The Economic Development of The Third World Since 1900

The Economic Development of the Third World since 1900
PAUL BAIROCH/Translated by
1900
1920
1940
1960
1980
>ngan hak cipta
THE ECONOMIC DEVELOPMENT OF THE THIRD WORLD SINCE 1900
T h JL s On DK3W-2QH-YDXB
DK3W -2QH -YDXB

M aterial com direitos autorais
M aterial com direitos autc
The Economic Development o f The Third World Since 1900
Paul Bairoch translated by Cynthia Postan
UNIVERSITY
OF
CALIFORNIA
PRESS
B E R K E L E Y AND LOS A N G E L E S
Bahan dengan hak cipta
University o f California Press Berkeley and Los Angeles, California 1975 Paul Bairoch California Paperback Edition 1977 ISBN: 0 520 03554 2 Library of Congress Catalog Card Number: 74 16706 Typeset in IBM Journal Roman by Preface Limited, Salisbury, Wilts and printed in Great Britain
NOTE TO THE USA EDITION Throughout this book the word billion has been used in the sense of 1000,000,000
Contents
L1ST_QJELSTAT1ST1CAL-TABLES_________________________ ix
NOTE TO THE ENGLISH EDITION________________________xii
INTRODUCTION_________________________________________1 1 P-QP-ULA11QN____________________________________________5 A Changes in population o f the under-developed countries from 1900 to 1970__________________________ 5 B R ecent changes and future prospects___________________ 8 2 AGRICULTURE________________________________________ 13 A Agricultural production______________________________ 14 (i) Changes in production up to 1934/38 ____________ 14 (ii) Changes in agricultural production since 1 9 3 4 m ._______________________________________ 16 B Agricultural productivity____________________________ (i) Method o f calculating the index o f agricultural productivity ___________________________________ (ii) Productivity thresholds _________________________ (Hi) Index o f agricultural productivity: Results and comments _____________________________________ 21 22 26 22
C Recent changes: A green revolution?__________________ 45 3 EXTRAGI1YE.I_ND_ USTRY______________________________ 42

A J h e x o n n e x iQii-b.elw.eeii-ttie-extrac.tiy.e-aiid
m anufacturing industries____________________________ 49 B Changes in extractive industry: Causes and consequences_______________________________________ 51 (i) Changes in output ______________________________ 51 (ii) Causes o f growth in output ______________________ 54 (Hi) The consequences o f growth o f mining and extraction in under-developed countries __________ 57 v
vi 4
Contents MANUEACTJURMG-IND-UST R Y_________________________ 61 A Rates o f growth in m anufacturing industry (i) All non-communist , under-developed countries (ii) Rate o f growth in manufacturing industry by country B Level o f developm ent o f m anufacturing industry (i) General indicators based on the relative position o f manufacturing industry (ii) Simple indicators C Agriculture and m anufacturing industry (i) Levels o f development in agriculture and manufacturing (ii) Changing conditions in agriculture and industry 5 FOREIGN TRADE A General changes in foreign trade B The pattern o f foreign trade by commodities 92 98 64 64 72 77 78 83 88 88 90
C The geographical pattern o f foreign trade 101 (i) General changes 101 (ii) The place o f the under-developed countries in the foreign trade o f the western developed c m nmes_____________________________________ 104 D Note on th e foreign trade o f China 6 THE TERMS OF TRADE A The term s o f trade 18701938 (i) Critique o f the estimates proving a secular fall in the terms o f trade o f primary products (ii) Additional proofs o f the reality o f a secular improvement in the terms o f trade o f primary products B The terms o f trade 19261970 (i) Changes in the prices o f raw materials and manufactures 1926/29 to 1950/54 (ii) Changes in terms o f trade, 19481970 109 111 112 113
116 123 123 126
Contents 7 J .
vii
THE LEVEL_QILED_ U_CAT1QN__________________________ 115 M . -mrfj.A ______ JLAX/l_______________________________ A The level o f literacy and prim ary schooling B Secondary and higher education 135 140 144 145 148 151 156
URBANIZATION A Changes in urbanization B Urbanization and levels o f economic development C Causes o f urban inflation D Urbanization in China
IHEJLABQ11EJQRCE-&-EMELQYMENT________________118 A_Th.kb.Q.iir-Qcc___________________________________ 118 (i) Past changes and future projections o f total labour force 158 (ii) The structure o f the labour force by bra?ich o f activity 159 B Unem ploym ent and under-em ploym ent (i) Rural unemployment and under-employment (ii) Urban unemployment and under employment 165 167 168 172 172 172 175 177 177 182 183 183 186 188 190
10
M AOR O-ECON O MIC DATA A Capital form ation and im ports of capital equipm ent (i) Changes in the rate o f capital formation (ii) Capital equipment as a share o f imports B Financial aid (i) General changes (ii) Geographical distribution o f financial aid C Growth and level o f per capita product or national income (i) Rate o f growth o f Gross Product in non communist , less-developed countries (ii) Comparisons with developed countries (iii) Mainland China (iv) Changes in per capita income and output
Contents GENERAL CONCLUSIONS POSTSCRIPT (1974) NOTES APPENDIX: SYNOPTIC TABLE BIBLIOGRAPHY INDEX 194 210 211 244 250
Material com direitos autorais
List o f Statistical Tables

Page 1
2
Changes in the population o f the less-developed countries and of the world Projection o f population for less-developed countries an.d-th.e_w.Qrld-19-7-Q"2.Q .Q .Q ________________________________ 11 World o u tp u t o f the principal tropical commodities from 1880 to 1970_______________________________________L5 Agricultural production indices in less-developed countries________________________________________________ 1_2 Annual rate o f growth o f total and per capita agricultural production___________________________________ 18 Changes and annual rates of variation in the index of total agricultural production______________________________ 20 Variations in yields 1947-1962____________________________ 28 Level o f agricultural productivity: averages 19.6-Q/.64 19.6.8/.7.2_______________________________________ 20 Comparison between agricultural productivity and availability o f agricultural land____________________________ 33 Weighted average o f index numbers o f agricultural productivity_____________________________________________ 36 Index numbers o f agricultural productivity by country_______38 Change in the male working agricultural population_________ 39 Comparisons between levels o f agricultural productivity_____ 40 Annual rates o f growth o f output, m anufacturing and extractive industry_______________________________________ 50 Index o f o u tp u t o f extractive industries o f all non communist, less-developed countries_______________________ 52 Changes in the less-developed countries share in world production o f some extractive commodities 19131970 55 ix
3 4 5
6
7
8
9
10
n
12
n 14 15 16
x 17
List o f Statistical Tables Index and rate o f growth o f o u tp u t of m anufacturing industry in the less-developed countries (excluding com m unist countries) Annual rate o f growth o f m anufacturing industry Comparison o f annual rates o f growth in m anufacturing o u tp u t according to stage o f industrialization and economic system Index o f o u tp u t in m anufacturing industry o f all non com m unist less-developed countries Indices and rates o f growth o f m anufacturing industry Percentage of the male working population occupied in m anufacturing industry Percentage o f the male working population em ployed in m anufacturing industry in some developed countries 1 8 4 1 -1 9 7 0 Gross domestic product generated by m anufacturing (percentages in relation to total) Supplem entary indicators o f the level of industrial development in the less-developed countries, 1970 Indicators o f the level o f industrial development of selected developed countries 17601970 Comparative ranking o f the levels o f industrial and agricultural development Comparison between the growth o f purchasing power generated by agriculture and the industrial o u tp u t in France Foreign trade o f less-developed countries 19001970 Change in foreign trade and trade balance of non com m unist, less-developed countries according to geographical regions Foreign trade by com m odity classes o f non-communist, less-developed countries Changes in the regional origins and destinations of the foreign trade o f the non-communist less-developed countries 19381970
65 67
18 19
69 71 75 79
20 21 22 23
81 82 84
86
24 25 26 27 28
89
90 93
29 30
97 99
31 32
102
List o f Statistical Tables 33 The share o f the less-developed countries in the exports o f the developed countries as a percentage o f total exports o f the various products Exports from developed countries intended for the non com m unist Third World in 1970 com pared to total exports and to G.N.P. at m arket prices Relationship between freight costs and prices o f selected prim ary products 1876/801926/29 Changes in prices o f some prim ary products and m anufactures between 1876/80 and 1926/29 Index o f export prices o f primary products from lessdeveloped countries Changes in the terms o f trade Changes in the export prices o f prim ary products (for all non-comm unist countries) Changes in the share of synthetics in world o u tp u t o f textile fibres and rubber Changes in the rate o f illiteracy Changes in numbers o f students at secondary and higher level o f education Average annual rate o f growth o f urban population and rate o f urbanization Comparative changes in rates o f urbanization 1850 2000 Comparison between the levels of urbanization and percentages o f the active population em ployed in m anufacturing industry Levels o f urbanization in non-com m unist less-developed countries
xi
105
34
107 119 120 124 127 130 131 136 141 146 147
35 36 37 38 39 40 41 42 43 44 45
149 150
46 47
Changes in the structure o f the active population o f the less-developed countries between 1900 and 1970 according to type o f em ploym ent 160 Changes in rates o f capital form ation 19501968/70 Share o f im ports in the gross capital form ation o f machinery and transport equipm ent 174 176
48 49
xii 50 51 52 53
List o f Statistical Tables N et flow o f financial resources to non-comm unist less-developed countries Annual rates o f changes in the Gross Domestic Product per capita Long-term changes in national income per capita Gross Domestic Product per capita Synoptic Table
179 184 191 193 246
Note to the English Edition

The present edition, though based on the fourth French edition, differs from it in a num ber of respects, such as the addition o f two Chapters (Urbanization and The Labour Force & Em ploym ent), the updating o f practically all the figures, the numbering of the tables and notes, the presentation o f certain tables and the contents of the bibliography, which has been specially adapted to the literature of th e subject in the English language. Unless otherwise stated the m etric system o f weights and measures has been used throughout.
y^- >-7^ yy
Introduction
The subject o f this book is the nature o f under-developed economies and their behaviour over the long term . A study o f economic evolution cannot ignore structural factors since the tw o are partly interdependent, for just as such factors are essential to the explanation o f economic circumstances so changes in these circum stances will m odify the structural factors themselves. I have discussed structural issues more specifically in my study Revolution indus trielle et sous-developpementd and the research I undertook then made me aware th at there is a notable gap in the large and often distinguished literature on the Third World the lack o f com parative studies o f its economic development. While there are several case studies on the developm ent of individual countries, or small groups of countries, and also some studies o f broader scope dealing w ith certain aspects o f this developm ent, there are none (as far as I know) which might help to single out the various characteristics of economic developm ent in the Third World over a long enough period (i.e. the last seventy years) or for a large enough num ber of countries to make the analysis applicable to all the less-developed countries, or at any rate all th e larger ones. In writing this book I have attem pted to fill this gap and to produce a study the aims of which are accordingly: 1. To extend the diagnosis over a sufficiently long period, which I have defined as from 1900 to 1970. In every case where the nature o f the data makes this necessary annual averages have been calculated. For agricultural data averages have been calculated on a five-year basis, while for sectors less subject to fluctuations three-year averages have been used. 2. To include a large num ber o f cases. Whenever index num bers or series relating to the entire Third World, or to large groups of less-developed countries, are available or can be calculated, these have been given precedence. In the case o f individual countries, since it is obviously impossible to deal with every country or territory now in existence (the balkanization o f form er colonial empires has produced nearly 170 economic or political entities), I have thought it best to base the study m ainly, b u t not wholly, upon a sample of tw enty-four countries, chosen in the following manner: 1
The Economic Development o f the Third World since 1900 (a) I have included a few countries from each o f the larger geographical regions. Africa (both Black Africa and N orth Africa); Latin America; Asia; and the Middle East. (b) A few countries from different econom ic and political regimes. (c) A few countries w ith a large population.
The countries selected, whose population represents 80 per cent of th a t o f the whole Third World, are:
AFRICA AMERICA ASI A M I D D L E EAST
Ghana Kenya
Argentina Brazil
Madagascar Chile Morocco Cuba Nigeria Mexico Tunisia Peru Zaire (until Venezuela 1972 Dem. Rep. Congo)
China Sri Lanka (Ceylon) India Pakistan Indonesia Philippines Thailand
Egypt Iraq Turkey
The dem ographic significance o f the country and the methodological im portance o f the Chinese way of developm ent dem anded the inclusion o f mainland China in this list, even though this entailed research on a massive scale o u t o f all proportion to the addition o f a single country. This list o f tw enty-four countries is not treated as exclusive and will often be added to , or equally, in the absence o f statistics, reduced. Finally, th e Appendix contains tw o synoptic tables giving the principal economic and social d ata available for the 94 under-developed territories each possessing more than 400,000 inhabitants in 1965 and representing, from the economic and social point o f view, more than 99 per cent o f the Third World. 3. To differentiate various aspects o f econom ic growth in the under-developed countries. The chapter titles o f this study indicate its range o f interest: 1. Population; 2. Agriculture; 3. Extractive industry; 4. M anufacturing industry; 5. Foreign trade; 6. The term s of trade; 7. The level o f education; 8. Urbanization; 9. The labour force and em ploym ent; 10. Macro-economic data. These chapters are not o f equal im portance. In the chapter on
Introduction
population, for instance, it is enough to give a brief summary o f the quantitative material, while the chapter on the developm ent of agriculture shows th at the subject is o f greater im portance than alm ost any other. I have also attem pted to fill in certain gaps in international docum entation, in particular by calculating an index of agricultural productivity,2 and an index o f the production o f the extractive industries. Given the num ber o f countries included and the im portance o f the factors under consideration, an in-depth exam ination o f the con ditions and situation peculiar to each country could n o t be expected of the present study. It has only been possible to make a diagnosis o f the econom ic developm ent o f the Third World in its entirety. Needless to say, the exceptions to this diagnosis will be dealt w ith as they affect the different countries. I have tried to keep the argument as concise as possible by excluding any m aterial which is superfluous or o f doubtful usefulness. To increase the value o f the diagnosis, systematic comparisons have been drawn with the economic progress of the developed countries during the take o f f period o f their history always distinguishing as far as possible between those countries which industrialized early and those which entered the race at a later date. The following remarks m ay clarify the m ethodology used. 1. Terminology. The term inology referrring to under-developed countries has been the subject o f some controversy and there is no intention here o f entering th e lists. In this study I have used the term s less-developed,, developing*, under-developed*, non industrialized* and Third World* synonym ously, and have used the word developed* solely in its econom ic sense. 2. Geographical definition. It has n o t been thought necessary to reclassify the territories considered to be under-developed. The classification used is based on th at o f the statistical services o f the United Nations which, though n o t perfect (could any be?), has the advantage o f being widely em ployed and understood. The U.N. statistical services distinguish three geographic regions: (a) developed countries; (b) developing countries; (c) com m unist countries (or centrally-planned economies). Among the non-com m unist countries the following are considered to be under-developed: all those in Asia w ith the exception o f Japan; all those in Africa w ith the exception o f South Africa; all Latin-American countries; and all those in Oceania w ith the
The Economic Development o f the Third World since 1900
exception o f Australia and New Zealand. This group will be referred to here as the non-com m unist under-developed countries.3 In addition to the countries m entioned above, the under developed world will, for the purposes o f this study, include the com m unist states o f Asia (China, Korea, N orth Vietnam and Mongolia). Readers are rem inded, though, th at till recently the statistical services of th e United Nations included Cuba amongst the non-com m unist group o f under-developed countries. 3. How up to date is this stu d y? While very recent figures would n o t norm ally have been given special treatm ent in a long-term study of this kind, I have m ade every effort to include as m uch recent material as possible. This English edition (based on the revised fourth French one) incorporates all facts th at are available up to the middle o f 1973. Even so, it m ust be remem bered that the statistical time lag varies greatly from topic to topic and country to country. 4. The provisional nature o f recent statistics and the margin o f error. M ethods used in compiling statistics, especially those based on estimates or sampling techniques (national accounting, agricultural and industrial production and population) m ean that statistical institutions have to revise their figures periodically. Thus, as a general ru le, it is w ise to trea t sta tistic s rela tin g to th e last fe w y e a rs as provisional. This is in any case a necessary precaution w ith all published statistics. It may be assumed, therefore, th a t an im portant part o f the statistics used here contains margins o f error. Wherever these are significant th e readers attention will be drawn to the fact. But to om it all statistics containing a margin of error would am ount to renouncing any attem pt at econom ic analysis.
Bahan dengan hak cipt
I
INTRODUCTION
Population
Both the causes and consequences o f demographic expansion afflicting the under-developed countries are so well known that it will suffice in th e present chapter to deal briefly with a few quantitative facts. The chapter is divided into two parts. The first is devoted to the long-term changes o f population betw een 1900 and 1970, continent by continent and for the whole o f the Third World. Details relating to separate countries will be found in the synoptic table which appears in the Appendix (p. 244) and which includes population in 1970 and the rates o f demographic growth from 1960 to 1970 for m ost of the under-developed territories. The second part of the chapter will deal mainly with recent changes in birth rates and, very briefly, with population projections. A. Changes in population of the under-developed countries from 1900 to 1970 Table 1 has been constructed on the basis o f inform ation published by th e statistical office o f the United Nations. With the exception of data prior to 1950, the table is based on the latest issue (for 1972) of the dem ographic year book (New York, 1973). It should be noted (a) th a t figures prior to 1950 contain a margin of error which gets progressively larger the earlier the period to which they refer and (b) th a t the data from 1900 to 1940 has been adjusted in order to take account o f recent corrections made in the 1950 figures. D ata relating to the com m unist countries m ust be considered as approxim ate only, since so little is known about the population of China which, in fact, dom inates the total. The figures reproduced here are taken from the demographic year books o f the United Nations and assume a rate o f population growth of 1.8 per cent per annum for the period 1950 to 1970. However, it appears after exam ination o f the literature relating to Chinese demographic problem s1 th at actual rates o f growth were higher. From 1950 to 1957, a period during which figures were published by the Chinese authorities, demographic growth was of the order o f 2.1 per cent. From 1957 to 1970 the indications are fragm entary and very crude. It appears th a t after the census o f 1953 new censuses or counts 5
6 Table 1
The Economic Development o f the Third World since 1900 Changes in the population o f the less-developed countries and o f the world
Non-communist, less-developed countries Asian com m unist countries2 Total less developed countries
Dates Population (in millions) 19001 1920* 1930* 19401 1950 1960 1970 Annual rate of growth (%) 1 9 0 0 -2 0 1 9 2 0 -3 0 1 9 3 0 -4 0 1 9 4 0 -5 0 1 9 5 0 -6 0 1 9 6 0 -7 0
A frica
Asia
America
Total inc. Oceania
World
130 135 155 180 205 254 324 * 1.2 1.5 1.4 2.2 2.5
440 480 540 625 716 888 1,152 * 1.2 1.4 1.4 2.2 2.6
65 90 110 130 162 213 283 * 1.6 1.9 2.3 2.8 2.9
635 710 805 935 1,086 1,358 1,763 0.5 1.3 1.5 1.5 2.3 2.6
445 480 500 525 556 664 801 * 0.4 0.5 0.6 1.8 1.9
1,080 1,190 1,305 1,460 1,642 2,022 2,564 0.5 1.0 1.1 1.2 2.1 2.4
1,645 1,850 2,055 2,275 2,486 2,982 3,632 0.6 1.1 1.0 0.9 1.8 2.0
* T h c m a rg in o f e r ro r is to o g r e a t fo r th e se fig u re s t o b e w o rth in c lu d in g .
R ound figures. See te x t fo r com m ents on usefulness o f these figures. Sources: V arious ed itio n s o f D em ographic Y earbook o f the U nited N ations; World P opulation P rospects as assessed in 1963 (U nited N ations, New Y ork, 1966); J. D urand, World p o p u latio n estim ates, 1 7 5 0 - 2 0 0 0 , World P opulation C onference 1965 (U.N., 1967).
were undertaken in 1958 and in 1964; but, up till now, no results for 1964 have been made public. Since the cultural revolution the data is obviously even more uncertain. The only official figure recently divulged was, first o f all, 750 million, then 700 million, w ithout in either case indicating the year to which the figure relates.2 This naturally leaves us to imagine th at it refers to the 1964 census. Klatt is probably right when he writes th a t nobody inside or outside China is likely to know the true size o f the countrys population; he himself keeps the figure o f 750 million for 1970/71. But in Fields view th e figure should lie somewhere between the lower and the higher hypothesis resulting from the calculations in Airds welldocum ented study, th a t is to say, betw een 788 and 814 million in 1970.3 In m y opinion the figure o f 750 million is probably too low, since if the censuses or counts o f 1953 (582.6 million) and 1958 (646.5 million) are taken as a basis, we obtain an annual rate of growth between those tw o dates o f 2.1 per cent, and, betw een 1958
Population
and 1970 of only 1.2 per cent, which is obviously too low. For this reason a figure for 1970 slightly below the average of A irds two hypotheses, i.e. 800 million (in mid-year), is preferred here. In my view, therefore, instead of the figure o f 801 million inhabitants for the to tal population o f the com m unist less-developed countries in 1970, as accepted by th e United Nations, a figure o f the order o f 840 million would be nearer reality. In th e absense o f any more valid inform ation, the statistical office o f the United Nations, until the M onthly Statistical Bulletin for August 1971, were content to add 10 million each year to the figure of Chinas population in the preceding year. These 10 million represent an annual growth for 1958 o f the order of 1.6 per cent which is rather low, b u t for 1970 they represent no more than 1.3 per cent, which is m ost certainly to o low. Since August 1971, however, th e data has been revised upwards and the statistical office now postulates an annual rate o f growth o f 1.8 per cent, which is probably nearer the truth. Always allowing for margins o f error, there are three phases of population change in the Third World th at can be distinguished between 1900 and 1970. The first, which begins well before 1900 and lasts until about 1920, is characterized by w hat one might call the demographic evolution o f a traditional or ancien regime economy, i.e. by a population movem ent m arked by short-term variations of a positive or negative kind related to harvests or epidemics, and whose average rate o f growth over longer periods was very low (well below 1 per cent per annum ). During this period the factors of demographic change were birth and death rates which were equally high, i.e. crude rates of the order o f 40 per thousand. It should be noted, however, th a t in a large num ber o f regions population change was being, or had already been, influenced by the introduction o f western medical techniques. From 1910 20 until about 1950, a second phase can be
d istin g u ish ed w h ic h m ark ed a break w ith tra d itio n a l d em o g ra p h ic
trends. During these decades appeared the first signs o f that demographic inflation which has recently been so widely and justifiably discussed. Population growth reached about 1.2 p e rc e n t, which, relative to previous change, constituted a m arked acceler ation. It was caused, as already noted, by a heavy fall in m ortality due to medical progress and by a largely unchanged birth rate. Around 1950 the crude m ortality rate was of the order o f 25 per thousand.
Towards the end o f the second world war a third phase began, m arked by a genuine demographic inflation: the rates o f growth of the non-com m unist under-developed countries, which were around 2 per cent in th e first half o f the fifties, exceeded 2.2 per cent in the second half o f the decade, rose to slightly less than 2.5 per cent between 1960 and 1965, and a little over 2.6 per cent betw een 1965 and 1970. As we shall see shortly, it is probable th a t they will rise still higher in the years to come. Population growth in Latin America has been noticeably higher since the very beginning o f the period under consideration. In the early decades the difference was largely due to migrations, while in th e more recent period larger available food resources probably account for the difference. A t the present m om ent the population in th a t continent is growing at the rate o f 2.9 per cent, as against 2.5 per cent for Africa and 2.7 per cent for Asia. In the countries which began their development in the course of the eighteenth and nineteenth centuries, the forty or sixty years preceding th e start o f their industrial developm ent experienced an average rate o f population increase o f less than 0.5 per cent per annum, while during th e first sixty years o f their take o f f the rate was about 0.7 per cent per annum. There is thus a very wide gap between the rate o f demographic growth o f the developed countries at the tim e o f their take o ff and the rates experienced currently by the undeveloped countries. And to th e extent to which such comparisons are valid, it is significant th a t the difference between the population growth o f countries which had their industrial revolution in the eighteenth and nineteenth centuries and the present dem o graphic inflation o f th e less-developed countries is quite wide less than 0.5 per cent on the one hand and 2.6 per cent on the other. B. Recent changes and future prospects We have ju st seen th a t a very m arked acceleration of population growth has recently taken place. However, when the earlier estimates are exam ined it will be seen th a t these tended generally to under-estim ate population growth. To illustrate this situation we shall take the case of the m ost populated o f the non-comm unist, under-developed countries. From 1951 to 1961 the official estimates for India postulated a dem o graphic growth o f 1.4 per cent per annum. However, if we refer to th e census o f 1961, we find th a t the actual progression registered was 2 per cent.4 The calculations o f the Central Statistical Organization
Population
of India, m ade in 1959, anticipated a population o f 480 million in 1966; new forecasts, m ade in 1962, gave a figure o f 492 m illion.5 The same sort o f under-estim ation can be seen in m ost of the less-developed countries. However, it appears th at the situation is slightly different in China (although the evidence is too unreliable for us to be able to pronounce confidently on this aspect o f the problem ) where population growth does n o t seem to have acceler ated. As regards government policy, the following changes occurred; a violent anti-birth policy was succeeded in 1958 (after the Great Leap Forward) by a m arked pro-birth policy: there was propaganda in favour o f births, and there were perhaps even (although the facts are n o t very clear) measures in favour o f large families. With the bad harvests o f the years 1959 62 there was a return to an anti-birth policy which appeared to be still in force at the time o f the cultural revolution. Inform ation since then is not any easier to come by. Suffice it to say and this does n o t indicate any change of official policy th at, according to visitors to the Canton fair,6 the pill was on sale at a reasonable price in Chinese towns and villages and th a t it was m anufactured locally. However, demographic expansion in non-communist, lessdeveloped countries does n o t necessary imply th at no recent progress has been made in reducing fertility. Taking the present age structure o f the Third World population, even a considerable reduction of fertility would n o t retard population growth in the short term. The lack o f even relatively com plete statistics o f recent date certainly does not help one to form an objective appreciation of the situation. Even for crude birth rates regular annual data is available for only a few countries. The absence of recent censuses in m ost under-developed countries makes the calculation o f fertility rates even more hazardous. On the basis o f statistics or estimates o f crude birth rates, it would appear th at a small num ber o f countries have experienced reductions in this rate sufficiently im portant to establish the probability o f a measur able reduction in fertility. But it has taken place only in small countries, such as Taiwan, Puerto Rico, Hong Kong, Mauritius, Trinidad and Tobago, etc, which have already benefited from some economic and social developm ent and have received a large volume o f aid. They have also undertaken birth-control campaigns o f a thoroughness which would be o u t of the question in the larger countries. It is in these larger countries, o f course, th at the problem is really crucial, and it does n o t appear from an analysis o f the crude
ial com direitos autor
10
birth rates th a t any significant progress has been achieved, since any plans for birth control have usually encountered many obstacles, particularly social ones. The difficulties m et w ith by campaigns to regulate births can be better illustrated by reports o f actual experience than by global statistics. West Bengal, where the Indian authorities had made a special effort to prom ote a programme o f fitting the intra-uterine device, will be taken as an example. According to the latest inform ation this programme has largely failed. There has been a steady fall in the num ber o f wom en accepting insertion. The num ber fell from 121,000 between O ctober 1965 and March 1966 to 51,000 between April 1966 and Septem ber 1966, while it was only 72,000 for the whole of the following year, i.e. from October 1966 to September 1967.7 This failure was mainly due to the fact th at the effectiveness of the device was discredited through the personnel responsible n o t being sufficiently trained to deal with the many attendant medical problems. Even the vasectomy programme experi enced a certain degree o f failure in this region, because men undergoing the operation were not told th a t sterilization would not be effective until about three m onths afterwards. This was un doubtedly the cause o f m any births reported subsequently and it tended to throw d o ubt on w hat is otherwise a m ost reliable m ethod. These failures, or partial failures, are echoed from m any directions and show us how urgent it is to speed up the search for a contraceptive m ethod which is geared more closely to the economic, medical and, above all, social conditions o f the under-developed countries. It is, in addition, m ost desirable th a t studies of the psychological aspects should be set in m otion. Only when propa ganda techniques are well attuned to local m entality will the introduction of family planning have a real chance of being effective. The pessimism expressed in earlier editions o f this book has been confirm ed by the m ost recent population forecasts which are given below. The figures anticipated for the less-developed countries have had to be revised upwards once more. If we confine ourselves to the short-term prospects we have to remember th at even if these birth-control campaigns are successful they can influence population levels only very slightly, since growth in the foreseeable future depends much more on the age structure of a population than on changes in its fertility. As Henry and Pressat pointed out when speaking in 1956s o f the demographic outlook for the Third World, th e age pyramid has such a large base th at if the
ial com
Population
11
population is n o t to increase there m ust be either an extrem ely low fertility rate or an extrem ely high death rate. This remark is even more pertinent now, when the changes which have taken place since the middle o f the fifties have further enlarged the base of the Third W orlds age pyram id. Allowing for this factor and assuming th a t the surpluses o f western agriculture (and o f N orth America in particular) would be capable of fending o ff for at least ten to fifteen years the risks o f famine in the less-developed countries, we can expect with reasonable certainty th a t the rate of population growth o f the Third World will increase in the course o f the next ten years, probably up to 1980.9 For this period we can assume as probable an annual rate of growth o f around 2.8 to 2.9 per cent for the non-com m unist, under-developed countries, at least. Table 2 Projection o f population fo r less-developed countries and the world 1 9 1 0 -2 0 0 0
1970 Population millions M edium variant Non-comm unist less-developed countries Africa Asia America Com munist less-developed countries 1980 Population * millions 1990 Population * millions 2000 Population * millions
1763 324 1152 283
2329 430 1517 377
2.8 2.9 2.8 2.9
3036 5 79 1950 500
2.7 3.0 2.5 2.8
3828 770 2397 652
2.3 2.9 2.1 2.7
801
943 3272 445 7
1.6 2.5 2.1
1091 4127 5438
1.5 2.4 2.0
1231 5059 6494
1.2 2.1 1.8
Total less-developed countries 2564 World Non-comm unist less-developed countries: variants L ow variant M edium variant High variant Constant fertility variant 3632
1763 1763 1763 1763
2272 2329 2375 2370
2.6 2.8 3.0 3.0
2863 3036 3213 3280
2.3 2.7 3.1 3.3
3475 3828 4243 4655
2.0 2.3 2.8 3.6
A n n u al grow th rate com pared to th e previous decade. Sources: Derived from U.N. projections.
12
It will be noticed th a t the rates I have quoted relate to the non-comm unist countries only. For the com m unist countries it would be rash to make any forecast in view of the uncertainty o f the past and present data. I have, however, reproduced in Table 2, for inform ation only, the figures compiled by the United Nations experts which summarize th e population forecasts made by them for the period 1970 2000. It should be noted th a t the future population growth rates I have retained (2.8 and 2.9 per cent) are a little above those derived from the m edium variant o f the United Nations projection (2.8 per cent). If we accept this medium variant the countries now reckoned as less-developed will account for 78.3 per cent o f to tal world population by the end o f the century. But if we postulate a constant fertility rate the figure rises to 82 per cent. In 1930 the under-developed countries accounted for only 63 per cent o f world population10 and in 1970 70.5 per cent (71.5 per cent if the more realistic figure for China is used). It is only to o obvious, therefore, th at population growth has a burdensom e part to play in the Third World problem s to be discussed in the following chapters. Before this chapter closes attention m ust also be drawn to the recent emergence in certain quarters in the relatively under-populated countries o f a sentim ent hostile to the slowing down of demographic growth. This hostility usually results from a regrettable confusion between the pace o f growth and population density. The real danger o f demographic growth lies in too rapid an increase and n o t at least n o t in every case, nor even in the short term in the absolute level o f population which th at growth creates. It is true th a t for some under-developed countries it is possible and even probable th a t their present population density is n o t sufficient to perm it a higher level o f developm ent.11 But in seeking to achieve a higher density too quickly the very possibilities o f developm ent are themselves jeopardized.
2
INTRODUCTION
The im portance o f agriculture in the econom y o f the under developed countries as well as in all traditional economies is such th a t a larger am ount o f space has been devoted to its study here than to any other topic. The crucial p art played by agriculture in the economic life o f the under-developed countries is revealed in two ways. Firstly, at the present time the sector occupies between 65 and 75 per cent o f the total active population, provides about 35 per cent of the Gross Domestic Product and accounts for about 65 per cent of the foreign trade o f these countries, excluding trade in oil.1 Secondly, the im portance o f agriculture, spotlighted by the per centages quoted above, decrees th a t it m ust play a major role in the process o f take o ff. The latter would be inconceivable w ithout the participation o f such a substantial part o f the population, who are, of course, also consumers. The study o f the industrial revolution2 seems to indicate th at it was this sector which gave the crucial im petus th a t started the western countries off on th e road to industrialization. It is obvious, though, th a t structural conditions have since then been subject to profound changes which, on th e one hand, make the take o f f more difficult (for population growth is, after all, the special affliction of under-developed countries), while, on the other hand, allowing a degree o f disengagement from agricultural determinism which, in the earliest phases o f the industrial revolution, was almost complete. But the disengagement is in reality only partial and the m om entum of demographic growth o f a rural population alone makes a rise in agricultural productivity essential for growth in the econom y as a whole. The dom inance o f agriculture which was param ount in w e ste rn c o u n tr ie s, will, in th e course o f th is study, be shown to exist also in the under-developed countries, especially when we make comparisons betw een levels o f agricultural productivity and growth of m anufacturing industry. Since in the last resort productivity is the best indicator o f progress in agriculture, the main emphasis has been placed on this aspect. In the absence o f com parable indices on an international level which would allow particular changes in one co untrys productivity to be com pared w ith those o f others, it has been found necessary to 13
14
calculate3 a special index o f this kind for the tw enty-four under developed countries which represent 80 per cent of the Third W orlds population. The index is worked o u t for seven dates during the period 1909/13 to 1968/72, whereas the index num bers o f produc tion constructed by F.A.O. go back no further than 1934/38, and then only for a small num ber o f countries. The core of this subject will be the study of productivity, b u t I propose to preface it by a short analysis of the growth o f aggregate agricultural o u tp u t so as to com plete my assessment of agriculture considered as a vital segment of the econom y as a whole. A. Agricultural production For statistical reasons it is only possible to deal in detail w ith the changes in agricultural production after 1934/38 (see section II below). However, a relatively com plete picture o f changes occurring before th a t time can be sketched o u t briefly. For th at period, even more than for recent years, it is preferable to treat subsistence cultivation and export cultivation separately. We have to remember th a t colonization usually endowed agriculture with a dualistic character: on the one hand there were the export crops generally, b u t n o t invariably, dom inated b y a plantation system employing advanced technology. On the other hand there were the subsistence crops, often excluded from the best land and alm ost always cultivated by m ethods which were both traditional and primitive.
I C H A N G E S IN P R O D U C T I O N UP T O 1 9 3 4 / 3 8
Let us see, first o f all, w hat happened in the field of export crops in which there is an abundance o f statistical data o f a fairly reliable nature. (a) Export crops It should be rem em bered th at since the sixteenth century differences in climate have m ade it possible to produce certain specialized South American crops at costs so low and profits so high as to bear the high transport charges. The fall in transport costs, the growing dom ination o f Asia and Africa by colonial powers, and the industrial revolution, together m ade it possible to enlarge this system o f exploitation considerably. The nineteenth century (especially the second half) saw a great expansion in the production and export o f tropical commodities. Each year European m arkets absorbed an increasing quantity o f sugar, coffee, tea and cocoa, as well as the more
S a c h c o b a n ci
Agriculture
15
traditional spices and dyes. In the tw entieth century the expansion even accelerated with other exotic goods, such as citrus fruits, bananas, rubber, oil seeds, etc., being added. Table 3 shows the increase in production o f nine products, or groups o f products, representing the m ost im portant export crops.4 For comparative purposes the 1970 figures have been included.
Table 3
World output o f the principal tropical commodities from 1880 to 1970 (in thousands o f tons: annual averages for quinquennia centring on dates in table, except 1970 where the average is for 196971)
1880 1900 300 102 970 53 1,200 1,220
1910 1,8004 227 1,090 87 1,7704 1,560 2,700 6,320 360
1936 8,100 560 2,110 790 3,280 1,960 12,060 14,200 450
1970 28,000 1,430 4,420 2,960 7,230 3,290 22,400 41,100 1,220
Bananass Cocoa Coffee Natural rubber C otton fib re1 Ju te 6 Oil seeds2 Raw sugar cane7 Tea3
30 60 550 11 950 600
1,850 175
3,340 290
1E xcluding U.S. and U.S.S.R. 2G round nuts, copra, palm and palm oil 3 Excluding U.S.S.R. 41911 5B efore 1910: w orld ex p o rts 6A nd allied fibres (K enaf and C ongo ju te) C entrifugal sugar N o te: Figures rounded up in u nits o f ten to the nearest 1000 units. Sources: 1880 A u th o rs estim ates based in general on various sources, b u t m ainly the retrospective sections o f the statistical yearb o o k s o f France, and M. G. M ulhall, D ictionary o f Statistics (L ondon, 1899). 1900 to 1936 M ainly A nnuaire international d e statistique agricole o f the International In stitu te of A griculture, R om e, several years. 1970 F.A .O ., and U.S. D ept, o f A griculture.
As can be seen, production of agricultural products for export grew very greatly before 1936. Weighting the volume of production by taking average export prices for 1926/295 we obtain annual rates of grow th (based on five-year averages for the periods under consideration) for the nine export crops listed in Table 3 as follows: 18801900 19001910 19101936 19361970 2.6 4.2 3.6 2.7 per per per per cent cent cent cent
16
According to F.A.O. figures the annual rate o f growth o f total agricultural production in the period 19361970 was 2.1 per cent, which confirms, if there is indeed any need to do so, th a t the products selected are representative. (b) Subsistence agriculture Inform ation on the o u tput o f subsistence agriculture is much more fragm entary. On the basis o f existing inform ation it can be estim ated that, between 1900 and the eve o f the second world war, food production grew more slowly than population in these regions. But it is difficult to be more precise about the gap between the pace of growth. All th at can be said is that it is n o t a large one probably less than 0.5 per cent per annum and th a t in 1934/38 the countries of the Third World still had a positive trade balance o f cereals, i.e. as a whole, they exported m ore cereals on average each year than they im ported. And this holds good even if we exclude Argentina, which in this respect, as in some others, is n o t really a part o f the Third World. As we shall see later, in about 1936 this surplus (excluding Argentina and China) was o f the order o f 11 million tons o f cereals, th at is about 7 per cent o f the production o f such comm odities by these countries. A nother point to be noticed is the more favourable growth of production in Latin America (and probably also in Africa) relative to th at in Asia, a difference which is n o t wholly invalidated by the margin of error present in these figures. We shall now turn to changes in more recent years.
II C H A N G E S IN A G R I C U L T U R A L P R O D U C T I O N S I N C E 1 9 3 4 / 3 8
F.A.O. (Food and Agriculture Organization o f the United Nations) calculate index num bers o f agricultural production6 for the principal under-developed countries (with, until very recently, the im portant omission o f China, mainly because o f the absence of com plete statistics). The statistical services o f this organization also calculate index num bers for the main geographical regions.7 However, the F.A.O. index goes back no further than 1934/38. (a) Changes in agricultural production b y main geographical regions In th e tables below, in order to reduce the effect o f fluctuations in yields, I have calculated five-yearly averages o f the F.A.O. index num bers o f agricultural production, as well as the annual rate of
A i X a
Agriculture
17
variation. As is indicated in Table 4 there are tw o series o f index num bers, one relating to production o f food and the other to total agricultural production. F.A.O. also calculates these same indexes per capita. Table 4 Agricultural production indices in less-developed countries 1 (196165 = 100)
Food products only Total 1 9 3 4 -3 8 1 9 4 8 -5 2 1 9 5 3 -5 7 19 5 8 -6 2 1 9 6 3 -6 7 1 9 6 8 -7 2 58 66 79 92 105 121 Per capita 99 90 96 99 100 101 A ll agricultural products Total 59 66 78 92 105 121 Per capita 101 89 96 100 99 101
1E xcluding co m m u n ist countries. Sources: Derived from P roduction Y earbook 1972 (F.A .O ., R om e, 1973); The S ta te o f F o o d and A griculture 1970 (F.A .O ., Rom e, 1970), and M o n th ly B u lletin o f A gricultural E conom ics and S ta tistics (F.A .O ., various issues). E xcept 193438: a u th o rs estim ates on basis o f F.A .O . regional data.
If we consider the entire period 1934/381968/72 it appears that the annual rate o f growth o f total agricultural production is around 2 per cent. The rate reaches or exceeds 2 per cent for Africa, Latin America and th e Middle East, while it is about 1.9 per cent for Asian countries.8 Grow th in the African countries and the Middle East was slightly more rapid for non-food products, while the opposite was true o f Latin America and the Far East. All the same growth was not uniform throughout th e period. It was rather m oderate from 1934/38 to 1948/52, more rapid during the following decade, while in the decade 1960/70 a slackening o f overall growth can be detected. If the periods o f shorter duration are considered at a regional level, changes in the growth o f agricultural production seem rather more homogeneous: slow growth from 1934/38 to 1948/52 (near 1.0 per cent), acceleration from 1948/52 to 1958/62 (near 3.5 per cent), then from 1958/62 to 1968/72 a slackening o f the rate (about 2.5 per cent). The high rate o f population grow th is responsible for absorbing nearly all this increm ent o f production and even results in a lowering of per capita production. For Latin America this is relatively im portant since the figure fell by 0.3 per cent per annum from
18 Table 5
The Economic Development o f the Third World since 1900 Annual rate o f growth o f total and per capita agricultural production (per cent)
1934/38 to 1948/52 1948/52 to 1958/62 1958/62 to 1968/72 1934/38 to 1968/72
TOTAL Africa Far East L atin America Middle East Less-developed countries PER CAPITA Africa Far East Latin America Middle East Less-developed countries
L ess th a n 0.05% . Sources: See T able 4.
1.7 0.3 1.2 1.0 0.9 - 0 .2 - 1 .0 - 1 .0 * - 0 .9
3.1 3.3 3.5 4.0 3.4 0.8 1.2 0.7 1.4 1.1
2.5 2.5 2.5 3.1 2.7 * - 0 .2 - 0 .4 0.3 0.1
2.4 1.9 2.2 2.5 2.1 0.1 - 0 .1 - 0 .3 0.5 *
1934/38 to 1968/72. Only in the Middle East was a regression really avoided and slight progress registered. It is possible, of course, that these figures contain a margin of error, b u t the error would n o t be big enough to invalidate the changes described. The large num ber of countries involved would tend to reduce the margin o f error contained in each o f the indices taken individually.9 We shall now compare these rates with those o f the western countries, firstly during their take o f f , and secondly at the present time. Because agricultural production is faced w ith structural rigidities more obdurate than those experienced by industry,10 rates o f growth in this sector were never very high. Thus, in England during th e eighteenth century agricultural o u tp u t rose at an average of 1 per cent per annum. The comparable figure for France was 0.6 per cent in the eighteenth century and 0.9 per cent in the nineteenth century, while in Belgium the rate was 1 per cent during the second half o f th e nineteenth century. The rate was higher in Russia and especially in the United States where it reached 2.5 per cent between 1870 and 1910.1 1 In these two cases, o f course, growth was due to th e continuous colonization o f new lands made possible by the influx o f manpower. Thus, the rates o f growth achieved by agriculture in the under-developed countries appear favourable com pared to the rates o f western countries during their take o f f period. But th e real difference betw een the perform ance o f
A i X a
Agriculture
19
the tw o is caused by the growth of population which, as we have seen, was m uch lower in the western countries than in the under-developed countries. As for the present rate o f growth of agricultural production in all the developed countries taken together, it is higher than th at achieved in the course o f the nineteenth century, as well as th a t in the under-developed countries in recent years. Thus, for the period 1948/52 to 1968/72 the annual rate of growth of to tal agricultural production was 2.9 per cent in Western Europe, 1.7 per cent in N orth America and 3.9 per cent in Eastern Europe and the U.S.S.R. Therefore in this essential sector o f the Third World econom y the diagnosis o f the level o f production m ust be an unfavourable one: the speed of growth has been to o slow, and generally slower than the growth o f population. This negative diagnosis will be reinforced by the study o f productivity; because, o f course, if actual agricultural growth could have been accompanied by a growth in productivity, th e global diagnosis would have been different. In fact, as we have seen, rates in absolute terms are n o t so low. If we look at global rates rather than per capita rates, they are slightly above those achieved by the developed countries at the m om ent o f their take o f f . Before tackling th e problem of productivity, however, we shall briefly exam ine th e growth o f agricultural o u tp u t in some countries. (b) Changes in agricultural production by countries It is once again necessary to draw the reader's attention to the fragm entary nature o f m any statistics relating to the under-developed countries, especially w ith regard to agriculture. A nyone who is aware o f the difficulties experienced by statisticians in this field, even in the developed countries with their relatively elaborate statistical organization, will accept this paucity o f m aterial as inevitable. All we can do here is to fill in for some countries the diagnosis already form ulated for the larger geographical regions. In Table 6 five-year averages an d th e an n u al rates o f v a riation h ave been c a lc u la ted for three periods. Because F.A.O. has neither converted the pre-1952 figures to the form er 1952/56 base nor to the new 1961/65 base for separate countries, I have been forced to m ake these calculations myself, which introduces an additional margin of error since these index num bers are in round figures.12 For the annual rate of growth of agricultural production over a recent period for all countries for which F.A.O. calculates such an index readers should refer to the synoptic table in th e Appendix-
Copyrighted
20
Table 6 Changes and annual rates o f variation in the index o f total agricultural production
Index 1961/65 = 100 1934 1938 AFRICA Algeria Ethiopia Morocco Tunisia LATIN AMERICA Argentina Brazil Chile Colombia Cuba Guatemala Mexico Panama Peru Uruguay Venezuela ASIA Burma Ceylon China (Taiwan) India Pakistan Indonesia Malaysia (West) Philippines Thailand MIDDLE EAST Egypt Iraq Iran Syria Turkey 1948 1952 1968 1972 A nnual rate o f variation 1934/38 1948/52 1948/52 19 6 8 /72l
105
61 71 74 49 61 43 56
99 783 74 73 75 58 76 71 88 563 51 623 70 91 623
107 124 140 115 109 129 114 133 111 131 124 147 115 98 139 108 116 156 117 128 121 149 123 130
- 0 .4
2.0 0.2 0.1 1.2 1.6 3.6 3.2
0.4 3.1 3.2 2.3 1.9 4.1 2.0 3.2 1.2 5.8 4.5 5.9 2.5 0.4 5.5 1.8 2.8 4.8 2.7 2.5 2.8 3.5 3.5 5.5
33
3.2
53 75
2.0 1.4
86 50 56 68 71 74 54 51 36
75 66 61 69 75 762 74 62 44
-1 .0 2.0 0.6 0.1 0.4 0.1 2.3 1.5 3.0
57

