+C{2
The effects of fluctuations in foreign exchange rates
HI
second term will be unchanged if imported producer goods are noncompetitive imports, but will increase if they are competitive imports. However, this increase and the increase in the sterling value of imports due to a rise in r will be cancelled out by a decrease in the volume of imports DM of finished products, and the total value for the second terms as a whole may well decrease. The reason is that, although the price of domestically produced goods will rise at a rate which is less than the rate of increase in r, the prices of imports will rise at the same rate as r; therefore there will be a large switch in demand from imports to domestically produced goods and DM will decline very elastically. As for the third term, r will rise, but k will be unchanged or will decrease, and DD will also decrease. Unlike the demand for agricultural products, we may regard the demand for industrial products as being very elastic with respect to price, and therefore the changes in DD will probably dominate the whole in the third term. Therefore, the third term will probably also decrease. That is, we can conclude that if the pound weakens and r rises the trade balance (in the above formula) will probably increase. The workshop for the world or an empire of finance Thus a weak pound will increase exports and improve the balance of payments. Yet Britain has held firmly to a strong pound since the First World War. At the time the value of Britain's overseas investments vastly outstripped that of other countries, and rather than the 'workshop of the world', Britain had become a great Imperial nation dominated by finance capital. British capitalists who had continuously supplied funds to all parts of the world since the end of the nineteenth century wanted the interest they received for the loan of these funds to be large. As we shall see later, the exchange rate for the currency of a country which has a strong overseas investment sector and a good balance of payments on capital account will have a tendency to be set at a rate higher than the rate appropriate to produce a balance in trade. In addition to this the British rate of interest was kept high. This increased the attraction of the pound, and since people wished to save in the form of pounds rather than in the form of other currencies, the desire to swap other currencies for pounds was strong, and the exchange rate was decided in a way which was advantageous for sterling. The exchange rate for the pound (the rate, q, expressed in foreign currency) was thus kept high, but for that reason exports were held down and the balance of trade worsened. Moreover, the massive funds gathered by reason of high interest rates were not invested in domestic industry, but were invested abroad. Since exports were depressed and investment
112
Determination of foreign exchange rates
insufficient, the domestic economy stagnated and unemployment became chronic. Roughly the same state of affairs continued from the First World War until the beginning of the 1930s. Those who profit by high interest rates are the rentiers who live by interest. On the other hand those who clearly lose by high interest rates are industrialists, exporters (those who suffer because of the strong pound brought about by high interest rates), and workers (those upon whom is inflicted massive unemployment due to inadequate investment and exports). These were the people who were the support and driving force of Imperial Britain, the 'workshop of the world'. Given the choice between 'the workshop of the world' and 'an Empire of Finance', Britain chose the latter, and the prosperity this brought to the City brought decline and depression to industry. It was Keynes who suggested removing the pressures from those who live on interest and cutting the chains of high interest by which Britain was bound, and Keynes' thinking long held sway among economists until the policy of 'high interest rates, a strong pound, high unemployment' was revived by the Thatcher Cabinet in 1979-1981 under the name of Monetarism.8 3 How exchange rates are determined (I)
The inter-bank market In the market for foreign exchange, unlike stock markets and commodity exchanges, there is generally no special building used, and no special institutions exist whose purpose is to deal primarily and in large amounts in foreign currencies. Therefore exchange rates are not determined by auction. There was once a foreign currency exchange in the old London Royal Exchange, but it was abolished in 1921 and at present the foreign exchange banks in each country trade amongst themselves in foreign currency by means of the telephone and the telex.9 In any capital city there are many main and branch banks which 8
9
Keynes' problem was a problem of the conflicting interests of two camps; the camp of those who live off interest (rentiers) and the camp composed of industrialists (entrepreneurs) and workers. In Marx the rentiers and entrepreneurs were considered to be one and the same person, and the conflict of interests was considered to be between the capitalist (rentier-entrepreneur) and the worker. Keynes drove a wedge between the rentier (saving) and the entrepreneur (investment). Even at present, in European cities like Paris and Frankfurt, there is an area in the corner of the securities exchange where currency transactions take place (exchange bourse), but on the European continent too the status of the bourse is declining. The central bank participates in the setting of exchange rates on the bourse, and the rates fixed there are preliminary or basic rates in deals with customers. But because the bourse and the outside world is connected by telephone, the principle of one price for one good is almost wholly solid in this case, there being scarcely any difference between the bourse price and that outside (for example, the inter-bank exchange rate).
How exchange rates are determined (I)
113
handle foreign exchange and they form the world market, usually keeping in close communication with the foreign exchange banks and currency brokers in the other cities of the country and in cities abroad. Despite the fact that a world currency exchange nowhere exists as a legally incorporated organization, exchange rates are adjusted rapidly, moment by moment, in response to the prevailing demand. Moreover, since London, New York, San Francisco, Tokyo, Singapore and Dubai have their nights at different times, the world market is always open somewhere around the world. Thus when morning comes in Tokyo, and the windows open in the foreign currency section of banks at 9 a.m., the bank is instantly able to know all of the prevailing exchange rates on the world market, and despite a night when no business was conducted, the bank can begin work right away furnished with almost complete information about the state of the market.10 The reason why auction does not take place in exchange markets is that, before the banks open each day, they have already obtained almost complete information on the current world equilibrium prices, and there is thus no need to discover the equilibrium price through an iterative process. The two-country case Let us now assume the world consists of two countries: America (the dollar zone), and Britain (the pound zone), and that the pound does not circulate in America, and people do not hold dollars in Britain. When an exporter a in the American zone requires payment for the price of his export goods, he issues a bill of exchange addressed to the British importer b (a collection bill) which states: 'pay American bank A £10000 (the equivalent amount in dollars) as the price of exported goods', and takes it to bank A attached to the shipping documents which prove he has sent the goods.11 A sends the bill and these documents to a 10
The distance between Tokyo and London, or Tokyo and San Francisco is too great and so when the banks close in Tokyo, the banks in London have not yet opened. (It is night-time in America and Canada.) The reason why Hong Kong, Singapore and Dubai play important roles in international finance is because they relay the Tokyo market to the London market, in addition to the fact that they have their own important roles in international trade. In addition, after the banks have closed in San Francisco there is a short period before the banks open in Tokyo when the international exchange market is open nowhere, and if a great event occurs unexpectedly during that interval the Tokyo market will be thrown into confusion when it opens. 11 This kind of bill of exchange is known as bill receivable in foreign currency. As against this a bill of exchange which receives payment of a fixed sum (for example, 20000 dollars) of the domestic currency (dollars) is known as bills receivable in the domestic currency. If b who has received a bill of exchange designated in pounds, pays the number of pounds specified, the bank will change them into dollars at the market rate and pay the amount to the issuer of the bill; but in the case of a bill of exchange (contd.)
114
Determination of foreign exchange rates
British bank B. B collects £10000 in exchange for these from b. If for the sake of simplicity we assume that bank A is the only bank in America and bank B the only one in Britain, all of the foreign exchange bills issued by American exporters a, a', a", . . . will be collected together at B, and similarly the foreign exchange bills issued by British exporters will be collected at A. Two problems now arise. The first is how many dollars bank B must pay to bank A as the sum in dollars which corresponds to the £10000 received from b; that is the problem of how the rate of exchange between pounds and dollars is determined. The second is by what means bank B will exchange these £10000 for dollars and how it will send them to bank A. These two problems are intimately related. Let the total amount of pounds to be changed into dollars that have been collected at bank B be YA, and the total amount of dollars to be changed into pounds that have been collected at bank A be Z 5 . I indicate the number of dollars, Y the number of pounds and the subscripts A and B the banks to which the money is to be sent. As we shall see later, where there is a third currency - the 'Yen' - the total number of pounds collected at bank B will be divided into YA and Yc, and the former will be changed into dollars and sent to bank A, and the latter changed into yen and sent to bank C in the yen zone. The sum total of dollars X will also be divided into XB and Xc, and sent to banks B and C respectively. The sum total of yen Z which has been collected at bank C will be divided into ZA and ZBy the former will be changed into dollars and the latter into pounds and they will be sent to banks A and B respectively. The role of dealers
The exchange of dollars and pounds will take place between banks. Each foreign exchange bank has a dealing room where dealers engage in the exchange of currencies by sending telephone or telex messages to each other. In our case bank B's dealer sells YA pounds and buys specified in the domestic currency (dollars), b must pay in pounds which, at the market rate, are the equivalent of the specified dollar amount on the bill. Besides Collection Bills (Bills Receivable), there are remittance drafts or bills which specify acceptance of the sum of money sent to bank B (which orders bank B to pay the bearer of the bill of exchange that same amount), and in this case too there are deals both in foreign and in the domestic currency. In the case of deals in a foreign currency the issuer of the bill bears the exchange risk - that is a profit or loss due to exchange ratefluctuations- but where deals are made in one's own currency the other side bears the risk. There is a tendency for export prices to be fixed in terms of the foreign currency when the export competitiveness of a country's industries is poor, and in one's own currency when it is strong. In what follows we shall continue the explanation chiefly in terms of bills receivable in the foreign currency.
How exchange rates are determined (I)
115
dollars, and bank A's dealer attempts to sell XN dollars and buy pounds, but what makes it certain that A and B will, respectively, be sellers of dollars and pounds is the assumption that bank A only handles exchange transactions designated in dollars and B handles only those designated in pounds. In reality A may perhaps sell a demand draft designated in pounds (from country B to country A) and try to buy dollars, and for the same reason B too may be a seller of dollars and a buyer of pounds. The dealers have to cope with this demand and supply, but the complication stemming from the fact that there is no clear distinction between those demanding dollars and those demanding pounds is not a difficulty of principle. Now just as competitive trading in the various exchanges takes place according to the rules of those exchanges, so the dealers in the various banks conduct exchange transactions in accordance with certain rules.12 Bank A's dealer (to be known in what follows simply as A except in cases of possible confusion - bank B's dealer will be referred to as B) will call up B, who is expected to be selling pounds, by telephone or telex, and enquire the price from him. If B replies 'from 1.95 to 1.97', this means that B will buy pounds at 1.95 dollars per pound and will sell pounds at 1.97 dollars. When B replies in this fashion he is simultaneously accepting an obligation to buy pounds at 1.95 dollars per pound if that is the price, and to sell pounds if the price is 1.97 dollars per pound, up to a fixed upper limit designated in dollars (for example, 1 million dollars).13 A is attempting to buy pounds, but because B does not know whether A is a buyer or seller of pounds, B must tell A both the selling and the buying price for pounds. B will probably try to get as high a selling price and as low a buying price as he can, but there is a customary agreement that the difference between the selling and buying prices must not exceed a given limit (for example, 2 cents). In our example, there are only the two banks, A and B, but because in actual exchange markets hundreds of banks participate, banks which break conventions lose trust and are forced out of the market. Therefore custom is strictly observed. If a high selling price is set then the buying price too must be set high, but with a difference not exceeding two cents; and if the buying price is low, a low selling price must also be quoted. In the case where A is a seller when B quotes a buying price which is too high, B will be confronted with the dilemma whereby he has to buy pounds at a higher 12 13
Dealers are regular employees who get their salaries from the bank, and who do not attempt to obtain profits themselves from the exchange transactions they handle. There is, however, no obligation on A, who enquired about the prices, that he sell pounds or that he has to buy them.
116
Determination of foreign exchange rates
price than the market, and conversely, if he quotes a selling price which is too low, he will have cheap pounds bought from him by A. Thus the buying price must not be too high (and therefore the selling price which is two cents higher must not be too high), and the selling price must not be too low (and therefore the buying price which is two cents lower must not be too low). For these reasons B cannot quote too high a selling price and too low a buying price, despite his desire to sell as dear and buy as cheaply as he can. Based on the information he has gathered, B must thus quote the selling and buying prices which he thinks are the current market prices. The principle of clearing orders by the end of the day Let us assume that A has bought £500000 at 1.97 pounds per dollar (that is, for a sum of $9850000). The total of pounds B is trying to sell will decrease to YA — £500000, and the dollars A is trying to buy will decrease to XB — $9850000. When A or B next telephones, the price of the pound may have risen or fallen reflecting this new state of demand and supply.14 During the day new demands for pounds may find their way into bank A in the form of requests to change dollars into pounds, and so XB is likely to increase. Similarly, new orders to change pounds into dollars may be received by bank B and therefore YA too may well increase. Letting the values of B's successive sales of pounds be Y\, Y\, . . ., the values of A's successive purchases of dollars be XB, X2By . . . and the exchange rate for successive transactions be /?$/£, /?$/£, . . . (dollars per pound), then we have: There is also the opposite case where the dealer of bank B calls up A to buy dollars. Here the following is valid: Since the inverse of pounds per dollar pi/% is dollars per pound p%/i (that is P%/£ = 1//*£/$) > w e c a n a l s o write the above equation as 14
In order to simplify our explanation we assume that A demands no dollars and B no pounds, but in reality every bank demands and supplies both dollars and pounds. If A receives a large number of orders to buy dollars at the quoted exchange rate it will judge that there is a heavy demand for dollars on the market and may well raise the dollar price; and conversely if there is a large number of orders to sell dollars A will judge that there is a plentiful supply and that the price of the dollar ought to go down. The reason for this is that unless A ensures no further orders arrive by raising (lowering) the price, it will have to buy or sell more dollars than is necessary, when orders to buy (sell) are numerous.
How exchange rates are determined (I)
117
Dealers are directed to dispose of orders received on any given day during that day, and therefore, the sum of XlB, X}B, . . ., XlB, XlBx, . . ., made during the day will be equal to the total demand for pounds on that day XB (since this will include additional orders it will be larger than the orders XB which were initially received). Similarly, the total of YlA, Y\, . . ., 1^4, Y'/\ . . ., will be equal to the total demand for dollars on that day YA.15 Therefore we have Here p$/£ will be the average price during that day for pounds in terms of dollars. This will be high when the total demand XB for pounds is great, and it will be low when the total demand for dollars YA, forthcoming in the form of supplies of pounds seeking dollars, is great. The price of the pound /?$/£ will rise and fall during the course of the day, but its average value will be regulated by total demand and total supply. We have assumed up to now that there were only banks A and B in the respective dollar and pound zones, but there is no change in principle if there are banks A, A', A", in the dollar zone, and B, B', B", in the pound zone. Bills of exchange issued by exporters in the dollar zone will be collected at banks B, B', . . . in the pound zone, and bills issued by exporters in the pound zone will be collected by A, A', . . .. Banks must make telephone calls and transmit telexes to all quarters in order to match up the demands for dollars and pounds generated by these bills and to validate transactions. They may also get the help of foreign exchange brokers - who intervene between foreign exchange banks or between banks and customers (for example importers) and take a commission - who speedily effect the successive transactions. Of course the banks too charge a commission on the foreign currency they handle. Banks will probably carry out the transfer of the pounds and dollars they have purchased as follows. Suppose a foreign exchange bill of collection for £10000 is sent from bank A to bank B. B will hand over the bill (and the shipping documents) to the recipient b in exchange for £10000, but let us assume the £10000 has been sold to bank A'. In that case B will instruct A' to pay $19700 to A, the recipient of that bill of exchange, rather than to B. Similarly, A' will probably instruct B to pay the £10000 B has bought to the recipient B' of the corresponding bill. All the bills of exchange issued will thus be settled, and the prices paid by the importers will be received by the exporters in the currencies of their respective countries.16 15 16
Of course, there are exceptional cases where pounds or dollars left unsold are brought forward to the next day. See Additional Note f.
118
Determination of foreign exchange rates
4 How exchange rates are determined (II) The three-country case Let us now assume the world is made up of three zones: the dollar zone, the pound zone and the yen zone, and that in these zones there are respectively banks A, B and C. We assume that bank A is trying to exchange a total of X dollars into foreign currency (pounds or yen) in a single day, and that of this sum XB dollars are to be exchanged into pounds and Xc dollars into yen. On the same day bank B exchanges a total of Y pounds into dollars or yen (YA pounds into dollars, Yc pounds into yen), and bank C exchanges Z yen for dollars or pounds (ZA yen into dollars, and ZB yen into pounds). Where there are three currencies three exchange rates will be established between dollars and pounds, between dollars and yen and between pounds and yen, but thefirsttwo are the important ones, that is /?$/£, /?*/$. The last exchange rate, between pounds and yen /?¥/£> ma Y be calculated according to the arbitration formula from the current p$/£ (basic rate) and /? ¥ / $ (cross rate) as />¥/£=/>¥/$/>¥/£
(1)
For, if the actual yen-to-pounds rate is smaller than the value calculated in this way, it will be more profitable to convert pounds into dollars and then dollars into yen rather than directly exchange pounds into yen. Conversely, if the actual /? ¥ / £ is higher than the value given by the formula, it will be advantageous to do a roundabout exchange from yen to dollars, and from dollars to pounds rather than directly exchange yen into pounds. Therefore, in order to bring about direct exchange between the yen and the pound the rate p ¥ / £ must satisfy the arbitration equation. This rate is known as the arbitration rate. When bank B remits YA pounds to bank A, it can either convert the amount directly into dollars and send it like that, or it can send it by the indirect method of converting it first of all into yen and then into dollars. Let YAA and YAC be the amounts of pounds sent by these respective methods. Similarly, when bank B remits Yc pounds to bank C it can either do so directly (to the amount of Ycc), or else via dollars (to the amount of YCA). Naturally YA = YAA + YAC, Yc = Ycc + YCA
(2)
If the same symbols are used for bank A remitting dollars to banks B and C, and bank C remitting yen to banks A and B, then we will have XB
~ XBB + XBC, Xc = Xcc + XCB
ZA
=
%AA + ZAB* ZB = ZBB + ZBA
(3) (4)
How exchange rates are determined (II)
119
There are three markets which exchange (i) dollars and pounds, (ii) dollars and yen and (iii) pounds and yen, respectively. In market (i) the pounds which need to be converted into dollars will meet the dollars to be converted into pounds; the former can be expressed in terms of pounds as YAA + YCA+P*/*ZAB
(5)
while the latter in terms of dollars as XBB
+ XCB + P%/ ¥ %BA
(6)
The first term of (5) represents the amount of pounds converted directly into dollars, the second the amount of pounds converted initially into dollars for subsequent conversion into yen, and the third term represents the amount of pounds which has been converted from yen with the intention of being further converted into dollars. Expression (6) can be construed in the same way. The dollar-pound exchange rate is determined so as to equalize the demand for dollars (5) and the supply of dollars (6). Hence P$/i(YAA +
YCA
+ P£/*ZAB)
= XBB + XCB + P$/*ZBA
(7)
Similarly in markets (ii) and (iii) respectively we get P$/¥ (ZAA + ZBA + P*/£YAC) = XcC + XBC + P$/£YCA ZAB + p M/%XBC) = Ycc +YAC + P£/$XCB
(8) (9)
In view of the fact that /? $ / ¥ = l/p ¥ /$ and p £ / ¥ = l/p¥/£> the arbitration formula (1) enables us to rewrite (7) as P$/i( YAA + YCA) + p $ / ¥ ZAB = XBB + XCB + p $ / ¥ ZBA
(7')
If we substitute for (2), (3) and (4) in this formula and consider (9) and the arbitration formula, then (7') can ultimately be written as P%,i{YA + Yc) = p$/ ¥ Z B + XB
(10)
In exactly the same way p$/*(ZA + ZB) =P%/iYc + Xc
(11)
can be obtained from (8) and XB + Xc = P$/£YA + p%l*ZA
(12)
from (9). As will immediately be seen, these three formulae are not independent of each other. Given two of them - for example (10) and (11) - the third equation, (12), automatically follows.
120
Determination of foreign exchange rates
How are the basic and cross-rates determined? This means that if two of these three markets (i), (ii) and (iii) are in equilibrium, the remaining market will also be cleared. We will now concentrate our attention on the dollar-pound exchange market (i) and the dollar-yen market (ii), and attempt to view the pound-yen exchange market as a reflection of these first two markets. Providing demand and supply in markets (i) and (ii) moment by moment counterbalance each other, then (10) and (11) will obtain, also moment by moment, hence the value of pounds and yen expressed in terms of dollars -p%/£ and p $ / ¥ respectively - will be determined so as to fulfil both equations. For one whole day we therefore end up with ps/£X(YA + Yc) = p$,*2ZB + XXB
(13)
p$/*X(ZA + ZB) = p$/£XYc + XXf
(14)
where p%/£ and p $ / ¥ represent the average value through one complete day of the moment by moment prices, p%/£ and p £ / ¥ , but strictly speaking p%/£ and p $ / ¥ in (13) are not equal to those in (14). This is because in (13) the average p%/£ is calculated by taking the distribution of YA + Yc as the weights, whereas in (14) the average is obtained using the distribution of Yc only as the weights. The same is true of p$/ ¥ . We will below disregard the differences in average prices resulting from such differences in weights. Since during the course of a day bank B, while converting pounds into dollars, must send only YA to bank A, YA must be equal to XYA.17 Similarly for Yc, XB etc. we end up with Yc = XYC, XB = XXB and so on. Hence (13) and (14) can be written as YC)=XB+P%I*ZB
(15)
ZB) = Xc + P$/iYc
/
(16)
Similarly from (12) we arrive at ZA
(17)
If we divide both sides of the equations (15) and (16) by YA + YB and ZA + ZB respectively, we obtain the following (15') / where
/ y A~*
17
(16') y
*C
*A1-IC
Y ^
~T ^B
v ^A~r
^B
YA is the total volume of pounds which bank B has to exchange for dollars in a given day, and XYA is the total volume of pounds exchanged for dollars in the same day.
How exchange rates are determined (II)
121
P%it
Measuring the price of the pound p$/ £ on the horizontal axis, and that of the yenp $ / ¥ on the vertical axis, (15') and (16') can be shown by straight lines rising to the right. These are, of course, the partial equilibrium curves for the pound and the yen. The slope coefficient of (15') looked at from the horizontal axis (i.e. 1/6) is greater than that of (16') (i.e. d), and hence the lines will certainly intersect. The intersection represents the prices (general equilibrium prices) which equate the demand and supply of the yen and the pound (and thus of the dollar). (See Fig. 22.) The following may be discovered from the figure. When the demand for pounds XB by bank A increases, the straight line (15') will move to the right, and the point of the general equilibrium will move from p° to p1. Thus we can obtain the three laws with which we are familiar, (i) The average price of the pound expressed in dollars will rise, (ii) The average price of the yen will also rise, (iii) The rate of increase in the price of pounds will be higher than that of the yen. Similarly, when demand Xc for yen due to the dollar increases, not only the price of the yen but also the price of the pound will increase, and the rate of increase in the former will be higher than that of the latter.
122
Determination of foreign exchange rates
Next, where the demand for dollars YA stemming from the pound increases, a and b will decrease. In response to this the dollar price of the pound and the yen will fall, but the rate of decline in the price of the pound will be the greater. Similarly, where the dollar demand ZA stemming from the yen increases, the same sort of decline in prices will occur, but in this case the rate of decline in the price of the yen will be the greater. Lastly, when the demand for the yen Yc stemming from the pound increases, a, b, will decrease and d increase. But it is shown (i) that both ad and bd increase, and (ii) that a + be decreases at a greater rate than the rate of decline in (1 — bd). It is then seen that/?$/£ falls and/?$/¥ rises (try solving (15') and (16') with respect top$/ £ and/?$/¥ to confirm this result). Conversely, where the demand for pounds ZB stemming from the yen increases, p$/ ¥ will fall, while /?$/£ will rise. Equilibrating operations The governments and central banks of all countries use these laws of comparative statics to try and stabilize their foreign exchange rates. All countries establish an exchange equalization account or an exchange stabilization fund, and intervene in the market through exchange banks and exchange brokers. For example, if the price of the pound in terms of the dollar falls too much, Britain's 'exchange equalization account' releases the dollars it holds onto the market and buys up pounds, thus raising the dollar price of the pound. Again, if the price of the yen in terms of the dollar becomes much too high, Japan's exchange equalization account will release yen and buy dollars to prevent an increase in the price of the yen. All these measures are price operations designed to raise XB and ZA, respectively. In the real world economy, exchange operations take place mostly in exchange against the dollar, and there is almost no intervention as regards other currencies (in Britain as regards the yen, or in Japan as regards the pound). 5 Changes in exchange rates
The automatic balance-of-payments adjusting function of exchange rates Let us consider the following situation. Suppose Britain finds her exports are depressed and her imports excessive, whilst Japan's exports are flourishing in comparison with her imports. According to the model of the previous section, this means that the demand XB for pounds from the dollar zone to pay for exports to that zone, and the demand ZB from the yen zone, will both be small; in contrast, the demand for
Changes in exchange rates
123
yen from the dollar zone Xc and from the sterling zone Yc will be great. Then by virtue of the comparative statics laws derived in the previous section, the price of the pound in terms of yen, p%/Jp%/^ = /?¥/£> wiH be fixed at a lower level than before. (It will probably happen that/?$/£ falls and /? $ / ¥ rises.) This situation is extremely desirable for British exporters, and due to the cheap pound British exports will be comparatively cheap in Japan and are likely to be given a boost. On the other hand, Japan's exporters will have difficulties with their exports to Britain due to the expensive yen. Since imports will become comparatively expensive in Britain and comparatively cheap in Japan, competitive imports are likely to decrease in Britain and increase in Japan. On the other hand, non-competitive imports will increase in Britain but their rate of increase is likely to be lower than the rate of increase in her volume of exports,18 while in Japan non-competitive imports are likely to decline at a rate slower than the rate of decline in the volume of exports. Total imports to Britain, including both competitive and non-competitive imports, are likely to increase, and those to Japan are likely to decrease, and Britain's exports will increase over and above this increase in imports, while Japan's exports will decline over and above the reduction in imports. Thus, the ratio of the total sterling value of British exports to the total value of her imports will increase when p%/l falls. That is, the British balance of trade will improve. Similarly, the ratio of the total yen value of Japan's exports to the total value of her imports will decline when p $ / ¥ increases, so that there will be a fall in the surplus in the Japanese balance of trade. Even if the British balance of trade were to worsen while Japan's continued to be favourable, as long as the pound was correctly devalued and the yen revalued in response to this state of affairs, the cheaper pound would rectify the British balance of trade and the dear yen would reduce Japan's surplus. That is to say, changes in the exchange rates have the effect of eliminating both extremely unfavourable and extremely favourable trade balances. Yet for this adjustment mechanism to work effectively it is necessary for exports to change elastically when the exchange rate falls. If this assumption is not fulfilled, the British balance of trade will have little improvement and the Japanese surplus will hardly be removed; py£ will continue to fall and /?$ / ¥ will continue to rise. Yet if this process continues then British products will be as good as free and Japanese 18
Non-competitive imports will change in proportion to changes in a country's GNP. When the volume of exports changes, a change in GNP is induced, but the rate of the latter change is smaller, in absolute value, than the rate of change in the volume of exports.
124
Determination of foreign exchange rates
products prohibitively expensive. Ultimately the demand for British exports and the demand for Japanese exports must become elastic with respect to price. Thus in the end the self-adjustment mechanism of exchange rate fluctuations will begin to function, but the situation would be a tragic one if by that stage British industry had completely lost its ability to export. Purchasing power parity theory As we have already stated, it is the dealers of foreign exchange banks who are in the front line in the work of determining exchange rates, and they are directed as a matter of principle to make sure all demand and supply accepted on a given day is satisfied during that day. That is, they buy up in a given day all the orders to buy placed that day, and they sell off what they have to sell. The result is that equilibrium conditions for demand and supply such as (15), (16), (17) are valid each day for all currencies (dollars, pounds, yen). Whilst rates will rise and fall quite violently during the day, the average values of p%/i and/? $ / ¥ through the day will be determined so as to satisfy these equilibrium conditions. However, where these average rates are too high or too low, exports and imports will change elastically, and the flow of international payments will also change in response. Where the rate was too high (too low) it will be revised down (up), but we may well ask what is the basis for the judgement that the rate is too high (or too low); it is obviously this judgement which regulates the movement of foreign exchange markets through its influence on exports and imports. To discuss this problem we shall have to take into consideration forward markets in foreign exchange which we have hitherto ignored, and consider, too, the differences that may occur between rates of interest in different countries. But even at the present stage where we have limited discussion to the spot market, we can discuss which rate is too high and which too low with some degree of clarity. It is the purchasing power parity theory which deals with this question. In Europe, where intercourse between neighbouring countries is frequent and costs relatively little, everyday phenomena are explained by the purchasing power parity theory. Let us assume M = (ml5 m2. . .) to be the quantities of the various goods consumed by the standard Briton, and M' = (raj, m'2 . . .) to be the quantities consumed by the standard French person. Since there is no great difference between the way of life of the British and the French, the composition of M (the ratios m2/m1, m3/mu . . .) will closely resemble that of M' (the ratios m'Jm'i, m'3/m'1, . . .). However, when comparing the Japanese and the British, the composition of the standard bundle of goods or services
Changes in exchange rates
125
consumed, M, may be quite different. Even comparing the British and the French, the composition is, strictly speaking, different - for example, if we let good 1 be wine, the weight of m[ in M' will be higher than that of mx in M - and there may perhaps be some noticeable large difference in the absolute level of M and A/'. The French, for example, may eat well, drink well and generally enjoy themselves well. Let PB pounds be required to live in Britain at a standard M, and let the same lifestyle cost PF francs in France. The French may well make the same calculation, so that if it costs P'F francs for a standard of living M' in France it would cost PB pounds in England. Let us assume one pound exchange for pF/£\ it is clear that when we exchange one pound for francs it is possible to live in France at the level oipf/JPF of living M, while it is only possible to live at the level of l/PB of the same living for that one pound. If the former is the greater, that is if we have
p
T
(18)
the British will go shopping in France, or live in France and come to work in London. Conversely, the condition required for the French to go shopping in England is <
p,
p,
given that p£/F is the price of francs shown in pounds; it is the inverse of pFji - the price of the pound expressed in francs. If Britons go shopping in France pounds will be converted into francs, the supply of pounds will increase, and the price of the pound shown in francspF/£ will fall. Similarly, if the French go shopping in Britain, p£/F is likely to fall, and this means that pF/£ will rise. Thus, if the peoples of the two countries are both simultaneously engaged on shopping expeditions, forces tending to raise the exchange rate pF/£ will operate simultaneously with forces tending to lower it, and the rate can remain unchanged. However, in cases like those of France and Britain where lifestyles differ little and the composition of M and M' is almost the same, then even if there is a discrepancy in the absolute level, the validity of (18) implies the invalidity of (19), and vice versa. That is, if (18) is valid only Britons will go on shopping trips abroad (France) with the result that pF/£ will fall. If conversely (18) holds with the reverse
126
Determination of foreign exchange rates
inequality, only the French will go shopping abroad (to Britain) andp F/£ will rise. Thus ultimately pF/£ will settle down to a value, where
Since l/PB shows the purchasing power of a single pound and \/PF the purchasing power of a single franc (for the British), the above theory, which asserts that the exchange rate settles down to where it is equal to the ratio of the purchasing powers of the two currencies, is known as the Purchasing Power Parity Theory.19 In the above example where consumers themselves go out directly to buy, purchasing power was at the level of retail consumer prices, but for exporters and importers the level of purchasing power is likely to refer to the level of wholesale prices. Moreover, for importers who only handle a specific commodity (like motor cars), the purchasing power of the pound is merely the inverse of the price of the car. Let commodity 1 be cars and let the sterling price of a British-made car be pXB and the yen price of a Japanese-made car be pu. The conditions for a British importer to import cars from Japan are />¥/£
1
PlJ
PlB
Similarly for other goods, if the purchasing power of the yen and pound are compared good by good and more of a good can be bought by converting pounds into yen and buying in Japan rather than by buying the goods directly in Britain, then the good will be imported from Japan. In the opposite case, Japan will import goods from Britain. Having said all this, let us divide goods into three groups. Group I are those goods which are in demand in both Britain and Japan (automobiles, tape recorders, etc.); Group II consists of goods which the Japanese demand a great deal of, but for which the British have hardly any demand (fish sausage, kimonos, tatami flooring, etc.); and Group III contains those goods which the British demand a great deal of, but for which the Japanese have hardly any demand (smoked herring, cheese, salami, Western-style furniture, etc.). For goods in Group II and III respectively, the following are likely to be valid
P*l±>±PU 19
(II)
PiB
The Purchasing Power Parity Theory is old and can be traced back at least to Ricardo and Mill, but it was Cassel who was its most powerful advocate and who gave it its name. See Cassel, G., Money and Foreign Exchange after 1914, 1923.
Changes in exchange rates PU
127 (III)
PiB
This will be because fish sausage made in Britain will be expensive, and the price of Japanese-made salami will also be comparatively high. Let us assume for the time being that for all goods in Group I the following holds: P
-^>±
PU
(I)
PiB
Looked at from the point of view of purchasing power, Britain will import goods of Groups I and II, and Japan will import goods of Group III. Therefore forces which tend to raise the exchange rate and forces tending to lower it will operate simultaneously. If both forces are strong, exporting and importing between Britain and Japan will be carried out under comparatively stable exchange rates. However, in the areas where lifestyles are very different as between Britain and Japan, British people will import almost no goods of Group II even if the terms of trade are favourable, and Japan will hardly import any goods of Group III. Only goods of Group I will be exported and imported, and in that area Britain alone will be the importer. The price of the pound is likely to fall quite markedly, but as long as the purchasing power relations of (I), (II) and (III) continue and the Japanese lifestyle does not change and thus generate a large demand for goods of Group III, JapaneseBritish trade will remain one-sided for a considerable period. However, a continued decline in price of the pound, p#/£, will eventually violate the conditions (I) and (II). Trade war For the above discussion we have generally ignored elements in the international balance of payments other than the balance of trade. However, the international balance of payments is made up of trade and non-trade balances of payments, the international balance of payments of capital account, and that of cash account. Thus what governs exchange rates is the entire international balance of payments excluding its cash accounts, and certainly not merely the balance of trade. Consequently, as in Britain for example, where the balance of trade is bad but the country is one whose non-trade balance of payments and capital account are comparatively good, the exchange rate for that country's currency (pound) will be fixed at a comparatively high rate (that is, at a value higher than the rate for the pound which would bring the balance of trade into equilibrium). As opposed to this, in countries
128
Determination of foreign exchange rates
like Japan and West Germany with poor non-trade balances of payments and weak capital accounts, the yen and Deutschmark exchange rates tend to be set rather low. Industry in countries where the exchange rate is comparatively high will tend to have poor export competitiveness, so effective demand from abroad will tend to be inadequate, the employment level will be low, and the unemployment rate will tend to rise. By contrast, countries whose exchange rate is comparatively low will be able to export competitively and will have low rates of unemployment. Considered in this way, we can see that the fact that a country's trade and employment are depressed may not necessarily be due entirely to those working in export industries (employers and workers). Where the country has, relative to its export sector, disproportionately strong shipping and insurance industries and holds a large volume of overseas capital, the pressure of a comparatively high exchange rate will come to bear on those in its export industries. Seen from this viewpoint, trade disputes are not collisions of interest between the trading sectors of the various countries involved, but discord between a particular country's trading sector and its shipping industry and insurance industry, as well as its overseas investment sector.
6 The futures market The inevitability of forward trading
As in other markets, the foreign exchange market has spot (actuals) and forward (futures) markets. Since with overseas trade in particular there is quite a gap between the despatching of a commodity and its arrival at its destination, it is normal for payments for imports and exports to be made through the futures market. For example, when a Japanese exporter a who has made an export contract designated in yen completes the shipment of his goods, he will at the same time draw or issue a foreign exchange collection bill, with Tokyo foreign exchange bank A as recipient, which specifies that 150 days after the time the shipment is made the London importer b will make payment. He will then take it along to bank A together with the shipping documents. A will then send it with the shipping documents to bank B in London, and 150 days after the shipment was completed B will hand over the foreign exchange collection bill, together with the shipping documents, to b in exchange for payment in sterling. B will then convert this sterling into yen in the foreign exchange market and remit it to bank A in Tokyo. Given this mechanism for making payments for foreign trade, if there
The futures market
129
were only a spot market in foreign exchange, payments for goods would have to be made by one or other of the following methods. (i) Bank A will calculate the pound equivalent, at the exchange rate of that day's spot market, of the amount of yen a is due to receive in payment from b. Bank A will then get a to draw (issue) a bill of exchange for the same amount of sterling, and will then purchase this bill with yen. One hundred and fifty days later bank A will get the amount of yen obtained by converting into yen on the spot market the pounds received in exchange for this bill. (ii) The second method is where bank A takes charge of the bill of exchange designated in yen which a has drawn (issued) for 150 days, and then when this period is up b will convert yen to the value of the bill of exchange into pounds on the spot market and pay them to a. Lastly, (iii) bank A will buy up the yen bill of exchange issued by a, using pounds to the value indicated by that day's spot price. At the same time bank A will demand payment in pounds from b. In this case, b will not receive the imported commodities until 150 days later but must make payment for them in advance. By the first of these methods, bank A will either gain or lose depending upon the spot rate when the bill was accepted and the spot rate 150 days later. That is, the bank accepts the exchange risk. By the second method, the bank will avoid the exchange risk but b will have bought his goods at an undetermined price for the pound, and therefore importer b has to accept the exchange risk. By the third method neither trader a or b nor bank A will sustain an exchange risk; but since b receives his goods 150 days after the deal is made we should regard the third method of payment as a kind of variant of futures dealing. That is, unless the third method is adopted, either the trader b or the bank must accept an exchange risk by having the bill of exchange settled in the spot market. What happens then if the bill is settled in the futures market? Since the export contract was concluded in yen, exporter a has the right to receive a fixed payment in yen after 150 days. But if a were instead to obtain b's agreement that when the 150 days were up, b would pay to a a sum in pounds, at some mutually agreed rate of yen to the pound, which is the sterling equivalent of the sum in payment owed to a by b, then b will avoid the danger which stems from the fact that the future market rate of exchange between the yen and the pound is as yet undetermined, and hence uncertain. Moreover, the bank acting as intermediary merely has to ensure that the agreement concluded is carried out, and therefore
130
Determination of foreign exchange rates
it sustains no risk. Trading in futures consists of this kind of forward trading and the settlement of export (or import) deals where time elapses between the despatch and the receipt of goods normally takes place through the foreign exchange futures market. Premium and discount Let the spot exchange rate for pounds per yen be p £ / ¥ , and the exchange rate for 150-day futures be p£/ ¥ . These prices usually have a systematic discrepancy between them and this systematic difference between them depends on the level of interest rates in the two countries at the time. If we let the interest rate in Britain for a 150-day period be r£, and that for Japan in the same period be r ¥ , a person with m yen who changes his cash into pounds on the spot market and deposits these pounds is able to obtain h = (1 4- r£)/?£/¥ra pounds 150 days later. If, on the other hand, he deposits m yen in the form of yen and immediately sells the principal and 150 days' accumulated interest (1 + r ¥ )m for pounds on the futures market, he obtains 150 days later k = pf£/^ x (1 + r ¥ )m pounds, h and k must be equal, because if h > k, people will change the yen they have into pounds on the spot markets, and since it is more profitable to make deposits in pounds than yen, the supply of yen on the spot market will increase and the spot price of the yen against the pound will fall. Consequently p £ / ¥ will decline. By contrast, if h < k, then since it will be more profitable to make deposits in yen and to convert the principal and interest into pounds on the futures market, the supply of yen futures will increase, and the price of yen futures p£ / ¥ will fall. Thus eventually h = k, that is the following will become valid: = p $ / ¥ ( l + r¥)
(20)
If the value by which the futures price /?£ /¥ deviates from the spot price p £ / ¥ is positive, it is known as the premium, and if it is negative, it is known as the discount. If we consider the above equation, then ty¥ " P£/¥ -
1 ~r r^
and therefore the premium will be proportional to the difference between the British and Japanese rates of interest. However, this kind of equation will be valid only in the case of certain special expectations regarding prices. A person who has m yen can, rather than changing them into pounds on the futures market and depositing the pounds for 150 days, instead deposit the yen as they are, and 150 days later can change the principal and interest into pounds on the spot market. The spot pound price per yen 150 days later can only be
The futures market
131
anticipated at present. If we let the anticipated or expected spot price be /?!/ ¥ , the person who has changed yen into pounds by such a method may 150 days later be able to have; = p\j ¥ (1 + r ¥ )m pounds. If such ay is larger than k, that is if / = p | / ¥ ( l + r ¥ )m > k = /?£/ ¥ (l + r ¥ )m, people with yen will judge it more profitable to change these yen into pounds later on in the spot market, rather than change them into pounds now on the futures market. Therefore, converting in sterling futures will disappear (the demand for pounds futures will decrease) and the price of pounds expressed in yen (that is, p^/£) will fall, and therefore /?£/¥ will rise. By contrast, if/ is smaller than k> /?£/¥ is likely to fall. Price expectations depend on the individual and whatever information mechanism develops not everyone will view the future in the same way. For one group of people j> k, and they will behave so as to raise Pfi/y above the value set by (20). Another group of people for whom j
The /?$/¥ we have calculated in this way will be referred to as 'the reasonable expected value of the future spot price deduced from the present spot rate and rates of interest in both Japan and Britain'. People who expect that the price of the yen will in future be lower than this 'reasonable price' will sell off yen for pounds on the spot market and deposit the pounds thus obtained in Britain. By doing so and using the principal and interest to buy yen with when the spot price of yen really has fallen in the future, they will obtain more yen than if they had continued to keep their yen on deposit in Japan. Conversely, people who expect that the spot price for yen in the future will be higher than the 'reasonable price' will try to change pounds into yen. Whereas there is no particular need for such people to buy (or sell) pounds, they will do so in order to achieve their profit margins, counting upon selling or
132
Determination of foreign exchange rates
buying in a reverse direction at a more advantageous price in the future. This is speculation of the first type - speculation which uses the spot market twice at two different points in time. Speculation by the second method is carried out by persons who expect that the spot price in the future will be higher (or lower) than the present price of futures p£/ ¥ . Such people obtain their profit margins by currently selling pounds and buying yen (or vice versa) on the futures market. Then if their expectations are proved correct and the spot price of yen expressed in pounds does rise (or fall), they sell off their yen (or pounds as the case may be) and thus make their gains. Speculators will suffer a loss if their expectations are not proved right, and since foreign exchange cannot prevent speculative trading, exchange rates may be distorted by a concentration of speculative selling (or buying). However, an appropriate amount of speculation is useful in ensuring that the futures price p£/ ¥ is a reasonable one. That is, when /?•£/¥ is regarded as low (high) compared with the future spot price, a speculative demand for the yen (or a speculative supply of the yen) will arise and the pound price of the yen p£/ ¥ will rise (fall) when it is underpriced (overpriced). Whether in the form of spot or futures contracts, possessing foreign exchange may prove profitable; but it can also cause unexpected losses. In order to remove this danger when foreign exchange has been bought either at spot or as futures, it is necessary to sell precisely the same amount of foreign exchange forward (or at spot) to make the net amount held zero. In so doing the loss on spot will be cancelled out by the profit on futures (or vice versa), and one can avoid the dangers associated with holding foreign currency. The action of selling forward what was bought at spot and of selling at spot what was bought forward with the object of avoiding risk, is known as 'hedge operation'. Seen from this point of view, speculation is nothing more than acting positively to build a positive or negative balance of foreign currency without resorting to hedging. If a foreign exchange bank carries out such actions, in the worst case the bank itself can go into liquidation, and therefore the dealers of each bank are directed as a rule to make the net amount of each foreign currency held at the end of a day equal to zero, before they close that day's work. This is the highest precept of economic morality which dealers must maintain, and if they go on strike or are otherwise unfaithful to this morality the capitalist economy would be exposed to uncertainty like a ship navigating in fog, and quivering with fear at the danger, would perforce steer a very hesitant course.
PART TWO
The working of the national economy
The modern industrial society
1 An outline of the model A model of the industrial state lacking natural resources Macro-economics is a theory which reveals the structure of national economies and teaches us about the way they operate. Since economic structures are complex and differ from country to country, it will be necessary in what follows to make both simplifying and characterizing assumptions. Simplifying assumptions are those which make discussion simple and the analytical perspective clear; characterizing assumptions are those which remove extraneous factors so that we can isolate the special characteristics of an economy. But if an assumption, however much it may simplify an argument, essentially alters the course of the reasoning and if it appears to produce a substantial change in the conclusions, then we can no longer regard this as a simplifying assumption. Simplifying assumptions do not alter the essence of the argument, nor do they affect the conclusions. They are neutral in the sense that they merely simplify a complicated discussion. In contrast to this, a characterizing assumption is one which, while illuminating the special features of the economy about which it is made, allows us to pursue a course of analysis that is only valid under the assumption, and to draw particular conclusions on the basis of it. The postulate, for example, that 'the economy operates under a free enterprise system (that it is a capitalist economy)', is a characterizing assumption. But if, on the assumption that capitalism operates, we are only able to draw conclusions which are also true of socialist economies, our assumption will merely function as a simplifying one, in the sense that it is neutral with respect to the conclusions, and it will not be an effective 'characterizing assumption'. The true mark of a characterizing assumption is that it is crucial for the derivation of conclusions which are valid under the assumption, but which cease to be valid once it is removed. Recognition that an assumption is neutral so that the same conclu135
136
The modern industrial society
sions are valid whether or not we make the assumption (especially perceiving afresh that a previously unconsciously made assumption was, in fact, a neutral assumption) itself has scientific value. But simplifying assumptions contribute mainly to science in that they make the publicizing, dissemination and interpretation of knowledge easier, and it is precisely when appropriate characterizing assumptions are made that theory is refined and makes substantial progress. In what follows we shall not merely characterize an economy simply as operating under a system of free enterprise, but we shall also consider the country as having an economy composed of manufacturing industry alone (producing both consumption and capital goods), where agricultural and mining products are imported from abroad (the raw materials and fuel for manufacturing industry). Our model ought to be valid for Britain, Japan and West Germany, but it ought not to be applied in the form in which it now stands to America, Canada and Australia which have powerful agricultural sectors (and perhaps not to France either). Furthermore, appropriate caution should be exercised when we come to analyse the United Kingdom's fuel problem with our model now that she is making use of North Sea oil. The assumption that a manufacturing country depends on other countries for raw materials and fuel is a characterizing assumption. We shall further assume for the purposes of simplification that there is only one variety of consumption goods and of capital goods. Thus the model is a 'two-industry model'. A three-class society
The economy is made up from three classes of inhabitant: workers (w), entrepreneurs (e), and those who live off interest, the rentiers (r). Normally economics, and Marxist economics in particular, does not consider entrepreneurs separately from rentiers, but lumps them together as 'capitalists'. In the early stages of the development of the capitalist economy, many of the people who supplied capital were also heads of firms, but with the modernization of society, the management of firms is more and more becoming independent of those who provide capital. Moreover, those who still combine both roles often find that their interests as capitalists and their interests as entrepreneurs are not necessarily one and the same. (As we have previously stated, the fact that the interest rate is high, for example, will be desirable from the rentier's standpoint, but is not something which ought to be welcomed by the entrepreneur who is about to invest.) In modern societies, it is not necessarily the case that, as Marx saw, capitalists (entrepreneurs and suppliers of capital) will combine together in class opposition to the workers; it is also possible that entrepreneurs and workers will have
An outline of the model
137
interests in common and join together in opposition to suppliers of capital. In fact when rates of interest are high and investment low, so that the economy is depressed and many people are unemployed, the unions may well support a movement for a lowering of interest rates in response to calls from entrepreneurs. To be able to analyse such a situation - such as when the Director of the Confederation of British Industry (CBI) Sir Terence Beckett at the annual conference of the CBI in 1980 raged in opposition to the policies of the Thatcher Conservative government - the model must not be the two-class model of the Marxist and neoclassicists, but the three-class (or if we allow landlords, the four-class) model which was suggested by Walras and which Keynes took as his premise.1 Corporate bodies -firms, banks, governments In its demand for funds, the firm must borrow money from the banks, issue shares, or issue company bonds. However, in what follows we shall assume that all funds are acquired by issuing company bonds. This is not a characterizing assumption but a simplifying one, although not a simplifying assumption in the absolute sense. When new openings for the acquisition of funds present themselves, such as bank loans and share issues, our money economy develops in quite a different way, but there will be no significant change in the real economy (that is, in the production of consumer goods, capital goods, consumption and investment). This assumption cannot therefore be a simplifying assumption for the analysis of the money economy, but it may be regarded as more or less satisfactory as a simplifying assumption as long as we confine ourselves to an analysis of the real economy. Furthermore we shall also ignore the services of land as productive services in manufacturing industry, thus completely ignoring the landlord class as well as the existence of agriculture. The assumption of the absence of a landowning class ought to be regarded as a characterizing assumption rather than a simplifying assumption. Of course, the landlord class is very powerful in Britain, and in Japan too the problem of land for industrial use is a very important one. However, we will try to examine the industrial problems of the British and Japanese economies from the point of view of effective demand, production and employment, and not from the 1
For Walras' view of society see Morishima, M., Walras' Economics, Cambridge University Press, 1977, and Morishima, M., 'W. Jaffe on Leon Walras: A Comment', Journal of Economic Literature, Vol. XVIII, Jan. 1980, pp. 550-8. For an alternative interpretation of Walras' view see Jaffe, W., 'Walras' Economics as others see it', ibid., in particular pp. 529-30.
138
The modern industrial society
viewpoint of the land problem. When we analyse an economy with this intention, it is entirely reasonable to develop one's theory with the characterizing assumption that land can be ignored, and that it is not a bottleneck or constraint on the operation of industry. In an economy such as Japan's, where land can be an important constraining factor in the expansion of production, it will be necessary to revise the theory by explicitly incorporating land; but this problem is one which ought to occur at the second stage of the development of the theory. Next with regard to the corporate bodies of which the model is composed, apart from firms there are another three to be found in the economy - government, banks and foreign exchange stabilization funds. There are a large number of firms in both the consumer goods industry and the capital goods industry, and they arrange their production in the form best suited to be successful in competition. Thus firms which belong to the same industry adopt the same technology, and so production coefficients will be the same for all firms in any given industry. Firms are made up of a production and sales section (p) which manufactures products and supplies them, and an investment section (i) which has responsibility for productive capacity for the future. The government has the traditional, particular responsibilities of administration, maintenance of order, and defence; but in addition modern governments are assigned the task of ensuring the smooth running of the economy - which tends to occupy a greater weight in the overall workload of the government than the traditional responsibilities. Banks can be divided into central and commercial banks. The central bank is the only organization which can issue bank notes, and commercial banks are able to borrow from the central bank against the government bonds which they hold, and then lend the money so obtained, together with private deposits, to firms. The central bank cannot issue bank notes without limit. Instead it is able to issue bank notes up to a limit which is a given multiple of the gold which it has in reserve, or else it has to issue bank notes which are counterbalanced by government bonds. In what follows we shall not distinguish between government and company bonds, and we shall assume that central banks can also use the latter as a basis upon which to issue bank notes.2 2
This kind of assumption in not a good one. To regard company bonds and national bonds in the same light means a recognition that each enterprise enjoys the same degree of confidence as the government, and for that reason any model resting OH this kind of supposition can no longer deal with the problem of the likelihood of bankruptcy of weaker enterprises. Nevertheless this assumption is of value for the purposes of simplification as long as strict attention is paid to restricting the scope of our analysis below so as not to enter into that sort of problem.
An outline of the model
139
Exchange stabilization funds and foreign countries Since the short-term flow of funds between countries is not steady, it is easily apt to change direction. There are cases where funds which hitherto flowed from country A to country B reversed themselves and flowed from B to A or else left both A and B and went in the direction of C. Foreign exchange rates will fluctuate accordingly, and in order to calm such fluctuations and stabilize foreign exchange rates an exchange equalization account or exchange stabilization fund will be established in the government (Ministry of Finance) or central bank. Either by buying or selling their own currency, these organizations intervene positively in the exchange market and prevent rises and falls in the exchange rates. Thus the exchange stabilization fund is the authority with the responsibility of conducting equilibrating operations with regard to the exchange rate, but the result of its pursuing this task will be that money will be poured into the market or conversely absorbed from the market and hence its role as regards the implementation of monetary policy is a very important one. Lastly, there are foreign countries. In the recent controversy over Japan's trade dispute with Western Europe and America, it is pointed out that the Japanese economy is almost completely open to trade (this is the Japanese story, but the Europeans and Americans see the Japanese market as still having rather high barriers against the outside world, and non-tariff barriers in particular), but that the extent to which the Japanese economy is open to foreign funds (capital) is very low (the Japanese too recognize this). However, we shall assume that our economy is open to both trading and fund transfers. Consequently, not only will all industries export products to and import raw materials from abroad, and not only will the suppliers of capital buy bonds issued by companies in their own country and bonds issued by their own government; by participating in foreign bond markets they will supply funds abroad and introduce foreign funds into their own country. As far as an analysis of the current state of the Japanese economy is concerned this is no more than an inappropriate idealization. However, the economy Keynes himself used in the General Theory was just such an open economy which was abstracted from the British economy. Therefore, leaving aside foreign countries, our model is composed of three classes of individuals (workers, entrepreneurs, rentiers) and the five kinds of corporate body (private enterprise, government, commercial banks, a central bank and an exchange equalization fund). Of these, rentiers and firms will participate directly in foreign markets, but the exchange equalization fund will only be linked with foreign countries via the domestic foreign exchange market (the London market in Britain,
140
The modern industrial society
the Tokyo market in Japan). But the foreign exchange market is already highly internationalized and since regional markets between which close links are maintained form a world market, to participate in the London or Tokyo market is no more than to participate in the world market. We assume that other private and corporate bodies have no direct connections with foreign countries. Amongst corporate bodies, the government is the main body responsible for fiscal policy, the central bank is chiefly responsible for monetary policy and the exchange equalization fund chiefly responsible for foreign exchange policy. 2 The structure of the model (I)
The economic linkage table By what kind of mechanism does such an economy work? Like a precision machine whose parts are interlocked by gears, the changes which occur in one sector of the national economy will influence other sectors via certain fixed routes, and the effects thus produced will then influence other sectors to produce secondary effects. In addition, where the first change has a far-reaching influence it will produce tertiary and quarternary effects, and as well as the range over which the effects are felt being very broad the aftermath will last for a long time. In order to show the structure of a national economy - just as one can explain the mechanism of a machine no better than by means of a blueprint - it will be best to show the organization of the national economy in diagrammatic form and explain it in terms of a table. Table 6 shows how the demands for different goods at a given point in time, for example, 18 July 1983, are inter-related. The vertical columns show sectors (individuals, firms, the government and the rest), and the horizontal rows show goods and items of cost (products, factors of production, and so on). Thus if the /-th column is the production sector composed of firms, the cell at the intersection of the /-th column and the /-th row will show the value of the output of good i produced byfirms,or it will show the value of the/-th good used by the firms in production. If consumers are the category covered by the /-th column, the cell at that point will give the value of the quantity demanded by this category of economic agent (or the value of the quantity supplied by them). The figure for demand will be entered as a positive value, and that for supply as a negative value. Since products are supplied by firms whilst firms procure inputs from other sectors, the figure for the value of output will be entered as a negative value, and the input figure as a positive value. There are 11 vertical columns and 12 horizontal rows. The first two columns show the production sectors composed of firms (consumer
I
I
I
I
I bo.
CTQ
D
i
3
31
oo
-
bo
S
-
J
"
i
bOi
'
I^ i. bo,
+
I
1 i 11 1
O so
Foreign exchange stabilization fund
Central bank
Bank b
Government g
Foreign trade
Investment sector i
Rentiers r
Entrepreneurs e
Workers w
Capital goods industry
Consumption goods industry 5^f
Ci*
1
142
The modern industrial society
goods firms and capital goods firms), the next three columns show individuals (workers, entrepreneurs and rentiers), the 6th column shows firms' investment sector and the next two columns the foreign trade and governmental sectors; the final three columns are the monetary sectors (commercial banks, the central bank and the exchange stabilization fund). The first two horizontal rows show products (consumer and capital goods), and the last four rows show bonds (government and company bonds), time deposit and money - foreign currency (foreign exchange) and domestic currency. The 6 rows in between (3rd row to 8th row) show categories of costs (wages, depreciation costs, imported raw material costs, taxes, profits and interest). The production sector To begin with the production sector which consists of firms. The columns which show their activities (columns 1 and 2) also show how much was paid out to each item of cost, and what was the value of the products produced. Since the cost categories also include profits, total costs equal the total value of output, and since this is shown in the table as a negative value, the sum total of the terms in the column of each production sector is necessarily equal to zero. That is, if we take the price-determination equations given by the full-cost principle3 Pi = (1 + ti) (1 + m) (wa3X +/?4fl4i + rp* a5X)
(for consumer goods)
p2 = (1 + t2) (1 + m) (wa32 + p4a42 + rp* a52)
(for capital goods),
multiply both sides by the production volumes Xl and X2 respectively, and re-arrange the terms, the following will be obtained -pxXx + wa3XXx + p4a4XXx + rp*5a5XXx + mcxXx + ^(1 + m)cxXx = 0
(1)
- p2X2 + wa32X2 + p4a42X2 + rp *5a52X2 + mc2X2 +12 {\ + m)c2X2 = 0
(2)
Each of these terms is then entered in the first row of column 1 in the case of —pxXx or the second row of column 2 in the case of —p2X2 and then respectively from rows 3 to 7 of the two columns. Here each a^ represents a coefficient of production, w the wage rate, p4 the price of capital service, p* the price of imported raw materials in foreign currency (e.g. dollars), r the pound/dollar exchange rate, and m the mark-up rate. px, p2 are prices inclusive of sales tax. If we express the cost of production per unit of output, wa3i + p4a4i + rp1-a5i, by ch then the pre-tax price becomes (1 + m)ch and if VAT is levied at the rate f,, we get a price including tax atp l; = (1 + tt)(1 + ra)ct.
The structure of the model (I)
143
Workers Whilst the columns for the production sectors thus show the formula for fixing prices under the full-cost principle, for the other sectors the corresponding columns show their respective budget equations. Firstly, the workers' sector (column 3) will obtain wages W as income. However, if after having paid income taxes twW (where tw is the tax rate on wage income), and having spentp\Dwx to buy consumer goods there is still a surplus, the cash that they hold will increase by the amount of that surplus; if the total of these outgoings exceeds wage income then their cash balances will fall by this amount.4 That is, if we let the volume of money balances be Mw before the workers receive wages - and thus before tax payments and consumption expenditure - and if we make Lw money balances after consumption expenditure and payments of taxes, we obtain W - twW - pxDwx=
Lw -
Mw
The left-hand side of this formula shows the surplus, and the equation shows that the size of money balances will change by the amount of this surplus. Since this formula shows how income W is distributed between taxes, consumption and money balances, it will be called the budget equation. Table 6 rewrites this equation as follows: pxD\ -W+twW+
(Lw - AT) = 0
(3)
and has merely recorded each term in rows 1,3,6 and 12 of the workers' sector. However, we must not forget that a characterizing assumption is concealed behind this budget equation. If workers have a positive surplus from wages after having paid their taxes and made their consumption expenditures, it may be that they will use a part of it to buy bonds or make bank deposits; and they may with another part of it change pounds into dollars to buy foreign bonds. However, we shall assume here that workers neither appear in the bond or foreign exchange markets, nor do they make deposits (time deposits). In fact, some workers may buy bonds, and, on occasion, remit money abroad. Moreover, workers in dire poverty will have to borrow money from someone, so this is not a realistic assumption. Yet, trading in bonds and foreign exchange is not conduct which is essential to the workers. What makes a worker a worker is a way of life whereby he supplies his labour, obtains wages in recompense, and survives by buying consumer goods out of his wages. Our budget equation assumes workers confine 4
The reader is reminded that here D\ is the volume of demand for consumer goods of workers, and tw is the tax rate.
144
The modern industrial society
themselves only to the essential and carry out no acts which are not essential to them. This is done because we think that by doing so we shall reveal the fundamental characteristics of workers. Entrepreneurs The entrepreneur receives as his income a part all of the current recorded profits 77= mcxXx + mc2X2
and retains the remainder 77— a77 within the firm to provide funds for investment. The entrepreneur pays income tax ^a77 (where te is the tax rate on entrepreneurial income) in the same way as the worker, and makes consumption expenditures of p\D\, retaining the surplus in the form of cash. We can therefore write the family budget equation for the entrepreneur as pxD\ + teall- a77+ (Le -Me) = 0
(4)
Here, D\ is the volume of consumption goods the entrepreneur purchases, and Me is the entrepreneur's initial cash balance (that is, before he has received income, paid taxes, and made consumption expenditures). Le is the cash balance after making savings in the form of cash. This budget equation also hides a characterizing assumption. We have assumed that, like workers, entrepreneurs do not buy domestic or foreign bonds nor make deposits. However this assumption is much more extreme and unrealistic than in the case of workers. Since the surplus in the family finances of the entrepreneur is clearly larger than the worker's, the loss arising from keeping such a large sum in cash without its earning any interest is a large one. Naturally then the entrepreneur will buy bonds and make deposits with part of the surplus, and as a result he will have two kinds of income: the profits from the firm, and the interest from deposits and bonds. The entrepreneur will no longer be merely an entrepreneur; he will become a rentier too. There are amongst entrepreneurs many who intend when they are old to continue no longer as entrepreneurs but to become pure Rentiers. Therefore they accumulate their surpluses as far as possible in forms which bear high interest rates, and with the passage of time come to look more and more like rentiers, until in the end they do become pure rentiers. Since our assumption that entrepreneurs neither buy domestic or foreign bonds nor make deposits does not permit the existence of complications such as 'entrepreneurs who double as rentiers', it is an unwelcome assumption in that nearly all entrepreneurs in the real world are complicated creatures of this kind. However, for the view which considers that the interests of entre-
The structure of the model (I)
145
preneurs and rentiers are opposed in real-world economies, it is necessary to examine a model where entrepreneurs are only entrepreneurs and rentiers only rentiers, in order to bring the characteristics of real economies into sharper relief.5 Rentiers Financial markets in the real world are complex, but we shall assume for simplicity that apart from time deposits only one kind of fixed interest bond exists. That is, when we refer to bonds in what follows we shall mean bonds issued in perpetuity, so that once a bond is issued the issuer has to pay its owner one pound per year per bond in perpetuity. Such bonds will be negotiable and have a price, and if we make the current price of a bond pb, bonds purchased with pb pounds invested will earn interest of £1 per period, and therefore the rate of interest will be ib = n/pb If we assume rentiers owned Br bonds such as these, the interest income they get from the bonds will be £1 x Br. Assuming time deposits mature after one year, and that Qr is the total value of rentiers' deposits on that given day a year previously (thus maturing today), the interest income from the deposit will be hdQr. (Hence hd is the annual interest rate on deposits of one year ago.) Therefore rentiers' total interest income will be Ar = Br + hd(2r, and having paid income tax Tr (=trAr) on it and made consumption expenditures pxD[ out of it, their surplus will be Ar-trAr-PlD[. Unlike entrepreneurs and workers, rentiers do not participate directly in firms. Thus if they hold their surpluses in the form of money which does~not bear interest - as do entrepreneurs and workers - their incomes as individuals will not increase even if firms prosper. Therefore rentiers must aim to invest a part of their surplus in the purchase of bonds, and expand their sources of income by increasing the volume of their bond holdings or increasing the amount of their deposits. That is, if we let the volume of bonds purchased be 8Br, and the increase in their deposits be 6<2r, they will divert pb8Br and 8Qr from their surplus to expand the volume of bonds they hold and to increase their deposits. 6 5
6
By thinking in terms of this kind of model, it becomes very clear how the real entrepreneur, who is entrepreneur and rentier combined, is frequently placed in something of a dilemma by conflicting interests. If Br is the amount of bonds held by rentiers after they have purchased company bonds, then it is the sum of the bonds held prior to purchase (say, Br) and the amount of bonds newly acquired; therefore we get i.e. 8Br = Br-Br Br=Br+8Br Also if Qr is the new time deposits now commencing, and Qr those reaching maturity, then 8Qr = Qr - Qr.
146
The modern industrial society
Nor do they stop at that, because rentiers will try and hold not only bonds originating in their own country but also bonds from abroad, if such a course is advantageous to them. Assuming they hold a number of foreign bonds B£ which bear interest on a given day, their dollar income will be $1 x B£, provided that the issuer of the foreign bond promises to pay the owner 1 dollar per bond per period. On the other hand, assuming the rentier purchases 8B£ of foreign bonds during that period, he must pay their price pi 8B£ in dollars (p% shows the current price of the foreign bonds in dollars). Part of that payment will be offset by the interest income in dollars &£, but the remainder (p%8Bl - B£) must be paid in pounds. If the foreign exchange rate from dollars to pounds is r, the amount which must be paid in pounds will, of course, be r(p%8Bl — B£). The tax to be paid on the income from abroad, rB£, will amount to tr{rB£). These amounts too are subtracted from the surplus, and the remainder will be held in the form of cash. Assuming the volume of money held by the rentier at the beginning of the period is Mr, and that U is the volume of money held at the end of the period after adding on the new acquired cash sum, then7 U - Mr = [Ar - tr(Ar + rB;) -
PlD[]
- [ pb8B'] - 8Qr - [r(pt8B: - B;)]
The part within the first set of brackets on the right-hand side of this equation shows the rentier's surplus, the term inside the second set of brackets shows the sum paid out for the purchase of domestic bonds. Following this is the increase in deposits, and within the final set of square brackets is the sum paid to foreign countries. The equation shows that the remainder left from the surplus after these payments are made is assigned to an increase in cash balances. If we rearrange this equation - which is nothing more than the budget equation for rentiers we can write it as follows PlD\
- Ar + Tr + pb8Br + 8Qr + r{p*b8Br* - BQ + (U - Mr) = 0
(5)
where Tr = tr{Ar + rB*). In column 5 of Table 6 we record each item of this equation in order from the top downwards in its appropriate place. Their sum total is thus zero. 7
Assuming that time deposits cannot be withdrawn before their maturity.
The structure of the model (II)
147
3 The structure of the model (II)
The firms' investment sector The firms will purchase machines to replace worn out ones and to expand the scale of production. Let the volume of capital goods purchased be 72, and the amount of money required for the purchase be p2l2. The firms will reserve sufficient funds for this purpose. They will put aside depreciation allowances 77 for replacing machines, and the firms will retain as reserves what remains out of profits after the entrepreneurs have received their share all. The firms will also have money to the amount of M\ part of which they may mobilize in the purchase of capital goods. Thus the total amount of funds which the firms themselves have at their disposal is
where V is the balance remaining with the firms after they have disposed of a part of their cash in hand for the purchase of capital goods, and thus Ml - V is the total of money spent out of the amount of money in hand. If the total figure is not as large as the amount required for the purchase of capital goods, p2l2, the enterprises will have to borrow funds externally and make a new issue of company bonds for the purpose. Let O represent the total of the bonds issued by the firms in the past, on which they will have to pay interest of C. Then the total of newly issued bonds pb8Cl must be sufficient not only to purchase the capital goods but also to pay the interest on these bonds. That is8 Pbsa
= [p2i2 + c ] - [H+(n-
an) + (M< - L1)]
Here we assume no corporate income tax is levied upon the retained profits (1 - a)n. If for convenience of symbolization we now write this as C' = —B\ SO = — 8Bl (which signifies we consider bond-issuing as the owning of a negative quantity of bonds),9 we can re-write the above equation by re-ordering the terms as follows p2l2 - 77 + (an - 77) - & + pb8Bj + (V - Ml) = 0 8
If O is the total number of bonds already issued by an enterprise, then the total number of bonds after the new issue will be
0=0 9
(6)
+ 80
i.e.
8O =
O-O
The same formula holds for bonds issued by the government. For this reason &, 8Bl have a negative value.
148
The modern industrial society
This is the budget equation for the investment sector of the firms, and each item is entered in its appropriate place in order from the top downwards in column 6 of Table 6.10 Foreign trade
With foreign countries not only does trade take place but also the loaning of funds through the buying and selling of bonds. Let us assume that i^is the quantity of bonds which carry interest on the given day out of all the bonds of the particular country (the United Kingdom) owned by foreign rentiers (therefore, their interest income on that day will be Bf)\ and if we let the volume of bonds which they newly buy on the day be SBf, then in order to pay the UK foreigners will have to change dollars into pounds to the value equivalent of the total of exports purchased from the UK plus the value equivalent of the bonds newly purchased from the UK. That is, the following amount of dollars must be exchanged for pounds
Sj=q(p1El+p2E2+pb8Bf) Conversely the home country, the United Kingdom, must acquire the following amount of dollars in order to pay interest on bonds and the price of imports Here Ex and E2 show the volume of consumption goods and capital goods exported, and F indicates the volume of raw materials imported; q indicates the exchange rate of the home country's currency (the pound) into the foreign currency, the dollar.11 Since q is the reciprocal of r, the exchange rate from dollars into pounds, the above two equations can be written as follows rSf = pxEx + p2E2 + pb8&, rDj =rp5*F+& If we subtract the second equation from the first and rearrange terms we obtain, pxEx + p2E2 - rp$F -Bf + pb8Bf + r(£>/ - S$ 0 = 0
(7)
There are in addition the interest payments received by rentiers from abroad B£ and dollar payments made abroad for the purchase of foreign 10
11
In the real economy enterprises deposit a certain amount of money with banks either to provide for future investment or as an entree to borrowing funds from banks. However, it will be assumed below that the time deposits of enterprises are 0. It will also be assumed for the sake of simplicity that retained profits are untaxed. As has already been explained, /?J is the price of imported raw materials abroad. For that reason it is given in dollars.
The structure of the model (II)
149
bonds pb8B£. If the supply and demand for dollars including these figures are written as 5$ and D $ respectively, then
If we then substitute these into (7) we get pxEx +p2E2 - rp$F- Bf + rB'^ + pb8Bf-rp*b8B^ + r(D$ - 5$) = 0 (7') This formula shows that balance of international accounts is divided into the current account, the capital account, and the cash account. The first five terms represent the current account, the following two the capital account and the final term the cash account. Equation (7') shows that the total of all these three accounts must necessarily equal 0. It is noted that the current account is further divided into trade balance shown by the first three terms of (7') and non-trade balance by the next two. Government In this model it is assumed that taxation (T) is the only source of revenue for the government; but in order to carry out its public duties the government not only employs civil servants but also purchases consumption goods and capital goods. If we let the number of government employees be N8, and the total of all the goods it purchases be Gx and G2, the government's total expenditure will be p\Gx + p2G2 + wN8, and where the government's income does not cover this expenditure it must issue government bonds and thus borrow money from the public or the central bank, or from abroad. If we let the cumulative total of national bonds issued by the government in the past be C8, the government will have to pay interest of C8 pounds, and therefore the sum the government must budget for is piGx + p2G2 + wN8 + C8. The government must issue new national bonds to the value of the difference between this figure and tax revenue T. That is, pb8C8 = PlGx+ p2G2 + wN8
+C8-T
where 8C8 is the number of newly issued government bonds, and the above formula shows that the value of these bonds must exactly offset the deficiency in government revenue. If, as we did with firms, we consider issuing the number of bonds as owning a negative number of bonds and write this as C8 = -B8, 8C8 = -8B8, we can then re-write the above government budget equation as pxGx + p2G2 + wN8 -B8-T
+ pb8B8 = 0
(8)
Column 8 in Table 6 records each of these terms in order from the top.
150
The modern industrial society
Banks: commercial banks and central bank As we saw above, the borrowers of funds, that is the issuers of bonds, are the government and firms (the investment sector), whilst the lenders of funds, that is the buyers of bonds, are rentiers both at home and abroad. Where more bonds are issued than are demanded, the surplus is taken up by the banks (commercial and central), and firms or the government obtain money from the banks in exchange for these bonds. That is, if we let Bb be the number of government and company bonds taken up by the commercial banks in the past and 8Bb be their newly accepted bonds, then the commercial banks must supply the government and/or firms with funds to the value of pb8Bb.12 On the other hand the amount of time deposits in the commercial banks will increase by 8Rb (where 8Rb is the difference between new deposits Rb made on the given day and Rb which is the amount of deposits maturing on that same day), and they will have to pay hdRb on the maturing deposits. Therefore even if the banks immediately recycle in the form of loans the interest Bb which they are paid on bonds that they took up in the past, still they will have newly to supply money 8Mb = pb8Bb + hdRb + wNb - Bb - 8Rb
To do this, the commercial banks must reduce the amount of money they hold as cash reserves. (wNb here shows wage payments in the commercial bank sector.) If we let the cash reserves held by the commercial bank before supplying this money be Mb, and Lb be the value of cash reserves after supplying it, then 8Mb = Mb — Lb. Considering this relationship and putting Rb = — Qb, 8Rb = —8Qb (deposits received by the banks are negative holdings of deposits, i.e. claims on the banks), we may rewrite the above equation to obtain the budget equation for commercial banks wNb -Ab+ pb8Bb + 8Qb + (Lb -Mb) = 0
(9)
where Ab — Bb + hdQb. Each term in this equation is recorded in column 9 of Table 6 in order from the top.13 The central bank, as banker to the commercial banks, will not only lend them money when their cash reserves are inadequate; it will enter the bond market to carry out buying or selling operations in bonds. That is, where there is an excess supply of bonds it will carry out buying 12
13
The superscript b applied to B refers to the city banks, while the subscript b applied top below refers to the bond. As explained in an earlier chapter, foreign exchange dealings must always be carried out through banks; therefore the banks must be paid a handling charge for dealing with the exchange. In equation (9), as well as in all other equations relating to foreign exchange dealings, this handling charge is totally disregarded.
The structure of the model (II)
151
operations and take up the excess bonds, and where there is an excess demand the central bank will release the bonds it has in hand and conduct selling operations. Let the central bank purchase 8BC bonds (where the bank sells bonds, 8BC will have a negative value), and let us assume that the number of bonds it owns which bear interest on any given day is Bc. Let the increase in lending from the central bank to the commercial banks be 8QC, and let the loans which mature on that same day be Qc.u Since the money received by the central bank on the given day is AC = BC + hdQc,15 and money paid out by the bank is pb8Bc + 8QC, the central bank creates money by the amount equal to the difference between these two sums. That is 8Mc = pb8Bc +
8Qc-Ac
Re-writing this equation, we can obtain the budget equation for the central bank as follows - Ac + pb8Bc + 8QC -8Mc = 0
(10)
The terms of this equation are recorded in the appropriate cells of column 10 in Table 6. In addition to the money created by the central banks, commercial banks will release (dishoard) a quantity of money Mb — Lb, and the exchange stabilization fund will also release money, as we shall soon see in what follows. The total new supply of money for the whole economy is the sum of three elements. Of these, outside money is that part which offsets the government bonds and claims designated in foreign currencies accepted by the central bank and the commercial banks; and inside money is money supplied to offset the acquisition of company bonds. We shall discuss this point in detail later. Exchange stabilization fund If there is excess supply of a foreign currency (dollars) in the market, the rate of exchange for the 'dollar' against the domestic currency - the 'pound' - will worsen; and conversely if there is an excess demand for the dollar, the 'dollar' will strengthen against the 'pound'. If the domestic currency gets too strong, exporting will become difficult, and if it becomes too weak it will be more difficult to import. Nothing is more important than the stability of exchange rates in order to avoid such situations, but for that purpose a fund must be set up either by the 14 15
If Qc are the advances made on that day the net advances 8QC will be equal to Qc - Qc. The money rate of interest charged by the central bank on its advances is known as the official interest rate. This is not the same as the rate of interest paid by the city banks on time deposits, though the two are related. We will below assume, again for simplicity, that the two are, in fact, equal.
152
The modern industrial society
government or within the central bank which will buy dollars when there is an excess of them on the market, and later release them out of stocks held when there is an excess demand. The Exchange Stabilization Fund was established to meet just such a necessity, and if we express the demand for dollars of the fund as D|, then pounds equal to rD| will be required to buy dollars of that amount. That is the fund must supply pounds to the amount 8MS = rDf
and since this money is issued to offset foreign currency it is therefore outside money. Where the fund released dollars we should consider D | as having a negative value. At such times 8MS in the above equation will take on a negative value of course, and pounds will be taken up by the fund from the market in exchange for dollars. That is, the supply of domestic money will have decreased by the amount -8MS. Since the total of dollars held by the fund will have decreased through releasing them onto the market, then naturally that portion of outside money issued to offset dollars will also have contracted. Whichever is the case, we can rearrange the above equation - which we call the budget equation for the exchange stabilization fund - and re-write it as follows rDs$-8Ms = 0
(11)
Both terms in this equation are recorded in the bottom two spaces of the last column of Table 6.16 4 Equilibrium in real commodity markets - the principle of effective demand Walras' Law Up to now our reading of the Table of Economic Linkages (Table 6) has been vertical. Since the sum of the items in each column - which we will call the column total - is zero, the sum of all the items in the table, i.e. the sum of all the column totals, will also be 0. This does not necessarily mean, however, that the total of the items in each horizontal row which we will refer to as the row total - will all be 0 as well. This is because even though there may be several rows whose total is positive, as long as there are one or more rows with negative totals to offset them, the whole can always end up as zero. 16
For simplicity employment by the central bank and the exchange stabilization fund will be assumed in this table to be 0.
Equilibrium in real commodity markets
153
In this section we are going to try to read the table horizontally. Reading horizontally the table consists of three separate parts, namely the first and second rows representing the demand for and supply of consumer goods and capital goods respectively; the last four rows representing the demand for and supply of bonds, time deposits, foreign exchange and money; and the six other rows in between. Since the demand and supply are not necessarily always equal in the case of each commodity the row total of each of the first two and last four rows will likewise not necessarily be 0; however, the sum of each of the intervening six rows must always be 0, as will soon become apparent. To take row 3 first, W here represents the total amount of wages currently being paid to those employed in industry, the government sector or the city bank sector, and not the amount of wages corresponding to the volume of labour which workers would like to supply. Consequently the row sum of this row is not the difference between the amount of wages which will accord with the total demand for labour and that according with the* total supply of labour. It is the difference between total wages paid out by each industry, government and banks (i.e. the sum of wa3lXu wa32X2, wN8, wNb) and wages actually received by workers employed in these sectors (i.e. W). Therefore the sum of the items in row 3 must naturally always be 0. Similarly, in the case of row 4, since total depreciation reserves H will be the sum of depreciation reserves in each industry (i.e. pAaAlXx and p4a42X2) the sum of the items in row 4 will be 0 as well. The rp*F in row 5 is the total value of imports, i.e. in our model the sum of imports of raw materials for each industry, and the government's total gross tax revenue T in row 6 is the sum of indirect taxes on the products of each industry and the income tax levied on workers, entrepreneurs and rentiers. Since U in row 7 is the sum of the profits recorded by each industry (mcxXx and mc2X2), so the sum of each of all these three rows will be 0. To look finally at row 8, if we take into account that A> = &i + hdQi(j = r, b, c) the sum of the items in row 8 can be re-written as follows - Ar - & -Bf-&-Ab-Ac = r - [B + & + # + B8 + Bb + Bc] - hd[Qr + Qb + Qc] The contents of the first set of brackets on the right-hand side of this formula represent total holdings (by rentiers, enterprises, foreign countries, government, city banks and central bank) of bonds previously issued (company bonds and government bonds). Of these holdings the volume of bonds issued is entered as a negative volume of holdings, and
154
The modern industrial society
the sum is zero. This is because all bonds which have been issued must be held by someone, and it is impossible for someone to hold bonds which have not been issued by anyone. The contents of the second set of brackets are the sum of the total volume of time deposits and the advances made by the central bank. Since the time deposits received by the city banks and the sum borrowed by the city banks from the central bank are entered as Qb and have negative value, they are offset by the time deposits of rentiers Qr and advances from the central bank Qc respectively, therefore the sum of the contents in this set of brackets also equals 0. For that reason the row sum of row 8 is also zero. This means that where we have a horizontal reading of the table the central section of the table (i.e. rows 3 to 8) is constructed in such a way that the positive and negative items in each row offset each other. It is in the first two and the last four rows that the row total does not necessarily equal 0. However the sum of the totals of these six rows i.e. what remains when we deduct the central section (rows 3-8), which totals 0, from the whole table, whose total is also 0 - is 0. That means we get + Dl + D[ + E1 + Gi- Xx)
(sum of row 1)
+ Pi(h + E2 + G2- X2) r
(sum of row 2)
1
b
C
+ pb(8B + SB + 8ff + 8B8 + 8B + 8B ) r
b
C
+ (8Q + 8Q + 8Q )
(sum of row 10)
r
s
+ r(pb* 8B\ - B * + D / - 5 / 4- D$ ) b
b
+ (Lj -Mx + L -M =0
(sum of row 9)
C
S
8M - 8M )
(sum of row 11) (sum of row 12) (13)
where Lx = Lw + Le + U + L\ M1 = MW + Me + Mr + M. Since in this formula the contents of the brackets represent the volume of excess demand (i.e. volume of demand — volume of supply) for the respective commodities, the formula shows that the sum of the values of excess demand (estimated according to the prices of the respective commodities) of these six goods (consumer goods, capital goods, bonds, deposits, foreign exchange and money) must always be 0. This kind of relationship is termed Walras' Law, and the above formula demonstrating it is known as the Walras formula. The Walras formula inevitably operates even when the economy is not in a state of equilibrium. That is to say even when there is excess demand for some goods, this will be counterbalanced by excess supply of others, and, as the formula
Equilibrium in real commodity markets
155
makes apparent, total excess demand in the economy as a whole will always be 0. Inverse relationship between the real commodity economy and thefinanceeconomy Walras' Law immediately tells us the two following things. If we now divide goods up into real commodities (consumer goods and capital goods) and financial commodities (bonds, deposits, foreign exchange and money), then if real commodities are in an overall state of excess demand (i.e. if the sum of the totals of the first two rows is positive), according to Walras' Law financial commodities must be in an overall state of excess supply (i.e. the sum of the totals of the last four rows must be negative). This means that if there is excess demand for real commodities the result will be inflationary pressure in the financial markets. Conversely if there is an overall state of excess demand in the financial markets (i.e. if there is financial restraint), real commodities will be in overall excess supply, leading to stagnation and depression. Thus the movements in real commodities and in the financial field are the converse of each other, and it is the existence of Walras' Law which produces this kind of inverse relationship. Next, it is apparent from the Walras formula that provided that demand and supply are equal to each other for each of any five of our total of six kinds of real and financial commodities, then supply and demand will also be equal in the case of the remaining one good. For example, if supply and demand are equal in the case of all goods apart from money, then the contents of the first five brackets in formula (13) will all equal 0, and (13) may simply be written as (Lx-Mx + Lb-Mb- 8MC- 8MS) = 0 This formula means that the demand for and supply of money are equal to each other. Thus the market situation for any single given good reflects the market situation in the other five; consequently should there be a state of equilibrium in the markets for the remaining five goods the remaining one market which is a reflection of them is also in a state of equilibrium. For this reason, when we examine closely whether or not the markets of the economy are all in a state of equilibrium, there is no need to examine each of these various markets individually. It is possible to disregard that market (e.g. the money market) whose role is that of a mirror emitting reflections of the other markets on the grounds of superfluity, and to concentrate on the remaining markets. Many economists leave out completely the demand and supply equation for money, while others omit those for bonds and deposits.
156
The modern industrial society
Production period Let us take first of all the markets for real commodities. It is evident of course that a certain period of time is necessary to produce such commodities. The production period for any good may be defined as the time taken from when production is first started until that good ends up as a finished product. In our economy, since the consumer goods industry purchases and uses machinery and equipment made by the capital goods industry, the period needed for the production of machinery and equipment necessary to produce consumer goods is not included in the production period for consumer goods. On the other hand, capital goods are also needed to make capital goods (i.e. machinery and equipment). As far as capital goods are concerned as well, what we will call the production period will be the period from when the machinery and equipment for the production of capital goods has been installed and the production of individual capital goods has commenced, until the time as such goods have been completed. The period required to construct a factory and equip it with machinery will not be included in the production period for individual capital goods. This kind of production structure is called a double track revolving structure.17 It is double tracked in the sense that the consumer goods producing sector and the capital goods producing sector run in parallel with each other. It is revolving in the sense that of the machinery and equipment produced with capital goods one part is utilized in the consumer goods producing sector, while the remainder flows back to the capital goods sector and is used in the reproduction of the capital goods themselves. Since it is in the nature of certain kinds of products, notably agricultural, forestry and marine products and things processed from them (e.g. cheese, wine, etc.) to be highly dependent on the conduct of 17
In contrast to this the structure of production assumed by Bohm-Bawerk and Wicksell is one of single-track progression. As far as they are concerned capital goods are no more than intermediate products for the production of consumer goods. In the first stage of production labour alone is used to produce intermediate products; then these products are in turn used to produce second-stage intermediate products. Gradually in this way wefinallyend up with the finished consumer good. This kind of perception of the structure of production has recently been rendered more complex by Hicks, but in view of the fact that in an industrial country today, machinery or other capital goods must always be used at any stage of any production process (and therefore even at the first stage), any assumption of single-track progression which takes as its starting point labour alone is quite unrealistic. See Bohm-Bawerk, E. von, Positive Theorie des Kapitales (1889), Wicksell, K., Uber Wert, Kapital und Rente (1893), Hicks, John, Capital and Time (1973), Oxford, Clarendon Press. For the multi-track, revolving production structure see Burchardt, F., 'Die Schemata des Stationaren Kreislaufs bei Bohm-Bawerk und Marx', Weltwirtschafliches Archiev, 34 Band, Heft 2,1931. Marx was one of thefirstexponents of the multi-track, revolving theory.
Equilibrium in real commodity markets
157
nature, there are goods where it is essential for a specific period of time to pass. Once the production process for this kind of good has commenced it is very difficult either to advance or to postpone the time of completion; the product is ready after the passage of a more or less fixed period of time from the inception of the production process.18 By contrast, in the field of manufacturing industry, the production period for a large proportion of products is very flexible. In making an aircraft carrier, for example, if a war should suddenly break out it is possible for work to go on day and night to complete the ship several months in advance of schedule so that it can be sent to the battlefield. Conversely where there is excess production of a certain product, workers can be laid off and production halted for a time, then recommenced when the period of slump is over. It is therefore characteristic of manufacturing industry that the production process can be freely halted while it is actually in train and then started off a second time. As a result the output of manufacturing industry is very flexible, and it needs very little time to increase output. It is just as if the production period were 0. Such a thing becomes absolutely clear when the various processes of production are all brought together in an assembly line. Once production has been started no product will be obtained until it has gone right through the production pipe line, but once a product has completed the whole process products will immediately appear one after the other with each shift of the conveyor belt, just like water running immediately we turn on a tap. Should an increase in the day's output be desirable, production work just has to be extended by an appropriate length of time and the necessary extra products will be obtained during the course of that day. Even where the production period is T days there is no need to wait for T days to have those products. Since it is possible to gain within a day the extra amount of products decided on that same day, in that sense the production period is 0. However even in the case of the same production formula where demand is not very great, producers start to fill up the pipline only after having received orders, such as happens in the case of shipbuilding, so there exists at least some time lag between the decision to produce and supply of the finished product, though the production period may be flexible. Therefore the output of this sort of good is, like that of agricultural products, an inheritance of past production and is completely inflexible in the short term. 18
Recently, however, there has developed a farming sector where the period of production can be fairly freely controlled, as in market gardening, pig-rearing and chicken farming.