41
66 863 743 69s 55
124 143 131 102 126
1.0

2.1
3.2 3.4 3.9 2.6 4.2
1 F o r countries w here th ere are no figures for 1948/5*2 the annual rate o f grow th is calculated for th e period 195 3 /5 7 to 1968/72. 21 9 5 1 /5 3 3 195 3/5 7 Sources: Derived from various issues o f P roduction Y ea rb o o k and M o n th ly B ulletin o f A gricultural E conom ics and Sta tistics (F.A .O ., R om e).
Copyrighted material
Agriculture
21
So far as Africa is concerned long-term figures are available for only a few countries, none o f which is p art o f Black Africa. In the Maghreb growth is very unequal: in Algeria regression is followed by stagnation; in Tunisia progress is lim ited and in Morocco expansion is relatively fast. The rate for th e whole o f Latin America conceals a very wide variation betw een individual countries. Countries registering very little growth in o u tp u t (less than 2 per cent) include Argentina, Cuba, Uruguay and Chile. On the other hand production in Brazil, Venezuela, Guatemala, Panama and Mexico has risen fast enough to overtake the increase in population. There is much less differ entiation in Asia. Amongst the countries experiencing a favourable change are Thailand, Sri Lanka, the Philippines, Malaysia and Taiwan. But countries w ith an unfavourable change include three of the four largest in th e continent: India, Pakistan and Indonesia. As we have seen, the F.A.O. does n o t calculate long-term indices for the fourth large country, China, b u t on the basis o f the data collected for the remaining sections o f this chapter, we m ay conclude th a t any change occurring has been more favourable than in m ost o f the other Asian countries, particularly India, even according to western estim ates. As far as the Middle East is concerned, for the long-term index (1934/381967/71) we possess figures only for E gypt13 and Turkey. The latter has during this period experienced a rise in agricultural production (3.3 per cent) which is noticeably faster than th a t o f population. On th e other hand, E gypts rate of about 2 per cent has n o t kept pace w ith population growth. O ther countries, apart from Syria, have experienced a relatively high and relatively hom ogeneous increase in the short term (about 3.5 per cent). B. Agricultural productivity As stated at th e start o f this ch a p ter, th e agricultural sector is of such im portance th a t an index which would enable us to trace the growth of agricultural productivity in the various countries o f the Third World would be well w orth while constructing. In another study devoted to econom ic developm ent in the nineteenth cen tu ry 14 I was able to calculate such an index o f agricultural productivity for twelve countries which are nowadays considered developed. An index for under-developed countries constructed by means o f a similar m ethod would have a double objective:
Copyrighted
22
The Economic Developm ent o f the Third World since 1900 1. To measure and com pare at an international level changes in agricultural productivity in the under-developed countries. 2. To compare the agricultural productivity of under-developed countries with th a t o f developed countries before and during their take o ff period.
I. M E T H O D O F C A L C U L A T I N G T URAL PRODUCTIVITY
THE
INDEX
OF
AGRICUL
One im portant prelim inary rem ark m ust be made concerning the exiguous nature o f the basic figures used in this study. I have already drawn attention to the substantial margins o f error in Third World agricultural statistics. Within the limits of possibility some of the more obviously erroneous figures have been corrected by means of accepted statistical techniques, while some series are so fragm entary th at they have been rejected altogether. But however much such statistical caution may have reduced the margins o f error, they still remain considerable. We m ust accept the fact th a t errors are impossible to eliminate entirely, for if all quantitative material which included any margin o f error were to be rejected this would lead to the rejection o f almost all other data used in economic analysis, especially in such m odern concepts as national product or income, capital ou tp u t ratio, etc. The index o f agricultural productivity as it has been calculated here is in fact an estim ate o f the net agricultural production by male labour em ployed in agriculture expressed in d irect calories.15 The exclusion o f women from the population em ployed in agriculture is justified because the criteria used in different countries to determ ine their em ploym ent vary too widely. The stages adopted in calculating the index will very briefly be explained: (a) Total production o f gross calories 1. Vegetable products The following crops have been taken into consideration: all cereals (including millet and sorghum), soya beans, manioc, potatoes (including sweet potatoes) and all dried vegetables. Calorific values have been obtained for each o f these products and applied to output. An allowance has been made for the difference in calorific content between rice and other cereals so as to establish an average appropriate to each country as a function o f the respective im portance of the various cereals.
Copyrighted
Agriculture
23
2. Meat Statistics for m eat production are practically non-existent for m ost under-developed countries, so, in order to make the series hom o geneous, production had to be estim ated. This has been done in the following way: (a) The annual num ber o f animals slaughtered for consum ption at home and for sale has been estim ated on the basis o f total livestock existing at different dates. The coefficients used to determ ine the num ber of beasts slaughtered have been based on figures for countries with a similar or related structure for which valid statistics were available. I have also had to allow for changes over time, for as agricultural technique improves animals are slaughtered at a younger age. This means the coefficients m ust rise. I have, o f course, used different coefficients for each kind o f animal, country or group o f countries, as well as for different periods. The averages for the different livestock species at the end o f the period were as follows: oxen and cows 15 per cent; buffaloes 10 per cent; sheep and goats 22 per cent; pigs 70 per cent. (b) Total weight for each species has been estim ated on the basis of figures available for countries with a related structure (statistics in this field are actually more abundant). Here allowance has also been made for an increase in the average weight o f animals. (c) 2,300 calories per kilo have been postulated for all kinds of m eat except pig meat, for which the figure o f 3,200 calories has been tak en .16 However as these figures represent derived calories, they had to be converted back into direct or initial calories. The choice o f coefficients used (since, in fact, they differ widely between one kind o f livestock and another) has posed problems o f a technical and methodological nature. Im provements in animal husbandry lead, in fact, to m odification of these coefficients for, on the one hand (as pointed o u t above) the tendency to slaughter animals at an earlier age entails a reduction in the calories absorbed by animal m aintenance, while, on the other, the improvement o f breeds and techniques o f animal husbandry lead to an increase in the energy yield (i.e. in the num ber o f m eat calories produced by a given quantity o f forage and other feeding stuffs absorbed by the animal). Consequently as agricultural technique improves th e coefficient in question is lowered. The problem of m ethod which it poses is th a t o f a choice between a variable or a
24
fixed coefficient. A variable coefficient would be justified on the purely form al plane b u t it would lead to a considerable lowering o f the effects o f improvements, narrowly speaking, since for a given production o f m eat fewer and fewer vegetable calories are needed. But since th e main object of this study is to measure changes and differences in productivity it seems as if the least inaccurate solution is to use a fixed coefficient. As for the technical* problem, it is simply how to determ ine the average coefficients, since no w ork has been done in this field. Briefly, the following coefficients have been adopted here: oxen and cows 9 (nine vegetable calories to produce one m eat calorie); buffaloes 12; sheep and goats 24; pigs 5 .17 (b) Total production o f net calories The results o f the calculations presented above provide an estim ate of the production of gross calories, i.e. including seed.18 These figures have been changed into net calories (excluding seed) by using for each period and country considered a different coefficient of correction. These coefficients have been estim ated from yields per hectare of the various cereals under consideration and o f the corresponding seed requirem ents, th a t is the seed-harvest ratio for the main cereal in each country. These coefficients have had to be adjusted for countries where manioc forms a large proportion of agricultural production because seed for this crop does n o t signifi cantly reduce th e harvest available. (c) Male labour employed in agriculture The total production of net calories obtained has been divided by the num ber of male workers em ployed in agriculture: as already stated, women have been excluded because there are im portant variations in the criteria used to determ ine their participation. Insofar as valid statistics from recent censuses are available, figures for the total num ber o f male agricultural workers have been used. In all other cases this figure was estim ated in the following way: on the basis of total male population the male working population was derived by using global activity rates extrapolated either from the same countries for dates where such figures exist, or from the activity rates in other countries at a similar level o f development. From this male working population the num ber employed in agriculture was estim ated from inform ation about the relative share of this sector. In m any cases I have calculated these shares myself. I
Agriculture
25
have then gone on to extrapolate on the basis o f earlier known data and general changes in this field derived from inform ation about other countries. (d) Eliminating non-subsistence crops, particularly plantation crops The fundam ental difference between subsistence and plantation agriculture makes it necessary to exclude the latter when calculating our index o f agricultural productivity.19 Plantation agriculture should in fact be considered as totally outside the agricultural economy of most of the under-developed countries properly defined, partly because o f the techniques em ployed and partly because o f the nature of land-ownership. Excluding this sector o f agriculture does n o t pose a problem from the production point o f view, b u t it is m uch harder to separate em ploym ent in the two sectors. For all practical purposes statistics do not exist, so I have been forced yet again to make in each case an estim ate based on the two following approaches, either together or separately. (1) By using the various norms o f labour required per hectare for each crop. (2) By making an estim ate o f labour based on the value of production, on the average coefficient which labour represents in the cost price and on th e average level o f agricultural wages. Both these m ethods o f estimating, especially th e first, m ust be examined in a little more detail. 1. Norms of labour required per hectare Very little w ork has been done in this field.20 The only general study th at I know, entitled Normes de main-d'"oeuvre pour les travaux agricoles au Congo beige. 21 relates to the form er Belgian Congo and refers m ostly to tropical crops. On the whole we may accept these norms as valid if we treat them as general averages. A system atic search for norm s in o th er periods or territories in studies dealing either w ith the products or the countries we are concerned w ith has resulted in some supplem entary m aterial.22 It should be noted, however, th a t th e figures calculated in the I.N.E.A.C. study on the Belgian Congo could not be used as they stood, because average norm s had to be com puted from figures relating to investm ent period, upkeep during the period when there was no yield, as well as the period o f yield. 2. Estimate o f labours share in the value o f production From such fragm entary m aterial as is available it is possible to establish th a t the average share of labour in the selling price of
26
plantation crops is about 42 per cent.23 This is, of course, only a very rough figure, b u t th e margin o f error has, nevertheless, been kept to a minimum. From the ratio of the level of agricultural wages to th e value o f production o f the crops under consideration, I have been able to form an estim ate o f labour requirements. (e) Choice o f countries: sources The wide scope o f this study2 4 has rendered it impossible to make calculations for all th e under-developed territories. It will be remembered how the fragm enting o f form er colonial empires has resulted in th e creation o f some 170 economic and political entities. In addition, satisfactory statistical inform ation on agriculture for m ore than a small num ber o f these separate entities simply does not exist. Consequently, I have been forced to use as my base here only the tw enty-four countries which, it will be recalled, were selected according to three criteria, b u t which nonetheless represent 80 per cent of the population o f the Third World. The fundam ental figures of agricultural production are mainly based on inform ation collected by F.A.O., either in the Production Yearbook or (for the m ost recent material) in Monthly Bulletins o f Agricultural Economics and Statistics. For th e period before 1946 the Year Books o f the International Institute o f Agriculture were used. Nevertheless for certain countries and certain products2 5 additional research in the statistical year books o f th e countries, or in com m odity studies, has been necessary.
II. PRODUCTIVITY THRESHOLDS
Before looking at the results o f calculating our agricultural pro ductivity index, I have attem pted to give this index a slightly m ore realistic character. In addition to the inform ation the index provides about the relative levels o f individual countries, or o f different economies, or about th e changes which have taken place betw een different periods, we w ant also to be able to say th a t the agriculture o f these countries has reached definite stages o f developm ent. With this object in view I have calculated a series o f productivity thresholds expressed in th e same units which are used in the main index. The reader is rem inded th a t the index o f productivity is the average annual production o f a male agricultural w orker expressed in millions of net direct calories. A pproxim ately 0.9 units, or 900,000 direct calories, are needed annually to feed a man. This figure is based on a m inimum o f 2,100 calories per day2 6 and allows on the one hand for an under-estim ate,
Agriculture
27
calculated as about 10 per cent, because certain products (milk, vegetables, oils and fruit) have been excluded and, on the other hand, for the fact th a t animal calories are here expressed as direct calories. On th e basis o f a working population representing 40 per cent of total population2 7 and of a male working population representing 28 per cent o f to tal population, we arrive at a figure o f about 3 units as the minimum physiological productivity required to support the entire population, assuming the total absence o f other activities. By postulating a daily consum ption o f the order o f 3,200 calories (which should be considered an optim um figure) and also in the total absence o f other sectors the minimum productivity reaches 5.4 units (allowing for the rise in animal calories inherent in such a change). These levels are obviously purely theoretical, if only because they assume th e to tal absence o f other activities and because they om it losses and, above all, fluctuations in yields. It leads us on to calculate tw o, more realistic, thresholds o f productivity in which these factors are accounted for. The first, which can be called the physiological threshold, is also based on a daily consum ption o f 2,100 calories per head, which, as we have seen, implies a productivity o f 0.9 units in the absence of other sectors. By postulating th at th e non-agricultural sectors represent 15 per cent o f the working population28 and th a t various losses (deterioration, waste through animal pests, milling, etc.) represent 10 per cent of production,2 9 the physiological minimum am ounts to 3.8 units. Therefore, at this level productivity can be considered sufficient to cover food requirem ents w ithout noticeable quantitative scarcities.30 But this threshold does not allow for a factor too often neglected in the diagnosis o f the changes in agricultural econom y: wide fluctuations in yields. It is for this reason th a t I have thought it essential to define and calculate a second threshold which would give the average level at which agriculture can be considered to have surm ounted the
dan gerlin e b e lo w w h ic h fa m in e s are still p o ssib le . It is tru e d ia t in
determ ining this threshold it would be necessary to take account of several factors which will be ignored here, especially the im portance o f animal husbandry7 as a buffer against the effects o f fluctuations in agricultural yields.3 1 In practice animal husbandry regulates the quantity o f calories available for hum an consum ption: during periods o f good harvests the num ber o f livestock is increased because o f the surplus cereals available, while in years when harvests are insufficient or less good, animal slaughter allows food supplies to be increased
28
The Economic Developm ent o f the Third World since 1900
n o t only by providing m eat, b u t also by reducing the demand for cereals to feed livestock. However, it m ust be adm itted th a t in general this mechanism is of little significance in under-developed countries because the low productivity o f agriculture reduces the consum ption o f derived calories. Consequently the second thresh old, which I shall call productivity potentially free from the risk of fam ine, will take account only o f fluctuations in yields. Table 7 Variations in yields (100 kg per hectare), 1 9 4 7 -1 9 6 2
Low est annual yield Crop and country RICE Ceylon India1 Philippines Madagascar WHEAT Argentina Brazil Egypt Morocco MANIOC Brazil Indonesia3 Peru Zaire2 Average yields (1 9 4 7 -6 2 ) Variation fro m the mean
Figure
15.4 12.8 11.7 14.2 12.5 7.4 21.4 6.3 13.0 8.2 12.5 10.6
12.4 10.1 10.8 12.6 7.8 4.1 15.3 3.9
20% 21% 8% 11% 76% 95% 29% 38%
12.4 6.7 9.9 8.4
5% 18% 21% 21%
1195 2 62 2village o u tp u t 3 194 8 62 Sources: These averages have been calculated o n th e basis o f m aterial provided b y th e P roduction Y earbook for th e years 19 4 9 63 (F .A .O ., Rom e).
Table 7 shows variations in the yields of three food products which are im portant to the under-developed countries. It will be seen th a t w ithin the period, according to produce and even country, there is a striking gap between average yields and lowest yields. The gap varies on average from about 20 per cent for rice and manioc to as much as 40 per cent for wheat. Bearing in mind th a t in under developed countries both rice and m anioc are of equal or greater im portance than w heat we can roughly com pute the probable
Agriculture
29
average gap between average agricultural production and a very bad harvest as 2530 per cent. This rate applies to gross production (including seed): the gap is even wider for net production, since it is obvious th at the am ount o f seed rem ains the same no m atter what the variation o f yield w ithin the country. Allowing for this factor, the gap can be taken as 30 per cent, which gives us a figure of about 4.9 units for the productivity threshold free from famine risk. But once more this is only a theoretical figure, for it is clear th a t in the hypothetical case of a country where agriculture accounted for only a small fraction of total em ploym ent it could be m uch lower w ithout being followed by a famine. Before examining the productivity index itself the different thresholds can be recapitulated as follows: 0.9 units: 3.0 units: Food requirem ents per head on the basis o f 2,100 calories per day. Threshold o f m inimum physiological productivity for the whole population assuming the absence o f other sectors. Threshold of physiological productivity. Threshold o f productivity potentially free from famine risk.
3.8 units: 4.9 units:
I ll I N D E X O F A G R I C U L T U R A L P R O D U C T I V I T Y : R E S U L T S AND COMMENTS
We shall briefly examine the results of the com putations made according to the m ethods described above under the following three headings: (a) Average levels (1960/641968/72) (b) Changes between 1909/13 and 1968/72 (c) Comparisons w ith the developed countries, now and before their take o ff.
(a) A v e r a g e lev els ( 1 9 6 0 / 6 4 1 9 6 8 / 1 2 )
Although the indexes have been calculated for five-yearly periods, it seemed best, nevertheless, to use the average for the two recent periods when comparing levels of production, in order to eliminate as far as possible the effects o f fluctuations due to variations in yields. One preliminary conclusion springs to the eye immediately. This is the m arked divergence between productivity in the Latin American countries and in th e countries o f Asia and Africa. The average for the form er is around 10, for the latter 5, while the average for the Middle
30
Table 8 Level o f agricultural productivity: averages 1960/641968/72

C O U N TR Y AFRICA Ghana Kenya Madagascar Morocco Nigeria Tunisia Zaire Weighted average 1 excluding Madagascar AMERICA Argentina3 Brazil Chile Cuba Mexico Peru Venezuela Weighted average 1 excluding Argentina Level o f productivity C O U N TR Y ASIA China2 Ceylon India Pakistan Indonesia Philippines Thailand Weighted average 1 excluding China MIDDLE EAST Egypt Iraq Turkey Weighted average 1 Level o f productivity
3.9 5.3 10.6 6.3 3.6 3.6 5.5 4.7 4.5 75.6/38.1 11.9 17.0 7.0 6.6 6.6 8.6 13.5 9.8
5.4/6.4 5.4 4.0 4.3 4.5 4.7 7.0 4.8 4.3 5.5 7.7 11.3 8.6
1W eighted according to th e active m ale agricultural p o p u latio n em ployed in agriculture (excluding plantations). 2T he first index num ber is based o n w estern estim ates o f p ro d u ctio n , th e second o n in form ation derived from th e Chinese authorities. 3T he second index n um ber excludes stock-raising. Sources: A u th o rs estim ates (see te x t for m eth o d o f calculation). N ote: T he small e x te n t to w hich the figures have been ro u n d ed up does no t im ply any corresponding m argin o f error.
Eastern countries lies somewhere in between. It should be noted, however, th at because Turkey is included, the Middle Eastern figure is n o t fully representative. Obviously within each group there are countries whose level o f productivity differs considerably from the others and is nearer th a t o f other groups, or even o f the developed countries. Argentina is such a country. Argentina is usually included in the Third World because according to m ost classifications it is considered under-developed and because, together w ith Brazil and Mexico, it is one o f w hat m ight be called the Big Three o f Latin America. But it becomes more and more obvious th a t such a classification is w rong.32 All classifications m ust be arbitrary to some extent and they can only be accepted if we recognize this fact.
Agriculture
31
But this said, the extrem ely high level o f agricultural productivity in Argentina m ust surely exclude it from the list o f those countries traditionally called under-developed. 1. Agricultural productivity and thresholds of productivity All th e countries included in the table are above the level of physiological productivity (3.8). But several of the African and Asian countries are below th e productivity threshold (4.9) which marks the potentially free from fam ine level: these are Ghana, Nigeria, Tunisia, India, Pakistan and Indonesia to which we should add (always allowing for margins o f error in the productivity estimates) the Philippines. Thus overall a large m ajority (nearly four-fifths) of the population o f under-developed territories are supported on a level of productivity which still implies a risk of fam ine.33 If we accept, as will be explained later, th at China has been able to pass beyond this threshold, the proportion falls to about one-half o f the entire Third World. Thus as F.A.O. often declares through its directors, Mr Sen and more recently Mr Boerma judged by the calorific value of available foodstuffs, the non-comm unist, under developed countries, especially the countries o f Asia, and to a lesser degree, o f Africa, still live under a very real threat of famine. In Asia only a few restricted areas, such as Sri Lanka and Thailand, are free from this terrible risk. Even fo r those countries whose productivity has risen above th e threshold, the threat o f localized famine still exists, for th e figures given here are national averages and there are very m arked regional differences o f productivity w ithin countries, added to which difficulties o f internal transport often hinder redistribution. Consequently, for m ost under-developed countries, priority for agriculture is n o t so much a choice as a compelling necessity. The existence o f an agricultural surplus in the developed countries (in particular in N orth America) has until now staved o ff a serious c a ta stro p h e. But s to c k s do n o t fo r m m ore than a fraction o f world production,3 4 and it would only be necessary for a series of bad harvests to occur simultaneously in both developed and under developed countries for the problem o f famine in the Third World to become insuperable.3 5 It is obvious therefore th at a rise in agricultural production beyond the threshold where the Third World would be potentially free from famine risk should be an absolutely essential objective. In this connexion I would like to discuss a factor which, so far as I
32
know, has n o t been reckoned w ith in analysing developm ent difficul ties in those countries where religion requires its followers to practise vegetarianism. To help us understand this problem we should remind ourselves of the role livestock plays in reducing the fluctuations in available foodstuffs caused by variable yields. In good years a part of the surplus harvest is used to increase the num bers of livestock; on the other hand, after bad harvests, livestock is reduced and provides more food both through additional quantities of m eat and through a fall in the quantity o f cereals consumed by animals. But with a vegetarian population, in view of the high cost o f food stocks, there is a much greater fluctuation in the quantity o f available foodstuffs, with adverse consequences for the peoples health. A vicious circle is thus set up, for bad health means a lowered capacity for work and thus a smaller agricultural output. But more im portant still is the fact th a t a diet w ithout meat inevitably lessens the incentive to increase agricultural productivity, since consum ption soon comes up against a physiological barrier. In vegetarian diets the gap betw een average and maximum consum ption of calories is fairly small, no more than about 20 per cent. In m eat-eating societies the gap is m uch greater due to the fact th at a calorie o f m eat needs about 7 vegetable calories for its form ation. Thus, at present in developed countries the num ber of calories consumed daily can be as much as 9,000 while the maximum the body needs daily is approxim ately 3,500. Faced w ith such a small elasticity o f dem and, it is n o t surprising th a t vegetarian populations should m ake so little effort to increase their food supply. Furtherm ore, by raising those animals which graze on vegetation which otherwise would be useless to man, total agricultural pro ductivity is increased; these animals are in addition tended by members o f the population (elderly people and young children) whose physical strength does n o t fit them to participate in agricultural production properly defined. In view of all these factors it seems m ore than likely that vegetarianism has hitherto hindered developm ent, and still continues to do so. 2. Agricultural productivity and availability o f agricultural land Allowing for inherent differences in the level o f development, narrowly defined, there is also a correlation betw een the level of productivity and the agricultural area per male agricultural w orker.36 The level o f productivity is higher where the availability o f
Agriculture
33
agricultural land is greater. The correlation is n o t, of course, absolute, because several factors, such as soil fertility, type o f crop or presence o f irrigation, affect the relationship. The correlation coefficient o f the two series in Table 9 is 0.92. Table 9 Comparison between agricultural productivity and availability o f agricultural land
Hectares o f agricultural area 1 per male worker occupied in agriculture 2
C ountry AFRICA Ghana Kenya Madagascar Morocco Nigeria Tunisia Zaire AMERICA Argentina Brazil Chile Cuba Mexico Peru Venezuela ASIA China Ceylon India Pakistan Indonesia Philippines Thailand MIDDLE EAST Egypt Iraq T urkey
Level o f productivity 1 9 5 3 /5 7 -1 9 6 0 /6 4
3.7 5.0 10.5 7.3 4.0 4.2 5.7 64.4(31.8)* 10.9 15.0 6.9 5.9 6.7 7.5 4.8 (5 .8 )t 5.2 4.1 4.4 4.7 4.9 6.7 5.3 8.1 10.9
5.3 4.2 43.2 10.6 2.4 8.7 20.2 102.5 13.0 9.8 7.8 16.7 12.2 26.6 2.7 1.2 1.9 1.4 1.1 2.6 1.8 0.6 13.6 11.3
A rable land; land under p e rm a n en t crops; p erm anent m eadow and pastures. 2 In censuses tak en b etw een 1956 and 1962, except C uba 1946, and C hina 1954. E xcluding cattle farm ing tO fficial estim ates Sources: Productivity, a u th o rs estim ates; agricultural area derived from F.A .O.
34
The lim ited am ount o f agricultural land available in the Asian countries is certainly a m ajor obstacle to any substantial increase in productivity. The figure of 1.8 hectares o f agricultural land37 per male agricultural w orker (the average for Asia, excluding China) is of course very low indeed. As a m atter o f fact the figures for Asia (as can readily be seen from Table 9) are n o t only much lower than those for Africa and Latin America, b u t are also totally unrelated to those o f th e developed countries (France, 13.1; Belgium, 9.4; Italy, 5.8; U.S.S.R. 34.0) n o t to m ention the N orth American and Oceanic countries o f European settlem ent, where the figure is about 100 hectares (United States, 99; Canada, 90; New Zealand, 111; Australia, 1,100. The last figure is explained by the predom inance of sheep farming). Furtherm ore, as we shall see in the comparative section, th e am ount of land available in the currently developed countries before their take o ff was also higher.
3. Agricultural productivity and plantations Productivity in subsistence agricultural is generally higher in coun tries where plantations are relatively unim portant. This situation is illustrated in Africa by Ghana and Nigeria, which both have a large num ber o f plantations and where subsistence productivity is below th at of th e Congo and Kenya where plantations are fewer. There is the same difference in Latin America betw een Brazil and Chile and in the Middle East between Egypt, Iraq and Turkey. The difference is explained by th e fact th a t plantations (or export crops) have m onopolized the best land.38 We should realize moreover th a t the m ore productive m ethods used on plantations do n o t seem to have had an apparent influence on subsistence agriculture.
4. Agricultural productivity and European settlem ent The present point is particularly valid in Latin American countries. It is in just those territories where European settlem ent has been proportionately largest th at agricultural productivity is highest, as is well illustrated in the cases of Argentina, Brazil and Chile. The difference is, o f course, partly attributable to the influence o f a population which has brought with it b etter techniques, b u t th a t is n o t th e only explanation. Had the fertility o f the land been low to begin w ith, or at least n o t obviously superior to the regions abandoned in Europe, the land would n o t have been settled.
I I Agriculture 35
5. Average productivity index for all under-developed countries Finally, around 1970 the weighted average o f the agricultural productivity index for all under-developed countries under con sideration here is about 5.5 including China, according to the western estim ate, and 6.1 including China according to the official estimate. Since these countries represent 80 per cent o f the to tal population of the Third World and since, from the available inform ation, the productivity o f all th e om itted countries together is probably a little lower than th at o f those studied, it is reasonable to set the average level o f agricultural productivity o f the whole non-com m unist Third World at about 5.4. (b) Changes between 1909/13 and 1968/72 While a very great increase in agricultural productivity can be observed in the developed countries in the course o f the seventy or so years which separate us from the beginning o f the century (productivity increased more than five-fold during this period),39 there has on th e contrary been a falling o ff o f agricultural productivity in the under-developed countries. This is especially true in the Afro-Asian countries. In Latin America there has been either a slight advance or at least a m aintenance of the status quo. This is, of course, taking the region as a whole; there are m arked differences between country and country. Between 1909/13 and 1922/26, while there are divergences, the overall change in productivity is minimal. The Latin American and African countries experienced, on the whole, a m oderate growth, but overall agricultural productivity in Asia suffered a slight regression. Between 1922/26 and 1946/50 there was an alm ost general fall which was greater for Asia than for Latin America. Productivity fell in Asian countries by 19 per cent (th at is 1.7 per cent per annum) from 1922/26 to 1934/38 and by 10 per cent (0.8 per cent per annum ) from 1934/38 to 1946/50; while for the Latin American countries in the same period the decline was no m ore than 5 per cent (0.4 per cent per annum ) and 6 per cent (0.5 per cent per annum ). The change in Africa and the Middle East is less clear, although the indications are th a t the status quo was m aintained on the whole, w ith only a slight tendency to decline. From 1946/50 to 1953/57 the general tendency, with the exception o f m ost African countries, was one of growth in productivity. For the whole o f Asia growth was 17 per cent (2.3 per cent per annum ); for Latin America 10 per cent (1.4 per cent per
36
annum ), and for the Middle East 38 per cent (4.7 per cent per annum ). It is true th a t th e classic problem o f margins o f error is still present: estim ates of hom e consum ption', which represents a major portion o f agricultural production in these countries, are likely to have improved in accuracy. It is also possible th a t part of the growth registered should be im puted to progress in m ethods o f estimating. Table 10 Weighted average 1 o f index numbers o f agricultural productivity
Country AFRICA 1909/13 6.9 1922/26 8.7 13.3* 7.6*
1934/38 7.1 12.6 7.5 4.2 4.4 4.1 6.0
1946/50 7.3 11.9 7.8 3.7 3.7 3.7 6.1 4.4
1953/57 5.2 13.1 9.0 4.4 4.4 4.3 8.3 5.1
1960/64 4.7 12.9 9.2 4.7 5.1 4.4 8.4 5.4
1968/72 4.6 14.2 10.4 4.8 5.6 4.2 8.8
AMERICA 11.9 exclud. Argentina 7.2 ASIA China ASIA excluding China MIDDLE EAST Total for 24 countries studied D itto excluding China
5.1* -
5.0
4.9
5.5
5.9
5.9
5.3
5.0*
5.7
5.6
5.5
A fte r th is d ate th e n um ber o f countries included in th e calculation o f th e w eighted average is changed. 1W eighted according to th e num ber o f m ale agricultural w orkers occupied in agriculture, excluding p lantation. Sources: A u th o rs estim ates (see te x t for m eth o d o f calculation). N ote: T h e small e x te n t to w hich the figures have been ro u n d ed up does n o t im ply a n y corresponding m argin o f error.
From 1953/57 to 1960/64 the general tendency was reversed and was once m ore a declining one, especially in the Afro-Asian countries, w ith the notable exception o f China, to which we shall return later. The under-developed countries o f Latin America, due to the decline o f Brazil (3 per cent, 0.4 per cent per annum ) and the relative im portance o f th a t country, experienced a slight fall in the weighted average (1 per cent or 0.1 per cent per annum). Finally, between 1960/64 and 1968/72 productivity rose again if only very slightly. F or all under-developed countries taken together this growth was 0.2 per cent per annum. But if China is excluded there was a slight decline. In relation to the general average, the change has been relatively favourable for Latin America and the
Material com direitos autor
Agriculture
37
Middle East, and unfavourable for Asia. On the basis o f adm ittedly fragm entary inform ation, agricultural productivity in China has grown by 1 to 1.5 per cent per annum. We shall have to return later to this. As far as the effects o f the 'green revolution' are concerned, these will be discussed in section (c) below. Meanwhile let us examine the changes in some im portant countries. For Latin America, stagnation in Brazil during the first half o f this century should be noted and contrasted w ith the favourable changes taking place in Mexico40 and Venezuela, even though the absolute level o f productivity in these tw o countries still remains rather low. In Asia th e m ost significant phenom enon is o f course the growth displayed since 1950 by China, a growth which contrasts with decline, or at least no major change, on the p art o f m ost other Asian countries. Even if we retain the western estimates, the annual rate of growth in productivity has been from 2.0 to 2.5 per cent per annum from 1946/50 to 1968/72, while the index of productivity rose from 3.7 to 5.6.4 1 A t this level China has broken through the barrier and is potentially free from famine risk. It is, of course, probable th at the harvests lie between the estim ates based on western figures and those based on government inform ation. It could then be postulated th at the index o f productivity around 1970 lies above 6. This would allow China to enter a process o f wider industrialization, since at this level o f productivity it is possible to free a significant proportion of the agricultural labour force from w ork on the land. As far as th e second Asian giant, India, is concerned, we are witnessing a deterioration in the level o f productivity. It has at the present m om ent fallen to a level which is about 25 per cent lower than th a t o f th e period 1909/13 to 1922/26. With productivity at 3.9 India finds herself once more below the threshold o f potential freedom from famine risk. The gap betw een the agricultural productivity o f these two countries which have opted for such different modes of development is lik e ly to h ave p r o fo u n d rep ercu ssio n s o n th e w h o le u n d er developed world, especially if the divergence continues in the future. A cautious extrapolation o f th e trends o f Chinese growth and an assumption th a t India, contrary to past decline, will have a future growth o f som ething like 0.5 per cent per annum, results in a prognostication th a t by 1980 there will be a difference o f about 100 per cent betw een the agricultural productivity o f the two countries. No significant divergences can be observed for the African and Middle Eastern countries analysed here.
38
Table 11 Index numbers o f agricultural productivity by country

C ountry AFRICA Ghana Kenya Madagascar Morocco Nigeria Tunisia Zaire AMERICA Argentina excluding cattle Brazil Chile Cuba Mexico Peru Venezuela ASIA China (West, estim.) China (Offic. estim.) Ceylon India Pakistan Indonesia Philippines Thailand MIDDLE EAST Egypt Iraq Turkey 1909 1914 1922 1926 1934 1938 1946 1950 1953 1957 3.7 4.6 10.0 8.5 4.2 4.6 6.4 60.6 30.0 11.1 14.5 6.7 5.5 6.8 6.8 1960 1964 3.8 5.3 11.0 6.0 3.9 3.8 5.0 68.2 33.3 10.8 15.6 7.0 6.3 6.5 8.1 1968 1972
9.7 6.0
13.8 7.0
11.1 7.6
6.1 11.9 7.3
4.3
4.5
5.4 5.6 59.1 36.1 10.1 13.8 5.6 3.0 6.0 5.2
4.4 6.9 50.9 24.3 10.0 13.1 6.5 3.7 7.8 4.9
3.9 5.3 10.3 6.5 3.4 3.4 5.9 86.9 43.0 12.9 18.4 7.0 6.8 6.6 9.2
61.2 33.8 9.4 12.5
65.4 37.0 10.4 12.3
3.9 5.5
3.0 5.7
4.4 5.0 3.9 4.5 3.7 4.7 6.2
3.7 3.6 3.3 4.5 3.7 4.5 6.5 4.6 7.6 7.0
4.4 4.9 4.1 4.2 4.7 5.1 6.7
5.4
5.1
3.5
7.3 5.7

3.3 5.5 8.5 4.8

5.1 6.5 5.5 4.1 4.5 4.6 4.8 6.8 5.4 7.1 11.1
5.6 7.1 5.3 3.9 4.2 4.5 4.6 7.2 5.7 8.3 11.5
4.4 8.9 7.1
5.1 9.2 10.7
Sources: A u th o rs estim ates (see te x t for m eth o d o f calculation). N ote: T he sm all e x te n t to w hich th e figures have been ro u n d ed up docs not im ply any corresponding m argin o f error.
It is quite certain th a t the decline in agricultural productivity to be seen in m ost o f the under-developed countries can be attributed largely to the population explosion. It has activated the old law o f diminishing returns.4 2 The num ber o f male agricultural workers has, on average, more than doubled betw een 1900 and 1970. Such a growth, coupled with the low availability o f land, w hether of average or o f low fertility, obviously poses alm ost insoluble problems. A t this point it is useful to rem em ber th a t agriculture during the industrial
Agriculture
39
revolution developed with a relatively stable working population (see Table 12), the m oderate increase in population (four or five times less than th a t o f the under-developed countries) being relieved by the transfer of part of the working agricultural population into industry (see Chapter 9). Table 12 Change in the male working agricultural population (in round numbers)
Belgium Dates 1846 1856 1866 1890 1910 1930 1961 (1,000) 680 710 700 640 580 500 210 France Dates 1815 1830 1856 1866 1896 1936 1962 (1,000) 4,500 4,800 5,150 5,300 5,750 4,250 2,650 Great Britain Dates 1688 1811 1841 1861 1881 1921 1961 (1,000) 2,000 1,600 1,460 1,820 1,580 1,390 874
Sources: Population censuses o f th e respective countries and dates, except for France 1 8 1 5 - 3 0 ; J. C. T o u ta in , La p ro d u c tio n de I'agriculture franqaise de 1700 a 195S (C ahiers de 1I.S.E.A ., 115, 1961); and G reat Britain: m y ow n calculations based on G. Kings estim ates o f the stru c tu re o f the w orking p o p u latio n in 1688, and on th e 1811 census (fam ily results) and subsequent stru ctu ral data.
The under-developed countries could not possibly benefit from the same relief because th e required increase in the non-agricultural sector would need to be truly enorm ous to keep pace with the vast increases in population. On the basis o f a population growth o f 2.6 per cent per annum and of a non-agricultural sector of, say, 30 per cent of th e working population, a stable num ber of workers in agriculture would assume an average growth of about 8 per cent per annum in the non-agricultural working population during the first ten years o f such a change. Such a rate of growth assumes a pace of industrialization which would be q u ite impossible to a c h ie v e . (c) Comparisons with developed countries at the present time and before the \take o f f One o f the aims o f these estimates is to make it possible to compare the level o f agricultural productivity in under-developed countries with th at of the already developed countries before and during their take o ff. In Table 13 I have adapted some features o f my study on Levels o f econom ic developm ent from 1810 to 1910. The data has
40
been calculated by a som ewhat similar m ethod.43 To provide reference points for the present time I have calculated levels for France and the United States for recent years on the same basis. In considering these cases it is clear th a t the m ethods o f estim ation of productivity devised for economies in an early stage o f their developm ent contain pitfalls if applied to more advanced economies. Table 13 Comparisons between levels o f agricultural productivity
C ountry and 'stage* o f developm ent DEVELOPED COUNTRIES R ecent position France United States Position before or during take o f f France Great Britain Sweden Belgium Germany Italy Russia Switzerland United States Spain LESS DEVELOPED COUNTRIES R ecent position Africa Latin America* Asia Middle East TOTAL FOR ALL LESSDEVELOPED COUNTRIES Period Index num ber o f agricultural productivity
1968/72 1968/72
100.0 330.0
1810 1810 1810 1840 1840 1840 1840 1840 1840 1860
7.0 14.0 6.5 10.0 7.5 4.0 7.0 8.0 21.5 11.0
1 9 6 0 /6 4 -1 9 6 8 /7 2 1 9 6 0 /6 4 -1 9 6 8 /7 2 1 9 6 0 /6 4 -1 9 6 8 /7 2 1 9 6 0 /6 4 -1 9 6 8 /7 2 1 9 6 0 /6 4 -1 9 6 8 /7 2
4.7 9.8 4.8 8.6 5.5
E x clu d in g A rgentina. Sources: A u th o rs estim ates (see te x t for m eth o d o f calculation).
It will be seen the level o f productivity in European countries on the eve o f the industrial revolution was about 7. This is below th a t of Latin American countries, about the same as th a t o f the Middle East, b u t about 45 per cent more than th a t in Africa and Asia which are, from a dem ographic point o f view, the nucleus of the under developed world. A gap o f about 45 per cent is sufficiently wide for us to be able to assert th a t agricultural conditions in the currently developed countries before the beginning o f the industrial revolution
Agriculture
41
m ust have been very different from those o f the under-developed countries o f Asia and Africa today. A different picture emerges if we attem pt a comparison between productivity in under-developed countries and developed countries before their agricultural revolution. By this revolution we mean the rapid growth in agricultural productivity which, in m ost cases, preceded th e industrial revolution b y about fifty years and had itself been preceded, ever since the Middle Ages, by a very slow and irregular increase in agricultural productivity. It is estim ated th a t in the forty to sixty years preceding the industrial revolution agricul tural productivity had grown by about 40 per cent,44 which means th at the average level o f agricultural productivity in the developed countries before their agricultural revolution was about 5 a level which is still slightly above th a t o f the Third World in Africa and Asia today. Given the essential role o f agriculture, we can deduce th a t a similar gap also exists betw een the living standards of the under-developed countries and those o f the developed countries before their industrial revolution.4 5 This being so, the auspices for econom ic developm ent appear to be very bleak even if we were to disregard the other adverse structural factors from which the countries o f th e Third World m ight suffer. Figures of th e availability o f land have been com pared w ith the index o f productivity and here one should note th a t the figures for Asian countries, which are low in comparison with those for developed countries to-day, are also low if they are com pared with figures for the latter at th e beginning of, or before, their take o ff. It will be rem em bered th a t for Asia (excluding China) an average o f 1.8 hectares o f agricultural land per male agricultural w orker has been com puted. The corresponding figure for France at the beginning o f the nineteenth century was 7.5 hectares, and at the beginning o f the eighteenth century was probably nearly 9. For Belgium, a densely-settled country, the figure in 1846 was 2.7 hectares and at the beginning o f the nineteenth century was p r o b a b ly more than 3 hectares.46 For England at the beginning o f the seventeenth century th e figure was a little more than 5 hectares. As one can see, these divergences are very considerable and the gravity of th e situation is certainly n o t lessened by the climatic conditions in m ost Third World territories, which tend to lower both soil fertility and hum an exertion. Given th e stagnation or decline in productivity in under-developed countries, on the one hand, and the constantly increasing pro
42
ductivity o f the developed countries, on the other, the gulf between these tw o groups is now formidable. It can be seen from Table 13 th a t the productivity o f a farmer in the United States is over sixty times th at o f an average Asian peasant. The comparison with France shows the gap to be less, b u t at a ratio of 1 to 20, it is still great. Furtherm ore, this index under-estim ates the real state of affairs because it does n o t take account o f the production o f fresh vegetables, fruit and milk in the U.S. and France. These products play a m uch larger part in developed countries than they do in the Third World. Making allowances for this factor raises the index of productivity by about 15 per cent for the United States and by 10 per cent for France.47 This adjustm ent makes American produc tivity nearly eighty times and French productivity over tw enty times th at o f Asia. In this connexion it should be noted th a t the productivity of agricultural labour has increased very rapidly in the developed countries since 1940/50; indeed it increases noticeably faster than th at of m anufacturing industry a total reversal o f the secular trend. However, we shall return to this problem at the end o f Chapter 6, where the terms o f trade are discussed. But the m ost significant comparison in our present discussion is th at between the levels o f productivity in the under-developed countries and the western countries at the period when the latter began to industralize. As we have just seen, the present average level o f agricultural productivity in African and Asian countries (between them representing four-fifths o f the Third World population) is 45 per cent below th at reached by the developed countries at the start of the industrial revolution. In fact it is at the same level as th at of the European countries before their agricultural revolution. Now, m ost under-developed countries wish, consciously or un consciously, to by-pass this stage just when other structural conditions o f developm ent are making a take o ff more difficult than it was when m ost European countries and the United States were im itating Englands example. What makes the failure to adm it or even to recognize this problem all the more serious is th a t the problem itself is intractable. Leaving aside m ental attitudes, landownership and political considerations, it cannot be stressed too forcibly th at an increase in the area cultivated per agricultural worker is one o f the essential conditions o f an increase in productivity. But in view of the population explosion it is impossible to assume, even on m ost hopeful assum ption, th at the reduction in cultivated area per worker will be anything b u t slight.4 8
Agriculture
43
O f course, a consideration o f present international economic conditions does brighten the picture a little. The western countries could n o t begin to industrialize w ithout a growth in their own agricultural production because they relied on their own resources to feed the increasing num ber o f non-agricultural workers there being no other countries w ith a high enough level o f developm ent to be able to export large quantities o f food continuously and profitably. But today surpluses produced by western agriculture, especially in North America, and the low costs of transportation, make it possible for th e under-developed countries to count on imports o f food. Although, before the war, the under-developed countries (excluding Argentina) had a surplus o f about 11 million tons in their cereal trade, by th e end o f the war this had turned into a deficit which still continues to rise: from 6 million tons in 1950 to 16 million tons in 1960 and 25 millions around 1965. A few good harvests during recent years slowed down the increase in the deficit; it was around 29 million tons in 1972. But it is still too early to say w hether a real reversal o f the trend has occurred. The pre-war surplus represented a little more than 7 per cent o f the total production o f cereals, while the deficit in 1972 is more than 8 per cent. In spite o f the fact th at a significant share o f im ports o f foodstuffs have been provided in the shape o f gifts, or purchased at a reduced price and, in general, food prices have fallen, the cost of food im ports has increased very rapidly, Thus, it is estim ated th at, expressed in value c.i.f., the im ports o f food by non-comm unist, undeveloped countries has risen from S4000 million in 1955 to $6000 million in 1962 and to $8000 million in 1970. This trend cannot continue for two reasons: 1. The resources o f the developed countries are n o t unlimited. It is true th a t in m ost o f these countries because agriculture regularly produces surpluses, laws attem pting to restrict output are in force, thus making an increase in supply possible. But on the assumption th a t the Third World depends on foreign food, could supplies ever be sufficient to fulfil demand? It hardly seems likely. A simple com putation based on a rather m odest hypothesis (population growth lower than th at expected; a lower level o f consum ption at the end o f the period) will dem onstrate the size o f the problem. If we then suppose th a t from 1970 until the year 2000 the under developed countries (including China) wishes to im port foodstuffs am ounting to one-third o f their consum ption, we find the answer to be one which would be impossible to achieve. Basing the com puta tion on cereals alone, and assuming a per capita consum ption o f the
44
same order as in western Europe in 1900 (th at is 500 kilos per head per year), and a dem ographic growth o f 2.3 per cent per year from 1970 to 1990 and 2 per cent for the following ten years, consum ption would reach 2,500 million tons, o f which 800 million would have to be im ported. This latter figure is about three times as m uch as the to tal production o f the non-communist, developed countries and about th irty tim es more than the surplus (up to 1970) in these countries. To satisfy internal and external dem and over the thirty years from 1970 to 2000 (and taking into account the augmented internal dem and which would result from a probable 35 per cent growth in population during this period) production in the developed countries would have to increase more than threefold, and thus grow at an annual rate o f 4 per cent. Even if such a growth were technically possible, from a practical p o in t o f view it would seem quite unrealistic. O utput in these countries during the last tw enty years has only grown by about 2 per cent per annum, so th a t future growth w ould have to double. 2. Even if we assumed th a t growth in the western countries on such a scale were technically possible49 there is another serious economic problem : th at o f the means o f paym ent which might be available to the Third World to enable it to acquire this quantity of foodstuffs. It is absurd to imagine th a t exports o f m anufactured goods to developed countries could ever provide the means. It is not impossible, o f course, th a t such exports might grow in the future, as they have done recently, b u t the cost o f cereals, estim ated above, is five times more than the value o f the exports o f m anufactured goods from under-developed to developed countries in 1970, even based on present cereal prices. It is also obvious, assuming such an enormous increase in production, th a t land o f lower productivity would have to be brought under cultivation entailing a rise in the cost of producing the cereals. Where prim ary products and tropical agricultural products are concerned th e possibilities are perhaps wider, b u t remain limited even so by the size o f dem and in the developed world. Recently, moreover, prices o f these goods have suffered a sharp fall, clearly indicating th a t the present supply is already greater than dem and can absorb. Indeed dem and could hardly grow as fast as the export of agricultural products would have to do if the Third World were to be dependent on external agriculture. It should be pointed out th a t the cost o f the quantity o f cereals in our forecast is larger than the value o f all exports from the under-developed countries.
ial com direitos autorais
Agriculture
45
It is clear from all this th at the m ost im portant (although n o t the only) line o f attack is to raise the level o f agricultural productivity. How to do so is, alas, far from straightforw ard, as has been so amply dem onstrated. As Rene D um ont puts it in his direct way, the real tragedy lies in the length o f tim e it has taken the world to become aware o f th e gravity of th e situation, and this does n o t m ean merely ordinary people, b u t responsible adm inistrators and politicians as well. But we m ust accept the absolute necessity for action if we are to face the austerity and arduous effort which will be required. The prospect of blood, sweat and tears is a disagreeable one.50 But he has also shown us th a t the problem is n o t technically insoluble. We m ust believe this, for otherwise past events might convince us th at the outlook was hopeless. C. Recent changes: A green revolution? Recently there has been much talk o f a new factor in the terrible equation o f food versus population. This has been called by some people th e green revolution, and even by some the green m iracle, brought about by breeding varieties o f high-yielding cereals adapted to conditions in the under-developed countries. The story o f the green revolution is as follow s.5 1 In 1944 th e Rockefeller Foundation financed a programme in Mexico to select new varieties o f w heat suitable for th at country. It was originally intended to find a variety resistant to stem rust (blight). But in 1955 attention was directed also to yields. Norman E. Borlaug, the future Nobel Prizewinner, and his team developed several new varieties by crossing the dw arf Japanese wheat, Norin No. 10, with other wheats. A fter being tried o u t in Mexico, where spectacular improvements in yields were achieved, the new varieties o f w heat were successfully introduced in the fifties into the Middle East under the auspices of F.A.O. In 1964 India and Pakistan im ported some Mexican seed to
sow sm all areas. In 1968 India so w e d 3.5 m illio n h ecta res and
harvested a to tal o f 16.6 million tons, which was 50 per cent more than in 1963, while Pakistan, where 1.2 million hectares were sown with high yield varieties, the total w heat harvest at 6.5 million tons was 1.1 million tons m ore than the national objective. In 1971 the area sown in India rose to 59 million hectares and in Pakistan to 3.0 million. O ther Asian and Middle Eastern countries followed suit, especially Turkey, Egypt, Afghanistan, Iran, Iraq, Kenya, Nigeria, Syria and Ethiopia, and several countries even developed their own varieties.
46
The adventure began somewhat later for rice, the basic cereal o f so much o f the Third World. In 1962 the Rockefeller Foundation and the Ford Foundation together founded the International Institute for Research on Rice a t Los Banos in the Philippines. Research was concentrated on improving the type known as Indica. Within about one year th e Institute developed IR-8, a variety resulting from a cross between an Indonesian rice called Peta w ith a dw arf Taiwanese variety, and this was christened the miracle rice because o f its fantastic yield. This variety and others, similar b u t more recent, were first sown on a large scale in 1966/67. It is estim ated th at by 1971 about 13 per cent of the area devoted to rice in southern and eastern Asia was sown with the new varieties o f seed. To begin w ith, therefore, progress m erited the epithet miracu lous, since new varieties of rice and w heat produced a great increase in yields, sometimes more than 100 per cent. The epithet was probably bestowed all the m ore readily since a com bination o f circumstances resulted in its total effect in the years 19678 being over-estimated. It will be rem embered th at in 1965 and particularly in 1966 agricultural production in the non-com m unist, under-developed countries had practically stagnated, mainly due to bad harvests in India, w ith a consequent decline in o u tp u t per head. But in 1967, food production leapt ahead by nearly 5 per cent and m uch o f this increase was attributed to the use o f new seed varieties. It is certainly possible th a t the new seed played its part, b u t n o t to any great extent. The real im pact is difficult to measure since there is usually rapid growth in o u tp u t after a series o f bad harvests when the land has had tim e to recover some of its fertility. This phenom enon together w ith m ore favourable w eather would in itself bring a rise in yields, which, com bined w ith stable or, m ore probably, increased areas under cultivation, would in turn lead to further rises in output. Similar, or even greater, increases had occurred earlier in the Third World countries, particularly in 1953, 1958 and 1962, to m ention the period 1950/70. In spite o f an extended use o f new varieties of seeds, food production for all the non-comm unist under-developed countries between 1962/66 and 1971/75 improved only by a little less than 2.7 per cent per annum a growth rate slightly above that o f population and even below that o f the 1952/561962/66 period (2.3 per cent). It is, therefore, obviously too soon to speak o f a clear break w ith earlier trends. Although it is true, as we have seen above, th a t new cereals can theoretically increase yields and productivity, experts warn us that
ial com
Agriculture
47
the pace o f dissemination is limited by the need to em ploy fertilizers and pesticides more intensively. In some cases also irrigation has to be extended. Furtherm ore, it is often hard to persuade the people for whom the new varieties are intended (and this applies particularly to rice) to m ake th e necessary change in their traditional eating habits. The real problem , therefore, is n o t so m uch the discovery o f new ways to increase o u tp u t and productivity as the ability to adapt such discoveries rapidly and widely to the conditions o f ecology and soil chemistry to be found in the different regions. After all, in nineteenth-century Europe new m ethods of agricul ture advanced by no more than a few miles a year. Modern com m unications should certainly help to speed the pace, and have already done so to some extent. But it m ust never be forgotten th at country people to-day are still largely illiterate. For instance, if nearly 70 per cent o f th e to tal population o f 14 years and over in India is illiterate, the level o f illiteracy amongst the rural population would certainly be over 90 per cent. In such conditions the main problem is n o t the existence o f innovation b u t how to disseminate knowledge o f it. A part from this there are tw o other problems o f a different kind involved in spreading th e green revolution\ The first is one of social structure. It seems increasingly as if the introduction o f improved seed only serves to accentuate the divergence o f income between rich and poor farmers, since th e larger landowners and well-to-do farmers can m ore easily use th e new seeds and pay for the increased input needed. The increase in production resulting from the use of improved seeds leads, it is tru e, to a fall in cereal prices. But in the case o f th e larger farmers th e increase in o u tp u t generally compen sates for the fall in th e price per unit, while the poor farmers, not having th e opportunity to use the new seed, do n o t obtain higher yields, so the fall in th e price per unit goes w ith a m ore or less stable output. The rise in th e to tal revenue o f rich farmers, therefore, may be and g e n e ra lly is a c c o m p a n ie d by a fa ll in the total income of the poorer ones.5 2 TTie second problem is perhaps even more serious because o f the risks which it reveals. The new seeds are m ore vulnerable to disease, and as large num bers o f local varieties are being replaced by single ones, th e consequences o f a new disease could be catastrophic. For all these reasons it is m uch to o soon to deliver a considered judgem ent on w hether the experience has been a green miracle or a green mirage. Such a judgem ent will probably n o t be possible
48
before the 1980s at the earliest. In the meantime the Third World, which by 1980 will have to feed nearly one-third more m ouths than in 1970, still lives under the threat o f severe famine.
Extractive Industry
This chapter and the next will be devoted to industrial activity, dealing respectively with changes in the extractive and in the m anufacturing industries. Any such study m ust, however, start with a brief outline of the connexion between the tw o, since the extractive industries depend largely on the level of activity in m anufacturing industries. I shall then enquire what changes have taken place in the exploitation of natural resources, and discuss the causes o f growth. Finally I shall take a look at the consequences of this growth. I shall be dealing only w ith the non-communist, developing countries because the structural characteristics which interest us operate only in these countries. The o u tp u t o f extractive industries in non-comm unist, under-developed countries, as we shall see, is destined alm ost exclusively for export, which is not so in China or the other com m unist under-developed countries. A. The connexion between the extractive and m anufacturing industries In a general, if somewhat simplified way, one could say that extractive industry exists to produce raw materials and energy intended alm ost entirely for m anufacturing industry. If this view is accepted then extractive industry m ust be seen merely as a sector contributing to m anufacturing industry with the latter determining absolutely the level and pace o f its activity. But this economic determinism is affected by technological developments and, in particular, by two factors which m odify the relationship between the tw o industries. 1. Firstly there is the tendency o f m anufacturing industry to produce goods of a m ore and m ore complex nature. This is so obvious th at we need not linger over it. 2. Equally there are im provements in technique which lead to a better use o f raw materials. This may be observed in several fields. Thus, taking coal as an example, we can see th a t technical progress has led to increased productivity from its use. A t the end o f the eighteenth century about six tons o f coal were needed to produce a ton o f cast iron, b u t now the corresponding figure is less than one ton (in some m odern blast furnaces it is even less than 800 kgs,1 although, o f course, consum ption varies according to the ore used 49
50
and its m ethod o f preparation). A similar im provem ent is apparent in the generation o f electrical energy; thus, in Belgium, thanks to the increased efficiency o f generating stations, net consum ption in kilo/calories per kilow att/hour has declined from 4.375 in 1939 to 4.033 in 1950, 2,995 in 1960 and 2.418 in 1970.2 Technical progress has also led to larger yields o f iron and electricity from lesser quantities o f raw m aterial, through the improved resistance of steel and the increased productivity o f machines using electrical energy. One could, o f course, quote many other examples of this kind. In economic term s such changes in technology are generally accompanied, particularly in the developed countries, by a slower growth in the extractive industry relative to m anufacturing industry. Thus, in western developed countries between 1958/60 and 1968/70 the annual rate of growth in m anufacturing industry has been 6.0 per cent, while for extractive industry it has been no m ore than 2.7 per cent. This m arked divergence is the result of a long drawn o u t evolution, as we can see in Table 14. The pace o f growth in extractive industry at th e start o f the century was actually slightly above th at o f m anufacturing industry, b u t the gap between the rates
o f g ro w th in th e d e v e lo p e d c o u n tr ie s grad u ally w id e n e d in fa v o u r o f
m anufacturing.3 Table 14 Annual rates o f growth o f output

M anu facturing industry U nited States 1 8 9 9 -1 9 1 4 1 9 1 4 -1 9 2 4 1 9 2 4 -1 9 2 9 1929 -1 9 5 1 N on-com m unist developed countries 1938 -1 9 4 8 19 4 8 -1 9 5 3 195 3 -1 9 5 9 1 9 5 9 -1 9 6 6 1966 -1 9 7 2 E xtract ive industry 6.0 4.3 3.7 2.4 Ratio 2:1 %
5.5 4.3 4.8 3.4
109 100 77 71
4.0 7.2 3.9 6.3 4.5
2.2 3.5 1.9 2.6 2.1
55 49 49 41 47
Sources: U.S. figures derived from H istorical S ta tistics o f th e U nited S ta te s (N ew Y ork, I9 6 0 ); developed co u n tries derived from various issues o f M o n th ly B ulletin o f S ta tistics (U .N .); recent figures derived from M ay 1974 issue.
Extractive Industry
51
What happened in the developing countries was exactly the opposite. N ot only was the rate o f growth in extractive industry higher than th a t in m anufacturing industry, but the gap tended to increase. It is this phenom enon, as well as general changes in the extractive industry, which will be discussed in the following pages.
B. Changes in extractive industry: Causes and consequences

I. C H A N G E S IN O U T P U T
The statistical office of the United Nations calculates an index o f production for extractive industries as part o f their index of general industrial production. The latter, which does not go further back than 1938, is based on the aggregation o f several national indices, and mainly because o f this it does n o t adequately reflect changes which to o k place before 1960. The picture o f the pre-war period is even more inadequate because very few o f the developing countries possess indices going so far back. For this reason we have had to calculate our own index in a m anner which will be briefly described. But th e United N ations index is m ore accurate than this one for recent years because the statistical coverage has improved. For this reason I have retained m y own calculations in the present edition only for the pre-1960 period, and have linked my index o f earlier data with th at calculated by the statistical office o f the United Nations. This also gives readers an opportunity to update the figures. (a) Method used to construct an index o f production o f extractive industries in the non-communist, under-developed countries For th e years 1900, 1913, 1926/28, 1936/38, 1948 and 1953 to 1965 to tal o u tp u t in the non-comm unist, developing countries was calculated for the following products: coal and lignite, natural gas, petroleum ; iron, copper, zinc, lead, tin, chrome, tungsten, nickel and aluminium (bauxite) ores; gold and silver. Two separate indices have been c o n str u c te d , o n e for fu e ls (c o a l and lig n ite , n atu ral gas and petroleum ), the other for various minerals (i.e. all other products w ith the exception of those listed as fuels). In the fuel index each o f the three products has been weighted according to its calorific value. Natural gas and oil have been given an equivalent in coal on the basis o f the coefficients adopted by the statistical services o f the United Nations to determ ine the consump tion of energy (1 ton o f crude oil = 1.30 ton o f coal; 1000 cu. m of natural gas = 1.33 ton o f coal).
52
The ore index and the total index (fuel and ore together) are weighted according to th e estim ated value in 1963 o f the o u tp u t of each product on the basis o f prevalent world prices.4 (b) Results and comments Table 15 shows th e results of these calculations: it attem pts to provide an index o f th e mining o u tp u t o f the entire non-comm unist Third World. Table 15 Index o f output o f extractive industries o f all non-communist , less-developed countries (1963 = 100)
Fuel 1900 1913 1 9 2 6 -2 8 1 9 3 6 -3 8 1948 195 3 -5 5 1 9 5 8 -6 0 196 3 -6 5 1 9 6 8 -7 0 0.9 3.3 8.3 12.4 25.2 45.1 69.8 108.2 173.7 Minerals 9.1 17.9 30.6 37.8 37.6 56.1 79.3 106.3 132.3 A ll 1.8 4.8 10.6 15.1 26.5 46.2 70.8 108.0 162.7
Sources.- S tatistical office o f th e U.N.; and fo r figures up to 195860, a u th o rs estim ates (see te x t). Note.- T he small e x te n t to w hich th e figures have been ro u n d ed up does no t im ply a corresponding m argin o f error.
Here I m ust point o u t th at there is a margin o f error which, though not im portant, is nevertheless greater than the apparent precision o f my decimal point might suggest. My intention in not rounding o ff the figures was to ensure as small a margin o f error as possible in any calculations based on my indices th a t readers may like to make for themselves. The margin o f error in the index is fairly small between 1948 and 1970. On the other hand, because the basic data is so much more unreliable, the margin o f error in the pre-1948 indices (specially from 1900 to 1928) is larger, and therefore the figures should be taken merely as orders o f magnitude. Furtherm ore for those years an additional distortion is introduced due to the weighting which is based on relative values of 1963. But on the whole the general trend which emerges from the index m ay be accepted as fairly realistic.
Extractive Industry
53
The trend of the global index between 1900 and 1970 reveals two distinct periods. The first, lasting from 1900 to 1948, shows a rate of growth o f 6 per cent per annum in the extractive industries as a whole. Such a rate, in its historical context, may be considered very high, especially since it takes in the great depression o f the thirties. An intensive exploitation of the mineral resources o f the Third World thus began m any years ago. However, another rapid acceleration occurred during the second period commencing about 1948. Between 1948 and 1970 annual rates of growth rose to an average o f 9 per cent. There is no need to point o u t th at such rates are extrem ely high. Before proceeding to study the causes and effects o f growth in the extractive industry, we m ust take a brief look at the com ponent sectors. We can see th at the m ost rapid expansion took place in the o u tp u t of fuels, mainly oil, which by 1963 form ed 67 per cent of the total. The annual rate o f growth of fuel production was about 7 per cent between 1900 and 1948. But after 1948 this rate increased noticeably, reaching more than 9 per cent per annum between 1953 and 1970. Where minerals were concerned there was an irregularly changing rate o f growth from about 4.6 per cent per annum (for the period 1900 to 1927), to 2.1 per cent (1927 to 1937), to a stagnation (1937 to 1948), to 7.0 per cent (1948 to 1953), and to 8.0 per cent (1953 to 1957). Between 1957 and 1970 the changes were also irregular, alternating between years of regression (1958), relative stagnation (1961, 1962 and 1967), and heavy expansion (1959 and 1960). But on average the annual rate of growth between 1958 and 1970 was of the order of 6 per cent. The trend of the index of mineral production conceals some very divergent movements in its com ponent parts. Broadly, these move m ents can be separated into tw o periods with the years 1953/55 as the dividing line. Between 1900 and 1953/55 there was a vast expansion in the o u tp u t o f iron ore, together with copper, bauxite, tungsten and nickel ores; a more m oderate expansion in the output o f zinc, lead and chrome ores; and relative stagnation in the o u tp u t of tin, gold and silver. From 1953/55 to 1970 there was also a vast expansion in the extraction of ores in the developing countries, but it was concentrated on iron and, until 1967, bauxite, with only a m oderate expansion of copper and tin, coupled with stagnation, or even a fall, in the production of other ores. From the geographical point of view, growth in o u tp u t during the last two decades has been more rapid in the middle eastern countries
54
and Africa, mainly because of the enorm ous expansion in oil production and (especially in Africa) iron ore. In Latin America growth has been slow, even slower than th a t o f m anufacturing industry.
II. C A U S E S O F G R O W T H IN O U T P U T
It m ust be emphasized th at the enorm ous increase in the o u tp u t of the extractive industries in the developing countries was in no way due to the demands of local industry. Indeed local industry absorbed only a fraction o f production. For proof o f this it is only necessary to com pare either globally or country by country o u tp u t with exports of the products of mining and extraction. Thus, for countries like Brazil, Chile, Liberia and Malaysia, exports of iron ore vary from 80 to 100 per cent of production. Further, while in 1970 the under-developed countries produced 39 per cent of world o u tp u t of iron ore (excluding China and the U.S.S.R.), the same countries produced slightly less than 5 per cent o f the w orlds steel. Thus, some 90 per cent o f the iron ore m ined in the under-developed countries goes to feed the blast furnaces o f the developed countries. I say some 90 per cent despite the fact th at the ratios (39 per cent and 5 per cent) produce an export rate o f 80 per cent, since our calculations have to allow for the use of scrap iron. Naturally we also find evidence o f the process at the other end of the scale: for example between 1952 and 1970 the French iron and steel industry increased its consum ption of foreign iron ore by about 2700 per cent. Starting in 1952 at 359,000 tons (that is 1.3 per cent of Frances total consum ption) by 1970 the figure had risen to 9.800.000 tons (21 per cent of her total consum ption). In th at year 2.240.000 tons o f iron ore came from as far away as Brazil and a further 640,000 tons from Australia. Im ports by all E.E.C. countries of iron ore originating in the under-developed countries rose from 15 million tons in 1960 to 46 million tons in 1970. Most o f the other products o f mining and extraction underw ent a similar experience with the result th a t the share o f under-developed countries in world o u tp u t has gradually grown. In Table 16 changes in the share in world production of three minerals produced by the less-developed countries have been calculated. Developments o f this kind, moreover, tend to perpetuate them selves. The construction o f larger and larger ore-carrying ships leads to a reduction in the cost o f transport, while the relocation near the ocean of the iron and steel industry in the developed countries
Extractive Industry Table 16
55
Changes (191370) in the less-developed countries' share in world production o f some extractive commodities (in 1000s o f tons)
1913 1928 1948 1960 1970
IRON ORE (F.E. content) World production1 Less-developed countries ou tp u t As % o f total BAUXITE World production Less-developed countries ou tp u t As % o f total CRUDE PETROLEUM World production Less-developed countries output As % of total
158,0002 4,0002 3% 550 2 0.4%
7 3,000 4,900 7%
87,300 7,300 8%
172,300 49,500 29%
285,000 112,000 39%
2,100 450 21%
8,400 5,100 61%
25,100 17,000 68%
5 3,500 31,800 59%
44,800 6,900 15%
171,000 43,000 25%
438,000 156,000 36%
901,000 497,000 55%
1,964,000 1,314,000 67%
E x c lu d in g C hina and th e U.S.S.R. 2 W eight o f o re and n o t iron co n ten t Sources: Derived from various num bers o f S ta tistica l Y earbook (League o f N ations); S ta tistica l Y earbook (U .N .), and Eisen und S ta h lstatistik and P etroleum Press Service.
(which is both a cause and an effect o f this change) merely hastens the tendency. The same developm ent can be foreseen for oil tankers. In 1970 the largest tankers in service had a capacity o f 325,000 tons (dw t), 16 times greater than the average tanker in 1958 (while needing a smaller crew) and about 150 times greater than the first tanker of 1868 (which, incidentally, was English). Tankers of 500,000 or even 1,000,000 tons are envisaged in the near future. In addition, speeds o f navigation and o f handling are also m uch greater. However, as a result, on the one hand, o f recent discoveries of rich deposits of ores, oil and natural gas in the West (see below) and on the other of threats to the security o f the supply lines created by the increased dependence o f the w estern world on im ported fuels, it is probable th at th e rapid growth of the Third W orlds share in world production will slow down. And it is not even impossible th a t the upward trend will be reversed, as has already happened in the case of bauxite mining. For ores with a high specific value, the exploitation of mines located a long way from the site o f conversion has been profitable
56
for a relatively long period because transport costs, high though they may be, represent only a small part of prime costs. But for ores and fuels with a low specific value (like iron) the physical separation of mines and blast furnaces is only possible if the cost o f transport is very low. The price of iron ore was, in 1965, about 510 a ton for ore of 50 per cent iron content. By way o f comparison the price of Venezuelan oil was about $20 (also the average 1965 price in $ per m etric to n ); o f zinc ore $70; of lead m ore than $100 (relative to its metal content, o f course: in this case 30 per cent), and of tin about $2000 (for a m etal content o f about 70 per cent. Prices o f tin ore fluctuate very widely and follow closely the price of the m etal itself. In the course of 1964 the highest m onthly average price was as much as 50 per cent above the lowest.) It is not easy to determ ine average transport costs. Here too there are heavy fluctuations in prices. Even if annual averages are used, differences of the order of 50 per cent between one year and another are possible, and a difference o f more than 250 per cent between maximum and minimum average annual prices has been known to occur within the space of five or six years. Furtherm ore, prices also vary according to the kind of contract (liner service, time charter, trip charter, etc) and th e nature o f the product. But to fix some sort of order o f magnitude, we may accept th at to move such products as ores or oil by sea a distance of 6000 8000 km (40005000 miles) in the mid-sixties would have cost about $6 per ton. By rail, assuming a com plete trainload, the cost per ton per 100 km (60 miles) would have been $0.8. Consequently while transport costs were high, long-distance movem ent of goods such as iron ore was economically impossible. Thus, in 1930, according to Ivar Hogbom s estimate* for the under-developed countries, maximum distances (allowing for trans p o rt costs) separating mineral workings from export ports were as follows: In normal circum stances, he says, coal, iron ore and phosphates can only be mined within a radius o f 100 km (60 miles) and mineral oils within a radius o f 250 kms (150 miles) of the coast. Manganese and chrome ores may be mined further from the coast than iron ore, and zinc and lead ores at an even greater distance. Nevertheless, the only com m on m etals which can be mined econo mically anywhere in the world, th at is a long way from the sea, are copper and tin . A t the present time, o f course, taking iron ore as our example, the deposits being exploited in Mauretania lie 675 km (420 miles) from the p o rt of export, i.e. seven times further than was considered normal before the war. The same applies to the deposits
Extractive Industry
57
at Rourkela in India (720 km or 450 miles), at Cassinga in Angola (630 km or 390 miles), at Vale de Rio Doce in Brazil (560 km or 350 miles), at Knob Lake in Canada (570 km or 355 miles), etc. The fall in transport costs, therefore, has given a powerful impetus to the growth of extractive industries in the under-developed world. But there are other factors, such as the gradual exhaustion o f the richest and m ost profitable deposits in the developed countries and the design and production o f efficient equipm ent for earthmoving and for public works, which, especially since the last war, has made possible the exploitation o f mines and quarries and the building of transport lines in many under-populated regions, using small teams of technicians from the developed countries and local unskilled labour. This has been the case in Mauretania for example. But it remains nonetheless true th at w ithout low international transport costs the exploitation o f minerals in the under-developed countries could not have been organized on such a scale and, m ost im portantly, the o u tp u t o f such an expansion could not have been exported. We m ust now analyse the consequences o f these special circum stances of the mining and extractive industries.
I ll T H E C O N S E Q U E N C E S O F G R O W T H O F M I N I N G AND E X T R A C T I O N IN U N D E R - D E V E L O P E D C O U N T R I E S
(a) Mining and manufacturing industries The exam ination o f the causes of growth in the extractive industries of the under-developed countries has revealed that one of the normally expected consequences of such growth (i.e. the creation of nearby industries to convert the minerals and ores into a more elaborate form ) has not in fact occurred, since almost all o u tp u t is exported. Additionally it appears th at the great extension o f mining, so far from encouraging the developm ent o f m anufacturing industry, has actually hindered it. This conclusion emerges from an exam ination of the rate of growth in the tw o sectors. In fact, where one could expect to find a positive correlation between high rates of growth in extractive and m anufacturing industries, brought about by the generally beneficial effects o f the developm ent o f mining on the economy of Third World countries, one finds quite the opposite occurring. For the period 1958 70, rates of growth in the m anufacturing and extractive industries for nineteen under-developed countries for which relevant statistics are available have been compared. From an exam ination of these figures it is apparent th at, among the countries registering a considerable expansion in the mining sector, most show
58
only slow progress in m anufacturing industry, and vice versa. Clearly, then, there is a correlation betw een high rates of growth in the extractive industry and low rates in m anufacturing industry. Al though it might be rash to assert th at the developm ent o f mining and allied activities acts as a brake on the general growth o f other sectors, it appears nonetheless from a comparison o f the pace of growth that a rapid expansion of the mining sector is certainly n o t sufficient in itself to engender vigorous development in m anufacturing industry. To support this opinion I shall outline the conditions under which growth in mining might lead to regional development, and show how present circumstances in m ost under-developed countries prevent this from happening. I shall then briefly discuss the negative factors exhibited by the growth of extractive industries. (h) Theoretical possibilities arising from expanding extractive indus tries The discovery and exploitation o f a source o f raw materials opens up the immediate possibility of establishing secondary industries to make use o f these materials. As we have seen, during the nineteenth century it was unthinkable in economic term s that raw materials should be processed at a site more than a short distance from the deposits themselves. In practice, therefore, mining was usually a prelude to the establishm ent of secondary industries. But when transport costs began to fall, geographical separation a t once became possible, and today as we have seen above a large proportion of extractive industries in the Third World feeds secondary industries in the developed countries. A m ost im portant opportunity to prime the pum p is thus lost. But this is not normally the only opportunity theoretically provided by th e development o f mining. Broadly speaking other opportunities might arise in the following ways: 1. Induced effects of activities connected with the construction and equipping of the mine workings themselves. 2. Induced effects o f activities connected with the transport infrastructure built to remove the o u tp u t of the workings. 3. Local investment of profits accruing from mining. 4. Effects o f aggregate wage outlay. Each of these points will be dealt with in turn. 1. The induced effects arising from the construction and equip m ent of mine workings are, practically speaking, non-existent. Given
ial com
Extractive Industry
59
the low level o f industrial technique in under-developed countries and the sophisticated nature o f modern extractive enterprises, almost all the machinery needed m ust be im ported from the developed countries. This also applies to almost all capital equipm ent, as we shall see in the chapter devoted to m acro-econom ic data. 2. Induced effects o f tw o kinds m ust be considered in connexion with the infrastructure built to remove production. The first is connected with the actual construction of the transport network. Here one lim itation is immediately obvious: any such effects would be of short duration. But even if this aspect were disregarded, we should still be confronted on the one hand by the factors just discussed in the case o f mining equipm ent, which apply equally to transport equipm ent, and on the other by aggregate wage outlay, the effects of which will be investigated below. The second kind o f induced effect is the more interesting. It arises from the use o f the transport netw ork once it is in being. The orientation o f the lines of com m unication tow ards the export ports does not, in my view, produce positive effects. Quite the contrary, for it favours increased im ports of m anufactured goods from developed countries m uch m ore than it encourages the creation or developm ent o f undertakings serving local consum ption. By lowering transport costs the infrastructure, intended to make mineral exports profitable, at the same tim e favours the im port o f industrial or agricultural goods coming from countries with a higher productivity. And this effect is even more pronounced since the availability o f a return freight reduces still further the cost o f transport in the direction of im ports coming from developed countries. 3. The possibilities of investing the profits from mining and extraction. W ithout entering into a discussion of all the various cases (which would go beyond the scope of this study) I think it is obvious th at in m atters o f investm ent both the nature of the firm and the policy o f the government involved are o f prime im portance. In the present discussion we may generally assume th at we are talking about countries dom inated by foreign companies whose main aim is the realization, and especially the repatriation, o f as large a proportion of the profits as possible. Such companies do not prom ote investment in an under-developed country except when by doing so they would earn profits higher than those produced by investing in a developed country and high enough, moreover, to compensate adequately for the additional risks of nationalization and political troubles. The latter consideration also partly accounts for the reluctance o f these companies to extend their activities into secondary industries, since
60
the under-developed countries offer no special comparative advan tages in this respect. It is solely the wealth of their subsoil which offers such a high return and allows additional profitability on capital sufficient to com pensate for the m arket being so far removed from the mines. There is therefore small likelihood th a t the proprietors o f extractive enterprises would spontaneously diversify their activities outside the sector in which these countries present obvious natural advantages. The only way to change such a situation would be to subtract somehow part of the profits in order to invest them locally. But it is obvious th at whatever form such retention of profits took (e.g. royalties, or legal agreements requiring all or part o f the profits to be invested locally) it could not be very substantial, since this would to some extent limit the profitability of the enterprise and thereby discourage investors. And even if profits were retained, the problem of how they were to be judiciously invested would remain. The example of oil royalties which have usually been invested in residential and administrative building, within and, particularly, outside the country, dem onstrates th a t the problem is not simple. These reservations apart, retention o f profits in some form offers real possibilities for countries whose wealth lies in the subsoil. But it is essential th at political conditions should be such as to make the imposition o f such a policy practicable, and in m ost countries in which extractive industry (either oil or mineral) predom inates this is far from being th e case. But the success achieved for m ajor oil producers by the Teheran Agreement o f January 1971 is encour aging. 4. Aggregate wage outlay, i.e. the possible effects on developm ent o f an increase in demand resulting from wages spent. Here three factors combine to reduce the likely effects. 1. Em ploym ent in extractive industries (particularly in oil pro duction) is generally not very substantial because only the m ost profitable deposits are exploited (Latin America, perhaps is an exception to this rule). 2. Skilled personnel is usually recruited in the developed coun tries, which has the effect o f reducing considerably the demand for local consum ption goods. 3. The advantage offered by return freight tends to favour goods im ported from the advanced countries. What makes the advantage all the greater is th at mines and oilfields are usually located far from the centres of agricultural production.
Extractive Industry
61
Finally, therefore, it seems th at few o f the factors which might, in the developed countries, aid developm ent through the exploitation o f natural resources play a similar part in th e Third World. Furtherm ore to the lack o f these positive factors m ust be added the existence o f certain negative ones. (c) Negative aspects o f the development o f extractive industries By attracting a high proportion o f qualified labour, already limited in such countries, mining severely restricts the opportunities for other sectors to recruit such labour. The same argum ent applies to capital in so far as local capital is to be found in such enterprises. Lastly, and this is the m ost significant of the negative factors, mineral and oil deposits m ust to a great extent be treated as irreplaceable assets. And from this point o f view the prem ature exploitation of the m ost profitable workings, whose o u tp u t is exported, may profoundly handicap the country in question ten or fifteen years later, just at the m om ent when it might have been able to achieve the conditions necessary to process its own raw materials. A t such a juncture, the earlier skimming off of the m ost im portant deposits will considerably reduce average profitability, and, conse quently, increase the costs of production. Thus, just when the country might need raw materials at low cost to facilitate the launching o f secondary industries it may no longer be able to lay its hands on them . In my opinion this factor is not sufficiently taken into consideration by local powers. Seeing th at extraction by itself does n o t create conditions favourable for general development, it is dangerous to encourage too rapid an expansion in this sector w ithout taking account o f probable reserves and, above all, the economic value o f those reserves. Thus, it may be assumed in general th at far from creating favourable conditions, the expansion of extractive industries in under-developed countries, as carried o u t at present, produces an unfavourable rather than a favourable economic climate. Nevertheless the problem is far from simple, because rapid progress in developing and utilizing synthetic materials may mean that certain raw m aterials may lose their econom ic value in the future. Despite (and maybe because of) the recent debate on the energy crisis this m ay also be true of oil; indeed the fall in the cost o f generating nuclear energy and th e perfection of electric traction for m otor cars, b o th of which are becoming increasingly feasible, m ay have substantially reduced th e econom ic utility o f oil reserves by 1 9 8 0 -9 0 . And this does not take account o f growing com petition in the
62
foreseeable future from the expansion o f reserves in the developed countries. In the East, Russia is beginning to have a surplus problem, thanks to a great expansion o f oil production which betw een 1953 and 1970 m ultiplied more than six times, rising from 8 to 16 per cent o f world production. The surplus is even more obvious in the field of natural gas, where o u tp u t has m ultiplied more than thirty times between 1953 and 1970. In the West, natural gas from Europe is beginning to tu rn the whole structure of energy consum ption upside down. French deposits at Lack, and others in the North Sea and in Holland are increasing prodigiously the volume o f energy reserves available to the Old World, where the under-developed countries principal custom ers for oil are at present to be found. O utput has already reached an appreciable level. Thus, in the six principal producing countries, Germ any, Austria, France, Italy, the N ether lands and Great Britain, production of natural gas has risen, between 1960 and 1972, from less than 12 to 126 thousand million cubic metres. In 1972 locally-produced natural gas contributed some 10 per cent of the total prim ary energy consum ption o f western Europe compared to one per cent in 1958. And this is no more than the start of an expansion which will accelerate enorm ously in the years to come. Furtherm ore, if recent indications are to be believed, we cannot exclude the possibility th at the N orth Sea will prove to have very rich oil deposits as well. It now seems probable th a t Australia will become an im portant oil producer in the antipodes, and elsewhere Alaska is in the process of becoming a new Texas. In general, undersea deposits are almost certain to augment the volume o f reserves in the developed countries. Thus, to repeat, the problem is certainly not straightforw ard. The margin between th e danger of a prem ature exploitation, which might compromise the future chances of an easier industrialization, and a delay, which entails the risk o f seriously reducing the economic value of the deposits, is both narrow and hard to determ ine. In addition, in many cases the interests of one country run counter to those o f the larger region o f which it is but a part.
Copyrighted
4
INTRODUCTION
Manufacturing Industry
The subject o f m anufacturing industry is of particular im portance because the developm ent o f secondary industries is the principal objective o f all Third World policies aimed at economic growth. Moreover the degree o f industrialization attained by under-developed countries is frequently used to mark the stages in their upwards progress from unindustrialized beginnings. This chapter is divided into three sections. The first examines rates of industrial growth (total, sectoral and by country) and compares these rates with those of developed countries at different stages in their growth. In the second section I shall try to define the level reached by industry in the main Third World countries and here again there will be an attem p t at a comparison with the developed countries at various stages in their growth. The tentative nature of this exercise will become clear later when we see th at precise definition of absolute, or even relative, levels of developm ent is extrem ely difficult. The existence of structural variations and dissimilarities make realistic comparisons a m atter of delicate judgem ent. However, by using six different indicators I hope to have reduced the margin o f error to an acceptable scale. Finally, the third section will deal briefly with the problem of the relationship between levels o f agricultural and industrial development, from the point of view of actual levels attained as well as that of changing economic conditions. I originally intended to deal at length with changes in produc tivity, but the paucity of statistical material has forced me to abandon this project, because the results would be too limited and too tentative to provide valid conclusions. It should only be noted th at, according to calculations by the statistical offices of the United Nations, the productivity of labour in m anufacturing industry has grown more slowly in under-developed countries than it has in the rest of the world. Thus, between 1955 and 1970 productivity grew at the rate of 3 per cent per annum in the non-com m unist lessdeveloped countries (3 per cent for Asia and 2.5 per cent for Latin America) com pared to 4 per cent in western developed countries and 6 per cent in com m unist developed countries. 63
64
A . R a te s o f g ro w th in m a n u fa ctu rin g in d u stry
In this section the pace o f growth in m anufacturing industry will be examined. The relevant facts, while not com plete (m any under developed countries possess no indices o f production in m anu facturing industry), are nonetheless sufficient to establish a valid diagnosis o f the speed of growth in Third World countries in recent years. But, with few exceptions where figures go back to 1937/39, indices for individual countries are totally non-existent for earlier years. To put alongside these indices for countries taken individually we also have, thanks to the U.N. statistical office, an index of production for groups of countries, as well as one for all under developed countries taken together. It should, however, be noted th at mainland China and the other com munist countries are excluded from this index, as they are from most o f the aggregates calculated by the U.N. The main reason for such an obvious omission is the lack of statistical material. Because an index covering the whole of the under-developed non-comm unist world is available, the analysis will be made on two levels. First o f all it will be made for all under-developed countries, and there will be at the same time an attem pt to fit what we know about the pace o f growth into the larger historical context. Then changes taking place in different countries will be examined.
1. A L L N O N - C O M M U N I S T U N D E R D E V E L O P E D C O U N T R I E S
(a) The facts The U.N. statistical offices index number of industrial production for all under-developed, non-com m unist countries is an arithm etical average of the indices for different countries by sector o f activity, weighted according to the value added. However, because for many countries, especially in Africa, index numbers for m anufacturing production do not exist, the index is incom plete from a geographical point of view. Nevertheless as it includes a num ber o f countries, such as India, Pakistan and the big Latin American countries, with a large population, it is an acceptably faithful reflection of changes in under-developed countries taken as a whole. Moreover, I have tested the accuracy of the index by comparing it with the changes in energy consum ption between 1950 and 1963 in non-comm unist, Third World countries. The United Nations in its estim ates for geographical regions provides more comprehensive figures for energy consum ption than for industrial production. As is well known, the connexion
65
between industrial production and energy consum ption is close, although energy consum ption usually grows m ore rapidly giving a difference of about 5 per cent between the curves o f the two series.1 As, therefore, energy consum ption in under-developed countries between 1950 and 1963 rose by 7.1 per cent per annum, allowing for the above correction, the annual growth in m anufacturing industry was 6.7 per cent. The U.N. index actually shows a rate o f 6.9 per cent. This difference can attributed to the U.N. index om itting the African countries, because if we look at changes in energy consum p tion in all African under-developed countries, the conclusion is that these countries experienced a smaller growth in this field: i.e. not m ore than 4.6 per cent per annum. Table 17 Index and rate o f growth o f output o f manufacturing industry in the less-developed countries (excluding com m unist countries)
Index o f output A nnual rate o f grow th %
per
Year 1938 1948 1950 1953 1958 1960 1965 1966 1967 1968 1969 1970 1963 = 100 29 40 44 49 73 86 118 123 128 138 149 158 Period 1 9 3 8 -1 9 5 0 1 9 5 0 -1 9 6 0 1 9 6 0 -1 9 7 0 Total 3.5 6.9 6.3 capita 1.8 4.5 3.6
1 9 3 8 -1 9 6 0 1 9 3 8 -1 9 7 0 1 9 5 0 -1 9 7 0
4.9 5.3 6.6
3.0 3.1 4.1
Sources: Derived from The G row th o f World In d u stry 19 3 8 1961 (U .N ., 1963); and M o n th ly B ulletin o f Sta tistics (U .N .). F o r recent figures M ay 1973 issue.
The index of industrial production for all under-developed countries is, however, only calculated for 1938, 1948, and all subsequent years. It is thus not possible to indicate growth in this sector before 1938 by this means. Table 17 shows such indices as are available. We are immediately struck by the very rapid growth which took place after 1953. Thus, between 1953 and 1970, the annual rate of growth was 7.2 per cent (betw een 1938 and 1950 it was only 3.5 per cent). Even allowing for the large growth of population, the
66
The Economic Developme?it o f the Third World since 1900
annual rate is still high, since it rose for the period 195070 by more than 4 per cent per capita. Looking at changes over shorter periods, the following phases are apparent: 1938 48. This period m ust be considered as a whole since figures for interm ediate years are not available. Only a m odest developm ent of m anufacturing industry took place during these years, the annual rate of growth being 3.5 per cent, and the per capita rate being 1.8 per cent. Between 194853 the pace of growth accelerated 4.4 per cent or about 2.2 per cent per capita. It was, however, irregular, ranging from a maximum of 7.3 to a minimum of 1.5 per cent.2 Between 195356 the rate of growth was very high, reaching as much as 9.5 per cent (or 7.3 per cent per capita). Between 195670 growth was again high, even though the rate was below th at of 195356: 6.5 per cent, or about 4 per cent per capita for the whole period. But the pace was again irregular.
(b) Comparison with rates o f growth in developed countries The positive trend o f growth rates in m anufacturing industry in all the under-developed, non-comm unist countries is obvious from Table 18 where these are juxtaposed with data from selected developed countries for the m odern period as well as for take o f f periods in the eighteenth and nineteenth centuries. Indeed, it m ay be accepted th at a per capita growth rate o f 4.0 per cent in m anufacturing is fairly close to that attained by the western countries since the end of the last war and during times between the wars when econom ic conditions were favourable. By contrast rates in the eighteenth and nineteenth centuries were lower, a t about 2 to 3 per cent. This may appear surprising because there is a widespread belief that growth was more rapid during the earlier phases of industrialization.3 It should be pointed out, however, th at a distinctly higher rate of growth has occurred in eastern European countries. This has been about 10 per cent per capita, but falling in recent years to 8 to 9 per cent. The same is also true in some other countries with uncentral ized economies which began to industrialize towards the end o f the nineteenth century. This suggests th at industrial growth may be m ore rapid in countries where industrialization began later. The point is sufficiently im portant to w arrant further exam ination before con tinuing the main exposition.
Manufacturing Industry Table 18 Annual rate o f growth o f manufacturing industry

O utput manufacturing industry T otal non-com m unist, less-developed countries 1 9 3 8 -1 9 5 0 1 9 5 0 -1 9 6 0 1 9 6 0 -1 9 7 0 Total non-com m unist, developed countries 1 9 3 8 -1 9 5 0 1 9 5 0 -1 9 6 0 1 9 6 0 -1 9 7 0 Selected developed countries Great Britain 1700/04 -1760/641 1 7 6 0 /6 4 - 1 8 00/041 1800/04185 0/5 4 1 1850/541900/04* 1900 /0 4 -1 9 1 0/14* 1919-1929* 1929-1938* United States 1 8 6 0 /6 4 -1 8 8 0 /8 4 1 8 8 0 /8 4 -1 9 0 0 /0 4 1 9 0 0 /0 4 -1 9 1 0 /1 4 1 9 1 9 -1 9 2 9 1 9 2 9 -1 9 3 8 France 1781/901803/12 1803/121835/441 1835/441865/741 1865/741895/0 4 1 1 8 9 5 /0 4 -1 9 0 5 /1 31 1920/241 9 2 5 /34l 1925/3419 35/381
67
Population
O utput m anufacturing industry per capita
3.8% 6.9% 6.3%
1.7% 2.3% 2.6%
2.1% 4.5% 3.6%
4.5% 5.0% 5.6%
0.7% 1.2% 1.1%
3.8% 3.8% 4.4%
0.8% 2.4% 3.6% 2.2% 1.8% 2.9% 2.7% 6.9% 5.8% 4.7% 4.3% -0 .1 % 0.5% 2.7% 1.9% 1.7% 2.0% 4.4% 0.5%
0.4% 0.8% 1.4% 0.8% 0.8% 0.4% 0.4% 2.6% 2.1% 1.9% 1.5% 0.7% 0.6% 0.5% 0.3% 0.2% 0.2% 0.6% 0.1%
0.4% 1.6% 2.2% 1.4% 1.0% 2.5% 2.3% 4.2% 3.6% 2.8% 2.8% -0 .8 % -0 .1 % 2.2% 1.6% 1.5% 1.8% 3.8% 0.4%
1 Including extractive industry. Sources: ALL DEV ELO PED AND LESS-D EVELOPED C O U N TR IES derived from U.N. statistics. G R EA T BRITAIN derived from W. H offm anns Index ( British In d u stry 17 0 0 1950, O xford, 1955), and K. S. Lom as. P ro d u ctio n and P roductivity m ovem ents in the U nited K ingdom since 1900, in J. R o y. S ta t. S o c ., 1959 (cited in B. R . M itchell, A bstract o f British H istorical S tatistics, C am bridge, 1962); U N ITED STATE'S calculated from several indices tak en from H istorical Statistics o f th e U nited S ta tes, New Y ork, I9 6 0 ; FRANCE derived fro m T . J. M arkovitch, Lindustrie franâisc de 1789 a 1964. Cahiers de V l.SM .A ., 163, Ju ly 1965.
68
(c) Period o f industrialization and rate o f growth o f manufacturing industry Table 19 gives us at least a partial answer to a whole series of interesting questions. The table gives the annual rates o f growth for the periods 192857 and 1958 70 which I have been able to calculate from the index numbers o f m anufacturing industry. The character o f these two periods in very different, b u t the starting and finishing points occur at m om ents when conditions were rather similar. Grow th rates have been calculated for two groups of countries, the first comprising those in which industrialization began early, i.e. approxim ately before the second half o f the nineteenth century, and the second comprising several countries which came late to th e process late meaning after the second half of the nineteenth century. It should be observed th at the under-developed countries have been excluded from this group but not from the table. The countries have been divided according to the predom inating economic system, and the order in which they are listed approxi mates roughly to the order in which the take o ff occurred.4 The following are the most im portant points to emerge. 1. The growth rate in the second period (1958 70) is distinctly higher for the more recently industrialized countries than it is for the countries which developed earlier, irrespective o f region or prevailing econom ic system. 2. Rapid growth in the newly industrialized countries is, however, a very recent phenom enon, since for the period 192857 (and for earlier periods not included in the table) their growth rates are very close to those o f the older industrialized countries, with the notable exception of the U.S.S.R., the only country in the period 1928 45 to have a centralized econom y.5 Grow th rates in this period for the older industrialized countries are much lower than those achieved in 1958 70. It should be noticed that, although the term inal years occur at m om ents when conditions were rather similar, the period 195870, unlike the period 192857, did not include any major econom ic crisis nor any international war. 3. By comparing the results o f Table 19 with those o f Table 18 in which some growth rates achieved in the course of the nineteenth century are shown we can see that growth rates in m anufacturing industry in the countries coming late to industrialization are higher for the last tw enty years (but for no other period) than those of the older industrialized countries enjoying a similar level o f development. 4. The recent developm ent of under-developed countries would thus seem to proceed at the same pace as that o f recently
ial com
or
69
Table 19 Comparison o f annual rates o f growth in manufacturing output according to stage o f industrialization and economic system
1 9 2 8 -1 9 5 7 O utput manufact. Popula industry tion hariy industrialized country: (a) m arket econom y Great Britain United States Belgium France West Germany Sweden (b) centralized econom y East Germany Czechoslovakia R ecently industrialized country: (a) m arket econom y Italy Spain Japan Greece Portugal Ireland (b) centralized econom y U.S.S.R. Bulgaria Hungary Poland Roumania Yugoslavia (c) non-communist
le s s -d e v e lo p e d
1 9 5 8 -1 9 7 0 O utput per capita O utput manufact. Popula industry tion O utput per capita
2.4 3.9 1.1* 1.9* 2.8* 3.8*
0.4 1.2 0.4 0.3 0.3 0.6
2.0 2.7 0.7 1.6 2.5 3.2 3.8
3.6 5.3 6.1 6.0 6.6 6.3 6.8 6.8
0.6 1.3 0.4 1.0 1.1 0.7 0.0 0.6
3.0 3.9 5.7 5.0 5.4 6.2 6.8 6.2
3.8*
0.0
2.9* 4 .4 1 3.8 4.0 3.54 3.52 12.0*
0.6 0.9 1.3 0.9 0.8 -0 .1 1.1
2.3 3.5 2.5 3.1 2.7 3.6 10.9

8.1 10.4 15.9 7.6 8.1 6.6 9.0* 12.5 8.1 8.8 13.1 9.7
0.7 0.9 1.0 0.7 0.8 0.3 1.3 0.8 0.4 1.1 1.0 1.1
7.3 9.4 14.8 6.9 7.2 6.3 7.6 11.6 7.7 7.6 12.0 8.5
countries
4.7
1.8
2.9
6.8
2.5
4.2
T o ta l industrial o u tp u t 119 4 0 1957 2 1 9 2 9 -1 9 5 7 3 1 9 3 8 -1 9 5 7 4 Based on gross dom estic p ro d u ct generated b y m anufacturing industry and in c o n sta n t prices for th e period 1 9 4 8 -1 9 5 7 . Sources: Derived from Statistical Y earbook (U .N .); M o n th ly Bulletin o f Statistics (U .N .); and statistical abstract o f several countries.
70
industrialized countries, as far as growth rate is concerned. However, it seems th at instead o f an acceleration in th e pace o f grow th, as happened in countries beginning their industrialization at the end of the nineteenth century, we m ay expect a slowing down. The reason for this is stagnation in th e hom e m arket due to the slow growth, or even decline, in the purchasing pow er of the agricultural population, which, accounts at the present time for 70 per cent of the total active population. The relative unim portance o f industrial o u tp u t has prevented the expected factors from operating, because domestic production up to now has been a substitute for im ports rather than an increase in local consum ption, as is confirmed by various national studies6 and by the figures for total im ports in different regions. In Chapter 5, on foreign trade, we shall see how slowly im ports of m anufactured goods other than capital equipm ent have expanded. However, we m ust, as always, make allowances for regional differ ences as well as for changes at different periods. Placing the growth rates of im ports of m anufactured goods by volume (excluding capital equipm ent and chemical products) along side those of production and of estim ates for local consum ption of these com m odities during the period 1955 70, it appears that im port substitution grew more rapidly in Latin America than in Asia (considered as a whole). Within the period the growth o f im ports by volume expanded very rapidly betw een 195357 (8 per cent); stagnated between 195763; and finally expanded again between 196370 (9 per cent). Although such considerations as the availa bility of foreign exchange m ust have played an im portant part it is probable, nonetheless, th at im port substitution occurred mainly between th e years 195657 and 1963 64. According to my estimates for all th e under-developed, non-comm unist countries the relationship between the value of im ports f.o.b. of m anufactured goods (I.S.I.C. 6 and 8, i.e. goods other than capital equipm ent and chemicals) and the value added in m anufacture changed as follows: 1953 1955/57 37.5% 38.0% 1961/63 1968/70 22.0% 24.0%
These percentages are clearly not a real index of the substitution of dom estic production for im ports, b u t they enable us, nonetheless, to plot the trend with a fair degree o f accuracy. When discussing the problem, we m ust obviously allow, as always, for regional and national differences. Taken by itself this im port substitution is clearly a positive factor, b u t if it is not accom panied by an enlargement of the local m arket the growth o f dom estic production will be held back by lagging demand. Given th a t rural population
ial com direitos autors
71
represents on average 70 per cent of total population, industrial expansion cannot keep up its current pace for long w ithout an increase in rural consum ption. If the decline in agricultural pro ductivity (and, consequently, rural incomes) posited in Chapter 2 continues, we m ust expect a consequent slowing down in industrial growth, especially in those countries with a predom inantly rural population. I shall return to this problem later in the chapter, and also, partly, in the next section where differences in the growth rates of various sectors will be shown to confirm my hypothesis. (d) Changes in manufacturing industry in all under-developed countries by sectoral activity Table 20 sets out changes in the index of industrial o u tp u t of the under-developed, non-comm unist countries since 1938 as seen in some of the main sectors of activity. It appears that heavy industry, and especially steel-making, has been the main cause o f the immense growth in m anufacturing industry. Between 1938 and 1970, rates of growth in heavy industry are more than 70 per cent above those of light industry 7.4 per cent and 4.4 per cent per annum respect ively. Table 20 Index o f output in manufacturing industry o f all non-communist _________ less-developed countries (1963 = 100)________________________
ALL 1938 L 'jh t industry Heavy industry Total m anufacturing industry of which: Food Textiles Clothing footw ear Wood products Paper, printing Chemicals Non metallic minerals Basic metals Metal products 37 18 29 36 44 53 31 11 21 18 24 14 1953 60 39 50 63 63 65 47 35 41 49 42 33 1958 82 67 73 83 84 80 70 69 70 75 59 61 A ll 144 179 158
146
1970 Latin A m erica 1 146 185 163 144 139 144 134 161 178 174 164 203 E. and S. E. Asia 141 175 153
148
140 133 149 157 174 171 159 193
138 123 166 154 172 171 157 182
11969 Sources: The G row th o f W orld In d u stry 19 3 8 1961 (U .N ., 1965), and M o n th ly B ulletin o f Statistics. (U.N.) For recen t figures May 1973 issue.
72
The gap is less evident for m ore recent years, although it still exists. The high rates of growth are due partly to the almost total absence o f heavy industry in these countries before the war and generally before their political independence, and partly to the priority accorded to these sectors, with more or less justification, by those responsible for such decisions. The m yth of the steel works as a symbol of independence and economic progress was widespread amongst politicians and even economists in these countries. The main cause o f the enormous expansion of steel-making is the creation of centres of production in Asia. The conversion of other metals, although fairly rapid, has risen more slowly. There has also been a great expansion of papermaking, although, again, it is true th a t the original level was extrem ely low, production in under-developed countries in 1958 being only 4.4 per cent of world total. The creation and extension of oil refineries are responsible for the relatively high rates of growth in the chemical sector. By contrast, the slow extension of light consumer industries, e.g. food and textiles, m ust be emphasized, especially in Latin America where this sector is hardly keeping pace with the growth in population. This is not unusual for these industries normally grow more slowly than other sectors, but even so the rates attained in Latin America are exceptionally low, especially when set alongside population growth. These low rates of growth in sectors where the starting point is relatively higher than in other sectors, go some way to confirm the hypothesis propounded above th at stagnation or decline in agri cultural productivity in the under-developed countries m ust lead, in the medium term , to a decrease in industrial growth. And this is especially true because the practical limits in substituting domestic production for im ports have, in m any cases, almost been reached.
II R A T E O F G R O W T H COUNTRY IN M A N U F A C T U R I N G I N D U S T R Y BY
We shall now examine changes in the m anufacturing industries of different countries, including mainland China. Recent figures for the latter country, especially where growth rates are concerned, are not only exiguous but also contradictory. Here, as with all statistical material connected with China, we have two sets of data at our disposal: the first compiled from official figures and the second from estimates by w estern experts, who are usually Chinese refugees working in the United States. These experts d o u b t the credibility of the official figures which they consider too optim istic, and attem pt
ial com
73
to re-evaluate changes on the basis of figures which are thought to be m ore realistic. We shall examine both series briefly,7 b u t before doing so it should be noted th at the index is one o f total industrial production, i.e. it includes extractive industry and energy product ion, but excludes construction. As, however, the export rate of extractive products is very low, their inclusion does not distort the results and we may treat the index o f Chinese industrial production as very nearly com parable to the indices of m anufacturing in other Third World countries. So far as the official figures are concerned it should be recalled th at the Chinese authorities stopped issuing com plete data in I960: such inform ation as they have provided is fragm entary and applies only to certain sectors. But the general impression for 1962 and 1963 is th at production continued to grow, although total rates were not provided. For 1964 a total rate o f 15 per cent was given, and for 1965 20 per cent. M atters are further com plicated by the Cultural Revolution. According to the survey by Field,8 based on very scanty inform ation, official statem ents imply th at average growth between 1957 and 1968 was about 9 to 10 per cent per annum. From the beginning of 1971 the quantity o f statistical inform ation p u t out by the Chinese government and officials m arkedly increased. This is probably because o f the re-opening o f China to the outside world accompanied by her admission to the United Nations, as well as because the end o f 1970 m arked the end o f the third five-year plan, the quantitative aims o f which, it will be noted, were never made public. In this connexion the dates o f the previous five-year plans should be m entioned. The first plan covered the years 195357, the second the years 1958 62 and the third only began after the econom ic re-organization resulting from the disturbances of 1961 62 and the break with the U.S.S.R. in 1960 in 1966 and ended in 1970. The fourth plan will cover the years 197175. These plans, of course, never had the centralizing and coercive force o f the Russian five-year plans. Early in 1971, for the first time in ten years, China itself gave out some statistics o f quantitative production, first in an interview which Chou en Lai granted to E. Snow,9 but mainly by means o f a study in New China which was com pleted by the traditional New Year editorial article.10 The overall impression from these statem ents is one o f industrial growth o f the order o f 18 to 20 per cent per annum betw een 1969 and 1971. If, from these incom plete figures, we may hazard an estim ate o f the approxim ate level o f industrial production in 1970, this might be (taking 1963 = 100) about 200, as against 140
74
in 1959, 61 in 1957, 15 in 1950 and 11 in 1949. It is not, however, impossible th at the authorities might claim an even higher level for 1970. To tu rn now to the western estimates. Because com plete official statistics after 1960 are lacking, m ost western experts term inate their series at 1959. For the period 195059 I have made a survey o f the different estimates available,1 1 which give for the years 1950 to 1959 an average annual rate o f growth for industry of 15 per cent. For the years 1960 71 the material is even m ore fragmentary. The only western estimate for a long period at present available is, so far as I know, Fields,12 which puts the level o f industrial production in 1968 hardly above the level o f 1959 at m ost and 11 per cent below th at level at lowest, i.e. if we settle for the average, about 5 per cent below the level of 1959. It will be rem em bered that the latter date was the peak of the G reat Leap Forward. On the basis of this material, supported by other figures for the years 1968 to 1970,13 western estimates place industrial growth between 1959 (the peak of the Great Leap Forward) and 1970 at the annual rate o f about one per cent, rising to 5 per cent if we take into consideration the years 1957 to 1970 which form a more significant period. It m ust be emphasized th a t this figure is no more than an indication and we m ust await the time when the Chinese authorities see fit to divulge fuller statistics before one can construct a more accurate estimate. When it comes to a choice between the m erits o f official figures and western estimates, we m ust accept th at such a choice is, o f course, no m ore than a very subjective one for there are as m any opinions in this field as there are economists working on the problem. My own view is th at the official figures certainly over-estimate growth and th at some western estimates under-estim ate it. The truth probably lies fairly close to the figures th at we retained as an average of western estimates, w ith the probability th at this is a slight under estimate. From the data on industrial growth in the under-developed countries presented in Table 21 the following conclusions may be drawn: 1. We find here a good illustration o f the example given in courses o f statistics about th e uncertain nature o f index numbers which relate to countries where th e level o f production is extrem ely low. An extrem ely high level of expansion may occur during a short period, as may be seen here to have been the case in Zaire, Venezuela, South Korea, Pakistan and Taiwan. In some of these countries industrial expansion, having reached very high growth
75
rates, then falls. It must be remembered that this statistical distortion applies only to countries with a very low initial level, to small countries with an average level or, obviously, to countries which combine both elements, because, in such a situation, the establish m ent o f a few m odern units of production leads to a very great increase in the total production o f the sector. China and Turkey are
Table 21 Indices (1963 = 100) and rates o f growth o f manufacturing industry

Indices 1963 = 100 1938 AFRICA Algeria Kenya Morocco S. Rhodesia Senegal Zaire AMERICA Argentina Brazil Chile Guatemala Mexico Peru Venezuela A ll Latin America ASIA China (Taiwan) China (official)8 China (western) S. Korea India Pakistan Philippines A ll Asia 9 MIDDLE EAST Egypt Syria Turkey 1950 1957 1970 Annual Rate o f Variation 1938 1950 1950 1957 1957 1970
51
28
53
8 38 15 31
47 80 34 47 61 44 36 23 51 8 15 32 344 40 11 444 37
86 77 72 80 842 105 87 57 70 78 66 63 63 75 57 61 82 56 62 44 69 67
168 135 163 115
5.6

44.0
7.8 4.6 12.3 1.6 7.5 5.7 3.6 6.1 8.2 15.3 5.5
5.1 4.9 5.6 2.6
24 28
172 164 121 129 189 164 145 163 342 190 190 390 135 186 141 153
4.2 7.0 3.8 5.1 5.1 0.8

5.4 8.5 4.3 3.9 8.4 7.6 6.6 6.1 14.8 9.1 6.7 16.1 6.2 11.7 5.6 6.6
361 30
15.9 22.0 14.0 10.5 6.6 22.3 3.4 8.8
1.6
16 29
33 712 69
123 244 176s
5.4
13.1
11.6 10.8 8.8
1 1939 2 1958 31958 = 100 9C om m unist countries, Japan and Sources: Derived from Statistical an d M o n th ly B ulletin o f Statistics
4 1953 s 1966 61967 ?1 9 6 9 8 G eneral index o f industrial o u tp u t Israel excluded Y earbook and M o n th ly B ulletin o f Statistics (U .N .); and Statistical Y earbook o f th e countries concerned.
76
not included since in their case the starting level is n o t a dom inant factor. Egypt is in a situation halfway between the two. 2. Where larger groups of countries are concerned, there are the following divergences: from 1938 to 1950 growth rates have clearly been higher in Latin America than in other groups. But between 1950 and 1970 the situation is reversed, Latin America then experiencing a rate which is clearly lower than the average in other countries. The Latin American rate is strongly influenced by th at of Argentina which occupies a leading place in the field o f m anufacturing. The recent very rapid growth o f Brazil has not yet influenced the overall average o f Latin America. 3. Where individual countries are concerned the following facts should be pointed out. In Latin America, the m ost developed country, Argentina, has since the war gone through a growth crisis. Chile is, to some extent, in the same position: her growth rate is lower than th at of other countries, although her level o f developm ent is higher, as we shall see later. The rate o f expansion in other countries (and especially Brazil) has been very rapid, around or above 8 per cent per annum, which, while allowing for a very high population growth, still gives an increase o f about 5 per cent. The m ost im portant and interesting question in Asia is the contrast between China and other countries, especially India, since the starting point of these two countries was close. When dealing with China the tw o sets o f figures the official and the western estim ates m ust be remembered. However, even if we take the western estimates, it seems th a t the Chinese growth rate has been definitely higher than th at in India. The respective figures for 195057 are 14.0 per cent and 6.6 per cent and for 1957 70 5.0 per cent and 6.2 per cent. For the whole period 1950 70 the figures for China are 8.5 per cent and for India 6.0 per cent. In terms o f the more significant per capita rates o f growth the gap is even wider since Indias population increased between 1950 and 1970 at an annual rate of 2.2 per cent while the comparable rate for China was probably below 2.0 per cent. If we take the official data for China, the figures for total growth are respectively 14.0 per cent and 6.0 per cent. In the confrontation betw een these tw o Asian giants all will ultim ately depend on future events in China. If industrial growth recovers the pace o f th e years 195057 th e advantage will incontestably lie w ith China. On the other hand, if the conditions th a t prevailed after the Great Leap Forward and during the Cultural Revolution recur the gap between the tw o countries will shrink. To judge from recent inform ation the first alternative appears the more likely.
11
Pakistan, South Korea and Taiwan have experienced a growth rate close to, or higher than, th at of China, but, as I have already pointed out, developm ent in these countries started from a very low level.14 The som ewhat laggard pace of the Philippines should be pointed out, at a time when industrial developm ent in this country seems to be at a slightly higher level than in other Asian countries, China excepted. This is one more example of the problem o f the relationship between the level of developm ent and the pace of growth, to which attention has already been drawn in connexion with Argentina and Chile. Many of the factors involved (statistical bias due to low levels of production, effects of domestic demand, etc.) do not allow us to conclude positively that, having achieved a certain stage o f development, under-developed countries experience as a result o f this factor alone a slackening in industrial growth. In my view the essential factors here are the limits o f im port substitution and a weakening dom estic demand arising from insuff icient growth in agricultural productivity. This brings us to the problem of die relationship between industrial and agricultural growth, which will be the subject o f a later section in this chapter. But before passing to it we m ust consider the problem of the level of development achieved by m anufacturing industry. B. Level o f developm ent in m anufacturing industry In this section I shall attem pt to define the level o f development achieved by m anufacturing industry in th e individual under developed countries. Where agriculture is concerned we have seen th at it is possible to measure, by means o f productivity, both the changes taking place and the levels of developm ent achieved in this sector, and also to make a comparison with developed countries in the present day and in the eighteenth and nineteenth centuries. This approach is feasible because we can reduce production in this sector to a common physical unit w ithout relying on an expression of its m onetary value. U nfortunately we cannot apply the same m ethod to m anufacturing industry. We shall therefore attem pt to attain our objective by means of certain data on per capita o u tp u t and consum ption. For this purpose the following indicators have been used*. I. Two general indicators based on the relative position o f m anufacturing industry: (a) em ploym ent in m anufacturing industry; (b) gross product generated by m anufacturing industry.
ial com
or
78
Figures for China will not be included in either o f the tw o indices because the data is too unreliable for comparisons to be valid, b u t I have tried to supply retrospective data for these two indicators in the case o f the other countries. II. Four particular indicators referring to per capita data: (a) energy consum ption (b) steel o u tp u t (c) industrial consum ption of cotton (d) tonnage moved by railways. It goes w ithout saying th a t none of these indicators alone would be sufficient to reflect the level o f developm ent achieved in m anufacturing, their selection having depended on available statistics. However, all six indicators taken together perm it a reasonably accurate diagnosis. Comparing the Third World with the developed countries at various stages in their industrialization is a very hazardous business, and the conclusions reached below should n o t be regarded as too precise.
I. G E N E R A L INDICATORS BASED ON THE RELATIVE
POSITION OF M A N U F A C T U R I N G INDUSTRY
For the first set of indicators, especially the relative level of em ploym ent, I have tried to assemble as long a series as possible, so th at it is possible to make a diagnosis o f economic conditions. (a) Employment in manufacturing industry Ever since Colin Clark and Fourastie published their studies, the structure o f the working population has rightly been accepted as reflecting a co u n try s level of development. Although classifying the active population into three sections is in some respects unsatisfac tory, it does reflect fairly accurately changes resulting from the process o f development. A fortiori , therefore, it is obvious that, in so far as it is an indicator o f the level o f m anufacturing development, the proportion of the working population so em ployed can provide useful inform ation about the early phases o f development. The indicator based on em ploym ent data has the added advantage of providing longer and more extensive series for some countries than one which relates to o u tp u t only. So far as the under-developed countries as a whole are concerned, changes in the active population engaged in m anufacturing industry will be discussed in Chapter 9. But some details about individual countries are set out here.
79
Table 22 Percentage o f the male working population occupied in manufacturing industry

1 9 3 0 -1 9 4 6 Date AFRICA Algeria Morocco Tunisia Zaire AMERICA Argentina Brazil Chile Costa Rica Mexico Peru Venezuela ASIA Ceylon India Pakistan India & Pakistan Indonesia Malaysia Philippines Thailand MIDDLE EAST Egypt Syria Turkey % about 1950 Date % about 1960 Date % about 1970 Date %
1948 1952
5.2 6.2
1952
5.7
1954 1960 1956 1955 1960 1960 1960 1963 1960 1961 1961
5.6 7.0 6.8 5.7
1966
6.0
1966

7.7
1940 1940
6.5 14.8
1947 1950 1952 1950

20.3 8.5 17.1 10.1

1930 1940 1941
10.3 10.1 7.0
1950
9.4
25.4 8.1 16.5 10.3 14.1 12.1 11.2
1970 1971 1970 1970
11.0 23.4 15.3 16.4
1971
19.2
1946

7.3

1931
10.1
195 3 1951 1951 1951
9.4 8.2 6.2 7.8
1947 1939 1937
5.8 6.0 2.1
1957 1948 1954
7.0 5.7 2.9
1964 1961 1961 1961 1961 1962 1961 1960
9.2 10.1 8.4 9.8 4.9 6.7 7.1 4.1
1971