158
The modern industrial society
Realizing equilibrium through quantity regulation Assuming that the volume of goods supplied is the same as that produced, where output is inflexible demand must be regulated to meet supply if the two are to tally with each other. Where demand exceeds supply, supply and demand can be balanced by the rather forceful measure of ignoring the amount of excess (for example on a first come first served basis); where demand is insufficient to meet supply the excess supply cannot be removed unless demand is created by certain methods. (In this instance any forcible reduction of supply means making the supply flexible, and demolishes the supposition that the volume of supply is inflexible.) In order to get some consumers spontaneously to cut down their demand at a time of excess demand, and to give a fresh stimulus to spontaneous demand at a time of excess supply, some scheme must be operated which will bring together demand and supply through the agency of price regulation. As has repeatedly been said, agricultural products and some other goods are provided with a 'perfectly organized market' (exchange), but even in the case of goods such as this not all the total amount is always handled in the exchange; large amounts are subject to dealings outside it. Normally the price decided in the exchange is announced outside it as well, and as a fairly determined price it tends to dominate outside the exchange too. Thus as long as dealings within the exchange fairly reflect dealings as a whole then an equilibrium price will be established on the exchange, and at the same time demand and supply will balance both inside and outside the exchange. Where this is not the case, i.e. where exchange dealings are biased and fail to reflect fairly dealings as a whole, an imbalance between demand and supply outside the exchange will become apparent. Even in such a case as this, however, demand and supply within the exchange must be regarded as reflecting to a greater or lesser degree demand and supply outside, albeit in a distorted form; therefore the degree of imbalance between demand and supply outside the exchange at exchange prices will not be very great. The excess amount of demand or supply is forced to be given up, but the scale of this kind of 'forcible command' is not very large. Thus where the volume of supply is inflexible the price mechanism operates to bring demand in line with supply. Where demand is flourishing far in excess of supply the price of the commodity will rise; where demand is very slack it will fall. When the price has risen (fallen), each enterprise is likely to expand (contract) its output of the goods; so after one whole production period has been completed a greater (or lesser) volume of the goods will appear on the market. If the level of demand after the increase (decrease) in the price is too small (too great)
Equilibrium in real commodity markets
159
by comparison with this expanded (contracted) supply, then the price will be forced to fall (rise). The price will thus fluctuate up and down; so we obtain the so-called cobweb theorem.19 Where output is flexible, however, the situation is quite different. If output should fail to meet the demand for each good at the product price determined by each enterprise according to the full-cost principle, then each enterprise works overtime just for the length of time necessary to increase production by the amount needed to make up the shortfall. If the reverse situation obtains, and planned output is greater than demand, production is halted when the amount equal to demand has been produced. Thus it is possible to balance demand and supply for both the consumer goods industry and the capital goods industry without any change in prices by regulating output Xu X2.20 This means that where output is inflexible over the short term prices are flexible. Whereas where output isflexibleprices are fixed. Thus where Xx, X2 can be promptly regulated in this way excess demand and excess supply will not appear either in the consumer goods industry or in the capital goods industry, and the sum of the items in rows 1 and 2 of the Table of Economic Linkages will both be 0. Therefore the first two rows of the Walras formula (13) disappear, and only those items showing excess demand for financial commodities remain in formula (13). As we have seen above, in an economy where the volume of supply is inflexible the price mechanism operates to bring demand in line with supply, whereas in an economy where the volume of supply is flexible supply is adapted to a given level of demand. This kind of formula whereby demand determines supply is known as the principle of effective demand. The traditional economic theory of what is called the neoclassical school normally assumes operation of the price mechanism for all goods, but we will below be concerned with an economy where the principle of effective demand prevails for real commodities. The economy below, therefore, is one where as far as real commodities are concerned prices are inflexible. Our adoption of this kind of model as opposed to that of the neoclassical school is because in an economy where manufacturing industry constitutes the main industry, mass 19
20
Where goods are produced to order, like ships, demand and supply for the ship itself are always in accordance with each other, so there is no need to regulate prices so as to bring about a balance of supply and demand. Even in this case, however, the demand for the use of ships is not necessarily equal to the amount of ships available for use (including new ships), therefore prices are adjusted so as to bring the two together. The price for using a vessel will fluctuate up and down in line with the cobweb theory. These industries in reality have goods in store, and satisfy excess demand by releasing these stores, while excess supply is absorbed by increasing that store. Production is then adjusted to bring back these depleted (inflated) stores to their regular level.
160
The modern industrial society
production methods based on the assembly line formula are predominant; therefore balance of demand and supply is achieved largely through regulation of output. In analysing the case of 'a country lacking natural resources', which is engaged domestically only in manufacturing industry, importing agricultural products and other raw materials from abroad, it is not the neoclassical model which is appropriate, but a model based on the principle of effective demand. 5 The structure of the financial sector Short term bonds, central bank lending and the inter-bank market There exist in our model four kinds of financial goods; bonds, deposits, foreign exchange and money. Bonds are issued by different bodies, and depending on their different terms of maturity they have to be dealt with as separate kinds of commodities. However, for simplicity's sake we shall here make no distinction between government bonds issued by the government and the various company bonds issued by individual enterprises; we will assume that they are all bonds of a uniform character, that they are all fixed interest bearing bonds which promise to pay regularly a fixed amount of interest over an indefinite period. In the real world, central and local government issue all kinds of short and medium term bonds to enable them to carry out public investment, and enterprises, too, issue short term bonds (bills etc.). In our model such short and medium term bonds are disregarded, but since it is possible, even in an economy where only long term bonds exist, to achieve the same effect as that produced by the issue of short and medium term bonds by issuing long term bonds and then buying them back after an appropriate period of time, our disregard of short and medium term bonds is not serious. Moreover in our model no distinction is drawn between central bank advances to city banks and city bank advances to households (time deposits). If we rewrite the budget equation for the city banks we get -8Qb = [pb8Bb + (Lb - Mb) + wNb] - Ab
(10')
The contents of the square brackets on the right-hand side of this formula represent the sums required by the banks for the purchase of bonds (first item), to increase their holdings of currency (second item) and for the payment of wages to their employees (third item). A part of these sums is supplied from the bank's income from interest (Ab), so the city banks will try to raise the remainder from rentiers in the form of time deposits. It is therefore possible to regard (10') as expressing the
The structure of the financial sector
161
city bank's demand function for time deposits; but should this volume of demand exceed the time deposits of rentiers 8Qr the difference must be made up by advances from the central bank. Lending by the central bank is therefore a last resort to rescue city banks suffering from a shortage of funds. To repay the central bank for this relief activity the city banks must pay it interest in line with the official rate. If the sum required in aid by the city banks, [—8Qb] — 8Qr, should be greater than that which the central bank feels inclined to supply, the official interest rate will be raised. Where the opposite is the case the rate will be lowered. Since the rate of interest on time deposits is decided on the basis of the official rate of interest, should the latter rise (fall), the former will also rise (fall). Therefore, when -8Qb - 8Qr > (<) 8QC i.e.
8Qr + 8Qb + 8QC < (>) 0
the rate of interest on time deposits rises or falls; when 8Qr + 8Qb + 8Qc = 0
there is no change in the interest rate. Thus city banks suffering from a shortage of funds borrow the shortfall from the central bank, but in the real world banks in this situation also borrow their amount of shortfall from other city banks. This is what we mean by the inter-bank market. Advances between banks are normally for under 3 months, and very short term advances for half a day or until the following day (so-called call loans) are also frequent. The rate of interest prevailing in the inter-bank market is known as the inter-bank rate. When carrying out any analysis of individual city banks it is impossible to disregard the existence of the inter-bank market, but in view of the method of analysis adopted here, where the city banks as a whole are compared with the central bank, simplicity would suggest that it is permissible to disregard this market.21 Stock market The other important financial market we have not taken into account is the stock market. Looked at historically, each individual enterprise was established and developed on the basis of the personal capital of the entrepreneur himself or that of one or a small number of capitalists, but once mechanization came to necessitate huge amounts of fixed capital such means of raising capital at the individual level proved quite unable to satisfy the demand for capital. In the modern joint-stock company the enterprise sells transferable securities of small denominations, i.e. 21
For a more realistic model which does not disregard such items as call loans and short and medium term bonds, see, for example, Suzuki Yoshio, Kinyu (finance), Nihon Keizai Shinbunsha, 1980, pp. 88-93.
162
The modern industrial society
shares. By this means it can not only mobilize vast amounts of petty savings, but also apportion part or all of the profits to shareholders in proportion to the size of their holdings. Shareholders are the owners of the enterprise, and the shareholders' general meeting is by law the highest decision-making body. In fact, however, day-to-day decisions are made by those who plan the enterprise and those entrusted by the shareholders' meeting with its management (i.e. entrepreneurs and managers). Almost the whole amount of the capital raised by the enterprise in this way (i.e. its share capital) is used as long term capital and goes toward such things as machinery, other forms of fixed capital and land (for building factories). However, as an enterprise develops and expands it is necessary for it to engage in new investment for such things as the purchase of new equipment. Such new investment is frequently supplied by bank advances and bonds. However, if an enterprise continues to raise its new investment from purely external liabilities the ratio of the enterprise's own capital to its total capital will fall, as will confidence in that enterprise; so enterprises have to replenish their own capital by coping with new investment through new share issues. Since existing shareholders normally have the right to be given priority in the purchase of new share issues, the price of the old shares tends to rise as soon as the decision to issue new shares is announced. If, on the other hand, a holding of shares is disposed of, that person is no longer an owner of the enterprise. One can easily sell shares if they are listed on the stock exchange. The stock exchange is a conspicuous example of the 'perfectly organized market' explained in Chapters 1 and 2. On it the prices of shares are regulated by the minute to equate supply with demand. For those holding shares, therefore, not only dividends but the capital gain and loss resulting from fluctuations in share prices, are of major concern. The yield on shares is compared with the yield on other financial commodities, not merely in terms of the dividend,22 but also in terms of the predicted overall yield,23 which includes predicted capital gain or capital loss. Since share prices fluctuate fairly violently it is difficult to predict accurately their overall yield. Predictions tend to be able to give no more than the probable rate of yield. The variance (around the mean) of the probability distribution of the predicted overall yield on shares is 22
23
More precisely, in terms of the ratio of the dividend payable on each share to the market price of that share. That is to say, the ratio when the sum of the dividend payable on each share and the expected capital gain per period resulting from a rise in share prices is divided by the market price of that share.
The structure of the financial sector
163
generally regarded as being greater than the variance (around the mean) of the yield on other assets (e.g. company bonds). In the case of company bonds a fixed rate of interest is paid year by year regardless of the results of the enterprise, and when the term of repayment is completed the original sum is repaid. By contrast, in the case of shares, when a company's results are expected to be very favourable in the future shares not only provide a high rate of dividend but can be expected to rise in price. The predicted overall yield can therefore be expected to be a very high one. However, where an enterprise's results are expected to be not so favourable the price of shares will fall and there will be no dividend, so the expected overall yield will have a negative value. In the extreme case where an enterprise is disbanded the shares themselves will end up with no value at all. In this sense shares are risky assets. Some people like this sort of risky assets, while others dislike them. Those who like them purchase shares even when the average value of the predicted yield on the shares is lower than the yield on bonds, while those who dislike such risky assets will not purchase them unless the reverse relationship obtains. This kind of risk involved in the holding of shares can be avoided by 'hedging'. Take a person who holds 50 shares of A of £10 each. Should their price fall by £1 the value of his assets will fall by £50; but such a loss can easily be avoided by taking out an option dealing on the sale of 50 A shares. An option dealing means the purchase of an option on being able to sell a fixed number of shares (in this case 50 A shares) at a fixed price (in this case £10 per share) within a fixed period (for example 3 months). If the shareholder purchases this sort of right, if the price falls by £1 to £9 he will exercise his right to sell 50 A shares at £10. If he at the same time purchases 50 shares at £9 he will have £50 profit. Needless to say, the loss of £50 in the value of his assets is compensated for by the profit from his dealings. If the opposite should happen and the price of A should rise, he has no need to exercise his right to sell under the option dealing. He loses the money he has paid to secure the option and profits by the amount of increase in the assets value of the shares he holds. Thus the purchase of such an option renders shares an extremely safe form of assets. The money needed for option dealing can be regarded as the insurance premium necessary to avoid the risk which goes with the possession of shares. Since in our model shares do not exist no further explanation of the stock market is really necessary. Of the remaining financial commodities foreign exchange has been dealt with in Chapter 4, so no repetition is needed here.
164
The modern industrial society
What is currency? Last there is currency. Currency serves as a standard of value, a general medium of exchange and a repository of value. In the case of good A, its price pA expresses the amount of currency which is of the same value as one unit of A (i.e. it is expressed in terms of so many pounds per item). This is where it serves as a standard of value. Consequently whenever any person purchases any amount xA of good A, as long as he pays the volume of currency which accords with the price of that good pAxA, he will get xA. Here it is a general medium of exchange. In this case currency does not serve as a medium of exchange which can be used only in particular cases; its general character of being able to serve as a means of exchange for the exchange of any good with any person at any time is extremely important. For currency to possess this special character it either has to possess a particular quality which means that no person will ever refuse to accept any quantity of it (like gold and silver), or alternatively its circulation must be guaranteed by law, as is the case with fiat money or legal tender.24 As a result of this kind of general circulation, then, the person who has received currency can afterwards use it to purchase the things he wants, so holding of currency becomes a way of storing transferable value. Here it is used as a repository of value. Of course, the amount of value which is stored (i.e. the purchasing power of the currency) will decrease (increase) as the price of goods rises (falls), but there is no change in the nominal value of the stored currency during the period it is kept, whether expressed in pounds or dollars. It is not merely cash currency which combines these three currency functions. For certain kinds of deposits (current deposits) it is always possible to draw a cheque to take out money from one account and pay it into another; such accounts therefore possess the function of a means of payment. In return for this service such accounts bear no interest. Furthermore, as long as deposits of this kind are placed with city banks where there is no fear of bankruptcy they operate as repositories of value. For that reason such deposits are called deposit money and are included in regular currency. What we call Mx is the sum of cash and deposit money held by the private sector excluding the banks. By contrast time deposits do not possess all the qualities of currency in the 24
It goes without saying that this quality of general acceptability is restricted to within certain limits (e.g. to within the economy of that nation). Furthermore, under rationing currency alone cannot purchase wanted goods, therefore money is by no means all powerful as a medium of exchange. This means that money is not the perfect medium of exchange. In order to obtain things not only is money necessary, but also the permission of the state (and proof of this, e.g. coupons). In this way the state not only guarantees the general circulation of money, but also limits it.
The structure of the financial sector
165
same way as current deposits do.25 Since time deposits are not payable on demand, and cannot be withdrawn at any time, it is not possible to use a cheque to make a payment into someone else's account; therefore it does not serve as an immediate means of exchange. It is, however, a means of payment which can befirmlyrelied on, since when its full term is complete it can be used for payment or the settlement of accounts. If we regard time deposits as functioning more strongly as a repository of value - since time deposits are interest bearing - while having a more restricted function as a means of payment than currency, it is quite natural for them to be included in currency as quasi-money (what is known as Af2), interpreting currency in a broad sense. In our model, currency includes both cash currency and deposit currency. Hence the current deposits placed with the city banks are included in either M or L in our Table of Economic Linkages. The volumes of such deposits held at the beginning of the period and and those possessed at the end are included at their respective positive values in the volume of currency at the start of the period (M) and volume of currency at the end of the period (JJ) of workers, entrepreneurs, rentiers and enterprises (y = w, ey r, i). They are recorded as minus values in the accounts of the city banks Mb, Lb. Apart from this there are the current deposits placed with the central bank by the city banks, i.e. central bank deposits. Since when city banks engage in the creation of credit they do so on the basis not only of their cash currency but also of their deposits with the central bank, such deposits are as powerful as cash. Our currency includes the deposits placed with the central bank by each of the city banks, and these appear in the accounts of the city banks Mb, Lb with a positive value, while the value of the increase in central bank deposits during the period appears in the accounts of the central bank — 8MC with a minus value (hence a positive value in 8MC). Cash and central banks deposits are together referred to as high powered money. Supply of high powered money and the currency multiplier Let us now explain the items in the bottom row (the currency row) of our Table of Economic Linkages which are summed up as [(Lw + U + U+ V) + Lb] - [(Mw + Me + Mr + M') 4- Mb + 8MC + SMS] 25
Between current deposits and time deposits, there are some deposits which, unlike current ones, bear a low interest, but which, unlike time deposits, are always negotiable on demand. Cheques cannot be drawn directly on such deposits, but because they can always be transferred to current deposit accounts such deposits must be included as money (Mj), and this is in fact done.
166
The modern industrial society
First of all the contents of the parentheses within the second square brackets show the cash and value of deposits payable on demand (i.e. current deposits) held at the beginning by the private sector excluding the banks. The next item Mb is the cash in the vaults of the city banks at the beginning of the period plus central bank deposits of the city banks minus the amount of current deposits placed with them, therefore the sum of all the My's in the second square brackets is cash and central bank deposits held at the beginning of the period by the whole private sector, including the city banks, i.e. the volume of high powered money M at the start of the period.26 Against this 8MC is the sum of cash released by the central bank during the period and the value of the increase in central bank deposits during the period, while 8MS shows the cash released by the exchange stabilization fund. This means that 8MC + 8MS is the amount of high powered money released during the period by the central bank authorities including the exchange stabilization fund. Similarly the contents of the first set of square brackets - now represented by L - is the volume of high powered money which the private sector, including the city banks, aims to be holding at the end of the period. When the money market is in equilibrium the sum of the items of the currency row (row 12) of Table 6 is 0. We therefore get (Lw + Le + Lr+ V) + Lb = (Mw + M' + Mr + M') + Mb + 8Me + 8MS(U) or L = M+8M (14') This formula shows that in a state of equilibrium the demand for high powered money L equals the supply M + 8M. Furthermore the part within the parentheses on the left-hand side of formula (14) represents the volume of Mx which it is intended to hold at the end of the period, while the corresponding part on the right-hand side (i.e. the part in the parentheses) gives the volume of Mx at the beginning of the period. The difference between the two, therefore, is the volume by which the public wishes M1 to increase during the period, which we will write as 8MX. Lb - Mb is equal to increase in central bank deposits + increase in cash held in vaults of city banks — increase in deposits payable on demand during that period. For the city banks to conduct their day by day payment of deposits smoothly they are obliged to hold a certain percentage of their deposits 26
Note that as far as deposits payable on demand are concerned the amount possessed by the non-bank sector and the amount held by the banks exactly cancel each other out.
Equilibrium on financial markets
167
either on deposit with the central bank or in their own vaults, therefore the ratio of [increase in central bank deposits + increase in cash held in city bank vaults] to the increase in deposits payable on demand (a ratio we will term a), must have a more or less fixed value. Furthermore, let 8MX be the amount by which Mx is expected to increase during the period and e the ratio of that part of 8M1 which will be held in the form of deposits payable on demand to 8MX\ then Lb — Mb can be written as L-Mb = (a-l)e8M1 If we substitute this formula into (14), and bear in mind the fact that the difference between the parts in the parentheses on the left-hand side and right-hand sides of (14) is 8MU then (14) or (14') can be written as 8MX + (a- - l)eSMl = 8M
(14")
and from this we get 8Ml/8M= 1/(1
-e+eo)
This formula shows the amount of increase in currency (Mj) which will be stimulated in a situation of equilibrium by a supply of high powered money of the magnitude of 8M. This magnifying power is known as the money multiplier. 6 Equilibrium on financial markets - the determining of the interest rates The bond market Compared with markets in real commodities financial markets are well consolidated. Within the securities exchange there are two places for dealing in securities - the share sector for dealing in shares and the bond sector for dealing in public and company bonds. There are in addition other dealings which do not take place through the medium of the securities exchange, i.e. direct dealings between seller and buyer and over-the-counter dealings by securities companies. But prices in the exchange are published in newspapers and elsewhere, and prices in the over-the-counter market are rapidly notified to dealers by means of information exchange centres, etc.; therefore prices on the exchange and those paid in private dealings tend not to be very different. If they were very different arbitration dealings would be engaged in and the prices standardized. The price of bonds is determined on the bond market. (The prices of shares are determined on the share market, but since we are disregarding shares we do not need to give any consideration to their prices.) If
168
The modern industrial society
the demand and supply of bonds should be balanced at the bond price thus determined, the sum of the items in row 9 of the Table of Economic Linkages will be 0. The rate of interest on time deposits is fixed in accordance with the official interest rate, and if the deposits sector should be in equilibrium at that rate of interest the row total for row 10 will be 0. Furthermore if the foreign exchange rate r has been fixed at a level which will equalize the demand for and supply of foreign currency, the sum of the items in row 11 will also be 0. Then, providing the total of each of these three rows is 0, because the totals of rows 1 and 2 have already been fixed at 0 (a situation of equilibrium prevails in the real commodity markets), according to the Walras formula the row total of row 12 will also be 0. This means that demand and supply in the money market will be equal. It is thus possible to regard the money market as a reflection of the other financial markets, and as long as we are able to perfect our analysis of the other three markets the money market does not require any direct consideration. Of what sort, then, is the mechanism which will establish a balance of demand and supply in the other financial markets. To take bonds first of all, the market has both a morning session and an afternoon session. In Japan, for example, the morning market is 9 to 11, while the afternoon one is 1 to 3. Since the first (opening) and last (closing) deals in each market must be carefully carried out, there exist both cases where dealings are made by bidding, like dealings in Tokyo before the last war, and cases where prices are determined through chalking up offers according to the principle of balancing supply and demand. Bidding is conducted on the basis of very much the same formula as that explained in Chapter 1. As far as the 'chalking up' method is concerned, a representative of the market gathers together the demand and supply entered in the order book (i.e. on a board - in fact white cardboard), finds out, on that basis, the point where demand will be brought in line with supply, and carries out dealings at that point. Because of this there is no bidding, but the result is pretty much the same as if there had been. When the opening price has been decided in this way, agreements on dealings are carried out continuously until the start of the closing deals. There is no need for the price to remain fixed during the session, nor any need to equalize supply and demand at the respective prices on each individual occasion. The market representative notes offers to sell or purchase in the order book and the members of the market deal with them on the principle of 'first preference according to price, then preference according to time', and by reaching some sort of compromise between the lowest priced offer to sell and the highest priced offer to
Equilibrium on financial markets
169
buy, agreements are concluded at a level where price and quantity coincide.27 If demand and supply are not equated the surplus will remain on the market. Where it is supply that remains it cannot be cleared from the market until new demand at the same price appears or until there is a new supply at a lower price. When this latter is the case the new supply coincides with the demand at a lower price than that which previously existed, and contracts are agreed at a new price. The same thing happens when it is demand which is left over. Each time a new supply appears which the seller is willing to dispose of at a cheaper price the price will fall, and each time there is new demand prepared to pay a higher price, an increase in price will result. Price fluctuations of this kind occur repeatedly during the duration of dealings and when the session of dealings finally comes to an end the price will again be decided by the 'chalking up' method at a level which balances supply and demand. Over-the-counter dealings are conducted privately by the securities companies between themselves and their clients, just like the case of buying stationery at a store. Therefore the price for the same bond can be different according to the securities company. However, unlike the stationers, the securities company buys bonds from other securities companies which are selling them cheaply, and tries to sell them to other companies willing to pay a high price for them: so any company which is selling cheaply will soon be left with no bonds, while one buying in at a high price will end up with a surplus. For that reason a company of the first kind will try and supply itself with bonds by buying at a slightly higher price, while the latter will try and rid itself of its surplus by selling cheaply. The prices are thus rapidly brought to one level throughout all the securities companies. It must be noted that securities companies play very much the same role as that performed by each bank's dealers on the foreign exchange market. Although over-the-counter dealings are those between the customer and the securities company, the securities companies are in effect forming their own market for dealings between them, since they engage in the mutual exchange of informa27
Because normally no one will try to buy at a price higher than the sales value, we obtain the inequality: the lowest sales value 5= the highest purchasing price. When this formula holds with the pure inequality sign l > \ it means that no-one will be willing to purchase even at the very lowest selling price, therefore dealings will be 0. With the equality sign '=' people will be willing to buy at the selling price therefore dealings will take place. However the volume of demand may well exceed that of supply, or else supply will be the greater. In the former case the unmet demand, and in the latter the remaining supply, will remain on the market.
170
The modern industrial society
tion, and engage in the buying from and selling to other companies if there is a cheap supply or a demand at a high price.28 When successive new supplies appear one after the other during the duration of dealings the price of bonds pb will gradually fall as long as there is no one wishing to buy. If there are buyers even when supplies increase, new suppliers do not need to dispose of their supplies at a price which is any lower; as long as there exists an organization which adjusts its demand to accord with the supplies which appear, the downward progress of prices can be halted. This means that the price of bonds can be supported by purchase. Similarly if new demand continues to appear a price rise can be checked as long as there is someone wishing to sell. The body which conducts such buying and selling to support or restrain prices on the bond market is the central bank. This kind of intervention on the part of the central bank by regulating 8BC is known as bond purchasing (or selling) operations. The central bank releases by means of selling operations bonds previously bought up in buying operations; without this kind of intervention the price mechanism would prevail in the bond market and the balance of demand and supply would be reached through adjustments in the price of bonds. If, however, the central bank should once intervene in the market through manipulative buying or selling the price mechanism is immediately suspended and the bond market turns into a fixed price market. The time deposits sector We will next look at the time deposits sector. This consists of two connected sectors. The time deposits entrusted to the city banks by individuals (rentiers) and private enterprises (though in our model the deposits of private enterprises are disregarded) belong to the first section, while the advances made to the city banks by the central bank belong to the second. The interest rate governing the first of these two sections is the interest rate on time deposits id, while that prevailing in the second section is the official rate ie. The two are, in our model, given in terms of the rate per period (day or week) but not per year. They are connected, and if the official rate should fall the rate of interest on time deposits will sooner or later fall correspondingly. The two sections are also linked. Let us now suppose that for some reason the time deposits of an individual were to be of a small amount. 28
In Britain the jobber engages in dealings within the market, playing more or less the same role as that fulfilled by the Japanese securities companies in their non-market dealings. That is to say they engage in buying and selling in response to fluctuations in market prices and thereby mitigate fluctuations in the price of securities, but they also add considerably to the volume of dealings. The following explanation of the securities market and over-the-counter dealings applies exactly to the stock market.
Equilibrium on financial markets
171
If this was to happen the city banks would have insufficient capital and would be forced to borrow from the central bank to make up for the shortfall.29 When the central bank increases its lending 8QC there will be no change in the official rate of interest as long as it is possible to vary advances to comply with the demand for lending from city banks; but where 8QC is inflexible the official rate will have to be raised to curb the excess demand for advances. A policy which adapts 8QC flexibly (therefore one which does not change the official rate of interest) is known as a pure lending policy, while one which holds 8QC unchanged and regulates the official interest rate is known as a pure official-interestrate policy. In fact central banks often use an appropriate combination of these two policies. This means that they may increase 8QC by an appropriate amount while trying to confine any increase in the official rate within a suitable range. The time deposits sector is normally a flexprice sector regulated by the official interest rate, but when a pure lending policy is implemented it becomes a fixprice sector. Similarly the foreign exchange market takes the form of a flexprice market regulated by the exchange rate, but where the exchange stabilization fund acts in a totally passive manner in adapting its own demand Df, it becomes a fixprice market. Thus in the case of the three markets for bonds, deposits and foreign exchange, either the respective price variables are regulated, or the central bank or exchange stabilization fund regulates the policy variables 8BC, 8QC, Df, so as to establish an equilibrium in the respective markets; then balance of demand and supply will also be brought about in the money market which is a reflection of these three markets. The interest system If the price of bonds is pb and its reciprocal is ib, then pb is equal to the total capital value of the series of the interests of £1 for every year in perpetuity, and ib is the rate of discount for the capitalization.30 ib is the 29
30
Before banks with insufficient capital borrow from the central bank they will borrow from other city banks on the inter-bank market. In that sense the central bank is the 'lender of last resort'. In our model we will disregard the inter-bank market (and therefore also the inter-bank rate which prevails there). Analysis of the way in which city banks act, the theory of banks' behaviour, is a somewhat backward field in current economics. The capital value v of the perpetual interest income discounted at ib is £1 £1 £1 V= + ;+ + ... i + i> (i + ;,) 2 (i + ;,) 3 Putting k = 1/(1 + ib) we get v = £1 x (k + k2 + A:3 + . . . ) = £1 x =— \-k ib Bearing in mind our definition of yield ib(ib = £l/pb) it is clear that pb = v.
172
The modern industrial society
annual yield on bonds held in perpetuity. This will in our model below be termed the long term interest rate. Since bonds can be transferred at any time there is the yield appropriate to holding for a limited time, i.e. the short term or medium term (or more generally n period) interest rates. Let us now suppose that bonds are held for one period only and are sold on the bond market in the next period. Since the sales value of bonds has not yet been decided in the current period any estimated price can be no more than a prediction. Assuming now that the expected sales value is p'b, when the bonds are sold in the next period interest (of £1) will have by then been acquired; therefore when a bond of the current value pb has been held for one period and disposed of at the start of the next period, its short term rate of return ix will be given by the formula Pb
Where the same bond is held for two periods and is disposed of in the period after next (with an expected sales value of p"b), it will get interest of £1 at the beginning of the second period and at the beginning of the period after that. If bonds are bought by the first interest of £1 which is obtained at the beginning period, one can buy \/p'b units of bonds, and if they are kept for a further period and then sold they will be sold at P'blp'bi w ith additional interest of l/pb. Moreover, one unit of the bond originally held will be sold at p"b in the period after next, producing interest of £1. For that reason if bondholders dispose of their bonds in the period after next they will get a total sum of i.e. ( Thus where a bond is held for two periods the yield per period i2 is supplied by the formula
Here the first item on the right-hand side is 1 + iu the second is 1 + /,' (i[ is the expected yield where a bond is held for one period from the second to the third periods). This means that Similarly, when a bond is held for period n the yield in is given by (1 + !„)» = (1 + I,) (1 + i{) (1 + *',') . . . (1 + ,V» " »)
Equilibrium on financial markets
173
This means that the yield over period n is given by the yield in the short term and its expected values. Thus the various long and short term yields are tied to each other. In particular, when the short term yields are expected to be constant to each other (i.e. when il = i[ = i'[= . . .), the yields on bonds will be equal to each other regardless of how long they are held (i.e. ix = i2 = . . . = /„ = . . .)> a n d will in turn be equal to The interest rate on time deposits id will normally be lower than the yield on bonds it held during the same period, but the difference between the two must be neither too great nor too small. If it is too great (i.e. if the interest rate on time deposits is too low) rentiers will try to hold their surplus not in the form of deposits but in the form of bonds. The price of bonds pb will rise (therefore their yield it will fall), and deposits will decrease; so the city banks will try to borrow from the central bank (hence the official rate of interest will rise, resulting in a rise in the interest rate on time deposits). If, by contrast, the difference between it and id is too small, banks will get no profit from lending the funds they have amassed to enterprises by purchasing company bonds (and, in fact, with such a difference they cannot pay their administrative expenses). Consequently banks cease to engage in the purchase of bonds and as a result pb falls and yield it increases. If, furthermore, the banks' purchase of bonds decreases, their demand for advances from the central bank also falls and the official interest rate falls as well. This then results in a fall in the interest on deposits. In this way the difference between /, and id must be of an appropriate magnitude, but what is an appropriate magnitude of difference depends on (a) the city banks' administrative expenses required for being entrusted with one unit of time deposits and using it for the purchase of bonds (the difference between it and id must at the very least cover these expenses) and (b) the elasticity of time deposits with respect to the interest rate (i.e. if id should fall what proportion of their deposits will rentiers try and convert into bond holdings) as well as other factors. The smaller the amount of bonds bought, the greater the cost attendant on the purchase becomes per each unit. By contrast in the case of time deposits the cost of entrusting the banks with an amount of money does not get higher even where very small sums are involved. Thus where a balance is small it becomes time deposits, and where it is large it tends to be held in the form of bonds, but if the differences in interest rates between the two gets larger, it covers the difference in the two costs, so that small denominations of time deposits will more and more be changed into the form of bond holdings. It is thus the difference between the cost attendant on dealings in bonds and that required for
174
The modern industrial society
time deposits which decides the interest-rate elasticity of rentiers' time deposits. The yields on various financial assets must in this way fulfil mutually appropriate relative conditions, and not only that, their absolute level must retain an appropriate level in terms of its relationship with money.31 Money (whether cash currency or deposits currency) is not interest bearing, or even if it is the rate is very small; but it has instead the advantage that it can be used at any time. Thus if the yield on bonds and other items is too low, people will think that it is not worth losing the convenience of money for this kind of low yield; they will stop holding their balances in the form of bonds and convert them into money. Conversely, if yields are too high money will be changed into bonds. Not only that, interest rates must keep to an appropriate relationship with the foreign exchange rate, just as we have seen in section 6 of Chapter 4. For the yield on eachfinancialasset to fulfil these kind of relative and absolute conditions, it is necessary for the central bank and the exchange stabilization fund to conduct delicate market operations in the bond market and the foreign exchange market respectively, so as to hold the official interest rate and the exchange rate at appropriate levels. Each of these various kinds of financial assets can to a high degree be substituted by another, and demand and supply react sensitively to small differences in yield. The smooth operation of the financial market is like manipulating a fighter with acute rotational performance, and the slightest failure to control the joystick can result in disaster. Determination of prices Finally, we will look at the way in which prices pu p2 of consumer and 31
However, this 'appropriate level' is likely to be different depending on whether or not there is inflation. The rate obtained by subtracting the anticipated rate of price increases from the rate of interest on time deposits (or the yield on bonds) is termed the real rate of interest on time deposits or the real yield on bonds, but when the expected rate of increase in prices is very high such real rates or yields will either have a low value, or, in extreme cases, a negative one. Hence those holding time deposits or in possession of bonds become virtually non-existent; such people are likely instead to invest in land or property, for example, or in stockpiling. Thus a rate of interest on time deposits or a yield on bonds which may have been at the 'appropriate' level at a time when there was no inflation, ceases to be appropriate the moment inflation appears. The choice of whether to hold savings in the form of cash, bonds or time deposits, will depend on the yield on bonds and the rate of interest on money, but the choice of whether to hold savings in the form of financial commodities, such as cash, bonds, or time deposits, or in the form of land, property, stocks or other physical commodities, is likely to depend also on the rate of price increases or else the expected rate of such increases. Therefore the demand for cash balances on the part of individual and enterprises U (;' = w, e, r, i) is a function not only of the rate of interest and the yield but also of the rate of price increases (or the expected rate of such increases).
Equilibrium on financial markets
175
capital goods are determined. As has already been mentioned on several occasions, prices are calculated according to the formula of the full-cost principle; the mark-up rate ra, which we have up to now regarded as given in the calculation process, in fact differs in size according to the level of interest rates. It may, at first sight, seem that m is decided by each enterprise individually. However, an enterprise which calculates its prices using a mark-up rate which is too high is forced to sell its own products at a high price; so it will either be outdone in competition or will have no choice but to use a smaller m. Thus competition pushes down m to a limiting value which is fixed in accordance with the level of interest rates. That is to say when m drops below its limit an enterprise will choose to go out of business, investing its operating capital in the purchase of bonds or in time deposits, rather than continuing to operate and receiving profits with m at a low level. This, therefore, results in either excess demand for bonds (whenp^ rises and ib falls) or in an increase in city bank deposits (in which case advances from the central bank decline and the official interest rate and rate of interest on time deposits id both fall). Whatever the case, a smaller m is only compatible with lower interest rates. We can below write the relationship between the interest rates and the mark-up rate as m = m(ib, id)
(15)
If we look next at the depreciation reserve, let us assume that each unit of capital goods uses up its 6 x 100% each time one unit of product is produced. To produce one unit of consumer goods (or capital goods) a4i (or a42) of capital goods are used. Since each of these consumes 6 x 100% of a41 (or a42), each time a unit of consumer goods or capital goods is produced,32 then the price p4 of the capital goods consumed in terms of wear and tear must fulfil the relationship P4=0p2
(16)
If we substitute (15) and (16) in the equations for determining prices we get 32
The rate of wear and tear of capital goods used in the production of consumer goods is not necessarily the same as that of capital goods used in the production of other capital goods, but for the sake of simplicity it will be assumed below that both are equal. For administrative reasons of taxation the statutory rate of depreciation allowance is determined for capital goods over a single period, but it does not necessarily reflect actual technical capital wear. However in order to calculate the average cost which constitutes the basis for calculating the price of its products, and in order to calculate the amount of depreciation for which money must be set aside for the renewal of capital goods in the future, an enterprise has to make up cost accounts which are based on actual technical wear and tear.
176
The modern industrial society
px = (1 4- tx) (1 + m(ib9 id)) (wa3l + p20a4l + rpla5l)
(17)
P2 = (1 + '2) (1 + rn{ih, id)) (wa32 + p26aA2 + rp5*052)
(18)
In these formulae the rates of indirect tax, tu t2 are policy variables determined unilaterally by the government, p* is determined by foreign markets, r by the foreign exchange market, while m is obtained when the interest rates are determined in the financial markets. Furthermore 0 and atj are fixed numbers determined according to the technology of production. Thus if the wage rate w can be given as well the capital goods industry will determine product prices p2 so as to fulfil equation (18), and the consumer goods industry will calculate its cost of production on the basis of that p2, and determines its own prices px after adding to it profits and taxes (equation 17). When the price of each good has been decided in this way, the demand for consumer goods on the pa'rt of workers, entrepreneurs and rentiers is determined. Exports, on the other hand, are determined by foreign demand, while government demand Gx is determined as a matter of policy by the government. Output is regulated so as to be equal to the sum total of these demands. Similarly the output of capital goods is determined to make it equal to l2 + E2 + G2, where investment demand I2 as well as foreign demand E2 and government demand G2 may be regarded as given. The interest rates and the foreign exchange rate are determined on the financial markets and the foreign exchange market respectively, but the level of the interest rates and the exchange rate are largely dependent on the volume of bond purchases 8BC and amount of lending 8QC by the central bank, and on the exchange stabilization fund's purchases of foreign currency Df, both of which are determined as matters of policy. As has thus been seen, our system includes many policy variables. Tax rates tu t2, tw, te, tr are, first of all, determined by taxation policy, while government expenditure G1? G2 is determined by fiscal policy. 8BC is determined by the bond purchasing operations of the central bank, 8QC by lending policy, while Df is determined by the exchange stabilization fund. Given that these policies have been decided, once enterprises' investment plans have been decided, as long as the wage rate w is given, all other economic variables are determined according to our model. That is to say prices are determined by the mark up principle, output by the principle of effective demand, and the interest rates and the exchange rate by the law of demand and supply in competitive markets.33 33
This system is an extension of the conventional IS-LM model described in Additional Note g.
Is full employment possible?
1 The labour market
The ethos of the people and the labour market The labour market is where dealings in labour service are carried on. Labour service, unlike a normal commodity, cannot be stored. That labour which for any length of time whatsoever has been unemployed or squandered on enjoyment has been lost in perpetuity. Moreover labour service has to be furnished by people; there is no such thing as labour without people. Consequently a labour contract for the supply of 'prescribed labour over a fixed period of time' is essentially something restricting that worker during that period, and a bad labour contract is bound seriously to harm the freedom of the individual. For example a labour contract which engages 'to work for one's whole life for a certain individual (or for a certain company)' must essentially be regarded as something very close to a contract for the purchase and sale of slaves, and not merely as a contract governing the purchase and sale of labour service. In the case of the selling and buying of slaves not only does the payment for the sale of the slave's body not go into the hands of the slave himself, but a slave has no freedom even outside his working hours. If instead that payment was to come into the slave's own hands and with it he became a free slave, free to spend his private life as he wished outside his working hours, there would be no difference between such a free slave and a person who, while selling his own current labour service on the spot labour market, at the same time sells on the futures labour market any labour service he may offer in the future. The difference between a free slave of this kind and some kinds of worker e.g. workers in the lifetime employment system and professional athletes - is minimal. Since modern capitalism must not embrace elements of slavery even in the smallest degree, great care must be taken in organizing the labour market so as to prevent such elements from emerging. First of all, because workers must be completely free individuals, all labour contracts which restrict the freedom of the worker are illegal. 177
178
Is full employment possible?
Therefore even where there is a lifetime employment system there must be nothing akin to a legal enforcement of lifetime service for the worker. Even where an employment contract has been agreed on the understanding that an employee will continue to work over a long period, the worker must be free to resign if he should submit his resignation. If this were not the case the worker's right to freedom of occupation would be violated. Therefore while there may be a lifetime employment system the system is not enshrined in law, and anyone who deviates from or goes against this system is only going against a social custom whose contravention leads to no worse than moral censure and economic disadvantage. For that reason this kind of labour custom is not prevalent in societies which put their trust only in legally guaranteed contracts; moreover in societies where individuals in contravention of such customs (i.e. individuals who have left a company midway) are not subjected to social censure on the grounds that they have shown insufficient loyalty towards their company, lifetime employment systems are not common, even as a custom. Similarly, as far as the employer (entrepreneur) is concerned, as long as the economy within which he operates is a free enterprise system, his freedom of management must be preserved. Consequently labour contracts whose provisions would contradict his freedom to manage are acceded to only as matters of mutual understanding, and such contracts are, strictly speaking, entirely unlawful. One labour market of a certain specified type may be approved according to the historical environment, national ethos and perception of values of the people, but under a different historical environment it would completely lose all its support, and would not be able to function. For English workers believing their occupation to be their God-given destinies, loyalty towards their occupation takes precedence over all other things (for example over loyalty to the employer).1 In a society such as this electricity workers regard it as their calling to sell their labour on the labour market for electricity workers, so feel no guilt at changing to one employer after another, and they continue to supply labour service as electricity workers on their own specific labour market. They closely resemble workers in Japan before modern capitalism was firmly established. If, by contrast, one takes a modern Japanese worker (for example an electricity worker) who regards his company as the master he must serve, and who believes that he must love his company in the same way as he would love his country, his prime object of concern is whether or not he will be able to continue to give his labour to 1
As has been pointed out by Max Weber, this kind of religious sentiment is well expressed in the word 'calling' used in English to refer to an occupation.
The labour market
179
the enterprise to which he belongs, and as long as he continues to work for that company he will not mind in the least if the kind of work he does there changes a bit. For that reason a labour market for electricity workers in the sense of electricity workers appearing as suppliers of labour on the one hand and various companies seeking to employ electricity workers on the other, i.e. a labour market where supply and demand confront each other, is not really very developed in Japan. Japan has instead developed what might be called an intra-enterprise market, which means that on the one hand all the employees of an enterprise present themselves as suppliers of labour, and on the other the enterprise demands from them various types of labour. Each enterprise has its own synthesized labour market. Where such a labour market as this operates labour unions are likely to take the form of enterprise unions, as is in fact the case in Japan, while where the labour market is of the English type there is a tendency for labour unions to become trade or industrial unions. As has been made clear by Max Weber and others, the fact that British people have a strong loyalty to their occupation rather than the enterprise for which they work, while the reverse is true of Japan, has something to do with the religions of the two peoples. Therefore the structure of the labour market becomes in the final analysis dependent on the religious perceptions of the people. Whatever the case the labour market is conspicuously a product of history and is without doubt a very 'human' market. 2 The basic structure of the labour market in modern industrial society The structure, constitution and function of the labour market thus differ tremendously from country to country, but despite this it is an undeniable fact that the labour markets of all modern industrial countries possess a common fundamental structure. That is to say labour can be divided into regular labour which is in demand not just once but repeatedly, day after day, and temporary labour which is in demand on only one occasion, or seasonal labour which is required only at certain periods during the year; the labour which constitutes the main force in industrial society is regular labour in need of systematic work. Therefore a worker who is today working in a certain factory can expect to be employed there tomorrow as well, as long as he does not possess any serious shortcomings. Consequently we have a situation where there is within each factory (or each company) a labour market where the 2
Weber, M. The Protestant Ethic and the Spirit of Capitalism, London, George Allen and Unwin, 1978; Morishima, M. Why has Japan 'Succeeded'?, Cambridge University Press, 1982.
180
Is full employment possible?
workers employed by that factory (or company) constitute the suppliers of labour and the respective factories (companies) are providing the demand for labour (i.e. an internal labour market); in these markets labour contracts are not operative on a daily basis, but are valid for longer periods of time - weeks, months, years or even several years. Since in this sort of labour market contracts are valid until a specific point of time in the future, it becomes a kind of futures market and the wage rate decided upon there is fixed throughout the term of the contract. In this way the dominance of the internal labour market in the labour market as a whole contributes to a lack of flexibility in wage rates over a certain period, but in the supplementary labour market outside there is a certain degree of flexibility in wages. Factories and companies cannot secure all the labour they require from the internal market. Enterprises do not necessarily always operate on a fixed scale, and there will always be some enterprises which are expanding while others are contracting. Expanding enterprises will have to recruit labour from outside the internal market, while those reducing their scale of operations will be forced to release workers whom they have hitherto been employing outside the internal market as unemployed. Apart from this enterprises will also have to employ new workers to replace those workers who have left because they have reached the age of retirement, or for other reasons. There is in addition, the demand for labour from newly established enterprises. Demand for labour from such quarters appears on the labour market which must be used in search of labour outside the internal market, what we will now call the general market. Suppliers of labour in this market comprise such people as individuals who have just left school or college, the unemployed and those who are currently employed with one enterprise but who wish to move to another. In some cases these suppliers of labour compete on equal terms to find employers on the general market, while it is also frequently true that they are differentiated from each other and the general market is subdivided into more segmented markets. In the case of England, for example, where an enterprise is seeking to purchase on the general market a labour skill of a specified quality those possessing that skill must compete on an equal basis regardless of whether they have just left school, whether they are unemployed, or any other such conditions. In the case of Japan, where an enterprise is purchasing not only the skills of a worker, but also his loyalty to the enterprise,3 an individual who 3
Morishima, M. op. cit. pp. 117-28.