10.0

1965
4.7
1970 1970
7.6 3.7 13.2 10.3 10.5
1937 1935
8.1 8.2
1947 1950
10.3 9.9
1960 1960 1960
9.7 9.6 11.0
1966 1969 1965
N ote: T he sign represents an estim ate based on th e global figure for em ploym ent in m anufacturing industry plus co n stru ctio n and, occasionally, electricity, w ater, gas and san itary services; th e categories have been deduced by m eans o f an estim ate derived from earlier or later d ata. Sources.- Derived from The W orking P opulation and its S tru c tu re, Vol. I o f International H istorical Statistics, under th e d irectio n o f P. B airoch, b y T. D eldycke, H. G elders and J.-M. L im bor (New Y ork and Brussels, 1968); Y earbook o f L abour Statistics, 1973 (I.L .O ., G eneva, 1973).
Since the criteria defining an economically active person fluctuate more for wom en than for men betw een one census and another, the data for individual countries relates solely to the male working population. Table 22 is therefore confined to percentages of the male working population engaged in m anufacturing industry. These percentages show how unequal is the level o f developm ent achieved by m anufacturing industry in the various under-developed
80
countries, and further, how irregular is the rate of expansion in this sector. We find alm ost the same differences here as in our survey of the level of agricultural productivity. The relationship which exists between these tw o aspects of developm ent should lead us to expect this; unproductive agriculture cannot support much expansion of industry. As in the case of agriculture, the countries fall into three groups. The Afro-Asian countries, where em ploym ent in manufac turing, at less than 10 per cent of the whole working population, is very low (excluding China this makes an average in 1960 o f about 9 per cent); the Latin American countries taken together, in which em ploym ent in m anufacturing is slightly larger, the average being about 13 per cent; and, between these tw o groups, b u t nearer Asia, the Middle Eastern countries with an average of about 10 per cent. Within these groups there are some marked differences. In Latin America we find once more th at the proportion of workers in m anufacturing industry in Argentina and Chile is noticeably above the average in the other countries. In the case of Chile we m ust also take note of the fact th at an extrem ely high proportion, 4 per cent o f all workers, is em ployed in mining. In Asia, although the gap is not large, India has a higher proportion o f workers engaged in m anufacturing than other countries in th a t continent as well as in m ost Middle Eastern countries. Table 23 allows us to compare the present level o f m anufacturing em ploym ent in the under-developed countries with th a t o f the developed countries at various stages in their industrialization. It should be noted th at because there are no statistics the precise percentage o f the working population em ployed in m anufacturing during the very early phases o f industrialization in the developed countries cannot be determ ined. Nevertheless, it can be assumed from extrapolations and some fragm entary figures, th at at the m om ent of take o f f the percentage was between 10 and 12 per cent. The low level attained in under-developed countries in similar occupations today is thus all to o evident. However, the differences between the tw o types of economies m ust n o t be forgotten: the productivity of the 10 per cent o f workers engaged in m anufacturing industry in present-day under-developed countries is probably m uch higher than th at of an identical percentage in the traditional western economies before the industrial revolution, since, in the under-developed countries, a large proportion o f the 10 per cent is employed in factories with a productivity similar, if n o t equal, to th at of m odern developed countries. This point will be raised later in discussing the other indicator when I shall attem pt to assign the
A i X a
Manufacturing Industry Table 23 Percentage o f the male working population employed in manufacturing industry in some developed countries , 1841 1970
G R E A T B R IT A IN Dates 1841 1861 1891 1921 1951 1961 1966 Percentage 35.7 35.9 34.2 34.0 37.2 37.8 37.0 FRANCE Dates
_
81
UNITED ST A T E S Dates 1850 1870 1890 1920 1950 1960 1970 Percentage 16.4* 18.0 20.1 29.4 26.3 29.0 29.1
Percentage
_
1856 1896 1921 1954 1962 1968
20.6 23.1 25.3 27.3 28.3 29.1
T o tal w orking population Sources: Derived from The W orking P opulation and its S tru c tu re , op. cit.; Y earbook o f Labour Statistics 1972 (I.L .O ., Geneva, 1972).
relative levels o f development reached in the Third World and the industrialized countries. I am not going to discuss the changes in this index at length because the problem will be taken up again in the chapter on the active population and em ploym ent. Here it is only necessary to point out th at before 194050 the proportion of the male population employed in m anufacturing industry was generally declining and that progress since then has been far from uniform in all countries. (b) The share o f Gross Domestic Product generated by manufacturing This indicator has the advantage over the one based on em ploym ent in th at it embodies the notion of productivity. Indeed, assuming m ethods of estimating to be correct, the share o f m anufacturing in the Gross Domestic Product represents the relative share of this sector in final production, for the latter is roughly the sum of e m p lo y m e n t and p r o d u c tiv ity in th e se cto r . N e v erth eless, as it is here a m atter o f relative proportions (difficulties o f conversion making comparisons between absolute values very uncertain), productivity in other sectors is also relevant. Thus, a country whose industrial sector is limited and productivity low, but whose agriculture is even less productive, could, because of this relationship, register in its Gross Domestic Product a higher relative share of m anufacturing industry than could an industrially advanced country with a highly productive agriculture.15
82
The Economic Development o f the Third World since 1900 Table 24 Gross domestic product generated b y manufacturing (percentages in relation to total)
1 9 4 7 -1 9 4 9 AFRICA Morocco Nigeria Tanzania Tunisia3 AMERICA Argentina Brazil Chile5 Colombia Equador Mexico Peru Venezuela ASIA Ceylon India6 Indonesia Pakistan Philippines6 Thailand MIDDLE EAST Iran Turkey 1 9 5 9 -1 9 6 1 1 9 6 8 -1 9 7 0
13.5 5.0 3.2 14.02 32.3 23.64 24.4 17.0 15.42 26.31 16.4 12.91 5.2 14.22 12.8 9.7 19.7 11.0 25.3 12.0
13.8 7.010 8.5 12.9 28.0 19.010 27.7 22.2 17.5 23.3 15.5 1 20.0 9.3 13.3 9.5 12.1 15.5 15.5 31.5 14.4
22.6 20.3 23.3 15.8 16.0 17.9 15.7 8.9s 4.2s 16.27'8
7.89/ s 11.3 10.4 10.6
1On th e basis o f co n stan t price 2 1 9 6 0 -6 1 3Excluding basic m etals 4 195960 5 1 9 5 0 - 5 2 6N et dom estic p ro d u ct for 1943 and 1961 in c lu d in g m ining and co n stru ctio n 8 19 4 8 49 i n c l u d in g electricity, gas and w ater 1 0 196769 Sources: Derived from data collected in: Yearbook o f N ational A c c o u n ts Statistics, 1961, 1964 and 1971 (U .N ., New Y ork); Sta tistics o f N ational Incom e and E xp en d itu res, 1954 (U .N .); and N ational A c c o u n ts o f Less D eveloped C ountries (O .E.C .D ., Paris, 1968).
In Table 24, because o f revisions in m ethods o f estimating, the figures for 1947 49 are not in every case strictly comparable with the other tw o series. But if these lim itations are accepted and especially where they refer to a possibly over-hasty comparison with corresponding percentages for developed countries it appears that the calculations made here confirm in general both the changes and the differences suggested by the indicator o f em ploym ent (a) Nevertheless, it is w orth underlining some o f the differences and attem pting to explain them.
83
In the Philippines 20 per cent o f the Gross Domestic Product is generated by m anufacturing industry, although this sector employs only 7 per cent of the working population. This situation is the result o f an unproductive agriculture combined with a relatively developed industry (relatively developed especially in comparison with th at o f India and Pakistan).16 The divergence between India and Pakistan cannot be explained in the same way. Although the m ethods o f estimation used for these tw o countries appear to be similar, it seems as if there m ust be a significant difference in the m ethods to account for the widely dissimilar results produced. A comparison between Tables 22 and 24 confirms that there has been a general increase in productivity, as the share o f m anufacturing industrys contribution in G.D.P. has grown faster than the percen tage o f the male working population employed in the industry (see my final remark in the introduction to this chapter).
II. SIMPLE I N D I C AT ORS
Table 25 brings together indicators which are rather different, but which express in varying degrees the level of development achieved by m anufacturing industry. The first (per capita energy con sum ption) is to my mind th e m ost significant, although it is strongly biased in favour o f certain countries, especially those producing oil, where the unlim ited availability o f the product encourages relatively high private consum ption. This is the case in Venezuela, and to some extent also in Iraq and Iran. The second indicator (gross steel production per capita) is the least significant being of less value than that giving the consum ption of cotton by industry and handicrafts. However, the reasonably good correlation between these tw o last series, and also between all four indicators, will be noticed. The fourth indicator (tonnage moved by rail) is more indirect. Yet, in spite of the distortions introduced by the difference in railway development, it seems to reflect the level o f industrialization fairly faithfully, except in so m e u n u su al cases. W hile each o f these indicators taken separately is not a sufficient measure o f the level of development, y et all four taken together provide a valuable yard stick. The discussion which follows will be conducted on two levels: we shall first consider the differences in the level of development reached by various under-developed countries and then try to establish an average for the various groups o f countries and relate it to the level achieved by developed countries.

i
84
Table 25 Supplementary indicators o f the level o f industrial development in the less-developed countries , 1970
ENERGY per capita consum ption in kgs o f coal equivalent COTTON per capita consum ption by industry and handicraft (kgs) TRANSPO RT per capita tonnage m oved b y rail (ton/km s)
C O U N TR Y AFRICA Algeria Ghana Madagascar Morocco Nigeria Senegal Tunisia Zaire AMERICA Argentina Brazil Chile Colombia Cuba Mexico Peru Venezuela ASIA China (official) China (western) Ceylon India Pakistan Indonesia Philippines Taiwan Thailand MIDDLE EAST Egypt Iraq Iran Syria Turkey
STEEL per capita output (kgs) 2

462 164 67 194 45 149 247 73 1,688 472 1,208 578 1,039 1,205 609 2,498
38
0.2 0.6 0.5 0.5 0.1 0.8 0.6 0.6 4.3 3.1 3.2 3.3 2.3 3.0 1.4 2.1
98 36 35 171 29 45 258 89 575 2801 259 55
80 58 62 11
78 14 89 23 22
466 43 1
526 156 191 98 111 279 925 245 268 597 939 48 3 487
10 1
21 1 7
2.1 0.2 2.0 3.3 0.3 0.1 9.3 0.1 0.6 0.1 0.2 0.3 5.1
29 223 66 32 3 187 65 100 241 96 16 159
31
1 1969 2 1968 Sources: EN ERG Y : Statistical Yearbook (U.N.). STEEL AND COTTON: Derived from U.N. d ata. TR A N SPO R T: Derived from Statistical Y earbook a n d M o n th ly Statistical Bulletin (U .N .).
85
(a) Levels reached by individual countries Once more the countries fall into the same three groups, with the Afro-Asian countries at the bottom of th e ladder, the LatinAmerican countries at th e top, and those o f the Middle East in between. We shall not dwell on the divergences within these three groups, since these are clearly shown in Table 25. The m ost advanced countries in Africa are those in the Maghreb. And once again in Latin America the high level attained by Argentina, and to a lesser degree by Chile, is obvious. Indeed, to apply the adjective under-developed to these two countries is hardly accurate. The only point over which we shall linger, by reason o f its implications, is th e level reached by China, since no inform ation about this country has been given for the two previous sets of indicators. If the level o f industrial developm ent achieved by China is set against th at of other Asian countries, it is clear th at about 1970 she easily surpassed the average, even by the lower western estimates. Her industrial level is clearly higher than Indias, notw ithstanding the fact th a t the starting levels o f 1938 49 were about the sam e.17 The divergence seems to justify those who attrib u te to China a higher rate o f industrial growth than to India. (b) Comparison with developed countries at different stages o f industrialization We shall now endeavour to make a comparison betw een the level reached by under-developed and developed countries a t various stages in their industrialization. With this end in view identical indices for six developed countries were calculated and these countries have been divided into tw o groups; one (including the United Kingdom, the United States and France) representing the older industrialized countries, and th e o th er (comprising th e U.S.S.R., Italy and Japan) representing the newly industrialized countries. The choice o f the tw o groups o f countries was dictated by the o p p ortunity thus offered of a m ore refined interpretation, based as it is bound to be, on a clearer definition o f the level of development. As far as the difference in the pace o f growth between the earlier and the more recently industrialized countries is concerned, the figures in Table 26 confirm w hat has been said before, nam ely th at until the post-war period one can see no obvious gap betw een the rates o f growth as a function of the date o f take o ff. The countries which began their industrialization in the second half o f the nineteenth century did not undergo, at least until after the 1939 45
ial com
86
Table 26 Indicators o f the level o f industrial development o f selected developed countries 17601970
ENERGY per capita consum ption in kgs o f coal equivalent COTTO N per capita consum ption b y industry 8c handicraft (kgs) TR A N SP O R T per capita tonnage m oved b y rail (tons/km s)
C O U N TR Y Early industrialized countries Great Britain 1760 1810 1860 1900 1929 1950 1960 1970 U nited States 1760 1810 1860 1900 1929 1950 1960 1970 France 1760 1810 1860 1900 1929 1950 1960 1970
STE E L per capita output (kgs)
600 2,600 4,200 4,110 4,420 4,920 5,360 250 500 3,900 6,570 7,510 8,010 11,140 200 500 1,300 2 ,4 2 0 3 2,030 2,400 3,790
3* 20* 130* 220* 210 330 470 500 2* 8* 25* 190 470 580 500 580 2* 4* 25* 65* 240 210 380 470
0.2 2 15 19 15 9 5 3 2 6 12 14 13 8 0.1 0.3 3 5 9 6 7 5
550s 660 710 580 480 1,0902 2,720 5,400 5,670 4,650 5,440
85 430 1,220 930 1,250 2,980
R ecently industrialized countries U.S.S.R. 1860 1900 1929 1950 1960 1970 Italy 1860 1900 1929 1950 1960 1970
250 400 1,300 2,850 4,450
5* 25* 35 150 300 480 2* 1* 53 51 170 320
0.5 2 34 5 6 7 0.2 4 6 4 5 4
75 3,300 7,020 10,270
200 610 630 1,190 2,680
80 290 210 320 340
Manufacturing Industry Table 26 ( continued)

ENERGY per capita consum ption in kgs o f coal equivalent 1860 1900 1929 1950 1960 1970
_
87
C O U N TR Y Japan
STE E L per capita output (kgs)

_
COTTON per capita consum ption b y industry & handicraft (kgs) I1 4 9 3 8 7
T R A N SP O R T per capita tonnage m oved by rail (tons/km s)

_
150 740 780 1,160 3,210
1* 18 58 240 900
30 200 400 570 610
*Pig iron *1850 2 1882 3 Including th e Saar 4 1913 5A ccording to G. S undbargs estim ate (Aperqus statistiques in te m a tio n a u x , S to ck h o lm , 1908), based on freight :raffic receipts. Sources: Pre-1929 d a ta derived from m aterial collected for m y b o o k , R evo lu tio n industrielle e t souslevelo p p em en t and m y article, Niveaux de developpem ent econom ique de 1810 a 19 1 0 ; d a ta fo r 1 9 2 9 - 7 0 ierived from Sta tistica l Y earbook and M o n th ly B ulletin o f Sta tistics (U .N .) and statistical yearbooks and bulletins o f th e respective countries.
war, a more rapid developm ent than those which took off* earlier. These levels o f developm ent will be compared to the average level of development in the main continental groups o f Third World countries: Africa, Asia, Latin America and the Middle East. Such comparisons can only be tentative, but I think the attem pt is of genuine interest. Judging from the four indicators compiled here and, in part, from the two preceding ones, we can conclude th at the African countries are at a lower level of industrialization than the western countries were when they were still in their traditional econom y stage. Present levels in Asian countries (excluding China) correspond to those in the developed nations in the first few decades o f industrialization, i.e. about 1770 90 in England, 180010 in France and the United States, and before 1900 in the more recently industrialized countries. Chinas progress since 1952 places her at present approxim ately 15 to 25 years ahead o f the average o f the other Asian countries if we take western estimates, and 30 to 40 years ahead if we take the official figures. In Latin America Argentina and Chile have achieved an industrial developm ent which corresponds to th at o f France in approxim ately 190010 or England 1850 60, while the other countries are another 20 to 30 years behind. The Middle East occupies an interm ediate position, as it does where agriculture is concerned; somewhat closer to Asia than to
88
Latin America, i.e. about the level o f France in 184050 or England, in 1 8 0 0 -1 0 . But, I repeat, these comparisons are only very approxim ate and cpnsequently m ust n o t be taken too literally, although the margin of error is unlikely to exceed tw enty years either way. Expressed in a tim e scale reflecting the rate o f growth of the nineteenth century, the older industrialized countries o f the West are m ore than two centuries ahead of Africa, about a hundred and fifty years ahead of Asia (an approxim ate average), fifty years ahead of Latin America and about a century ahead of the Middle East. But these comparisons do not take account of structural changes. C. Agriculture and m anufacturing industry I do not intend to discuss the part agriculture played in the western countries while they were going through the industrial revolution, since it is not germane to the present study.1 8 Suffice it to say that it was all-im portant in providing the initial impetus. Here I shall try to determ ine the present relationship betw een agriculture and industry in less-developed countries and com pare this with the equivalent situation in the older industrialized countries. The relationship will be discussed briefly under two heads: I The level of development in both sectors. II The connexion between changing conditions in both sectors.
I L E V E L S O F D E V E L O P M E N T IN A G R I C U L T U R E A N D
MANUFACTURING
The influence o f agriculture was not limited to ju st the initial take o ff period of industrialization, it also manifested itself in subsequent phases o f industrial development. This is illustrated by the close correlation, during the nineteenth century, between agricultural productivity and industrial developm ent in twelve o f the advanced countries for which I have calculated figures.1 9 If the coefficient o f correlation between the tw o sets of figures is com puted, the positive correlation is m ore than 0.9, and even m ore if agricultural develop m ent in one period is compared with th a t of industry in the following period. It is probable th a t calculations for other developed countries in the m odern period would produce a similar relationship. The situation in the under-developed countries is analagous. When we consider the larger geographical regions, the agricultural productivity ratings (see Chapter 2) are reproduced in the level o f industrial
Bahan cjengan hak cipta
89
Table 27 Comparative ranking o f the levels o f industrial and agricultural development

I860 Agriculture Germany Austria Belgium Spain United States France Italy Japan Great Britain Russia Sweden Switzerland 6 -7 9 4 -5 4 -5 1 3 11 12 2 10 6 -7 8 Industry 6 -7 6 -7 2 -3 9 2 -3 5 1 0 -1 1 12 1 9 -1 0 7 4 1880 Agriculture 3 8 5 9 -1 0 1 4 11 12 2 9 -1 0 7 6 Industry 5 7 -8 2 -3 9 2 -3 6 10 12 1 11 7 -8 4 1910 Agriculture 2 8 4 10 1 5 -6 11 12 3 9 7 5 -6 Industry 4 -5 8 3 9 1 6 -7 10 12 2 11 6 -7 4 -5
Sources: P. Bairoch, Niveaux de developpem ent econom ique dc 1810 a 1910, op. cit.
development, the ratings for which are given in this chapter. Thus Latin America is at the to p , followed by the Middle East, with the Afro-Asian countries coming last. A t the national level, the same correlation between agricultural productivity and industrial developm ent, although not perfect, is once again evident. Thus, the tw o countries in Latin America with the highest level of agricultural productivity are Argentina and Chile. And amongst th e seven Latin American countries for which calculations have been made the one in which agricultural produc tivity is lowest is also the one in which industrialization has advanced least, i.e. Peru. In the three Middle East countries for which calculations o f agricultural productivity have been com pleted, both agricultural productivity and industrialization are highest in Turkey. In the Afro-Asian countries th e relationship is not so obvious, since the differences between countries, both in agriculture and manufac turing, are not so great. However, we can see a correlation in at least a few instances. In Asia, amongst the big four China, India, Pakistan and Indonesia it is China, with the highest agricultural productivity which also has the m ost developed industry. In Africa, Tunisia proves rather an exception to the rule, with a low level of agricultural productivity and a relatively high degree of industrializ ation. But in Morocco, and m ost of the other countries, the relationship between these tw o sectors accords m ore nearly with w hat has been observed before.
90
II C H A N G I N G C O N D I T I O N S IN A G R I C U L T U R E A N D INDUSTRY
In a closed econom y dom inated by agriculture it is natural th at industry should be strongly influenced by changes in agricultural conditions, for a decline in agricultural production reduces the outlets for industrial production through a general fall in purchasing pow er.20 A situation o f this kind occurred in m ost European countries in periods when their industrialization had n o t proceeded very far. Thus in France, right up to 1870, every bad harvest was followed by an industrial crisis. The various statistical series by which the parallel m ovem ent o f the curves o f agricultural and industrial ou tp u t can be dem onstrated will not be given here, b u t figures taken from Professor Marczewskis interesting and suggestive study o f the quantitative aspects o f French economic history (Table 28) show the close relationship between the tw o sectors. Given the part played by industrial im ports in under-developed countries and assuming also th a t the growth o f local production has hitherto been based largely on im port substitution im ports being much m ore sensitive to econom ic fluctuations than is production when it forms so small a part o f consum ption it is difficult to find
a sig n ifica n t co rrela tio n b e tw e e n agricultural and in d u strial p ro d u c-
Table 28 Comparison between the growth o f purchasing power generated b y agriculture and the industrial output in France (annual averages in millions of current francs)
Growth i?i absolute value in purchasing pow er generated by agriculture 797 464 577 159 793 2,076 1,024 -5 6 6 -6 0 9 324 2,730 G rowth in absolute value o f industrial output 1,090 208 1,072 1,444 948 1,824 1,031 451 227 1,274 3,809
Period 1 7 8 1 /9 0 -1 8 0 3 /1 2 1 8 0 3 /1 2 -1 8 1 5 /2 4 1 8 1 5 /2 4 -1 8 2 5 /3 4 1 8 2 5 /3 4 -1 8 3 5 /4 4 1 8 3 5 /4 4 -1 8 4 5 /5 4 1 8 4 5 /5 4 -1 8 5 5 /6 4 1 8 5 5 /6 4 -1 8 6 5 /7 4 1 8 6 5 /7 4 -1 8 7 5 /8 4 1 8 7 5 /8 4 -1 8 8 5 /9 4 1 8 8 5 /9 4 -1 8 9 5 /1 9 0 4 1 8 9 5 /0 4 -1 9 0 5 /1 3
Sources.- Jean M arczcwski, 'L e pro d u it de Icconom ie franâise de 1789 a 1913 (com paraison avee la G rande B retagne) in Cabiers de I'l.S M .A . A F 4, no. 163, Paris (1 9 6 5 ), p. lxvi.
91
tion. iMoreover as the available series are so few and cover so short a period it is improbable th a t we should be able to fine one. For these reasons I have endeavoured to establish the influence of agriculture on industrial conditions by comparing the change in the gross product generated by agriculture with th at generated by m anufac tures. This comparison clearly shows a strong relationship between the two sectors in under-developed countries: the curve o f agricultural out put powerfully influences th at o f m anufacturing industry. The U.N. statistical services calculate index numbers of gross agricultural product and gross industrial product (manufactures) for the con tinents o f Latin America and non-com m unist Asia (excluding Japan). The movement o f these curves is evidence of agricultural influence and th e correlation coefficient between the tw o series reveals very high ratios: 0.99 for Latin America and 0.97 for Asia. It should be noted, nevertheless, th a t the real correlation is over-estimated due to the progression of the two curves. But the exam ination o f the correlation for individual countries is far more illuminating. The curves o f the w o series have been examined for fo rty under-developed countries between 1950 and 1970. The correlation is clear for nearly all the countries with a relatively im portant m anufacturing sector. Each decline in, or slowing down of, agricultural production is followed by a similar movement of m anufacturing o u tp u t, and similarly each acceleration in agricultural production leads to an acceleration in industrial growth. In m ost cases a delay of approxim ately one year separates the tw o curves, which, since harvests occur m ostly in mid-year, is what might be expected. I have attem pted to verify the relationship by comparing the curve o f im ports of m anufactures with th a t of agricultural o u tp u t (F.A.O. figures) for Asia, Latin America and the Middle East. The trend of these tw o curves indicates the influence of agriculture on im ports of m anufactures in the case of the tw o first groups of countries. But the figures for the Middle East are less clear cut because receipts from oil exports distort the relationship. Thus, as we m ight have expected, given the proportionate size of rural population, it is quite clear th a t agriculture strongly affects the development of m anufacturing industry and economic conditions generally in the Third World countries, as it did in the western economies at the time of their take o f f . I shall leave further discussion of this most im portant aspect o f econom ic development to the concluding chapter.
5
INTRODUCTION
Foreign Trade
I shall devote two chapters to the foreign trade of the Third World, the first dealing with general quantitative aspects and the second focussing on the problem o f term s o f trade, where, I shall attem pt to show the real changes which have taken place in these over the whole period 18701970. For the purposes o f the present study it is unnecessary to analyse the pattern of foreign trade for each individual country, as com plete and reliable statistics are available for the Third World taken as a whole. However, th e reader will find in the appendix a synoptic table of figures relating to 1970 im ports and exports (and the annual rate o f variation o f exports betw een 1960 and 1970) for m ost of the under-developed countries. L iterature on th e problems o f foreign trade abounds,1 so I shall make my com m entary brief. O f the four sections in this chapter the first three deal with th e non-comm unist, under-developed countries and the fourth with China, under the following headings: A. B. C. D. General changes in foreign trade The pattern of foreign trade by com m odities The geographical p attern o f foreign trade A note on the foreign trade of China
A. General changes in foreign trade Thanks to the statistical office o f the U nited Nations we have at our disposal tables which assemble data about the foreign trade o f the entire under-developed world from 1938 onwards, excluding only the com m unist countries. The omission of the latter is o f no significance for the present study since their foreign trade is very limited. Thus, while the com m unist under-developed countries have a population nearly half the size o f th e under-developed non com m unist countries, their share in term s o f exports is no more than about 4 per cent. Their omission can also be justified on m ethodo logical grounds, for foreign trade in the so-called 'm ark et economies plays a totally different role to th at which it plays in m ost com m unist econom ies and especially in China. (F or a brief summary of Chinese overseas trade the reader should turn to the end o f this chapter.) 92
Foreign Trade
93
I have supplem ented the United Nations series for the period 190028 from the figures of the individual countries. It should noted th a t th e United Nations statistical office counts Turkey as a developed country in all their trade data (total trade, geographical structure, price, etc.) while in reckoning other aggregates, th at country is usually considered part of the Third World. But these figures have not been adjusted for tw o reasons. The first and the m ost im portant is th at the overseas trade of Turkey represents no more than one per cent o f the total for the whole non-comm unist, under-developed world (and less than 0.3 per cent of that o f the non-comm unist world). The second reason is th at these very m inor corrections might confuse any one wishing to update my figures by reference to published data. The reader should also bear in mind that in this chapter, unless otherwise indicated, am ounts are expressed in current U.S. dollars. Table 29 Foreign trade o f less-developed countries 1 1 9 0 0 -1 9 7 0
EXPO RTS f.o .b . IMPORTS c.i.f. in millions US dollars 1900 1913 1928 1938 1948 1950 1953 1958 1960 1963 1965 1968 1970
Sources:
B A LA N C E OF TRAD E in millions US dollars
in millions US dollars 1,600 3,800 7,600 5,900 17,100 18,900 21,100 24,800 27,300 31,500 36,400 44,100 55,600
as percent o f world exports 16.0 19.0 23.0 25.0 29.7 31.0 25.5 22.9 21.3 20.4 19.5 18.4 17.8
as percent o f imports
1,600 3,500 6,500 5,800 18,600 17,500 21,500 27,800 30,200 32,900 38,000 46,000 57,600
+ 300 + 1,100 + 100 -1 ,5 0 0 + 1,400 - 400 -3 ,0 0 0 -2 ,9 0 0 -1 ,4 0 0 -1 ,4 0 0 -2 ,2 0 0 -2 ,0 0 0
+ 9 + 17 + 2 - 8 + 8 - 2 -1 1 -1 0 - 4 - 4 - 5 - 4
1 Excluding com m unist countries 1 9 0 0 2 8 , a u t h o r s estim ates; 1 9 3 8 7 0 d e r iv e d f r o m Y e a r b o o k o f I n te r n a tio n a l T ra d e S ta tis tic s (U .N .), various issues; recen t figures from M o n th ly B ulletin o f Sta tistics (U .N .), M ay 197 3.
Long-term changes in the foreign trade of the under-developed world, i.e. since 1900, can be divided into three fairly distinct periods. The first which to all intents and purposes began in the m id-nineteenth century and ended in 1950 saw a gradual growth in the im portance of th e foreign trade o f these countries in the world economy. In 1900 their overseas trade form ed about 16 per cent of world trade: by 1913 the figure had reached 19 per cent: by 1948 30
A i X a
fcsV*
94
per cent. The period was also characterized by favourable trade balances since exports easily outpaced im ports (in spite o f the fact th at, as is the general custom, exports are expressed in f.o.b. prices, while im ports are expressed in c.i.f. prices).2 But, in fact, the term trade balance is hardly applicable to the countries we are considering since m ost o f them were still part o f a colonial system. The second period ran from 1950 to 1962 and saw an overall reversal of the trend established between 1900 and 1950. There are three points to be made about the general developm ent o f the foreign trade of the Third World during this period. 1. Although in absolute volume th e overseas trade of the under-developed countries had definitely increased, the rate of growth was nevertheless lower than th at o f Gross Domestic Product. Thus, taking 1950 = 100, exports were 148 in 1962, while the index number of the G.D.P. was 174. 2. The foreign trade of the developed countries grew much faster than th at o f the Third World, thereby bringing about a continuous and significant reduction in the latters share o f world trade. Thus, Third World exports which in 1948 had form ed 30 per cent of the world total, had fallen to only 20 per cent by 1962. The settingup of the Common M arket a n d , to a certain e x te n t, th e E u rop ean F ree Trade Association, by favouring trade betw een their members, helps to explain this development. However, the foreign trade of industria lized countries who were n o t members also grew faster than th at o f the under-developed countries. The deterioration in the terms of trade o f the under-developed countries (to which I shall return in the next chapter) obviously played a part in reducing the m om entum of their international trade. Indeed, if expressed in current prices with the level of 1950 as 100, the index o f their exports in 1962 was 148, whereas expressed in constant prices it was 156 (the index num ber o f the un it value of exports having fallen from 100 in 1950 to 95 in 1962). As we shall see in the next chapter, this worsening of the term s of trade occurred mainly between 1955 and 1962. It m ust be emphasized th a t w ithout an increase in the volume of oil sales (and in spite o f the intervening fall in prices) the Third W orlds trade would have suffered even more. I shall return to this point later when discussing the structure o f foreign trade. The m arked falling off is mainly evident after 1954. The whole question o f the under-developed countries position in world trade, especially in th at o f the western world, has sufficiently im portant implications to m erit separate discussion (see below, section C,II).
jU - X
l_ lik . J A l
i X A
Foreign Trade
95
3. The period is m arked in general by a relatively large deficit in the trade balance of the under-developed countries. This deficit reached and even, during some years, exceeded 3 billion dollars (i.e. 10 to 11 per cent of im ports), thereby reinforcing the necessity for financial aid. I shall return later to the problem of the trade deficits. The third period began about 1962 and lasted until after 1970. Its essential characteristic was a much faster expansion in both the value and the volume o f exports. While from 1950 to 1962 exports in current values increased at an average annual rate of 3.4 per cent, from 1962 to 1970 th e rate m ore than doubled, reaching 8.4 per cent. This acceleration is linked with a halt in the fall in unit prices o f exports (measured by volume, the respective rates o f expansion are 4 per cent and 7 per cent) and with the generally expansive movement in international trade which was characteristic of the sixties. Indeed during those ten years and especially after 1959, the expansion o f international trade was quite unique: never before had there been a continuous upw ard m ovem ent of such am plitude. From 1958 to 1970 the value o f world exports leapt from 108 billion dollars to more than 312 billion dollars, which represents an annual rate of growth of some 9 per cent. The expansion o f exports measured by volume was slightly less about 8.5 per cent3 b u t even so it was remarkably high, representing as it did a doubling of volume every nine years. A nother cause of the rapid growth in the Third W orlds exports was the great increase in their sales o f m anufactured goods, especially after 1963. Between 1962 and 1970 sales practically tripled, and I shall have to return to this im portant aspect of foreign trade below (section B). However, even after 1962 world international trade grew faster than the foreign trade o f the under-developed countries; so their share of exports in the world total still fell, although more slowly than formerly, from 20 per cent in 1962 to 19 per cent in 1966 and 18 per cent in 1970. The figure for 1970 is close to th a t for the beginning of the century. The peak of 31 per cent had been reached in 1950, this percentage being exceptionally high because the prevailing prices o f tropical com m odities were also high, so th at we should perhaps consider the m aximum norm al share o f the less developed countries in th e world to tal of international trade to be about 2528 per cent w hat it in fact was around 1950. A slower growth o f im ports in relation to exports, especially after 1963, led to the gradual reduction o f trade deficits which appeared after 1948. If we ignore the distortion introduced by the difference between c.i.f. values of im ports and f.o.b. values o f exports this trade
96
deficit has practically disappeared in recent years. When the figures are corrected to take account o f this difference4 there is even a slight surplus in trade balances from 1963 to 1970. However, since the foreign trade o f the under-developed countries is carried by m erchant shipping of the developed countries,5 th e difference between c.i.f. value o f im ports and f.o.b. value o f exports, which is essentially the cost o f transport, is n o t recovered by the under developed countries. This is why, when considering resources created by exports, we should only take account of f.o.b. values. This particular equilibrium is only valid for the under-developed countries taken as a whole, for there are obviously substantial variations between individual countries (see synoptic table, Appendix p. 244) and groups of countries. It m ust also be emphasized th at equilibrium was only achieved through the increasing volume of oil exports which, by 1970, form ed more than 33 per cent o f exports as against only 20 per cent in 1953. Thus, for example, the trade deficit for the years 1969 and 1970 for all the under-developed countries together was about 2 billion dollars, i.e. 3.6 per cent o f imports, b u t if we were to exclude the main oil-producing countries, the deficit would rise to about 10 billion dollars, i.e. 20 per cent o f imports. This is even more striking when we consider th at the countries exporting oil have less than 3 per cent o f the total population of the Third World. The deficit o f 10 billion dollars is a record, the m ore normal figure between 1957 and 1966 fluctuated around 6 billion, b u t in relation to im ports the deficit (20 per cent o f im ports ) is n o t so exceptional, for it was 21 per cent on average between 1957 and 1966, rising to a peak of 24 per cent in 1961. Before discussing the structure of foreign trade, the changes in the trade balances of the main geographical areas since 1913 will be briefly examined. Latin America had a positive trade balance for the whole o f the period up to 1967, except for a few untypical years. Since the end of the war the surplus has been due in large measure to Venezuelan exports o f oil. Yet, in spite of the Venezuelan contribution, there has been a growing deficit since 1968. Similarly oil exports are responsible for the very great surplus in the trade balance of the Middle East which began to m anifest itself in the 'fifties and which, by 1970, formed nearly 60 per cent of imports. In th at year the surplus was about 4.2 billion dollars, although w ithout the oil trade, o f course, there would have been a deficit of about 7 billion dollars.6 The rest of Asia suffered a rude reversal of its trade balance, which
ial com
or
Foreign Trade
97
until the war had been a strongly favourable one (24 per cent in 1913 and 26 per cent in 1928); after the war the balance swung alm ost equally far in a negative direction. The trade deficit, which was less than one billion dollars up to 1953, reached 1.8 billion dollars in 1958 (22 per cent of im ports), 2.2 billion dollars in 1960, 3.5 billion dollars in 1965 and close to 4.5 billion dollars in 1970 (i.e. 23 per cent of imports). Lastly, Africa's deficit fell steadily, though interm ittently. In 1913 it form ed a quarter o f its im ports; b u t since 1964 the trade balance has been either in equilibrium or very slightly in surplus.
Table 30 Change in foreign trade and trade balance o f non-communist , less-developed countries according to geographical regions (in millions o f dollars)
1913 LATIN AMERICA Im ports 1,450 1,600 Exports + 150 Balance as %age of + 10.3 imports ASIA Im ports Exports Balance as %age of imports 1928 1938 1948 1953 1960 1970
2,450 3,100 + 650 + 26.5
1,540 1,710 + 170 + 11.0
6,180 6,520 + 340 + 5.5
6,530 7,620 + 1,090 + 16.7
8,350 8,560 + 210 + 2.5
18,920 17,180 - 1 ,7 4 0 9.2
1,250 1,550 + 300 + 29.0
2,500 3,140 + 640 + 25.6
2,370 2,650 + 280 + 11.8
6,010 5,200 - 810 13.5
7,010 6,050 - 960 13.7
9,830 7,630 -2 ,2 0 0 22.4
18,770 14,430 4,340 23.1
MIDDLE EAST Im ports Exports Balance as %age of im ports AFRICA Im ports Exports Balance as %age of im ports
410 330 80 19.5
630 630

720 550 170 23.6
2,230 2,060 - 170 7.6
2,500 2,810 + 310 + 12.4
4,520 5,130 + 610 + 13.5
10,150 14,300 +4,150 + 40.9
330 250 80 24.2
690 590 100 14.5
670 660 10 1.5
2,750 2,260 - 490 17.8
3,750 3,290 - 460 12.3
5,320 4,440 - 880 16.5
8,840 9,030 + 190 + 2.1
Sources: 1 9 1 3 28, a u th o rs estim ates; 1 9 3 8 70 derived from Y earbook o f International Trade S ta tistics (U .N .), various years, and M o n th ly B ulletin o f S ta tistics (U .N .).
aterial com direitos autor,
98
B. The pattern of foreign trade by commodities Only since 1955 have we had a hom ogeneous series o f statistics enabling us to study the structure of foreign trade in th e Third World as a whole. This account o f changes in the com position and direction of trade com m odities is based on the inform ation given in these statistics (prepared by the U.N. statistical office). In the first place the percentage distribution o f the various products confirms the lowly position o f industry in the under developed countries. In 1970, in spite o f trem endous growth in recent years, m anufactured goods form ed only 23 per cent of exports while forming 69 per cent o f im ports. By way o f comparison, in th e same year m anufactured goods form ed 76 per cent o f the exports from developed countries. Though we m ay not find this difference surprising y et we should remember th a t the industrialized countries o f the present-day had, at the outset of their take o f f , a very different pattern o f foreign trade to th a t now prevailing in th e Third World, with m anufactured goods forming a m uch greater proportion o f their exports. Thus m anufactured goods form ed about 43 per cent o f French exports in 1715; 45 per cent in 1787; 56 per cent in 17981800 and 70 per cent in 182730.7 In other countries the pattern was similar. However, these statistics m ust n o t be taken to mean th a t the difference betw een the levels of developm ent attained by the developed countries then and the Third World now, was correspond ingly great. In the past exports form ed a very small part o f Gross Dom estic Product (in the traditional western economies before industrialization it was probably around 35 per cent) whereas in the under-developed countries o f today exports can in some countries be as much as, or even m ore than, 30 per cent o f G.D.P. (1970). For all under-developed, non-com m unist countries the average percentage of exports in G.N.P. for 1970 was 15 per cent; in the early fifties it was about 20 per cent; while at the beginning of the century it had already reached the level o f 1517 per cent. Thus on average and overall the relative share o f exports in the G.N.P. is three to five tim es greater in the under-developed countries in this century than it was for traditional economies before their industrialization. These statistics help to explain, at least partially, the startling difference in the distribution o f export com m odities outlined in the previous paragraph. To return now to a m ore detailed exam ination o f the structure of the foreign trade o f the under-developed countries and the modifi-
Foreign Trade
99
cations it has undergone. As has already been emphasized, m anufac tured goods form an im portant share o f im ports, b u t only a m inor share of exports. Nevertheless changes in this field betw een 1955 and 1970 have been favourable. Although the percentage of m anufac tured goods in the total im ports has increased from 59 to 69, this was due mainly to im ports of capital goods; the absolute growth in the value o f other m anufactured goods im ported has n o t kept pace with the growth o f total im ports. We have already seen in an earlier chapter th a t im port substitution resulting from the relatively fast developm ent o f m anufacturing industries can account for this decline. In the field of exports, rapid growth in tw o groups o f products m ust be singled out: fuels and m anufactures. Where fuels are concerned, growth began well before 1955, b u t the changes in production o f m anufactured goods have been much m ore recent. Until 1962 sales o f m anufactured goods increased at nearly the same rate as total exports; b u t between 1962 and 1970 a m ost spectacular expansion occurred the value o f these goods rising from 4.3 billion dollars to 12.7 billion dollars, i.e. an annual average increase of 14.5 per cent. The incidence o f benefits arising from these changes needs close inspection. The expansion in exports o f m anufactures has actually Table 31 Foreign trade b y com m odity classes o f non-communist, less-developed countries
IMPORTS 7955 TOTAL (in billions of USS) D istribution of products as percentage of total 1. Food products 2. Primary products 3. Fuels
4. Chemicals
EXPO RTS 1970 57.7 1955 23.7 i9 6 0 27 A 1970 54.3
I9 6 0 29.1
23.1
15.4 8.1 11.8

7.0
16.2 7.4 10.0

7.5
5. 6. 1 -3 4 6
Machinery & Transport O ther m anufactures Raw materials Processed goods
23.3 29.0 35.3 59.3
27.6 27.8 33.6 62.9
12.9 6.2 8.0 8.9 33.4 26.4 27.1 68.8
32.5 29.4 25.2 1.0 0.5 11.4 87.1 12.9
29.6 27.9 27.9 1.1 0.7 12.4 85.4 14.2
24.3 18.2 33.3 1.5 2.4 19.5 75.8 23.4
N ote. T he above to tals are no t strictly com parable w ith to tal im ports and e x p o rts q u o te d elsew here since no account has been taken o f ships stores, bun k erin g fuel and o th e r special ty p es o f m erchandise, figures for which are n o t available. In addition th e figures for im ports are based on w orld e x p o rts destined fo r these countries w hich, consequently, im plies th a t the values are f.o.b. Sources: Derived from th e Statistical O ffice o f th e U.N. (revised scries in roneo form ) and M o n th ly B ulletin o f Sta tistics (U .N .), Ju ly 1972.
rial com direitos autor,
100
taken place in a few countries with small populations: thus the share o f these goods exported from Hong Kong, Nationalist China (Taiwan), Israel and the Republic of Korea com bined, has risen from 29 per cent of the total Third World exports o f these goods in 1959 to 42 per cent in 1968. In 1968 these countries provided only 8 per cent o f the total exports o f the under-developed world and form ed only 3 per cent o f its population.8 Hong Kong's case, which is often taken as an example o f the possibilities theoretically open to under-developed countries in the export o f m anufactures, can serve to illustrate the limits to such development. Simple arithm etic will show us th a t if the under developed countries (always excluding the com m unist countries) had exported in 1970 as m any m anufactured goods per capita as Hong Kong did, these exports would have risen to $1040 billion, i.e. nineteen times m ore than the total Third World exports and over three times more than total world exports. Exports of fuels' alm ost entirely oil grew from 6 billion dollars in 1955 to 18 billion dollars in 1970 (in spite o f the fall in prices already m entioned above) which pushes up the relative percentage of this sector from 25 per cent o f exports to 33 per cent. These tw o groups o f products together form more than half the total exports o f the Third World (55 per cent). The remaining exports are composed of a huge assortm ent o f prim ary and agricultural products. A round 1970 the tw o principal com m odities were coffee and copper, each with about 5 per cent o f total exports, followed by sugar, cotton, oilseeds, cereals and rubber, each with about 3 per cent, while tea, cocoa and m eat formed no m ore than one per cent each. To sum up, though it may seem as if, overall, the structure of foreign trade in the Third World reflects the condition o f under developm ent, yet changes occurring since 1955 have been favourable, with im ports of capital goods increasing while those of other goods decline relatively, and exports o f m anufactured goods expanding rapidly. These changes confirm the diagnosis I m ade when discussing m anufacturing industry, when I pointed out the favourable move m ents in this sector, noting at the same tim e th a t this positive developm ent might be short-lived because industrial production has been out-stripping dom estic consum ption the latter being rela tively sluggish because agricultural productivity is stagnating if not actually declining. Industrial expansion has so far been sustained by the substitution o f hom e-produced goods for im ported ones. Before going on to examine the geographical structure o f foreign
Foreign Trade
101
trade a few words m ust be said about the concentration o f com m odity-export in under-developed countries and about fluctu ations in export prices. Extrem e cases of concentration o f com m odity-export in m any Third World countries have been the subject of much discussion, and I shall lim it myself here to estimating its im portance for th e under-developed world as a whole. To this end I have calculated a weighted average (by dollar value of exports) o f the share o f a principal product relative to total exports. In 1963, for the entire Third World, 55.6 per cent of all exports consisted o f goods which were at the top o f the list in individual countries; in the developed countries the proportion was only 18.5 per cent. It is obvious th at this difference is crucial and that diversification o f foreign trade should be one o f the objectives of j economic policy in th e under-developed countries, although I naturally it m ust depend to a great extent on industrialization. Diversification will also help to resolve the problems caused by excessive fluctuations o f export prices in under-developed countries, since such fluctuations arise mainly from the predom inance of a small num ber of agricultural and mining products. C. The geographical pattern of foreign trade
I GENERAL CHANGES
In Table 32 I have calculated the geographical distribution of the foreign trade o f the non-comm unist, under-developed world on the basis of U.N. statistics. The figures upon which these percentages have been based are not strictly comparable to those which will be , found in preceding tables. The differences are due to the fact that while the data is assembled from the figures o f exports classified by destination (im ports therefore being expressed in f.o.b. prices), it has remained impossible to treat all figures according to this classifi cation, as the destination o f some exports is n o t specified. Additionally, some figures are revised less frequently than the totals, which also affects the results. It is impossible to eliminate these differences, but they are not im portant enough to introduce significant bias into this table. Scrutiny o f this table immediately reveals one im portant point: the very m inor role played by trade between the under-developed countries themselves. Indeed such exchanges form ed less than 20 per cent o f the total foreign trade of the Third World in 1970. The percentage has tended, moreover, to fall. In the neighbourhood o f 25 per cent before the war, it rose to 30 per cent immediately afterwards due to the relative decline in the
102
foreign trade o f the European countries; b u t by 1953 it had fallen back again to 25 per cent and by 1965 it was about 21 per cent. Thus the overseas trade of the non-comm unist Third World is conducted mainly with the developed countries, and more especially with the western developed countries. Their exchanges with the com m unist countries, although they have strongly increased recently, still remain very limited. Table 32 Changes in the regional origins and destinatio ns o f the foreign trade o f the non-communist less-developed countries , 19381970
Origin o f imports into less-developed countries (percentage) 1938 DEVELOPED COUNTRIES (non-com m unist) North America Common Market E.F.T.A. Japan Oceania LESS-DEVELOPED COUNTRIES (non-communist) COMMUNIST COUNTRIES Europe & U.S.S.R. Asia TOTAL in millions $ (f.o.b.) 1948 I9 6 0 1970 Destination o f exports fro m less-developed countries (percentage) 1938 1948 1960 1970
68.0 17.2 20.7 17.4 10.7 1.2
67.4 32.1 12.8 17.7 0.9 2.8
74.8 25.8 23.0 15.6 7.0 1.2
72.7 23.9 20.0 11.2 13.4 1.8
71.9 17.1 23.6 20.3 8.1 1.9
66.6 23.9 16.9 19.8 1.0 2.6
72.3 23.6 22.8 16.2 5.1 1.9
73.5 21.7 25.4 12.3 10.6 1.3
25.2 6.6 4.2 2.4 100.0 5,070
30.0 2.6 1.2 1.4 100.0 16,800
20.9 4.2 2.8 1.4 100.0 29,140
18.4 8.9 6.9 2.0 100.0 57,790
21.7 3.1 1.4 1.7 100.0 5,900
29.1 2.7 1.8 0.9 100.0 17,300
22.3 4.5 3.5 1.0 100.0 27,340
19.5 5.7 4.9 0.7 100.0 54,290
Sources- Derived from Yearbook o f International Trade S ta tistics , (U .N ., 1963) and M o n th ly B ulletin o f Sta tistics , (U.N., Ju n e, 1972). Sec tex t for differences in com parability betw een this table and preceding ones.
If the share o f all the developed countries in the overseas trade of the Third World has not, in global terms, altered very profoundly (rising from a total o f 70 per cent in 1938 to 73 per cent in 1970), there are nonetheless some individual changes to be observed. N orth America, which in 1938 had a smaller volume o f overseas trade than either the Common Market or E.F.T.A. countries as a group, overtook them after the end o f the war, especially where im ports coming from the Third World were concerned. The real reasons for this changeover were firstly the enorm ous increase in sales or gifts of American agricultural produce, and secondly (and this is also true for
Foreign Trade
103
foreign trade as a whole) the isolation of Europe during m ilitary hostilities so th at world trade turned towards the United States, where the immense growth in o u tp u t in the war period favoured trade with the less-developed countries who remained outside the grip of the Axis powers. But gradually continental Europe is regaining and even superceding its form er im portance in Third World trade. C ontrary to w hat might have been expected from its geographical location in the western hemisphere, Latin Americas share in U.S. trade with the Third World is not really predom inant. Thus, in 1970 U.S. trade with Latin American countries form ed only 51 per cent o f American exports destined for under-developed countries and 53 per cent o f im ports thence. It was only in the few years immediately after the war th at these currents of trade assumed great im portance; in 1948 they formed some 65 per cent of the trade between the U.S. and the under-developed countries. In western Europe the experiences of the Common Market and E.F.T.A. countries were divergent. In 1938 trade between these two groups and the Third World had been at a very similar level but the E.F.T.A. countries gradually lost ground in spite of the smaller impact of the war on their economies. By 1970 their share of trade was no more than 12 per cent com pared to about 19 per cent in 1938. The Common M arket countries, whose trade had been so much reduced by the war, experienced a m arked revival and the relative level reached by 1970 was about the same as that of 1938. Decolonization does not appear to have entailed a significant rc-orientation of trade, especially if the Common Market and E.F.T.A. countries are taken as a whole. On the other hand, in the case o f individual countries, changes usually m eant a reduction in the part played by the principal trading partner. Japan, which by the end o f the war had lost its place in the m arkets o f the under-developed countries, gradually recovered her position. Thus although in 1948 she had only one per cent o f Third World trade as compared to 9.5 per cent ten years previously, by 1965 her share was nearly 10 per cent again and in 1970 it had risen to 12 per cent. Eastern Europes volume of trade with the Third World was much reduced by the war and by the subsequent com m unist takeovers. Not until after Stalins death, and the changes in economic and foreign policy which followed, did these countries begin to assume a more im portant, though still limited role, as trading partners. In 1954 the volume of their trade with the Third World increased by 70 per cent
104
com pared to the level of 1953. The increase has continued, though the rate has hardly been m aintained, and these countries had regained their pre-war share by 1960. In 1970 th e European com m unist countries to ok over 5 per cent o f all trade with the under-developed countries: little enough compared to the western countries, but, even so, three times m ore than the pre-1953 figure. A similar change in the trade between the Asian com m unist countries and the under-developed world is evident, although the volume remains very low about 1.5 per cent o f total trade in 1970. In conclusion the following tw o points deserve re-emphasis: 1. The very small am ount o f trade betw een the under-developed regions themselves. Obviously the principal reason for this is th at their economies are not com plem entary. But as this is also due to the patterns of trade established by colonial and neo-colonial dom ination9 a partial re-orientation is certainly possible as well as desirable. 2. The pre-dom inance of the western and European countries in the trade of the under-developed countries.10
II T H E P L A C E O F T H E U N D E R - D E V E L O P E D C O U N T R I E S I N T H E
F O RE I G N T R A D E OF THE WESTERN D E VE L O P E D C OU NT RI ES
We have just seen th a t the western developed countries are taking a growing and largely dom inant place in the foreign trade of the under-developed countries, since at present some 73 per cent o f the latters trade is with them. Consequently, it would be interesting to see if the place of the under-developed countries in the foreign trade of the developed countries is equally im portant and if it has moved in the same direction. The answer is in the negative. The relative dimensions o f the two elements are quite disproportionate and the trend o f change is divergent. Indeed in 1970 the under-developed countries to o k no more than 19 per cent o f the total exports o f the developed countries, and this percentage has for some years shown a distinct tendency to fall, especially since 1958, at which time it was still nearly 28 per cent. In 1938 the under-developed countries provided 23 per cent o f the developed countries foreign m arkets; in 1948 the percentage was 31 and remained approxim ately the same until 1958. But the decline began in th at year and continued until 1970, when, as we have seen, it was no m ore than 19 per cent. The share of im ports from Third World countries in total im ports o f developed countries is very similar.
Foreign Trade
105
Thus we are confronted with a pattern which, in global terms, places the under-developed countries in an unfavourable situation, for although the western m arket is o f prime, even vital, im portance to them , the place o f the under-developed countries in the trade o f the western world can by contrast be considered if not exactly marginal, then certainly far from being o f m ajor im portance. And if. the share o f the developed countries exports to the Third World in 1970 is expressed as a percentage o f their Gross Domestic Product the figure is no more than 2.0 per cent (see Table 34) hardly a substantial figure. In 1958, the figure at 2.5 per cent was slightly higher, b u t it has certainly never reached 3 per cent. On the other hand, if we express the exports o f the under-developed countries destined for the developed countries as a percentage o f their G.D.P., the figure for 1970 is about 11 per cent (Africa and the Middle East, 22 per cent; Latin America, 8 per cent; and Asia, 5 per cent). The size o f this divergence helps explain many o f the dom ination effects o f the developed countries on the Third World. The share of the under-developed countries in the foreign trade of the developed countries obviously varies from sector to sector and from country to country. Given th a t m anufactured goods form the larger part o f it, the volume of these goods exported to the Third
Table 33 The share o f the less-developed countries in the exports o f the developed countries as a percentage o f total exports o f the various products
1955 1. Food products 2. Primary products 3. Fuel 4. Chemicals 5. Machinery and transport 6. O ther m anufactures TOTAL 13. Raw materials 4 6. Processed goods 21.3 7.7 15.6 36.8 35.5 29.5 27.7 14.9 32.6 I9 6 0 25.0 8.4 15.8 30.8 31.8 24.6 25.5 16.6 28.3 1970 18.3 9.1 8.0 22.7 21.3 16.1 18.1 13.3 19.2
Sources: Derived from in form ation a b o u t th e special fea tu re o f th e M o n th ly B ulletin o f Statistics (U .N .), various num bers; recent figures in th e n um ber for July, 1972.
106
World by developed countries is proportionately more im portant but still not dom inant. Thus, exports of m anufactured goods to under-developed countries in 1955 form ed 33 per cent o f total exports of these goods. But it can be seen th at betw een 1955 and 1970 the Third World considered as a m arket for the developed countries declined rapidly in im portance both for m anufactured and for capital goods. For while in 1955, for example, it absorbed m ore than 35 per cent of total exports o f capital equipm ent from the developed countries, the rate has since dwindled steadily, being no m ore than 32 per cent in 1960 and only 22 per cent in 1970. A similar change occurred where fuel, chemicals and other m anufactured goods were concerned. The trend is obviously at least partly the result o f the phenom enon discussed earlier: i.e. im port substitution has caused an expansion o f m anufacturing in under-developed countries. But the m ain reason has surely been the immense increase in the trade o f the developed countries among themselves resulting from their rapid economic growth and the treaties of economic integration such as the Common Market and E.F.T.A. Thus, for example, from 4 billion dollars (4.9 per cent o f world trade) in 1953, inter-com m unity trade betw een C o m m o n M arket c o u n tr ie s rose to 6.9 billion dollars in 1958 (6.4 per cent o f world trade) and 43.3 billion dollars (13.9 per cent o f world Trade) by 1970. To turn now to the Third W orlds share in the developed countries imports. The only sector which is a really dom inating one, and increasingly so, is that o f fuel (in 1970 fuels originating in the Third World form ed 64 per cent o f total im ports o f these products). Where other products are concerned the im ports o f the developed countries have declined. In 1970 only 32 per cent o f to tal im ports o f foodstuffs and less than 29 per cent o f im ports o f prim ary products came from under-developed countries. The rapidly growing exports of m anufactured goods has led, since 1963, to a slight increase in the Third World share in total im ports of these products by developed countries a rise from 6 per cent in 1963 to close on 6.5 per cent in 1970. But, as we have seen, this growth profited only a very small part of the Third World. In Table 34 I have worked out, for the principal countries and groups o f countries, the 1970 percentages o f exports destined for the Third World, compared to total exports and to G.D.P. The estimate for the eastern European countries contains a margin o f error which, however, is not more than that which would result from rounding up the figure.
Foreign Trade
107
Table 34 Exports from developed countries intended fo r the non-communist Third World in 1970 compared to total exports and to G.N.P. at market prices
E xports Destined fo r Less-Developed Countries In millions of $ United States Canada C om m on M arket Germany Belgium & Luxem bourg France Italy Holland Free Trade Area (E.F.T.A.) Great Britain Sweden Switzerland Japan Australia, N. Zealand A ll western countries East European countries (including U.S.S.R.) 12,600 1,200 11,500 4,200 900 3,700 1,900 1,100 6,500 4,000 600 800 7,700 1,100 41,900 A s percentage o f total exports 30 7 13 12 8 20 14 9 15 20 9 16 40 19 19 13 A s percentage ofGNP 1.3 1.6 2.4 2.3 3.6 2.5 2.0 3.5 2.8 3.3 2.0 4.0 3.9 2.6 2.0
4,000
0.7
S ources: A u th o rs estim ates (see tex t).
Although, compared to total foreign trade, the share claimed by the under-developed countries varies considerably it will be noticed th at, when expressed as percentages of G.D.P. the differences are much less evident. The share o f exports directed towards the Third World lies between a minimum of 1.3 per cent and a maximum of 4 per cent o f the Gross Product of the industrialized countries, which is, even at m axim um, a very small proportion. In 1970 exports destined for the non-com m unist Third World from all industrialized countries taken together form ed, as we have seen, 19 per cent of their total exports and 2.0 per cent of their Gross Domestic Product. If we look at the larger groupings of countries or the great economic powers, we shall find that the Third World, relatively speaking, is the m ost im portant o u tlet for Japan (40 per cent of exports and 3.9 per cent o f Gross National Product). Japan is followed by E.F.T.A. (15 per cent o f exports and 2.8 per cent of G.N.P.) and then by the Common M arket (13 per cent o f exports
108
and 2.4 per cent of G.N.P.). Although the under-developed countries absorb 30 per cent of total United States exports (a quarter o f which are foodstuffs) these exports do n o t form m ore than 1.3 per cent of American G.N.P., owing to the low export rate of the American economy. * * * * * This chapter lacks an analysis o f the relationship betw een foreign trade and economic development. I started a study o f this problem after the first edition of this book was published, b u t the present state of my researches does not allow me to provide more than a few general and provisional conclusions.1 1 From an analysis o f th e period 195070 it is apparent that, contrary to w hat occurred in the industrialized countries, there was alm ost no relationship between the rate o f growth o f exports and th at of G.N.P. This holds good both for individual countries (with a few small and isolated exceptions) and for the Third World taken as a whole, even though in 1950 the latters exports form ed about 19 per cent o f G.N.P. Moreover, in the years 1962 66, which were the best ever experienced by the Third World until 1968 as regards the value o f exports, the G.N.P. grew m o re s lo w ly than it had d o n e since 1950. This resulted, o f course, m ostly from the fall in agricultural production, itself largely attributable to unfavourable weather conditions in India. But the limited connexion betw een exports and the rest o f the econom y was above all due to the fact th a t m ost goods exported by the under-developed countries were produced in w hat one might call economic enclaves (either plantations or mines) which had only a tenuous relationship with the rest o f the economy and whose effects on the econom y as a whole were consequently very limited. Furtherm ore these export sectors em ployed only a small part of the working population. R ather crude estim ates o f my own reveal th at for non-com m unist, under-developed countries 5 7 per cent of the working population produce goods o f which some 90 per cent are exported while the other 9395 per cent produce goods o f which only 1 per cent is exported. Such a total lack o f sym m etry is never found in industrialized countries. The m ost extrem e case is th a t o f oil production which, in 1970, formed 33 per cent o f total exports from th e under-developed countries and y et provided no more than 0.02 per cent o f total em ploym ent. This, o f course, is the main explanatory factor in the conclusions o f MacBeans study, which are at first sight so paradoxical. Taking as a working
Foreign Trade
109
hypothesis the commonly-held view th at annual fluctuations in export receipts have a negative econom ic influence, he has sought to measure the effects o f these fluctuations on some individual countries, and has concluded that there is no direct link between fluctuations in export receipts and those of the econom y as a whole. If I have stressed th e limited degree to which foreign trade can influence the econom ic life of m ost Third World countries, it is because there is a large body o f opinion which tends to exaggerate the possibilities. But we m ust beware o f falling into the opposite error. Foreign trade fulfils many im portant functions. At the m ost obvious level, export receipts are needed to pay for a whole range o f commodities which could never be produced locally in the early phases of development, b u t which are nonetheless absolutely essential for this development. A bout 70 per cent o f total external receipts of the Third World come from exports. Furtherm ore, because m any Third World countries are so extrem ely small, foreign trade is their sole avenue to developm ent. Seen from this angle regional integration becomes, for a large part o f the Third World, an absolute imperative. D. Note on the foreign trade of China The first thing to be said about Chinas foreign trade is th at the enorm ous size o f the country combined with the comparatively recent and far from com plete penetration hy western traders has m eant th at trade, even before the advant o f the present regime, had always been limited. Thus, in 1900 Chinese exports were 0.3 dollars per capita as against 1.2 dollars in India, and 3.7 dollars in the rest of the under-developed countries. This ratio was n o t altered by subsequent events since as the Chinese share in the total foreign trade o f the Third World remained stable, which it did until the outbreak o f the second world war. Chinese exports were w orth in current values about 120 million dollars in 1900, 290 million in 1913, 720 million in 1928 and 420 million in 1938.12 Before the Japanese occupation o f M anchuria began in 1931, the principal exports were leguminous plants, especially soya beans (about 20 per cent o f total exports), followed by raw silk (slightly less than 20 per cent). M anufactured silk goods formed less than 5 per cent, as did also tea, while the balance consisted of other agricultural products and some minerals, particularly coal.13 Japan had increasingly become Chinas main trading partner, closely followed by Hong Kong and the United States.
Bahan dengan hak cipt
110
Passing over the troubled years o f 1945 49, in which exports stagnated at no more than a few hundred million dollars, we come to the period 1950 70. These decades can be divided by the break with the Soviet Union into tw o periods of equal duration. The characteristic o f th e first period was a renewal and rapid developm ent o f Chinas foreign trade, in which exports rose from 620 to 2,205 million dollars14 between 1950 and 1959, or from 0.8 to 1.9 per cent of world exports. This, o f course, was still a m inute proportion, although it was n o t far short o f the 1928 figure o f 2.2 per cent. During this first period a little m ore than tw o-thirds o f the trade (exports as well as im ports) was conducted with other com m unist countries, especially the Soviet Union. Concerning the p attern o f exports, there was an increase in the sales o f m anufactured goods which in 1959 form ed 33 per cent o f total exports. The second period (196070) was m arked, until 1962, by a slump in foreign trade. During 1962 exports am ounted to no m ore than 1,525 million dollars (1.1 per cent o f the world total). This recession was due to a fall in the trade with m ost o f the com m unist countries. From 1963 onwards a slow revival got under way, thanks largely to a rapid increase in trade with western countries (which doubled between 1962 and 1965), while trade with the com m unist countries continued to decline. By 1970 the 1959 level had once again been attained and even slightly exceeded: Chinese exports in th a t year can be estim ated at 2,100 to 2,300 million dollars, i.e. 0.7 per cent of world total. Or, in other words, slightly less than $3 ($2.7) per capita com pared to $31 for the whole of the non-comm unist, under developed world (b u t only $3.7 for India), $311 for the western countries and $89 for the East European countries. By 1970 the geographical p attern o f trade was very different to what it had been before 1960, for only a third of die trade was conducted w ith com m unist countries. The principal trading partner was Japan (about a fifth of total trade), followed in order of im portance by Hong Kong, West Germany, G reat Britain, Australia and Canada. The principal exports in 1970 were foodstuffs, (about 30 per cent of the total) and, close behind, textiles and clothing (slightly less than 25 per cent). Amongst im ports first place was occupied by metals (about 25 per cent), followed by cereals (about 15 per cent, but the level fluctuated wildly from one year to another), m achinery (about 15 per cent), chemical fertilizers (10 per cent) and rubber (8 10 per cent).
The Terms oj Trade
It will be recalled th a t the term s o f trad e expresses the relationship existing betw een th e prices o f exports and im ports, either o f one country or group o f countries, o r o f one product or group of products.1 Hence an im provem ents in the term s o f trade o f a country, i.e. a relative rise in its export prices com pared to its im port prices, increases its m oney receipts and vice versa. Though from the point o f view o f Third World countries a worsening o f the term s of trade usually has an adverse effect, it should be noted th a t such a change could reflect a favourable econom ic situation, since lower ex p o rt prices could be due to a rapid growth in productivity. In such a situation the lower export prices need not necessarily reduce the total rem uneration accruing to local factors o f production. The problem o f the term s o f trade o f less-developed countries has attracted the atten tio n o f economists in general and even more o f the officials responsible for policies in the under-developed countries. As developing countries mainly export prim ary products and im port m anufactured goods, their term s o f trade have usually been treated as if they were identical with the terms o f trade of prim ary products against m anufactured goods. Even if it is obvious th a t these tw o concepts (i.e. th e term s o f trade o f prim ary products and the term s o f trade o f developing countries) are n o t altogether identical, their interchangeability can be accepted as a valid working hypothesis. The lack o f relevant statistics has m eant th a t terms o f trade of developing countries before 1938 have usually been measured and discussed on the basis o f the term s o f trade o f prim ary products. But from 1938 onwards the U.N. have brought o u t data from which a relatively valid series o f term s o f trade o f the Third World can be constructed. The different nature o f these tw o sets o f data and the rather unusual changes which have taken place since the end of World War II m ake it advisable to study the tw o periods separately. Thus I shall discuss first the term s of trade for the period 1870 to 1938 (concentrating on the years 1876/80 to 1926/29, because the depression years o f 1930/39 were exceptional) and then proceed in the following section to the period 1926/29 to 1970 and will thus link the early years w ith th e post-war period. In the first part I shall examine the widespread and generally
111
112
accepted assum ption th at during the earlier period the terms of trade o f prim ary products as com pared to m anufactures deteriorated. In the second p art I shall try to single out w hat were the probable causes of the deterioration o f the terms of trade which undoubtedly occurred after 1954. A. The terms of trade 18701938 As I have just said, the lack of statistics on the terms o f trade o f the developing countries necessitates an indirect approach to the subject through a comparison o f prices o f primary products with those of m anufactures. I shall first summarize the origins o f the m uch repeated and widely accepted assertion th at long-term price changes, between the last quarter o f the nineteenth century and the eve o f the second world war, resulted in a 43 per cent reduction in the price o f prim ary products com pared to those o f m anufactures. What this assertion implies is th at a given quantity o f prim ary produce would in 1938 have exchanged for only 60 per cent o f the quantity o f m anufactured goods th a t it would have exchanged for at the beginning o f th e period. From this it was concluded th a t the under-developed countries or territories had experienced an equivalent deterioration in their terms o f trade and a reduction in their ability to im port th e m anufactures, and above all the m achinery, necessary to their economic development. These estimates were made by the statistical office o f the League of Nations; they were first published in a book entitled Industrialization and Foreign Trade (Geneva, 1945) and were repeated later in a United Nations publication Relative Prices o f Exports and Imports o f Under-developed Countries 2 (New York, 1949). It is this latter, m uch-quoted study which has been mainly responsible for propagat ing th e notion o f a long-term fall in the prices o f prim ary produce in relation to industrial goods, although it should be noted th a t the study concentrates m ostly on the period 1938 48. On the evidence o f the generally available inform ation about the changes in productivity in the various sectors it seems doubtful to me th at such a change ever occurred. I have, therefore, decided to examine the statem ent critically from two separate angles: from the validity o f the date selected as a term inal point, and from the price indices used. I believe th a t a crticical exam ination o f the figures will disprove the theory th a t there was a fall in the prices of raw materials, and I shall present further evidence to show th a t in fact a
The Terms o f Trade
113
real secular im provem ent occurred in the terms of trade o f primary products vis a vis m anufactures. This I shall do on tw o fronts. 1 shall begin by analysing the share o f transport costs in the price of prim ary products, since the fall in costs during the period concerned is mainly responsible for the erroneous interpretation o f the relative price changes, and I shall proceed to scrutinize the price series of prim ary produce and industrial goods in various countries.
I. C R I T I Q U E O F T H E E S T I M A T E S P R O V I N G A S E C U L A R F A L L IN T H E T E R M S O F T R A D E O F P R I M A R Y P R O D U C T S
(a) The terminal date o f calcidations Ignoring for the m om ent the fundam ental problem o f the price series itself, 1 shall first try to show th a t the date chosen as a term inal point for comparing the relative change in the price o f primary products and m anufactures is an irrelevant one. For my immediate purpose I am therefore assuming th at the League of Nations estim ate, and others similarly constructed, are correct. We have seen th at the deterioration in the terms o f trade appears m ost obvious when th e years 187080 are com pared to 193038. But the thirties were the years of the Great Depression, and as far as prices are concerned they were totally abnormal. For proof o f this it is only necessary to calculate the depreciation of the prices of prim ary products com pared to the years before the crisis. For instance, according to the League o f Nations figures, the deterior ation for the period 1876/80 to 1936/38 was 43 per cent and for the period 1876/80 to 1931/35 59 per cent. But if we take only the years up to 1926/29 the deterioration was no more than 20 per cent and up to 1911/13 no more than 7 per cent. Furtherm ore, if a change in the term inal date so reduces the degree o f the so-called secular deterioration in raw material prices an even more striking change can be dem onstrated by choosing a different starting date. Between 1896/1900 and 1926/29 there was actually an im provem ent o f 3 per cent. Irrespective o f the indices used this shows how extrem ely cautious one should be before drawing any conclusions or constructing any theories based on a comparison of series, the term inal date o f which falls during the depression years of the thirties. By substituting the term inal date 1926/29 the deterioration is cut by half (20 per cent). We shall now see how far this deterioration corresponded to reality by examining the validity o f the price indices used.
ial com
114
(b) The validity o f the price indices Before proceeding to a critique of the index used by the League o f Nations I m ust m ention another estim ate o f the value of world trade in m anufactures and prim ary products, m ade by A rthur Lewis.3 An exam ination o f the basic data this author uses will reveal th a t the two estimates are almost similar (at least for the period 18701920). For wholesale prices between 18701920 the index is the same as th at o f the League o f Nations except some differences of weights between the two series used in constructing the total index of primary products (including food). For the prices o f m anufactures the series are the same up to 1913; after th a t date, instead o f using only the British export and im port prices for these goods, Lewis gives an average o f British and American prices. Therefore there are no great differences betw een th e tw o series. Thus according to the League of Nations figures th e relative deterioration in the prices of prim ary products for the period 1876/801936/38 is 43 per cent as against 31 per cent in Lewiss figures, while the deterioration for the period 1876/801926/29 is 20 per cent as against 13 per cent. Conse quently to simplify m atters I shall confine m yself to an analysis o f the League o f N ations estim ate, which is in any case m uch more widely known. In addition, the analysis will be limited to the period 18701929, so as to elim inate th e abnorm al effects o f the depression. The first, and a very surprising, point to be noted is th a t the price indices of world exports are only those for British exports: thus, if there is a deterioration in th e term s of trade this concerns only one country, n o t th e whole world. Let us now look at the indices themselves and consider how far the British figures can be used to reflect the world situation a hypothesis th a t cannot be excluded a priori. Beginning with prices o f m anufactures, we see th a t the index is an average o f the im port and export price indices constructed for G reat Britain by Schlote.4 In comparing this index with those of the other great economic powers it appears th at there is a rise o f 33 per cent registered for 18721928,5 b u t taking the average o f G reat Britain and the United States, as Lewis does, the rise is only 26 per cent, and according to Kindlebergers index for industrial E urope,6 we find th at there is actually a fall of 13 per cent. Therefore, insofar as an index o f world export prices of m anufactures is required at all, the one upon which the League o f Nations estim ate is based is obviously badly distorted. It seems m ore likely th a t during the period 18721928 world industrial prices remained relatively stable than
The Terms o f Trade
115
th at they rose by 33 per cent as the League o f Nations index indicates or fell by 13 per cent as Kindleberger would have us believe as his figures did not take account o f the U.S., for which there is a rise o f about 20 per cent. If therefore, as seems probable, there was relative stability in export prices o f m anufactures between 1872 and 1928 (expressed in gold values) rather than a rise o f one-third, the relationship between the prices o f raw materials and m anufactures would turn out to be radically different from th at commonly assumed. The prices used in the League of Nations indices o f prim ary products between 1870 and 1930 are those o f Sauerbecks index of wholesale prices o f prim ary products in Britain,7 corrected to allow for the fall in the value o f sterling between 1921 and 1925.8 Thus again the index is based on purely British figures. It shows a rise in prices o f 10 per cent (18721928). When this index is com pared with other indices o f wholesale prices for raw materials we see th at there was little difference between English and European prices (expressed in gold values), the English rising slightly m ore slowly. On the o th er hand the difference with the United States is m ore im portan t since prices o f raw materials there rose by about 20 per cent. It seems therefore th at 10 per cent may be too low an estim ate of the rise, and th at an interm ediate figure o f betw een 10 and 20 per cent should be adopted. Thus in global terms for the period between 1872 and 1928 we find (again, in gold values): (1) Relatively stable or slightly falling prices for m anufactures; (2) A rise of 10 to 20 per cent in prices o f raw materials. The conclusion is thus th at the terms of trade for prim ary products improved by between 10 and 25 per cent instead o f worsening by about 20 per cent. Even more significant than the disputable validity o f the price in dices o f m anufactures and raw materials, is another even m ore im p o rtan t distortion responsible for the impression of deteriorating raw material prices. Judging from an analysis o f the m ethods o f calcula ting the various price series for raw materials it appears th at these in dices are nearly all based on im port prices,9 which are c.i.f.; while those for m anufactures are based on im port and export prices thus only 50 per cent are c.i.f. The distortion introduced by this factor is im portant for the three following reasons: (1) Between the years considerably; 1870 and 1928 transport costs fell
ial com
116
(2) The share o f transport costs is obviously more significant for raw materials where the average specific value is m uch lower than for m anufactured goods; (3) Finally the share of transport costs in prices o f m anufactures is lower not only because of the latters higher specific value, but also because im ports o f m anufactures m ade in industrial countries move on average a m uch shorter distance than do prim ary products im ported into these countries. Consequently a comparison betw een tw o price series so con structed is, in the last analysis, directly affected by the fall in transport costs, since by comparing these tw o series we have on the one hand the price o f raw m aterials plus 100 per cent o f transport costs, and on the other hand the price o f m anufactures plus only 50 per cent of such costs. And on top o f this transport costs for prim ary products are proportionately higher than those for m anufactures. By using such a biased index it is possible to exhibit a relative worsening in the price o f raw materials. The bias is particularly m arked for Britain because locally-produced raw materials compared to total consum ption in th a t country were at a very low level and shrank considerably during the period. These elements confirm my hypothesis o f the im provem ent in the relative prices of raw materials a hypothesis which is directly opposed to the one generally accepted for the period 1876/801 9 26/29,10 and which I now propose to substantiate with additional argument.
II. A D D I T I O N A L P R O O F S O F T H E R E A L I T Y O F A S E C U L A R I M P R O V E M E N T IN T H E T E R M S O F T R A D E O F P R I M A R Y PRODUCTS
The first proof is based on an estim ate of the proportionate share of transport costs in raw m aterial prices for the period 1876/801926/29. The second entails a comparison of the prices of certain pairs o f commodities (prim ary products and manufactures) in three countries for the same period. The third, which is m etho dologically the m ost im portant, consists in presenting some data on the term s of trade o f under-developed countries, or those exporting raw materials, during the period 18701929. It is possible to com bine th e results o f these calculations w ith some others o f the term s o f trade derived from indices o f dom estic prices o f primary products com pared to those of m anufactures: they will be seen mainly to re-inforce those produced by our second argument.
The Terms o f Trade
117
(a) The part played b y transport costs in the prices o f primary products (1 8 7 6 /8 0 -1 9 2 6 /2 9 ) This is a field which has been very little studied; consequently the facts offered here are rather fragm entary. The problem will be approached in tw o ways: (1) from th e m acro-economic aspect by comparing world values o f the exports and im ports of raw materials, and (2) from the m icro-economic aspect by comparing the relative transport costs of some prim ary products. (1) A comparison of world values o f prim ary products im ports and exports Since im ports are expressed in c.i.f. prices while exports are expressed in f.o.b. prices, the difference between the world values of im ports and exports gives us an approxim ation o f the relative cost of transport and associated charges. But it is, o f course, a rather crude m easurem ent with a wide margin o f error, mainly because o f the incom plete nature of the statistics. Besides (and in the present context this is significant) th e fall in transport costs stim ulated growth in the trade in com m odities which had a low specific value. As a consequence the transport charges measured as a percentage of the total value o f im ports tended to fall m ore slowly than the absolute cost of transport. However, in spite o f this bias, which does not favour our thesis, it can be seen, if we look at existing statistics for the period under discussion (1876/80 to 1926/29), th a t the gap between the f.o.b. value o f world exports of prim ary products and the c.i.f. value o f world im ports o f similar goods fell from 23.1 per cent to 10.9 per ce n t.11 By way o f comparison we should note that the similar gap for m anufactures fell from 8.3 per cent to 4.2 per cent during the same period. These two proportions represent a relative fall in transport costs of 53 per cent, which is enough alone to explain a drop o f about 12 per cent in the im port prices o f prim ary products w ithout any fall in export prices. However, remembering th a t a drop in transport costs leads to a growth in the trade in goods with a low specific value, the percentages m ust be greater, and certainly great enough to account for a drop o f more than 15 per cent at the least in the price of im ports o f prim ary products w ithout a drop in export prices. What follows will confirm this conclusion.
ial com
118
(2) A comparison o f the relative transport costs of some prim ary products The figures in Table 35 enable us to appreciate the real fall in freight charges which occurred between 1876/80 and 1926/29. Of course freight charges are only one elem ent in transport costs since the cost o f insurance also enters into the total charge and this also fell m arkedly during the period. Thus we can see from the table th at in 1926/29 rice costing 227.0 shillings in Burma cost 284.5 shillings in Great Britain a difference o f 57.5 shillings o f which only 26.8 shillings were actual freight charges. It will be seen from the table th a t a m arked drop in freight charges occurred between the years 1870 and 1920. It would be pointless to try to extract a simple average from these scanty figures; however, they are sufficient to give us an idea of the size o f the relative drop in transport costs. Thus, we see that in 1876/80, in transporting coal from Great Britain to Constantinople, freight charges alone am ounted to 157 per cent o f the cost of the coal its e lfwhile freight charges to Hong Kong would have am ounted to 300 per cent. This percentage fell to 74 in 1926/29. For wheat transported from the United States to Great Britain the percentage cost fell from 22.2 to 5.1 per cent and for rice coming from Burma to G reat Britain from 84.3 to 11.8 per cent. Insurance12 and other costs would probably have increased the percentages by about 100 per cent for 1876/80 and 1926/29. These relative declines lend confirm ation to the earlier estim ate based on th e macro-economic figures. It appears probable th a t the drop in transport costs alone is responsible for the relative 1525 per cent fall in im port prices of raw materials from 1876/80 to 1926/29. For m anufactured goods with their higher specific values, freight costs played a smaller role. We can assume th a t the freights of m anufactures would be responsible for a drop of only 5 per cent. All things considered this makes it probable th a t the fall in transport and allied costs which occurred between the years 1870 and 1920 masked a relative 10 20 per cent rise in the export prices o f prim ary products com pared to those o f m anufactured goods. This rise, combined with the changes in the underlying price indices them selves, probably led to improved relative prices o f raw materials of about 20 to 40 per cen t.13 This estim ate is supported by the following brief analysis of price changes and by the analysis o f the term s o f trade for some specific countries which I shall m ake in section (c) below.
The Terms o f Trade
119
rable 35 Relationship between freight costs and prices o f selected primary products 1 8 7 6 /8 0 -1 9 2 6 /2 9
A verage 1 8 7 6 -8 0 WHEAT (U.S. S per bushel) (a) Price in U.S.A. (b) Freight from U.S.A. to G.B. % relationship b:a COTTON (U.S. S per 100 lbs.) (a) Price in U.S.A. (b) Freight from U.S.A. to G.B. % relationship b:a COAL (G.B. shillings per long ton) (a) Price in G.B. (b) Freight from G.B. to Constantinople % relationship b:a RICE (G.B. shillings per long ton) (a) Price in Burma (b) Freight from Burma to G.B. % relationship b:a SUGAR (G.B. shillings per long ton) (a) Price in Java (b) Freight from Java to G.B. % relationship b:a COTTON SEED (G.B. shillings per long ton) Freight from Alexandria to G.B. Average 1 9 2 6 -2 9 Variation 1 8 7 6 /8 0 1926/29
1.35 0.30 22.2%
1.78 0.09 5.1%
+ 32% - 70% - 77%
9 .3 1 0 .7 1 7.5% 9.7 15.2 156.7%

A
18.6 0.6 3.2%
+ 100% - 14% - 57%
17.1 12.6 73.7% 227.0 26.8 11.8%
+ 76% - 17% - 53%
70.0 59.0 84.3%
55% 86%
325.02 73.3 22.6%
380.02 26.4 6.9%
64% 69%
19.4
14.2
27%
11 8 5 5 -6 0 ^Prices calculated from prices in G.B., less freight costs plus handling, insurance and m iscellaneous charges. Sources: FR E IG H T (excluding c o tto n ), average calculated from tables o f m axim um and m inim um freight charges for 1 8 7 6 8 0 , given in the w eekly journal Fairplay o f 1920, and for 192629 in the annual supplem ents o f the sam e jou rn al fo r th e years 1 9 2 7 30. COTTON, D. C. N orth, T he E co n o m ic G row th o f th e U n ited S ta te s 1 7 9 0 1860 (E nglew ood, 1961), p. 258; derived from N ew Y ork Ship p in g a nd C om m ercial Freight Prices, W heat and C o tto n , H ISTO RICA L STATISTICS O F TH E U.S. (W ashington, 1960). COAL, B. R. M itchell and P. Deane, A b stra ct o f Rritish H istorical S ta tistics (C am bridge, 1962). RICE and SUGAR, 1 8 7 6 - 8 0 see coal; 1 9 2 6 29, International Y earbook o f A gricultural S ta tistics (R om e various issues).
120
(b) Comparative changes in price o f some raw materials and manufactures As we have seen, the overwhelming incidence of im ported products in the samples o f m ost wholesale price indices renders the latter useless as a measure o f the changes in export prices of raw materials. This bias is increased by the wide availability o f statistics relating to im port prices (especially in indices stretching back to the nineteenth century), which gives them a disproportionate im portance in the Table 36 Changes in prices o f some primary products and manufactures between 1876/80 and 1926/29
Average 1 8 7 6 -8 0 GREAT BRITAIN Coal: export f.o.b. (long tons, shillings) Pig iron: Scottish (long tons, shillings) FRANCE (Fr. tons) Coal Wheat Potatoes Meat4 Cast iron Steel rails Cem ent2 Hand tools3 UNITED STATES Coal: anthracite (short tons) Corn (bushels)1 Raw cotton (100 lbs) Raw wool (100 lbs) Steel rails (long tons) Nails (100 lbs)1 C otton cloth (100 yds) Average 1 9 2 6 -2 9 Variation 1 8 7 6 /8 0 1926/29
9.7 52.5 13.7 292 75 1,828 92 239 70
17.1 77.8 107.4 1,548 520 11,580 460 821 252
+ 86% + 48% +684% +430% +593% +533% +400% +244% +260% +407%
3.38 1.35 11.7 88.5 52.6 2.85 79.0
11.10 1.67 18.6 115.4 43.0 1.81 125.8
+228% + 24% + 59% + 30% - 18% - 36% + 59%
1 A djusted to allow for variations o f q u ality or ty p e o f p ro d u ct 2 1883 and 1926 respectively 3Averagc grow th o f w holesale prices o f 14 h an d to o ls for 1879 and 1881 com pared to 1926 and 1931; derived from th e individual series in F o u rasties b o o k , q u o ted below . 4 Averages o f beef, veal and pork Sources.- G R EA T BRITA IN , B. R. M itchell & P. Deane, A bstract o f British H istorical S ta tistics (Cam bridge, 1962). FRANCE, coal, w heat, p o tato e s and cast iron, A nnuaire statistique de la France, retrospective section; o th e r com m odities, J. Fourastie, D o cu m en ts p o u r I H istoire et la Theorie des p r ix (Paris, n.d. 1958?). U.S.A., H istorical S ta tistics o f the U nited S ta te s (W ashington, 1960).
The Terms o f Trade
121
price structure. Indeed alm ost all the raw m aterial price series available for the nineteenth century originate in statistics of foreign trade. Administrative requirem ents and the necessity of levying customs dues fostered greater sophistication in th e statistical apparatus o f foreign trade as com pared to other fields o f economic activity. The resulting distortion o f economic records explains the non-representative character o f m ost wholesale price indices o f th at era. Because o f this, in order to provide a supplem entary proof o f the hypothesis th at no relative deterioration in raw material prices took place, I have attem pted to assemble for each of three im portant countries (Great Britain, France and the United States) a series of prim ary-product prices which are not those o f im ports, and to com pare their developm ent with th a t o f series for m anufactured goods. The figures in Table 36 are rather few, as hom ogeneous series are not easy to find, b u t the extent to which they agree is very significant. It can be seen th at in each case the prices o f m anufactures have increased less than those o f raw materials. One could n o t attem pt to calculate a numerical difference on the sole basis of the commodities represented in the table, b u t th e facts confirm those p u t forward earlier which reveal an im provem ent in prim ary-product prices compared to those o f m anufactured goods between the third quarter o f the nineteenth century and 1926/29. (c) Data on the terms o f trade o f under-developed or o f primary-goods producing countries and on the terms o f trade o f primary products in domestic trade (18701929) The relatively small num ber o f examples quoted here m ust be stressed. The main reason for this paucity is the absence o f basic statistics and the very few researchers working on the problem Six cases of external term s o f trade and three cases of terms o f trade based on domestic price indices will be analysed. For terms of trade proper the analysis will deal with three non-European prim ary-product exporting countries: India, Australia and Canada. Around 1900 these three countries provided about 35 per cent o f the combined to tal exports o f under-developed and prim ary-product exporting countries.14 To these three cases we shall add those o f Norway, Finland and Denmark. This analysis being the best means o f determ ining the probable change in the term s o f trade o f the Third World, an attem p t will be made whenever possible to
122
push the inquiry back to 1870 (in some cases back to 1860). Now let us look at the facts. For India the data is taken from Bhatia's stu d y 15 which reveals th a t between 1866/75 and 1926/29 there was an im provem ent o f 31 per cent (between 1861/65 and 1926/29 o f 39 per cent) in the barter terms of trade'. The num ber and choice o f the commodities used to construct the index o f export prices (28 commodities) and im port prices (11 com modities) gives a real value to these con clusions. For Australia the im provem ent in the term s o f trade was no more than 7 per cent betw een 1870/74 and 1926/29,1 6 b u t in this case the index of im port prices used has n o t been clearly defined. The Canadian figures, which seem to have been calculated in a m ost com plete way, indicate th a t betw een 1869/73 and 1926/29 there was an im provem ent in the term s o f trade o f 84 per ce n t.17 Alongside these figures we can place those o f the term s o f trade o f some European countries specializing in the export o f agricultural goods and raw materials. Thus Norway's foreign term s of trade (excluding services) improved by some 55 per cent between 1867/73 and 1930,18 while for Finland the im provem ent between 1867/73 and 1909/14 was 44 per cen t.19 This probably indicates an average improvement betw een 1867/73 and 1926/29 o f m o re than 50 per cent. Finally in Denmark, where the agricultural products were more sophisticated, the term s o f trade improved by 39 per cent between 1875/79 and 1926/29.20 One need hardly add that the bias attendant on th e fall in transport costs also affects these figures, and in the examples quoted here w ould in fact exaggerate the improve m ent in the terms o f trade, although its effect would be reduced by the fact th at im ports consisted mainly o f m anufactured articles o f a high specific value. To move on now to an analysis o f some dom estic price indices. For France we have the index o f prices calculated by LevyLeboyer2 1 according to which the term s of trade betw een agricul tural and industrial prices improved by 6 per cent betw een 1866/75 and 1926/29 (by 17 per cent betw een 1856/65 and 1926/29). In Germ any2 2 the term s o f trade o f agricultural products compared to machinery and agricultural tools improved by 27 per cent and 41 per cent respectively for the same periods. In Japan, betw een 1878/81 and 1926/29, the im provem ent was about 330 per cent for prices in the agricultural sector compared to industry.2 3 In this last case the range o f com m odities considered is relatively limited, and the agricultural m arket was protected to some degree against the im port
The Terms o f Trade
123
o f cereals. Thus these figures confirm those put forward earlier. It goes w ithout saying th at there is no question o f drawing from these indications a more precise figure o f the degree o f the im provement of the term s of trade between prim ary products and m anufactures, but they provide additional confirm ation o f the results o f my own estimates suggesting a probable im provement between the years 1870 and 1926/29 of 20 to 40 per cent in the export prices o f primary products relative to export prices of manufactures.
B. The terms o f trade 19261970 We have seen that, contrary to generally accepted notions, the secular change of prices up to 1926/29 favoured primary products; and we have further concluded th a t there was an im provem ent of anything from 20 to 40 per cent in the export prices o f prim ary products relative to the export prices o f m anufactures. We m ust now enquire to w hat extent this tendency has been m odified in more recent times. I shall again om it the exceptional years 193039 from the analysis and concentrate on the years 1948 70. Valid (U.N.) series for term s o f trade o f the under-developed countries exist from 1948 onwards and reveal th a t the terms deteriorated after 1954. Before discussing this, however, we must determ ine to w hat ex ten t the price structure o f the early 1950s differed from th a t o f the years 1926/29. This is im portant because if the price structure in this period (compared to 1926/29) was characterized by a relatively high level o f the prices o f primary products, this would imply th at the fall after 1954 could have been an adjustm ent.
I. C H A N G E S IN T H E P R I C E S O F MANUFACTURES, 1926/29-1950/54 R AW MATERIALS AND
The Bureau of General Economic Research and Policies of the United N ations24 has calculated for some periods a price index o f raw materials exported from under-developed countries. Despite im perfections this index25 is nevertheless o f consider able value, largely because o f the num ber o f com modities included. From 1924/28 to 1950/52 the index o f prim ary-product export prices grew by 14 per cent, while th a t o f m anufactures fell by 10 per cent, giving a comparative im provem ent in prim ary-product prices of
124
The Economic Development o f the Third World since 1900 Table 37 Index o f export prices o f primary products from less-developed countries (1924/28 = 100: prices converted to constant 1934 U.S. dollars)
1934/38 Cereals Meat Bananas, oranges, tangerines Sugar Coffee, Cocoa, tea Tobacco Oilseeds and oils Cotton, wool, jute, sisal Rubber Non-ferrous metals Petrol, coal TOTAL M anufactures1
1 E xports to th e U.S.A.
1950/52 117 108 134 105 129 96 121 131 63 123 117 114 90
1959/61 89 139 121 86 102 109 103 81 50 114 117 95 98
34.6 46.8 49.4 36.6 29.1 47.9 32.2 32.1 21.8 41.1 41.6 34.8 46.3
Sources: Derived from S tu d ies in W orld E c o n o m y , Vol. 1 (U.N., 1963).
about 27 per cent. This should be lowered to about 24 per cent to allow for the great rise in prim ary-product prices caused by the Korean War in 1950/52. This estim ate m ust now be com pared with others. An index of prim ary-product prices calculated by the GATT26 shows an even more marked relative growth: about 50 per cent from 1926/29 to 1949/51, expressed in constant U.S. dollars. (It should be re m arked th a t GATT also used the export prices o f the U.S. to calculate prices of m anufactures.) Kindlebergers index for this period is of little significance since in 1952 the economy o f m ost European countries was still restricted by post-war reconstruction, and this was reflected in the price structure. On the other hand the great rise in prices occurring in m ost o f these countries did exert considerable influence.27 These factors explain the 11 per cent deterioration shown in this index. Lewiss index, however, shows an improvement of 14 per cent between 1928 and 1950, and changes in the domestic price index o f the U.S. also show a relative improve m ent of about 10 per cent for prim ary products com pared to manufactures. From another index o f export prices calculated by the statistical office of the United N ations,28 based on a weighted average o f national indices o f countries representing more than 85 per cent o f world trade, we find a relative improvement o f 21 per cent betw een 1926/29 and 1950/52 and of 24 per cent betw een
The Terms o f Trade
125
1926/29 and 1950/54. Exam ination of some individual prices of both primary and industrial products suggests a similar change. Consequently it is likely th at the U.N. estim ate of a 25 per cent improvement in export prices o f Third World primary products is realistic though perhaps slightly overestimated. The range was probably somewhere between 15 and 25 per cent. O f course, this applies specifically to th e period betw een 1926/291950. Compared to the years 1936/38 the im provem ent in prices of prim ary products would turn out to be still greater, since in the years between 1926/29 and 1936/38 the prices o f primary products deteriorated m ore m arkedly by com parison with those o f manufactures. Indeed, the prim ary product price index calculated by the U.N. indicates a rise in the price o f raw materials o f 228 per cent between 1934/38 and 1950/52, while for m anufactures this rise is only 94 per cent. This represents a relative im provem ent in prim ary-product prices o f 69 per cent. But if we direct our attention to the m ore significant change between 1924/28 and 1950/52 we also note th at the price level in 1950/52 (even excluding 1951) shows a very pronounced im provem ent in prim ary-product prices, i.e. as suggested above, o f about 15 to 25 per cent. In order to arrive at a m ore balanced judgem ent on the changes after 1950 we have to try to p u t the 1950/52 levels into a comparative perspective. With this in mind we m ust consider the changes which occurred betw een 1926 and 1950 in a wider context. In an earlier passage we estim ated that prim ary-product prices between 1870 and 1926/29 had improved by anything from 20 to 40 per cent, or by an average rate of 0.3 to 0.6 per cent per annum. Between 1926/29 and 1950/52 the im provem ent accelerated to a noticeably higher rate o f 0.6 to 0.9 per cent per annum. Thus if the changes occurring before 1926 were to be considered norm al it would follow that the level o f raw m aterial prices o f the year 1950 was relatively high. It is obvious th at such a m anner o f determ ining the norm al level of prices is arbitrary, b u t a truly objective definition o f a norm al price in this field is virtually impossible. Yet to neglect such a definition altogether raises difficulties, such as im plicitly assuming th at the 1950 levels was norm al. An acceptable solution, therefore, might be to regard the changes occurring between 1870/1926 as norm al, and from this to conclude th a t the level of prim ary-product prices was relatively high in 1950 by comparison w ith those o f industrial products about 5 to 10 per cent above the level to be deduced from an extrapolation of the long-term trend. Such an assum ption throw s a very different light on the fall which occurred from then onwards, allowing it to be seen, at least partially, as an adjustm ent of prices. The word partially is used
126
advisedly since, as we shall see in the following pages, the fall actually w ent far beyond the limits of such an adjustm ent.
II. C H A N G E S IN T E R M S O F T R A D E , 1 9 4 8 - 197 0
(a) The figures Thanks to improved statistical m aterial we have enough data for this period to measure the changes in the terms o f trade for countries or groups of countries as well as for groups of commodities. The availability o f this double series also allows us to confirm that the working hypothesis, in which the tw o series are regarded as equivalent, is valid for practical purposes at least till the early 60s. As far as changes in the terms of trade of developing countries are concerned, the years 1948 70 can be divided into three distinct periods. The first is from 1948 to 1955, and is characterized, apart from annual fluctuations such as the 1951 rise due to the Korean War, by relative stability both in developed and under-developed countries. The second period (1954/55 to 1962/63) is m arked by a considerable deterioration o f the term s o f trade of the under developed countries about 11 per cent between 1950/55 and 1962/63. During the same period the term s o f trade o f the developed countries improved by almost 10 per cent. Although these two contrary movements are connected to a large degree, the links are not entirely reciprocal, since while com m odities originating in the developed countries represented m ore than 70 per cent o f total im ports of Third World countries, the products from the Third World provided no more than 19 per cent o f total imports in developed countries. The third period, beginning in 1962/63 and going on till at least 1970 (and even until the beginning o f 1973), was m arked by relative stability in the movem ent o f the terms o f trade both in the third and the developed world. However, whereas in the Third World the trend was stabilized near the lowest level since the end o f the war, for the developed countries stability occurred at the top o f the curve during the same period. Moreover, the deterioration in Third World terms of trade since 1962 has been halted mainly by the growing part which m anufactures began to occupy in their exports. Thus, if one were to look only at the terms o f trade in primary products it would appear th at since 1962 the decline has been gently slowing down, b u t th a t there has been no actual stabilization. From 1952/54 to 1960/62 this deterioration was 15 per cent, while from 1960/62 to 1968/70 it was no more than 3 per cent. Naturally when we tu rn to individual countries we find th a t there are great divergences from the average figure quoted above, because on one hand export prices o f various raw materials have changed in different ways (not to speak of the actual size o f the swings in
The Terms o f Trade Table 38 Changes in the terms o f trade (1963 = 100)
LESS-DEVELOPED COUNTRIES E xport prices 1938 1948 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 34 109 102 128 117 109 111 111 110 1 10 105 103 103 100 97 100 103 102 104 103 103 106 109 Im port prices 43 113 91 107 109 103 99 101 102 105 101 98 99 99 100 100 102 103 103 103 102 105 109 Terms o f trade 80 95 112 119 106 105 112 110 107 104 104 106 104 102 98 100 101 99 101 100 101 101 100 DEVELOPED COUNTRIES Terms o f trade 99 95 91 88 91 94 92 92 93 92 96 97 98 99 100 100 100 100 100 101 101 101 102
127
Terms o f trade o f prim ary against manufactured products 1
128 128 116 114 119 115 111 111 105 104 101 98 97 100 101 98 97 95 96 96 94
1 T rade betw een developed and developing areas Sources: V arious issues o f Statistical Y earbook and M o n th ly B ulletin o f Statistics (U .N .); for recen t data M o n th ly B ulletin o f April 1973.
prices), while on the other, some countries possess a highly specialized foreign trade, with the result th a t swings in either direction tend to be more extreme. R eturning to the average for the Third World, we may assume th at a m arked fall in the term s of trade was a phenom enon limited mainly to the period 1952 62, and it was accompanied as already stated by a recognizable im provement in the terms for the developed countries, which in turn led to an even more marked comparative worsening o f the terms of trade between the two worlds. The divergence was about 14 per cent from 1950/54 to 1961/63; and the deterioration was even worse for the terms o f trade of raw materials compared to m anufactures (exchanged between under-developed and developed countries), reaching 19 per cent between these dates. Since the price level for raw materials in 1950/54 was quite high (from 5 to 10 per cent higher than the extrapolated curve of the
Copyrighted
128
long-term trend) we m ust assume th a t perhaps one-third or one-half of the drop was in fact an adjustm ent, as defined earlier in this chapter. Consequently we may conclude th at during this period there was a real fall of about 10 per cent in prices o f prim ary products compared to m anufactures, which led, given the structure o f foreign trade, to a similar real drop in the terms o f trade o f the under-developed countries. This fall may be thought to be rather exceptional, given the earlier changes and the economic level o f the period, thus justifying the interest taken in the problem by the United Nations Conference on Trade and Development (UNCTAD), which was held in Geneva in 1964 b u t had been convoked in 1961. For the less-developed countries themselves the consequences of the adverse trade balances were especially serious because amongst m anufactures it was capital equipm ent th at suffered the greatest rise in prices.29 The worsening in the terms o f trade between prim ary products and capital equipm ent was 38 per cent30 between 1954 and 1962 in the exchanges between developed and under-developed countries. Such a deterioration obviously posed formidable problems for m ost under-developed countries for whom, as we have seen, foreign trade forms a very im portant part o f their national product. On the basis o f this worsening o f the term s o f trade calculations can be made showing statistically th at the loss o f resources which ensued form ed a very im portant fraction o f the financial aid granted to the Third World. Taking the extrem e case (i.e. accepting that relative price levels in 1954 were norm al and th a t the prices of prim ary products compared to capital equipm ent worsened by 38 per cent) in 1962 there was a loss of resources by the under developed countries o f about 11 billion dollars, or about 130 per cent o f the total financial aid received by them in the same year. (According to existing data see Chapter 10 the am ount o f aid in 1962 can be put at about 8.5 billion dollars.) O f course, such an estim ate distorts reality. But even if we assume a more realistic figure of 12 per cent as the degree o f deterioration in terms of trade, we still arrive at an estim ated loss o f resources in 1962 o f around 3.5 billion dollars, or about 40 per cent o f the total aid (generously defined) given to under-developed countries during this year. For 1970 the loss could be estim ated at 5 billion dollars, or about 35 per cent o f aid. It is obvious th at this situation had a very strong im pact on the development possibilities o f the Third World, and fully deserves the interest which has been displayed in it. In the following pages the causes o f this deterioration in the terms o f trade will be examined briefly.
The Terms o f Trade
129
(b) A brief review o f the possible causes fo r the worsening terms o f trade Within the fram ew ork o f the present analysis I do n o t intend to study the causes of the deterioration in depth, b u t m erely to present a straightforw ard critical exam ination of the principal factors which have been advanced in explanation o f it.3 1 The several theses may be summarized in the following six propositions: 1. The demand for raw materials com pared to th a t for m anufac tures is relatively inelastic. 2. Technological progress has reduced the input coefficients of raw materials in m anufacturing industry. 3. The im port o f raw materials into the developed countries has been reduced by restrictive measures. 4. Synthetics have been developed and are p u t to ever-increasing use. 5. The allocation o f the gains accruing from a rise in productivity have differed in their effects on prim ary products and manufactures. In m anufactures the effect was rising prices and higher rem uneration of factors o f production; in raw materials the effect was to lower prices and to bring about a standstill or a reduction in rem uneration o f factors o f production. 6. The supply o f raw materials greatly increased. It is obvious th at a com bination o f some o f these factors is possible, and it has been suggested th at this was the case. Before passing to the separate exam ination o f each explanation, I should point out that m ost theories have been based on the hypothesis which postulates a secular deterioration in the relative prices o f prim ary products. But, as we saw earlier, there was, on the contrary, an im provement in these prices up till 1950/54. By eliminating the hypothesis o f a deterioration we should be able to scale down some o f the explanations, i.e. numbers 1, 2 and 3 and (at least partly) num ber 5. Indeed, a relative inelasticity of dem and for raw materials, a continuing reduction in the input coefficients o f raw materials and restrictions on im ports have existed since the beginning o f the industrial revolution and have not up till now (or had not at least until 1950/54) caused a relative worsening o f the prices o f raw materials. Moreover, we cannot exclude the possibility th at the same factors might have prevented a greater im provem ent in the relative price of primary products than the one th a t actually took place. On the other hand there are some im portant industrial products, especially semi-finished ones, whose share in finished products has
130
The Economic Development o f the Third World since 1900 Table 39 Changes in the export prices o f primary products (for all non-communist countries) 1958 = 100
Variation 1 9 5 0 /5 4 1961/65 7% 32% -1 1 % 9% -1 3 % -2 0 % -2 1 % + 3% + 3% + 3% - 8% 8%
1950/541 Food products Coffee, tea, cocoa Cereals N on-food agricultural products Fats Textiles Wool Minerals Metals Fuels GENERAL INDEX Natural rubber
1 E x c l u d i n g 1951
1961/65 98 76 102 103 97 108 116 93 102 91 982 98
105 111 115 118 111 137 147 90 99 88 106 107
2T he follow ing arc the changes in th e index for recent years 1963 1964 1965
100
103 103
1966 1967 1968
104
101 100
1969 1970 1971
104 108 115
Sources: E xcluding rubber, derived from various issues o f th e M o n th ly B ulletin o f S ta tistics (U .N .). F or recen t data see the B ulletin for M arch 1973. R ubber: a u th o rs estim ates based on ex p o rt prices o f ru b b er from th e follow ing countries: Malaya, T hailand, Ceylon and Nigeria.
decreased while their price has increased. This is especially true of iron and steel. Finally it should also be rem em bered th a t the elasticity o f demand for cereals is less than th a t for coffee, tea or cocoa, although the prices of the second group fell much more heavily than those o f th e first group (see Table 39). Similarly there are industrial products for which the dem and is very inelastic, b u t whose prices have nevertheless risen. To facilitate a critical exam ination of point num ber 4 I have worked out in Table 39 the variations in the export prices of different groups o f raw materials between 1950/5432 and 1961/65. The figures suggest th at the introduction o f synthetics did not by itself constitute a sufficiently valid explanation o f some o f the price falls. Primary products such as fibres, textiles, metallic metals and rubber which were directly threatened by these products have decreased less quickly than, or in a m anner n o t fundam entally different from, other products such as coffee, tea and cocoa for which a synthetic substitute does not exist. The fact remains that j
The Terms o f Trade Table 40 Changes in the share o f synthetics in world output o f textile fibres and rubber (1000 tons)
1938 TEXTILE FIBRES Total natural1 and man-made fibres Cellulose synthetic fibres2 Non-cellulose synthetic fibres3 Total man-made fibres Synthetics as % o f total RUBBER4 Total natural and synthetic Synthetic Synthetics as % of total
1 C o tto n , w ool (w ashed), flax and silk 2 R ayon, fibranne 3 N ylon, orlon, perlon, etc.
131
1948
1958
1970
9,300 880
880 9%
10,300 1,150 34 1,190 11%
14,900 2,270 420 2,690 18%
22,200 3,440 4,930 8,370 38%
931 6 0.6%
2,150 600 28%
3,520 1,550 43%
8,800 5,900 67%
4 E xcluding China Sources: A u th o rs estim ates derived from Statistical Y earbook and M o n th ly B u lletin o f Sta tistics (U.N.)-, P roduction Y earbook and M o n th ly B ulletin o f A gricultural E conom ics a nd Sta tistics (F.A .O .) and T e x tile Organon. (F o r natural fibres th e figures are five-year averages centring on th e chosen year, e x c ep t for 1970 w hen th e average is fo r th re e years.)
this factor m ust have had some effect since the expansion of synthetics has been very great. Table 40 above provides evidence for some such effects in the case of textile fibres and rubber. It should be remarked here th a t because they are lighter and more durable, a given quantity o f man-made fibres corresponds to a greater quantity o f natural fibres, so as long as we base our percentages on weights we are apt to under-estimate the real share o f man-made fibres. In addition to man-made fibres and synthetic rubber we m ust also consider synthetic detergents which com pete with fats, as well as plastic m aterials33 whose characteristics enable them to some extent to replace leather, wood and metals. As the share of these products before the war was extrem ely low and as today they form a significant, n o t to say the larger, fraction o f ou tp u t, it is obvious th at the prices o f the raw materials for which they are substitutes were bound to be affected. But this is not an adequate explanation by itself since, as we have seen, a deterioration in price had also occurred for m ost o f the products for which there are no substitutes. There remain two other explanations to analyse, num bers 5 and 6, i.e. the difference which exists betw een the under-developed and the developed countries in th e allocation o f the gains from increasing productivity and the great increase in the supply o f primary products.
132
The difference in allocation of productivity gains can be only a partial explanation because, as we have already noticed, a similar struc ture had also existed in the past w ithout lowering prices; and secondly because it is hardly likely th at the difference between the increase of productivity in tropical agriculture and th at o f m anufacturing industry in developed countries between 1954 and 1962 would have corres ponded to the deterioration in the term s o f trade which had taken place, i.e. 19 per cent. But these conclusions m ust once m ore be carefully scrutinized in relation to prevailing commercial policies. The developed countries have sought in the past, and seek now, to keep down raw m aterial prices: such policies would have more chance o f success if resistance and im pedim ents on the part o f prim ary producers were not sufficient to prevent the gains from productivity being translated into lower prices. The resistance was lacking m ostly because o f the small degree o f trade-union pressure and the absence o f well-organized com binations o f sellers. This thesis is the one p u t forward by Prebisch in particular and is implicit in the creation o f UNCTAD (United Nations Conference on Trade and Development). We m ust also call to m ind the figures compiled here for the chapter on foreign trade, from which it emerges that in 1970 Third World exports destined for the developed countries form ed 11 per cent o f their G.D.P. while developed countries exports destined for the Third World form ed only 2 per cent o f their G.N.P. We should note moreover th a t pressure o f this nature on the p art o f the developed countries ought not to be considered purposely contrived, b u t m erely as the logical consequences o f a m arket econom y and a liberal system o f international trade. What also supports the thesis connecting the worsening o f the term s o f trade with the different allocation o f gains in productivity accruing to developed and under-developed countries is the differences in price changes of com m odities with similar functions b u t different origins. The prices o f such com m odities produced in the Third World com pared to these o f com m odities produced in the developed countries clearly differed. Thus th e price o f beet sugar rose or remained stable while the price o f cane sugar fell. There were the same changes to be observed in the price o f fats a rise in b u tter, a fall in oilseeds and fruits a greater increase in th e fruits o f the tem perature zone and a slighter increase for tropical fruits. These changes are based on com putations o f price variations betw een 1952/54 and 1961/63 based on F.A.O. data made for 21 com m odities or groups o f commodities. A t the m ore general level the indices o f export prices, calculated separately by the U.N. for the developed and under-developed nations, show a similar trend. The general index o f raw materials and sub-groups has changed m ore favourably for the developed than for the under
The Terms o f Trade
13 3
developed countries.34 Of course the support policies for agricultural prices in the developed countries are usually the cause o f these differences, b u t the very existence o f such support policies confirms the validity o f the hypothesis o f a differentiation in the way the gains from productivity have been distributed. It should be noted, however, th at the com position o f the exports o f these groups of com m odities is very far from being identical in the developed and under-developed countries. To the extent th at political independence in a large num ber of colonial territories has coincided with the fall in prices we may be justified in enquiring how far independence increased the pressure of the purchasers o f primary products. Given the new political structure, the dangers (social as well as political) o f a fall in the rem uneration o f factors o f production have no longer concerned the buying countries, while this was certainly not so under the colonial regimes. This may seem a paradoxical hypothesis but it is one m ost w orthy of consideration. We come now to the last point, that of increases in supply. We can see th at in fact the o u tp u t of primary products in the under developed countries grew enormously. Where minerals and fuels are concerned, we saw in the chapter on the extractive industries that o u tp u t rose from an index num ber o f 12 in 1936/38 (100 = 1963) to one of 108 in 1963/65, i.e. an increase o f 800 per cent. For agricultural export crops the average increase in o u tp u t between 1934/38 and 1963/65 can be estim ated to be about 106 per cent.3 5 As the domestic consum ption o f these com modities in the lessdeveloped countries is in most cases very low, we can, roughly speaking, take the volume o f their domestic production to represent their supply on th e international m arket. On this basis and judging from the size o f the increased supply, it appears more than likely th at there occurred an excess of supply over demand, and th at this is a valid explanation of the fall in prices. The foregoing critical exam ination of the explanatory factors, although extrem ely brief, does allow us to suggest a variety of tentative answers. As so often happens in the social sciences, an explanation by one single factor is too simplistic. Reality can more often be seen in the conjunction o f several causes, and this is certainly the case here. We may conclude th at the relative worsening of prim ary-product prices which occurred after 1954 resulted from a com bination o f th e following factors, in order o f decreasing im portance: a great increase in the supply o f prim ary-products from
134
the less-developed countries; a difference betw een the less-developed and the developed countries in the m anner in which the gains from increased productivity accrued; the perfection and high o u tp u t o f a wide range o f synthetic commodities. Three other factors listed here also in order of decreasing im portance had a m uch m ore limited influence: the relative inelasticity o f dem and for raw materials compared to m anufactures; the reduction o f the input coefficient o f raw materials in m anufac turing industry due to technological progress; measures restricting the im port o f prim ary products into the developed countries. Research on the problem s of the term s of trade done since the first edition of the present study appeared shows proof on the one hand of the reality o f the secular improvement of the term s of trade of prim ary products up till 1950/54 and, on the other hand, highlights a factor capable o f explaining, at least in part, the break in the trend after th at date. The factor, which so far as I know has not y e t been taken into account as an explanation of the phenom enon, is the profound change which has recently occurred in the rate in the growth o f productivity in agriculture as compared to industry. In fact, up to 1940 in the U.S. and the fifties in Europe, productivity (b o th o f labour and o f o th e r fa c to r s) increased m u c h m o re rapidly in industry than it did in agriculture; while since then the opposite has been tru e.36 Such a radical change is obviously capable of causing a deterioration, even if only a partial one, in the term s of trade of agricultural com m odities in the tem perate regions as com pared to m anufactures. But as we do not have any valid figures for the changes in the productivity of export crops o f countries of the Third World it is difficult to be sure w hether this factor also occurred in their,case. We have seen th at for food crops productivity stagnated. But it is possible, even probable, th at the trend in productivity was different for ex p o rt crops because the prevailing systems of cultivation facilitated the swifter and m ore general introduction o f im portant innovations in agricultural equipm ent, in selection of plants, and fertilizers, pesticides and herbicides. As there are no figures for productivity in these sectors a large question m ark hangs over the causes o f the new trend a trend which could be m ost harmful to the Third World if it should turn out not to be due mainly to this last cause.
7
INTRODUCTION
The L evel of Education
In this short chapter, which is to some extent outside the main subject of our enquiry, I shall attem pt to diagnose briefly the changes in a field which is not strictly speaking economic b u t which nevertheless interacts with the econom y in an im portant way. If the take o ff in the m odern developed countries was able to get under way w ithout being handicapped by the low level o f literacy of the population, it was because conditions at the beginning of the nineteenth century were so totally different. Industrial technique was largely improvised and based on simple empirical grounds. Today conditions are totally different: science plays an overwhelming part in technology, and thus in econom ic and industrial life. For this reason the level of literacy has a m uch greater im pact now than it did in the early nineteenth cen tu ry ,1 and this is why of late, particularly in the last two decades, so much emphasis has rightly been placed on education in developing countries. This chapter is an attem pt to measure how far these efforts have been successful. The first part o f the chapter deals briefly with long-term changes in the level o f literacy and prim ary schooling; in the second part more recent changes in secondary and higher education are discussed. The delay in compiling educational statistics has m eant th at we cannot in all cases include, as we have done in other chapters, figures for very recent years. A. The level of literacy and prim ary schooling As statistics for the period before 1950 are so few and unreliable, I have sought to measure short- as well as long-term changes through differences in the level of illiteracy registered for the various age groups. Thus in Table 41, besides the level of illiteracy in the population over 15 years of age I have com puted, on the basis o f the last census, the level o f illiteracy for some older age groups. This provides a fairly valid estim ate o f the situation in earlier years. Before looking at the changes I shall com m ent briefly on the differences in the levels o f 1960. Leaving aside the differences, sometimes quite large ones, existing between countries, we should note the superior position o f the Latin American countries, in which the level o f illiteracy is half that in the Afro-Asian and Middle Eastern countries. We should also note that in this respect the Middle 135
136
East does not occupy the interm ediate position between Latin America and Asia which it does in certain other fields. In spite o f the weight o f population increase reflected in the enorm ous rise in the num ber o f children of elem entary school age,2 great progress in education has been made in m ost of the Third World. A comparison of the percentages of illiteracy for the age Tabic 41 Changes in the rate o f illiteracy
% o f illiteracy amongst over-15s* 1950 AFRICA Madagascar Morocco Nigeria Senegal Tunisia AMERICA Argentina Brazil Chile Cuba Mexico Peru Venezuela ASIA Ceylon India Pakistan Indonesia Philippines Thailand MIDDLE EAST Egypt Iraq Iran Turkey 1960 % o f illiteracy by age group in 1960 census 1 0 -1 4 2 0 -2 4 3 5 -4 4 4 5 -5 4 5 5 -6 4
66.5
86.2
67.2
85.0
87.4
87.8
8 8.55