The labour market
181
claims that he wishes to move to another enterprise because he is not satisfied with the one at present employing him is at a very great disadvantage compared to a recent school or college leaver who remains as yet 'untainted'. The market for school or college leavers is quite distinct from that for the unemployed and movers from one firm to another, which is in turn distinct from the market for temporary workers. In the market for temporary workers, the demand for very short term, irregular labour is met by those who require seasonal work, students and the long term unemployed. The demand for the labour of school and college leavers is communicated to those intending to graduate or leave school through their respective educational institutions, and supply is adjusted to demand through the medium of the company entrance examination. (Alternatively, when the supply is small the demand is left unmet.) In contrast to this the demand for the labour of the unemployed and those transferring from one employment to another and the willingness of these people to make their labour available are communicated to each other through the mediums of employment agencies and individual personal connections. In Japan the unemployed and those moving from one job to another are normally employed to a greater or lesser extent merely on a supplementary basis. For that reason their position within the enterprise labour union after their unemployment is low, therefore the assistance they receive from the union will also be small. In the market for school and college leavers where it is not a worker's skill but his loyalty which is being purchased, the question of what level and what kind of skill the suppliers of labour possess hardly comes into consideration; the main criteria by which the employer determines success or failure in the company entrance examination are the applicant's desire to acquire skills and the latent ability he or she might possess. In the market for the unemployed and transferring workers, by contrast, an employer will be looking not for loyalty, but rather for skills, so this market will be subdivided on the basis of each specialized skill on offer. Also, in a society such as Britain, where the skills possessed by a worker become the main criterion for a decision whether or not to employ that worker, it is natural that the general labour market, as manifested in such places as Jobcentres, should be subdivided into various sectors on the basis of skill, but in an economy such as this the dividing line between the internal labour market and the general labour market is not a very considerable one. Individuals supplying labour on the internal market always have an eye to the situation on the general market outside as well, and if they see a good opportunity will
182
Is full employment possible?
present themselves on the general market as individuals wishing to change employment. Flexibility of wages
As was said before, in the labour market in a modern industrial society any treatment of workers in the manner of slaves is as far as possible avoided, so there is a ban on all acts reminiscent of the auctioning of slaves, horses or cattle. Consequently the sale and purchase of labour by auction is never undertaken, and even where employment agencies act as intermediaries in putting those seeking employment in touch with available work considerable attention is paid to keeping in mind the characteristics and desires of the individual. Public employment agencies try to introduce the right person to the right job, trying to preserve both the worker's freedom of choice of work and the employer's freedom to employ, and standing midway between worker and employer, siding with neither and acting in a neutral manner. However, agencies are not the only intermediary organizations. In each enterprise the personnel section is also likely to engage in its own labour-seeking activity, and in such cases will undertake positive recruitment measures through newspaper advertisements or personal connections. Out of those who apply the employer select those he regards as most qualified. This method of selection is in essence a cross-trade; the supply of and demand for labour are brought together not by changing wages but by regulating its volume. The demand side for labour is not likely to revise fundamentally the wages which were announced in the first place during the process of seeking employees. It may on occasions modify the wage rate to a greater or lesser degree in response to the wishes of applicants, but a drastic change is out of the question. Even where the number of applicants exceeds the number of vacancies, neither the enterprise nor the applicants will try to bid down wages, and even in the reverse case any bidding up of wages - at least after the selection has already started - is as a rule avoided. It is essentially a regulation of volume, where applicants in excess of the number of vacancies are thrown out, and, where vacancies exceed applicants, the attempt to fill the vacancies which remain is abandoned. Where labour is in short supply and enterprises have been unable to secure all the labour they need they are likely even after that to continue to make efforts to fill their remaining vacancies. Since they will once again be unable to find suppliers of labour if they have only offered the same wages as before, enterprises will almost certainly offer wages higher than the previous ones. However, though an enterprise may be able to fill its vacancies by doing this, if it should become clear that there
Labour unions
183
is a disparity in wages between the workers who were recruited initially and those who entered the company subsequently, those new members of the company who entered it first are likely to call for equal treatment and to demand a rise in wages to remove the disparity. Furthermore all those workers who entered the company in the past will start to agitate for an increase in wages. If the enterprise should accede to such an across the board rise in wages it will have no choice but to increase the price of its products. Should it do this, demand for its products will then decrease, which will provide a severe blow to the enterprise, so rather than all this happening the enterprise is more likely to choose the alternative of leaving its vacancies unfilled. Therefore even where the demand for labour exceeds its supply wages will not rise and the excess demand will remain unmet. Thus regardless of the existence of an excess demand for or supply of labour there is no outward change in the money-wage rate. In that sense wages are inflexible upwards and downwards. Nevertheless in real terms (i.e. if we bear in mind the quality of the labour which is being secured) wages are flexible. By this we mean that when labour is in excess supply and unemployment is high the recruitment side is able to take its choice of the available labour, so that only good quality workers will be employed. Conversely when the demand for labour is increasing and the supply is tight enterprises will be forced to employ even the lower quality workers. Therefore, the wages which are to be paid per unit level of quality of work are low at a time of high unemployment and high at a time when unemployment is low. 2 Labour unions Trade unions and relative wages
We have already said that the structure of the labour market differs greatly according to whether the object of loyalty which the worker must serve is his occupation or his enterprise. Let us now call a society where the mainstay of a worker's morality is his loyalty to his occupation, a British type of society, and one where loyalty to the enterprise is the worker's supreme ethical virtue, a Japanese type of society. These societies possess labour markets whose structures are respectively as follows. First of all in a British type society there is hardly any barrier between the internal market within each enterprise (the market where contracts for the labour of those already employed by the enterprise are renewed) and the general market (that for school and college leavers, those who wish to change their employment and those who are unemployed).
184
Is full employment possible?
Moreover both the internal market and the general markets are subdivided according to trade. The internal labour markets in all enterprises for electrical workers for example, are tied up with the general labour market for all electrical workers, therefore for each individual electrical worker the question of what enterprise he should work for is hardly a problem. If he is dissatisfied with the enterprise which is currently employing him all a worker has to do is to move to another enterprise. It is, however, extremely rare for an electrical worker to change his occupation and to work in another kind of employment (for example as a miner). Where the structure of the labour market is of this kind electrical workers have to form a labour union consisting only of electrical workers, on either a national or regional basis; it would be senseless to form an electrical workers' union for each enterprise. Trade unions covering the whole nation do not serve the interests of workers of any specific enterprise in particular, so as long as they belong to that specific occupation those who are unemployed as well are qualified to be union members. Since labour unions are organized on the basis of their respective trades it goes without saying that in a society such as this each enterprise will have to negotiate with a very large number of labour unions. Wages are decided as a result of negotiation between each enterprise and the labour union, and the existence of labour unions with the characteristics described above contributes to the establishment of uniform wages over the nation as a whole. Labour unions are not so irrational as to ignore completely the management situation in each individual enterprise, so that there may be some differences in wages for the same trade in different enterprises; however, such inter-enterprise wage differentials are very small by comparison with those which can be found where there are enterprise labour unions, as described in the next subsection. Since, however, there is a difference in the bargaining power of the various trade unions, an occupation which has a powerful union will invariably negotiate successfully with the enterprise, and for that reason the members of such a union will always obtain more advantageous wages than those in other occupations. Consequently one matter frequently under discussion in societies such as this is whether the relative wages for different occupations are, in fact, fair or not. However, if the wages for a certain trade continue at a comparatively high level over a long period, then there will be a tendency to regard this situation as fair per se, so a movement for just wages is no more than a movement which supports the wage relativities which currently exist, and there is a tendency for relative wages, which reflect the real power of labour unions, to be upheld in the name of 'fairness'.
Labour unions
185
In such a society as this, therefore, if one labour union should be successful in raising its wages, then this will set in motion other labour unions as well, and they too will demand wage increases. This means that the movement for a rise in wages is easily spread from one occupation to another and brings about across the board wage rises in all occupations, though such rises cannot be said to be strictly commensurate with each other. Nor will the relative wages in various occupations be in proportion to the marginal productivity of the respective workers. Nor will they be in proportion to the degrees of scarcity of each kind of work. A significant influence in the determination of relative wages is also exercised by factors usually regarded as outside the field of economics, such as what sort of scale of strike can be backed up by the funds of the respective unions, the degree of leadership possessed by the leading members of a labour union, as well as how skilfully they are able to enlist the support of public opinion. And wages determined in this fashion become the norm for recruitment on the general market. Enterprise labour unions and wage differentials between enterprises Where each worker feels strong sentiments of loyalty towards the enterprise to which he belongs, once a worker has become employed he will very rarely move to another enterprise, so the labour market is subdivided on the basis of each individual enterprise. For such a society as this, enterprise labour unions are more appropriate than trade-based labour unions. Within each individual enterprise a union is formed which has as its members all the workers employed in that enterprise (or all those among them who are interested); wages for all the different occupations embraced by that enterprise are then determined as the result of negotiations between the enterprise and its union. Some enterprises possess in addition to their labour unions staff unions whose membership is the white collar workers employed by the respective enterprises, while other enterprises make no distinction between workers and staff members and have only a single employees' union where both groups can become members on an equal footing. In cases such as this the union participates in the determination of the salaries of white collar employees as well, negotiating on their behalf. An enterprise labour union (or staff union or employees' union) of this kind is not involved merely in the determination of wages and salaries, it also has a significant interest in promotion within the enterprise, and there is a tendency for the relative wages of a worker to depend on his position within the enterprise rather than on the occupation of that worker. Temporary workers, retired workers or those made redundant as a
186
Is full employment possible?
result of personnel retrenchment are in principle lacking the qualifications to become members of an enterprise union. Such enterprise labour unions frequently come together on the basis of industry to form national union councils and pursue a joint campaign, but during such a struggle the individuality of each individual union remains dominant, which means that it is not difficult for considerable wage disparities between enterprises to appear.4 As has been said, we cannot decide what sort of relative wages are fair between different occupations (e.g. miner, train driver and automobile worker), and there is ultimately a tendency to regard as fair wages those relative wages which have historically prevailed over a long period. With regard to relative wages between enterprises, however, there, is one universally acceptable perception for deciding what sort of relative wages are fair. That is to say, what could be termed fair is for the relative wages for the same work offered by enterprise A and enterprise B to be 1:1. Despite this, where each enterprise decides wages as the result of negotiation with its own labour union, relative wages for the same work in different enterprises often deviate considerably from the 1:1 standard.5 Thus under the enterprise union system notable wage disparities tend to exist between outstanding enterprises (large enterprises) on the one hand, and enterprises where business is poor (medium and small enterprises). Since those enterprises which can pay only low wages are unable to ask loyalty of their employees, the internal labour market of these enterprises is unable to construct an effective barrier against the outside, and workers in these enterprises will easily change their jobs when it is advantageous for them to do so. They present themselves on the general market used by the unemployed and other workers changing from one job to another, and mutually compete for wages which are, even just a little, higher. However, the workers who appear on this sort of market are those who are excluded from any enterprise union, so they receive no union protection whatsoever. For that reason this market is completely a buyer's market. Employers offer extremely low wages calculated on the basis of wages offered by not very prosperous enterprises, and those suppliers of labour who have not been able to get 4
5
Professor Peter Wiles' comparison of Japan and Yugoslavia has revealed that there is a tendency in countries where enterprise unions exist for there to be very large wage differentials between enterprises regardless of the nature of the system; it makes no difference whether a country has a free enterprise system or whether it is a socialist country. Concerning this point see Morishima op. cit., p. 174. This demonstrates that relative wages depend very greatly on the enterprise's management situation and the power of the labour union. Similarly where craft unions exist relative wages depend largely on the strength of each individual craft union. As a result they are far removed from any ratio of the pure marginal productivities of each occupation.
Labour unions
187
themselves selected are thrown out. On the general market for school and college leavers enterprises offer wages in line with those agreed to by the labour union in their respective internal markets. For that reason the wages of those who have been able to gain employment with the best companies are high, while those who have only been able to gain employment with the less prosperous enterprises will receive very low wages. Those who have been unable to obtain employment with any enterprise will become unemployed and enter the general market for the unemployed and those who wish to change their employment. Here they will be forced to compete with others for work which is, to say the least, unattractive. The latent tendency towards slavery in modern society Whether in a society like Japan or in one like England, the role played by the labour union in wage determination is a considerable one. A union which is low in funds cannot commence a strong strike, and ultimately will be forced to give way to pressure from capitalists and management. Even a union which possesses ample funds to make it strong enough to continue a strike over a long period will not only lose the support of public opinion but may even tend to find that its own membership is split into factions at odds with each other if the union leadership is excessively aggressive. An enterprise depends for its success not just on its strength in terms of capital; it also depends greatly on such characteristics as the resolution of the entrepreneur, on his leadership and on his administrative and managerial ability. In exactly the same way a labour union's success or failure in winning good wages could without exaggeration be said to be tied up with whether or not those at its head offer outstanding leadership. Moreover a first class union president will make his name as an entrepreneur as well. Consequently any enterprise which possesses its own union is likely to regard intra-company conflict between the union leadership and management divided into two camps - labour and capital - as a waste of the valuable commodity of leadership capacity. Seen from this point of view, the fact that in many Japanese enterprises the president and directors look for their future president among the union leadership ceases altogether to be a strange 'Japanese custom'. The participation of the union leadership in the enterprise's management meetings which takes place in West Germany must also be seen not just as a sop to workers but as an effective means of making the most of their leadership ability. It is a fact that the lives of workers have been conspicuously improved through the power of labour unions, but despite this under the surface
188
Is full employment possible?
workers remain in just as much of a weak position as they were before. Because workers have no means of living other than by working for wages, they are in a very disadvantageous position, and may be forced to put up with unfair, low wages. Although in modern industrial societies great efforts have been made to try and sweep away the residue of the 'slave market' from the labour market, there is no change in the fundamental characteristic of capitalist society whereby the capitalist side is in a favourable position to impose low wages on the worker. In a society like England a low level of unemployment means that the capitalist is in a less advantageous position, and even inferior workers have a considerable chance of being employed at the wage demanded by the labour union; when the unemployment rate is high, however, only the best workers have a considerable chance of employment, so unions are likely to be unable to demand high wages. In a society like Japan, however, there is no assistance from labour unions in the general labour market for the unemployed and those wishing to change their employment, so only very low wages are on offer. Even though it may be something of an exaggeration to do as Marxists do and designate modern factory workers as semi-slaves, it is a fact that, in modern society, just as Marx pointed out, while workers may be legally free they remain no more than individuals who are compelled to work, in the final analysis by the iron law that 'he who does not work must not eat'. That this kind of latent tendency towards slavery does not apparently become an observable tendency in present-day society is due to the provision of unemployment benefit based on a system of social security. 3 Unemployment (I): Marx
Kinds of unemployment Unemployment is usually categorized as one of three kinds on the basis of the reasons behind it. These are frictional, cyclical and structural unemployment. As we have already seen, the unemployed gain information on job vacancies through such media as employment bureaux, newspaper advertisements or talking with acquaintances. Consequently information concerning the situation vis-a-vis vacancies in a different region is hard to come by quickly, and even if such information is obtained employment opportunities are frequently let slip on the grounds of factors such as inability to prepare for moving or family circumstances. However, unemployment resulting from this sort of reason would, given time, probably gradually disappear. Information regarding vacancies would eventually be passed on to a large number of people, and some of the unemployed would become sufficiently pre-
Unemployment (I): Marx
189
pared to move. For this reason this sort of unemployment is called transitory, or frictional unemployment, and with an economic theory which supposes an ideal situation without any friction, such unemployment can be either completely ignored or left undiscussed. Cyclical unemployment is that which is produced when business is bad, and is largely removed when the upturn comes. Structural unemployment is that resulting from a change in the structure of the economy. It includes unemployment based on such changes as, for example, a reduction or total disappearance in the demand for a certain kind of labour due to the appearance of a labour saving machine (e.g. robots), workers becoming redundant due to the exhaustion of natural resources (e.g. miners and coal), or the decline in the demand for agricultural workers as a result of the progress of industrialization. Unemployment resulting from the loss of overseas markets and the extension of monopoly are also categorized as structural. Furthermore where there exist large disparities in wages between enterprises or between industries, there will exist in one part of the economy individuals who work for very low wages; such individuals are working because they have to do so to stay alive, but must really be regarded as unemployed rather than employed. This kind of half-employment half-unemployment is called disguised unemployment.6 Three views of unemployment: Marx, Keynes and the classical school We shall below disregard frictional unemployment, and for the purposes of our analysis of unemployment divide it into the three following kinds. First of all we have unemployment originating in insufficient machinery and capital equipment. Since people originally had no machines and began to produce things with their bare hands, it is possible even today to produce goods using only labour which almost exclusively uses primitive methods of production. In a modern industrial society, however, if an entrepreneur no longer has machinery he no longer needs labour.7 As long as there is capital (taken below to include both 6
For disguised unemployment see, e.g. Robinson, Joan, Essays in the Theory of Employment, Oxford: Basil Blackwell, 1947 (2nd edn); Nurkse, Ragnar, Problems of Capital Formation in Underdeveloped
7
Countries, Oxford: Basil Blackwell, 1953;
Morishima, M., The Economic Theory of Modern Society, Cambridge University Press, 1976. Also see Additional Note h for Keynes' voluntary unemployment. The relationship between machinery and workers in a modern industrial society resembles that between warships and sailors in a modern navy. If the warship should sink the soldiers are either saved and assigned to another ship, or are defeated; similarly a modern worker without machinery is either re-employed in a new factory possessing machinery or made unemployed. Workers can no more initiate the production of goods with their bare hands than sailors can swim to the attack on an enemy vessel.
190
Is full employment possible?
machinery and capital equipment), labour will be in demand, and the volume of capital in existence will determine the maximum limits of the demand for labour. For that reason, looked at from this point of view, unemployment comes into being because the amount of capital which exists is insufficient to provide work for all workers, and the accumulation of capital may be regarded as likely to remove unemployment if such accumulation should progress sufficiently. However, machinery is designed so as to require fewer workers to operate it the newer it is. That is to say the more advanced a machine is, the less its capacity to absorb labour, which means that the amount of capital sufficient to absorb all labour will be enormous and, therefore, will never be able to be accumulated. Unemployment resulting in this way from a shortage of capital, in the light of capital and labour being complementary, is known as Marxian unemployment (i.e. the reserve army of labour). However, an analysis of unemployment along these lines tacitly assumes that 'there exists sufficient effective demand to put to work all the capital which exists'. Yet it would be foolish to put to work all the capital which exists and produce up to a level where sales are impossible at a time when this sort of demand does not exist, so in such a case part of the capital is bound to lie idle. Simultaneously the number of workers employed will be reduced, and the rest become unemployed. That is to say unemployment will arise even where capital still remains to be used, as long as there is insufficient demand for goods. This kind of unemployment is Keynesian unemployment. On this occasion whether or not the demand for products is inadequate is determined by its magnitude relative to the size of the capital actually in existence.8 Where the given demand is such that even the full operation of all available capital will not fulfil it there will be excessive use of capital, production costs will mount and product prices will rise. Since immoderate use of capital means at the same time workers having to work outside their normal hours (either by overtime or by the introduction of night shifts), wages too will rise. The inflated demand will be absorbed by a rise in prices and the real volume of demand for products will be equated with the amount of output produced by the over-used capital, establishing an equilibrium in the product market. In contrast to this the classical school of economists regard unemployment as resulting from wages not being at their equilibrium value. Unemployment appears because the demand for labour is less than the 8
We will suppose, however, that there exists the right number of workers to operate all the capital in actual existence. Without this number of workers wages would begin to rise at the point of full employment, producing inflation, and the excess demand for products would be absorbed through price increases.
Unemployment (I): Marx
191
available supply and for unemployment to be zero the demand for labour must be equal to the supply. Since both supply and demand for labour are regarded as being functions of the price of labour (i.e. wages), wages must be regulated in order to balance supply and demand. Where wages are higher than the equilibrium value, demand will be below supply and unemployment will result. Unemployment is therefore a transitional phenomenon prior to a rationalization of wages, and is likely to disappear before long. Despite this the permanent and chronic existence of unemployment in the actual economy is attributed by the classical school to the fact that as a result of external stimuli affecting the economy from one moment to the next, the equilibrium value of wages is forever changing, and even if wages are adapted in accordance with this the speed at which adaptation takes place will be less than the rate at which the equilibrium value itself fluctuates.9 Preparations for analysis In this section and the following one we will attempt to analyse the three kinds of unemployment - Marxian, Keynesian, and that of the classical school - within the framework of our model of modern industrial society. To take Marxian unemployment first, we may write the equations for equality of demand and supply for consumer goods and for capital goods as Xx = D\ + D\ + D[ + Ex + Gj
(1)
X2 = I2 + E2 + G2
(2)
respectively. If we assume that the consumption expenditures of workers and entrepreneurs, pxDwu pxD\, are linear functions of their respective disposable incomes (1 - tw)W and (1 - te)allv/e obtain PiZ)ir = c w (l-r M ,)W + p1yM,, PiDi = ce(l-te)an
+ Plye
(3)
where cW9 ce are workers' and entrepreneurs' marginal propensities to consume, and yw, ye are the quantities of consumer goods which workers and entrepreneurs would demand if their incomes were zero. 9
The neoclassical school, which currently occupies the position of orthodoxy in the world of academic economics, has assumed the mantle of the classical school with regard to many fundamental points, including the question of unemployment; at the same time it has refined and modernized the theory of the classical school. For a classic exposition of neoclassical theory see, for example, Hicks, J. R., The Theory of Wages, London: Macmillan, 1963 (2nd edn).
192
Is full employment possible?
Throughout the following all these coefficients are assumed to be constant. In addition we have W = w(a3lXx + a32X2) + w(M + Nb)
(4)
all= am{cxXx + X )
where a, which is also assumed to be constant, represents the rate of distribution of profits to entrepreneurs. If we substitute (4) into (3) and then (3) into (1), we can write (1) as Xx = bxXx + b2X2 + ux
where bx = cw(l - tw) — a3X + ce(l - te)a ^
Pi
(5)
Pi
b2 = cw(l - tw) — a32 + ce(l - te)a — Pi Pi
(6)
ul = yw+ye + D\ + El + G1 + cw(l-tw)-(^
+ Nb)
(7)
Pi
In among these the demand for consumer goods of rentiers D[ depends on their disposable income, i.e. their after-tax interest income from the bonds and time deposits which they held at the beginning of the period, (1 — tr)(Ar + rBQ, so that it is independent of the output of consumer and capital goods during the period, Xx and X2. In the following it is also assumed that the government's and city banks' employment, N8, Nb, are zero. Therefore, ux is independent of Xl9 X2; similarly we may also regard u2 = I2 + E2 + G2
(8)
as independent of Xu X2. We shall below call uu u2 exogenous demand and the total demand including that part of consumption which depends on income, i.e. bxXx + b2X2 + ux and u2, the effective demand. The maximum level of employment which can be associated with a given capacity of capital equipment Marxian unemployment may be elucidated by solving the problem of what is the maximum amount of employment which can be created with a given stock of capital goods (machines and capital equipment) K, if there is no constraint on exogenous demands, uu u2. Any excess of labour supply over this maximum employment must inevitably remain
Unemployment (I): Marx
193
unemployed, as long as there is no change in the capital stock K. Marxian unemployment therefore results. This maximum level of employment can be determined by solving the following problem: maximize the amount of employment, a3XXx 4- a32X2, subject to the conditions a4XXx 4- aA2X2 ^ K
X2 = u2
(**)
where Xx ^ 0, X2 ^ 0, ux ^ 0, u2 ^ 0. Since there is no other constraint on ux and u2 than that they should not take on a negative value, the conditions (*) and (**) are equivalent to
respectively. Therefore, the above problem of determining the maximum level of employment is reduced to a problem of linear programming: maximize C13XXX + C132X2
(9)
subject to /C ^ dAXJCx ~\~ uA2/v2
\10)
Xx 2* &,AT, + b2X2
(11)
X2^0
(12)
(Note that the coefficients a's and b's are all positive.) In order to solve this problem it is important to see that the coefficient bx is a positive number less than 1. This can be shown in the following way. In the definitional expression (5) cw and ce must not exceed 1 because they are workers' and entrepreneurs' marginal propensities to consume. l — tw and \ — te take on their maximum value 1 when the tax rates tw, te are zero. The rate of distribution is greatest at a= 1 where profits are distributed in their entirety among entrepreneurs. Therefore bx cannot exceed (wa3X 4- mcx)/px, and this upper limit of bx is less than 1 because the numerator wa3X + mcx is only a part of (1 4- m)cx and the denominator pj is obtained by expanding the latter at the rate 1 4- tu so that the ratio is less than 1. Hence the constraint (11) concerning Xx may be rewritten in the following form ^ i ^ ~ — T x2
(ir)
194
Is full employment possible?
Since 1 > bx > 0 andfe2> 0? the right-hand side of the above expression gives a straight line rising to the right (i.e. 06 of Fig. 23). The constraint (11') requires that Xx should be either on this line or on its right. We can next give the constraint (10) on Xu X2 due to the shortage of capital stock the following geometrical interpretation. Letting K/aAX = a and K/a42 = fl',we will make a triangle by joining a and a' (see triangle 0aar in Fig. 23). It can then easily be shown that any point within this triangle satisfies the constraint (10) imposed by K.10 It is therefore clear that in Fig. 23 the edges and the inside of the shaded triangle give the set of all points which fulfil both these two constraints and (12). Which point within this set, therefore, is the one which maximizes the total amount of employment (9)? Let N be an arbitrary number which is positive. Dividing it by a3l and fl32, we obtain l = N/a31 and l' = N/a32, respectively. Joining / and /' which are positioned on the two axes we get a triangle 0//' (see Fig. 24). In just the same way as was demonstrated in the case of triangle Oaa' in 10
That at any point on the line aa' (e.g. point x) aAlXx + aA2X2 = K can be proved as follows. As is apparent from the figure, X2/(0a - X{) is equal to the slope of aa', i.e. to a4l/a42. Therefore given that 0a = K/a4l we get (K/a4l) - Xx
a42
therefore a4xXx + a42X2 = K. Since Xx = X[, X2>X2, at any point within the triangle Oaa', e.g. x' = (X[, X'2) the following inequality will obtain a4XX[ + a42X'2
Unemployment (I): Marx
195
Figure 24
footnote 10, it can be proved that for any point on the hypotenuse of 0//' the following equation holds a^Xx + #32^2 = N That is to say, the line //' is an equal-level-of-employment line. The higher the position of the line, the greater the total employment; it reaches a maximum at the vertex x° of Oax0.11 A point giving higher employment than this must always be outside the shaded triangle; it therefore violates at least one of the constraints (10), (11) of our linear programming problem. We shall denote the coordinates of the maximum employment point x° by X\ and X\. Unemployment due to a deficiency in capital Let the existing amount of labour be N and the maximum amount of employment be N° = anX\ + a32X°2; the difference N-N° then measures the degree of Marxian unemployment.12 This can be explained 11
12
We have above assumed that the slope of line aa' (a4l/a42) to be steeper than that of line //' (a31/a32). This means that we have assumed the degree of capital intensity in the consumer goods industry (a4l/a3l) to be greater than that of the capital goods industry (a42/a32). Conversely if in fact the capital goods industry was more capital intensive than the consumer goods industry it is line //' that would be steeper than line aa', and point a would be the point of maximum employment. This means that if Xl = a, X2 = 0 the maximum volume of employment for a given amount of capital K will be realized. Joan Robinson defined Marxian unemployment (i.e. the reserve army of labour) in the following manner. If A represents the total of labour available and N the amount of employment required to work the existing stock of capital at its normal capacity then A — A/" is, according to her definition, Marx's reserve army of labour. The similarity and dissimilarity between her and my definitions are apparent. See Joan Robinson, The Rate of Interest, London: Macmillan, 1952, p. 110, f.2. Also see Additional Note h.
196
Is full employment possible?
graphically in the following manner. Labour employed is measured along the horizontal axis and capital utilized along the vertical axis. a3l and fl41 stand for the labour and capital employed when one unit of consumer goods is produced; this point of employment is denoted by kx. Similarly, to produce one unit of capital goods, labour and capital of the amounts a32 and a42 respectively are employed, and this point of employment is denoted by k2. Since it is assumed that the consumer goods industry is more capital-intensive than the capital goods industry, the slope of 0kx is steeper than that of 0A:2, as shown in Fig. 25.13 Let us now multiply the unit vector 0kx by X\ and 0A:2 by X\\ we then obtain the points/and h, which represent the employment of labour and capital in the consumer goods industry and in the capital goods industry, respectively. The total employment of labour and capital in the whole economy is represented by point g which is the sum of 0/and Oh. Since point x° is on line aa' in Fig. 24 we have aAlX\ + aA7X\ = K so the ordinate of point g equals K. The abscissa of g, on the other hand, equals N° which is the greatest value N can take on subject to the condition that the amount of capital utilized does not exceed K.u The 13 14
The opposite case must have the following argument verified by construction of the appropriate figure. At first glance it would seem possible from Fig. 25 for there to be employment greater than N° where Xx is smaller than X\ and X2 greater than X\. However since
x\ = 1-b,
(contd.)
Unemployment (II): Keynes and the classical school
197
difference between the supply of labour N and maximum employment N° gives us the Marxian unemployment; this is the minimum level of unemployment inevitable, as long as the stock of capital is set at K whatever the magnitude of the exogenous demand uu u2. In the case of N being small enough to be located to the left of N{\ there is no Marxian unemployment. 4 Unemployment (II): Keynes and the classical school
Unemployment due to a deficiency in effective demand In the above we have entirely disregarded the exogenous demands for consumer and capital goods. In the actual economy where such demand plays an active role, the outputs of the two industries are determined at X\ and X\, which satisfy the equations for equality of demand and supply X\ = bxX\ + b&\ + ux X\ = u2 Because ux>0 and u2>0, (13) shows that X\ and X\ satisfy the constraint (11) of the linear programming problem. Furthermore, if the exogenous demands ux, u2 are not large, the effective demands X\, X2 are not large either, so that the other constraint (10) of the problem aAXX\ + aA2X\ ^ K
will be fulfilled; therefore the level of employment at X\, X\, which is does not exceed maximum employment N°. Thus N - Nl gives the actual amount of unemployment which is obtained when outputs X\, X\ are produced with exogenous demands set at uu u2. One part of that unemployment, N° — N\ is the unemployment due to a deficiency in effective demand (i.e. Keynesian unemployment) whilst the remainder N — N° is that due to a deficiency in the stock of capital (i.e. Marxian unemployment). We thus obtain actual unemployment = Keynesian unemployment + Marxian unemployment However, where the supply of labour is not that large and Nl
15
where Xl and X2 have this sort of value the condition (IT) for our linear programming ceases to be fulfilled. For that reason the maximum value is N°. The situation analysed by Keynes himself was one where the amount of capital was extremely plentiful, and therefore N°>N. His disregard of Marxian unemployment, therefore, was entirely consistent with theory.
198
Is full employment possible?
N
In the case of Keynesian unemployment being present both unemployed workers and idle capital K— K1 may co-exist as indicated in Fig. 26. As will be discussed in more detail later, classical and neoclassical economists consider that the capital stocks which exist are fully utilized and that the wage rate is automatically adjusted in such a way that full employment is realized. A state of affairs where capital is operating at full capacity and labour fully employed is often referred to as the state of Walrasian general equilibrium. In Fig. 26 the point W at which the horizontal line passing through K (the ceiling) intersects the vertical line passing through N (the wall) gives the Walrasian equilibrium point. From this sort of perspective, a Marxian state where workers are in surplus while capital is fully utilized, as well as the actual state of affairs A (a Keynesian state) where idle capital and unemployed workers co-exist, are regarded either as abnormal or as states of disequilibrium in transition towards the Walrasian equilibrium. It is clear, however, that the economy will stay at point A as long as exogenous demands uu u2 remain unchanged. Consequently, whether or not the actual point A will move and converge on Walras' point W depends upon whether ux, u2 will change so as to bring about such a convergence. Is unemployment a temporary phenomenon? From this viewpoint let us examine the components of ux and u2. First, exports Eu E2 and government demands Gl9 G2, N8 depend on the decisions of traders abroad and on those of the government, so that they
Unemployment (II): Keynes and the classical school
199
are regarded as given for the economy. Secondly, the demands yw, ye which workers and entrepreneurs need to subsist when their incomes are zero are regarded as constant in the short term although they may change in the long run. In addition it is inconceivable that in the presence of unemployment the city banks will increase their employment Nb with the intention of absorbing unemployment in the economy. Nb is therefore regarded as constant. Consequently, whether or not the 'modern industrial society' has an automatic adjustment mechanism towards the Walrasian general equilibrium is entirely dependent on whether investment 72 can carry out that kind of function, because, as we shall see in section 5 below, the wage rate has no power to create such an effect, even though it may be perfectly flexible. First, investment I2 which has so far been assumed to be constant depends, in the real world, on the rate of interest. If this is low, the costs of borrowing money for investment are cheap, and investment projects which would have been discarded as unprofitable will be carried out. The effect of a reduction in the rate of interest on investment may be considered to be rather large in economies such as Britain and Japan where investment relies on external funds rather than on being financed by the internal funds accumulated by the enterprises themselves. However, there is a limit to such an effect; in particular, where economic activities have cooled down and it is believed that there is no prospect of a boom for a considerable period in the future, no enterprises will invest since they will make a loss even if the interest rate drops considerably (and even if in extreme cases the rate is zero); there will therefore be only a few investments which can be increased. Thus the modern industrial society is not furnished with an adjustment mechanism which makes the economy move automatically towards a state of full employment (a point on the N wall), to say nothing of the Walrasian general equilibrium (the intersection of the N wall and the K ceiling). Consequently, to achieve full employment the government must constantly make a conscious effort to adjust its demands G l 5 G 2 , N8 so that the actual point A approaches point W or a point on the N wall. Full employment is not produced naturally as the result of competition and laissez faire; it is something which can be realized only through a government's well-organized expenditure policy. However, such a Keynesian policy is not a panacea for establishing full employment. It is true that where there is an abundant capital stock K, N will be smaller than N°, and only by manipulating Gu G2 can point A be brought to point W or a point on the N wall. But where K is comparatively small A will bump against the K ceiling before it reaches the N wall. As a result a shortage will arise in the stock of capital and, as
200
Is full employment possible?
will be seen later, inflation will occur. This is because Marxian unemployment cannot be relieved by a Keynesian expenditure policy. The classical (neo-classical) view of unemployment The neoclassical school, which is a modern version of the classical school, derives the function of demand for labour from the marginal productivity theory of the firm. Ignoring raw materials, let us assume that commodities are produced only by labour and machines. We then obtain the production function X = f(N, AT), where Xstands for output, TV for labour input, and K for the number of machines employed. The available number of machines is given as K. If we assume that enterprises use as many machines as are available, then as far as the production function is concerned K may be regarded as fixed, so that X is a function of a sole variable, labour. It can therefore be written as f(N). Profits will be pX—wN, where p is the price of products and w the wage rate. Since = f(N)=\
f'(N)dNandN=\
dN
we can write profits as n=
(pf'(N)-w)dN
(14)
f'(N) is known as the marginal productivity of labour because it shows the rate at which output increases for each additional unit of labour whilst pf(N) is called the marginal value-productivity of labour. In order to increase labour input by one more unit, it is obvious that additional wages w must be paid; if the marginal value productivity of labour exceeds the wages (i.e. if pf(N) - w>0), the additional labour will give rise to an increase in profits; in the converse case it will bring about a decline in profits. Thus, as long as pf'(N) - w > 0 the enterprises will employ more labour, and such an expansion stops when the equation pf'(N) - w = 0, or f'(N) = w/p
(15)
is established.16 Since the classical school considers that enterprises employ sufficient labour so as to maximize their profits their demand for 16
It is assumed below that as the volume of employment N increases the marginal productivity of the final unit of labour declines, i.e. that/'(TV) is a decreasing function of TV (the law of diminishing marginal productivity).
Unemployment (II): Keynes and the classical school
201
labour is obtained by solving (15) with respect to N. N is a decreasing function of the real wage rate w/p, because if (15) holds at No for (w//?)o, then we have f(N0)<(w/p)i where (w/p)l is larger than (w/p)0; this implies that the employment of labour up to the level of No is over-employment. Enterprises will halt their input of labour at Nu which is less than No. If we thus assume the neoclassical theory of the firm, it is clear that the demand for labour is a declining function of the real wage rate; therefore, full employment is established when the real wage rate adjusts itself so that the demand for labour becomes equal to its supply N. According to neoclassical doctrine, when the product price/? and the wage rate w are given, the supply of products X and employment of labour to produce them N are determined independently of the effective demand for products; it may therefore seem at a glance as if full employment is realized independently of effective demand by adjusting the real wage rate. Even with the neoclassical school, however, matters are not that simple. Except in a totally unrealistic world it is impossible to imagine a labour market where workers bid the real wage rate up and down; in a more realistic economy where, even though trade unions may not yet be formed, labour contracts are made in terms of the money wage rate it is extremely difficult to regulate the real wage rate. Even if the money wage rate falls because unemployment exists, the real wage rate will in fact increase if the price of products p falls at a rate higher than the fall in w, so that the demand for labour will shrink bringing about a growth in unemployment. Thus, in neoclassical theory, the problem of unemployment depends on the movement of the price level p. Quantity theory of money In neoclassical theory the level of prices is regarded as being determined by the equation of the quantity theory of money MV = pX
(16)
where M is the existing amount of currency, while V represents its velocity of circulation over a fixed period. Since goods (products) are handed over in exchange for money there is no reason why MV cannot be regarded as representing the demand for products in terms of an amount of money. Assuming that X is total production over a fixed period, since output must be sold, the price at which this output is supplied will be pX, and (16) can be regarded as the equation for
202
Is full employment possible?
S' 0
X1
Figure 27
balancing the supply of and demand for products in money terms. Therefore the balance of the volume of supply and the volume of demand for products is given as (17) Output X is an increasing function of employment N, and since this latter is a decreasing function of real wages, X becomes an increasing function of p/w. If we now regard the amount of money in existence and the velocity of circulation as given, the total volume of demand for products (the left-hand side of (17)) will be a decreasing function of the price, and if we regard the wage rate w as being fixed, the supply of products will be an increasing function of the price p. In Fig. 27 these functions are shown by DD and SS - the total demand and total supply curves respectively, and the level of prices is determined where the two curves intersect. Now if the amount of labour needed to produce the equilibrium output determined in this way is less than the available supply of
Unemployment (II): Keynes and the classical school
203
labour, and if the money wage rate w falls due to the resulting unemployment, then the amount of output for the same price will increase, moving the supply curve 55 after the fall in the money wage rate to the right, e.g. to the position of 5'5'. At the new equilibrium point the price p1 will be lower than the old equilibrium price p°, while the new equilibrium output X1 will be greater than the previous one X°. For that reason the volume of employment in accordance with the new equilibrium point will be greater than the previous one, and unemployment will decrease. That is to say the fall in wages w gives rise to a decrease in the real wage rate and hence it serves to expand the production, resulting in the absorption of unemployment.
The influence of interest rates The emergence of such a favourable result, however, rests on the assumption that the rate of circulation Fis constant. However, Fcan be regarded as an increasing function of the interest rate because if interest rates are high people-will try and keep their assets in other financial forms which produce a high interest rather than in money which gives them no interest, so they will part with their money all the more easily, and the rate at which money circulates V will increase. If this should be the case, the way in which the volume of employment changes will alter depending on what sort of influence the fall in employment has on the interest rate. For example, if V should decrease in conjunction with a fall in wages (with interest rates also falling), then curve DD in Fig. 27 will shift to the left, resulting in its point of intersection with curve 5'5' being located to the left of X1, or even, should the reduction in V be considerable, to the left of X°. In the latter case, the volume of employment in line with the new equilibrium level of output is less than the old volume of employment, and unemployment in fact increases despite a fall in wage rates. That means that what effect a change in the wage rate will have on unemployment will depend on the repercussions of the change in the wage rate upon the interest rate, and for that reason the problem of employment is not an isolated problem of the labour market but a problem of the whole economy, including the machinery for the determination of interest rates. Yet although an economist of the classical school might believe that unemployment is produced because wage rates are too high and that greater competitiveness in the labour market might remove unemployment, such economists have been unable to entrust everything to the flexibility of wages and unable to look on the unemployment problem as an isolated labour market problem.
204
Is full employment possible?
5 Wages and unemployment
The modified classical system We have above been interpreting the quantity theory of money equation (16) along income theory' lines. That is to say we assume that MV is equal to total income Y in that period,17 and that if all income during that period is directed towards the purchase of goods then the demand for products during that period will be Y/p. Since (16) can be written Y/p = X
(18)
the quantity theory of money equation can be interpreted as an equation balancing the supply of and demand for products: it therefore hides behind it the proposition (hypothesis) that 'the prices of products are determined on the market so as to equalize the demand and supply for such products'. However, there exists no such market for industrial goods; if we bear in mind the fact that when such commodities are shipped from the factory to the market their sales price has already been more or less decided, it must be recognized that although the above hypothesis has been traditional to economics, it is not one appropriate for an explanation of the movements of the economy in a society where the main products are industrial ones. So once we have made the decision to reject this hypothesis it is essential to find an alternative mechanism for the determination of prices. One revised version which comes easily to mind is to combine the determination of prices according to the full-cost principle and the Phillips curve and use them to reconstruct the wage adjustment mechanism of the classical type, but in this kind of 'modified classical system' wages and prices are likely to fluctuate in the following manner. If we look first at the Phillips curve, if the rate of unemployment is u and the rate of increase in wages vv, then vv will be a decreasing function of u. That is to say the higher the rate of unemployment the lower the rate of increase of wages, and when inflation is expected the rate of wage increases will be greater than is the case when inflation is not expected, by a margin equal to the expected rate of inflation. Therefore the wage rate will be regulated in accordance with the formula w=f(u)+pe, e where p represents the expected rate of inflation, i.e. the expected value of the rate of increase in the price level. 17
In his paper 'The Quantity Theory of Money - A Restatement', in Studies in the Quantity Theory of Money (ed. by M. Friedman) 1956, The University of Chicago Press, pp. 9-11, Milton Friedman proves that Y= MDV holds identically, where MD is the volume of demand for money. Hence when the demand for and supply of money are equal (MD = M), we get Y= MV.
Wages and unemployment
205
Friedman believed that wage rates are regulated on the basis of this sort of formula,18 but we shall below consider the wage rate as regulated not according to the expected rate of inflation but in accordance with the actual rate of increase in prices at the time, i.e. w=f(u) + £p (£ is a given number)
(19)
This is because when the actual rate of price increases is 10% the employers' side is likely to emphasize that wage negotiations be conducted on the basis of the real 10% rise in prices even where labour unions predict that prices will rise by 12%. Conversely, where the union side has expected a rate of increase in prices lower than the actual rate of increase, when the time for wage negotiations comes round the union side will adhere to the actual rate and will not be inclined to back down. Predicted value plays a significant role when an individual comes to making up his mind, but in the case of negotiations to determine wage rates if such predictions are not to the benefit of the unions they will not be introduced into the discussion. If they are advantageous they will be rejected by the other side on the grounds of having no confidence in such predictions. Negotiations are conducted on the basis of actual values which have to be universally recognized. In formula (19) £ is a coefficient of realization, and represents the percentage of the actual rate of increase in prices which is reflected in a rise in wages. In usual cases £ has a positive value of 1 or less than 1. Where, on the other hand, prices are decided according to the full-cost principle, the rate of increase in prices will remain smaller than the rate of increase in wages. That is to say, since wage cost will be no more than a part of average cost, an increase in the wage rate gives rise to a less than proportionate increase in average cost, and since prices are in addition proportionate to average cost the rate of increase in prices will ultimately be less than the rate of increase in wages.19 Now if the ratio of the former to the latter is 77, we get p = T)w 18 19
(20)
Friedman, M., Inflation and Unemployment: The New Dimension of Politics, Occasional Paper 57, London: Institute of Economic Affairs, 1977. In our modified classical model the neoclassical theory of the enterprise and the full-cost principle are brought together in the following manner. If p and w are now given as p° and w° respectively, according to the neoclassical profit maximization principle the volume of employment N will be determined at NQ. Since p°f(N°) = w°N° + other cost + profits (*) we get • "* cost I ( v ) (contd.)