94.4 9 84.27 8.6 16.4 16.2 34.6 39.4 9 34.2
88.6 12 73.4
93.0 85.0 11.1 28.0 29.4 25.1
95.3 87.5
95.6 89.2 20.0 38.0 10 47.3 46.1
88.9 97.1 91.5
13.3 50.6 19.8s 2 2 .16 43.2
47.8 37.0 80.74 8 1 .14
5.2 9.7 28.4 2 6 .314 22.9 1 2
7 .113 16.6 3 3.310 42.3 38.9
15.7 13 25.9 4 4 .9 10 52.3 53.5
40.03 48.02
32.36 72.29 81.29 57.19 2 8 .19 32.39
57.7 72.3 27.9 27.1 14.3
66.4 76.9 42.5 15.4 16.0
74.6
7 8 .2 11
62.6 33.4 39.6
73.4 40.2 56.7
80.1 52.0 68.8
80.12

68.1
80.5 85.5s 87.27 61.9
59.6 67.4 71.7 44.8
77.3 78.0 84.1 49.1
80.8 89.0 88.5 63.7
84.7 90.6 91.4 73.3
86.3 93.6 92.5 83.0
E x c e p t Peru + 17 years; M adagascar, Senegal, A rgentina +14 years; Tunisia +10 years; Nigeria +7 years; Mexico (1950) + 6 years. 1 1946 2 1947 31948 4 1951 s 1952 61953 7 i956 81957 9 J96i 103 0 39, 4 0 49 and 5 0 - 5 9 years respectively 1 *4559 years 1 2 1 5 - 1 9 years 1 3 1 4 - 2 9 , 3 0 - 3 9 and m ore than 50 years respectively 1 4 1719 years Sources: Derived from figures collected from D em ographic Y earbook (U .N .), for 1955, 1963 and 1964, and Statistical Y earbook (UNESCO ), for 1964 (Paris, 1966).
The Level o f Education
137
groups 10 to 14 years and 20 to 24 years, shows a great improvement. Thus for India th e percentage fell from 66.4 to 57.7 per cent; the fall for other Asian countries is less obvious, b u t is nevertheless substantial. The Philippines, where we can see a marked increase, are som ewhat o f an exception. However illiteracy in this country is lower than in other Asian countries. A great drop in the percentage o f illiterates can also be seen in the African and the Middle Eastern countries, particularly in Egypt where the level of illiteracy for the 20 to 24 year-olds is 77.3 per cent and for the 10 to 14 year-olds is 59.6 per cent. Progress in Latin America has been m ore m oderate, and this can be explained by the relatively high absolute level o f literacy. Over th e longer term it can be shown by comparing the levels of the age groups over 35 years of age th a t educational progress in earlier periods was much slower than it is at present. One should remember, o f course, th a t since there is a correlation between the level of education and standard o f life there is usually a higher m ortality among illiterates. This results in an under-estim ation o f the level of illiteracy among age groups above 45 years, and reduces the apparent extent of real progress. Allowing for this factor the changes in recent years (say from 1948 to 1960) are significant compared to the earlier period (say from 1910 to 1948). The years 1948 60 witnessed great progress with the pace quickening betw een 1969 70, as we shall see. For the whole non-comm unist, under-developed world the estim ated level of illiteracy amongst the population of 15 years and over fell over the long term as follow s:3 1900 1930 1950 1960 1970 80 76 74 65 56 per per per per per cent cent cent cent cent
These average levels, o f course, conceal profound differences between countries and even between the wider regions. In 1970 the level o f illiteracy was highest in Africa with 76 per cent (86 per cent in 1950), followed by Asia with 50 per cent (78 per cent in 1950) and by Latin America with 24 per cent (42 per cent in 1950). Progress will appear even m ore rapid when numbers of pupils enrolled in prim ary education are considered. Even though too m uch im portance should n o t be attached to these figures, they reflect, nonetheless, real changes. In 1950 there were slightly fewer than 65
138
million pupils enrolled in prim ary schools in the non-communist under-developed countries; by 1960 this figure had risen to 119 millions, and by 1970 to 202 millions. According to my calculations and estim ates for the same region the num ber o f pupils was about 7 million in 1900, 11 million 1910 and 37 million about 1937/38. A t the regional level the change was marked by an equalizing trend in the rate of school attendance (i.e. the proportion of children attending school). It was in Africa where th e rate o f school attendance was the lowest th a t the num bers of pupils and the rate of school attendance grew the fastest; in Latin America which had the highest starting level, progress was the slowest. Around 1970 the rate of attendance at the elem entary level for the whole Third World was about 57 per cent (31 per cent in 1950 and 45 per cent in I9 6 0 ); for Latin America the figures were 78 per cent (46 per cent in 1950); for Asia 56 per cent (32 per cent in 1950) and for Africa 43 per cent (18 per cent in 1950). This progress is amazingly rapid, much m ore than in the developed countries at any stage in their development. During the nineteenth century in Europe the annual rate of growth in the num ber of pupils at the elem entary level was about 1.8 per cent.4 The annual rate of growth in school attendance can thus be estim ated to be 1 per cent, or less than one-third o f th a t of developing countries at the present day. It is a harder m atter and a somewhat arbitrary exercise to try to make comparisons between the enrolm ent ratios since in this respect we can observe great differences which had nothing to do w ith the corresponding stages o f econom ic development. Thus in the mid nineteenth century Germany, Switzerland and m ost o f the Scandinavian countries had rates o f school attendance well above th at of England, which was economically further developed.5 The differences become progressively less visible with the app roach of the tw entieth century because, w ith the widespread introduction in Europe o f compulsory elem entary schooling around 1880, school enrolm ent ratios tended rapidly to approach 9598 per cent.6 It can be estim ated th at this rate o f 95 98 per cent had been reached in alm ost every European industrialized country in about 1910. The level of school enrolm ent in the rest of the under developed world in 1970, i.e. 57 per cent, is the same which Europe (excluding Russia) achieved in about 1860. At th a t time the percentage o f the labour force em ployed in agriculture was about 5560 per cent as against a little below 70 per cent for the under-developed countries today, and the proportion of the popu lation em ployed in m anufacturing industry was about 1820 per cent as against 10 per cent for the under-developed countries now.
rial com direitos autor,
1 39
Such an extrem ely swift progress in elem entary schooling created problems by its very speed: in particular it encouraged a rural exodus, as we shall see in the next chapter. The pace of progress and the present level o f education docs not mean th a t all problems o f the adult working population in the primary or even secondary sectors have been solved by this means. Given that tertiary em ploym ent, especially in the public services, absorbs a high proportion of the educated, it follows, that, in spite of educational levels of about 60 per cent, levels o f illiteracy o f about 60 80 per cent (excluding Latin America) will be likely to be found amongst those em ployed in industry and especially agriculture for a long time to come. This rate is high enough to obstruct the diffusion of innovations into those sectors. Thus any effort aimed at adult education should clearly receive top priority. Such an effort to concentrate on adult education seems to have been attem pted in China, at least in some periods. But it is difficult to distinguish at all clearly how far it was successful, since, as in the field o f primary education in general, we possess no figures for any period more recent than 1958 when we know that some 90 million children attended prim ary schools. This would form a school enrolm ent ratio of about 60 per cent (compared with 43 per cent in the non-comm unist under-developed countries). It appears probable th a t the ratio has not noticeably increased and that the campaign for increased literacy undertaken in the popular communes during the Great Leap Forward obviously did not succeed and has not been pursued with the same enthusiasm since th en .7 Although it is rash to hazard figures, it is probable th at the level of illiteracy in the adult population o f China (about 1970) lies somewhere between a minimum of 25 per cent and a maximum of 45 per cent. This is definitely lower than the level for non-comm unist Asia which for the same period is 50 per cent, although the disequilibrium between the urban and rural proportions is probably less marked in China. Thus, there is no d o ubt th at im provement is taking place in the field o f prim ary education and it deserves to be recorded because, as I remarked earlier, it reflects a great educational effort. However, it has to be pointed out th a t the diagnosis made here deals solely with the quantitative side of education. There are obviously qualitative problems which are impossible to analyse within the framework of the present study. These arise as much from over-crowded classes, the over-expansion of school population and the shortage of qualified teachers as from the content o f w hat is taught. The curriculum is often badly adapted to local conditions because it is closely copied from th at o f the form er colonial powers and was
140
intended for pupils w ith a different m entality and standard o f living. The last point would not of course apply to China. This problem together with the one related to the inefficacy o f the school system, which is becoming increasingly serious, form ed the m ajor themes at UNESCOs recent international conference on education.8 In fact it is not only th a t the content o f prim ary education using te x t books from form er m etropolitan countries is so often badly adapted to local needs, b u t also th a t its effect is limited. A large num ber o f pupils never finish their course of study. The drop-out rate is considerable and rises, for example in Brazil and Algeria, to as m uch as 60 per cent, with a consequent increase in the cost per pupil completing his education. B. Secondary and higher education Progress in literacy and prim ary education has been followed by even m ore rapid progress in education at the secondary and higher level. Details about the changes in the total num ber o f students in these tw o educational levels will be found in Table 42. In educational establishm ents up to secondary level there were 43 million students in 1970, i.e. nearly six times more than in 1950. To p u t it in another way the num ber was not far short o f th a t in the western developed countries (55 million), while in 1950 there had been only one-third as m any. The Third World will surpass the western countries in absolute num bers o f students by 1975. But in term s of attendance rates the gap is still an im portant one; in 1970 the rate is about 26 per cent in the Third World and about 75 per cent in the developed countries. As is the case with other indicators, regional differences are considerable, as also are differences between countries. Ranking is about the same as for the econom ic factors. The attendance rate is highest in Latin America (about 37 per cent in 1970), followed by Asia (about 26 per cent) and lowest in Africa (about 17 per cent). In general, the levels of attendance achieved by the Third World in 1970 are much higher than those in the industrialized countries at the time when they had attained a similar stage o f development. Thus a secondary education attendance rate o f about 26 per cent was only achieved by the developed countries in the decades between 1930 and 1950, i.e. at the m om ent when their per capita standard of living was about six to eight times as high as th a t in the currently less-developed countries. This shows how pressing is the problem facing the under-developed countries regarding the need for m odern technology.
The Level o f Education Table 42 Changes in numbers o f students at secondary and higher level o f education (in thousands)
1950 SECOND L E V E L 1 Less-developed countries 5 Africa Asia Latin America China Developed countries 3 World5 TH IRD L E V E L 2 Less-developed countries5 Africa Asia Latin America China Developed countries 3 World5 7600 480 5410 1710 1310 30500 38000 940 40 620 280 139 5380 6300 1960 18200 1660 12190 3890 85204 45700 63900 2100 140 1390 570 900 9090 11200 1970 42500 4420 27250 10720
141
70700 113200 5600 350 3740 1520
20440 26100
1 G eneral, vocational and teach er training 2 Universities and o th e r in stitu tio n s o f higher education 3 Including co m m u n ist countries 4 1958 5 E xcluding C hina (m ainland), N orth K orea and N orth V ietnam Sources: D erived from UNESCO Statistical Y earbook 1972 (Paris, 1973).
As it was in relation to prim ary education, the relative nature of the figures and the problem of the content of the education has also to be considered. The relative nature of the figures emerges mainly from the fact th a t the num ber o f pupils actually completing their secondary studies is very low, well below th a t o f the average rates in the developed countries. The problem o f the educational content is similar to th a t found in prim ary education. But, in addition, as we shall see in Chapter 9, there is the problem in urban areas of high unem ploym ent amongst young people with secondary education. Expansion has been even m ore rapid in education at the third, mainly university, level. From about 900,000 in 1950 (350,000 in 1937), student num bers in the non-communist under-developed countries rose to 2.1 millions in 1960 and the figure stands a t 5.6 millions in 1970. To this m ust be added the 200,000 students who are following university courses in the developed countries.9 The regional differences are, as in other fields, considerable (see Table 42). In spite o f such a very rapid expansion the gap betw een the
142
enrolm ent ratios in the Third World and the developed countries still remains a large one: a ratio of about 1 to 6 in 1970. But here also even m ore than for other types of education, the rates achieved in 1970 were, relatively speaking far higher. In fact, an enrolm ent ratio of 4 per cent, which was th a t of the Third World in 1970, had not been achieved in m ost European countries until after the second world war. It had o f course been reached much earlier in the U.S. (about 1920). Valid international comparisons o f this kind are, however, rather difficult to make for a num ber of reasons. In m any countries and disciplines such a rapid expansion o f higher education produced a serious disequilibrium betw een supply and dem and, encouraging qualified students either to emigrate, or at least not to return to their own countries if they had received their education abroad. This is the notorious brain drain which has been such a controversial topic in recent years. If the problem of students n o t com pleting their courses is less serious for higher education than for the lower levels, th at o f the orientation o f the studies themselves has proved to be a continuing one. In fact the total num ber o f students in technical and scientific disciplines is quite small. There are no world statistics available, b u t
from c o m p u ta tio n s I h ave m ade m y s e lf b a sed on 101 u n d er
developed countries whose students form m ore than 95 per cent o f the total, the distribution as a percentage o f the total in 1966 68 was as follow s.10 Literature, education, fine arts Law, social sciences Exact and natural sciences Engineering Medicine Agriculture 40.3 22.2 6.0 15.3 9.8 4.5
Comparisons between regions are not very useful in this field, since local needs are so different. However scientific disciplines absorb about 45 per cent o f students in the western European countries, about 65 per cent in th e East European countries, while we can see from the list above th at the figure in the under-developed countries is only 36 per cent. But there is a further problem o f adjusting supply and demand which, even in the developed countries, is proving hard to solve. Not the least part of the problem is th at it creates yet another vicious circle in the Third World. Few scientists are produced since the demand for them is limited, while the setting up of
143
enterprises needing highly qualified personnel is handicapped by the shortage o f specialists. So far as China is concerned such data as is available from 1950 to 1960 will be found in Table 42. During this period, and in quantitative term s, progress was clearly faster than in the remainder o f Asia. But we know th a t the cultural revolution (beginning in the autum n o f 1965 and intensified in 1966) brought the whole problem of secondary and higher education into question and even led to an alm ost com plete closure o f educational institutions at these levels. It was only in 1967 th at such education was gradually re-started.1 1 On the basis o f adm ittedly exiguous inform ation it appears th at student num bers in 1970 were fewer than in 1960. Furtherm ore the kind o f education was very different to w hat was form erly given, or is given now in the West, or even in the Soviet Union. This fact, combined with the shortage o f inform ation, makes further enquiry into such questions very uncertain.
8
INTRODUCTION
Urbanization
Urbanization in its early m anifestations is m ore often treated as a factor in the progress of civilization than as an outcom e o f economic developm ent which made it possible. But historians, looking at the ebb and flow of urban population from the early Dark Ages onwards, saw th a t it provided a useful way to measure the rise and fall of prosperity in western Europe. With the coming of the industrial revolution the issue grew in size and kind: in size because in a large economically integrated u n it1 urban population for the first time could exceed 10 to 15 per cent o f total population to reach at first 30 per cent and then 50 per cent within a historically short period; in kind too, because the links betw een urban population on one hand and industrialization and econom ic developm ent on the other were becoming closer and m ore direct. It is true th a t there is not always perfect correlation betw een the level of economic developm ent and the degree of urbanization. Many diverse factors, geographical and social as well as political and historical, play their part in the pace and form taken by the concentration o f population in an urban setting.2 Thus, by choosing, say, pairs of countries with a similar general structure within the confines o f Europe we can see a t least three cases in which the m ore urbanized country is the less economically developed.3 But after about 1930 a new phenom enon which m ight be term ed urbanization w ithout industrialization began to appear in the Third World. This phenom enon very rapidly acquired an inflationary character and in the early sixties began to present m ost serious problem s of urban unem ploym ent and under-em ploym ent. A discus sion o f these particular problem s will, however, be deferred until Chapter 9. In th e first section o f this chapter I shall examine the question o f urban developm ent in the narrow sense of the term ; in the second section I shall consider the question o f w hether the towns of the Third World are really, as they are sometimes described, cities that came too soon.4 The causes o f the rapid acceleration in the process o f urbanization will then be reviewed in the third section. The analysis of the m any problems raised by the different types o f urban centres in the various under-developed countries and the m anner of their form ation will be om itted. It is difficult to give a short account 144
Urbanization
1.45
of the research done in this field because of the enorm ous num ber o f such studies. By 1966 at least 300 scholars had devoted themselves to the problem o f urbanization in Africa alone,5 and Verhaegen gives no less than 2,544 references in his bibliographical study relating to Africa published in 1961.6 All the three sections will be confined to the non-comm unist under-developed world for the simple reason that the problem s raised by the growth o f tow ns in com m unist countries are so totally different. However, I shall as before, take a very brief look at the changes which have taken place in China (section D below). The consequences o f all these changes will be discussed in the next chapter which is devoted to the problem s o f em ploym ent. The basic m aterial upon which this chapter is based (with the exception of section D) comes from a study I recently com pleted at the request o f th e International Labour Office.7 Readers with a special interest in these particular problem s m ay find there a more detailed discussion o f the issues and m any m ore references than I have had room for here. A. Changes in urbanization Thanks to recent studies by the D epartm ent o f Economic and Social Affairs of the United Nations, we have at our disposal a fairly exhaustive statistical analysis o f changes in urban population in the different countries and regions of the w orld.8 But as the figures do not go farther back than 1920 it is impossible to make use o f them for a comparison between the pace o f urban growth in under developed countries and developed countries at similar stages of development. For this reason I have had to com pute some figures for Europe covering the period 18501920 by adopting the same criteria to define urban population as those used in the United Nations survey.9 Two indicators are used in Table 43 to measure the pace of urbanization in different regions. The first is simply the annual average rate of growth of the urban population. However, this indicator (which is widely used) operates under a shortcom ing here because to a great extent it reflects the differences in the pace of growth in total population. The second indicator shows the average rates of growth in the level o f urbanization, i.e. change in the percentage o f urban population as com pared to total population. Between 1920 and 1960 urban population in under-developed countries grew at an average rate o f 4.0 per cent per annum, i.e. at a rate which was twice as high as th a t of the developed countries in a
146
Table 43 Average annual rate o f growth o f urban population and rate o f urbanization (percentage)
1 8 5 0 -1 9 2 0 GROWTH OF URBAN POPULATION Non-communist, less developed countries Africa America Asia Developed countries 1 GROWTH OF RATE OF URBANIZATION2 Non-com m unist, less developed countries Africa America Asia Developed countries 1 1Including East European countries
2 R a t e o f u r b a n iz a t io n = p e r c e n t a g e o f th e t o ta l p o p u l a t i o n liv in g in a g g l o m e r a t io n s o f
1 9 2 0 -1 9 6 0
1 9 6 0 -2 0 0 0
4.0 4.3 4.4 3.8 1.9
4.1 4.6 4.2 4.0 1.7
2.5
1.5
2.3 2.6 2.1 2.2 1.1
1.6 1.9 1.2 1.6 0.8
20,000 inhabitants or more

Sources.- A fter 1920 derived from G row th o f th e W orld's Urban and Rural Population, 1 9 2 0 2 0 0 0 (U .N ., 1969). Before 1920, a u th o rs figures (see te x t).
similar period (even when the exceptional pace registered in this field by the U.S.S.R. is included). The divergence is even m ore noticeable if we look at variations which occurred in the rates o f urbanization. Urban population in th e Third World countries grew in succeeding decades by the following annual rates: 19201930 19301940 19401950 19501960 1 9 6 0 -1 9 7 0 2.9 per cent 3.4 per cent 4.1 per cent 5.1 per cent 4.1 p e r c e n t10
The concentration o f population in towns, like m ost phenom ena which have an upper limit, is a logistic curve. Consequently to be significant all comparisons m ust relate to similar stages o f develop m ent. This has been attem pted in Table 44. Before passing to a short com m entary on the figures in this table, tw o points m ust be made. The first concerns changes betw een 1960 and 1970. The figures for 1970 are projections m ade by the U.N.
jU - X
l_ lik . J A l
i X A
Urbanization
147
departm ent o f Economic and Social Affairs. These assume a marked slowing down in the pace of m igration to towns in the less-developed countries. Although we m ust o f course await the results o f censuses taken around 1970 for definitive figures, it nevertheless transpires, from such partial data as we have, th a t such a slowing down is m ost unlikely to have occurred. The alternative figures which I p u t forw ard in to Table 44 have been com puted on the basis o f changes which will probably occur in the 34 countries forming 53 per cent of the total population in the under-developed world. The second po in t concerns changes in the urbanization o f under-developed countries before 1920. The few figures at our disposal suggest th at the pace of urbanization before 1920 was m arkedly slower than between th a t date and 1930. Table 44 Comparative changes in rates o f urbanization 1850 2000
Rates o f urbanization 1 % age Less developed countries 3
_

Average annual growth o f rate o f urbanization % age Less developed countries 3

_

Europe 1850 1880 1900 1920 1930 1940 1950 1960 19702 19802 20002 15.0 22.0 31.0 34.7 37.2 39.5 40.7 44.2 47.1 49.5 55.0
Europe
_
6.7 7.8 9.7 12.9 16.7 19.7 (21.0) 23.2 31.7
1.2 1.7 0.6 0.7 0.6 0.3 0.8 0.6 0.5 0.5
1.6 2.2 2.9 2.6 1.6 (2.2) 1.6 0.8
P e rc e n ta g e s o f th e to ta l p o p u latio n living in agglom erations o f 2 0 ,0 0 0 o r m ore in h ab itan ts 2 U n ited N ations predictions. T h e figures in brackets are th e a u th o rs estim ates for 1970. 3 N on-com m unist Sources: See Table 4 3 , and tex t.
When in about the years 1860 70 Europe had reached a level of urban concentration o f 15 per cent the rate was increasing by 1.2 per cent per annum. In the under-developed countries at the same stage, in about 1950, the increase was at the rate o f 2.6 per cent. The divergence was even greater in the earlier stages. Thus in Europe, not counting England and Wales, urbanization rose from 11 per cent to 16 p e r cent betw een 1850 and 1880, which is an increase o f 1.3 per cent per annum , while at the same stage in the under-developed countries the rate was 2.8 per cent. It should be said here th a t there
148
is a tendency to exaggerate the effect emigration had on relieving demographic pressure in Europe. Between 1850 and 1880 the net balance o f emigration from Europe can be estim ated at no more than 6 per cent o f the to tal increase in population; and, if the United Kingdom is excluded, the percentage falls to 4 per cent. This means th at E uropes population, excluding the United Kingdom, would in 1880 have been only one to two per cent higher had the possibility o f emigrating n o t existed. We can, thus, conclude th at the pace of urban concentration in the under-developed countries was in reality very rapid and not at all com m ensurate w ith th at o f the European countries at the time when they had reached a similar stage in their economic development. B. Urbanization and levels of econom ic developm ent The rapidity of urbanization in the Third World one is tem pted with reason to describe it as urban inflation poses an im portant question. Was it, or was it not, the consequence o f the speeding up of econom ic growth? The tw o phenom ena had in the past been closely connected; and some developed countries, notably Russia, had experienced an even higher pace of urbanization than have the under-developed countries precisely because o f the very rapid transform ation o f their economies. We m ust repeat the reservations already made about correlating levels o f development and urban ization: but, nonetheless the conclusion dem onstrated in m any studies, both geographical and historical, is th a t such a correlation does exist, if the comparison is restricted to economies o f one particular type. We now have to see w hether the m ovem ent of population into towns which has recently taken place in the Third World can be considered as conform ing to a traditional pattern or w hether it should be treated as a genuinely inflationary process producing a state of hyper-urbanization. 1 1 Table 45 shows comparative changes in urbanization rates and proportions of the labour force em ployed in m anufacturing industry both for Europe (excluding England) from 1850 to 1930 and for the under-developed countries from 1920 to 1970. The percentage o f the population em ployed in m anufacturing is considered to be the least objectionable o f the straightforw ard indicators o f the level o f industrialization, when making comparisons on the historical and international plane. In continental Europe the level o f urban concentration remained lower than th at of em ploym ent in m anufacturing industry until
A i X a
Urbanization
149
about 1890, i.e. the m om ent when the percentage o f persons em ployed in m anufacturing had reached 18 per cent. Although the level of urbanization did n o t exceed the level o f m anufacturing Table 45 Comparison between the levels o f urbanization and percentages o f the active population employed in manufacturing industry 1
Percentage o f urban population (A) EUROPE2 (excl. England) 1850 1880 1900 1920 1930 Percentage o f active population in Relationship manufacturing industry (B) o f A to B
11 16 24 29 32
16 18 20 21 22
+ + +
30 10 20 40 45
LESS DEVELOPED NON COMMUNIST COUNTRIES 1920 6.7 7.8 1930 1940 9.7 1950 12.9 1960 16.7 19703 19.7 19704 21.0 AFRICA 1960 13.4 32.8 13.7
8.5 8.5 8.0 7.5 9.0 10.0 10.0 7.0 14.5 9.0
- 20 - 10 + 20 + 70 + 85 + 100 + 110 + 90 +125 + 50
AMERICA 1960 ASIA 1960
1 B rought up to th e nearest u n it for rates o f urban p o p u latio n and percentages o f active E uropean p o p u latio n ; to th e nearest half-unit for percentages o f active p o p u latio n in the less-developed c o u n tries; and to th e nearest five u n its for th e differences. 2 E xcluding U.S.S.R. 3 U.N. projection 4M y ow n estim ates Sources: U rban p o p u latio n rates, see T able 4 3 ; active p o p u latio n , a u th o rs estim ates.
em ploym ent until then, from th a t tim e onwards the gap betw een the tw o widened progressively until at the present m om ent it is nearly 100 per cent. In the less-developed countries this stage (i.e. equal level o f urbanization and o f m anufacturing em ploym ent) was reached between 1930 and 1940, at a tim e when the percentage o f workers in m anufacturing industry was less than 9 per cent. Thus the gap between the tw o levels widened to 100 per cent in less than 30 years whereas in Europe a similar change had been spread over m ore
150
than 80 years. The degree o f urbanization in under-developed countries was the same in 1960 as it had been in Europe in 188085 when the latters percentage o f working population engaged in m anufacturing was twice th at o f the less-developed countries. Some part o f the gap arises from the differing structure o f foreign trade in the tw o regions. Whereas, about 1890, Europe had (as it still does) a favourable trade balance in the exchange of m anufactures, under-developed countries have, for the whole period under consider ation, had a large deficit in this field. Commercial activity associated w ith the exchange o f such a large quantity of m anufactures and raw m aterials obviously requires additional urban working population, b u t the addition w ould be fewer than th a t needed to produce the im ported m anufactures locally. To enable comparisons to be made, therefore, the percentage of the labour force engaged in m anufactur ing in the under-developed countries would have to be increased by a certain proportion. Exactly w hat the proportion should be is very difficult to determ ine, b u t it m ay be assumed to be roughly 20 per cent. This adjustm ent would raise the level of the labour force engaged in m anufacturing in 1970 to 12 per cent of total population. Thus, even allowing for such an adjustm ent, the gap between the position o f the under-developed countries in 1970 and E uropes position at the m om ent when it was at a similar stage of urban concentration would still remain substantial. The gap is no less evident if we use o u tp u t per capita as a yardstick.12 Here is the com parison expressed in this way: Level of urbanization Under-developed non-comm unist countries C ontinental Europe Period National product per capita (1970 US $ and prices)
21
21
1970 1890
340 650
The gap in th e level o f income per capita at approxim ately similar stages o f urbanization is about 90 per cent, which is very m uch m ore than any margin o f error which can be attributed to the d ata.13 The comparisons could be refined even m ore, b u t the conclusions would be no different because insofar as the Third World is concerned the situation is one o f genuine hyper-urbanization. An analysis o f the causes of such a situation will be made in section C. But first o f all we m ust take note of the very im portant regional differences in urbanization levels, which were evident as
A i X a
Urbanization Table 46 Levels o f urbanization in noncommunist less-developed countries (percentages o f total population in agglomerations o f 20,000 or more inhabitants)
Africa 1920 1930 1940 1950 1960 19701 19801 20001 4.8 5.9 7.2 9.7 13.4 16.0 20.0 28.3 America 14.4 16.8 19.3 25.1 32.8 37.8 43.1 53.6 Asia 5.6 6.6 8.6 11.6 14.5 17.1 20.6 28.1 TO TAL 6.7 7.8 9.7 12.9 16.7 19.7 24.2 32.6
151
P r o je c tio n s Sources: Derived from G row th o f the W orld s Urban and Rural Population, 1 9 2 0 - 2 0 0 0 (U.N., 1969).
early as 1920. Regional data can be found in Table 46 and data for individual countries in the synoptic table in the appendix. C. Causes of urban inflation It will be recalled th at this chapter is mainly based on a study which I undertook at the request o f the International Labour Office. What follows is a summary of the second chapter in th a t study. The search for causes o f urban inflation can be narrowed down to an investigation o f the exodus from the countryside, because, even though natural population growth in cities had an im portant role to play, immigration has had an even greater impact. Thus we can estim ate th at betw een 1950 and 1970 migrations from rural areas have been responsible for 45 to 55 per cent o f the increase in urban population in the non-communist, under-developed countries. The influence o f migrations has had even greater significance on the size o f the working population since from 55 to 65 per cent o f the growth in the num ber o f urban workers resulted from this cause. Classical theories about m igrations have always underlined the significance of the so-called push-pull factors; and amongst the push factors over-population in agricultural areas takes pride of place. If this factor is one o f the m ost plausible explanations for internal or overseas migration in industrial countries, it is obvious th at in view of high demographic pressure in the under-developed world, it would play an infinitely greater part there. International and regional differences would lend additional weight to this factor.
152
Which regions w ould be m ost likely to be subject to over-population cannot be revealed by simple m easurem ents based on agricultural area per farmer. Nevertheless it is a reasonable presum ption th a t the tw o phenom ena are correlated even though the m any other variables prevent us from isolating th e correlation. If over-population o f agricultural land has pride of place amongst the push factors, the higher level o f urban incomes forms an essential elem ent in the puli factors.14 An analysis of the three m ain m ethods of measuring the gap (and its changes) betw een incomes in tow n and country (national accounting, wages and household budgets) shows indeed how im portant it has been. The average difference between wages o f agricultural and unskilled industrial labourers is roughly 80 to 150 per cent. If average agricultural wages are com pared with average industrial wages the difference is between 100 and 200 per cent. But if we take incomes, the difference is only about 60 to 120 per cent. Here it should be pointed o u t th at these figures designate not the extrem e limits o f the statistical spread on these three differences (and least o f all the upper limit) but merely the approxim ate range o f the average difference. It is difficult to be sure how the gap betw een rural and urban
in c o m e s changed ov er a p erio d o f tim e . M o st stu d ies, w ith o u t a lw a y s
advancing soundly based argum ents,15 conclude th at the divergence increased. So far as I have been able to test this hypothesis, there is a distinct probability, although no absolute certainty, th a t it is correct. The sizeable gap between urban and rural incomes is peculiar to the under-developed countries, since it is much smaller, and even in some cases non-existent, in the m odern industrialized countries after their take off period. W ithout going into th e reasons for this here, it is clear th at the dualistic structure o f m ost under-developed economies is th e m ost im portant cause. The general adoption of m inimum wage legislation appears also to have contributed to this situation. The m aladjustm ent o f education to real econom ic needs, both in a quantitative and a qualitative sense, is an alm ost universal problem b u t is accentuated in the Third World by three factors: the advanced state o f contem porary technology; widespread illiteracy; and a legacy of an educational system modelled too closely on th a t o f the erstwhile m etropolitan power. Rural emigration is hastened by this educational m aladjustm ent, as is confirm ed by m any studies. Rene D um onts conclusion is typical: The num ber of young people who are prepared to go back to the land after m ore than three or four years in the classroom is infinitesim al. 16 The following figures give,
A i X a
Urbanization
153
for one p art o f the Ivory Coast, the percentage o f persons intending to emigrate, according to their educational level.17 Illiterates Those who can read and write Those with school-leaving certificates Men 8 42 61 Women 11 55 75
And it is more than a declaration of intent, because over 60 per cent o f young people between the ages of 15 and 29 who had graduated from school had already left the area while those who remained considered their stay to be no more than a tem porary one. The connexion between educational level and propensity to emigrate, especially to towns, emerges from m ost studies into the causes of migration in the Third World, although it is a characteristic which is not, o f course, peculiar to under-developed countries alone.18 The significance o f this factor is also obvious when we recall from the previous chapter how immensely rapid expansion o f education in the Third World has been. Among other factors favouring a rapid movement o f population into towns the following m ust be singled out: the im pact of colonialism, as well as the opposite process; the inherent attractions of tow n life; and the existence in rural com m unities of social constraints o f one sort or another. The latter certainly m otivates a substantial num ber o f migrants, although it ought n o t to be thought o f as peculiar to the Third World alone, and therefore as a significant cause o f its urban inflation. However, it cannot be totally ignored, particularly as the special character o f many traditional rural societies in the Third World is often p u t forward (wrongly in my opinion) as a contrast to contem porary society in the West. It would be better if the comparison was drawn betw een traditional societies everywhere. But this point of view certainly does not mean that some social constraints, such as tribalism, do not favour rural emigration. The existence o f tribalism makes migration betw een rural com m unities harder. It indirectly encourages m igration into cities which are o f course de-tribalized com m unities par excellence. The im pact o f colonialism on the process o f urbanization is a complex one, because it is the result of phenom ena working in opposite directions. As has been rem arked above, until the 1950s increasing im ports o f m anufactures by causing handicrafts in the Third World countries to shrink helped to p u t a brake on the growth of urban centres. In m any countries this state o f affairs was re-inforced by administrative action taken by colonial authorities to
ial com
or
154
limit immigration into cities. On th e other hand the creation of colonial adm inistrations and the developm ent o f export crops were factors which led not only to the creation o f new urban centres,19 but also to the growth of older towns. Furtherm ore the im position by m any colonial governments of certain duties and taxes tended to favour migration. Again, in some countries, for instance in the Maghreb, th e policy o f colonial settlem ent led to the indigenous population being replaced by Europeans and this favoured a drift from th e countryside. In general it appears th a t until 1920 the working o f contrary forces m ore or less counter-balanced each other because th e level o f urbanization could n o t be considered excessive by th e standards o f the Third World as a whole. But the compensating effects did not, o f course, operate at the same level in individual countries. The breakup of colonialism, on the other hand, seems quite unequivocally to have stim ulated the m ovem ent o f population into tow ns which was the result o f five main causes: 1. The breakup o f colonialism was accom panied alm ost every where by an excessive administrative expansion 2. Policies of industrialization, especially those based on im port substitution, caused industrial em ploym ent to increase more rapidly than hitherto. 3. In nearly all countries in which colonial powers had instituted adm inistrative measures to lim it emigration into selected urban areas, such measures were abolished or m ade m ore flexible after independence. 4. The balkanization of some colonial empires, o r parts of empires, encouraged urban growth by creating new political and adm inistrative capitals. It should be noted in this connex ion th at in the case o f form er capitals whose geographical range of influence was reduced this gave rise to unem ploym ent. 5. In some cases political troubles caused refugee movements which enorm ously increased the population o f some cities. This brings us to th e last factor which has been term ed the inherent attraction of tow n life'. The im portance o f this factor has often been com m ented on, and rightly so. From the Greek city state to to d ay s megalopolis via the medieval city the life o f th e tow n has attracted people in all walks o f life, each m an or wom an finding in it some satisfaction th at country life could not provide a t all or could provide only partially. But w hat makes th e m odern situation so different, in the West as well as in the Third World, is th a t prodigious
ial com
Urbanization
155
advances in com m unications have enabled the existence of urban facilities and satisfactions to be m ore widely know n and therefore to be much more attractive. Aspirations towards w hat is thought to be a b e tte r life m ust often play their p art in the decision to move to a city, w hether or not such an aspiration can be fulfilled. By abandoning rural society for th at o f the tow n, peasants hope to leave an undeveloped world, often confused with an inferior civilization and to join an advanced world equally wrongly associated with a superior civilization. The com bined effect all these factors have on population movements is heavily influenced one way or another by such considerations as the existence o f channels o f m igration, geographical distribution o f urban nuclei, types of agriculture, etc. We have finally to return to the natural growth o f urban population. This was very rapid in the Third World because o f two sets o f factors. To begin with there were the factors which explain general population growth in the Third World, i.e. a large reduction in m ortality due to the use o f medical techniques w ithout a parallel reduction in the birth rate. To these general factors we may add another: contrary to w hat happened in western countries, a move to a tow n m eant in m ost Third World countries an increase in the expectation o f life. In the West during the nineteenth century, and even in some cases during th e early decades of the present century, a move to the city m eant a considerable reduction in th e expectation of life. The difference in the death rates by ages betw een urban and rural societies was so pronounced th at one could almost say th at city life itself became a brake on urban expansion. Throughout the nine teenth century (and before) m ortality was higher in urban areas, this was due to a variety o f reasons (amongst which living and working conditions were th e m ost im portant). Differences betw een countries make it difficult to lay down any average figure for differences in m ortality, b u t it m ust lie somewhere approxim ately w ithin a spread o f 20 to 40 per cent.2 0 A fter all, was n o t the city considered a couple o f hundred years ago to be th e graveyard o f the countrym an?2 1 The high deathrate affected even small children, and its prevalence in city life was one o f the factors th at delayed a general fall in m ortality.22 In fact a higher urban m ortality is still to some extent present in some western countries, although the differences betw een tow n and country at the present tim e are very slight. In under developed countries, on th e other hand, because m odem medicine is m ore efficacious in an urban setting, urban m ortality is noticeably lower than rural m ortality.
Material com
156
Thus, as we have seen, there are m any im portant factors which help to explain the growth o f tow n population in th e Third World. It has not been possible w ithin the scope o f th e present brief survey to enquire into th e causes and the pace o f urbanization in individual countries since this would obviously have taken us into too great detail. However, regional evolution and projections were shown in Table 46 and recent levels o f urban population for all countries for which figures are available will be found in the synoptic table (see appendix). The following section provides a very rapid sketch of urbanization in China. D. Urbanization in China The first point to notice is th at at the beginning o f the century China was far m ore rural than the rest o f the Third World, m ost probably because it had never been directly colonized. Thus, in 1900 it possessed no more than eleven cities o f m ore than 100,000 inhabitants w ith a to tal population of about 6 million people altogether. On th e other hand, India, another relatively rural country w ith only 70 per cent of Chinas population, had eighteen cities o f m ore than 100,000 inhabitants, totalling 5 million people. The population living an all agglomerations o f 20,000 inhabitants and over form ed 5 per cent of total Chinese population in 1920, rising to 6 per cent in 1930 and to 9 per cent in 1950; this latter figure compares w ith one o f 12 per cent for India and one o f 13 per cent for the whole of the non-com m unist under-developed world. Two phases can be distinguished in the period betw een 1949 and 1970. The first, which came to an end w ith the G reat Leap Forward, was m arked by rapid growth. B y l9 6 0 population in tow ns o f 20,000 inhabitants o r m ore is estim ated to have been 13 per cent o f the total, com pared w ith 14 per cent for India and 17 per cent for the non-com m unist Third World. Between 1949 and 1961 the urban population o f China rose at an average rate o f 5.5 per cent per annum ,2 3 i.e. at a pace not so very different from th a t o f the rest o f the Third World. A fter 1961 62 it became m ore and m ore apparent th a t a new policy was being p u t into effect which, to begin w ith, i.e. during the first years years after the G reat Leap Forward when agricultural production fell, took the form o f returning m any tow nspeople to the villages. From th a t time on the accent was on th e need to re-distribute population in favour o f the countryside and to establish industrial production in a large num ber of rural centres. Because o f this policy and the measures taken to prom ote it, helped
ial com
Urbanization
157
by th e small difference (n o t m ore than 910 per cent) betw een urban and rural incomes, it is probable th a t there was actually a reduction in th e share o f urban population. In absolute terms the num ber o f people living in cities in 1970 was m ost probably near w hat it had been in 1960, o r even slightly less. Therefore, in view o f the growth in to tal population it is likely th a t th e rate o f urbanization (defined as agglomerations o f 20,000 inhabitants or m ore), which for 1970 can roughly be estim ated at about 1012 per cent, actually fell;24 th e com parable rate for India was 16 to 17 per cent, and for the whole o f th e under-developed, non-com m unist w orld was 20 to 22 per cent. Likewise it is possible to say th a t while the non-com m unist, under-developed world was becoming a towndwelling society w ith o u t industrialization, China was making strenu ous efforts to achieve industrialization w ithout urbanization.
The Labour Force &
INTRODUCTION
This chapter will be lim ited to the non-com m unist less-developed countries, firstly because o f the lack o f valid statistics for China and other Asian com m unist countries, and secondly because the prob lems of em ploym ent and unem ploym ent in these countries are so very different from those of other countries. I shall divide the chapter into tw o sections, one dealing with the labour force as a whole and the other with the questions raised by unem ploym ent and under-em ploym ent. But before I begin to analyse the figures, I should point o u t how relative and arbitrary all notions o f em ploy m ent and unem ploym ent are when applied to societies o f the Third World. The relative notions familiar to m ost readers are based on conditions which prevail in industrialized societies and differ in m any respects from those in under-developed countries. I shall have to return to this point in greater detail at the beginning o f section B. A. The labour force W ithout going into greater detail, we should note th a t the generally accepted definition of labour force, or active population, refers to those persons who are either gainfully em ployed or else unem ployed b u t seeking work, and it does n o t include students, housewives, those living on their own means, or retired people. The inclusion of the armed forces varies from country to country, b u t in the m ajority o f cases and for the purposes of international statistics they are included. One o f the main causes of lack o f uniform ity, however, is the inclusion or exclusion o f some wom en in th e agricultural labour force in consideration o f which the exact criteria to be used are a m atter o f nice distinction.
I. P A S T C H A N G E S A N D TOTAL LABOUR FORCE FUTURE PROJECTIONS OF
The increase in the size o f the labour force in the under-developed countries has been less than th a t o f total population mainly because the base of the age pyram id is getting larger as a result o f the population explosion. From 1900 to 1950 the size o f th e economic158
ial com
The Labour Force & Employment
159
ally active population in th e non-communist, under-developed countries grew at an average rate of 0.8 per cent per annum. From 1950 to 1970 the rate rose to 1.7 per cent (Africa 1.6 per cent; Latin America 2.5 per cent; Asia 1.7 per cent). These are extrem ely high rates. By way o f comparison th e rate by which the European labour force rose during th e nineteenth century was only 0.7 per cent per annum and the rate o f growth during the second half o f the eighteenth century was even lower. Because o f these extrem ely high rates the Third World labour force m ultiplied 2.5 times betw een 1900 and 1970 rising from some 270 millions to some 660 millions. In the decades to come growth is likely to be even faster. Thus, according to recent I.L.O. projections1 the labour force will probably increase at an annual rate o f 2.4 per cent betw een 1970 and 1985, resulting in th a t period in a 43 per cent enlargement o f the labour force in the under-developed countries. Grow th as betw een wider regions is likely to be more uniform than it was in the past: 2.3 per cent in Africa, 2.4 per cent in Asia and 2.6 per cent in Latin America. This great expansion o f th e labour force cannot be m odified by any o f th e contingent measures for population control since em ployable persons who will enter the labour m arket betw een 1970 and 1985 were already born in 1970. Thus there are certain to be great problem s ahead, although of course it is possible th a t our estim ate of th e exact scope o f the problem is subject to a margin o f error, more especially to mistakes in projecting death rates for adults and children and activity rates (i.e. th e percentage o f active population to to tal population) for different groups. But the margin cannot be very large, and the annual rate of growth o f the labour force between 1970 and 1985 will alm ost certainly be w ithin the range o f 2.12.7 per cent, equivalent to a global increase o f betw een 37 and 50 per cent in fifteen years. In terms o f jobs required, however, this growth presents enormous problems. Some 130 to 170 million new jobs will have to be created between 1970 and 1985 in order to absorb the increase in the labour force alone, w ithout making any inroads into the existing num bers of the unem ployed or under-em ployed. We shall have to return to this problem at a later stage.
II. T H E S T R U C T U R E O F T H E L A B O U R F O R C E BY B R A N C H O F ACTIVITY
Details o f changes in th e structure o f the labour force by branch of activity in th e Third World (i.e., as always throughout the chapter, in
160
the non-com m unist Third World) will be found in Table 47. The reader's attention is drawn to the inevitable margins o f error, m ostly arising from gaps in basic statistics,2 and it m ust also be pointed out th a t the figures for 1970 are only provisional since the results of censuses made in 1970 71 are n o t y et available fo r a certain num ber o f countries. Table 47 Changes in the structure o f the active population o f the less-developed countries between 1900 and 1970, according to typ e o f em ploym ent
1900 Percentages: Agriculture Extractive industries M anufacturing industries Construction Trade, banking T ransport, comm unications Services TO TAL A bsolute figures (millions): Agriculture Extractive industries M anufacturing industries C onstruction Trade, banking T ransport, com m unications Services TO TAL 1920 1930 1950 I9 6 0 1970
77.9 9.8
12.3 100.0
77.6 0.4 8.5 1.0 5.4 1.6 5.5 100.0
76.6 0.4 8.5 1.1 5.4 1.8 6.1 100.0
73.3 0.6 7.6 1.8 5.8 2.0 8.9 100.0
70.7 0.6 8.9 2.0 5.9 2.2 9.6 100.0
66.0 \ 13.0 1 > 21.0 1 100.0
213.0 26.5
2 3 8 .0
2 4 9 .0
3 0 4 .0
3 6 6 .0
4 3 5 .0
33.5 273.0
1.1 26.0 2.9 16.4 4.9 16.9 306.0
1.3 27.7 3.6 17.6 6.0 19.9 325.0
2.5 31.5 7.2 24.2 8.3 37.0 415.0
3.2 46.0 10.6 30.8 11.4 49.9 518.0
, 85.0 1 x 140.0 1 660.0
Sources: P. B airoch, La structure d e la p o p u la tio n active du Tiers-M onde, 1 9 0 0 1 9 7 0 ; b u t th e absolute figures have been revised so as to m ake th em correspond w ith th e revised p o p u latio n figures (Chap. 1). N ote: T he small e x te n t to w hich the figures have been ro u n d ed up does n o t im ply any corresponding m argin o f error.
(a) The structure b y branch o f activity around 196070 Before examining the main changes in the structure o f the Third W orlds labour force by branch of activity and discussing the problem o f how to absorb surplus agricultural labour, we shall analyse very briefly the situation as it has been in recent years.3 As might be expected it reflects the low level o f econom ic development in 1970. The labour force engaged in agriculture is slightly less than two thirds o f the total: in the developed countries a com parable proportion was so occupied at the beginning o f the nineteenth century, i.e. a century and a half earlier. But for more than a century
161
the Third World has been in an active economic relationship w ith the already industrialized regions, w ith the result th a t the econom ies of the various less-developed countries have been heavily affected and distorted. The distortion has been passed on to the structure of em ploym ent and its m ain m anifestation has been w hat is described as a hypertrophy o f th e tertiary sector of em ploym ent. It is present to a greater or lesser ex ten t in all countries, b u t th a t it is present in the Third World as a whole is m ost evident. Thus, although in 1960 Third World econom y, measured by the percentage of the labour force em ployed in agriculture, was at the same level as th a t of the in dustrialized countries o f western Europe and N orth America around 1810, th e Third World possessed the same proportion of persons em ployed in trade and banking as had been reached by the industria lized countries around 1890. In transport and services the corresponding period is th a t o f 184050. On the other hand the proportion o f the labour force engaged in the m anufacturing and extractive industries is lower than it was in the advanced countries in about th e year 1800. But the percentage em ployed in the construc tion industries corresponds fairly closely to the level o f developm ent reached. The high level of tertiary em ploym ent can be explained by the excessive developm ent o f commercial activity and public services in m ost of the countries concerned and by the impossibility of absorbing in the secondary sector all the surplus labour from agriculture. The small proportion in m anufacturing already noted results from a com bination o f tw o factors. Firstly, the less-developed countries are large net im porters of m anufactured goods*, secondly, these countries possess factories em ploying m id-tw entieth century techniques in which productivity is high while at the same time having a low per capita consum ption, equivalent to early or m id-nineteenth century levels o f consum ption in the West. Where the extractive industries are concerned the low level is also explained by the advanced technology o f mining enterprises m ostly due to the use of m odern equipm ent developed in the course o f the last war. It is obvious th a t the situation in the various countries and in the larger regions differs m arkedly, the best example o f the difference is provided by the comparative figures o f Latin America and the remainder o f the Third World.4 A round 1970 about 40 per cent of the labour force in Latin America was engaged in agriculture as against 70 per cent in Asia and 75 per cent in Africa. Tertiary hypertrophy was very advanced in Latin America: in 1970 m ore than 35 per cent o f the labour force was em ployed in this sector the
162
same proportion as for Europe (excluding the U.S.S.R.) in about 1955. For figures relating to individual countries, the synoptic table on page 244 should be consulted. (b) Changes between 1900 and 1970 A few words first about the changes which took place prior to 1900. The alm ost total absence of censuses in this period (with the exception o f India and certain Latin American countries) permits only very crude extrapolations to be made, based on 1900 figures and the know n facts o f economic development. Thus we can postulate th at in the m id-nineteenth century and the preceding decades the relative im portance of the prim ary sector should have been about the same as it was in 1900, whereas the secondary sector, not yet much affected by the im port of industrial products, should have occupied a slightly higher percentage o f the labour force to the detrim ent of the tertiary sector. Turning to th e outstanding points o f developm ent between 1900 and 1970, we find a slight fall in the relative position o f agriculture in spite of the fact th at population growth led to a great increase in the absolute num bers o f the agricultural labour force. Between 1900 and 1970 the labour force in agriculture more than doubled. This fact has the m ost weighty econom ic consequences, since with so little agricultural land available it m eant th at the area o f cultivated land per agricultural worker fell by almost half. This largely explains the problems encountered in the Third World in attem pting to increase agricultural productivity (see C hapter 2). We should note, however, th at the reduction of the relative level o f the agricultural labour force accelerated after 1950, even though because o f population inflation the increase in absolute num bers o f rural labourers was even higher. There was a greater rise in these tw enty years than during the preceding fifty years. Where the secondary sector is concerned we can discern a different trend in the extractive and m anufacturing industries. In extractive industry em ploym ent grew in both absolute and relative terms. The change can be explained by the growing part the Third World played in world production in this sector (see C hapter 3). It will not, however, escape notice th at even in 1960 this sector o f em ploym ent remained relatively unim portant: 0.6 per cent o f the total labour force. This explains in large p art the few linkage effects which the increased exports o f mineral products, which form ed more than 40 per cent of total Third World exports, had on th e total econom y. It
163
should also be noted th a t a large proportion o f the labour employed in the extractive industry was engaged in activities connected with the dom estic m arket sand and gravel, building stone, etc. This proportion was certainly more than 50 per cent o f all em ploym ent in the extractive industries. The border-line case of a sector w ith a low em ploym ent rate, b u t a high value in world exports is oil. According to my estim ates, about 0.02 per cent o f the to tal labour force o f the T hird World is thus em ployed, b u t it produces m ore than 33 per cent of the total value o f exports around 1970. The relative im portance o f m anufactures shrank betw een 1900 and 1950. W ithout taking into account the possible distortions arising from the different criteria of censuses,5 the fall largely reflects th e disappearance o f the large num ber of craft workshops condem ned by im ports o f m anufactures from the developed coun tries whose productivity was very m uch higher. Such growth in em ploym ent as there has been since 1950 reflects the great increase in the dom estic o u tp u t o f the Third World in this sector (see Chapter 4). But the low relative share o f m anufacturing industry (probably slightly above 10 per cent in 1970) underlines the difficulties encountered by this sector in the past, which will continue to be encountered in the future in absorbing a substantial portion of th e surplus rural labour force. The difficulty o f absorption, moreover, helps to explain the rapid grow th in the labour force in the tertiary sector. Between 1900 and 1970 absolute num bers have alm ost quadrupled. This brings us to a consideration of the problem s o f th e surplus population o f the countryside. (c) The problem o f the absorption o f the surplus rural labour force In considering the changes th at have taken place betw een 1950 and 1960, and on the basis of demographic trends we can estim ate th at the rural labour force between these tw o dates has grown by about 81 million persons. But during the same period m anufacturing industry has provided no more than 14 million extra jobs, while the labour force in this sector has itself, through natural increase, grown by some 7 million persons. Even by assuming, quite arbitrarily, th a t a quarter o f the natural increase in the industrial labour force has moved on into th e tertiary sector, the conclusion emerges th a t the growth o f em ploym ent in m anufacturing has not provided jobs for m ore than 10 per cent o f the surplus rural labour force. Moreover, even this level, m odest as it is, is high when com pared to the change which occurred in th e same field betw een 1920 and 1950. During those th irty years the growth o f em ploym ent in m anufacturing
164
industry had been on such a small scale th a t it had n o t even been capable of absorbing th e natural increase in th e labour force already at w ork in the sector. In fact the grow th in em ploym ent had corresponded to only about one-half of the natural increase o f the working population. This was bound to contribute largely to the hypertrophy of the tertiary sector o f em ploym ent, though a certain am ount of labour was absorbed by m ore rapid grow th in em ploy m ent in mining and particularly in construction. C om putations of the same kind made for th e period 196070 (although the figures are m ore provisional) allow us to conclude th at, during this decade, the new jobs in m anufacturing industry had absorbed about 10 to 12 per cent of the surplus labour force o f th e countryside. I have made similar com putations for the European countries during the first tw enty years for which valid censuses o f the labour force are available (usually 1840/50 to 1860/70). These reveal th a t the rate of absorption was from 40 to 100 per cent, w ith the average lying around 50 per cent. For the w hole o f early industrialized E urope6 betw een 1880 and 1900 this rate was about 80 per cent. But in 1880 no m ore than 46 per cent o f the labour force in this p art o f the world was em ployed in agriculture. On the basis o f the above figures we can estim ate th a t in the advanced countries the rate of absorption since th e earliest phases o f industrialization m ust have been ab o u t 30 40 per cent. The difference betw een rates such as these and those being achieved at the present tim e by the Third World is not, however, th e result o f faster expansion in m anu facturing industry in th e developed countries (on th e contrary the grow th o f industrial production in under-developed countries is, as we saw in C hapter 4, higher at present th an it was in th e developed countries during the nineteenth century), b u t is due essentially to differences in population grow th. Moreover, a theoretical calculation for the Third World betw een 1950 and 1960 based on th e hypothesis o f a population increase o f 0.6 per cent per annum, in economic conditions like 1950 and postulating increases in o u tp u t and productivity of th e same order as occurred betw een 1949/51 and 1959/61, would result in a hypothetical rate o f absorption o f 70 per cent. By reducing th e rate o f growth o f industrial o u tp u t by 1.7 (i.e. the gap betw een th e real growth o f population: 2.3 per cent and the one assumed here 0.6 per cent) we still come to an absorption rate of m ore than 30 per cent. The gradual im provem ent in the rate w ith which th e surplus rural population o f the Third World was being absorbed betw een 1950 and 1970 com pared to th e years 192050 should not, however, let us
The Labour Force Sc Employment
165
assume th a t this trend will necessarily continue in the future. On the contrary it seems to me that, in the m edium term (1015 years) we m ay find th at the rate, already a limited one, will decline. Three factors will com bine to cause this trend. First o f all there is the certainty th at, whatever success is achieved by the campaigns for b irth control, there will still be an increase in the labour force due to the existing age structure o f the poplation (see section I o f this chapter). Then there is the strong probability th at the rate of increase in industrial production, largely due to im port substitution will slow down: indeed the limits to w hat is possible are already close to being achieved. Finally, there is the probability th a t productivity in m anufacturing industry will rise faster than it has done in the past.7 These changes explain why, since the beginning o f the 1960s the problem s o f unem ploym ent, b u t m ore particularly o f urban unem ploym ent, have reached such overwhelming proportions. These problem s themselves will now be examined. B. U nem ploym ent and under-em ploym ent Before considering th e various points in this connexion we must remind the reader th at concepts o f unem ploym ent and under em ploym ent as they have been form ulated in the West cannot be applied to such very different societies except in a very crude and approxim ate way. In m any parts o f the Third World the extended bonds o f kinship and the im portance o f alternative activities greatly am eliorate the social and economic plight o f an unem ployed person w ithout taking account o f the incom e a large part o f the urban population receives from activities outside the law. On the other hand o f course, since there is generally no unem ploym ent benefit the economic plight o f urban unem ployed labour is worse. A further, and im portant, reservation is th at in m any, one might alm ost say all, traditional societies in the Third World the lack o f a regular job is not necessarily felt to be a disadvantage. As Jacques Berque has rightly pointed o u t it is the intrusion of Western values th at has led to a situation in which the traditional leisureliness of a peasantry becomes the joblessness o f underdevelopm ent.8 It does n o t do, of course, to idealize the life style of the traditional society, where norm al condi tions for the m ajority were punctuated by periodic harvest failures which were as often as n o t followed by famine and death, and only a m inority was always able to live in plenty. But it is also true th at in m any cases the process o f colonialism caused a fall in the standard of life o f a large fraction o f the population, who were either dispos sessed o f their land or forced into cultivating plantation crops.
166
colonialism caused a fall in the standard o f life of a large fraction of the population, who were either dispossessed o f their land or forced into cultivating plantation crops. Moreover, the notion o f unem ploym ent as we understand it today is relatively new even in the West. The very use o f the word unem ploym ent in English is comparatively recent, appearing in a te x t for the first tim e in 1888.9 In French the w ord for an unem ployed person ( chomeur) in the sense we use it today was first recorded slightly earlier in 1876.10 The verb chomer actually comes from the low Latin caumare deriving from the Greek kauma =burning heat. Thus chomer originally m eant to rest during the heat o f the day'. As for the concept o f under-em ploym ent, the criteria by which it is determ ined are totally subjective and strongly conditioned by historical situations peculiar to m odern western society. While in m ost countries it is true th a t a certain am ount of under-em ploym ent exists, it should be rem em bered th at the criteria generally used to determ ine under-em ploym ent, and more especially the average norm o f w ork which a labourer can perform , are those appropriate to a particular society at a particular stage o f development. Indeed, leaving aside the changed attitudes introduced by the R eform ation, it should be rem em bered th at the industrial revolution was responsible, amongst other things, for a substantial increase in the num ber of hours worked during the year in both tow n and co u n try .1 1 In the countryside the introduction o f m odern m ethods o f agriculture brought about an intensification of labour, th e disappearance o f the fallow, the diffusion o f new crops and the integration o f livestock rearing and agriculture; all these factors united to increase consider ably not only productivity b u t also the average load o f labour. The factory established in the towns had replaced the craftsm ans workshop and had thereby enorm ously added to the num ber of hours worked, not to m ention also lowering o f the age at which the individual com m enced work and increasing the pace at which he worked. During the nineteenth century, particularly the first half of the century, a working day of 1416 hours became custom ary and, Sundays apart, the num ber of days taken o ff for holidays was reduced to ten or even fewer. This added up to about 300 working days per annum, or p u t in another way, about 4,600 hours per annum com pared to less than 2,800 which had been usual in the same societies a century or tw o earlier. Gradually, through a reduction in the num ber of hours worked per day and by introducing th e annual vacation, which had replaced the
167
many saints and holy days o f pre-industrial Europe, the advanced countries returned to an annual num ber of working hours approach ing or even below that which had been customary before the industrial revolution. But, in assessing the level of under-em ploym ent which exists in the under-developed societies today, there is a tendency to regard the reduced num ber o f hours now worked in the advanced countries as a situation w ithout precedent, arising solely from higher productivity made possible by m odern technology. A nother consideration which is generally overlooked is the limitations imposed by climate. W ithout accepting a narrow geo graphical determinism, there is no doubt th at the tropical or semi-tropical climates of the larger part o f the Third World make prolonged sessions of work less acceptable than they are in the tem perate regions where m ost o f the advanced countries are situated. This applies especially to certain seasons o f the year. There are in addition m any other problem s th at arise from the lack o f statistics or any other way of evaluating the extent of un- and under-em ploym ent, as well as from the fact th a t such estimates as do exist cannot be used to make comparisons. This last point, and more particularly the existence of differences between urban and rural conditions, makes it essential to examine the two sectors separately, as has been done by Turnham in his recent excellent stu d y .12
I. R U R A L U N E M P L O Y M E N T A N D U N D E R - E M P L O Y M E N T
Unem ploym ent in the countryside is relatively low in the Third World, as it is in m ost societies. The lack o f rural em ploym ent is usually reflected to a greater or lesser degree by a state o f under-em ploym ent, rather than by actual overt unem ploym ent in the m odern sense of the term. If we leave out of account some isolated cases in which, often because of the way they are measured, unem ploym ent rates are between 8 and 12 per cent, estimates for the m ajority of the less-developed countries m ostly p u t rural unem ploy m ent betw een 3 and 5 per cent of the labour force,13 with the average for the Third World as a whole probably lying between 3.5 and 4.5 per cent. It is pointless to make a comparison with the developed countries since figures for the eighteenth and nineteenth centuries are much too unreliable to enable us to postulate any rates which are even relatively accurate. The m ost one can say is th a t the rates should have been about the same as, o r slightly below, those of the Third World at the present time. Under-em ploym ent in the rural areas is clearly m uch more extensive. But in trying to find o u t just how extensive it is we come
168
up against some difficulties which have not yet been satisfactorily resolved. W ithout returning to the basic issue o f concepts we can say th at there are three m ain ways in which the problem could be approached. The first is by trying to calculate the num ber o f hours worked. The second approach is by determining the rural labour surplus. In this approach labour would be treated as a factor of production and an estim ate of the normal labour requirem ents for a given agricultural o u tp u t w ould be made. This estim ate would then be set against the available labour force. There is a third approach which is, however, hardly ever used; to determ ine w hat is a norm al agricultural income and to compare it w ith actual income received. We will not review the meagre results obtained from these three m ethods o f investigation. Here I shall merely try to fix a possible order o f m agnitude for existing rural under-em ploym ent. On the basis o f a rather conservative estim ate a figure o f 20 to 30 per cent may be p u t forw ard for th e Third World as a w hole,14 although some writers, notably A rdant, p u t the figure as high as 50 per ce n t.15 Here again, even m ore than for unem ploym ent, any comparison w ith conditions in advanced countries at the time of their take o ff is quite out o f the question, and it should also be rem em bered th at this chapter is restricted to the non-comm unist world.
II. U R B A N U N E M P L O Y M E N T A N D U N D E R - E M P L O Y M E N T 16
The great speed w ith which the process o f urbanization has proceeded in the Third World was described in the preceding chapter. The m ost im portant result has been the very great deterioration of the possibilities for em ploym ent in m ost urban centres.17 The num ber o f additional workers who came onto the labour m arket as a consequence o f migrations to towns between 1950 and 1970 can be estim ated as some 60 to 70 millions. This figure is the same as, or slightly higher than, the total num ber o f urban jobs known to exist in the year 1950. The doubling o f the urban working population occurring betw een 1950 and 1970 has created an em ploym ent situation so unfavourable th a t it was bound to be reflected by the under-em ploym ent in the tertiary and other sectors of the labour force, and by an unem ploym ent rate which climbed in m ost countries to an unacceptably high level. I have been able to assemble some figures for the 34 less-developed countries for different dates m ostly betw een 1964 and 1968 although they are unfortunately both incom plete and not strictly comparable. Urban unem ploym ent in four o f these countries
169
was 20 per cent and over; in eight countries it was 15 to 20 per cent; in eight more it was 10 to 15 per cent; and in the remaining fourteen it was alm ost invariably above 5 per cent. These rates would probably be enhanced if the statistics were m ore uniform. Thus in nine cases the urban rates are those for the capital cities, whereas it seems to be a general rule th at unem ploym ent rates in capital cities are lower than they are in other urban centres.18 Due to the scarcity o f available statistics, it is difficult to determ ine the exact changes in the urban unem ploym ent rates in recent years. What is almost certain is th a t betw een 1950 and 1970 the rates definitely increased, since the situation at the beginning o f the period was, generally speaking, m uch better. The period when conditions deteriorated m ost was probably betw een 1950 and 1960/65. It should be noted, neverthe less, th at beyond a certain point a deterioration of the labour m arket will be reflected m ore by an increase in under-em ploym ent than by rising unem ploym ent rates. It is possible to try to fix a rough order of magnitude for the absolute num ber of urban unem ployed in the under-developed countries. Assuming an average urban unem ploym ent rate o f 12 per cent for 1970 and o f 11 and 10 per cent respectively for 1960 and 1950 and an activity rate in urban areas o f about 50 per cent, we find the following changes in the num ber o f urban unem ployed: 1950 1960 1970 68 millions 1113 millions 2024 m illions19
It is m ore than likely th at the real changes were more dram atic than these figures suggest. The actual range could lie between the lowest figure for 1950 and th e highest figure for 1970. It should also be pointed o u t th at, taking a som ewhat wider definition of an urban area i.e. one based on the various national definitions, the unem ploy m ent figures quoted above would be even greater: from 20 to 32 per cent (according to w hether we postulate the unem ploym ent rates of the smaller towns representing either 70 or 100 per cent o f the unem ploym ent rates prevailing in the larger towns). If we include these urban areas the estim ated num ber of urban unem ployed in 1970 goes up to between 25 and 32 millions. (It should be recalled again th at the chapter is limited to the non-com m unist Third World.) The situation is aggravated by the fact that unem ploym ent bears particularly heavily on young people. Indeed, on the basis of figures which are very significant although adm ittedly not com plete, it
170
appears th a t unem ploym ent in the 15 to 24 age group is twice the average, which is itself, of course, affected by the massive inclusion of the youthful unem ployed. O ut o f seventeen countries for which T urnham 20 has assembled such figures, three have unem ployed rates amongst young people near to or above 40 per cent; five have rates o f 20 to 29 per cent; another five have rates from 10 to 19 per cent; and only tw o countries have rates less than 8 per cent. A nother serious feature is that the urban unem ployed person in the Third World is nearly always one who has had an average education, i.e. a young person who has com pleted six to eleven years o f schooling. The highest unem ploym ent rates are to be found in this catagory. N ext comes the group who has had only a prim ary education, or one to five years o f schooling; then comes the illiterate group and only lastly the group whose education has continued beyond the secondary level. Before passing on to discuss urban under-em ploym ent it should be emphasized that the m agnitude o f the problem is unique to the Third World countries. Leaving aside the m odern period when urban em ploym ent in industrialized countries is, o f course, very low, it should be noted th at even at the end of the nineteenth century the earliest period for which we possess relevant figures the situation of em ploym ent in western towns was always much better. In round figures urban rates o f unem ploym ent in Great Britain were about 6 7 per cent; in Germ any 4 5 per cent; in France 5 6 per cent; and in the United States probably 3 4 per cent. W ithout in any way attem pting to idealize the deplorable conditions which prevailed amongst the labouring classes of the towns during the take o ff period in western societies, it appears likely, in the light o f the figures presented above and corroborated by other inform ation of a non-quantitative kind, th at the average degree o f unem ploym ent was much lower than th at which now prevails in the cities o f the Third World. However, unem ploym ent rates probably fluctuated more in the course o f the nineteenth century than they do in the Third World at the present time. One might almost say th at urban unem ploym ent during the take o f f o f the western world was cyclical rather than structural. In considering the problem of urban under-em ploym ent we m ust first o f all recall how arbitrary it is to apply such a concept to under-developed societies. But, even ignoring this, it is clear th a t the definition and criteria which have to be accepted before we can estim ate levels o f under-em ploym ent are almost as varied as the estimates themselves. This is why all the facts have to be treated with
The Labour Force Sc Employment
171
such extrem e reserve and w hy we have to be particularly on our guard when it comes to drawing international comparisons from the very few figures we are going to provide. The available estim ates reveal very high rates of urban under developm ent in Latin America, ranging from 20 per cent for Panama to 28 per cent for Chile, with one exception, Argentina, where the rate is probably only 17 per cent.21 Estimates o f urban under em ploym ent for other areas are even fewer, b u t in general the rates are lower. On th e global level there is one good general indicator o f the degree o f urban under-em ploym ent.2 2 This is the hypertrophy of the tertiary sector which I have already m entioned, and which is such a feature o f the vast m ajority of under-developed countries. It is rather unwise to attem p t to attach any order of m agnitude to urban under-em ploym ent in the Third World as a whole, but in very broad term s one m ight say th at the possible range lies between 10 and 25 per cent.
* * *
Thus the labour situation is very grim and the outlook for the future is even grimmer. We may assume th at in 1970 genuine total unem ploym ent, i.e. unem ploym ent in the narrow sense of the word, together w ith under-em ploym ent, expressed as involuntary in activity, probably affected some 20 to 30 per cent o f the labour force, i.e. betw een 130 and 200 million persons in the non com m unist Third World. But betw een 1970 and 1985 the labour force is likely to increase by another 270 290 million souls. Therefore, if we wish to reduce real unem ploym ent to a tolerable level (say 10 per cent) no less than 350 400 million new jobs will have to be created in th e Third World within the next fifteen years. This means a num ber similar to th at which was created during the previous 80 to 120 years. The staggering im m ensity of the problem cries o u t for some original solutions and in this respect, at least, it is well to note that, mainly due to the efforts o f th e International Labour Office,2 3 there are some signs of an awakening o f the w orlds conscience to the terrible implications for the econom ic future o f the under-developed world.
io
INTRODUCTION
Macro-Economic Data
I had originally itended to devote m ore space to this chapter than I have finally done, b u t after scrutinizing the statistics in greater detail I have decided to shorten my study. Indeed it has seemed to me th a t because the estim ates for some countries are uncertain and for m any others entirely lacking, any lengthy treatm ent o f the subject m ust be largely arbitrary. The chapter will consist o f three parts: A. Capital form ation and im ports o f capital equipm ent B. Financial aid C. Growth and level of national p roduct or national income A. Capital form ation and im ports o f capital equipm ent The tw o following topics will be briefly considered under the headings: changes in the rate o f capital form ation; and capital equipm ent as a share o f imports.
I. C H A N G E S IN T H E R A T E O F C A P I T A L F O R M A T I O N
Rates of capital form ation should, theoretically speaking, provide us w ith an indicator of th e effort m ade by various countries to achieve econom ic growth, b u t unfortunately several factors com bine to reduce the value o f this indicator. First o f all there is a considerable margin o f error, especially where agricultural investm ent is con cerned. Knowing the difficulties o f trying to compile measurements o f this kind for the developed countries, one has to be even more cautious when it comes to the under-developed countries. F urther more, although the capital form ation rate gives us some indication of the relative size o f the collective effort, the extent to which the effort has been successful m ust also be m easured and here the gaps are even m ore obvious. Again, speaking theoretically, the capitalo u tp u t ratio should give us a yardstick o f this success, b u t it turns out to be difficult to detect to w hat degree the differences betw een the countries are due directly to the different margins o f error. We calculated some years ago the capital-output ratio fo r twenty-seven under-developed countries during the period 1954 60. The range lay between 2.6 and 10.0 w ith an average o f 4.4 for nineteen o f the countries where the margin of error was least. For these reasons great 172
A i X a
Macro-Economic Data
373
caution m ust be exercised when considering the data relating to capital form ation rates for the three different periods beginning with 1950, which have been com puted in Table 48. (Data prior to 1950 is lacking for almost every country.) Making due allowance for these warnings, the following points can be deduced from the table. So far as changes are concerned we can see a divergence between th e Latin-American countries and the rest o f the Third World. The capital form ation rate o f the Latin-American countries rises from 1950 to 1960 and tends to fall from 1960 to 1970. The changes for the other under-developed countries are slightly m ore favourable, for in them the num ber of cases o f a rise have exceeded the num ber of cases of a fall. But, again generally speaking, the change betw een 1950 and 1960 appears to have been m ore favourable than the one betw een 1960 and 1970. Because o f the reservations made above, I shall not attem p t to draw any conclusions as to the actual rates o f capital form ation. In order to put the problem of the m odern Third World in its historical perspective I shall simply note th a t the advanced countries were able to take o f f with much lower capital form ation rates than the under-developed developed countries o f today, since their rates ranged from 5 to 10 per cent. Since population growth was very low these investm ent levels were sufficient to generate economic growth. On the other hand, the capital form ation rates in the developed countries are, at the present m om ent, generally higher than in the under-developed countries, as w ith a few exceptions they exceed 20 per cent. The higher rate o f capital form ation com bined with the lower population growth help to explain the widening gap between per capita incomes in the developed countries and in the Third World countries. This gap will be discussed in section C (IV) below. Before passing on to the next section we shall deal very briefly with capital form ation in China. On th e basis o f Ecksteins com putations made to fit Chinese statistics into the fram ework of western accounting, it appears th at fixed capital form ation formed 10.3 per cent o f the Gross Domestic Product in 1952 and about 13 per cent in 1957.1 In the period o f the G reat Leap Forward rates were higher, as is evident from Hollisters estim ates.2 These show gross fixed capital as a percentage o f Gross National Product to be 12.4 per cent for 1952 (5.5 per cent for 1950); 15.9 per cent for 1957; 23.9 per cent for 1958; and 25.7 per cent for 1959. Pre-1958 rates are similar to those of other under-developed countries. Thus it w ould seem, assuming the faster growth achieved (see below), th at
174
Table 48 Changes in rates o f capital form ation 19501968/70 (rates o f gross fixed capital form ation as percentages o f GDP in current prices)
1950 A FR IC A Ghana Kenya Morocco Nigeria Sudan Tunisia Zaire Zambia AM ERICA Argentina Brazil Chile Colombia Ecuador Guatemala Mexico Peru Venezuela A SIA Ceylon India Pakistan Indonesia5 Philippines Taiwan Thailand M IDDLE E A S T Egypt Iran Syria Turkey
11951 21952
1 959 -1 9 6 1
1 9 6 8 -1 9 7 0
I9 2 6