206
Is full employment possible?
where 77 is a positive number less than 1. In the following we assume that both £ and 17 are constant. Now if real wages are
The volume of demand for labour is a decreasing function of the real wage rate, and since the volume of supply of labour takes a fixed value N, the rate of unemployment u will be an increasing function of the real wage rate. Since, as was explained earlier,/is a decreasing function of u, the right-hand side of (21) ultimately becomes a decreasing function of the real wage rate 0, and for a OJ greater than of it must be f(u(oj)) < 0. In the light of (21), when of > OJ, 6>> 0, that is, o> increases towards OJ*, and if of < o), then
-
L
—
/W)
r
.
}
J
When w1 changes to w2, then given that m1, Nl and other cost remain constant, under the new equation (***) prices will be determined as /?2, and so on. Since the 17 calculated from (**) is not the same as that calculated from (***), in (20) below 17 is not fixed. For the sake of simplicity, it is however assumed in the discussion below that 17 is constant.
d (w\ — — 20
21
,:» = !—1 P
dw —p-
\W _ &
w
dp — dtp
dw dp — -JL dt
dt
\P) Friedman, M., 'The role of Monetary Policy', American Economic Review, Vol. 58, 1968.
Wages and unemployment
207
Figure 28
classical model unemployment can be removed by adjusting the wage rate, and ultimately there remains only the natural rate of unemployment, which as frictional unemployment cannot be absorbed. If one thus abandons the quantity theory of money and replaces it with the full-cost principle and the Phillips curve, the viewpoint of balancing the supply and demand of products which it contained now disappears completely from classical theory. If product prices and the wage rate are given, real wage rate will also be determined, and in accordance with this also the volume of output and employment which will maximize profits will be fixed; then the rate of wage and price increases are determined in accordance with the resulting rate of unemployment. Since in this kind of 'modified neoclassical' process (see footnote 19 above) the rate of price increase is determined by the full-cost principle alone, there is no mechanism which equalizes the demand and
208
Is full employment possible?
supply of products, which is the purpose served by the quantity theory of money interpreted along 'income theory' lines (18). According to the neo-classicals enterprises go along with the theory of marginal productivity in executing academic 'profit maximization production plans' drawn up with no regard whatsoever to whether or not they can sell their products, and it is foolish to accumulate mountains of stocks by this kind of sightless management. Enterprises must produce only those products for which there is effective demand, and out of the theory of marginal productivity and the principle of effective demand it is the latter, and not the former, to which enterprises should first and foremost adhere. The pursuit of efficient productive activity with the aim of achieving maximum profits, in order to produce goods for which there is no demand, is mere stupidity, of the same order as trying to satisfy one's hunger with cakes drawn in a picture.22 The effect of a cut in wages Exponents on 'income theory' tied up income Y with the total volume of money in circulation MV; in contrast Keynes developed this in the direction of national income expended. Looking at gross national income at market prices from the point of view of expenditure, it is equivalent to the sum of individual consumption, investment, government expenditure and exports, minus imports, and if we look at the Table of Economic Linkages (Table 6) the items in columns 3 to 8 of the first 2 rows do no more than represent Keynesian effective demand subdivided into the demand for consumer goods and the demand for capital goods. Since product prices pu p2 are all decided according to the full-cost principle equations, there is no alternative to regulating the volume of output Xi9 X2 if a balance in the supply and demand for both consumer goods and capital goods is to be achieved. The volume of production is determined according to the equations
22
Xx = bxXx + b2X2 + ux
(22)
X2 = u2
(23)
My own view is that the Phillips curve is neither essential nor useful to economic analysis. As will be explained subsequently if the wage rate changes then both the unemployment rate and the price level will change as well. The Phillips curve has been used to show the empirical relationship between these three rates of change, but such a curve tends easily to shift according to the vagaries of the situation, its position and slope being dependent on such things as the level of interest rates and exchange rates, and the size of export demand and government expenditure. Since these three rates of change are determined by the whole economic system, and not merely by their own inter-relationship, the very fluctuation of the Phillips curve is a normal state of affairs, and a curve of the kind assumed by Phillips and Friedman can obtain only in particular circumstances.
Wages and unemployment
209
and employment is decided in accordance with these. The coefficients bu b2 of these equations are given by formulae (5) and (6), and apart from the rates of income tax on workers and entrepreneurs tW9 te these also depend on other factors such as wages and product prices. (For exogenous demand uu u2 see (7) and (8).) Now where the volume of employment determined in this manner is less than the supply of labour, and unemployment is produced, and where there exist no labour unions obstinate and persevering enough to prevent workers from going along with the resulting fall in wages, is unemployment then, as a result, likely to decrease? In this sense can it perhaps be said that labour unions must, even if only a little, bear some responsibility for the existence of unemployment? In the following analysis of the effects of such a cut in wages we will assume that the items of exogenous demand wl5 u2 - (all excluding the final item of uu u[ = cw(l-tw)(w/pl) (N8 + Nb)) - have not been affected by the change in wages. However, wj, the final item of exogenous demand ux, and the coefficients of effective demand bu b2 are without doubt influenced by the change in wages and the change in prices stimulated by it. Is the influence on these factors likely to be that of a favourable opportunity for reducing unemployment, or alternatively, is the cut in wages likely to bring about a vicious circle which causes unemployment to increase all the more? In order to try and solve this problem, we will rewrite the formula for bu b2 in (5) and (6) in the following manner. If we first of all bear in mind the price equation pt = (1 + ^)(1 + m)ch (i = 1, 2), we get
T£i = Pl
« EL (i + *,)(i + m ) P l
(i = 1 2) '
and if we substitute this into (5), (6) we get bt = cw{\ - tw)a3i - + ce(l - te)a ™ ^ Pi (i + 0 ( 1 + m) Pi Of course, we have
(i = 1, 2)
Now if the wage rate w rises (or falls) prices px will rise (or fall) at a lesser rate, so the rise or fall in money wages will bring about either a rise or a fall in the real wage rate w/px. For that reason if money wages should fall in conjunction with the existence of unemployment bx will decrease and b2 will also decrease - as long as the effect on b2 of the
210
Is full employment possible?
change in relative prices pjpi is disregarded.23 Then as long as (N8 + Nb) is not very elastic - and there is no possibility of this being the case - ux will decrease as well. Now as long as 'exogenous demand' remains unchanged the output of capital goods X2 from (23) will also be the same, but because ux,bu b2 in (22) decrease this will give rise to excess supply of consumer goods, so in order to get back to a balance of supply and demand output Xx must be reduced. If money wages thus fall in conjunction with unemployment, a fall in prices less than the fall in the wage rate will result, which will mean that workers' real purchasing power will fall, sales of consumer products will be poor, and consequently there will be a contraction of production in the consumer goods industry. This means that even where the 'exogenous demand' u2 remains the same, the lower the money wage rate the weaker becomes the power to create employment - i.e. the employment multiplier falls and Keynesian unemployment increases. Conversely when there is an excess demand for labour and money wages rise the demand for labour is stimulated even further and the labour market becomes even tighter. This conclusion, namely that the more wages fall the greater the increase in unemployment, has proved hard to accept for those who regard labour unions as monopolizing the labour supply and who believe that it is the very inflexibility (downward rigidity) of wages which accompanies the existence of unions which is the reason for unemployment. Therefore even of those who have come to believe in Keynes' policy of employment through the promotion of effective demand, for many the conjecture that unemployment will decrease if wages are not completely inflexible downwards is an undeniable one. Our analysis above, however, demonstrates that a downward inflexibility of wages plays the role of an automatic stabilizing mechanism just 23
In fact if this effect is compared with the effect of w/pt it is very small. Now in the production of capital goods if expenditure on wages wa32 is 40% of total expenditure c2, even given that the mark-up rate m is as high as 40%, we obtain wa32 = mc2. On the other hand the workers' propensity to consume cw will be high (probably 0.9) and the entrepreneurs' propensity to consume ce will be low (let us say 0.4). The rate of tax on wage income will be low, while the tax rate on income from profits will be high (let us suppose for the time being that tw = 0.1, te = 0.3). Assuming that 10% of all profits are distributed to entrepreneurs (a = 0.1), these figures (which while imaginary may be regarded as completely appropriate) will mean that the first term of b2 will be 29 times the size of the second term. Whatever the case the second term will be very small in comparison with the first. For a change in p2/px to have a sufficiently large influence on b2 to reverse the effects of w/plf p2/pi must change over 29 times as much as the value of w/plf in the opposite direction. If we take the line that such a situation is, in the first place, impossible, we can see that b2 will change in the same direction as that of the change in w/p1.
Wages and unemployment
211
like unemployment insurance and progressive income taxes.24 Not only that, but the movements for wage increases conducted by labour unions at an appointed time every year even when unemployment exists, while it may well be difficult for an individual enterprise to comply with their demands, are the best and most rapid means of sustaining and promoting effective demand in the economy as a whole.25 They are not, as many people regard them, reckless, self-destructive acts on the part of workers which will serve only to bring about the destruction of industry. If wages can be raised sufficiently the consumer demand of workers depending on their wages will expand and the demand for employment will increase up to the level where capital (machinery) can be fully operated, and this will happen even where the fixed part of effective demand, that part we called 'exogenous demand', is not sufficiently high. Then Keynesian unemployment becomes zero, but Marxian unemployment based on an insufficiency of capital may remain unalleviated. Because these people are unemployed as a result of a shortage of capital goods, the promotion of effective demand by itself is not enough to put them to work. Prior to this, a start must be made by providing capital goods (machinery) for these people. Whether or not it is possible to make an accumulation of capital of this sort is no longer a question of whether wages are high or low. Influences on exports and the exchange rate We must not forget, however, that we have above assumed that export demand E1 and E2 do not alter even when wages have changed. This 24
25
When business conditions are deteriorating and unemployment has increased, unemployment insurance benefits will prevent a sharp decline in effective demand. Conversely at a time of boom, unemployment insurance payments are collected as a percentage of wages, thereby curbing effective demand and mitigating inflation. For the stabilizing effects of progressive taxation see exercise for Chapter 7. Even where a labour union is not particularly aggressive, the process explained below may enable wages to rise even when unemployment exists. As we have seen already the labour market is not homogeneous; it is to a certain extent split up on the basis of occupation or enterprise. Let us now suppose some part of the economy to be faced with some bottleneck. Let us now assume that a certain enterprise is fortunate enough to have sufficient demand to enable its productive activity to flourish more than usual. The labour unions in that particular enterprise will demand an increase in wages and the enterprise is likely to comply with their demand. This being the case unions in enterprises close to the original one in either occupation or regional terms will demand an increase in wages on the grounds of 'justice' irrespective of the financial situation in those enterprises. Many of these other enterprises will have no choice but to agree, albeit in a modified form. There will thus be wage increases in many enterprises, and when this happens the desire for fairness among workers will grow stronger and stronger. Wage increases arising in one part of the economy will produce wage increases across the board, stimulated by workers' demands that fairness should be upheld, and this will happen regardless of whether or not unemployment exists.
212
Is full employment possible?
supposition means that we have from the beginning failed to take into account one significant effect of a fall in wages. That is to say if wages w should fall product prices px, p2 calculated according to the full-cost principle, will also fall, which will result in a strengthening of the competitiveness of that country's products in foreign markets, and an increase in exports Eu E2. The increase in exports will then bring about an increase in 'exogenous demand' uu u2 and for that reason it follows from (22) and (23) that output Xu X2, and therefore the volume of employment, will increase. In view of the fact that this increase in effective demand through the medium of foreign markets resulting from a fall in wages is likely to be sufficient to offset the decline in effective demand on the domestic market, the fall in wages will end up by reducing unemployment. We should not believe, however, that the effect on exports is always strong enough to produce this favourable outcome (favourable at least as far as the classical school is concerned). The reason for this is that if the effect on exports is a significant one, the increase in exports will have a favourable influence on the foreign value of that country's currency (i.e. the value, q, of the pound vis-a-vis the dollar). Product prices in dollars qpl9 qp2 will rise and exports Eu E2 will not increase as much as had just been thought. Changes in exchange rates must particularly be borne in mind in the case of long term export effects. So where there can be little prospect of an effect on exports the direct effect of a fall in wages on effective demand within the country will be dominant, and the fall in wages will have an adverse effect on employment. Whatever the case, we must acknowledge Keynes' conclusion that even if wages are downwardly flexible full employment will not necessarily be realized.26 6 Labour and machinery
Luddism During the time of the industrial revolution in Britain, especially during the years 1811-1816, workers initiated a movement to destroy machinery in the belief that mechanization would deprive them of their jobs. Ricardo, too, in his famous 'On Machinery' posed the question of whether or not the substitution of machines for human labour was to the benefit of the working class.27 Since the effects of such a substitution 26 27
See J. M. Keynes, The General Theory of Employment, Interest and Money, London: Macmillan (1936), pp. 257-71. The Luddite violence reached a peak in 1811-1812. David Ricardo's On The Principles of Political Economy and Taxation was published five years later, in 1817. See The Works and Correspondence of David Ricardo, ed. P. Sraffa, pp. 386-97. Cambridge: Cambridge University Press (1983).
Labour and machinery
213
appear initially as a change in the labour input coefficient a3i and the machine use coefficient aAi (7 = 1,2), the above argument, which assumes these production coefficients to befixedmust admittedly be one which totally ignores the question of substituting machines for human labour. When the wage rate falls due to the existence of unemployment each enterprise will change to production methods which are all the more labour intensive. What we must ask is whether or not, when the wage rate has fallen due to the existence of unemployment enterprises change to more labour intensive production methods (where a3i is large), thereby increasing the demand for labour, and also whether or not the reverse occurs, with a rise in wages providing an incentive to each enterprise to cut back on labour. The most simple formal way of dealing with this problem is to measure the volume of machinery in use per unit of output by enterprise i, a4h along the vertical axis, and the volume of labour employed a3i along the horizontal axis, and to draw an isoquant on this plane. The isoquant is a collection of all the points, i.e. all the possible combinations of men and machinery, where it is possible for one unit of a product to be produced. As Fig. 29 shows the isoquant is described as curving up to the left convex to the origin. The fact that it ascends as it moves to the left is due to the fact that in order to decrease the amount of labour the amount of machinery in use must be increased, and its being convex to the origin is because the greater the reduction in labour is, the more difficult mechanization becomes, and the amount of machinery needed to reduce labour by a single unit becomes greater and greater. Now, given this kind of isoquant, any attempt to curtail labour as a result of an increase in wages must involve the use of a greater amount of machinery: conversely if wages fall (and therefore the use of machinery becomes relatively more expensive), the volume of machinery used a4i will decrease and the amount of labour used a3l will increase. However the relationship between machinery and labour is not a simple one of confrontation. Once a kind of machine has been decided on, a decision is made on the number of workers needed to operate it. Take the case of an airline flight, for example. The number of the flight crew (pilot, engineer, wireless operator, steward, stewardess, etc.) is more or less technically determined in accordance with the type of aircraft. An aeroplane cannot be flown without a pilot on the grounds that pilots' wages have increased, and however low wages may be it would be foolish to fly a plane with unnecessary engineers on board. In the case of stewardesses it may be possible to reduce their number a little, but here too a prescribed number is essential to guarantee the safety of the passengers in case of emergency, and their numbers cannot
Is full employment possible?
214 041
Figure 29
break that minimum on the grounds of an increase in wages. If the total number of crew is reduced, even if the number of aeroplanes is increased, the number of planes which can be operated will fall and output (the total number of passengers) will also fall. Conversely, however much of an increase there may be in the number of crew, if the number of aeroplanes declines this will result in crew members not working and output will simultaneously fall. The relationship between the number of aeroplanes and the number of crew is not the relationship of increase on the one hand meaning decrease on the other, such as is depicted in Fig. 29. It is the relationship where as long as the number of aeroplanes remains at aAi production will not increase even if the number of crew increases above a3h and as long as the number of crew remains at a3i output will not increase even if the number of aeroplanes is increased beyond a4i. Therefore the isoquant will not curve down to the right (or curve up to the left) but form an L shape taking a 90° angle at point (a3h a4i).
Similarly in the case of other machines as well the tie-up between machinery and labour is a fixed one. It isfixednot simply on the basis of quantity, but the workers operating a machine are severely restricted by the movement of that machine. Modern workers must fit in with the
Labour and machinery
215
rhythm of their machines, and work mechanically as if they themselves were just another part of that machinery. Though they may be regarded as free workers, while they are actually working they are completely restricted by the machinery, and in that sense could be said to be even more tightly controlled than were slaves. For that reason if there is to be any change in the relationship between labour and machinery it is the pattern of the machine which must be altered. This means that, in terms of the same example used above, the number of crew can only be changed if and when there is some change in the type of aircraft. Thus in order to utilize machinery instead of labour (or vice versa) the machinery in current use has to be discarded and new, more laboursaving (or more labour-intensive) kinds of machinery must be introduced. This necessitates investment in machines. Substitution between labour and machinery becomes possible only when one kind of machine has been substituted for another, and in that sense the fact that the Luddites chose not to demand that employers should hire a large number of workers, but rather to break the machines and thus prevent their use (or alternatively to call for the use of more labour-intensive kinds of machinery) demonstrates that they understood the essentials of the situation. It suggests that at the very least they had an appreciation of reality far closer to the mark than that of the neoclassical economists who assumed that the combination of labour and machinery could be altered with no problems whatsoever. Substitution by means of investment The production coefficients a3i, aAi are therefore fixed as long as machinery remains unrenewed. Though one may think that it is a good thing for wages to fall as a result of unemployment and for enterprises to take on more workers, in order for them to do this they must be equipped with machines of a kind which can absorb a large number of workers, and for this investment is needed. Investment decisions are not things which are made at frequent intervals, or on a daily basis. Moreover it is a matter of carefully judging whether that kind of machinery will really be the most advantageous well into the future. Consequently the effects of a fall in wages on investment are not large, nor do they appear very quickly. Only when it can be predicted that the situation of significantly low wages will continue for a considerable length of time into the future will there be any thought of changing to the kinds of machinery appropriate to such wages. Moreover even when the decision has been made a fair amount of time is needed until this kind of machinery can actually be installed and the changeover completed.
216
Is full employment possible?
On top of this a cut in wages does not have the opposite effect of an increase in wages. Since the employment of a large number of workers makes personnel management more difficult entrepreneurs basically tend to prefer 'labour-saving machinery' to 'labour-intensive machinery'. Therefore even though wages may fall, if the fall is only a small one a change will not be made to labour-intensive machinery. The use of this kind of machinery, even though it may bring about a reduction in total direct production costs, does not necessarily mean that total direct and indirect expenditure inclusive of personnel management costs and expenditure on dealings with labour unions will be lower than it was prior to the conversion of the machinery, since it will necessitate the supervision and work of a larger number of workers and more dealings with the labour unions. Consequently, even when wages have fallen and are expected to remain at a low level into the future, enterprises may still undertake labour-saving investment. In contrast to this when wages are rising the stimulus towards 'labour-saving mechanisation' is all the greater. Even if labour unions complied with 'the laws of the market' and accept low wages at times of unemployment, the effect of such a cut in wages can at the most be regarded merely as weakening the tendency towards 'labour-saving mechanization'. They are not sufficiently strong actually to increase the enterprise's demand for labour. A cut in wages does not bring about the substitution of labour for machinery, even as a long-term effect by means of investment. In Britain the peasants who lost their land as a result of the landowners' enclosure movement (1760-1820) had no alternative if they were to survive but to sell their labour in the cities as members of the proletariat.28 Their descendants are now being driven out of the factories by machinery; they are destined in future to be expelled from the factories by hordes of robots. A change in the relative significance of sectors In our two-industry model - consisting of the consumer goods industry and the capital goods industry - it has been tacitly assumed that enterprises belonging to the same industry are all of the same type. In actual fact, however, both the consumer goods and capital goods industries are made up of many sectors producing various kinds of 28
Under the first enclosure movement, which took place in Britain during the fifteenth and sixteenth centuries, land was enclosed to create pastures for the sheep which would supply the wool, the raw material for the woollen textile industry rapidly expanding at the time. The second enclosure movement, which took place from the mid-eighteenth to early nineteenth centuries, marked the enclosure of agricultural land by landlords and farming capitalists in order to transform British agriculture into a large-scale, capitalist farming system.
Labour and machinery
217
consumer and capital goods, and it must be acknowledged that it is in actual fact a long way from reality to assume that everything is completely homogeneous within this collective whole. While in terms of our classification enterprises making cloth and those manufacturing television sets both belong to the consumer goods industry, these enterprises are qualitatively completely different, falling into different sectors both in terms of product and of machinery used. The labourcapital ratio wa3i/p2aM therefore, is also different for each sector. For that reason the labour-capital ratios for each industry wa3l/p2a^ and wa32/p2a42 a r e the averages of the ratios for each sector within the respective industries, therefore if there is a change in the relative significance of sectors within an industry the average ratio for the industry will also change.29 However, a change in the relative importance of each sector is not based on a change in the wage rate. Where enterprises draw up and execute production plans aimed at maximizing their respective profits on the basis of given wage rates and given prices for the products, ignoring completely how great the demand for their products is likely to be, a fall in wage rates will cause labour-intensive enterprises to increase their production by a considerable amount, while enterprises which use only a little labour will hardly be affected at all. Consequently in this sort of case the relative significance of each enterprise (or each sector) within an industry will depend upon the wage rate, and when wages fall the industry as a whole is likely to become labour-intensive, because the relative importance of labour-intensive enterprises (sectors) increases. However, as has already been explained, any profit maximization plan which fails to take into account the effective demand for products is, as far as an actual enterprise is concerned, no more than an empty, academic theory, and rather than trying to implement such an empty theory enterprises will attempt to produce the amount of goods which will satisfy their given effective demand. For this reason what determines the relative importance of each enterprise within an industry is the way in which effective demand is distributed between enterprises, and the wage rate has an influence on this relative importance only in as far as it influences this distribution. It is therefore only an indirect influence. 29
Since enterprises within the same sector do not always use the same kind of machinery, the labour/capital ratio within each sector depends upon the relative significance of each enterprise within the sector. Elsewhere in this book, however, each enterprise within a sector will be assumed to be of the same type. Should there exist within the same sector enterprises of a different type, i.e. enterprises having a different labour/capital ratio, then they can be split off to form a separate sector.
218
Is full employment possible?
This kind of indirect effect is small; even given that its effect is a considerable one, it is an uncertain one. For example, a fall in wages will have a significant influence on demand mainly for consumer goods used by workers, and virtually no effect at all on the demand for high quality goods. Now some of the consumer goods used by workers are labourintensive, whereas others are not. Consequently even if wages should fall and the distribution of effective demand be altered in favour of high quality goods, no clear assertion can be made as to whether the demand for labour in the consumer goods industry as a whole will increase or decrease. That is to say, even taking into account the influence of a fall in wages on the relative importance of sectors within an industry the effects of this influence on the demand for labour of that industry are uncertain, and it may be either beneficial or harmful to the working classes. Although workers may try to protect themselves in response to the trend towards mechanization by accepting low wages, results of this which are likely to occur are a fall in production due to a fall in the coefficients of effective demand bu b2 and the consequential contraction of employment. We must therefore conclude, with Ricardo - though by means of a very different course of reasoning from him - that 'the substitution of machinery for human labour is often very injurious to the interests of the class of labourers'.30 De-industrialization We have up to now carried on our discussion without taking into account tertiary industry (the service industry),31 but it is a fact that as modern industrialization progresses the service industries develop in parallel with it. First of all financial operations must develop to supply the huge amounts of funds factories need to purchase fixed capital goods. Furthermore, since the modern factory system ties workers and employees to the factory for a long period it rapidly puts them into a position whereby many household services which they had formerly carried out themselves, as well as such things as laundry and beauty care, have to be entrusted to someone else, and the demand for meals outside the home also increases. In addition although the mass produc30 31
Ricardo, op. cit., p. 388. Tertiary industry is normally regarded as being the same as service industry, but in addition t o the proper service industries (finance, commerce, property, hotels, advertizing, medicine etc.), it also includes sectors such as transport, communications, electricity, gas and water, which have vast amounts of capital equipment to an extent which is comparable with the extent of capital intensity of the large enterprises of the manufacturing industry. Where the tertiary industries are referred to below, however, w e will mainly be meaning the service industries in the proper sense.
Labour and machinery
219
tion of goods is carried out in a centralized manner at the factory, these products are sold to consumers who live in a large number of different places, and for this to happen wholesale and retail systems must be perfected. And then when the income level of workers rises in conjunction with industrial development not only does their demand for various kinds of services expand in quantitative terms, it gradually becomes more and more refined in terms of quality. Moreover, if the production of consumer durables increases the demand for repairs and after sales service will also grow. Thus the development of secondary (manufacturing) industry promotes the development of tertiary industry,32 and since tertiary industry tends to be labour-intensive tertiary industrialization measured in terms of the distribution of employment is far more conspicuous than tertiary industrialization measured in terms of the distribution of output. The delay in the development of tertiary industry after secondary industry has been a help as far as modern industrial countries are concerned. The capacity of secondary industry to absorb labour has decreased along with mechanization, and as a result those individuals who have been unable to get jobs in secondary industry have swarmed into the tertiary sector. In many service industries it has been possible to set up even very small-scale enterprises, so many of these people have begun work as proprietors of their own business. In the kinds of business where no special expertise or qualifications are necessary an influx of people results in excessive competition, and these fields as a whole may end up by becoming a pool of disguised unemployment. Furthermore, since in tertiary industry in general, labour unions are in many cases either conciliatory or lacking any power, wages tend to be far more flexible than is the case in manufacturing industry. In contrast to this, in many specialized occupations (e.g. doctors, lawyers, accountants, surveyors) the respective unions (e.g. the B.M. A.) often exert pressure to make the qualifying examinations more difficult, thereby preventing the number of people qualified from becoming too great, in an attempt to prevent any fall in income in these professions. Despite this there is a tendency as the economy develops for the ratio of specialized and technical occupations within the service industries to increase. That is to say, in conjunction with economic progress the tertiary industries not only become proportionately larger, 32
For the development of tertiary industry in the economy see Clark, C. G., The Conditions of Economic Progress, London: Macmillan, 1957 (3rd edn); Kuznets, S. S., Modern Economic Growth: Rate, Structure and Spread, New Haven, Connecticut: Yale University Press, 1966; Kuznets, S. S., Economic Growth of Nations: Total Output and
Production Structure, Cambridge, Massachusetts: Harvard University Press, 1971.
220
Is full employment possible?
but also improve qualitatively. The development of a knowledgeintensive, specialist service sector will raise the demand for education. It has already been pointed out that one of the characteristics of the tertiary industries is that enterprises are in general small-scale and labour-intensive, and they are not highly mechanized. This characteristic has meant that tertiary industry has been able to absorb a very large number of the working population, but recently this has become less and less the case. The trend towards a larger scale in many enterprises in the service industries, and the accompanying development of new, labour-saving forms of organization, are manifest in the supermarkets and chain stores of the retail sector. Furthermore in other fields, including the financial sector, there has been a massive shift towards computers as far as business machinery is concerned, while in the retail sector sales methods have also been mechanized. In the future robots and computers will play a major role in medical treatment. So if the mechanization of tertiary industry advances much further we will soon end up with vast amounts of labour being released from there as well. Unemployment will become the overriding political issue, and who knows whether a new generation of Luddites might not start to destroy robots. It would seem that the ascetic attitude to life which required a person 'to work regularly every day', an attitude newly acquired by people in conjunction with the rise of capitalist society, is going to be destroyed by the machines and equipment produced by that very capitalist civilization. For the government to preserve and nurture this disposition towards hard work, which still exists among people, it must as far as possible take measures to realize full employment, or at least a situation coming as near as possible to it. This is the government's supreme imperative. A people driven to despair are likely to attempt a major rebellion. Failing this a people lacking this sort of energy will fall into a state of total self-abandon, losing both will and discipline; ruin will then be the outcome.
Fiscal policy
1 The investment multiplier
Effects upon industries of an increase in investment We have hitherto proceeded with our analysis under the assumption that the amount of investment I2 is given. In taking account of the future of business each firm will decide whether it should make an investment and, if so, on what scale it should be made. If prospects in the future were to improve and the firm invested more, such an increase in investment would influence production. In order to single out the effects of investment we assume in the following that investment alone changes, regarding all other exogenous variables such as exports Eu E2 and government demand Gu G2 as being kept constant. Analysis of such effects can be made by referring to Table 6. As we stated earlier, it holds for any column that the sum of all the elements entered in that column is zero, while the sum of the elements entered in one row (such as rows 1, 2, 9, 10, 11, 12) is zero, only if the sector corresponding to that row is in a state of equilibrium, whilst the sum of the elements is identically zero for the other rows. Therefore, if investment increases by AI2, we find from the zero-sum condition for row 2 that the output of capital goods must increase by the same amount. Thus we obtain p2AX2=p2AI2
(1)
This expansion of the capital goods industry will bring about a proportional increase in its wages and profits (see Table 6, column 2). But if we assume that unemployment exists both before and after the increase in investment, we may assume that the wage rate w will be constant despite the increase in employment. If wages increase, workers' income taxes will increase, but because their disposable income will also increase their demand for consumer goods will increase too. Where a given percentage (say, a) of the increased profits AJJis distributed to entrepreneurs, such an increase in the entrepreneurs' income aATTwill 221
222
Fiscal policy
bring about an increase in their after-tax disposable income, with the result that their demand for consumer goods will be stimulated. (Concerning the above argument see Table 6, rows 3 and 7, and columns 3 and 4.) The increase in the demand for consumer goods brought about by such an expansion of production in the capital goods industry will induce a similar expansion in the consumer goods industry (see row 1). An increase in the output of the latter industry gives rise to an increase of the income of its workers and entrepreneurs, which in turn induces them to increase their demand for consumer goods. Then the consumer goods industry must increase its output further to satisfy this additional demand. Let ADWX be the increment of the total demand for consumer goods of the workers of both industries and AD[ that of the entrepreneurs; we then obtain from row 1 of Table 6: (2)
where we assume that the banks do not change their employment Nb in spite of the expansion of both industries. Supposing that both workers' and entrepreneurs' consumption change in proportion to their disposable income, we have and pxAD{ =
ce(l-te)aAII
where the tax rates tw, te are assumed to be constant. (Of course this will be an unrealistic assumption because progressive taxation is a normal state of affairs, but for the sake of simplicity we assume constant tax rates throughout the following argument.) The increments of wages and profits, AW and ATI, are given by AW = wa3XAXx + wa32AX2
(4)
AII = mc1AX1 + mc2AX2
(5)
respectively. Substituting (4) and (5) into the consumption functions (3) and then further substituting (3) into (2) we obtain PxAXx
= bxPxAXx + b{ p2AX2
(6)
where bx = cw(\ - tw) — ax + ce{\ - Qcan — Pi Pi w b2 = c ( l - t )
a3 + P2
e
(
e
) c(lt)am
c Pi
(7)
(8)
The investment multiplier
223
A comparison of these bx and b2 with the bx and b2 defined in the previous chapter enables us to find that b'2 equals b2(px/p2), while the two bxs are identical. Equation (6) tells us how much of an increase in output will be brought about in the consumer goods industry when production is expanded in the capital goods industry. Solving (6), we have ^
(9)
where 62/(1 ~ ^i)> which gives the ratio of the amount of increase in the value of output of the consumer goods industry (pxAXx) to that of the capital goods industry (p2AX2), is called the propagation coefficient of output expansion. Effect on GNP How much will the expansion of production in the capital goods industry, brought about by an increase in investment, stimulate production activities in the national economy as a whole? This may be well investigated in terms of the multiplier effect, by which each increase in investment is multiplied in order to give the resulting expansion of Gross National Product (GNP).1 The gross domestic product of the economy at market prices may be defined as the amount which remains after subtracting the value of imports needed for production from total gross output; that is, pxXx + p2X2 - rpt (a5xXx + a52X2)
where the part in parentheses, i.e. F in Table 6, represents the total amount of raw materials and others imported from abroad to produce Xx and X2 (see Table 6, row 5). In addition, the rentiers of the country have assets overseas of the amount rplB£, and, by making use of these assets for production, obtain interest income rB£. On the other hand the country pays interest income fl^to the foreign rentiers. Therefore the Gross National Product (GNP) of the country - that is, the total value created by the production activities carried out domestically and abroad by the people of that country - is given by2 Y = PxXx+p2X2 - r^a5X P\ 1 2
PxXx
- r^a52p2X2 Pi
For a definition of GNP see Additional Note g. See the first equation on p. 287 below.
+ rB^ - & (10)
224
Fiscal policy
If we write rPt
Pi
51
1,
Pi
these represent the proportion of the product prices px and p2 respectively accounted for by the cost of imported raw materials and fuel. In view of (1) and (9), we obtain from (10) (11) This equation shows how much of an increase in GNP will result from the increase in investment, AI2. By writing AI = p2AI2 which gives the increase in investment in terms of money value, we at once obtain the ratio of A Y to AI from (11) - that is from the part in the brackets on the right-hand side of (11) - which is called the investment multiplier. The formula of the investment multiplier can alternatively be written in the following form shown in (11'). First, let j8w be the share of AW in AY; then
_AW_w(a3lAXl AY AY Similarly, the share of a ATI in AY will be <xAII_ am(cxAXx + c2AX2) ~AY~ AY Therefore, the shares of workers' and entrepreneurs' after-tax income (disposable income) are (1 - tw)f$w and (1 - te)fie, respectively. Let mc be the average of workers' and entrepreneurs' marginal propensities to consume, cw and ce, weighted by these shares; then mc = cw{\ - tw)(3w + ce(l ~ te)pe If we also define the marginal propensity to import as mF (with respect to GNP), then rp*AF /XiPiAXi + fi2p2AX2 ttiF =
=
AY AY We then easily find that formula (11) can be rewritten in the form as is given in Additional Note i: AI
1 — mc +
The investment multiplier
225
Most textbooks of macroeconomics use formula (11') rather than formula (11). However, it should be noticed that (11') cannot be used as long as the shares, ($w, /3e, and the propagation coefficient b2'/(l — &i), remain unknown. To obtain them, all the procedures which we have gone through above have to be followed. The approximate numerical value of the investment multiplier Coefficients bx and b2 take on the following approximate numerical values. Let the workers' marginal propensity to consume cw be 0.9 and that of the entrepreneurs ce, 0.4. The tax rates on wage income and on entrepreneurs' income, tw and te, are assumed to be 10% and 30%, respectively. The indirect-tax rate on consumer goods and that on capital goods are both 5% (that is, ^ = ^ = 0.05), while the rate of mark-up m is 30%. Supposing that the capital goods industry is more labour-intensive than the consumer goods industry, let us be concerned with a case where the wages per unit of output in the capital goods industry amounts to 50% of the price of the product, while the same figure is only 40% in the consumer goods industry, that is, w w — 0 31 = O.4, — a 32 = 0.5 3
Pi
Pi
Given these numerical values we can calculate both cjpi and c2/p2 a t 0.73.4 We know that the percentage of profits to be distributed to entrepreneurs is somewhat unstable, depending on the economic situation at the time; where only 10% is distributed to entrepreneurs (a = 0.1), we obtain &! = 0.330, &£ = 0.411 Provided that the marginal propensity to import of each industry with respect to its output is 10% (i.e. /^ = /x2 = 0.1), then the value of the investment multiplier can be calculated at 1.45. In the extreme reverse case where the total amount of profits is distributed to the entrepreneurs (i.e. a = 1), we still obtain only bx = 0.385 and b2 = 0.466 and the value of the multiplier remains at 1.58. Thus we find that the value of the 3
4
These figures are not 'realistic'figureswhich have been econometrically confirmed, but are used throughout this chapter as a guide in the derivation of various economic propositions. There is, however, no question of producing any specific proposition such as is crucially dependent on these particular figures. The adoption of the form of numerical examples in the analysis which follows is to spare the reader complex algebraic formulas and the object of concern in this chapter, as in the others, is to elicit general propositions. If we substitute f, = 0.05, and m = 0.3 in the price formula /?, = (1 + /,-) (1 + ra)c, we get 0.73 as the value of cjp^i = 1, 2).
226
Fiscal policy
multiplier does not react particularly sensitively to a change in the distribution coefficient a. It will normally be the case, as has been assumed above, that workers' marginal propensity to consume will be much higher than that of entrepreneurs, and that the tax rate on wage income is lower than that on profit income. For the sake of simplicity, however, it is assumed in most textbooks that cw = ce and tw = te, and these are represented by c and t, respectively. It is also tacitly assumed in textbooks that the percentage of respective product prices accounted for by wages is the same throughout both industries (that is, w w — 031 = — <*32
Pi
Pi
in our symbols). If we do this we obtain bx = b'2\ and it can be shown further that these ft's are equal to
b = c{\ where
l-c(l-f)(ft>+<MT)
/|_1-M
1-/1
Taking into account the fact that 1/(1 — /x) = 1 + mF (see Additional Note i) and the equation 1 = G)+ V+ IJL+ 7T+ T
which follows from the price-cost equations (where v and T are the proportion of product price accounted for by depreciation costs and indirect taxation, respectively), we can further rewrite formula (11") as follows
^iA
i c ( i o
"+-
+mF]
di'")
AI I \_ (D+ V+ 7T + T J J In order to be able to write the formula in this form, it is necessary for a number of unrealistic assumptions specified earlier to be satisfied. Despite this limitation (11"') clearly and conveniently shows in what sort of circumstances the value of the multiplier will be large, and in what sort of circumstances it will be small. First, a rise in the marginal propensity to consume c will produce an increase in the value of the
The investment multiplier
227
investment multiplier. Secondly, if the proportion of indirect taxes to product prices r and the income tax rate t increase, the value of the multiplier will be smaller. Furthermore, if the percentage of depreciation costs to product prices v is larger, for example, by reason of the adoption of more capital-intensive methods of production, or if the marginal propensity to import mF is larger because industries become more dependent upon imported raw materials and fuel, then the multiplier will be smaller. An increase in the proportion of profits held within firms (that is, a decrease in a) would also be the cause of a decrease in the value of the multiplier. Finally (11'") is reduced to the textbook form AY
1
AI
1 - c(l - t) + mF
if we disregard v and r as well as all internally reserved profits, so that v — r = 0 and a = 1. Therefore, in countries such as Britain and Japan where industries have a high dependence on raw materials imported from overseas, the investment multiplier should cet. par. have a low value. Also, in a so-called welfare state, like Britain, the tax rate tends to be high in order to provide public goods and services for most of the people, so that there is a tendency for its multiplier to be smaller than is the case in Japan, where the tax rate is low. This is true as long as all other things between the two countries are equal. However, it must be remembered that there is a counteracting factor against this tendency. Since there is no great need for the individual to save for the future in a welfare state, the marginal propensity to consume will be larger in Britain than in Japan. This will serve to increase the British multiplier and offset a considerable part of the decrease in the multiplier resulting from heavy taxation. The employment multiplier An increase in investment brings about an increase in employment in parallel with an expansion of production. The amount of labour employed per unit of output in the consumer goods industry is a31, and in the capital goods industry is a32, so when these industries are expanded as in (1) and (9), total employment (measured not in terms of the number of the workers employed but in terms of wages paid) will increase by the amount
AW = wAN = I— a31 -^— + — a32]p2AI2 \Pi
l~bx
p2
f
(12)
228
Fiscal policy
With the numerical values given earlier the proportion of wages to the prices of their respective products is 40% for the consumer goods industry and 50% for the capital goods industry, so that bx = 0.330 and b'2 = 0.411, and the employment multiplier is calculated at 0.75. That is to say, an increase in investment creates new employment which requires an increase in the total wage bill amounting to 75% of the original increment in investment. Where the percentage of the price of the product accounted for by wages is equal in the two industries (i.e. w/px a3i = w/p2 a32 = co), the employment multiplier is reduced to AW AI
I
1 + bx
which can be further simplified into AW
co
AI
14-6
where coefficients bx and b'2 are equal to each other (i.e. bx = b2 = b). How tofinancean increase in investment In what way may funds be procured for increased investment? As we have already stated an increase in investment brings about an expansion of production and thus an increase in profits, and while part of this is distributed to entrepreneurs the remainder is retained within companies and appropriated to cover a part of the investment expenses. In the above numerical example of a 10% distribution of profits to entrepreneurs, if investment is increased by £10 million, retained profits will increase by £3.18 million. In addition, that part of the costs of production accumulated within companies to cover depreciation costs will increase when production is expanded. In our numerical example the proportion of depreciation costs to price is calculated at 23% for the consumer goods industry and at 13% for the capital goods industry, so that the increase in depreciation funds amounts to £2.71 million. Thus with an increase in investment of £10 million, funds to the tune of £5.89 million (= £3.18 m + £2.71 m) will be retrieved and the remaining £4.11 million has to be procured by other means. In order to procure the necessary funds, the firms will either (i) issue new bonds or sell those they have on hand (that is, they reduce 8B' of column 6, Table 6), or else (ii) reduce the amount of cash, V, they hold. Therefore, other things being equal, an excess supply will be created in either the bond or money markets, or in both.
The investment multiplier
229
However, other things never are equal. When investment increases and production expands, the family budgets of workers and entrepreneurs will be affected because wages and profits will increase. After having paid income taxes from their increased income, they will consume a fixed proportion of the remainder - the proportion equal to their respective marginal propensities to consume, cw, ce; however, because their marginal propensities to consume are less than 1, their savings will also increase. In our model it is assumed for simplicity's sake that workers and entrepreneurs save only in the form of cash. (See columns 3 and 4 in Table 6. It is assumed that only the capitalists (rentiers) listed in column 5 save in forms other than cash, i.e. in bonds, time deposits and so forth.) Thus the demand for money from workers and entrepreneurs will increase and an excess demand will appear in the money market. The income of rentiers is the interest they receive which accrues from the bonds and time deposits they hold at the beginning of the period, so that this income will not be influenced by anything which happens during the period. Rentiers will pay taxes from this given income and will allocate the rest to consumption and various forms of saving, though the allocation will depend on the prices, interest rates and foreign exchange rates which prevail in that period. As long as there is no change in these, the rentiers' demand for bonds, foreign bonds, time deposits and money, as well as their demand for consumer goods, will not readily change in the period during which production is expanded. Next we must look at the foreign trade sector. In the following we assume that demand from abroad (namely, exports Ex, E2) is fixed. However, since imports Fwill expand in proportion to domestic output, a demand for dollars will arise to pay for the increase in imports. If foreign rentiers' demand for bonds issued within the country (denoted by 8$) increases, the dollars which they pay for these will offset, at least in part, the demand for dollars due to increased imports; but if the demand for bonds from foreign rentiers remains unchanged,5 then the total amount of demand for dollars for payment of imports will put pressure on the foreign exchange market. Consequently, foreign currency (the dollar) becomes more expensive, and the exchange rate of the dollar in terms of the pound sterling, r, will rise. Finally we have the government sector, where it is assumed that there is no change in the government's demand for each good, Gu G2, and 5
Since investment in that country is increasing and production expanding, foreign rentiers take a bright view of prospects there, so their demand for that country's bonds SB^may be regarded as likely to increase. Below, however, it is assumed that there is no change in SB?.