21
20 17 10 12 12 183 15 21
11 19 14 154 124 22 26 25
20 15 9 14 9 10 12 15 26
21 17 16 18 14 10 15 18 20
19 16 15 19 15 14 19 12 24
9 8

11 n 1 122
15 14 12 10 13 16 14 16 16 15 15
41967/69
18 154 134 11 19 22 24 12 20 15 19
10
31960/61
5 I n c lu d in g in c re a s e in s to c k s Sources: D e riv e d fr o m th e fo llo w in g U .N . p u b lic a tio n s (v a rio u s y e a rs):
E co n o m ic S u rvey o f L atin A m erica, E tu d e su r la situation eco n o m iq u e de V A sie e t de 1E xtrem e-O rient, a n d M o n th ly B ulletin o f S ta tistics ; a lso N ational A c c o u n ts o f Less-developed Countries, 1 9 5 0 1966 a n d Latest In fo rm a tio n on N ational A c c o u n ts o f Less-developed C ountries
( D e v e lo p m e n t C e n tr e o f O .E .C .D ., P a ris, v a rio u s y e a rs).
A i X a
Macro-Economic Data
175
the efficiency o f investment was greater in China. Eckstein himself comes to this conclusion, especially when comparing Chinese and Indian figures. But the data is too uncertain for us to be able to extract any valid conclusions in this connexion.
II. C A P I T A L E Q U I P M E N T AS A S H A R E O F I M P O R T S
A good measure o f the level of developm ent a country has achieved is its degree o f dependence on the outside world for its capital equipm ent, although, as is always the case with individual indicators, it does not provide by itself a wholly balanced picture. The results of calculations made to estim ate the share o f im ports in the domestic consum ption o f capital equipm ent can be seen in Table 49. To express it in another way, I have p u t the total represented by capital equipm ent (including transport equipm ent, b u t not private cars) as it emerges from estimates of the structure o f fixed capital equipm ent, against the to tal represented by im ports o f the same products. To reduce the effects o f annual fluctuations and the delays betw een the date when equipm ent was im ported and when it was p u t into service, the percentage has been calculated on a three-year average. The omission from our calculations o f exports o f capital equipm ent can be justified by the negligible value o f this item in the foreign trade of the Third World countries. Thus in 1970 for all under-developed countries together, im ports of machinery am ounted to a total value o f more than 19 billion dollars, while exports form ed no more than one billion (of which more than 500 million dollars were destined for other under-developed countries). The num ber o f countries included in Table 49 is lim ited because com plete data for m ost under-developed countries is lacking. In interpreting the percentages quoted in the table we have to allow for the fact th at the unit value o f capital equipm ent as seen in estimates o f capital form ation is higher than appears in im port statistics, since it includes various supplem entary costs (profit m argins, transport, handling and erecting costs, etc.). The difference means th at im port rates should really be increased by a certain percentage, b u t to determ ine exactly w hat this should be is n o t easy, although one can safely say it would exceed 10 per cent. The figures in Table 49 show the overwhelming part im ported equipm ent plays in the capital form ation o f the Third World, especially if we exclude certain countries such as Burma, Tunisia, the Philippines and Uganda whose low rates should be ascribed to a statistical bias rather than to a m ore advanced level of development.
176
The Economic Development o f the Third World since 1900 Table 49 Share o f imports in the gross capital formation o f machinery and transport equipment
C O U N TR Y Argentina Burma Ceylon Chile Egypt El Salvador Ghana Honduras India Jamaica Korea S. Malaysia Mexico Morocco Nigeria Panama Philippines Sudan Syria Tunisia Turkey Uganda
! 1955~57 1 9 6 2 -6 4 2 1 9 6 0 -6 1 7 19 6 7 69
1 9 5 4 -1 9 5 6
1 9 6 1 -1 9 6 3 583 47 66
1 9 6 8 -1 9 7 0 21
57 122