230
Fiscal policy
that its personnel expenses wN8 too do not change. Its revenue from taxes T will, however, increase with the expansion of production. Indirect taxes will increase in proportion to output, while the income tax revenue will increase more than proportionately, wherever progressive taxation prevails. Since government expenditure is constant, such an increase in government income has to go toward the purchase of company bonds or the redemption of government bonds issued earlier. (See Table 6, column 8.) Influence onfinancialmarkets It is clear from the above that the following changes will occur in the financial markets. First, as far as the bond market is concerned, the demand for bonds from both domestic and foreign rentiers, 8Br and 8Bf, will not change unless prices, interest rates and exchange rates change. The enterprises' demand for bonds 8Bl will decline because they must raise money to make good the deficiency in their investment funds (or alternatively the enterprises' supply of bonds will increase), whilst the government's demand for bonds will increase. We cannot therefore a priori decide whether an excess demand or an excess supply will prevail in the bond market as a whole. In our numerical example, however, an increase in investment of £10 million increases indirect taxes by £0.77 million and the income tax paid by workers and entrepreneurs by £0.75 million and £0.11 million, respectively, provided that 10% of profits are distributed to entrepreneurs. On the other hand, the amount of capital which the firms must procure is, as we have already seen, £4.11 million; so the bond market will be in a state of excess supply as long as firms try to raise funds mainly on that market. Unless the central bank or city banks accept excessive bonds, the price of bonds will fall. Because the yield on a bond ib is a reciprocal of its price, this causes a rise in ib. In our model we suppose that only rentiers can have time deposits. However, the amount of their deposits will be kept constant as far as concerns the period during which investment has been increased, because the rentiers are not influenced at all in that period by the expansion of investment and production. On the other hand, in the foreign exchange market, as we have already seen, an excess demand for dollars is produced, so the exchange rate r of the pound against the dollar will tend to rise unless the foreign exchange stabilization fund releases dollars. If the yield on bonds ib rises, the mark-up rate applied by the enterprises will also be increased. Also, if the exchange rate r rises, imported raw materials and fuel become more expensive. Consequently the enterprises will ask higher prices, pup2, for their products; this will
The effect of fiscal expenditures
231
in turn stimulate trade unions to demand a higher wage rate w. If wages rise pricespx , p2 must increase once more. And where wages and prices change, the investment multiplier and other repercussion effects which have hitherto been calculated on the assumption of constant prices and wages will have to be re-calculated given the new values of prices, wages, interest rates and foreign exchange rates. We shall therefore assume below that the financial institutions will keep interest and foreign exchange rates constant by buying up the surplus bonds and releasing sufficient dollars to satisfy the excess demand, so that the prices of consumer and capital goods will remain unchanged. Where the city banks buy bonds (that is to say, 8Bb of column 9 in Table 6 increases), the cash in their possession must be reduced by just that amount, and where the central bank buys bonds (that is, 8BC of column 10 increases) it must create money (i.e. increase 8MC) by just that amount. Alternatively, if the foreign exchange stabilization fund sells a certain amount of dollars, the equivalent amount of pounds will be absorbed into the fund (see column 11). Whatever the case, in order for the consumer goods and capital goods sectors to constitute a fixprice economy, demand and supply in the financial sectors must also be quantitatively regulated so as to maintain fixed prices (i.e. fixed interest rates and fixed exchange rates). Finally, there is the money market (column 12). We have already seen that the cash balances held by workers and entrepreneurs, Lw, Le, will increase, while those of enterprises and the city banks, L\ Lb, will decline. On the other hand the central bank will create money, while the foreign exchange stabilization fund will absorb money from the market. Thus in the money market both demand and supply fluctuate, but in view of Walras' law, wherever all other markets (for consumer and capital goods, bonds, time deposits and foreign exchange) are in equilibrium, the money market too must establish a state of equilibrium where the total demand for money equals the total supply. 2 The effect of fiscal expenditures
The influence of government expenditures on production There are three elements in government expenditure. The first of these is the government's purchase of consumer goods, the second its purchase of capital goods, and the third its expenditure on personnel. The effects on production of an increase in the government's purchase of capital goods are exactly the same as the effects of an increase in investment by enterprises such as we described earlier; it will not merely produce an expansion of production in the capital goods industry, but
232
Fiscal policy
also give rise to expanding production in the consumer goods industry. Hence if we take the same numerical example as previously the multiplier effect of the increase in government expenditure on capital goods will, needless to say, be 1.45 (assuming a = 0.1). However, in the case of an increase in government demand for consumer goods and government employment the volume of production in the capital goods industry will remain unchanged. It will be only the value of production in the consumer goods industry which will increase. First of all, should the demand for consumer goods increase, from row 1 of Table 6 we will get PxAXx
= cw(l - tw)AW+ce(l - te)aAII+pxAGx
(13)
Since the volume of production of capital goods is unchanged we have AW=wa3lAXu
An=mc1AXl
(14)
If we now substitute (14) into (13) and bear the definition (7) of bx in mind, then (13) can be written as pxAXx = bxpxAXx -\- pxAGx
Therefore PxAXx=-^—pxAGx;
p2AX2 = 0
(15)
l-&i
Where there is an increase in government expenditure on capital goods we get PxAXx =
-\p2AG2; (16) PlAX2=p2AG2 1 — bx (see previous formula (9)); hence it is immediately apparent that the effects of these two increases in expenditure on both industries are quite different. Needless to say the multiplier effect on GNP is totally different as well. Bearing in mind the definition of GNP (10) it will be in the case of (15)
*
£
v
1
r
' \ - b
(17) x
and in the case of (16) AY
b2 =(l~Ml)
p2AG2
1" ( 1 — /Lt2>
1 - bx
(18)
The effect of fiscal expenditures
233
Here /L^, /^ a r e the proportions of product prices pu p2 accounted for by imported raw materials and fuel. In our numerical example the value of the multiplier (17) is 1.34, which is less than the value of the multiplier (18), 1.45. However, if ^ ! = /X2,
2,
— = —
Pi
(19)
Pi
then both will be equal. Here cox, o^ express the proportion of the product prices pu p2 accounted for by expenditure on wages. That is (w\
(w\
W
\Pi)
and provided that both are equal, and if, moreover, cl/pl = c2/p2, then according to (7), (8) we get bx = b2, and when /xx = \x^ (18) will be equal to (17).6 Where the government increases its expenditure on personnel it stimulates the demand for consumer goods through the consumer activity of its new employees. For this reason the total amount of its increased expenditure on personnel does not go towards the demand for consumer goods; the demand for consumer goods created directly by the increase in government expenditure will be equivalent only to the amount obtained by multiplying the disposable income which remains after these new employees have paid income tax by the marginal propensity to consume. Hence PxAXx
= Cw^
tw)
wAN*,
p2AX2 = 0
(20)
\-bx Therefore, where expenditure on personnel increases by wAN8 the multiplier of government expenditure will be
In the case of our numerical example (21) will be 1.09, far smaller than the values of the other multipliers. Three employment multipliers Of the three kinds of government expenditures, therefore, that on personnel is the smallest in terms of its power to create GNP, while 6
Conditions (19) demonstrate that the outlay structures in the consumer goods industry and the capital goods industry are identical. In looking at the multiplier effect of government expenditure Keynes did not consider increases in demand for consumer goods and increases in demand for capital goods separately, which means that Keynes tacitly recognized that the cost structure of both industries was the same.
234
Fiscal policy
being the most powerful in terms of creating employment. If we calculate the respective employment multipliers for each of the three kinds of government expenditure we get the following. Measuring, first of all, employment in terms of the sum paid out in labour, and not in terms of the number of persons employed, we have wAN =
OJXOJAA^)
+ (O2(p2AX2) + wAN*
where
1 l-b b'2 l
(22)
l-bx
respectively. Where expenditure on personnel increases we have (20) and as, moreover, AN8>0, we obtain = tox— —+ 1 (24) 1 K } \-bx In our numerical example tox = 0.4, to, = 0.5, so if we calculate the value of the employment multipliers (22), (23) and (24) we will end up with 0.60, 0.75 and 1.48 respectively. Thus government expenditure on personnel has a far greater influence in terms of creating employment than does expenditure on the purchase of goods, so where the employment situation is deteriorating badly and there is an urgent necessity for the creation of employment, the most effective means of doing this is for the government to expand its own personnel. However, of the employment multiplier in such a case, 1.48, 1 (i.e. the principal part) would result from the expansion in the government sector, while the increase in employment resulting from expansion in the production sector would be no more than the remaining 0.48. By contrast, where there is an increase in government expenditure on goods, the growth in employment occurs only in the production sector. For that reason the employment multipliers in such cases, 0.60 and 0.75, must be compared with the 0.48, and as this demonstrates, as far as the production sectors are concerned increased expenditure by the government on personnel produces no more than a minimal increase in employment. The aim of a government fiscal policy is not merely the expansion of the volume of employment. If it is to be used as a means of expanding
The effect of fiscal expenditures
235
and stimulating the producer sector, then the government must also work for an increase in expenditure on the capital goods industry, whose multiplier effect on GNP is considerable and where the employment multiplier is also considerable. Therefore when we refer to expenditure policy below we will mean expenditure policy in the sense of an increase in the government's demand for capital goods. Given this, in what way is the government likely to be able to raise the funds for this sort of expenditure policy? Deficit financing
As is clear from column 8 of Table 6, when government expenditure (piGi +P2G2 + wN8) exceeds government income (which according to our model is the sum of tax income T and the interest income from the bonds it holds Bg) the excess must be made up for by decreasing the number of enterprise bonds in the government's possession or by issuing government bonds (i.e. by increasing —pb8B8). Supposing that government demand for capital goods should increase by p2AG2 — £10 million, the resulting expansion in production, and therefore the expansion in employment and increase in profits, are exactly the same as they would be in the case of an increase in investment by enterprises. For this reason the increase in revenue from indirect taxes and the increase in income tax from workers and entrepreneurs, is also exactly the same as the increase in tax receipts which would follow an increase in investment, such as was explained in section 1. As we have already calculated in our numerical example, indirect taxes increase by £0.77 million, income tax paid by workers by £0.75 million and income tax paid by entrepreneurs by £0.11 million, so that total government income increases by £1.63 million. (The government's income from interest remains unchanged.) This produces a deficit of £10 million - £1.63 million = £8.37 million in the government's accounts, and to make good this deficit the government must either dispose of the bonds in its possession or issue new government bonds. (Therefore A(8B8)<0.) After the workers and entrepreneurs have paid their respective taxes from their increased incomes, part of what remains goes toward increased consumption, while the remainder is saved in the form of money. The volume of money in their possession therefore increases. (AL w >0, ALe>0.) In the foreign trade sector there is an increase in imports of raw materials and fuel, so the demand for dollars to pay for this increase inevitably increases. (AD/>0.) The influence on the enterprise investment sector is as follows. The amount of money set aside for depreciation increases along with the growth in production (AH> 0), and at the same time profits too increase
236
Fiscal policy
(AFI> 0). Part of these profits is distributed to entrepreneurs, while the remainder is withheld within the enterprise and becomes a part of the funds of the investment sector. Where an enterprise has increased its investment the increase in the depreciation reserve and in retained profits which this gives rise to is not as great as the original increase in investment, and the enterprise will have had to raise the shortfall on the bond market or the money market. However, where the government has increased its demand for investment goods the enterprise investment sector will have an excess of funds, and this excess must be retained in the form of bonds or money. Hence either A(8Bi) > 0 or
ALl>0.
Since the increase in the depreciation reserve is £2.71 million and the increase in profits retained within the company is £3.18 million then the total amount of capital excess in the investment sector will be £5.89 million. Even if the whole of this sum goes into the demand for bonds it will still not be equivalent to the total of £8.37 million worth of bonds which the government is attempting either to get rid of or to issue in order to make good its deficit, so the bond market will be in a state of excess supply. There will, however, be a state of excess demand for dollars on the foreign exchange market. Finally, there is the money market. The cash balances of workers, entrepreneurs and the enterprise investment sector will all increase, so the money market will be in a state of excess demand. As long as the city banks and the central bank fail to buy up the excess supply of bonds the price of bonds will fall and the rate of yield ib will rise. Moreover, as long as the exchange stabilization fund fails to release dollars on to the foreign exchange market the pound will become cheap compared to the dollar and the exchange rate r will rise. Then once both ib and r rise prices pu p2 will alter, and the various effects produced by the increase in government expenditure must be recalculated at the new prices. Let us assume that the central bank adopts a policy of fixed interest rates, and that of the excess supply of bonds the whole of what remains after the city banks have made their purchases is taken over by the central bank. In this case the total sum to the value of £2.48 million (i.e. £8.37 million - £5.89 million)7 will appear on the money market in the form of either (i) a reduction in the cash balances of the city banks (ALb < 0) or (ii) money newly created by the central bank (A(8MC) > 0). On the other side, if the exchange stabilization fund releases dollars equivalent to the excess demand for dollars in order to maintain the 7
This supposes, however, that the investment sector assigns the total amount of the surplus funds to the demand for bonds.
The effects of a reduction in taxation
237
exchange rate, then the amount of pounds set aside by enterprises for the increase in imported raw materials and fuel (£1.61 million) will be assimilated by the exchange stabilization fund. For this reason the net amount of money realized onto the money market will be £0.87 million = £2.48 million - £1.61 million, which is equal to the increase in cash balances of workers and entrepreneurs (ALW + ALe).s 3 The effects of a reduction in taxation A cut in the rate of income tax on wages tw The government has one more weapon which it can use to stimulate the economy. This is a reduction in taxes. In our model four taxes are levied. The first of these is income tax imposed on workers. (The rate of this tax is tw = 10%); the second is the income tax on entrepreneurs (the rate of this tax is te = 30%); the third is the indirect tax levied on consumer goods (tx = 5%), and the fourth that levied on capital goods Let us analyse what sort of stimulus the economy will receive if the income tax on wages tw is reduced by 1% to make t'w — 9%. If the amount by which the tax rate is reduced is now Atw = —0.01, then the disposable income of workers will increase by — AtwW as a direct result of tax reduction. W here represents total expenditure on wages. The workers' demand for consumer goods will therefore increase by cw(—AtwW), and this sort of increase will bring about an expansion of output in the consumer goods industry, resulting in an increase in employment and profits in this same sector. Income tax at the new low rate t^ is levied on the wages of the increased number of workers, who devote cw = 90% of their disposable income to the purchase of consumer goods. Entrepreneurs too will devote ce = 40% of their disposable income after paying taxes on their increased income from profits, to the purchase of consumer goods. Since the growth in production of the consumer goods industry based on the reduction in taxes must exactly cover the increase in consumer demand resulting from the original reduction in taxes for employees, cw{—AtwW), and the increase in demand for consumer goods resulting from the expansion of production, it must follow that PlAXx 8
= b[PxAXx + cw(-AtwW)
(25)
In our numerical model ALW = £0.67 million, ALe = £0.15 million, totalling less than £0.87 million. It should be noticed that the errors are based largely on the shortening of C i/Pi= C2/P2= 0.7326 to 0.73, and the consequent abbreviation of vx = 0.2326, v2 = 0.1326 to 0.23 and 0.13 respectively.
238
Fiscal policy
where b[ - cw (1 - Q - a31 + ce(l - Qam ^ (26) Pi P\ Assuming that 10% of profits are distributed to entrepreneurs b[ = 0.334. If we compare this b[ with the value of bx before the change in the tax rate, 0.330, we will see there is not much difference between the two; so, this minimal difference will be disregarded below, and b[ regarded as equal to bx. Obtaining p\kX\ from (25) as above, and writing the amount of the reduction in tax from employed workers (-AtwW) as —AT0, we end up with the multiplier effect of the tax reduction on GNP as AY —AT0
1 = (1 - MI)
~c i — ux w
= 1.21
(assuming
Ml
= 0.1)
(27)
The employment multiplier is
-ni0
-• — a31 —!— cw = 0.54 px l-bi
(28)
What sort of influence, then, is such a decrease in taxes going to have on the government's revenues? Since the reduction in taxes brings about an increase in production in the consumer goods industry wages and profits will increase and the revenue from income tax will increase. If before the reduction in the tax rate the value of output of consumer goods was £150 billion, and that of capital goods £80 billion, then the value of wages W prior to the tax reduction was £100 billion and the original gross tax reduction (—AT0) (i.e. not taking into account the tax revenues from the increased wage income) will be £1 billion. Consequently it is apparent from the employment multiplier above that wages will increase by £540 million, and by the same calculation that profits will increase by £297 million and the income of entrepreneurs by £29.7 million. Tax of ^ = 9 % is levied on these increased wages and the income tax for entrepreneurs remains te = 30% as before, so the government will receive income tax totalling £57.5 million from these increased wages and entrepreneurs' incomes. The indirect taxes levied on the consumer goods industry will increase in proportion to production by £64.4 million. Taking into account this increased revenue from tax, then the initial reduction in income tax on workers of £1 billion is a net reduction in tax of £878.1 million. That is to say AT= £878.1 million. This, therefore, causes the government's budget to run into deficit.
The effects of a reduction in taxation
239
This deficit can be made good by the issue of government bonds. Alternatively, since in the enterprise investment sector retained profits and depreciation reserves increase by £267.3 million and £310 million respectively, there will be a total surplus of £577.3 million, and these surplus funds can be directed towards the purchase of bonds. The value of the excess supply of bonds is £300.8 million (£878.1 million - £577.3 million). The increase in the demand for money from workers and capitalists (ALW > 0, ALC > 0) and the increase in the demand for dollars on the foreign exchange market (AD/ > 0) is exactly the same as that occurring in the case of the government's spending by deficit financing. In the financial sector there is excess supply on the bond market, and excess demand on the foreign exchange and money markets, and in the absence of any attempt by the banks and the exchange stabilization fund to stabilize the interest rate and foreign exchange rate through quantity adjustments, the interest rate ib will rise and the price of pounds vis-a-vis dollars will fall. A cut in the rate of income tax paid by entrepreneurs te The same thing is true when there is a cut in the rate of income tax paid by entrepreneurs, but since entrepreneurs' income is small by comparison with wage income a cut in te is unlikely to have much effect unless it is a fairly sizeable one. Let us suppose that before the reduction in tax the value of production of consumer goods was £150 billion, and that of capital goods £80 billion, hence before the cut in tax profits amounted to £5037 million. Since we assume that 10% of this is distributed to entrepreneurs the income received by entrepreneurs before the tax reduction a/7 will be £503.7 million. For that reason, unless there is a massive reduction, where te = 0.3 is reduced by 0.199, i.e. by nearly 0.2, to make a new entrepreneurs rate of income tax ^ of 0.1 there is no way in which taxes can be reduced by £1 billion. Thus a reduction in the income tax paid by entrepreneurs can be one of the weapons of government fiscal policy, but it can be no more than a secondary weapon, weak in influence when compared with the effects of a cut in the tax on wage income.9 Furthermore, the multiplier value of this kind of tax reduction is far smaller than that resulting from a cut in the income tax payable on 9
In our model it is supposed that only that proportion of profits distributed to entrepreneurs is subject to income tax, while that proportion retained within the enterprise goes untaxed. In reality retained profits too are subject to income tax. If a cut in the tax on retained profits is implemented in conjunction with a cut in the tax on entrepreneurs' incomes the government will easily be able to make a sizeable cut in tax, so such a combined tax reduction is a powerful weapon of fiscal policy.
240
Fiscal policy
wages. Now let the bx corresponding to the new rate t'e be b\\ b\ = 0.322, and is close in value to the bx prior to the change in te. Thus if we regard b'[ as equal to bx and devise a formula which corresponds to (25) we get PxAXx
= b^AXx + ce{-AteaU)
(29)
Finding the multiplier value from this formula we get AY _
1
-AT~
^
wAN
w = — a3X px
-ATX
l-bxCe~ 1
^ ^A ce = 0.24
l-bx
where -ATX represents the original gross tax reduction -ATeaTI, i.e. not taking into account the increase in tax receipts produced by the expansion in production stimulated by the cut in tax. Because the multiplier value is small the increase in wages and profits based on the expansion in production will also be small, and the increase in the government's revenues from the income tax on both wages and entrepreneurs' income will therefore be small as well. In our particular case, where entrepreneurs receive a large reduction of £1 billion in tax, the increase in tax revenues derived from these two groups will be no more than £24 million and £1.3 million respectively, and since there will be an increase in revenues from indirect taxation of £28.5 million the net reduction in tax will be £946.2, far larger than where there is a reduction of income tax on wages. A cut in indirect taxation If the rate of indirect tax is cut prices inevitably change. If we write down the price formulae once again just to remind ourselves, this must fulfil p x = (1 + tx) (1 + m) (wa3x + pAaAX + rp*5a5X)
(30)
Pi = (1 + h) (1 4- m) (wa32 + pAaA2 + rp%a52)
(31)
Since the price of capital service pA must be equal to 6p2 (0 here is equal to the rate of wear and tear of capital goods) we can solve (31) with respect to p2 to obtain _ (1 + h) (1 + m) (wa32 + rp%a52)
A cut in the rates of indirect tax tu t2, will have the following effect on prices. Supposing first of all a reduction in the rate of indirect tax
The effects of a reduction in taxation
241
imposed on consumer goods, tx, as becomes apparent from (32) the price of capital goods p2 (and for that reason their service price pA as well) remains unaffected. However, we know from (30) that the price of consumer goods px falls in direct response to a reduction in tx. However, where there is a cut in the rate of indirect tax on capital goods t2 we see from (32) that the price of capital goods will fall. Given this the service price of capital goods p 4 will also decline proportionately, and hence, as (30) makes clear, px will also fall. This means that in this case a reduction in t2 causes a fall not only in p2 but in px as well, though the amount of the fall in the case of px is far smaller than that in p2. As long as the change in prices produces no alteration in the demand for capital goods 72, E2, G2 the amount of output of capital goods X2 will not change either. However, the endogenously determined demand for consumer goods of workers, entrepreneurs and rentiers is influenced by the change in prices even where exogenous demand (Eu Gx) is constant. A fixed proportion of their income goes towards consumer expenditure, and should the price of. consumer goods fall the amount of consumer goods which can be purchased out of this consumer spending will increase. That is to say the fact that people continue to spend a fixed proportion of their income despite a fall in prices means that consumers are increasing their demand for consumer goods at the same time as prices of the goods are falling. In addition, there are the exogenous parts of their consumption, yW9 ye, yr, which are simply assumed to be constant. This kind of increase in the demand for consumer goods from a fixed income can be analysed in exactly the same way as an increase in the marginal propensity to consume while prices remain constant. That is to say the increase in demand brings about an increase in the volume of production, resulting in an increase in both the volume of employment and in profits in the consumer goods industry and, hence, an increase in the incomes of both workers and entrepreneurs, stimulating a further increase in the demand for consumer goods. The volume of production is further increased to meet this secondary increase in demand, and the multiplier process is initiated. In our model all the factors which cause friction and serve to delay completion of the multiplier process, such as the period of production and the lags of wage payments, are disregarded, so the multiplier effect is achieved instantaneously. If the new equilibrium production of consumer goods is X[ then AXx = X[-Xx>0. Should the consumer goods industry expand production in this way employment, and hence wages, will increase, as will profits and the income of entrepreneurs. This will result in an increase in the income
242
Fiscal policy
tax received by the government, and as a result of the expansion of production in the consumer goods industry, indirect taxes will increase at the new low, indirect tax rates. Such increases will clearly be small compared with the direct reduction in indirect tax receipts resulting from a lowering of the rate of indirect tax, so overall the government's tax income will decrease. The government is likely to have to issue bonds to cover its deficit. Alternatively, because the enterprise investment sector increases its retained profits, and the depreciation reserve grows as well, it now possesses surplus capital and that sector's demand for bonds will increase. Part of the bonds newly issued by the government, therefore, will be taken up by enterprises, but in view of the fact that the surplus funds accruing to enterprises as a result of the tax cut are not sufficiently large to absorb the whole of the new national debt, excess supply will persist on the bond market. As long as the banks do not take up this excess part there will be a fall in the price of bonds and rise in their rate of yield ib. Should ib rise the demand for bonds from rentiers will increase, and the state of excess supply on the market will be eased. On top of this the expansion of production in the consumer goods industry will increase imports of raw materials and fuel. Hence demand for foreign currency (dollars) will increase and unless the foreign exchange stabilization fund releases dollars there will be a shortage and the exchange rate of the country's own currency against dollars r will rise. This will provoke a rise in the price in pounds of imported raw materials and fuel, and lead to an increase in the price of products as well. Derivative price rises such as these partially offset the direct effect of the reduction in indirect taxes on product prices (i.e. the fall in prices), and the resulting expansion in production AXX too ceases to be as great as was first thought. And that is not all. Up to now we have assumed that the volume of exports remains constant even when there is a fall in product prices, but if px should fall the volume of exports Ex will normally increase. An increase in exports will produce an additional multiplier effect. Where the exchange rate has risen and resulted in an increase in pu the prices of exports in dollars (pi/r) will fall, not just those of consumer goods, but those of capital goods as well; this will lead to an across the board increase in exports of both industries (AEi>0, AE2>0). The effects of this increase in terms of an expansion of production will be considerable. Thus the maintaining of a low value for domestic currency vis-a-vis foreign currencies is a significant factor in maintaining a high level of production, but since this kind of policy is not to the advantage of rentiers who hope subsequently to invest abroad, the banking
The multiplier effect of balanced budgeting
243
authorities and exchange stabilization fund are likely to bear in mind these rentiers' interests and prevent a rise in the dollar by supporting the domestic currency.
4 The multiplier effect of balanced budgeting The catalytic effect of tax increases
As we have already seen, the government possesses two methods whereby it can increase employment. The first of these is an increase in fiscal expenditure; the other a reduction in taxation. Both of these have the subsidiary effect of increasing the government's deficit. Providing the situation whereby the government needs to increase employment lasts only for a short time the deficit will not be a particularly large one, and there is no need to worry about this side effect, but where there is a chronic state of under-employment and the only means of coping with it is repeated measures of massive fiscal spending and tax reductions over a long period, then the government's deficit will build up to high levels. Should this happen the government will find itself annually paying out vast amounts of interest to domestic and foreign rentiers. Without over-issuing currency it will be impossible to continue the policy of increasing employment by means of increasing expenditure and making tax cuts. This shows that these two policies, of increasing fiscal expenditure and reducing taxes, are effective as short term policies, but, like doses of medicine, can have a disastrous effect if resorted to repeatedly. The problem is, of course, whether this sort of side effect cannot be removed. This means what needs to be considered is whether or not it is possible to expand employment without an increase in the government's deficit - for example whether or not it is possible by means of combining the two policies of which we are already aware to transform a policy which is basically likely to increase the government's deficit into one which will serve to expand employment while maintaining a balanced budget. Except in very special circumstances this will always be possible. Should the reverse of either one of these two policies (e.g. not a policy of tax reduction, but one of increasing taxes) be implemented it will put the government's budget in the black by way of compensation for reducing employment. If this surplus is used to cancel out the deficit produced by the other policy (the policy of increasing government expenditure), this kind of policy mixture will not put the government into the red. The net increase in employment will be the difference between the increase in employment produced by the increase in
244
Fiscal policy
government expenditure and the reduction in employment resulting from tax increases. Should this net increase not be positive, but 0, the employment effect where the two policies have been combined to maintain a balanced budget, will be nil, and this policy combination will not be effective as a measure for increasing employment. The 'very special circumstances' mentioned above are exactly this state of affairs. However, where the net increase is, conversely, negative, what can be done is not to combine policies of increasing government expenditure and increasing taxation, but to combine a policy of tax reduction with one of reducing government expenditure. Should the deficit resulting from tax cuts be cancelled out by the surplus produced by cutting down government expenditure then on the employment side there will be a net increase in employment of the difference between the increase in employment resulting from tax cuts and the decrease caused by cutting expenditure. Also, a net increase of this kind will be the converse of the difference between the increase in employment resulting from increased government expenditure and the decrease caused by increased taxation. Hence when the latter is assumed to be negative the former must be positive. The use of one policy (e.g. taxation policy, either increasing or reducing taxes) as a subsidiary policy which can be combined with the other major policy (an increase in government expenditure or a policy of cuts) to avert one of the various effects produced by that major policy is known as a catalytic mixture of policy, and a subsidiary policy used with such an aim in mind is known as a catalyst.10 Whether an increase in government expenditure should be the main policy and tax increases the subsidiary, or whether the major policy for increasing employment should be tax cuts combined with a catalyst of cuts in government expenditure, will, as has been said before, depend on whether or not the increase in employment per unit increase in government deficit resulting from the adoption of an expenditure policy is greater than the employment increase per unit increase in deficit where there have been tax reductions. In normal cases the rate of increase in employment per unit increase in deficit is greater with an expenditure policy than with a tax reduction policy. So an increase in government expenditure must be adopted as the main policy and one of tax increases used as the catalyst, and looked at from this point of view, the policies adopted by, among others, the Thatcher government (i.e. of cutting government expenditure and reducing taxation) must be regarded as a balanced budget policy which will reduce employment. 10
This is a different definition of catalyst from that used in chemistry.
245
The multiplier effect of balanced budgeting
2S 3 2
increase in deficit
Figure 30
Diagrammatic analysis of the catalysis
Let the increase in employment be measured along the vertical axis and the increase in deficit along the horizontal axis in Fig. 30. The origin represents the situation prior to the implementation of policy, i.e. a situation where neither employment nor government deficit is increasing. (The increase in employment will be measured below not in terms of numbers of workers, but by the amount of wages paid.) There are three methods of increasing government expenditure, namely (i) an increase in government expenditure on consumer goods, (ii) an increase in government expenditure on capital goods, and (iii) an increase in its expenditure on personnel. There are also three means of reducing tax, namely (i) reducing the rate of indirect tax, (ii) cutting the income tax on wages, and (iii) cutting the income tax payable by entrepreneurs or rentiers. Since where indirect taxation is reduced the whole organization of product prices is inevitably disrupted, analysis becomes somewhat complex. Below, we shall therefore limit our analysis to the case of reductions in direct taxation, especially a reduction in income tax on wages, and limit it in the case of increased government expenditure on the other hand to consideration of increases in government expenditure on capital goods. It is up to each individual reader to find out more
246
Fiscal policy
about the more complicated cases of increases in government expenditure on other items and the implementation of changes in other kinds of tax rates. Now for every unit of increase in government expenditure on capital goods the volume of employment increases according to formula (23). In the numerical example calculated earlier the value of the employment multiplier is 0.75. Therefore every increase of £10 million in expenditure on capital goods will produce an increase in employment of £7.5 million in terms of wages payable. On the other side the same increase in expenditure will produce an increase of £8.37 million in the government's deficit. The increase in employment (measured according to wage rate) per unit increase in government deficit is therefore 0.90. In contrast to this if the rate of income tax on wages is reduced to cut the amount by £1 billion from that currently paid by employees (i.e. —ATO = £1 billion), the increase in employment measured by the amount of wages paid will be £540 million; the increase in the government's deficit will be £878.1 million, therefore the increase in employment per unit increase in deficit will be c. 0.61. In Fig. 30 the line 0a represents the increase in employment per unit increase in deficit where government expenditure has been increased, while 0b shows the ratio of the increase in employment to the increase in deficit when there has been a reduction in the rate of income tax on wages. Point b is in a position below point a. Let us now suppose that a £1 billion increase in taxation on workers has been politically possible. This will put the government into the black to the tune of £878.1 million, and assuming the government uses this to make up the deficit resulting from its increased expenditure on capital goods there can be an increase in fiscal expenditure of (878.1/8.37) x £10 million = £1049.1 million. From this sort of fiscal expenditure there will be an increase in employment (in terms of wages) of 0.75 x £1049.1 = £786.8 million. On the other hand, because a £1 billion tax increase loses £540 million of employment; so the net increase in employment will be £246.8 million. (In Fig. 30 point b shows the increase in employment and increase in deficit in case of a £1 billion tax reduction so if we extend 0b by exactly the same amount in the opposite direction we will get point d which will show the reduction in employment and increase in surplus resulting from a tax increase of £1 billion. Point a represents the result of an increase in expenditure on capital goods which will produce an identical increase in deficit. The point where vectors 0a and Od meet is 0c, and the height of 0c shows the net increase in employment from the balanced budget mixture of the two policies.)
The multiplier effect of balanced budgeting
247
An increase in taxes or a reduction in taxes? Should the line Oa showing the rate of increase in employment per unit of deficit caused by fiscal expenditure policy be steeper than line Ob showing the rate of increase in employment per unit increase in deficit caused by a policy of tax reduction, which was the case in our numerical example, then expenditure policy can become the main policy and tax increases be used as the catalyst, and employment in the economy as a whole can be expanded while the government maintains a balanced budget. If, however, Ob should be steeper than Oa tax reduction must be the main policy and cuts in fiscal expenditure used as the catalyst. It is therefore extremely important in terms of devising employment policy to be aware of whether it is Oa or Ob which has the steeper incline. In our numerical example above, as a result of the numbers we chose for our various coefficients, Oa is steeper than Ob, but this can be proved to be generally the case even where alternative figures are assumed.11 In all cases, therefore, it is possible to expand employment by using tax increases as the catalyst and increasing government expenditure, while still maintaining a balanced budget. If the government increases taxes up to the maximum politically possible and spends the amount gained in this way on making good the deficit in fiscal expenditure then it will be able to create the maximum amount of employment. In order to make an increase in the income tax on wages politically possible it will probably be necessary to increase the rate of income tax levied on entrepreneurs and rentiers as well. Needless to say these tax increases too would bring the government surpluses and provide it with the revenue for further increases in expenditure. The economic significance of a policy of tax reduction Despite this, in recent years many countries have been attempting to pursue a policy of simultaneously cutting taxes and reducing government expenditure, under the pretence of rationalizing the scale of government which they regard as having become excessively large through year after year of Keynesian policy. The machinery of government is operated in a bureaucratic manner, so there is no doubt that cutting down the government's budget is indeed a very difficult problem, but if this kind of policy is implemented what sort of situation does it give rise to? What is first and foremost absolutely clear, is that there will be a decline in employment and an increase in unemployment. Any government which adopts this sort of policy will inevitably be heavily criticized 11
See Additional Note j .
248
Fiscal policy
for forcing unemployment on the people. Since the amount of output of consumer goods will simultaneously fall, enterprise profits will decrease, and so will the income of entrepreneurs. The situation regarding both workers and entrepreneurs will clearly deteriorate. At the same time the enterprise investment sector will encounter difficulties in securing funds, and will be forced to issue bonds in order to secure capital. Since the government maintains a balanced budget there is no need to issue government bonds, but the bond market suffers from excess supply. In the absence of purchase of bonds by the banks the price of bonds falls, so the yield on bonds rises. The volume of bonds in the hands of rentiers will have increased by the beginning of the subsequent period and they will receive a greater amount of interest income. The decline in production in the consumer goods industry causes a cut in imports of raw materials and fuel. The demand for dollars to pay for these imports therefore falls, the foreign exchange market reaches a state of excess supply and therefore foreign currency (dollars) becomes cheaper in terms of domestic currency (i.e. the exchange rate of pounds vis-a-vis dollars r becomes less). Such an increase in the strength of the pound makes it cheaper for rentiers to purchase foreign bonds and their volume of holdings of foreign bonds will increase. As a result at the beginning of the subsequent period they will gain a higher amount of interest income from abroad than in the current period. On the other hand the fall in r will mean that imported raw materials and fuel become cheaper, stimulating a fall in the prices of both consumer and capital goods, but the rate of fall of these prices will be less than the fall in r. These export prices (i.e. the price of consumer goods and capital goods expressed in dollars) pjr, p2/r will therefore rise, causing a slump in exports. The fall in exports Eu E2 will force even more of a reduction in production in both of these industries, and the situation regarding workers and entrepreneurs will deteriorate even further. As a result of having to pay interest on the increased amount of the bonds they have issued, in the subsequent period the enterprise investment sector will be all the more starved of funds, necessitating a cutback in the volume of investment 72. It goes without saying that a reduction in I2 provokes a reduction in production in both industries. Thus any policy which combines reduction of taxes with cuts in government expenditure is blatantly deflationary. In the process of such a policy it is the rentiers (capitalists) who benefit and the entrepreneurs and workers who lose out. Similarly the reverse process (i.e. what happens where a policy of increasing government expenditure is implemented, backed up by a catalyst of an increase in taxation) will generate inflation; entrepreneurs and workers will benefit while rentiers
The multiplier effect of balanced budgeting
249
will lose out. Thus the problem of whether to increase or decrease taxation is essentially a question of whether to come down on the side of entrepreneurs and workers and give the rentiers the cold shoulder, or whether to sacrifice the entrepreneurs and workers in the interest of the rentiers. Marx believed that the interests of capitalists (entrepreneurs and rentiers) and workers were in conflict with each other, while Keynes looked at the economy from the standpoint of a conflict of interest between industry (workers and entrepreneurs) and rentiers.12 In this sense the model in this book is very Keynesian. In the real world, of course, it is frequently the case that entrepreneur and rentier are one and the same person, but functionally entrepreneurs tend to have common interests with workers, rather than conflicting with them, and it is the rentiers with whom they are at odds. Keynes made clear the conflict of interest between the entrepreneur function and the rentier function of so-called capitalists, and as far as policy was concerned he took the side of industry; he gave his attention to the means which should be used to promote industry. For that reason a 'policy combining tax cuts and cuts in government expenditure' which has been carried out by the Thatcher government, a policy to the benefit of rentiers, is an exceedingly anti-Keynesian policy. The multiplier effect on GNP It was Haavelmo who first carried out research into the effects of the above kind of balanced budget policy on GNP.13 Haavelmo's model was a very specific one in that it assumed, amongst others (i) that the propensity to consume of workers and of entrepreneurs was equal (cw = ce), (ii) that the rate of indirect tax on both industries was 0 {tx = t2 = 0), and (iii) that the rate of income tax on wages was equal to the rate paid by entrepreneurs (tw = te). The conclusion obtained by Haavelmo that the value of the GNP multiplier of a balanced budget = 1 is not a general one. In the case of our numerical example it is only 0.30. Such a value is very small by comparison with the simple government expenditure multiplier on capital goods, 1.45, but instead of this in the latter case the government suffers a deficit of 83.7% of the value of its expenditure. The difference between 1.45 and 0.30 is the amount of GNP which had to be abandoned by the government to wipe out its deficit, and must be regarded as the cost of dealing with its deficit. There is one thing which must finally be added. The previously 12
13
Keynes, J. M., The General Theory of Employment, Interest and Money, London: Macmillan, 1936. See particularly Chapter 24 'Concluding notes on the social philosophy towards which the General Theory might lead'. Haavelmo, T., 'Multiplier effects of a balanced budget', Econometrica, October 1945.
250
Fiscal policy
mentioned effect on employment of a balanced budget with £1 billion increase in taxes (i.e. a net increase in employment to the value of £246.8 million) will, assuming annual wages per worker of £4000, work out at a little over 60000 individuals. This cannot be regarded as a striking result, and even allowing for a parallel increase in the income tax paid by employers the increase in employment would be restricted to barley 90000. If, however, there were to be a balanced budget of an increase in the government's expenditure on personnel with £1 billion increase in the income tax on wages, the multiplier effect on employment would be as large as 278000 individuals, though its effect on GNP would be very small (only £7.1 million). If, in addition, there were to be a simultaneous increase in fiscal expenditure of the magnitude which would create a net government deficit of £1 billion, the total increase in employment would reach around 590000. This is a considerable figure. Even in Britain, where unemployment stands at around 3 million, if employment could be increased at a stroke by this amount, investment in private enterprises would increase accordingly, which would produce a further increase in employment. Keynesian policy is fundamentally this sort of pump-priming policy.
8
Monetary policy
1 Credit and business stimulation The role of banks Within the capitalist economy not only are new enterprises constantly appearing and becoming prosperous, but also enterprises which already exist are always making new plans and trying to improve their position. Investment is essential in order to implement such new plans. Part of these investment funds is furnished from the enterprise's own funds accumulated in the past, but most of it is raised in the form of credit advanced by banks. As mentioned earlier, banks have various kinds of deposits - in our model they have time deposits and current (or sight) deposits. When time deposits are used there is no question of the money being withdrawn within the prescribed period; even with current deposits there is no question of the total on deposit being withdrawn on demand, despite such deposits being used when it is not known when the money is to be withdrawn. Since whatever proportion it is (for example 90%) remains in the form of deposits, the bank for its part is able to set aside a certain amount of cash from current deposits (i.e. 10%) in preparation for deposit withdrawals by depositors. Not only can the banks feel easy about lending to entrepreneurs the total amount of time deposits and 90% of current deposits, but the banks themselves can use the interest on the money they loan to pay the interest to their time depositors and make a profit from these activities. When lending money bankers can investigate closely the investment plans of each enterprise and select those which they are willing to lend to; hence it would probably be no exaggeration to say that it is the banks themselves which determine the direction in which the nation's economy is to develop. In an economy without penetrating bankers, capital is squandered on inefficient entrepreneurs and unpromising projects, so that there is no efficient utilization of the nation's funds; hence there is no hope of progress and development over the long term. 251
252
Monetary policy
The ups and downs of the capitalist economy with its free enterprise system is largely dependent upon whether or not the country possesses to the full a flourishing entrepreneurial spirit. However, even if a country finds itself presented with numerous new plans, if the content of those plans is no more than mediocre there can be no expectation of outstanding success. Scrutiny of the quality of innovations is extremely important, both at bank level and at national level. This is not likely to be something which can easily be entrusted to the judgment of bankers. In Japan it is frequently the Ministry of International Trade and Industry (MITI) which takes the lead in bringing together the Economic Planning Agency and other involved agencies from each government ministry, as well as banks, economic organizations and involved enterprises, to look into and discuss which industries are to be developed as export industries, and what new industries are to be initiated. Such government-directed conferences of representatives of all sectors of the economy seem in the eyes of foreigners to be just like the board meetings of 'Japan Inc.', and are subject to severe criticism as joint conspiracies to give full play to 'Japan's egoism' on the international stage. However, if these meetings are viewed, in the way in which they should in essence be seen, as planning conferences to secure the most efficient use of capital, such conferences are necessary to some degree or other in any country.1 The creation of credit What is also important is not just the ability of banks to advance money currently deposited to enterprises, but their ability to lend sums in excess of these deposits, i.e. the creation of credit. This means that the money currently on deposit is enlarged to a sum several times the original amount, through the magic medium of the creation of credit; this then circulates within the economy. The maximum degree to which the sum can be multiplied - the value of the multiplier of the creation of credit is determined in the following manner. Let us suppose that bank A receives new current deposits to the tune of £10000. Bank A is unable to advance this entire sum. As was explained in Chapter 5, the city banks are under an obligation to retain a fixed proportion (a) of their current deposits in the form of cash either in their own bank vaults or as deposits with the central bank. As a result 1
However such discussions must take place under full public supervision. If the discussions are devoid of the consideration and self-control which will prevent any extreme deviation from the principle of the democratic distribution of capital and check increasing national egoism in the world's economy and society, such planning conferences will not only be subject to international censure; they will end up by being isolated within the country itself.
Credit and business stimulation
253
bank A leaves or x £10 000 (for example, £1000) for its cash reserves; it will then advance the remainder (1 - or) x £10000 (i.e. £9000) to an enterprise, for example, enterprise a. Bank A then pays the sum of £9000 into a's account. This means that A has supplied the sum of £9000 in the form of current deposits (hence it is a part of Mj).2 Part of this £9000 is soon converted into cash, while the remainder is paid out to workers and another enterprise b in the form of cheques as wages or as payment for the purchase of raw materials. These cheques are paid into the various accounts of the recipient b, producing an increase of a total of e x £9000 in the deposits of bank B. e here represents that proportion of the £9000 which is paid out as cheques; this is equal to the e, the proportion of M1 held as deposits, explained on page 167. If e = 0.7 the increase in deposits will be £6300 - that is to say, e (1 — a) x £10000. Bank B will retain part of this (i.e. 10%) for its cash reserves while advancing the remaining £5670 to enterprise c. The current deposits of enterprise c, to which the advance has been made, will increase by that amount, and this increase too will soon be paid out in the form of cheques or cash for wages or the purchase of raw materials. The increase in enterprise c's current deposits soon disappears in this way, and assuming that 70% of the payments have been made by cheque then the increase produced in the current deposits of enterprise d will be only 70% of £5670, that is £3969 (= e\\ - or)2 x £10000). This continues in the same way, making total current deposits £10000[l + e(l - o) + e2(l - (j) 2 + e3(l - o) 3 + =
]
x £10 000 = £27 027 1 — e + ea
and total loans £10000 [(1 -o) + e(l - a)2+ e2(l - a)3 + x 1~a x £10 000 = £24 324
]
1 — S + EC?
Total cash reserves of £10 000 [cr+ e(l - cr)cr+ s2(l - o x £10 000 = £2703 1 — e + ea is laid with banks A, B, C, . . . 2
Note that Ml, as has already been stated, is the volume of cash and currency on deposit held by the private sector excluding the banks.
254
Monetary policy
Thus if maximum advances are made the initial deposit of £10000 will ultimately create deposits of £27027 (hence deposit currency). This rate of increase, that is to say = 2.7027 1 - e + sois termed the multiplier of the creation of credit. Part of the current deposits created in this way (i.e. 10%) is retained by the banks in the form of cash reserves or deposited with the central bank, while the remainder is loaned to enterprises. Of the sum advanced to enterprises (£24324), 70% is deposited in current accounts with banks. (The remaining 30% circulates outside the banks in the form of cash.) Thus the amount of deposits in bank accounts is an increment of deposits derived from the initial deposit of £10 000. Needless to say, if we add the original deposit to the derived deposits we get total deposits of £27027. That is to say3 0.7 x £24 324 + £10 000 = £27 027 or e
1
~ ° " x £10 000 + £10 000 =
1 — e + ecr
x £10 000 1 — e + ea
Proof that the credit creation multiplier = money multiplier
The value of the multiplier of credit creation is exactly the same as the value of the money multiplier which we looked at earlier (see page 167). This can be proved as follows. When private individuals and enterprises receive an amount of cash 8M which is created through the dispersion of the same amount of high powered money, they will deposit part of this money, which is e8M, in bank accounts. The remainder (1 - e)8M
(1)
circulates in the non-bank sector. The first current deposits created in this manner are inflated through the credit creation multiplier into total deposits of l
-—e8M (2) 1 — e + ea This means there is this amount of increase in the amount of deposits payable on demand. The city bank, however, has to possess cash reserves of s8M 1 — e + ea 3
See Additional Note k.
(3)
Credit and business stimulation
255
of which cash of the amount e8M is in the bank's possession already at the time when the first account is opened so that the remainder must be provided as reserves within the bank during the process of creating credit. Therefore the amount of cash taken in by the banks from the non-bank sector is
[—z [1 - e + ea
il e8M
(4)
J
On the other hand, as we have already seen, deposit currency increases. Hence the net increase in Mx (the total of cash and deposit currency held by the non-bank sector) is equal to (1) + (2) - (4), that is 8MX = (1 - e)8M + \
[l-e+ecr 8M
[l-e+ea
1— £+ thus becomes clear that 8M
1 — e + ea
that is to say that the supply of a single unit of high powered money produces an increase in Mx of the same size as the credit creation multiplier. Hence the 8MX/8M obtained using the credit creation multiplier theory is exactly equal to the result in Chapter 5 which was obtained by the use of comparative statics method, i.e. the result established by supply and demand for high powered money ((14") in Chapter 5), what we earlier called the money multiplier. Multiplier process of credit creation Let us now explain in a little more detail, on the basis of our Table of Economic Linkages, the manner in which currency circulates between each sector and constituent of the economy during the process of credit creation. According to our model each enterprise is able to obtain capital from banks through the sale of bonds to banks. Now supposing for some reason the money on current deposit with the city banks was to increase by £10000 then, assuming there were no other changes in the situation, Lb in the bank column of the Table would merely decrease by £10000. (Note that Lb is cash held by city banks plus deposits to the central bank minus money on current deposit with the city banks.) Since the banks must possess cash reserves when they wish to lend out these funds on current deposit this generates a demand for cash on the part of
256
Monetary policy
the banks, and taking into account this attendant demand the net decrease in Lb will be £9000. (We assume the case of reserve rate a being 0.1) Therefore on the basis of the city banks' budget equation (the sum total of the elements of column 9 in the Table of Economic Linkages = 0) this will leave the city banks with a surplus of only £9000 with which to purchase fresh bonds. Hence, pb8Bb will increase by £9000. This money will be used for purchasing the bonds issued by enterprises. (Note that pbSBl has a minus value; so its absolute value -pb8Bl must increase by £9000.) Therefore, according to the budget equation of the investment sector (the sum total of the elements of column 6 = 0), enterprises are able to increase by £9000 the sum they invest, p2h. That is to say, enterprises are enabled to implement their investment plans as a result of being supplied with credit by the banks, and the increase in 72, according to the principle of effective demand, produces an increase in the volume of production of consumer goods Xx as well as an increase in the volume of production of capital goods X2. As a result the income of both workers and entrepreneurs (W and all) increases, and according to their budget equations (the sum total of the elements of column 3 = 0 and similarly for column 4) their money balances Lw and Le will also increase. Not only this, but profits (1 - a)TI will increase as well; this in turn will provide the enterprise with surplus funds, and its financial balance V increase. Part (say, 70%) of the increase in these balances (Lw, Le, V) assumes the form of an increase in current deposits, so the amount on current deposit with the city banks further increases, with Lb again decreasing to the same extent. In this way the banks come to hold surplus funds with which to carry out a second round of credit creation. The sum used by the banks to purchase bonds pb8Bb again increases and the capital released by the banks circulates between each sector and constituent of the economy by the same route as described above, producing increases in the money balances Lw, Le, V. Part of these increases (i.e. their 70%) takes the form of current deposits, so the amount on current deposit with the city banks increases, paving the way for a third round of credit creation. This credit creation can be repeated in the same way ad infinitum, but this kind of repercussion effect becomes smaller and smaller, until it eventually becomes infinitely small. The credit creation multiplier effect described above represents the total effect produced during this process of convergence. Assuming that e = 0.7, o-=0.1, an increase in the amount on current deposit with city banks of £10000 ultimately creates current deposits of £27027 and investment of £24324, and produces cash reserves of £2703 within the banks.
Deflation and inflation
257
2 Deflation and inflation
Imbalance of investment and the supply of capital This kind of amount of investment, however, is reached when funds are advanced to the limit of the banks' credit creation, and may not be an amount which tallies exactly with the level of investment desired by entrepreneurs. When there is little desire to invest and entrepreneurs wish to receive only small extensions of credit - or alternatively when, despite a wish on the part of entrepreneurs to receive large advances of credit, their investment plans are no more than indifferent and are for the most part judged by bankers to be unworthy of any extension of credit - the amount of credit created is likely to remain insignificant.4 In this sort of situation the bond market will be in a state of under-supply. Since there is little desire to invest there will be few investment plans, hence also few sales of bonds aimed at securing funds for investment. On the other side, banks will be left with capital which they would be quite happy to advance, consequently the banks will consider it desirable to purchase a greater number of bonds at current prices. In this sort of situation should each bank be actively looking for bonds to purchase the price of bonds will shoot up, resulting in their producing a lower yield ib. Should the yield on bonds become poor the rate of interest on time deposits id inevitably also falls. Needless to say if a bank pays a high interest rate while securing only a low yield on the bonds it has purchased it can hardly be a going concern. Furthermore, as it is assumed in our model that the interest rate on time deposits is equal to the official rate, a fall in the former implies a fall in the latter. In fact, under the conditions which we are now considering, where the banks do 4
The nature of the investment plans selected by the banks is highly significant, since it decides the direction in which the economy will subsequently develop. Supposing now that plans to invest in railways were proposed, but that plans for a massive expansion in the stagecoach system were also proposed. If bankers were to opt for the latter then the economy would be destined for collapse. It will doubtless be said that there is no question of such a silly case arising, but it is a well-known fact that when Japan had to choose between standard gauge and narrow gauge for her railways Japan's bankers and railway authorities opted for narrow gauge, a decision which led to long-lasting regrets. Such a problem as this, however, cannot be dealt with on the basis of a two-sector model such as ours. In order to be able to analyse properly Schumpeter's problem of innovation we must use a model which includes a plural capital goods sector. This must, moreover, be a model which enables us to analyse the obsolescence of various capital goods sectors as innovation occurs, and the appearance within the model of new capital goods sectors, and the process of movement of investment from the old sectors to the new sectors. As Schumpeter said, entrepreneurs and bankers are the helmsmen of a capitalist society. See Schumpeter, J. A., The Theory of Economic Expansion, Cambridge, Massachusetts, Harvard University Press, 1951, pp. 95-127.
258
Monetary policy
not make advances up to their maximum credit creation limit, the cash reserves of the city banks will be abundant, the loans they themselves receive from the central bank small, hence the official interest rate will also be low. Now if the rate of interest on time deposits should fall the volume of deposits is likely to decrease as well. The volume of increase in the time deposits of rentiers 8Qr shown in column 10 of the Table of Economic Linkages is small or negative, and the amount of time deposits entrusted to the city banks 8Qb is also minimal. Hence following a fall in the interest rate the banks' capital surplus is reduced, bringing about some improvement in the situation where capital has been in excess supply. Moreover, both the fall in interest rates and the increase in price of bonds serve to stimulate the desire on the part of entrepreneurs to invest. Businesses which have hitherto not been making a profit due to high interest rates and cheap bond prices will now begin to pay and so their volume of investment p2l2 is certain to increase. Thus only with the appearance of low interest rates can the situation be averted whereby the city banks are left with a large amount of surplus capital. When, by contrast, there is an overwhelming desire to invest, it is likely that sufficient capital cannot be provided even if the supply of capital is the maximum available through credit creation. In this case, either some investment plans are thrown out without having capital provided for them, or else the price of bonds falls as a result of the excessive demand for capital (i.e. the excess supply of bonds). In this latter case the yield on bonds ib will be improved. The banks then raise their rate of interest on time deposits to increase their supply of capital which they obtain via such deposits, and borrow from the central bank to supplement their depleted cash reserves. Such a fall in the price of bonds and a rise in the interest rate will serve to check investment. In this way the demand for and supply of capital are regulated, and any extreme shortages of capital are removed. All-round rise (all-round fall) in prices5
The mark up rate m of each industry increases (decreases) in accordance with the rise (fall) in the return on bonds and in the interest rate. Therefore the prices of products/?!, p2 are likely to rise (fall). This kind of inflation (or deflation) will disturb the capital market which has only just settled down. When inflation occurs there is an increase in the amount of funds needed for investment, capital again becomes tight, and both interest rates and the return on bonds rise. As a result m will 5
See Additional Note 1.
Deflation and inflation
259
rise further, producing the vicious circle of one price rise after another. (Deflation produces quite the opposite phenomenon.) Moreover, in the case of inflation, should there be any hint of a shortage of the labour needed to make the most of these vast amounts of investment then a struggle for labour will result and the wage rate w rise. This will stimulate all the more the rise in the prices of products. Also, in order to maintain their real wage rate in the face of ongoing price rises workers tend to demand increases in their money wage rate, regardless of the level of unemployment at the time. This, of course, gives a further spur to the increase in prices. This means that when the supply of capital is insufficient to meet the demands for investment a rise in interest rates results, and this rise brings about a rise in prices and wages whether or not unemployment exists. As we see above, where the requirements for investment exceed the limit set by credit creation, inflation will be produced, whereas if the desire to invest is the smaller deflation will result. In between a situation exists where neither inflation nor deflation occurs (and where prices are therefore stable), and the rate of credit creation which corresponds with this kind of situation can be called the natural rate of credit creation. Investment over and above that in line with the natural rate is an inflationary factor. Below we assume that the natural rate is set at the fullest possible extent of credit creation, whilst it is sometimes regarded as being considerably beneath this limit. Schumpeter, for example, seems to have regarded the extension of all credit as fundamentally inflationary.6 Stagflation
This kind of process, however, will not continue indefinitely. When the interest rate becomes very high no investment will make a profit, therefore the volume of investment I2 will become very small. As a result the demand for credit shrinks sharply. The pressure on credit which hitherto existed disappears at a stroke, producing just the reverse, an excess supply of credit. This will in turn lead to a fall in the yield on bonds, and at the same time a fall in the interest rate on time deposits (hence, a fall in the official rate). The economy thus begins to move in the opposite direction, and before long embarks on a process of deflation unless policy measures are taken to keep the economy on the right lines. However, in a period during which monetarists have control over economic policy things are unlikely to proceed along these lines. Since 6
Schumpeter, ibid, pp. 109-10.
260
Monetary policy
monetarists believe that it is the increase in the money supply which gives rise to inflation they try to rationalize the volume of money by restricting investment to a low level and accommodating the supply of credit to investment. The existing excess supply of credit is removed in this way, and this was the case in Britain under the Thatcher government 1980-81, when high interest rates continued to prevail in the financial market. When high interest rates prevail investment I2 is held at a low level. Moreover, since monetarist governments curb government demand Gu G2, each industry suffers from a lack of effective demand, giving rise to reduction or stagnation in the volume of production Xu X2. The volume of employment does not grow and unemployment tends to increase. The more puritan the government is in its monetarism the more serious the degree of business stagnation. However, because on the other hand interest rates are kept at a high level the mark-up rate in each industry will be high, and as a result the amount of wages wa3i will be a correspondingly smaller part of the price of consumer goods px. If we regard the labour input coefficient a31 as fixed then this means a reduction in the real wage rate w/px. Trade unions will demand an increase in the wage rate w to restore w/px to its value when interest rates were at a more normal level. However, the effect of such wage rises will partly be cancelled out by the price rise implemented to compensate for the increased wage costs, so that workers demand a further increase in wages giving rise to a vicious circle of price and wage increases. This kind of inflation combined with depression is known as stagflation. In an economy where a significant role is played by industrial sectors which determine their prices on the basis of the full-cost principle, i.e. in a modern manufacturing industrial society, stagflation is a phenomenon which can occur very easily. Moreover, stagflation can become even worse. With interest rates held at a high level and domestic bonds yielding a good return, those rentiers who formerly purchased foreign bonds are now likely to try to purchase domestic bonds instead. The demand for foreign currency (dollars) will therefore decrease. Furthermore demand from abroad for that country's bonds (pb8&) will also increase, serving to increase the supply of dollars. This leads to an excess supply of dollars on the foreign exchange markets, so the price of that country's currency (pounds in the case of Britain) rises against the dollar. This means that the price of dollars expressed in pounds r will become lower. If r falls then the price of imported raw materials and fuelstuffs in terms of pounds is reduced, and, because of the full-cost principle,
Deflation and inflation
261
the rise in the prices of products is restrained. In this sense real wages improve, and demands for wage rises from workers are likely to subside. However, a strong pound makes British goods comparatively expensive in foreign markets, causing Britain to lose her competitiveness in the international market. As a result British exports Eu E2 are likely to decrease, probably at a fairly sharp rate.7 A decrease in demand abroad makes the slump all the more serious; production Xu X2 decreases and unemployment grows. The decline in exports means a reduction in the supply of dollars on the foreign exchange market and the pound will fall in value against the dollar. A weak pound (a rise in r) is inflationary. Imported raw materials and fuelstuffs are expensive, causing a decrease in the proportion of the price of goods accounted for by expenditure on wages, and real wages will fall. Union demands for wage increases become more strident and inflation again advances. As a result British industry is unable to recover its competitiveness in overseas markets despite the pound being weak. Exports El9 E2 will remain low, and severe slump will continue in conjunction with inflation. The country will have been led into all the more chronic stagnation. The oil shock We have above described the stagflation which results from the monetarists' maintenance of high interest rates, but a similar state of affairs can also arise for quite different reasons. Let us now suppose that the oil-producing countries have made a sizeable increase in the price of their oil. For industrial countries which do not possess their own resources of oil this increase will mean an increase in the price of imported raw materials and fuel /?*. Given the full-cost principle this produces first of all a rise in the price of goods produced, and then a resultant fall in the real rate of wages. This gives rise to demands for a rise in wages to restore their real value, and wage increases produce further increases in prices. On the other hand a rise in p* will increase the amount of imports p*F, so demand for the dollar will increase, and its price expressed in terms of that country's own currency r will also rise. This kind of increase in r pushes up the prices of goods further, lowering real wages. This again results in demands for a restoration of real wage rates, and wage increases send up prices still further. Such price rises will bring 7
See above p. 109ff.We have in 'Britain' not only the effects of a strong pound but also inflation due to wage increases, so it is likely that the competitiveness of 'British' goods abroad will have been weakened all the more.
262
Monetary policy
about a reduction in exports El9 E2.H Depression will result, with a decrease in the volume of production XX,X2, but because the volume of production does not decline at the same rate as the fall in exports, imports (of raw materials and fuel) will decrease at a lower rate than exports. The balance of trade will go into the red and the price of dollars expressed in terms of that country's currency will again rise. Thus inflation will once again be stimulated, resulting in loss of exports and decline into even more serious slump. 3 The roles of the central bank and the exchange stabilization fund
The three monetary policies In the case we considered above no positive role was played by the central bank and the exchange stabilization fund. In such cases, as we have seen there is a tendency for the following sort of situations to arise: (1) Where investment is insufficient unemployment will arise, (2) When there is sufficient investment to realize full employment inflation will result, (3) Inflation can arise even where unemployment exists and the economy is in a state of depression. Stagflation is neither a strange phenomenon nor an exceptional one, (4) Where a slump results from slack in exports the balance of trade worsens, resulting in a fall in the value of a country's currency vis-a-vis other currencies (i.e. a rise in the exchange rate r), producing inflation at home. All of these are undesirable. Therefore in order to (a) bring about full employment, (b) stabilize prices and (c) preserve the balance in international payments, the central bank and exchange stabilization fund must not stand idly by, but must take positive action to guide the economy in a relatively desirable direction. Discussions on monetary policy will provide guidelines for activity for the central bank and exchange stabilization fund. There are traditionally three kinds of monetary policy: the open market operation for public bonds, lending policy and reserves requirements policy.9 The open market operation for public bonds is a policy by means of which the central bank engages in the sale and purchase of 8
9
A rise in r reduces the price of products in terms of dollars p/r and increases exports, but in the case we are currently analysing, where a rise in r stimulates an increase in the wage rate w, p/r may increase despite the rise in r, bringing about a reduction in exports Eu E2. There are apart from this such things as the central bank's regulation of its loans to the city banks, and the use of policy measures such as the regulation of consumer credit through exercise of the laws of hire purchase, but these are no more than secondary measures aimed at backing up these three basic financial policies, and are not of any particular theoretical interest.
Roles of the central bank and exchange stabilization fund
263
bonds on the open market, thereby supplementing any shortage of capital and takes up any surplus capital. If the central bank purchases company or government bonds, then the appropriate amount of currency is paid to the enterprise of the government, and the country's volume of currency increases by the same amount. Conversely if the central bank sells the company or government bonds in its possession the volume of currency will be reduced. By contrast, in the case of lending policy the central bank fixes the official interest rate, and whatever amounts the city banks wish to borrow are lent to them by the central bank at this official rate of interest. Since our model supposes the official rate to be equal to the interest rate on time deposits, the central bank's manipulation of the official rate means that it controls the interest rate as a variable of monetary policy, and that it can accordingly regulate the amounts which it lends to the city banks. Moreover, under the reserve requirement system, which places upon the city banks the obligation of depositing with the central bank cash in a certain fixed proportion to the amount of deposits they themselves have received, any change in the rate of reserve requirement (the reserve ratio) can regulate the limits on credit creation, and hence the supply of capital. Such manipulation is known as reserve requirement policy. However, we will attempt below to provide an explanation of monetary policy limiting our discussion to the open market operation and lending policy.10 Open market operation There are two means of manipulating the bond market. The first is the positive method of trying to regulate the economy through the price of bonds by fixing the amount of bonds 8BC which the central bank will try to buy (or sell), and getting the market to establish a price for bonds in accordance with the purchase (sales) of bonds laid down by policy. The second is the passive method whereby the central bank purchases (sells) bonds only to the level of the excess amount of supply (demand) which exists in the market in an attempt to keep the price of bonds unchanged. The former is known as the flexprice method, the latter the fixprice method. Our explanation below will concern only the fixprice method. Demand and supply on the bond market is represented in row 9 of Table 6. If, for the purposes of simplification, we ignore the demand for bonds on the part of domestic and foreign rentiers (i.e. providing 8Br=8Bf=0), then we have simultaneously on the market both the 10
For the sake of simplicity the explanation below will assume that the central bank hitherto engaged in no lending and no dealings in bonds, and that therefore its income from interest is 0, i.e. A = 0.
264
Monetary policy
supply of government and company bonds of government and enterprise (8B8 and 8Bl have a minus value) and the demand for government and company bonds from the city banks (8Bb > 0); let us then assume that there is an overall state of oversupply, i.e. 8Bl + 8B^ + 8Bb < 0. Supposing that the central bank takes up the total amount of excess supply of bonds and lets 8Bl + 8Bs + 8Bb + 8Bc = 0 this means that the central bank will have overpurchased bonds. The reason for this is that in order to purchase 8BC the central bank has to create currency by the amount pb8Bc = 8MC;11 given this, part (e8Mc) of the payment accruing to enterprises and government in return for the sale of bonds to the central bank (8MC) will be deposited with the city banks in the form of time deposits. Then according to the theory of the credit creation multiplier12 deposits t a the extent of 1
s8Mc
(5)
1 — 8 + EOT
and cash reserves to the extent of a s8Mc 1 — e + ecr
(6)
will be produced at the city banks, and the amount which can be advanced will be 1
~°"
s8Mc
(7)
1 — 8 + 8(7
That is to say a demand by the central bank for bonds whose money value is pb8Bc = 8MC creates a demand for bonds from the city banks to the value of (7), producing an overall demand for bonds in the banking sector as a whole to the value of 8MC +
^— 1 — 8+ 8d
c
e8M
=
8MC
(8)
1 — 8+8(7
What is needed is for this total demand to equal exactly the excess supply on the market (and if the central bank should demand as many bonds as will at a stroke wipe out the excess supply with which it is faced then the market will end up by having an excess of demand). 11 12
See column 9 of Table 6. This assumes, however, 8QC = 0. As has already been mentioned in Additional Note k, the multiplier effect below comes into being immediately, without any time lag.
Roles of the central bank and exchange stabilization fund
265
This kind of creation of money is likely to exert an influence on the money market (row 12 of Table 6). If we now suppose that even in the absence of any creation of money the foreign exchange markets are in balance without the intervention of the exchange stabilization fund, then the sum of the elements of row 11 will be 0.13 Therefore, according to Walras' Law there will exist in the money market an excess demand to the same value as that of the excess supply of bonds. 14 That is to say (Lw - Mw) + (Le - Me) + (Z/ - Mr) + (V - My) > 0 (However for the sake of simplicity let us suppose that prior to money creation Lb = Mb.) Now if the demand created for bonds on the part of the city banks is of the amount of (7), pb8Bb must be increased by that amount, hence Lb - Mb must be reduced by the same amount. (We assume that 8Qb and wNb remain unchanged.) Since it is assumed that prior to any creation of money Lb — Mb = 0, this means that after the currency has been created Lb-Mb=
1
~ ° " e8Mc 1 — e + sa
(9)
Since the excess supply of bonds on the market has been wiped out by the central bank's manipulative purchase of bonds, enterprises and government are now able to sell the amount of bonds they wished to do, and therefore carry out the investment and government expenditure regarded as desirable. As a result production will take place exactly according to plan, while the supply and demand of both workers and entrepreneurs will be realized without a hitch. Their cash balance plans will be exactly as they were before the currency was created. As the bond market is in equilibrium as a result of the central bank's manipulative purchasing, and all other markets are also in a state of equilibrium, so the money market too must balance. For this reason (Lw - Mw) + (Z/ - Me) + (Z/ - Mr) + {V - M') =
8MC
(10)
1 — s + ecr
That is to say if the central bank manipulates the purchase of bonds and is able to create money, this will produce in the private-non-bank sector an increase in cash balances which is equal to the amount of money created, 8MC, expanded by the credit creation multiplier. 13 14
Because of this Ds% = 0, therefore 8MS = 0. Assuming that the supply and demand for consumer goods, capital goods and time deposits are all in balance.
266
Monetary policy
This fact may also be explained as follows. Of the amount 8MC dispersed to the private sector, e8Mc is deposited with the banking sector, while the remainder (1 — e)8Mc stays in the various non-bank sectors in the form of cash. The city banks try to maintain cash reserves at the level of (6); since the city banks already received e8Mc of cash when the current account was first opened, the amount of net cash reserves which the banks have to create afresh is -e8Mc-e8Mc 1 - s + ea This amount of cash must be absorbed by the banking sector from the non-banking sector, hence the net increase in cash held in the nonbanking sector will be c
(1 - e)8Mlc -\ - I
\\-e+ea
c
c
e8M e8Mc --e8M e8Mc] I-•= — - — — 8MC
J
Since current account deposits increase according to (5) the increase in currency Mu including not just cash but also current deposits, must be 1 8MC (11) 1 — e + ea That is to say the cash balances held by the private-non-banking sectors increase according to (10). This kind of accumulation of cash balances on the part of the private sector will later on stimulate inflation, as will be explained below. For the time being, however, the inflationary pressure resulting from over-investment has been successfully removed by the central bank's manipulative purchase of the surplus government and enterprise bonds on the open market. The existence of excess supply on the bond market (i.e. of excessive demand for capital) would cause the yield on bonds to rise and stimulate price inflation by raising the mark-up rate of each enterprise, but the provision of this capital by the central bank avoids inflation through this sort of channel. However, this kind of policy does in the long term have a considerable influence, as will be analysed in section 4.
Lending policy There are two kinds of lending policy. In the first of these, the central bank alters the official interest rate and then lends to city banks the sums they require in accordance with this new rate. In the second, the official rate is left as it is, and the central bank then lends the city banks the
Roles of the central bank and exchange stabilization fund
267
amount of high-powered money which it could not obtain from the non-banking sector. Under the former it is the change in the official interest rate which is important; under the second the fact that the city banks are supplied with the amount of money which they require, with no change in the official rate. The latter case may be further subdivided into (1) where the whole of this required sum is lent out, and (2) where a limit is set to the amount to be advanced and where advances are made only up to, and not beyond, this limit. We shall below analyse the case of (1), i.e. where the central bank maintains the official interest rate at the current level, and sets no limit on advances to the city banks. Suppose the time deposits of rentiers are 8Qr(> 0) and the amount of time deposits desired by the city banks is 8Qb(<0); also suppose the latter will have a higher absolute value than the former, and deposits will therefore be insufficient. As long as the central bank does not exercise a lending policy (8QC = 8MC = 0) and there is no intervention in the market from the exchange stabilization fund (rZ)| = 8MS = 0), then supposing supply and demand in both the bond market and the foreign exchange market balance each other, then by virtue of Walras' Law there will exist an excess demand for money only of the same amount as the shortage of time deposits. This means (Lw - Mw) + {V - Me) + (Z/ - Mr) + (V - M) + (Lb - Mb) = -(8Qr+8Qb)>0
(12)
In this sort of situation the city banks are, as a result of the shortage of deposits, unable to make at least some part of their projected loans pb8Bb. Thus if a part of 8Bb no longer goes toward the purchase of bonds then the bond market will be oversupplied with bonds. To balance the bond market it is necessary to remove the deficiency in deposits at the city banks, and for this to happen the central bank must make to the city banks advances (8QC) of the amount equal to the shortfall in bank deposits, - (8Qr + 8Qb).15 Where such advances are made the high-powered money created amounts to 8QC = 8MC. Since - (8Qr + 8Qb) = 8MC, then (12) can be written as L = M + 8MC 15
In the actual economy the interest rate on time deposits is not equal to the official rate of interest, and the period of the loans made by the central bank to the city banks (usually three months, but sometimes repayable after a number of days) is shorter than that on time deposits. Our model, however, assumes that the interest rate on time deposits is equal to the official rate of interest, and also regards the money advanced by the central bank as a perfect substitute for time deposits.
268
Monetary policy
where L is the sum of L w , Z/, Z/, V, Lb, while M is the sum of AT, M e , M r , Af, M 6 . Thus the money market is in equilibrium with a higher level of high-powered money. Overloan Having received advances of 8MC from the central bank, what are the sizes of the advances made by the city banks to make good the demand for bonds, and how does money flow as a result between the privatenon-bank sector and the banking sector? The city banks first buy bonds with 8MC. One part eSMc of 8MC which initially flowed into the non-bank sector will find its way back to the city banks in the form of current deposits. This initial increase in current deposits will, according to the credit creation multiplier, be expanded into current deposits to the value of -8MC
(13)
1 — £ + ECT
The city banks will keep a part of this amounting to £0"
(14)
1 — e + Ed
as cash reserves, while the rest goes to finance the non-bank sector. Since this financing is carried out through the medium of purchasing bonds, then combined with the initial purchase of bonds to the value of 8MC, the total value of bonds purchased will be 1+ e ^ ~ ^ \8MC = 8MC |_ 1-e+eo-J 1-e+f-o-
(15) v
}
On the other hand, since of the total amount of money initially created 8MC, an amount to the value of (14) is kept in the form of cash reserves within the banks, the remaining 8MC
— 1 — £+
8MC = ECT
l E
~
8MC
1 — E+ ECT
is retained in the non-bank sector. The amount obtained by adding to this the amount of current deposits, i.e. 8MC 1 — E + Ecr
Roles of the central bank and exchange stabilization fund
269
is the amount of increase in the currency Mx in the non-bank sector. That is to say, if the central bank makes advances of 8MC to the city banks, then an amount equal to the original advances 8MC magnified as many times as the credit creation multiplier will be accumulated in the non-bank sector in the form of currency Mx (the so-called money multiplier). When the central bank makes flexible advances to the city banks in order to maintain the official rate of interest, rises (or cuts) in the official interest rate (and, hence, in the interest rate on time deposits) are avoided in spite of insufficient (or excess) deposits in the non-bank sector, and the mark-up rate remains unchanged. As a result there is nothing to disturb the price-cost equation of each industry, and for a time any change in price is averted. Such advances on the part of the central bank, however, do, on the other hand, cause money (Mx) to accumulate in the non-bank sector, so care must be exercised over any continuation of lending policy. A situation where advances from the central bank build up to large quantities is known as central bank overloan, and as will be explained later, such a situation will eventually inevitably get out of control and give rise to inflation. Balancing the foreign exchange market In the above explanation the foreign exchange market (Table 6, row 11) was regarded as being in a state of equilibrium. Below we shall try to look at the kind of manipulation exercised by the exchange stabilization fund in order to stabilize the exchange rate of a country's currency, where there is either an excess demand for, or excess supply of, dollars. Suppose first of all that there is an excess demand for dollars; in this case the sum of the elements of row 11 will have a plus value, so that by Walras' Law the money sector (row 12) will be in a state of surplus supply.16 Should no action be taken by the exchange stabilization fund and the whole thing entrusted to market forces, then quite clearly there would be an increase in the price of dollars expressed in terms of that country's currency (i.e. in the exchange rate). When the stabilization fund satisfies the excess demand for dollars by selling its own holdings of dollars in order to check this rise, then D$<0, and hence SMs<0. That is to say the foreign trade sector and rentiers who have purchased dollars from the stabilization fund pay the fund in pounds. These pounds find their way into the fund's vaults and the volume of currency circulating in the non-bank sector is reduced by that amount. 16
It is assumed here that the bond and time deposit sectors are in a state of equilibrium along with the consumer goods and capital goods markets.
270
Monetary policy
Then, because the volume of currency withdrawn in this manner is exactly equal to the amount by which currency is in excess supply on the money market, the money market will simultaneously return to equilibrium. Conversely, where there is an excess supply of dollars, the exchange fund must buy up dollars in order to keep r unchanged. That is to say Df>0, hence 8Ms>0. Currency is created and passes from the stabilization fund into the hands of those supplying the surplus dollars. This volume is exactly sufficient to mop up the excess demand for money which exists. The foreign exchange market is balanced by the exchange fund's support of the dollar, and at the same time the imbalance in the money market is also removed. In this way fluctuations in the exchange rate r through intervention on the part of the stabilization fund can be averted both when there is excess demand for foreign currency, and when it is in excess supply. Without such intervention the excess demand for dollars will raise the exchange rate r, bring about an increase in the cost of imported raw materials and energy, provoking cost-push price rises, but provided the stabilization fund releases dollars to meet this excess demand, a rise in r can be checked. At the same time, however, the volume of money held in the private-non-bank sector is reduced, so the likelihood of future deflation must be borne in mind. Conversely, where dollars are in excess supply there is a tendency towards cost pull deflation. If the stabilization fund intervenes to stabilize r, this kind of deflation can be avoided, but this is no more than a temporary avoidance policy, which will sow the seeds of inflation for the future. That this should be so is by no means surprising. All that has happened is that because we adopt inflationary (deflationary) policies to avert current deflation (inflation) the results of these policies will become apparent in the future. This contrasts with the other case which we have already seen; that is when the inflationary pressure exists in the bond market and time deposit sectors, then the manipulative purchase of bonds and lending by the central bank serves to keep this inflationary pressure in check for a while by postponing any actualization of inflation. By contrast, on the foreign exchange market, where the inflationary pressure takes the form of an excess supply of the country's own currency (i.e. an excess demand for dollars) the inflationary pressure is removed through the release of dollars and purchase of domestic currency - the policy of reducing circulation, in effect a deflationary policy - so while this does avoid inflation it brings about deflation in the future. It is worth looking at the differences in the long term effects of each of these policies.
The Wicksellian cumulative process
271
4 The Wicksellian cumulative process Discrepancy between the money interest rate and the natural interest rate
We will now consider a situation where not only each commodity market but each financial sector is in equilibrium without any positive action on the part of the central bank or the exchange stabilization fund. This means that for the stabilization fund we will have Z)| = 8MS = 0, and for the central bank 8BC = 8MC = 0, and hence according to column 10 of the Table of Economic Linkages Ac — 8QC = 0. The central bank will therefore advance to the city banks only the amount it receives as interest. Considering that under such conditions all financial sectors are in a state of equilibrium, then we will call that state a natural equilibrium, and the official interest rate (the interest rate on time deposits) in accordance with such a state the natural rate of interest; similarly, the yield on bonds which would prevail in the state of natural equilibrium is called the natural rate of yield. In such a state, furthermore, credit will be created at the above-mentioned 'natural rate of credit creation'. The actual official interest rate (the rate of interest on deposits) - and this will be referred to below as the actual rate of interest or the money rate of interest - is not necessarily the same as the natural interest rate. The natural interest rate is not a thing which can be publicly announced in some market or other,17 the only thing one can do is to do one's best to surmise whether the current money rate of interest is higher or lower than the natural rate, by looking at the direction in which market forces are operating. We will below first consider a case where the actual rate of interest is lower than the natural interest rate. Since in this situation the amount on time deposit from rentiers will be less than it would be in the 'natural' situation, the city banks will suffer from a shortage of time deposits. The city banks will therefore be unable to supply sufficient credit to enterprises and government (i.e. the city banks' demand for bonds 8Bb will be less than in the 'natural' situation), and an excess supply will result on the bond market. Such excess supply will be in part absorbed by the mechanism prevailing in the bond market (i.e. by the price of bonds pb falling below the natural price), but the city banks will also attempt to overcome the shortfall in time deposits by borrowing from the central bank an amount which exceeds the central bank's income from interest (i.e. by the 17
It must be said that this rate of interest is no more than an abstract concept for use in analysis of the economy, such as will enable econometricians to make calculations when an exact econometric model is provided.
272
Monetary policy
amount loaned by the central bank 8QC being at a level where 8QC > Ac). The result of this will be the creation of high powered money of the amount 8MC = 8QC — Ac. This volume of currency will be used by the city banks to back up further creation of credit,18 8Bb will be increased to supplement the hitherto inadequate demand for bonds, and the bond market will be turned into equilibrium. This means that the excess demand for credit is removed by credit created on the basis of advances from the central bank. This kind of process is inflationary. First of all, should part of the excess supply of bonds be absorbed by a fall in pb the rate of return on bonds will get higher, bringing about a rise in the mark-up rate m = m(ib, id) in each industry, producing in turn a rise in their prices pu p2. Secondary a rise in/?! causes the real wage rate w/px to fall, leading trade unions to demand an increase in money wages to compensate for this fall. Needless to say such wage increases stimulate price increases. Thirdly, the furnishing of credit bestows purchasing power on those who would not otherwise have it. The new enterprises and new projects from existing enterprises which are thus given a say in the market now appear in the producer goods market and compete in the scramble for producer goods. Where vast amounts of unused producer goods exist the competition is not that fierce, but even in this case some scramble is unavoidable. The setting up of new enterprises or the formation of new project teams employing only hitherto unemployed workers is likely to result in failure for such new enterprises or projects, so at least for their core members they are likely to try to obtain able workers from some other enterprise. Whatever the case, a scramble will, to a greater or lesser degree, be inevitable, and if we suppose, as Wicksell does, that there is from the first full employment,19 then the competition will be fierce indeed. The wage rate w rises, producing cost push price rises.20 Inflation from the creation of money The amount of money created anew, 8MC, will accumulate in the non-banking sector. In order to analyse the unadulterated effects of this accumulation of money on the volume of production Xu X2, on the lending by the central bank (and hence on the creation of money) and on the demand and supply of bonds (hence on the demand and supply of 18 19
20
For the process of credit creation see the earlier section on lending policy. Wicksell, K., Interest & Prices, 1936 (reprinted, 1962, by Augustus M. Kelley, New York), p. 132, p. 143 etc. For Wicksell's own explanation of the cumulative process see pp. 102-56 of the same book. Wicksell does not assume the full-cost principle as we do, so the mode of price fluctuations which follow from his model differs from the above.
The Wicksellian cumulative process
273
investment fund) during the next period (second period) of money creation, let us suppose that the price inflation described above did not occur.21 As we shall see later price and wage inflation create positive or negative forced saving. We will first of all analyse the simple case where forced saving is zero, and subsequently observe how this simple case is made more complex by the subsequent introduction of positive or negative forced saving. First of all, in view of workers', entrepreneurs', and rentiers' consumption functions we obtain from their budget equations:
(1 - te)all+ Me = ce{\ - te)an+pxye + Z/, (1 - tr)Ur + Mr = cr{\ - tr)Ur + Plyr + pb8Br+8Qr where Ur = Ar + rBr. At the beginning of this second period the amount of currency held by workers Mw increases, so the left-hand side of the above equation will increase, and as a result the right-hand side must increase as well. Thereupon each item on the right-hand side is likely to increase too. That is to say, it can be assumed that the propensity to consume cW9 the amount of consumption at zero income yw and also desired balance Lw will all increase in line with the increase in Mw. Similarly during the second period both Me and Mr increase above their level in this first period, so the entrepreneurs' and rentiers' propensity to consume and their amount of consumption at zero income ye, yr will also increase over their present level. Their demands for money balances Le, Lr and the amount of savings held by rentiers in other forms Pb^Br, 8Qr, rp%bBr^ are all likely to increase. This kind of increase in the propensity to consume and in the basic amount of consumption brings about an increase in the effective demand for consumer goods. (Note that prices are assumed to remain unchanged, so there is no forced saving effect.) On the other hand the official interest rate and the price of bonds (hence their rate of return as well) remain unchanged, so there is no reason to suppose that invest21
We now suppose that the excess supply of bonds in the current period is removed merely by the creation of money 8MC, and that no price mechanism works in the bond market. In such a case pb (and therefore the yield on bonds) is constant, so m is also constant. It is furthermore assumed that considerable unemployment exists among all kinds of labour, and that no scramble for labour has emerged despite the appearance of new enterprises and new projects. Given this w too will be constant, so price inflation of the kind explained above will not occur in the current period. Below, we shall in the first instance carry out analysis of this kind of simplified case.
274
Monetary policy
ment I2 alters.22 There is, therefore, no change in the effective demand for capital goods. The volume of production of consumer goods Xx increases, while the volume of production of capital goods X2 is held at the same level. As a result wages W, the depreciation reserve H, the amount of imported raw materials and fuel rp*F, profit 77 and government tax revenues T will all increase. This is the result of Xx and X2 adjusting to their respective effective demands, and this also means that excess demand for both consumer and capital goods is zero. Hence according to Walras' Law we get pb[8Br + 8Bl + 8Bf + 8B8 + 8Bb] + [8Qr + 8Qb + 8QC]
+ [(Lw - Mw) + (Z/ - Me) + (U - Mr) + (V - M) + (Lb - Mb)] = 0 (16) Here 8QC is the lending of the central bank in the second period which is made when no money has been created (i.e. 8QC = AC in the second period), and assuming that the amount of time deposits, -8Qb, which the city banks desire to raise is exactly equal to their time deposits received from the private sector 8Qr and their advances from the central bank 8QC, then 8Qr + 8Q0 = -8Qb, and for that reason the sum of the contents of the second square brackets in the above formula will equal zero. If we suppose an adjustment in the foreign exchange rate so as to remove the excess demand for foreign currency, the total for the third square brackets in the above formula will also be zero. For that reason if the sum of the terms in the first square brackets (excess demand for bonds) is negative the sum of those in the fourth square brackets (excess demand for currency) will be positive, and vice versa. Now the first two items within the fourth square brackets represent the savings in the form of money made by workers and entrepreneurs respectively. By virtue of the budget equations mentioned earlier these usually have a positive value except where the respective incomes are extremely small; likewise the third item (money savings of rentiers) is also positive. The fifth item too will be positive. If we rewrite the budget equation of the city banks (the sum of the elements of column 9 in Table 6 = 0), we get Ab - 8Qb = wNb + pb8Bb + (Lb - Mb) 22
Where prices fluctuate, investment (in housing and other construction, equipment and stocks) will depend upon the real interest rate (the difference between the money interest rate and the anticipated rate of inflation) and not the money interest rate. (Should the real rate of interest increase, then investment will decline.) In this particular case, however, the predicted rate of inflation is 0, and the money rate of interest is constant, so the real and money rates of interest will be equal and constant.
The Wicksellian cumulative process
275
where Ab must increase in the subsequent (second) period, since during the first period the central bank has implemented a positive lending policy, the city banks have accordingly created credit and advanced money to enterprises, and hence the income from interest accruing to the city banks during this subsequent period is going to increase. As has already been stated -8Qb= 8Qr + 8QC. The 8Qr of the subsequent period will be higher than that of the first period, and the same is likely to be true of 8QC (= Ac) where the positive advances made by the central bank will be reflected in Ac showing an increase over its value in the first period. With -8Qb therefore increasing, the left-hand side of the above formula will increase. This is likely to mean that each item on the right-hand side is positive, and that each increases to a greater or lesser degree. On the other hand the budget equation of the enterprise investment sector (the sum of the elements in column 6 of Table 6 = 0) can be written as [p2l2 -H-(l-a)II\-
& = -pb8Bl - (V - M<)
(17)
The second item on the left-hand side of this equation, -B\ will be positive, while the first item too is likely to have a positive value.23 Therefore the two items on the right-hand side of the equation, pb8B\ {V - M1) are both likely to take a negative value, but in view of the fact that V — Ml cannot amount to a negative sum of large absolute value were it to do so the enterprises' cash balances V would diminish, and they would encounter various difficulties in their transactions and in realizing their investment plans - even where (V - M) does have a minus value, it will not be one of any significant magnitude. We know from the above that all items within the fourth square brackets in formula (16), with the exception of V — M, will take a positive value, and even where V - M is minus its absolute value will be small. The contents of the fourth square brackets as a whole will therefore be plus, hence the contents of the first square brackets in the same formula will be minus. That is to say the bond market is in a state of excess supply. Should the central bank supply enough money at the prevailing official interest rate to meet the shortfall 8MC, with the aim of removing the excess demand for money, then the central bank's advances to the city banks will increase by the same amount; as a result credit will be created by the city banks on the basis of the loans made to them by the central bank, and the city banks' demand 8Bh for company and government bonds will increase. Following the creation of money 23
An increase in output Xl will produce an increase in depreciation reserves H and the amount of profits retained within the enterprise (1 - a)/7, but they are still unlikely to attain the amount of investment p2l2-
276
Monetary policy
both the money sector and the time deposits sector will be in equilibrium. Furthermore, in the foreign exchange sector the exchange stabilization fund makes adjustments so as to maintain the prevailing exchange rate, so equilibrium in all these sectors is maintained. This means that demand for and supply of bonds will also be balanced. In this kind of 'subsequent' (second) period the prevailing money interest rate will become lower than the natural rate of interest. The reason for this is that if the central bank only made advances equivalent to Ac (i.e. where 8Mc = 0), the time deposits and foreign exchange sectors would be in equilibrium, while the bond market would be in a state of excess supply and the money market in a state of excess demand, in just the same way as we have seen above for the first period. To amend this state of affairs there should be an increase both in the yield on bonds and in the official interest rate (the interest on time deposits).24 From the definition of the 'natural' rate it can be seen that following such a rise both the rate of yield and the rate of interest would be at their natural rates, so the actual rate of yield and the actual rate of interest (i.e. the money interest rate) will be lower than their natural counterparts. Hence all the phenomena described in the previous section - a rise in product prices, a rise in the yield on bonds and hence in the mark-up rate m, a rise in the wage rate and the accumulation of money (Afj) (and therefore purchasing power) outside the banking sector - occur in the second period. That is to say, the Mx which has been accumulated in the private-non bank sector, as a result of the creation of money in the first period, stimulates inflation in the second period. However, where footnote 21 applies this may be averted. The end of the boom As long as there is no rise in prices, wages and the rate of yield on bonds, the process whereby money is accumulated in the non-banking sector and production expanded accordingly is repeated, but it must sooner or later come to a standstill. The central bank cannot create an unlimited amount of money, nor can the city banks infinitely expand credit. (Nor, moreover, is it possible for the exchange stabilization fund to keep on buying and selling unlimited amounts of domestic and foreign currency in its attempt to maintain the exchange rate.) At some point, therefore, the central bank will be forced to raise the official 24
First of all the price of bonds falls due to the bond markets being in a state of excess supply, and because of that the rate of yield on bonds will rise. This sort of improvement in the return on bonds will lead rentiers to try and obtain bonds rather than putting their money in time deposits, so deposits will be insufficient. An increase in the rate of interest on time deposits (and hence in the official rate of interest) is essential to attain a state of equilibrium in the deposits sector.
The Wicksellian cumulative process
277
interest rate and refuse to create any more money. Then if interest rates rise investment I2 will decrease, resulting in a sharp decrease in the amount of bonds supplied by enterprises - SB1.25 A sharp downturn in the bond market will result, with an excess of demand for bonds, so that the price of bonds will be increased. The natural interest rate and the natural rate of yield will therefore be lower than the prevailing money interest rate and the prevailing rate of yield on bonds. (It is also possible for the same sort of situation to arise even before the limits of money creation are reached, as a result of circumstances on the investment goods side. The reason for this is that if money and credit are repeatedly created in each period, and investment continues to be held at a high level, then investment opportunities will probably before long dry up, and investment I2 will become very small even without any change in money interest rates. Thus the bond market comes into a state of excess demand and the natural interest rate and the natural rate of yield on bonds are lower than those rates which actually prevail.) So should the relationship between the money interest rate and the natural interest rate for whatever reason be reversed the direction being followed by the economy will also be reversed and the economy will embark on a process of deflation. Prices, wages, interest rates etc. will all begin to fall, and even if this phenomenon does not occur the volume of money held by the non-banking sector will decrease day by day, and the volume of production of consumer goods will go on diminishing in line with the reduction in purchasing power. Effects offorced saving In the above we disregarded completely the effect on the effective demand for goods of changes in prices and wages produced during the cumulative process. In the real world, however, as well as in our model, inflation and deflation exercise a considerable influence on effective demand, resulting accordingly in positive (or negative) forced saving and having a restraining effect (or a stimulating effect) on the volume of production of consumer goods. Forced saving is produced in the following way. If we take into account that W = w [a31Xx + a32X2 + A/» + Nb] Pi = (l + h) (1 4- m)cl9 25
11= m [cxXx + c2X2]
p2 = (1 + t2) (1 + m)c2
It is assumed that the range offluctuationof (U — Ml) is very small, hence according to (17) the decrease in I2 ends up by being shifted as a whole to — 8Bl.
278
Monetary policy
then the demand functions for consumer goods of workers and entrepreneurs can be written, respectively, D\ = cw(l - tw) - [a31X1 + a32X2 + N* + Nb] + yw Pi ~*
r
1
1
~
(18)
"i
• 7,
1+raLl + fi
(19)
1 + ^2 Pi
The demand function for consumer goods on the part of rentiers is
Forced saving means the amount of consumption by these groups which is checked by the increase in prices. First of all, it is clear that the rise in prices px forces down consumption by rentiers (positive forced saving). There are also influences on workers' and entrepreneurs' consumer demand which are exercised through indirect channels via their real-income changes which are brought about along with the change in prices. In order to ascertain these indirect effects let us now consider the changes in prices classified into two separate categories: (1) price changes resulting from a change in the wage rate and (2) those resulting from a change in the interest rate. Providing that throughout an upwards cumulative process the central bank keeps the official interest rate (and hence the interest rate on time deposits) constant, and that the city banks make good all their shortfall in high-powered money by means of advances from the central bank, then the second kind of price rise will not occur. If, on the other hand, a scramble for labour should arise during the cumulative process, then the first kind of price rise will occur. In this latter case the rate of increase in commodity prices px generated by the increase in wages will not be as great as the rate of increase in wages w, so the real wage rate w/px is likely to rise. Therefore according to (18) the consumer demand of workers will increase. This means that negative forced saving will take place. Such price increases do not, however, influence the consumer demand of entrepreneurs.26 As against this, in the case of the second category of price rises (with w therefore unchanging), real wages w/px fall, and the effect is to curb 26
It is assumed, however, that this kind of price fluctuation has no influence on relative prices of the capital goods, pjpv With thefirstkind of price change the interest rate is fixed, so that m/(l + m) in (19) will have a fixed value.
The Wicksellian cumulative process
279
the consumption demands of workers. That is to say, they have positive forced saving effects. In this case, however, m/(l + m) increases,27 so there is a positive effect on entrepreneurs' consumption demands (i.e. negative forced saving). Thus even where there is an identical rise in the price pi of consumer goods the effect in terms of forced saving can be very different depending on the reason behind a particular price rise (i.e. whether it originates from wage rises or increases in interest rates). According to the principle of effective demand, positive forced saving (i.e. a fall in consumer demand) reduces the volume of production of consumer goods Xx to a lower level than that prevailing when there is no positive forced saving, while negative forced saving further increases Xx. As has just been seen, in the case of the first kind of price rise real wages, which increase in conjunction with the price rise, produce a negative forced saving effect on workers through the indirect channel of their real income, while no effect upon entrepreneurs, and a positive effect upon rentiers, will be produced. By contrast, where price rises are of the second kind the real income of entrepreneurs will rise in conjunction with prices, and will therefore exercise a negative forced saving effect through the change in the real income of entrepreneurs. Any other effect on workers and rentiers has also been seen to be positive. Bearing in mind the importance of workers' consumption and the insignificance of entrepreneurs' consumption in terms of total consumer demand, in price rises of the first kind the overall forced saving effect could well be negative, and where the price rises are of the second kind the overall effect will probably be positive. Should a strongly positive overall forced saving effect arise during an upwards cumulative process, then the volume of production of consumer goods Xu which was increasing, will start to decrease. However, if, even where the forced saving effect is positive, it is not that strong, then the pace at which Xx is increasing may well slacken, but the increase will continue as before. Where the overall effect is negative Xx will increase at an even greater rate than it did when the forced saving effect was zero. Whatever the situation, as long as the excess demand for money at a time when the central bank is not creating money is positive, then an excess supply of bonds will exist on the bond market, and therefore the natural interest rate at that time will be higher than the money interest rate. There is therefore operating within the economy an impulse which drives the cumulative process along in an upwards direction. We must finally say a word concerning the effects resulting from a 27
With the second type of price change the interest rate rises; this will cause m to increase, and the value of ra/(l + m) will therefore be greater.
280
Monetary policy
change in the anticipated rate of inflation. When price rises occur in the current period, it can in many cases be predicted that such trends will continue into the future. Since a rise in the predicted inflation rate means a fall in the real rate of interest, investment I2 will increase. This increase in investment will bring about a rise in the volume of production and hence in the volume of employment, provoking wage rises and price rises. This means that anticipated inflation produces actual inflation. In contrast, should there be a fall in the anticipated rate of inflation, or any prospect of deflation, the economy will move rapidly toward recession or stagnation.
Additional notes
(a) The aggregate savings function If the savings of each group are expressed in terms of the symbols given in the Table of Economic Linkages (Table 6) in Chapter 5, we get (i) Workers:
W-pxDwx-twW
(ii) Entrepreneurs:
aU-pxD\-teaIl
(iii) Capitalists (rentiers): Ar + rBr*-pxD[ - tr(Ar + rBr*) On top of these there are corporate savings: (iv) Enterprises: (1 - a)77+ / / + & (v) Foreign trade sector: rp*5F+ &— pxEx — p2E2 — rB* (vi) Government: T+fit - wN8 -
pxGx-p2G2
(vii) Commercial and central bank: Ab + Ac — wNb In view of the fact that Ar + Bl + & + Bg + Ab + Ac = 0 (for explanation of this see pp. 153-54), then total savings (i.e. the sum of (i) to (vii)) can be written as [(W- wM-wNb)
+ 77+ / / + ( r - twW- teall- tr{Ar + rB$) + rp%F\
- [pxDY + pxD\ + PxD\ +pxEx+
PxGx
+ p2E2 + p2G2]
The part within the first round brackets is equivalent to the amount of wages paid out by the consumer goods and capital goods industries, the part within the second lot the indirect tax paid by these industries. The contents of the first square bracket, therefore, equal total value of production of these industries. (For a more detailed explanation see Chapter 5.) For this reason total savings are expressed as: 5 = PxXx + p2X2 - [px(D? + D\ + D\) + px(Ex + Gx) + p2(E2 + G2)]
(t)
Under this formula consumption by workers pxDwx and consumption by entrepreneurs pxDf are both functions of the wages paid to secure 281
282
Additional notes
production p\Xx +p2X2 and the profits margin accruing from it, so they are ultimately dependent on Xx and X2. The consumption by rentiers pxD[ on the other hand is a function of their income, i.e. interest on past savings and is constant as far as the present period is concerned. Furthermore the amount of exports Ex, E2 and also of government demand Gu G2 can be regarded as constant, therefore total savings S become a function of the total amount of production in both industries Xu X2. That is to say S=f(Xu X2). (b) The comparative-advantage theory of international division of labour Let a and b be two countries, and let a's agricultural production per capita be 100 units, and its industrial production 50 units. In country b let agricultural production per capita be 120 units and industrial production 160 units. (Then if those workers who can produce the 100 units of agricultural production in b are transferred to industry, they can manufacture 133 units of industrial goods, whilst in a the workers who produce 100 units of agricultural production can only manufacture 50 units of industrial goods. Therefore country b has a comparative advantage in industry over country a. Conversely the workers who produce 100 units of industrial goods in country b can only produce 75 units of agricultural production, while 200 units will be produced in country a; therefore a has a comparative advantage in agriculture over b.) In country a, transferring one worker from industry to agriculture will mean relinquishing 50 units of industrial goods for 100 units of agricultural production. In country b, transferring one person from agriculture to industry will mean relinquishing 120 units of agricultural production for 160 units of industrial goods. Thus if 100 units of agricultural production are exchanged for more than 50 units of industrial production (for example, 80 units), a will be happy to expand agriculture, and may well exchange the additional agricultural products thus obtained for industrial goods. In fact, in doing so a can obtain more industrial goods (80 units) than the 50 units it relinquished. In the same way, if country b industrializes it will gain 160 units of industrial goods and lose 120 units of agricultural production. However, if these 160 units of industrial production are exchanged for more than 120 units of agricultural production (for example, 200 units), b will be happy to industrialize. Thus an exchange rate of 100:80 for agricultural products to industrial products is one which is beneficial to both a and 6, and which makes specialization worthwhile. That is, an international division of labour increases the prosperity of both countries.
Additional notes
283
(c) Reselling, buying-back and the excess demand function When competitive trading first begins the excess demand is given by the simple difference between the volume of demand and that of supply, but because after that point re-sale and buying-back take place so as to cancel out trading at earlier times, we must take great care to define excess demand to include both buying-back demand and re-sale supply. Let us now suppose that in the white sugar market transactions of volume T took place at a previous time, and that since there was an excess demand the price was competed upwards. Since some of the people who purchased white sugar at the previous time may well now no longer want to buy at the higher price, there will probably be a resale of volume T. As T has already been sold from the supply volume 51(/?1, p2), the volume which must be sold anew is 5 1 (p 1 , p2) - T, in addition to which there is the re-sale volume T. Overall sales are SiiPi, Pi) -T)+ T. On the other hand, the demanders have already obtained T, and since they will re-sell T' from that, the volume of white sugar which will remain at their disposal is T- T'. Since the volume they desire at price Pi, p2, is £>i(/h, p2), they will try to obtain the difference between Dx and T— T in the market. Consequently the difference between overall buying and overall sales on the market is [Dx{px, p2)-(T-T')][(SiiPu Pi)~T) + T] = Dx(pu p2) - Sx{pXj p2), the simple difference between demand and supply or excess demand. That is, the complex excess demand (the left side of the above equation), which appears on the market where a certain volume of transactions have already taken place at a previous point in time and all or part of this volume is re-sold at the present point in time, is equal to the simple excess demand which ignores re-selling (right-hand side). Buying-back will take place where px is falling, but this case is identical to the foregoing. Because of this, the arrow graph (Fig. 5) showing the direction of price changes which was drawn on the basis of the signs of the simple excess demands, Ex and E2, also shows the direction of price fluctuations in the actual competitive trading process where re-selling and buyingback take place in the market. (d) Sham trading Selling which is not backed by real goods ('physicals' or 'actuals') is 'sham' selling, and purchases with no intent to deal in real goods and which are consequently later cancelled out by re-selling are known as 'sham' buying. In the case of transactions in futures it is sufficient if actual goods ('physicals') are got ready by the agreed day of delivery, and therefore it is very difficult, if not impossible, to distinguish between
284
Additional notes
suppliers who are 'sham' selling and those who actually have goods for sale. Once false selling has occurred buying-back must be carried out in the future in order to cancel it out, as we shall soon describe. As a result those who have 'sham' sold or bought will make or lose the cash difference. Those who make bold to 'sham' sell and buy with the intention of profiting from the cash difference - speculators - come to the market and play at least prominent if not the main roles. Such 'sham' trading is an essential feature of the futures markets, but Debreu assumes it does not take place, and has analysed general equilibrium which includes futures transactions. As a result his futures market became scarcely different from spot markets, and naturally he had to be content with a very static analysis. Debreu, G., Theory of Value, New York, John Wiley and Sons Inc., 1959. (e) Price war Price wars were first discussed by Bertrand, J., 'Review of Cournot, "Recherches"', Journal des Savants, 1883, p. 503. Edgeworth said that in the case of duopoly a price rise would occur after price-cutting competition and prices would fluctuate repeatedly. Against this Chamberlin countered that a lower limit would exist for price-cutting, and even if there were only two competitors, the competition to lower prices would cease at this lower limit. See Edgeworth, F. Y., 'La teoria pura del monopolio', Gionarli degli Economisti, 1897; translated in Papers relating to Political Economy, Vol. I, p. I l l , and Chamberlin, E. H., The Theory of Monopolistic Competition, Harvard University Press, 1933, Chapter 3. Although the grounds for my discussion are indeed different, in its conclusion it is closer to Chamberlin than to Edgeworth. (f) The demand for foreign currency due to non-trade causes It is not only trade which brings about international flows of funds. Payments to and from foreign investors for the sales and purchases of domestically issued public bonds and corporate debentures, and payments to and from home-based investors selling and buying foreign bonds and debentures are also an important item. In addition, we must include the payment and receipt of dividends, donations and gifts to and from foreign countries, and other remittances. The same is true for travellers' cheques which overseas travellers take out or bring in. Payment for a country's (Britain's) bonds sold by overseas investors, and interest payments made to them, will be brought to bank B in the form of pounds and converted into dollars. This will then form a part of YA. When they (residents in country A investing abroad) purchase
Additional notes
285
country B's bonds they have to send pounds to the value of the bonds from country A to country B. In order to settle the remittance bills of exchange designated in pounds which investors in country A send to country B, they have to buy the required number of pounds with dollars. The principle for the determination of exchange rates does not change; there are merely minor differences in procedure according to which bank the investors entrust their business. Assuming they opt for bank A, the dealer in bank A will confer with bank B over/?$/£, and if they agree a price he will buy pounds to that value (let us call it YB) and collect P%/$YB dollars from the investor. At this point, the investor can place an order (a limit price order or stop order) with bank A to the effect that ip%/i must not exceed such and such a value'. In this case it may take time to purchase the pounds required (a week or more perhaps). In exactly the same way, bills designated in dollars to be remitted to country B in payment for sales of bonds made by investors in country B, or as interest payments to these investors, will be brought to bank A and will form a part of XB. Conversely, when these investors purchase foreign bonds, they will bring bills of exchange designated in dollars to bank B in order to remit payment from country B to country A (let us call this sum XX ). Bank B will sell XX to bank A, and if the price at the time of the sale is p£/% it will collect Pi/%XX , from the purchasers of the foreign bonds. The total value of the dollars supplied by bank A is XB' + p%/iYB, and the total value of dollars demanded by B is p$/$YX +XX • In this chapter, for the sake of convenience, we shall set XX = YB = 0 and assume that the total of dollars supplied is XB and the total demanded is (g) GNP and national income We will now try to define GNP and national income. If we measure the value of production in the government sector in terms of wN8, i.e. the employment income created by that sector, and if the total value of production in the whole economy including the government is X, then X = p1Xl+p2X2+wN8
(lg)
Bearing in mind columns 1 and 2 of Table 6 it becomes clear that X = total wages (W) — government and bank wages (wN8 + wNb) + depreciation reserves (H) + imports (rp*F) + gross profits from consumer goods and capital goods industries (II) + indirect tax + wN8.
286
Additional notes
Since banking sector profits are the interest income of banks (Ab 4 Ac) minus bank wages wNb the above formula can be written X — total wages + total gross profits + depreciation reserves + imports 4 indirect taxes - (Ab 4 Ac) Total wages here include those paid out by banks and the government, while total gross profits include bank profits. Since from row 8 of the Table we get - (Ab + Ac) = Ar+ Bl + &'+ B* we obtain X = total wages 4 total profits 4 net interest income 4 depreciation reserves 4- imports 4- indirect taxes (2g) Here total profits are the net amount remaining after subtracting the enterprises' interest on debts (-#') from the total gross profits of the whole economy, while net interest income is the amount left when the interest on the debts of a country's government (-B8) has been subtracted from the interest income from that country accruing to rentiers both at home and abroad (Ar+ &). If we next suppose the demand and supply for both consumer goods and capital goods tally with each other, then we get from rows 1 and 2 of Table 6 X = consumption (C) 4- investment (/) 4- exports (E) 4 government expenditure (G)
(3g)
Here consumption C = P\(Di 4- D\ 4 D\), investment / = p2l2, exports E = p1El 4 p2E2, and government expenditure G = p\Gx 4 p2G2 4 If we subtract imports from X, what remains represents the Gross Domestic Product at market prices, while if we add to this the net income from abroad (rBr* — &) we have the Gross National Product at market prices. These are shown by (GDP)m, (GNP)m respectively. If we subtract depreciation reserves from these we get the Net Domestic Product (NDP)m and the Net National Product (NNP)m, while deduction of indirect taxes gives us GDP and GNP at factor cost, i.e. (GDP^ and (GNP)/. Deduction of indirect taxes from (NDP)m and (NNP)m gives us NDP and NNP at factor cost, which is shown as (NDP)/ and (NNP)/. Of these Net National Product is also called National Income.
Additional notes
287
Let us now put down the formula for (GNP)m subdivided into its various constituent parts. (GNP)m = X — imports + net income from overseas = total wages 4- total profits + net interest income + depreciation reserves 4- indirect taxes + net income from overseas = consumption (C) + investment (/) + (exports* — imports*) + government expenditure (G) (4g) Here exports* are exports (E) + income from abroad (rBr), while imports* are imports (rp*F) + income going overseas (&). It goes without saying that (GNP)/? (NNP)m, (NNP^ are made up of similar constituent elements. IS curve and LM curve In most elementary textbooks exports*, imports* and government expenditure are often ignored, and (4g) written merely as Y=C + I
(5g)
Y, of course, represents (GNP)m. In consideration of the financial market, on the other hand, textbooks do not subdivide it into the bond, time deposits and foreign exchange sectors as we have done, but regard it as a single entity which we may eliminate by virtue of Walras' Law. Providing 8MC = 8MS = 0, then a condition of equilibrium in the money market will be L =M
(6g)
After making these simplifications elementary textbooks further assume that Y — C, i.e. saving, is a function of Y, / a function of the interest rate i, and L a function of Y and i; then (5g) and (6g) may be written as S(Y) = I(i) , i) = M
(7g) (8g)
If we have a plane measuring Y along the horizontal axis and i along the vertical axis, and describe on it the partial equilibrium curves of (7g) and (8g), we can call these respectively the IS and the LM curves. The equilibrium values of income and interest rate are determined at the point where these two curves intersect. Curves such as these thus receive lengthy consideration in most elementary textbooks, but despite
288
Additional notes
this are hardly ever used in analysis of the real economy. Savings 5, for example, is regarded purely as function of income Y, while it is HI fact also dependent on the distribution of income throughout the economy as a whole (i.e. to workers, entrepreneurs and rentiers). Furthermore, there is no question but that the demand for money L is also dependent on the price level p. In such cases as this it is essential to have formulae clarifying distribution and explaining prices. These formulae are likely to include as variables both wages and the exchange rate, and in view of this further explanation of them is essential. Thus national income and the interest rate can no longer be determined by the IS and LM curves alone, and what is needed is to carry out, as we have been doing, a comprehensive analysis which bears in mind all sectors of our Table of Economic Linkages. (h) Voluntary unemployment Keynes himself subdivided unemployment into 'voluntary unemployment' and 'involuntary unemployment', but my feeling is that such concepts are inappropriate to an analysis of unemployment. This is first and foremost because the same categories of voluntary and involuntary will exist in employment as well as in unemployment. By involuntarily employed workers one means those who do not wish to work at those wages, but who work because they are compelled to do so. By the same token voluntary unemployment may be divided into those who do not mind if they do not work and those who do not wish to work at those wages but who have no choice but to do so in order to secure a livelihood. From the point of view of their not wanting to work this latter category is voluntary unemployment, but from the viewpoint of their being unemployed even though they have to work it must be seen as involuntary unemployment. For that reason voluntary unemployment in the true meaning of the term must be regarded as limited to those who do not wish to work at a given wage and who, moreover, have no need to do so. If one regards workers as those who have to work to secure a livelihood then voluntary unemployment will be 0 (or a negligible quantity), and all (or almost all) unemployment will be involuntary unemployment. My own view is that, rather than looking at unemployment from the labour supply side, as Keynes did, it is rather more fruitful to follow the examples of Marx and Joan Robinson in looking at it from the demand side and drawing a line between the unemployment based on a shortage of capital and that which originates in a deficiency of effective demand. See also Keynes, J. M., The General Theory of Employment, Interest and Money, Macmillan, 1936, pp. 4-22.
Additional notes
289
(i) The investment multiplier If we bear in mind the formula of repercussion effects (9) of Chapter 7, the formulae (7) and (8) which define bx, b2 give us mc = cw(l - tw)pw + ce(l - te)pe =
and m
_ „ /-I"". , „ P2^2_(n
mF
= Jlli
AY
t>'2
\P2AX2
h JU^
Mi
r- U7
m
1
n
AY
\* 1 - bx
)
n n
(ll)
AY
K
'
Therefore, b'2
,
M
\p24Z2
Since p 2 ^ G = ^ 2 ^ 2 = ^ ? a s a result of the multiplier formula (11) the right-hand side of the above formula is equal to AI/AY—1. We therefore get 1 — mc + mF = AI/AY, i.e. equation (11'). When jxx = jj^ = /A, it should be noted that
(Substitute formula (11) into (li), and take /JL1 = ^ = /x into account; then we obtain the above equation.) Therefore l+
mF=l/(l-fi)
(j) The slopes ofOa, Ob For the purpose of determining, in terms of general theory, which of Oa and Ob is steeper let us discuss the respective influences of these two policies on both employment and government deficit non-numerically and algebraically. First of all the effects on employment and government deficit of an increase in the government's demand for capital goods are shown as wAN
w
b2
p2AG2
px
l-bx
w +p 2
and
p2AG2
\-bx
. .. (lj)
290
Additional notes
where A = (1 + m)tx — + tw — a3X + team —
Pi
Pi
P\
B = (1 + m)f2 — + *w —«32 + 'e«W — /7 2
/7 2
/7 2
(lj) immediately follows from (23) and the definitions of (ox, o^. (2j) can easily be verified if we bear in mind the definition of government deficit as pxGx + p2G2 + wN8 - T, the definition of tax revenue T and formula (16). If, on the other hand, the rate of income tax on wages is reduced by AtwW= AT0, employment and government deficit will alter by wAN
-rzr = AT0 and 0
A
1
AT, \-bx respectively. (5j) is immediately obtained from (28), while (6j) is obtained from the definition of T and (25). It is clearly the ratio of (lj) and (2j) which gives the slope of 0
1 - A[b'2/{\ - bx)\ -B
(W/Pl)a31[l/(1
'
1 - A[(cJ(l
- bx)]cw
- bx)]
We will below suppose a standard case where (i) the proportion of the price of capital goods accounted for by wages (w/p2)a32 is greater than the same proportion in the case of consumer goods (w/pi)a3U and (ii) both consumer goods and capital goods are subject to the same rate of indirect tax (tx — t2). We will first compare the numerators of the two fractions of (7j). If we substitute (ii) into the price equation we get cjpx = c2/p2. If we apply this condition to (7) and (8) and make use of the bu b2 thus obtained we can write (w/Pl)a31[b2/(l - bx)] + (w/p2)a32 [(w/p2)a32 - {w/px)a3X][\ - ce(l \-bx
te)am(cjpx)\
Additional notes
291
According to (i) the part within the first square brackets on the right-hand side of the equation is positive, and the part within the second square brackets is also positive. (Needless to say, ce < 1, te < 1, a < 1, m{cjp^ < 1.) Hence (1 - 60] + (w//72)«32 > (w/pO«3l[l/(l ~ bl)] and because cw < 1 it can be proved that the numerator of the first fraction of (7j) is greater than that of the second fraction. In more or less the same way, it can be shown that ^[62/(1 — 60] + B > ^[ c w/(l - 60]. Hence the first fraction in (7j) (the slope of 0a) is greater than the second (the slope of 06). (k) The 'logical-theoretic' multiplier of credit creation The multiplier process has been explained in the text as something produced successively over time, but as Keynes has made clear, the multiplier process may be produced instantaneously, and multiplier theory may be explained on this assumption of immediacy. Let spontaneous current deposits be w, and total current deposits, including derived ones, be v. Derived current deposits will be produced in the following manner. Banks in receipt of v leave ov as cash reserves and lend the remaining (1 — o)v to enterprises. When the enterprises defray their loans they do so in part by cheque, so e(l — S)v of current deposits is accordingly formed in the banks. This is derived current deposits. We therefore get the equation w + e ( l - S)v = v and if we solve this we get the multiplier formula 1 v=
u 1 — e+ ecr
For the sake of convenience, in the main text below the various multiplier processes will be interpreted as successive, but even despite this the multiplier theory is regarded as instantaneous - i.e. all the effects of the multiplier appear in their entirety immediately. Such a theory is, strictly speaking, non-realistic, but since it is not unrealistic to expect some 80-90% of all the effects of the multiplier to appear within a fairly short space of time, instantaneous multiplier theory can be allowed as a first approximation to reality. We have already in a previous chapter analysed the multiplier effects of investment on GNP as instantaneous, so we must be consistent throughout our theory and think of the credit creation multiplier as instantaneous as well.
292
Additional notes
For Keynes' instantaneous multiplier theory, i.e. for his so-called 'logical theory of the multiplier', see Keynes, J. M., The General Theory of Employment, Interest and Money, Macmillan, 1936, pp. 115-23. (1) Inflation The standard view divides inflation into demand-pull inflation and cost-push inflation. The former is produced where there is excess demand for a large number of goods. The price of these goods rises to absorb the excess demand, leading to entrepreneurs' securing of windfall profits. Such profits may well be invested in the expansion of production, and if the economy is already in a state nearing full employment an excess demand for labour will result, leading to a rise in wages. In this case demand-pull inflation gives rise to cost-push inflation. In our model, however, the demand-pull type of inflation does not appear, as it is assumed that even where excess demand does exist it is absorbed not by an increase in prices but by the regulation of quantities. Even in such a case as this, however, if the production exceeds the capacity of the enterprise the average cost per unit of output will increase, and for that reason prices too will rise. By contrast the various kinds of cost-push inflation are produced where there is an increase in commodity prices as a result of a rise in the prices of the factors of production, e.g. wage rates. Wage inflation is the most significant of these, but in countries with a high degree of dependence on imports other kinds of inflation cannot be disregarded, for example inflation caused by increases in the prices of imported raw materials or fuelstuffs, and that stemming from a weakening of the value of a country's currency vis-a-vis the dollar. Keynes termed the inflation resulting from an increase in the price of factors of production (and hence in the income of those individuals who supply those factors) Income Inflation, while that provoked by the existence of windfall profits he called Profit Inflation. See Keynes, J. M., A Treatise on Money, vols I and II, London: Macmillan, 1930.
Exercises
Introduction 1. Substantiate whether Say's law is equivalent to the following propositions: (i) no general overproduction is possible, (ii) there is no obstacle to full employment, (iii) there is no independent investment function. 2. 'The ability of neo-classical growth theory (e.g. R. M. Solow, 'Contributions to the Theory of Economic Growth', Quarterly Journal of Economics, Vol. 70, No. 1, 1956, pp. 65-94) to generate a full-employment and full-capacity growth path depends on two hypotheses: (i) the substitutability of capital and labour and (ii) Say's law. The former precludes Marxian unemployment, the latter Keynesian unemployment.' Discuss. Chapters 1 and 2 3. In Table 1, leaving aside carte blanche orders, all buyers (sellers) place stop or limit orders. That is to say they demand (supply) a fixed amount provided the price is below (above) or equal to a designated price, and should the price be higher (lower) than the designated one there will be no demand (supply). According to normal demand theory demand changes elastically in relation to changes in price, e.g. if there should be a demand from the buyer of 100 bushels of rice when the price is £10, it will be of 70 bushels at £11, and 50 bushels at £12. The same applies to supply. This being the case, can the kind of exchange shown in Table 1 cope with elastic supply and demand? Verify that elastic demand and supply can be broken down into a series of stop orders, and draw the demand and supply curves, for the elastic case, as the sum of stop orders thus obtained. 4. Verify the following propositions: (i) If there are two groups of goods, with substitutability prevailing within the groups, and complementarity between them, then these goods will fulfil Hicks' chain rules: (a) complements of complements (or substitutes of substitutes) will be substitutes, (b) complements of substitutes (or substitutes of complements) will be complements, (ii) To the contrary, if complementarity prevails within the groups and 293
294
Exercises substitutability between them we will have the reverse of Hicks' chain rules: (a') complements of complements (or substitutes of substitutes) is complements, (b') complements of substitutes (or substitutes of complements) will be substitutes. (For (i) the conditions for stability and the comparative statistical laws have been investigated, for (ii) no investigation has as yet taken place.)
5. If the price of bonds pb rises (i.e. if the yield on bonds worsens) then the demand for bonds will decrease and people will increase their holdings of time deposits. If on the other hand the rate of interest on time deposits id is increased, then people's holdings of time deposits will also increase and the demand for bonds will fall. Corroborate the following propositions: (i) an autonomous increase in the demand for bonds increases pb and reduces id. (ii) if people have a higher propensity to put their savings in time deposits then pb will increase and id will fall.
Chapter 3 6. When the matrix of the coefficients of production .
is profitable at m = 0 then A is said to fulfil the Hawkins-Simon condition. Economists often assume the Hawkins-Simon condition in developing a theory of production, but this condition alone does not necessarily mean that the equilibrium prices are positive where a mark-up rate has been set at an appropriate value compatible with a given rate of interest. Verify this statement. Moreover, demonstrate that where some enterprises are producing not just one kind of good but also byproducts the price equations may produce negative solutions. Can these be interpreted as equilibrium prices? 7. According to the neo-classical theory of the firm each firm tries to maximize its profits on the basis of given cost and demand curves. However, profits gained in this manner will be eroded by the price war waged by the firm's competitors and new enterprises entering the fray, and each enterprise will ultimately have to be satisfied with the minimum mark-up rate. Thus the neo-classical theory of profit maximization may exist side by side with the full-cost principle and its minimum mark-up rate. Schumpeter believed that where enterprises are not satisfied with this sort of profit they have constantly to work for innovations and to secure
Exercises
295
monopolistic (or surplus) profits until other firms have made the same innovations. According to Schumpeter any situation where the minimum mark-up rate operates is one of slump, and since various innovations are carried out so as to break out of this situation, capitalism is constantly being stimulated by new innovations and is therefore essentially a dynamic system. Read J. A. Schumpeter, The Theory of Economic Development, Cambridge, Mass., Harvard University Press, 1934, and consider how this view of Schumpeter's can be incorporated into the framework of the present volume. 8. Read Alan MacFarlane, The Origins of English Individualism, Oxford, Blackwell, 1978, and consider whether or not there was in England a stage of 'simple commodity production'. If the answer to this question is 'no', does this then constitute a criticism of Marx?
Chapter 4
9. Clarifying the basis of the two following views of the trade war, discuss: (i) buying a good cheaper abroad than we can produce it at home is to our advantage, (ii) we are subject to the intolerable competition of a foreign rival, who enjoys such superior facilities for the production of goods that he inundates our national market and cause an enormous redundancy of workers in our factories. 10. The trade balances with Japan of both Britain and the US are deeply in the red. These two countries demand some concessions on the side of Japan, but Japan maintains that (i) as far as a country's trade balance is concerned it is not necessary for there to be a balance in its payment vis-a-vis any other individual country; it is enough for there to be a balance between that country and all other countries. Japan also maintains that (ii) it follows from this that it is not necessary to achieve a balance in trade alone, there is no problem as long as there is no serious imbalance in the overall balance of payments. Which of these assertions is correct? Japan's or that of Britain and the US? 11. (i) The yen is becoming stronger against both the dollar and the pound, (ii) The value of the pound against the dollar is falling. On the other hand (iii) Japan's trading balances with both the US and Britain are favourable, therefore Xc and Yc are increasing. Is it possible to explain (i) and (ii) by means of (iii)? If it is not possible, what further conditions would be necessary to do this?
296
Exercises Chapter 5
12. Derive Hicks' following equation concerning the firm by means of synthesizing the production sector and investment sector of the firm as shown in the Table of Economic Linkages (Table 6). Acquisition of cash = value of output — value of input — interest on debts — dividends -I- value of bonds issued (or sold). (See J. R. Hicks, Value and Capital, 2nd edn, 1946, Oxford, Clarendon Press, p. 158). What items in the table constitute 'value of input' in this formula? 13. Extend Table 6 so as to accommodate the North Sea oil industry. Discuss the view that North Sea oil has been a burden to Britain because it has meant that her exchange rate has risen relative to those of other Western countries and this has caused a fall in employment in manufacturing. What are the underlying assumptions for your argument? 14. Bearing in mind the definition of currency Mx, verify that Mx, Lx on p. 154 are its volumes at the beginning and end of the period respectively. How are the amounts of high-powered money at the beginning and end of the period respectively expressed? How is the amount of M2 expressed?
Chapter 6
15. Let u\ and u\ be the non-governmental exogenous demands for consumer and capital goods, excluding from them the government's demands Gx and G2 respectively. We may define a new Marxian unemployment as being the difference between the supply of labour N and the constrained maximum of employment subject to a4XXx + aA2X2 ^ K, Xx = bxXx +b2X2+
u[\ + Gx,
X2 = u(\ + G 2 ,
Show diagrammatically that this new unemployment is larger than the original in the text. Explain economically why it is so, by verifying: (i) that if the consumer goods industry is less labour-intensive than the capital goods industry, the employment policy to minimize unemployment
Exercises
297
amounts to Gj = 0 and G2 > 0; (ii) that if the labour-intensive condition is reversed, it amounts to Gx > 0 and G2 = 0, and (iii) that if the two industries are equal in labour intensity, Gx and G2 are indeterminate. 16. Total savings which can be written as (t) on p. 281 are also equal to the sum of all the elements of rows 9-12 in Table 6, plus/?2/2 (savings in kind). Using this relationship compare Keynes' definition of Say's law (total savings are identically equal to investment p2l2) and the Lange-Patinkin definition of it (excess demand for domestic currency is identically equal to 0). (O. Lange, 'Say's Law: A Restatement and Criticism', in his Papers in Economics and Sociology 1930-1960, Oxford, Pergamon Press, 1970; D. Patinkin, Money Interest and Prices, 2nd edn, New York, Harper and Row, 1965). Show that Say's law in the Keynes sense is incompatible with the existence of an investment function (so that the economy in Chapters 5-8 is freed from the law). Where Say's law prevails in the Lange-Patinkin sense is general overproduction impossible? 17. Examine the effects of a change in the pound sterling exchange rate upon production and employment. Discuss the following statement by Keynes: It is worse, in an impoverished world, to provoke unemployment than to disappoint the rentier.
Chapter 7 18. If you were the Chancellor of the Exchequer what kind of fiscal policy would you propose in order to decrease unemployment? Explain, and advance the reasoning to support your recommendation. Also, based on those numerical values of the coefficients of the model which are used in the text, estimate the effects of your policy on employment and GNP. 19. Give an analytical verification of the fact that progressive taxation plays the role of an automatic stabilizer (or built-in stabilizer) which reduces the amplitude of GNP by bringing about a fall in the value of the investment multiplier in times of boom and increasing its value in time of slump. Furthermore, given that the value of the multiplier is a low one of around 1.58 (see p. 225 above), is not the above assertion likely to exaggerate the automatic regulatory function of the economy? Discuss.
298
Exercises
20. The various effects of a change in the rates of indirect tax tu t2 have been discussed in the present volume only in a heuristic fashion. Using mathematical formulae, give a rigorous analysis of the effects exerted by such changes on px, p2, GNP etc., (i) disregarding any repercussions on ib, id, r, and (ii) taking into account such repercussions.
Chapter 8 21. 'Say's law implies a peculiar nature of the demand for money, namely, that the individuals in our system, taken together, are always satisfied with the existing amount of money and never wish to hold either more or less. Under these circumstances the money prices of commodities are indeterminate.' (O. Lange). Thus Say's law is inconsistent with the existence of a money economy.' (D. Patinkin). Discuss. Also discuss whether this kind of proposition remains correct where each firm determines its prices according to the full-cost principle, with particular reference to the case where the central banks adopt the policy of adjusting the supply of money so as to cancel any excess demand for (or supply of) money from the public. 22. Discuss the view that Keynesian policy is a policy for creating inflation, rather than a remedy for unemployment. 23. Discuss the effects of the creation of money upon prices, output, etc. in each of the two following cases: (i) where the exchange stabilization fund creates money to stabilize the foreign value of the country's own currency (the pound), (ii) where the central bank creates money by buying government bonds issued to finance an expansion in government personnel.
Index
Arbitrage dealing, 53 Arrow, K. J.,40n
Credit creation, 252, 291 multiplier, 254 multiplier process of, 255
Balanced budgeting multiplier effect of, 243 Bank central, 150, 262 central bank lending, 160 commercial, 150 Bertrand,J.,284 Bohm-Bawerk, E. von, 156n Bond annual yield of, 171,172 market, 167 Budget equation for banks, 150 city banks, 256 entrepreneurs, 144 exchange stabilization fund, 151 firm's investment sector, 147 foreign trade, 148 government, 149 investment sector, 256 rentiers, 145 workers, 143
Dealer foreign exchange, 114 Deane,P., 103n Debreu, G.,284 Deficit financing, 235 Deflation, 257 Demand curve, 21 estimated, 70 Devaluation of the pound, 109 Discount, 130 Disorderly bidding, 20
Cassel,G.,15 Chamberlin, E. H.,284 Clark, C.G.,219n Classical (neo-classical) school, 83, 84,189 Cob-web theorem, 159 Comparative-advantage theory of international division of labour, 282 of production cost, 36 Complementary goods, 41, 56, 58 Cost assessment current price approach, 74 historical cost approach, 75 Cost curve average, 68 Cournot, A.,54n,88n
Economic blocs, 102 Economic linkage table, 140, 255, 256 Edgeworth,F. Y.,284 Effective demand principle of, 6,152 Employment multiplier, 227 Equilibrium existence and stability of general, 47 existence of, 23 multiple, 24 partial equilibrium curve, 44 Excess demand function, 283 quantity of, 42 Exchange, 17, 39 commodity, 17 composite, 56, 56n general equilibrium within, 43 stock,161 Exchange rate and product prices, 107 automatic balance-of-payments adjusting function of, 122 basic rate, 120 cross rate, 120 export and, 211 fixed rate system, 105,106
299
300
Index
Exchange rate (contd) fluctuating rate system, 106 inter-bank market of, 112 under gold standard, 105 Exchange Stabilization Fund, 139, 151, 262 Fixprice economy, 32, 39, 39n historical background to making of, 34 Fiscal expenditure policy, 231 Foreign currency demand for (due to non-trade causes), 384 Flexprice economy, 32 Free trade theory, 35, 99,100 Friedman, M., 204n, 205, 205n, 206, 206n Full-cost principle, 27, 68, 91, 205, 205n price competition under, 29 Futures market of commodities, 61, 64, 74 of foreign exchange, 128 Government, 149 Gross National Product (GNP), 223, 285 Haavelmo,T.,249n Hahn,F. H.,40n Hall,R. L.,27n Hedging, 63, 65, 72, 75,131 Hicks, J.R.,156n,191n, 296 chain rules, 61n Law of prices, 48, 56, 82 Hitch, C.J.,27n Industrial country lacking natural resources, 4,135 large, medium-sized and small, 3 Industrialization de-industrialization, 218 tertiary, 219, 220 Inflation, 257, 272, 292 cost-push, 292 demand-push, 292 from the creation of money, 272 Interest money rate of, 271 natural rate of, 271 real rate of, 174n, 274n system, 171 Investment multiplier, 221, 224, 289 approximate numerical value of, 225 IS curve, 287 Jaffe,W.,137 Japan Inc., 252 Kalecki,M.,32n
Kantorovich, L. V., 2n Keynes, J. M., 7, 7n, 32-4,112n, 189, 197n,208,249,249n,250,292 Kuznets, S. S.,219n Labour market structure of, 179 Labour union enterprise, 185 trade union, 183 Labour theory of value, 92, 93 Lending policy, 266 LM curve, 287 Luddism, 212 Machinery, 212 Marginal productivity theory, 84 Marginal propensity to consume, 224, 226 to import, 224, 225 Mark-up rate, 27, 27n, 30 Marx, K., 13,13n, 93,188,192-7 Mitchel,B.R.,103n Money deposit, 164 high-powered, 165, 254 Mly 164,166 M2,165 multiplier, 167, 254 quantity theory of, 201 Monopolist, 95 Morishima, M., 7n, 61n, 94n, 137n, 179n Mosak,J.L.,41n National income, 285 Nurkse, R.,189n Oil shock, 261 Open market operation, 263 Overloan, 268 Phillips curve, 204, 208, 208n Premium, 130 Price-adjustment, 96 function, 22 Price-taker, 95 Price war, 88, 284 Production period, 72,156 price: existence of equilibrium, 78-81; repercussions of changes in, 82 structure of: single-track progression, 156n; multi-track progression, 156n Profitable at the mark-up rate m, 80, 294 Protectionism, 99
301
Index Purchasing power parity theory, 124 Quantity-adjustment, 96,158 Reselling, 284 Ricardo, D., 7, 91, 92n, 212, 218 Robbins, L.,2n Robinson, J., 189n Robot, 216, 220 Samuelson,P. A.,23 Saving aggregate saving function, 281, 297 forced, 277 Say's Law, 6, 7, 7n, 293, 297, 298 Schumpeter, J. A., 257n, 259n, 294 Sham buying, 61 buying back, 284 trading, 283 Small proprietor, 95 Speculation, 63,131 Stagflation, 259 Stock market, 161 Substitutive goods, 41, 43, 55 Supply curve, 21 Suzuki, Y.,161n Tax increase catalytic effect of, 243
Tax reduction, 237 income tax on wages, 237 income tax paid by entrepreneurs, 239 in indirect taxation, 240 Time deposits sector, 170 Trade war, 127 Trading by tender, 15 competitive, 39 cross-trading, 15 Unemployment classical (neo-classical) view of, 200, 201 disguised, 189,189n, 219 Keynesian, 190,197,198 Marxian, 190,192-7, 296 natural rate of, 206 voluntary, 288 Wage differentials between enterprises, 185 effect of a cut in, 208 flexibility of, 182 Walras,L.,7,7n, 13,14 Walras' Law, 152 Weber, M., 179,179n Wicksell,K.,156n,272n Wicksellian cumulative process, 271 Wiles,P.,186n Zimmerman, M., 30n