1031 85

84
69

61 64 96 86 35 62 6 56 110s 55 822 94 66 53 1044
647 40 427 707 101 1057 26 7 39 82
50 64 67
3 i 960 - 6 2 8 1 9 6 6 -6 8 *1962 $ 1 9 6 2 - 6 3 9 1 9 6 6 -6 7
447 88 718 65 92 617 68 689 39
Sources: A u t h o r s e s tim a te s (s e e t e x t f o r m e th o d o f c a lc u la tio n ) .
Statistical bias is also responsible for any rates th a t are higher than 100 per cent. If allowance is made for this factor and for the divergences in unit costs, it appears that about 90 per cent o f capital equipm ent used in th e Third World is at present im ported. The limited num ber o f dates in the series does not perm it a meaningful trend to emerge. Likewise the lack o f statistical uniform ity makes any conclusion based on the differences between one country and another too uncertain to be useful. The average rate quoted above (90 per cent) provides the only comparison w ith the developed countries o f any real value. As far as the m odern period is concerned, we can see th at nearly all essential capital equipm ent in the developed countries is o f local origin. For example, in France im ports form only 2 per cent of equipm ent in use and exports 2.4 per cent. Even if we look at the figures for some developed countries which are relatively less industrialized we can see th a t the deficit in the foreign exports o f capital equipm ent when compared to domestic
Macro-Economic Data
177
consum ption of these goods is still a small one. In Italy im ports form 26.7 per cent and exports 25.1 per cent; while in Denmark im ports represent 55.0 per cent and exports 44.9 per cent (both these examples are for the same period, i.e. 1961 63). As far as the take o ff period o f the developed countries is concerned I have made an estim ate for France3 which suggests that between 1825 and 1875 im ports form ed less than 2 per cent o f total consum ption of capital equipm ent. English export statistics (England was the principal source o f exports o f capital equipm ent at that period) suggest th at the French situation was typical o f all other European countries. Even in Japan the percentage between 1914 and 1936 was no more than 15 per cent. The gap thus revealed between the situation in the Third World countries and the developed countries makes it possible first to delineate the different paths o f development taken by these two groups o f economies, and secondly to lay bare the implications for econom ic growth inherent in foreign dependence. The m anufacture of equipm ent played a major part in the take o f f o f the western countries in diffusing among them the effects o f industrialization. I shall not elaborate here on why there was such differences; I shall merely point out that the causes can be sought in the developm ent of technology and the reduction o f transport costs.4 B. Financial aid In this section we shall concentrate mainly on changes th a t have taken place, although there will be a brief account o f how aid was and is distributed geographically.
I. G E N E R A L C H A N G E S
The more or less arbitrary way in which financial aid is defined obviously makes it hard to determ ine its precise volume. W ithout going into technicalities, however, it can be said that, apart from grants (in cash and in kind) narrowly defined, financial aid consists for the m ost part o f a net flow of loans of more than a years duration (approxim ately 30 per cent of total aid); investment and loans from the private sector (approxim ately 27 per cent); and guaranteed export credits (approxim ately 12 per cent). Thus, exclusion o f the three last categories, reduces the am ount of the aid given to the under-developed countries by about 70 per cent, leaving only about 30 per cent to be represented by grants and grant-like contributions. It should also be pointed o u t th a t whereas repaym ents
178
o f capital made by recipient countries are deducted from the am ounts o f net aid this is not true o f the interest charges on loans. Thus we can see th at the concept o f aid includes a collection of elem ents which it is very difficult to consider objectively as real aid, at least in their total am ount. The determ ination o f the size o f real* aid as opposed to w hat should m ore properly be called flow of financial resources to less-developed countries, is an extrem ely delicate and to some extent arbitrary m atter. Tibor Mendes recently published stu d y 5 explores the intricacies o f the question in a very suggestive manner. According to him, the rum p th at can genuinely be called aid is no more than about 2225 per cent o f the flow of financial resources to less-developed countries. This conclusion seems to me quite realistic and the figure should therefore be borne in mind throughout the discussion th at follows. The estimates o f the size o f the flow o f financial resources to the under-developed countries come from tw o principal agencies. The first is the United Nations and the International M onetary Fund and the second is the O.E.C.D. Material for both is based on question naires addressed to donor countries and on estim ates com plementing these. Slightly different m ethods are used to construct the estimates th e m se lv e s and th is lead s to d iv erg en ces b e tw e e n th e tw o se ts of figures, the O.E.C.D. total being about one billion dollars higher than the U.N. one. It seems likely th at if the definition o f foreign aid adopted by both institutions is adhered to, the O.E.C.D. estimate appears to be the more accurate o f the tw o because it does n o t leave o u t o f account transactions w ithin certain m onetary unions. It is this estimate, therefore, th a t I have mainly used in constructing Table 50. From this table it is possible to trace the development of foreign aid since its beginnings in the fifties. Owing to the fact th a t a large p art of the Third World prior to 1950 was still subject to a colonial regime, the notion o f aid is not applicable to that period, especially as the data relating to financial transfers in th at period is not available. Before going on to analyse the figures in the table it should be pointed out th a t they include foreign aid given to some European countries which O.E.C.D. considers to be under-developed (Greece, Spain, Turkey and Yugoslavia) and which are not included in the definition usually accepted here. The sum received by these countries has on average been 6 7 per cent of the to tal given in the table. The annual total o f the flow of financial resources rose from approxim ately 4.5 billion dollars at the beginning o f the fifties to about 16.6 billion dollars for 1969/71, i.e. an average increase o f
Macro-Economic Data Table 50 Net flo w o f financial resources to non-communist lessdeveloped countries 1 (yearly averages; billions of U.S. dollars and percentages)
179
Breakdown o f total per origin in percentage Western countries From public sector 3 1 9 5 0 -5 5 1 9 5 6 -5 9 1 9 6 0 -6 2 1 9 6 3 -6 5 1 9 6 6 -6 8 1 9 6 9 -7 1 2 1.9 3.8 5.7 6.8 7.6 9.3 From private sector 2.6 3.0 2.7 3.2 4.3 7.3 TO TAL 4.5 6.8 8.4 10.1 11.9 16.6 Direct 97 93 93 89 89 91 Multilateral agencies 2 5 4 8 8 7 Eastern countries 1 2 4 3 3 1.7
1 Including E uropean less-developed co u n tries (see tex t). 2 E xcluding grants m ade by private voluntary agencies, w hich a m o u n t yearly to som e 0.8 billion for this perio d ; no com parable d a ta available for previous periods. 3 Including m ultilateral agencies and com m unist countries. Sources: Derived from various issues o f D evelo p m en t Assistance o f th e O.E.C.D . and Flow o f Financial Resources and Liberalization o f Term s a n d C onditions o f Assistance (UNCTAD, Geneva, 1971, roneo-ed).
about 6.5 per cent per annum. This rise comes mainly from the public sector. Although the latters starting figure was very low, it increased from less than 2 to over 9 billion dollars during the same period, i.e. by 8.5 per cent per annum. Furtherm ore it should be noted th at the main part o f the rise occurred during the decade 1950 60. Since then the growth o f aid from the public sector has considerably slowed down and the main part o f the grow th in total aid has come from the private sector, a fact to which we shall return later. Public aid should n o t be lumped together with grants, since the latter provided only a fraction o f aid, a fraction, moreover, which has shown a m arked tendency to fall. Thus, for public aid from countries o f the D.A.C.6 'bilateral grants and contributions assimi lated to grants fell from 77 per cent in 1960/62 to 48 per cent in 1969/71. And it is once more necessary to insist on the relative nature of some grants (see in this connexion T. Mendes book referred to above). The figures quoted above and in Table 50 are in current dollars. In constant dollars actual perform ance appears more limited. The problem here consists in the choice o f deflator. O.E.C.D. calculated its deflator for the period 1960 707 on the basis o f the structure of aid and corresponding price indices which resulted in a price increase
180
o f only 19 per cent (or 1.7 per cent per annum ). This is a much lower rate than the rise o f domestic prices within the developed countries (about 3.5 per cent per annum ), b u t it is higher than the rise o f im port prices into the developing countries which increased by 1 per cent per annum in the same period. However the validity of price indices for foreign trade is questionable. If we apply the deflator used to calculate the Gross Domestic Product o f the developed countries, the volume o f total aid appears to increase between 1959/61 and 1966/70 at an average annual rate o f 2.3 per cent which is m uch lower than the rate o f growth of the volume of Gross Domestic Product in the developed countries. Moreover the relative size of the aid as a percentage o f the Gross National Product (at m arket prices) of the developed countries who are also m em bers of the D.A.C. was only 0.76 per cent for the years 1969/71 as against 0.88 per cent for the years 1960/62. This means th at the objective (one per cent of Gross Domestic Product) adopted by the second U.N. Conference on Trade and Development (U.N.C.T.A.D.) in New Delhi in 1968 has not been achieved, and, as will be pointed out in the next chapter this objective in itself was a very m odest one. The p e r ce n ta g e o b v io u sly varies fro m c o u n tr y to c o u n tr y ; th e highest percentage contributed by any great power is from France where it is alm ost twice the average. C ontributions from the other large countries are around the average figure.8 Even w ithout going into the question of w hat is the real cost o f the aid to the donor states, the level of aid is extrem ely low, especially if it is remem bered th at a large part o f the aid feturns to the donors either in the form of orders or o f salaries for the specialists and the adm inistrative and technical personnel. These factors, if taken into account, would depress the real level of aid even further. The rapid growth during recent years o f aid from the private sector will be noted in Table 50. This growth is due very largely to the rise in export credits which have m ounted from 0.5 billion dollars in 1960 to 2 billion dollars in 1970. As a consequence private investment has grown more slowly. The proportion form ed by aid in comparison with the Gross Domestic Product o f the under-developed countries m ust now be examined. We have estim ated9 the G.D.P. o f the non-com m unist developing world at m arket prices to be 390 billion dollars for 1969/71. This means th at to tal aid has am ounted to 4.0 per cent of the G.D.P. (and grants am ount to about 1 per cent). Thus, although aid does n o t appear much when compared to the resources o f the
Macro-Economic Data
181
donor states, it forms a significant fraction of the dom estic product of the under-developed countries, given the wide gap betw een average levels o f income. It is true th a t aid seems puny when set against the enorm ous needs created by the population explosion, since with a capital-output ratio o f 4.5 (which seems the m ost likely figure) existing aid would only be sufficient to absorb one-third of the current population growth of 2.7 per cent even if it were devoted wholly to investment. Here, of course, we are faced w ith the vast and insoluble problem o f determ ining how efficacious aid is. In some respects however, the problem is misconceived, since it would be arbitraty to try to isolate expenditures made within the total o f the m oney flowing from foreign aid. Thus the fact th a t aid has been used to purchase consum ption goods or for sum ptuary expenses certainly does not imply, a priori , th a t the aid has been o f no use, since it is easy to imagine a hypothetical case in which such transactions could release for profitable investm ent other resources which would have had to be used for expenditures c f this kind if the aid had n o t been available. It is also possible to envisage the reverse case. In actual fact the problem o f determ ining the efficacity o f foreign aid is indis soluble from th a t o f determ ining the efficacity of the allocation of total national incom e and public expenditure. It cannot, therefore, be discussed w ithin the fram ework o f this study, since a com plete analysis o f conditions in each country would be needed and this, for m ost o f the countries in question, is impossible because, as we know, the necessary statistics simply do not exist. Before passing on to a brief discussion o f the geographical distribution o f aid a few words m ust be said about the foreign public debt o f the developing countries. The debts mainly arise from past financial aid and are o f truly heroic proportions. For the whole of the non-comm unist, developing world the foreign debt (public debt or debts guaranteed by the state, n o t yet am ortized and repayable in one year or m ore) grew rapidly by 15 per cent per annum betw een 1955 and 1962 and by 12 per cent between 1962 and 1970.10 From less than 10 billion dollars at the end o f 1955 these debts have m ounted to about 60 billion by the end o f 1970. As debts continue to rise m uch faster than exports from these countries, the servicing of these debts which, in 1955, form ed 4 per cent o f export receipts, form ed 10 per cent by 1970. Thus the reverse of the medal is negative since the servicing o f foreign debt am ounted to over 5 billion dollars by 1970.11
182
II G E O G R A P H I C A L D I S T R I B U T I O N O F F I N A N C I A L AI D
An approxim ate picture of the geographical distribution o f aid provided to the developing countries by the West can be seen from the statistics o f the net flow received from the public sector o f the D.A.C. countries. Figures below relate to the annual average for 1969/71. We can see th at the distribution o f resources to the different countries, expressed in dollars per capita, is very unequal, b u t the average for all undeveloped countries is S4.4. For the three big non-com m unist countries o f Asia the per capita figures are $1.7 for India, $3.5 for Pakistan and $3.9 for Indonesia. The figures are m uch higher for South Korea ($10.6), and South Vietnam ($25.8); and lower for the Philippines ($2.2), Thailand ($2.6) and Sri Lanka ($4.2). In Africa rates for Botswana ($24.2), Congo (Brazzaville) ($22.8), and Tunisia ($24.6) are am ongst the highest. The rates for Tanzania ($4.2), Nigeria ($1.8), and Ethiopia ($1.9) are amongst th e lowest. Figures for some other im portant countries are: Ivory Coast ($11.2), Algeria ($8.7), Morocco ($7.0) and Zambia ($5.0). In the Middle East the tw o extrem es are represented by Israel ($22.3) and Jordan ($26.6) on the one hand, and by Libya and Saudi Arabia, where repaym ents are greater than receipts, on the other. O ther figures are Turkey ($6.7), Iraq ($0.9), Syria ($2.9) and Iran ($1.3). In Latin America the range between highest and lowest is narrower. The highest figures are Panama (16.1), Dominical Republic ($12.0) and Guyana ($24.4). The lowest are for Mexico ($2.9), Brazil ($2.8) and Argentina ($2.8). As a general rule we may form ulate the following law .' The size of public aid received per capita is a function of: (a) the size and population o f the country. The m ore populated a country is the smaller in general will be the level o f per capita aid. This situation seems to result from the fact th at in practice there is a floor and a ceiling to the total am ount o f aid. The floor favours little countries and the ceiling is unfavourable to the very big countries: and (b) the political com plexion of the government. In this case the law works in a m ore com plicated way. Minimal am ounts are obviously directed tow ards the countries totally opposed to the donor states; nevertheless some countries very closely connected w ith donor countries can also be found near the minimum level, Maximum am ounts are received either by the friendly' countries
Material com di
utor
Macro-Economic Data
2 83
whose political or military stability is thought to be precarious, or by countries who are more or less neutral and to whom aid is directed in an attem p t to prevent them from going over to the opposite camp. C. Grow th and level of per capita product or national income We shall now examine the econom ic growth o f the less-developed countries as expressed in changes in their Gross Domestic Product. In spite of some obvious im perfections in these estimates, it still remains true th a t the growth rate o f G.D.P. per capita at constant prices provides the least objectionable of all individual indicators capable of reflecting general economic expansion.12 The present analysis will be limited to the developing countries as a whole and to th e larger geographical regions. Where individual countries are concerned figures giving the annual rate o f growth of G.D.P. between 1960 and 1970 at constant m arket prices will be given in the synoptic table on page 244. The 1970 level o f the per capita aggregate expressed in dollars will also be given. We shall examine such figures for mainland China as are available in order to compare them w ith those for non-comm unist Asia. The sections will end with a description o f changes in levels o f per capita G.D.P. between 1900 and 1970.
I. R A T E OF GROWTH OF GROSS PRODUCT COMMUNIST LESS-DEVELOPED COUNTRIES IN NON
The U.N. statistical offices calculate indices o f G.D.P. for the larger groups of non-comm unist, developing countries, the exclusion o f the com m unist countries being explained on the one hand by the different m ethods o f calculating the aggregates and, on the other hand, by the lack of recent figures for China. The series goes back to 1950. I have com pleted the figures with estimates made by L. J. Zim m erm an.13 I have calculated weighted averages for Asia and the under-developed countries as a whole since such com putations were not made in the Zimmerman series. It should be noted th a t the total for the less-developed countries is, in this instance, formed by the sum o f the Latin American countries plus Asia; the Middle Eastern and African countries being left out. The omission does not distort the results to any great extent, for the countries left o u t form only 20 per cent o f the total and their average lies close to the average of the whole. We should note, however, th at the two series are not strictly
ial com
184
comparable since Zim merm ans series is based on national income and the U.N.s on G.D.P. Moreover the figures relating to the period 190054 are obviously subject to a larger margin o f error than the m ore recent U.N. ones. Nevertheless, as the point at issue is solely the rates o f growth, the tw o series are, for this purpose, reasonably comparable. It goes w ithout saying th at the margins of error in Zim merm ans series are larger since the figures available for the early period are less reliable. This is also true for the developed countries in the same period, although the am ount o f inform ation and statistical estim ates available for these countries are immeasurably greater and more precise than for m ost of the Third World countries. But this does not mean that the U.N. series does not also contain any margin o f error. The latter arises not so much from the m ethod of calculation, as from errors in the basic figures u sed.14 Furtherm ore, as we shall see below, there is the problem o f the accuracy o f the com putations. For all these reasons too m uch im portance should not be attached to the fact th a t the rates o f grow th I have calculated are given with one figure after the decimal point. The results o f these calculations are given in Table 51. The rates o f growth at constant prices in the table show th a t the growth fell into tw o quite distinct periods. The first stretches from the beginning of the century until the end o f the second world war and is m arked by a regular rate of growth in the under-developed countries o f about 2 per cent. But, as at the same time population grow th was accelerating, it resulted in a continuous fall in the per Table 51 A nnual rates o f change in the Gross Domestic Product per capita (at constant prices)
A ll non-com m unist less-developed countries LA TIN AM ERICA Total 190019131 1 9 1 3 -1 9 2 9 1 19291952/541 1 9 5 0 -1 9 6 0 1 9 6 0 -1 9 7 0 2.1 1.9 2.2 4.8 5.1 Per capita 1.2 0.9 0.6 2.4 2.4 Total 2.1 2.8 3.5 5.4 5.5 Per capita 0.3 1.0 1.4 2.5 2.5 A SIA excluding com m unist countries Total 2.2 1.3 1.1 4.1 4.8 Per capita 1.5 0.8 - 0 .3 2.0 2.1
1These are N ational Incom e series; A frican and M iddle Eastern countries are ex clu d ed (see tex t). S o u rces 1 9 0 0 - 1 9 5 9 derived from L. J. Zim m erm an, T he d istrib u tio n of w orld incom e 186019 6 0 in Essays on Unbalanced G row th Cs G ravehage, 1962); 19 5 0 1970 derived from Sta tistica l Y earbook for 1970 and 1971 (U .N .); W orld Hank and l.D .A . A n n u a l R ep o rt 1972 (W ashington, 1972) and D evelopm ent A ssistance 1972 (O .E.C .D ., 1972).
ial com
or
Macro-Economic Data
185
capita rates o f growth: 1.2 per cent from 1900 to 1913; 0.9 per cent from 1913 to 1929; and 0.6 per cent from 1929 to 1952/54. The latter rate would be a little lower if the years 1945/48 were taken as a term inal point instead o f 1952/54. Grow th during this period was definitely higher in the Latin-American countries than in the Asian countries except for the years 190013. During the second period there was a very great increase in the pace of growth beginning approxim ately at the end of the war. Global rates were nearly 5 per cent per annum and per capita rates about 2.4 per cent per annum. Here, we obviously encounter the problem o f how truthfully this change in the growth curve reflects reality. Thus, in the period 1955 70 and for the non-communist less-developed countries taken as a whole, there was, according to the U.N. figures, an annual increase of 2.4 per cent in the G.D.P. per capita at constant prices. But, I have already been able to show from my own calculations made in Chapter 2 th a t agricultural productivity remained stable in the developing countries between 1953/57 and 1968/72, and th at this probably led to a stagnation in the per capita income of the rural population which, as we have seen, forms about 70 per cent o f total population. In such conditions the increase in the real income of th e rem ainder o f the population (especially that part engaged in industry) would have had to be very large indeed, even allowing for the fact th a t the average income of the nonagricultural section of the population is higher than it is for the agricultural population. We saw in the chapters devoted to the m anufacturing and extractive industries th at both these sectors experienced a substantial expansion 4.1 per cent per capita for m anufacturing, and more for extractive industry (the latter however o f m inor im portance). If we assume th at growth in the sendees sector was as rapid as in industry and th a t average income in industry and services together was five times that in agriculture, this would account for the global rates o f growth exhibited in the U.N. figures. But this assumed gap between the agricultural and non-agricultural incom e is clearly too great; so it seems probable th at growth rates of per capita G.D.P. in the under-developed countries have been slightly over-estimated. This over-estimation does n o t originate in the com putations made by the U.N. statistical office but in over-estimates o f growth made in a large part of the under-developed world. The existence of this over estim ate is confirmed by K uznets recent stu d y ,1 5 in which he makes certain com putations and estim ates intended to remove some of the statistical biases appearing in th e U.N. calculations o f growth in per
186
capita o u tp u t for the period 195458 to 1964 68. So far as the developing countries are concerned Kuznets concludes th a t the annual growth rate in the volume of per capita o u tp u t was only 1.06 per cent instead of 2.21 per cent. On the other hand, he concludes that for the developed countries the growth rate was underestim ated: the figure should be over 4.0 per cent rather than 3.38 per cent. If we look at the difference in the (uncorrected) rates o f growth in the larger regions we can observe th a t (as happened earlier in the century) growth was slightly faster in Latin America than in Asia. In Africa, where before 1960 the figures are to o fragm entary to be used individually, growth between 1960 and 1970 was even more sluggish than in Asia, with the annual growth rate o f the volume o f o u tp u t per capita being 2.0 per cent. On the other hand the inequalities of growth as betw een individual countries and social groups m ust be underlined. Regarding the inequalities betw een countries it should be sufficient to refer to the synoptic table at the end of the book. The inequalities betw een social classes is a m uch m ore serious problem. Although it is not possible to form a judgem ent from statistics alone, it seems v ery lik e ly that in m o s t ca ses o n ly a fe w classes rela tiv e ly fa v o u red t o begin with profited from econom ic growth, thus making the distribution o f income even m ore unequal. The data on agricultural productivity which I have assembled is an indication o f changes of this kind. But we m ust await the collection o f m ore reliable and m ore abundant statistics on income distribution before we can make positive and explicit statem ents about it.16
II. C O M P A R I S O N S WI T H D E V E L O P E D C O U N T R I E S
Let us now look briefly at the relationship betw een growth rates of G.D.P. in the advanced and in the less advanced countries. It m ay be accepted th at in the first phases o f industrialization (roughly during the nineteenth century) growth rates were betw een 1.5 and 2.5 per cent per annum for to tal gross product and betw een 0.5 and 1.5 per cent for per capita gross product, with the average for the latter standing at about 1 per cent. These rates apply to the great m ajority of the advanced countries which began to industrialize during the second half of the eighteenth century or the first half o f the nineteenth. The growth rates o f countries such as Italy, Japan and the U.S.S.R., which began to industrialize later, m ust now be considered. It has already been pointed o u t in' the chapter on m anufactures th at in these countries, and in the countries with
Macro-Economic Data
187
non-planned economies, growth in recent years was more rapid when com pared to th at of th e older industrialized countries, b u t th a t in earlier periods the difference was so slight as to be alm ost non-existent, except in the case of the U.S.S.R. Here, however, the question of different economic regimes and systems o f national accounting is relevant. From fragm entary figures collected in order to com pute levels of industrial developm ent I have derived the same conclusions for growth over longer term. So far as the growth of national product (in per capita terms) is concerned, changes in different countries were very similar. Thus the annual growth rate of national income per capita between 1880 and 1960 was 1.7 per cent for Italy and 2.1 per cent for Jap a n .17 C ontrary to w hat is generally believed, growth rates in the developed countries have been higher in the last tw enty years than they were in the nineteenth century. Thus from 1950 to 1970 the growth rate per annum of G.D.P. has risen to 4.5 per cent (i.e. about 3.5 per cent per capita). Consequently it seems th at between 1900 and 1950 the growth rate of total dom estic product o f the less-developed countries was similar to th a t o f the advanced countries during their own take o f f . But the greater population growth in the Third World countries means the per capita rates have been less, approxim ately one-third less, than those o f the advanced countries. The rates achieved since 1950 insofar as they reflect reality, are actually higher than the rates for the take o f f period of the developed countries. This com parison, however, is not really valid since it is one betw een m edium-term rates of growth in developing countries and very long-term rates in th e developed ones. On the other hand, in view o f the acceleration o f more recent growth in the developed countries, actual rates o f increase o f dom estic o u tp u t per capita in the Third World are n o t so high as the present rates in the w estern developed countries.18 Consequently th e gap between developm ent levels in these tw o groups o f countries is constantly widening, as we shall see below. Before going on to consider w hat changes have occurred in China a global figure should be calculated for the level of G.D.P. at m arket prices. According to my own estim ates this am ounts in 1970 to 390 billions (in current dollars) for the non-com m unist under-developed countries and to 2090 billions for the non-com m unist developed countries.
188
III. M A I N L A N D C H I N A
It must first of all be recalled that the concept the com m unist countries em ploy to determ ine their gross product, which they call material product, differs quite noticeably from the one used in the West. The essential difference is that, contrary to w hat happens in the West, national accounting in the Eastern countries does n o t take into account activities which do not contribute directly to the material production. This excludes public adm inistration, national defence, personal and professional services and similar activities. It should be noticed, however, th a t distribution in the wider sense o f the term restaurants for instance enters into the calculations, insofar as it contributes to the distribution o f m aterial goods. M ethodological differences of this sort, quite apart from any possibility of voluntary over-estimation, present obvious difficulties for international comparisons. When considering growth rates, we have to ask ourselves w hether the bias would tend to raise or to lower the expansion rate. It would appear th a t the tendency would be to raise the estim ated rates since the com ponents of domestic product which are om itted would be expected to have a slower rate of growth than the whole. Nevertheless as it forms only a small p art of the whole (16 per cent), its im pact o n the fin a l figu re is a lim ite d one. Even assuming a rate o f growth for this part o f the to tal o f less than 50 per cent o f th at o f m aterial product, the rate o f growth o f the total would only be reduced by 8 per cent. There is also the recurring problem of the reliability o f the official estimates. As a scientific approach to this problem is obviously difficult, I shall give here, as I have on each occasion when Chinese figures have been discussed, tw o sets o f figures, one being the official set, and the other th at compiled by western experts. But both the official and the unofficial comprehensive estim ates of m aterial o u tp u t growth (and one explains the other) for all practical purposes stop at 1960, i.e. at the end of the G reat Leap Forward. I shall divide the changes betw een 195070 into two periods 195057 and 195770. Official statistics for the period 195057 reveal an annual rate o f growth o f net m aterial production at constant prices of 12 per cent. Western estim ates reduce this rate to about 7 per cen t.1 9 Assuming, therefore, a rate of population growth of 2.0 per cent, this gives us an increase in income per capita of respectively 10 per cent according to official estimates and about 5 per cent according to western estimates. The second period (1957 70) reveals a situation which in statistical term s is very hard to evaluate. We m ust rem ember th at
Macro-Economic Data
189
1958 saw the beginning o f the notorious Great Leap Forward, which was followed in 1960 by a period o f adjustm ent th at seems in 1961 and 1962 to have been m arked if not by a recession then at least by a status quo in total o u tp u t or a set-back in per capita output. From 1963 onwards progress again seems to have got under way. To judge from the evidence in miscellaneous sources, growth began to increase from 1964 65 onwards. According to official estimates the net material product at constant prices increased by 34 per cent in 1958 and 22 per cent in 1959. A t th at date all comprehensive statistics stop, and from then until 1973 only vague indications have been published in the form of a growth rate in certain sectors com pared to th a t of the preceding year, b u t generally w ithout any reference figures. From this incom plete data for the period 1960 70 and from the m ore complete data available for 1958 and 1959, we can conclude w ith our usual reservations th a t the Chinese authorities assume for the whole period 195770 an annual rate o f growth of the total volume o f m aterial production of about 67 per cent. There are obviously very few western estim ates for the period in which we are interested and m ost o f those stop in 1959. They give a rate o f growth for 1958 and 1959 which is less than half o f the official estim ates.2 0 On the basis o f western estim ates o f industrial and agricultural growth (reference should be made to the relevant chapters for the sources o f these estimates) it is possible to arrive at an average rate of growth o f th e total volume o f Chinese o u tp u t for the period 1957 70 o f 45 per cent per annum. This rate, 4 5 per cent, is obviously no more than an indication, since, as was pointed out above, m ost western estimates term inate in 1960. In fact we shall have to wait until the Chinese government sees fit to publish sufficiently detailed statistics for one recent year before we can resume our attem pt at estimates more accurate than the ones cited here. If the rates quoted here for the whole of the period 195070 are now com pared to the corresponding rates for the non-communist Third World countries (see Table 51) it appears, as the figures for agriculture and industry foreshadowed, th at growth in China has been faster, even if we take only the western estimates. For the whole period 1950 70 per capita o u tp u t in China increased at the rate o f about 6 7 per cent (according to official estimates) or about 3 4 per cent (according to western estimates). The corre sponding rate (not corrected for possible over-estimation) for the non-comm unist under-developed countries o f Asia is about 2 per cent, and about 2.4 per cent for th e whole o f the under-developed
190
world. If th e western estim ates are taken, rates as high as those o f China can be found and are even surpassed in some non-comm unist under-developed countries (for example in Taiwan, 6.3 per cent between 1955 and 1968; in Thailand, 5.3 per cent between 1958 and 1965; in South Korea, 6.2 per cent between 1960 and 1969; in Puerto Rico, 6.4 per cent between 1960 and 1967; in Zambia, 8.8 per cent between 1964 and 1967). But the exceptional nature o f the growth of Chinese national product resides in the combination o f the three following facts: 1. the growth took place over a fairly long period; 2. it took place in a country with an enorm ous population; 3. it took place in a country receiving very little foreign aid.2 1
IV. C H A N G E S IN P E R C A P I T A I N C O M E A N D P R O D U C T
Table 52 brings together the results o f our calculations about the changes in th e level o f per capita income betw een 1900 and 1970 expressed in constant dollars o f 1970. There are tw o distinct series. The first, which runs from 1900 to 195254 concerns per capita national income and is based on Zim merm ans figures,2 2 from which are made estimates concerning the under-developed countries of Latin America, South East Asia and the Far East. On the basis o f the tw o latter regions I have calculated a weighted average for non com m unist Asia, excluding the Middle East. In order to calculate the level of per capita national income for the whole of the under developed world I have assumed th at the countries for whom Zimmerman did not make any estimates (Africa and the Middle East) had, taken together, a per capita income similar to th a t o f Asia, as was the case in 1958. Insofar as the developed countries are concerned, I have been forced to recast the individual figures for the various countries in order to make the com position o f the whole similar to the definition adopted for the non-communist advanced countries which has been used in the second series and throughout this study. The second series from 1958 to 1970 is based on the calculations o f the statistical offices of the U.N. which have been supplem ented for 1970 by my own estim ates on the basis of the figures of each country. In bo th cases the results have been recalculated in terms o f 1970 dollars by using the im plicit price deflator resulting from the calculations of United States gross product at constant prices. It m ust be pointed o u t th a t the figures for 1900 to 1952/54 are not strictly comparable with those for the m ore recent period
ial com
Macro-Economic Data
191
especially where the under-developed countries are concerned. It is also likely th a t the series covering Latin America over-estimates the real developm ent: a likelihood revealed when the figures are p u t beside those for 1960. As the m odern tendency is to form ulate the evidence o f national accounts in terms o f per capita gross product, I have shown in Table 53 figures for the medium-term changes o f this aggregate in the larger regions and including China. Table 52 Long-term changes in national income per capita _________ (in U.S. dollars o f 1970)_______________________
Less Developed Countries 1 Total 19002 19132 19292 1952/542 1960 1970 110 120 130 150 170 210 Latin America 230 240 280 390 380 500 Asia 90 110 115 110 130 150 Developed Countries' Total 640 775 930 1360 1780 2610 United States 1360 1580 1840 2850 3350 4270 Europe 500 595 660 840 1270 2050
1 E xcluding C om m unist countries 2 N ot strictly com parable w ith recen t d a ta (see te x t). Sources: 1900 -1 9 5 2 /5 4 derived from L. J. Zim m erm an, T he D istribution o f W orld In co m e 18601960; 196 0 1970 derived from Y earbook o f N ational A c c o u n ts S ta tistics 1971 (U .N ., 1973).
The figures for China included in this table have been calculated in the following way: the base figure for 1953 is based on two estim ates Ecksteins,2 3 who concludes after a detailed study of Chinese national income, that G.N.P. per capita expressed in American prices for 1952 was betw een $80 and $120, and, if the median o f the spread is taken, $100 and W. W. Hollisters,24 who gives $145 in American prices for 1955, i.e. a figure which is 45 per cent higher than Ecksteins for 1952. Allowing for probable changes between 1952 and 1955, the average o f the two estimates gives us a figure for per capita national product expressed in American prices for 1953 o f about $120. But, as with other countries, the figures do not allow any correction for differences in purchasing pow er.2 5 I have therefore reduced Chinas figure to about $70 for 1953 or about $110 in 1970 dollars. This is the western estim ate; official ones give a figure 10 to 15 per cent higher, i.e. about $80 or $120 in 1970 dollars. I have applied annual growth rates given in the preceding section to these 1953 figures and reference m ust be made
192
to th at section for reservations expressed about the rates, especially those from 1960 to 1970. A ttention m ust also be drawn to the fact th at, as already pointed out, the figures given in the table are expressed in dollars based on m oney conversions determ ined by national rates o f exchange. But as we know, current exchange rates do not always represent the real relationship between local prices and those operating in the United States. Some research has been done to try to provide a statistical estim ate o f the differences,2 6 especially by G ilbert and Kravis. But as yet these estim ates have not yielded results which can be considered entirely satisfactory or which cover the whole period and all the countries in which we are interested. According to estimates made by Delahaut and Kirschen,2 7 the gap betw een the prices in under-developed countries and those in the United States was about 80 per cent in 1957. If this is so, we should have to increase the level o f income o f the less-developed countries by about 80 per cent to allow for the different structure o f prices. It should be noticed nevertheless th at the gap of 80 per cent applies specifically to the relationship o f the under-developed countries and the U.S. But as the price structure in the other developed countries differs from th at of the U.S., the rate o f correction sh o u ld be a b o u t 65 70 per cent if the average level of the under-developed countries is com pared to the advanced countries taken together. Some such implicit allowances for the differences m ust be made if the gap separating the average levels of income in the Third World and the advanced world is to be assumed. As can be seen from Tables 52 and 53, because of the slower rates o f growth taking place in the non-com m unist under developed countries, the gap which exists between them and the advanced nations is increasing steadily and rapidly. Thus the divergence between per capita income or product, which was about 1:6 in 1900 (and 1:5 in 1860), rises to 1:7 in 1929, 1:8.5 in 1953 and 1:13 in 1970. In my opinion the divergence for 1980 will be larger still, since I believe th at the next few years will bring a slowing down of the growth rate o f per capita national product due to the combined effects o f population growth (which in turn will lead to a slower pace o f agricultural productivity growth or, as this has already occurred, to an absolute decline) and a reduction in industrial expansion. But projections in this field are always extrem ely hazardous.2 s If we now make an allowance for the differences in the price structures, the gap between the level o f the Third World and the non-comm unist advanced world which was, as we saw, about 1:13 in 1970 becomes 1:8. The gap is adm ittedly smaller, b u t nonetheless
Macro-Economic Data Table 53 Gross Domestic Product per capita (in constant 1970 U.S. dollars)
1900 N on-com m unist less-developed co an tries TOTAL Africa Latin America Asia Middle East China (mainland) Official estimates Western estimates N on-com m unist developed countries TOTAL United States Europe 1953 1960 1970
193
105
155
220 90
350 105
180 130 420 120 330
230 160 540 145 470
120 110
3001 1702
350 160
705 1495 550
1455 3200 875
1965 3690 1375
2900 4730 2260
1 175 for 1957 2 1 3 0 f o r 1957 Sources: 1900, a u th o rs estim ates based on Table 52. 195 3 70 (except China, see te x t), derived from Yearbook o f N ational A c c o u n ts Statistics 1971 (U .N ., 1973) and previous issues o f th o se publications.
substantial. Under a relatively optim istic hypothesis, which assumes th at the non-comm unist under-developed countries would experience a future rate o f growth similar to th a t achieved between 1950 and 1970, these countries will need about 90 years to reach even the present level of the developed non-com m unist countries, allowing for this difference in price structure, and 110 years if the allowance is not made. But the gap which separates the average level in non-communist Asia from th at in China is more im portant in its implications. While in 1953 Chinas level o f per capita o u tp u t was about the same or even slightly less than th at o f the other Asian countries, by about 1955, thanks to a faster growth achieved since then, it had surpassed n o n -c o m m u n ist Asias. By 1960 it was higher by some 40 per c e n t, if we use western estimates, and by 150 per cent if we use government figures. The gap is a little smaller in 1970 (about 45 per cent and 140 per cent respectively), b u t it should be rem em bered th at 1960 was the peak year o f the Great Leap Forward and th at the figures for 1970 are much more unreliable. All extrapolations of this trend are hazardous where changes in recent years are concerned.2 9 I shall not pursue the econom ic and political implications opened up by these changes, since I shall discuss them in the next and concluding chapter of this study.
I I
General Conclusions
In the last chapter we were able to sketch o u t a global diagnosis of economic change in the Third World, and to lay bare the bones of the problem the painfully slow progress o f economic growth since 1900, which remains, despite a slight im provem ent since 1950, insufficient to prevent the gap betw een per capita incomes in the Third World and the developed countries from widening still farther. The sectoral analysis contained in the preceding chapters has helped to explain the causes underlying the sluggish economic growth and to fill in the details o f the global diagnosis. But the com ponent parts of the problem are m any and complex and it is an illusion to suppose th at the balance sheet could be either brief or straightforw ard. We have seen, for instance, th at the great expansion of the extractive industries, which on a superficial analysis m ight appear to be a good thing, has in fact had more unfavourable effects than favourable ones. On the other hand an analysis o f the term s o f trade enabled us to revise the unfavourable verdict on the secular changes in the terms of trade of raw materials th at has for so long been generally accepted. There is also the problem of regional differences, since (as was pointed out in the introduction) this study cannot claim to analyse econom ic change in all the 170-odd territories th a t go to m ake up the Third World. Nonetheless an attem pt has been made to provide some account of the more obvious variations. The question mark raised by the presence o f China, an undevelop ed country with some 800 million inhabitants th a t has opted for a path to developm ent different from th at o f the rest o f the Third World, has also been brought into the open, irrespective of the econom ic and political undertones which such a discussion m ust introduce. For all these reasons the conclusions which now follow will be treated in three separate sections. The first will give a sectoral diagnosis o f economic change in the non-comm unist Third World, with reference, where necessary, to larger regions. The second will be devoted to a comparison o f China with the other under-developed countries, but more particularly of China as against non-comm unist Asia. Lastly, the main outlines of a plan will be sketched out which might go some way towards solving the crucial problems of under-development. 194
General Conclusions A. Sectorial diagnosis of non-com m unist Third World
195
We shall begin with the population question. C hapter 1 presented the quantitative elem ents of the problem , which revealed tw o im p o rtan t factors in the under-development equation. The first relates to the magnitude o f population growth. While we keep as a reference for the past an annual rate of 2.2 per cent, this has today already risen to 2.6 2.7 per cent. The second factor concerns projections for the future. We m ust prepare ourselves for the rate to reach 2.8 per cent during the next ten or fifteen years, whatever the effort to be expended on the various birth control campaigns, and whatever the results to be achieved thereby. This population explosion no other expression will do bears no resemblance to the slow increase experienced in western societies before and during their take o ff, and is one of th e major explanations of the unfavourable trend in the production o f food, revealed by the index of agricultural produc tivity. The implications o f this m ost unfavourable diagnosis are all the more significant when one remembers th at the main reason why agriculture is so essential is th at more than three-fifths o f the Third W orlds labour force and nearly three-quarters o f the labour force in Asia and Africa works on the land. We shall return to this aspect of th e problem later. The num ber o f persons engaged in producing food in the Third World has doubled during the last half-century as a direct result of the population explosion. This fact, com bined with the small am ount of agricultural land available, has caused the old law o f diminishing returns to reappear in a m ost disagreeable form. According to my calculations, agricultural productivity in the Afro-Asian countries has fallen by some 20 per cent between 192226 and 1968 72. The greater availability of land in m ost Latin-American countries has helped, at least partially, to prevent a similar deterioration in th at part o f the world. It is, in fact, possible to establish a correlation between the availability o f land and the level o f productivity. Similarly a correlation exists between low agricultural productivity and the presence o f plantation agriculture, the explanation probably being that the plantations have occupied the b etter lands. But perhaps m ore significant is th at this correlation indicates that plantation agriculture has relatively few positive effects on food production technology. If we look at the actual level o f agricultural productivity we can see the difference between the Latin-American and the Afro-Asian countries is in a ratio of about 2 to 1, while the countries of the Middle East occupy a position between the tw o. The productivity o f
196
agriculture devoted to growing food in m ost o f the African and Asian countries is below the point at which we have estim ated they would be safe from risk o f famine. Even the relatively high level of agricultural productivity which has perm itted Latin-American countries to enjoy a m ore advanced industrial development cannot be considered wholly satisfactory, except perhaps in Argentina. Between 1913 and 1970 the situation has at least been relatively static. But even if the position is n o t as dram atic as it is in Asia, it is still a fact th a t rapid progress in the industrial field cannot be expected w ithout an increase in agricultural productivity. We shall have to return later to this aspect of the problem. Thanks to similar com putations, made for an earlier study, it has been possible to com pare the actual levels o f agricultural produc tivity achieved by the less advanced countries w ith those o f the more advanced countries at th e tim e of their 'take o f f . This has produced the following conclusions. The average level o f agricultural produc tivity in the African and Asian countries, which form four-fifths of the Third World population, is about 50 per cent below w hat was achieved by the developed countries at the start of the industrial revolution. Actual productivity achieved by the less-developed countries is about the same as it was in the European countries before their agricultural revolution. As was pointed o u t in R evo lution industrielle et sous-developpement , the western countries had undergone what Marc Bloch called the agricultural revolution before embarking on th e accelerating process o f industrialization traditionally know n as the industrial revolution. The earlier revo lu tio n had m ade it possible for agriculture to increase its produc tivity by 40 per cent w ithin a period of about forty to sixty years, and thus to transfer to expanding industry part of its labour force, purchasing power and savings. But m ost o f to d ay s developing countries wish, either consciously or unconsciously, to skip this stage even though other structural pre-conditions for developm ent have made the take o ff much harder now than it was in the nineteenth century when m ost European countries and the United States were trying to em ulate Englands example. Failure to understand these historical differences is all the m ore serious because increasing agricultural productivity is so particularly difficult. Agricultural productivity depends m uch m ore on structural factors than does industrial productivity. For instance, even if land ownership and political factors are left out o f consideration, an increase in the area under cultivation per agricultural labourer is an essential condition
General Conclusions
197
for rising productivity. But, faced with a population explosion on to d ay s scale, it is impossible for even the m ost favourable hypothesis to assume anything other than a slight increase in total area under cultivation, and a status quo as regards area per labourer. Recently much has been heard o f the green revolution and the possibilities of growing new varieties o f cereals, especially highyielding rice, selected to m atch the climate and soil chemistry o f the Third World. As so often happens, early optimism has now been suc ceeded by exaggerated pessimism. The favourable results to be expec ted were perhaps over-estimated, and w hat was probably no m ore than a normal rise in o u tp u t due to favourable weather conditions was interpreted as a revolution. The new seed has, it is true, increased yields and productivity on farms where it has been sown, b u t a wider use o f the seed, the experts tell us, would mean a more intensive application o f fertilizers and pesticides, which would reduce the pace at which the new seed could be introduced. A nother handicap is that the new varieties o f cereals (particularly rice) are not always acceptable to the people they are intended to feed. In addition more irrigation is required. Thus, the real problem is not so much the actual discovery o f new ways to increase o u tp u t and hence productivity, as o f b etter and faster ways of introducing the new techniques, given existing ecology, soil chemistry and social habits. Even then, there are dangers in concentrating on a limited num ber of seed varieties in countries which are ill-equipped to fight plant diseases. We know th at in nineteenth-century Europe new agricultural m ethods spread only a few miles a year. Modern m ethods of com m unication could certainly help to speed the spread, but we should not forget th at country people today are still largely illiterate. In these circumstances the main problem is not so much to discover innovations as to propagate inform ation about them. It is therefore m uch too early to pass a balanced judgem ent on w hat has so far taken place. A verdict cannot probably be delivered before 1980 and meanwhile there remains the possibility of a serious famine in the Third World which, by 1980, will have to find food for nearly one-third m ore m ouths than in 1970. We cannot as yet speak of a revolution, and can only nourish a fervent hope that we shall live to see the fear o f famine recede and to witness a worthwhile increase in agricultural productivity in the near future. An im provem ent in agricultural productivity is likewise absolutely essential for the prospects o f industrial development. For, if our study of m anufacturing industry has revealed unm istakable signs of
198
substantial gains in this sector, particularly since the end o f the last war, it has also shown us th a t agriculture has a part to play in the growth of industry, notably through its effect on the purchasing power o f the agricultural population. If the expansion of m anu facturing industry has been possible up to now w ithout a noticeable enlargement of the domestic m arket, it is due to the fact that domestic production has mainly been substituting for imports. As the limits of im port substitution are approached we may expect a reduction of industrial expansion, unless agricultural productivity increases. Besides a comparison of developm ent in m anufacturing industry has revealed that the level reached by the Afro-Asian countries in 1970 is very low, close to th at o f the western countries in the early years of their industrialization (1770 90, say, in England, or 180010 in France). Latin Am ericas level, on the other hand, is higher nearer to th at o f France in 1880 90; and in the case of Argentina and Chile nearer to w hat France had achieved by 190010. In this field, as in agriculture, the Middle East occupies an interm ediate place between Latin America and Asia, although nearer the latter. We may compare the level it has reached with th at of France in 184050 or England in 180010. The part played by the extractive industries in the industrializ ation and general developm ent of the Third World is a com plex one. As has been pointed o u t before, this sector underw ent, particularly after 1948, a very considerable expansion (an annual rate of growth between 1948 and 1970 of more than 9 per cent), b u t in fact the expansion, which was helped by the fall in transport costs, had a negative rather than a positive effect. Intended m ostly to serve consum ption in the m anufacturing industries o f the advanced economies, th e o u tp u t from extractive enterprises so far has hardly encouraged any extension of economic activity within their own lands. Accompanied usually by the use o f foreign capital equipm ent and export, of profits, the present rapid growth o f extractive industries could mean th at future developm ent is mortgaged, since there is a very real danger that by the time the economies o f which they form part have reached a stage in their developm ent when they could com pete with the developed countries in the transform ation of raw materials, their mineral resources will have been exhausted or at least much reduced in economic value through having had much of their potential skimmed off. This brings us to a discussion of foreign trade, since, contrary to the experience o f the western countries in the first phases of their take o ff, foreign trade is very im portant to th e developing
General Conclusions
199
economies. For instance, total exports form at the present time about 15 per cent of G.N.P. of the non-communist, under-developed countries, and for some o f them the rate exceeds 30 per cent. The structure o f foreign trade simultaneously reflects their low stage of development (a heavy reliance on raw material exports and m anu factured im ports) and their acute dependence upon m arkets in the advanced countries o f the West. This is revealed by the fact th at around 1970 exports from Third World countries to the developed countries represented 11 per cent of the Third W orlds G.N.P., while exports from developed countries to the Third World represented only 2 per cent of the G.N.P. of the developed countries a proportion which is continuing to decrease. Because of the relative preponderance o f foreign trade the terms of trade form one o f the major factors in a global diagnosis. For this reason economists and politicians in the under-developed countries have concentrated (and m aybe over-concentrated) on this problem A critical analysis of the indices of export prices used in the past to estim ate the term s o f trade has enabled us to conclude th at the generally accepted view th at there was a secular deterioration in the terms of trade o f raw materials as compared to m anufactures is erroneous. The great fall in transport costs, occurring between the third quarter o f the nineteenth century and the 1920s, masked an improvement in the relative prices of raw materials, and hence of the terms of trade o f the under-developed countries, o f between 20 and 40 per cent. We m ust remember, however, th a t in the short term (between 1954 and 1962) there was a deterioration of the terms of trade which went far beyond a mere re-adjustm ent o f the relatively high prices of raw materials during the early 1950s. This particular deterioration, which incidentally did not continue beyond 1962, obviously led given its magnitude (14 per cent approxim ately) to a reduced inflow of resources th a t partially offset international aid received during the period. On the basis o f a rough calculation o f the relationship between prices of raw materials and capital goods, one can see that the deterioration in the terms o f trade caused a loss of resources about equivalent to total aid received, defining aid in broad terms, b u t well above real total aid. Before we leave the sectoral diagnosis which reveals negative effects predom inating, and go on to a comparison between China and the non-comm unist Third World, the positive aspect o f educational progress should not be left o u t o f our balance sheet. Although the present state o f progress is not sufficient to provide for the requirem ents of m odern technology, there has been considerable
200
advance, in spite of a population growth resulting in even greater proportionate rises in the population o f school age. This conclusion applies, of course, to education in the quantitative sense; there are many other problems associated with the quality of education. In this co n tex t it should be emphasized th at the irrelevance of educational content and the rapid increase of school enrolm ent, combined w ith the gap between rural and urban wages, contribute to the drift from the countryside to the towns. Allowing only for what happened between 1950 and 1970, the migration balance sheet in the non-comm unist, under-developed countries shows a surplus of 100120 million people in favour o f the cities. The gathering speed o f this rural exodus together with the rapid natural growth of city populations has produced a veritable urban inflation. Urban popu lation rose from some 47 million persons in 1920 to 120 million in 1950 and to 370 million in 1970. This rapid growth of urban population has not been paralleled w ith developm ent in productive em ploym ent; has created a serious problem of under-em ploym ent which is one of the causes of the hypertrophy o f the tertiary sector o f em ploym ent; and has led to soaring levels of unem ploym ent. The unem ploym ent rate has reached 1020 per cent in m any Third World cities, and is particularly evident among young people with an average level o f scholastic achievement. Thus unem ploym ent in the cities has added to the traditional under-em ploym ent of rural society to enhance enorm ously the need for productive em ploym ent. To reduce real, i.e. full, under-em ploym ent (unem ploym ent properly defined plus under em ploym ent expressed as total inactivity) from the 1970 level of 2030 per cent to a figure o f 10 per cent by 1985, it will be necessary to create 350 4 00 million new jobs in the under developed world, i.e. more than were created during the entire century previous to 1970. B. China compared to the rest of developing countries We now turn to the second set of conclusions which concern the comparison between China and the non-communist Third World. The diagnosis for the non-comm unist, under-developed world is a generally negative one. The early effects of industrialization are more than offset by stagnation, if n o t regression, in the agricultural sector. The Third World has started badly and the chances o f a sudden acceleration into economic developm ent in the near future are slim. China, by contrast, appears to have achieved such a successful take-off. We are, of course, faced with the difficulty of assessing the
General Conclusions
201
statistical inform ation contained in two separate series of economic indicators. The first, coming from official sources, presents develop m ent in an extrem ely favourable light, whereas the second series, prepared by western experts, throws considerable doubt on these figures. But a successful Chinese take-off is obvious even if we base our diagnosis on the lower western estimates, and the success appears even m ore noticeable when compared with the perform ance of the non-com m unist, under-developed world. China appears to have recognized the priority th a t should, in our opinion, be awarded to agriculture, and it seems as if her efforts in this direction have borne fruit, not perhaps as abundantly as the jubilant announcem ents o f the Great Leap Forward would have us believe, b u t very considerably nevertheless. According to western estimates o f agricultural output, productivity has increased by about 30 per cent in this sector betw een 195357 and 1968 72. If this fact has im portant political implications for the present and the future, it also has positive implications on the purely economic plane in th at it dem onstrates effectively th at agricultural developm ent is not impossible within th e fram ework o f the conditions prevailing in the less-developed countries o f Asia, i.e. great population growth coupled with a limited quantity of land on which fertility is mediocre and where climatic conditions are m ostly unfavourable. It is theoretically conceivable th a t agricultural developm ent in such circumstances would be impossible and the regression in agricultural productivity suffered by countries in Africa and Asia suggests on practical grounds th at it has been so. Nevertheless we m ust remember th a t China, by comparison w ith the other larger Asian countries, has slightly more agricultural land available and th a t she was not subject to a colonial regime in the same direct way as India was. That a colonial regime could have had positive effects, such as the creation of a considerable educational, medical and transport infrastructure (India possessed 59,000 km of railway netw ork in 1920 as against Chinas 10,800 km) while at the same time have had such a depressing influence on the cultivation of food crops was, amongst other reasons, due to the large plantation econom y. But China through the grow th of productivity in her agriculture has progressed beyond the point at which nationwide famines are likely and has consequently been able to industrialize faster than the rest o f Asia. The com bined effect o f rapid growth in both sectors has m eant th at since 1955 China has had a higher level of per capita income than the rest o f Asia, whereas in 1950 she had about the same or even less. By 1970 per capita income was as m uch as 50 per cent
202
higher even by western estimates. If official figures are taken, growth w ould appear to have been even faster and the divergence still greater. We m ust not forget, however, th at the main part of the process of catching up and overtaking was achieved during the decade 1 9 5 0 -6 0 . Because of bad harvests and the disorganization of industrial production after the Great Leap Forward and disturbances to ou tp u t caused by the cultural revolution, Chinese economic growth slowed down considerably in the decade 196070. Even so, it remained slightly faster than th a t o f the rest o f Asia taken as a whole, even though not faster than th a t of some countries taken individually. If the next decade resembles 196070 rather than 1950 60 the difference will, however, no longer be so marked. Chinas success and the, at least partial, failure o f the non com m unist Third World pose the question as to w hat are the causes underlying the difference. U nfortunately there are no clear cut answers to this question, since if, to simplify the m atter, we take the two great Asian powers, China and India, we have to adm it that before 1949, while their levels o f economic developm ent were fairly similar, their structural conditions were much less so. In addition to differences in agriculture which have already been taken into consideration (availability of land and the existence of a plantation econom y), we also have to adm it th at the absence o f a direct and long-established colonial regime in China m ust have influenced the industrial sector: English m anufactures had penetrated the country sufficiently to cause the handicrafts o f India other than those engaged in a rt products largely to disappear, whereas they contin ued to exist in China and played their part in the process of industrialization. But the fact th at India received more financial aid than China m ust also be recorded. Structural differences like these, as well as others equally influential in the realms of religion and the social fram ework, prevent one from concluding th at the extent of Chinas economic success should be attributed to the prevailing political regime. But I think we may safely say th at the Chinese way is very likely to be a t the root of such economic success as this country has achieved, although it is by no means certain it would work equally well if applied to other societies. The effects o f sheer size are also left o u t o f consideration. China forms an economic unit with a larger population than th at o f 164 countries taken together out of the total o f 170 which makes up the Third World. The Chinese example, on the other hand, lends additional confirm ation, if that were needed, to our thesis th at agriculture is of
ial com
General Conclusions
203
prim e im portance for economic growth. In my book Revolution industrielle et sous-developpement I had already arrived at this
conclusion on historical grounds. It was possible to assemble
evidence o f the major part played by the increase in agricultural productivity in the genesis o f the industrial revolution in the western countries. The diagnosis o f the economic growth of the non-comm u nist Third World in m odern times has shown th a t agriculture has played, and still plays, a crucial role in industrial conditions. The relative success of China provides, therefore, this additional proof. C. O ptions for future actions This brings us to the last p art of our summing up in which we shall make certain recom m endations suggested by the foregoing diagnosis. Obviously these will be generalized options since the great num ber and variety of individual cases require strategies taking full account of local factors. The only distinction which it is possible to make here is betw een conditions in Latin America and in the rest of the Third World. The first and m ost im portant recom m endation relates to the priority which agriculture dem ands, a priority already stressed. If the existence o f an agricultural surplus in the West and the low cost of transport means the Third World is now less dependent upon agricultural o u tp u t than the advanced countries were in the early nineteenth century, it is nonetheless absurd to expect any industrial developm ent w ithout some prior growth in a sector which provides the livelihood o f 70 per cent o f Third World population (and hence of consumers), particularly in Africa and Asia. Appropriate ways of encouraging agricultural productivity locally need not be spelt out here, since they belong m ore to the com petence o f agronomists than to th at o f economists. But, as Rene D um ont points out, it is possible technically speaking to raise productivity. As for general recom mendations, given the observable differences between agricultural productivity in Afro-Asian and Latin-American countries, it is clear th at the problems of the latter are sociological and political rather than technological, since w ithout agrarian reform a rise in productivity could have b u t a m inor influence on developm ent because the mass of the agricultural population would not be in a position to share in any o f the benefits. The effects o f population growth on the agricultural sector are such as to block progress and even, as we have seen, to cause productivity to fall. This is in itself sufficient justification for a birth-control policy, if indeed any justification were necessary when
204
we call to mind actual rates of population increase in the Third World (2.6 2.7 per cent per annum ). With the exception of China, where population growth appears to be slightly lower less than 2.0 per cent and the territories w ith a high rate o f European immigration (where the problem is different in any case) we cannot quote the example o f a single large country which has up to now succeeded in taking o f f into economic development with a population growth rate o f more than 1.1 per cent. This is why, while giving priority to food-producing agriculture, it is at all costs essential to try to restrict long-term population growth. There is little hope, as we have seen, of any slowing down of population growth in the short term, because of the actual age structure. But if we desire to reduce future growth, and this is an imperative rather than a wish, then fertility m ust soon be substantially reduced. In this connexion it is m ost regrettable th at th e various United Nations agencies occupied with the problem s o f economic developm ent should devote so little effort to this end. Because m ost under-developed countries do not possess large enough scientific and medical infrastructures o f their own, it would be desirable to create an international agency which could concentrate on the following activities:
1. S tim u la tin g sc ie n tific research in to te c h n iq u e s o f birth c o n tro l
by improving the co-ordination and diffusion o f w ork already in progress; by subsidizing such; and possibly, by establishing an international research centre as has been done for cancer research. 2. Building up a corps o f specialists who could undertake local birth-control campaigns and study the psychological aspects of the problem s with the object o f using propaganda techniques th a t take local attitudes and customs into consideration. Let us now pass on to the industrial sector and exam ine the extractive and m anufacturing industries separately. So far as the latter is concerned my view is th a t it would be desirable to hold back on the hasty construction o f extrem ely m odern industries, however justified this might be on psychological grounds. Such a policy would be likely to prejudice th e m ore general progress o f developm ent because of its high cost and because it would n o t stim ulate local activities since nearly all the machinery and, later on, spare parts would have to be im ported. Policy-makers should not be afraid to set up tariff barriers to p ro tect local industries, even if this'w ould at the outset keep in operation factories w ith a low productivity m anufac turing articles o f a quality clearly inferior to those of the developed
General Conclusions
205
countries. At a later stage tariff barriers should be flexible enough to stim ulate some sectors o f industry by allowing enterprises whose perform ance was marginal, to disappear eventually. In addition, and I shall come back to this point later on, it seems essential for the state to play a leading role in industrial investment. It is unrealistic to expect a class o f capitalist entrepreneurs to emerge as they did in the West during the eighteenth and nineteenth centuries. Such a class was able to play its p art at th a t tim e because a com bination of circumstances existed then which no longer exists now. Amongst the favourable conditions were an elem entary technology; small units of production; a large body o f skilled craftsm en; greater intensity of capital per w orker in agriculture than in industry; a total absence of social legislation; no international or even inter-regional com petition; and high rates o f profit in industry. If state intervention is now accepted, then the choice o f types o f enterprises and sectors to be encouraged is easier if based on a logical plan. W ithout discussing which o f the various types o f investment are desirable I shall emphasize th a t in considering the broad choices the essential criterion should be the extent o f the induced effects on local production. Although I consider th at th e success o f industrialization in the Third World is prim arily dependent on dom estic demand, it is nonetheless true th at foreign trade can make a certain contribution. Given a low level of domestic wages, some industries in under developed countries could produce at com petitive prices, and, consequently, could find export m arkets. This is particularly so in industries such as textile m anufacture which em ploy a large am ount of unskilled labour. Lowering tariff or other barriers in the developed countries would help th e Third World to export larger quantities of such products. But, while encouraging trade with the developed countries, it is also essential for trade between Third World countries themselves to be intensified, particularly since the domestic market in m ost of these countries is to o small to provide outlets for a wide range o f m anufactures. Consequently a programme of integration, which in some cases has already been initiated, would be needed. The latter policy holds perhaps the greater prom ise for industrialization since dependence on exports to advanced countries invites the risk o f introducing a new division of labour which would confine the industries of the Third World to sectors characterized by m oderate growth rates. As has already been stressed, extractive industry faces the difficult question of expendable resources. Ju st because these resources are
206
non-renewable it m ight well be unwise to encourage too great an expansion, particularly since the locally-induced effects are generally very small. Detailed studies would have to be made into the implications o f all the choices available, and the long-term advantages and disadvantages of exploiting national resources m ust be carefully weighed in the balance. In m ost cases it would probably be more prudent to hold back developm ent o f mining until the national economy was in a position to use its own raw m aterials.1 It m ust also be rem em bered th at the political conditions o f today, which have fragm ented the old colonial empires, have created th e possibility that the interests of small countries possessing great mineral wealth might conflict with the interests o f the larger region o f which they form a part. The main trends of industrial history dem onstrate the need for a wide choice between m odes o f development or economic regimes likely to favour growth in the unpropitious conditions o f the Third World today. It does n o t seem possible th a t a liberal regime, such as the free m arket econom y which was the setting for the western industrial revolution, could nowadays be capable o f giving lessdeveloped countries similar opportunities for advance. It m ust be recalled th at growth in a m arket econom y, even if we take no account of the social problems caused by the early phases of industrialization, was painfully slow and was spread over a period of about a century, or a century and a half. As readers have been rem inded earlier in this study, the total Gross National P roduct of the m odern developed countries during the nineteenth century only grew at a rate o f about 2 per cent per annum, which is a far slower rate than th at of the population increase in the Third World today. To solve these difficult problem s some system o f planning will be required though this need not necessarily be totally centralized planning. Such a system is, in any case, wishful thinking where under-developed countries are concerned, since it pre-supposes an administrative, statistical and technological infrastructure th a t simply does not exist. Similarly, it would be foolish in the early phases to attem pt to modify the system o f private trading or to change existing systems of land-holding and types of cultivation too quickly and too radically. But the fundam ental choices should be subject to a national plan and the state should, in addition to the traditional fields of centralized powers, intervene extensively in the following ways: 1. by stimulating agricultural development on a vast scale: where this stim ulation requires an agrarian reform this reform should be a priority;
General Conclusions
207
2. by creating industrial enterprises which favour general economic development and by devising a tariff policy which encourages industry; 3. by instituting a policy o f birth control. Finally we m ust examine the question o f international aid. It is undeniable th at such aid is desirable, and if we consider the probable course of future population growth in the under-developed countries, it even seems indispensable. But it is the ways o f distributing it and the am ounts to be provided th at raise the problem. It has to be recalled, given the gap between levels o f income in the Third World and the advanced countries, th at the available aid would seem relatively substantial when com pared to the national product o f the Third World, whereas it would not form more than a small share of the national product of the advanced world. For example, the G.N.P. of the non-communist, under-developed world at current prices, am ounted to 390 billion dollars in 1970 com pared to 2090 billion dollars in the western developed countries, i.e. a ratio o f 5.4:1. The target laid down by the United Nations international financial aid am ounting to 1 per cent o f the national product o f the advanced countries though far from having been achieved, could be, and should be, greatly increased. A reasonable minimum level of international aid should be agreed which would allow the lessdeveloped countries to absorb four-fifths of their own population growth and thus raise these countries to the same level, in demographic terms, as the western powers during their take o f f . In these hypothetical conditions, and with a capital o u tp u t ratio of 4.5, aid should have am ounted to some 37 billion dollars in 1970, i.e. more than twice what was provided, taking the wider definition of aid. With a m ore realistic definition o f aid, the sum would be ten times higher than w hat is provided. This 37 billion dollars would form 1.8 per cent o f the G.D.P. at m arket prices o f the advanced countries, or, 2.0 per cent o f their national income. In future years, given the probable acceleration o f population and econom ic growth, aid (at constant prices) should rise by an annual rate of 4 per cent to a figure of about 55 billion dollars in 1980 (at 1970 values). Furtherm ore, and this is very im portant, the way the aid is distributed m ust also be modified. The aid should be m ultilateral and distributed by an international agency. It should comprise mainly grants and should have no strings attached. If this should not be feasible, then specific allocations of aid m ust be justified solely by the relative effort a beneficiary country is making tow ards economic development. It would, for example, be acceptable fo r financial aid
208
to be used for the purchase o f consum ption goods if this would help to stim ulate local investment. We should be aware th a t th e m ultilateral character of an international agency for distributing financial or technological aid to less-developed countries might not be able to prevent the money-lending country from influencing directly or indirectly th e distribution o f aid. Although this state of affairs is very regrettable, it is questionable if in practice it is possible to eliminate it entirely, given the current attitudes of both eastern and western powers. It is also essential th a t any aid provided by the advanced countries should n o t be recovered indirectly (either wholly or in part) as a result o f a deterioration in the term s o f trade of the Third World such as occurred after 1954. A problem of this kind could easily be solved if th e advanced countries so wished, and the creation of a special section o f U.N.C.T.A.D. or a re-orientation of the GATT, would be useful for this purpose. Obviously recom m endations in the field o f international aid such as these have little or no chance of actually being adopted. But the western world, which forms m ore than two-thirds of the advanced world, ought to realize th a t even such an effort is a m odest one am ounting as it does to barely one-third of w hat is spent on defence and is justified not only on hum anitarian and political grounds, but also on grounds of justice. For the colonial and trade policies of the advanced countries in form er times have, even if unw ittingly, upset the equilibrium of the Third World and had m ostly unfavourable effects. Besides, and this is perhaps the m ost unfortunate feature o f th e situation, so far as we can draw up a balance sheet, it seems th at the profit accruing to the West from its colonial and trade policies in the past was much smaller than the harm unw ittingly inflicted on the less-advanced countries. The obligation to provide aid with no strings attached is, however, incum bent not only on the western powers. Rapid economic growth in the Soviet Union and the other east European countries has resulted in their achieving a level o f development high enough to allow them to contribute a substantial share o f the total volume of financial aid needed by the less developed countries. The G.D.P. of the developed com m unist countries, expressed in western accounting term s, could be estim ated in 1970 at about 600 billion current dollars, i.e. 1700 dollars per capita, com pared to 2090 billion dollars and 2900 dollars per capita respectively for the western developed countries. Similarly there should be no objection to the com m unist under-developed countries being included am ongst the beneficiaries
General Conclusions
209
of international financial aid. Total aid required would thereby be increased by about 70 per cent. But if foreign aid is necessary in m ost cases, it is n o t sufficient itself to perm it the under-developed countries to get their economies moving. Efforts must be m ade w ithin a country to allow foreign aid to be deployed in the m ost efficient way. But since foreign aid can never provide more than a small part o f total investm ent required, it is even more im portant to mobilize all the local resources. On the assumption o f a growth rate o f 6 per cent per annum and a capital-output ratio of 4.5 (which is what the U.N. experts who are drawing up targets for the next decade of developm ent seem to expect), the total am ount of investm ent in the under-developed countries as a whole, including the com m unist ones, would be about 150 billion dollars in 1970. In my opinion the target of 6 per cent, although m ost desirable, is probably not likely to be achieved in the next decade. Besides this figure of 150 billion dollars, the new target (1 per cent of G.D.P. which would form 27 billion dollars of foreign aid from the entire advanced world) adopted a t New Delhi in 1968 by U.N.C.T.A.D. II, but neither achieved nor even accepted to be binding, seems to say the least very m odest. The same can be said for the target which I have proposed and which represents, for the whole o f the under-developed world, foreign aid o f 46 billion dollars. Quantitative juxtapositions o f this sort dem onstrate how im portant local efforts are likely to be if a growth rate o f 6 per cent is to be achieved. They also show th a t serious consideration o f the recom m endations made here is more than justified. To repeat, it goes w ithout saying th at these recom m endations are no more than generalizations, necessarily imprecise and based on average conditions. M athematically speaking, of course, an average need not apply to any o f the units form ing the statistical universe under consideration, especially if the statistical spread is as wide as is the case here. Thus it is obvious that restraints on population growth, which have been made a first priority, are less urgent, even though far from superfluous, for m ost o f the countries in Latin America and Africa because they com m and greater reserves of land. In the case of Latin America, o f course, this is also combined w ith a mo re advanced economic development than exists in Asia. On the other hand agrarian reform is m ore necessary and of greater urgency in the Latin-American countries than it is elsewhere. Similarly regional integration is not so imperative for m ost Asian countries because their dom estic m arkets are vast enough to provide outlets, even with
210
incomes at such a low level, for a wide spectrum of manufactures. But integration is desirable in Latin America, even m ore so in Central America, and in some South-East Asian countries, while in m ost African and Middle Eastern countries it is literally vital. As we have just said, it is unrealistic to expect all the recom m endations p u t forw ard in these pages to be considered with the same degree o f priority, or to be taken as relevant to each one o f the 170 different nations comprising the Third World. Each individual case deserves a detailed study (and it is interesting th a t m any o f these studies have already been made), and each needs its own remedy. But to the extent th a t m uch of w hat has been discussed is common to the entire Third World we have tried to show th at the recom m endations are m ostly the indispensable preliminaries upon which national schemes for economic developm ent should be based b u t which are all too often forgotten. D. Postscript (1974) Reading the proofs in Septem ber 1974 it is hard to resist the desire to add a brief postscript com m enting on the probable effects o f the sharp increase in oil prices in 1973. Should the present high price of oil be m aintained for a long period the main effects would probably be as follows: For the oil-exporting countries, representing some 3 per cent o f th e total population o f the Third World, there would be an opportunity to use the additional financial resources for investm ent purposes. But the very high price of oil will in itself keep down som ewhat the income from oil because the increase in the rate of consum ption will be slowed down and the process o f substitution will be speeded up. For the non oil-exporting, less-developed countries, representing over 95 per cent of the Third World, the new situation raises enorm ous difficulties. Briefly, these arise from the greatly increased cost o f im porting essential energy, fertilizers and food. It is also probable th a t the financial aid (in real term s) that they will receive from developed countries will decline and th a t the decline will not be wholly com pensated for by aid coming from the less-developed countries which are getting rich as a result o f their oil exports.
Notes
Introduction 1. 2. P. Bairoch, Revolution industrielle et sous-developpement , Paris, 1963 (4th ed. M outon, The Hague and Paris, 1974). These figures have already been the subject o f an essay, d ev o lu tio n de la productivity agricole dans les pays econom iquem ent sous-developpes de 1909 a 1964, in Developpement et Civilisations, 25 March 1966. It should be noted, however, th a t in some aggregates calculated by the United Nations T urkey is counted as p art of Europe and thus as a developed country. This point will be m entioned later.
3.
1. Population 1. In particular Ten Great Years (statistical abstract, Pekin, I960); G. Etienne, Quelques donnees recentes sur la population en Chine, Population (July Sept. 1962); W. F. Wertheim, La Chine est-elle sous-peuplee? Production agricole et maind oeuvre rurale, Population (May June 1965); discussion of demographic problem s in China News Analysis (24 and 31 July 1964). The analysis in this paragraph is based mainly on an excellent synthesis by R. M. Field, A note on the population of Com munist China, The China Quarterly , 38, (1969) pp. 15863; and on W. K latt, A review of Chinas econom y in 1970, The China Quarterly , 43 (1970) pp. 10020. J. S. Aird, Population grow th and distribution in mainland China, in An Economic Profile o f Mainland China, Vol. 2, Part III (Washington, Jo in t Economic Com m ittee of U.S. Congress, 1967), pp. 343 400. Aird, in projections up to 1985, uses four hypotheses. For 1970 (up to Jan. 1) these hypotheses postulate assuming th a t the census o f 1953 is correct a minimum of 788 and a maximum o f 814 million. Aird also admits the possibility th a t th e censuses under estim ate th e to tal from 5 to 15 per cent (p. 363). H. Gille, Acceleration dem ographique en Extrem e O rient, Population (O ct Nov. 1961). Differences o f the same order have been noted for other Far Eastern countries:
211
2.
3.
4.
212
The Economic Development o f the Third World since 1900 Pakistan (1951 61) 2.2% real growth instead o f 1.7% forecast Philippines (1948 60) 3.2% real growth instead o f 2.5% forecast Thailand (1947 60) 3.0% real growth instead o f 2.0% forecast
5.
6. 7.
8.
9.
10.
11.
Bilan et perspectives de leconomie indienne au seuil de Fannee 1965 (French commercial counsellor in India), Problemes Economiques, 898 (March 1965). New York Times , 20 May 1970. Figures extracted from the article by Sajal Basu and Sankar Ray, Im pact of intra-uterine contraceptive devices, Economic and Political Weekly (Bombay, 8 June 1968). L. Henry and R. Pressat, Perspectives de population dans les pays sous-developpes, in Tiers-Mondey sous-developpement et developpement (Paris, 1956), pp. 189213. Thus, according to projections made by F.A.O., we can expect a surplus in the western world in 1975 o f the order o f 27 to 33 million tons for all cereals (depending on the hypothesis adopted); a surplus resulting mainly from th a t forecast for N orth America (betw een 54 and 57 million tons), F.A.O. Agricultural Commodities Projections for 1975 and 1985 , Vol. I (Rome, Oct. 1966), We should note, however, th a t in the longer term (4050 years) the margin available to cover the growing deficits of the undeveloped countries by western agriculture is problem atical, as will be seen in the chapter on agriculture. (F or a 1980 forecast see note 35, Chap. 2.) This is the date in history when the relative share in the world total of th e population of these countries was the smallest; it should not be forgotten th a t from the beginning o f the industrial revolution up till about 1930 the population o f the developed countries grew more quickly in general than did that of other regions. Thus it can be estim ated th a t the present-day less-developed countries accounted for 75 per cent o f the world population in 1750 (73 per cent in 1850 and 65.5 per cent in 1900). It would, moreover, be im portant to know to w hat extent the estimates o f the optim um densities o f population take suffici ent account of to d ay s low costs o f transport and its greater speed. This, however, touches a problem outside the scope of this study.
Notes to 2, Agriculture 2. Agriculture 1. 2.
213
3.
4.
5. 6.
7.
8.
9.
These are o f course only approxim ate percentages: the differ ences between countries are considerable. See the au th o rs own studies, Revolution industrielle et sous-developpement; and Agriculture and the Jndustrial Revolution (London: Collins, The F ontana Economic History of Europe, 1969). The productivity index has been constructed in a very similar way to the one for m y study o f agricultural productivity in eleven countries during the nineteenth century. See my article, Niveaux de developpement econom ique de 1910 a 1910, in Annales (Economies , Societes , Civilisations), 6, 1965. In term s of export values these commodities represent about 80 per cent of total exports o f tropical com m odities and about 70 per cent of total exports o f agricultural commodities, fisheries and forest products of the Third World. Average export price for each com m odity for all the principal producing countries. The index of food production is based on the following commodities: cereals, starchy roots, sugar, pulses, edible oil crops, nuts, fruit, vegetables, wine, cocoa, livestock and livestock products. To obtain an index o f total agricultural production the F.A.O. adds the following commodities: fibres, rubber, tea, coffee, oil seeds (industrial oils) and tobacco. It should be noted th a t since the end o f 1967 the F.A.O. has considerably modified its ways o f calculating these figures. The indices of agricultural production are henceforth calculated for the civil year and for totals of the larger regions. There was a revision in 1972 and again in 1973. For these F.A.O. calculations the Middle East theoretically includes the following African countries: Libya, Sudan, Egypt, and these Asian countries: Aden, Afghanistan, Bahrein, Cyprus, Iraq, Iran, Jordan, Kuwait, Lebanon, Muscat and Oman, Katar, Saudi Arabia, Syria, Oman, Turkey and Yemen. Africa includes all the countries in the continent with the exception of South Africa and those included in the Middle Eastern group. The same applies to the Far East, except th at China and Japan are also excluded. Latin America includes all America except the United States, Canada, Bermuda, Greenland and St Pierre and Miquelon. Nevertheless it m ust be pointed o u t th at m ost of the indices
A i X a
214
The Economic Development o f the Third World since 1900 could possibly be biased in the same direction, in which case the margins o f error would not be reduced. A unilateral bias is quite probable and would have the effect o f exaggerating growth. The exaggeration would result partly from the fact th at commercial production increases m ore rapidly than total production (m ostly, but n o t entirely, through a rise in the level of commercialization) and partly through the desire o f the authorities not to provide excessively 'negative figures. Lastly, it is m ore than probable th a t the levels o f demographic growth are under-estimated. As we shall see in the chapter on m anufacturing industry, the annual levels of growth in this sector were nearly 3 per cent. These rates have been calculated, after taking account of various factors, in my study, Revolution industrielle et sousdeveloppement, p. 155. It is also possible th at certain series have been revised in the interval. I have endeavoured to avoid this pitfall by using, as far as possible, figures which are n o t provisional. The fact th at the United Arab Republic is referred to under its form er name has no political significance. The choice is justified on purely statistical grounds because the material concerns solely the Egyptian part o f the U.A.R., as is also the case for m ost o f the statistics. The title U.A.R. would have risked introducing an error into the geographical range o f the statistics, all the m ore since the union has changed in the course of time. See my article Niveaux de developpem ent econom ique de 1810 a 1910. English readers will find a very short resume of this article in Section lie of m y study, Agriculture and the Industrial Revolution. By direct or initial calories is m eant those which have n o t undergone a transform ation in th e course o f livestock rearing. Livestock produce from the calories contained in vegetable feeding stuffs in the form o f m ilk or m eat are called indirect or derived calories. I shall come back later to this problem. The calorie content o f m eat differs greatly according to the percentage o f fat which it contains, since muscle or pure m eat contains about 1600 calories per kilo, while p ure fat contains nearly 9000. F.A.O. has adopted a single coefficient o f 7 for all m eat in all countries (accepted also by M. Cepede and M. Lengele, Economies alimentaires du globe, Paris, 1953). I have preferred
10. 11.
12.
13.
14.
15.
16.
17.
A i X a
Notes to 2 , Agriculture
215
18.
19.
20.
21. 22.
23.
to use different coefficients for each kind. For the contem porary period there is sufficient data to allow these coefficients to be determ ined in the form o f weight gain per unit o f feeding stuff. From these contem porary figures which I worked out for th e earlier study in this field I have calculated coeffici ents, theoretically valid in the m iddle o f the nineteenth century for partially-developed European countries. (The choice was based partly on th e advice of specialists whom I consulted and whom I m ust thank for th eir kindness, and partly on fragm entary m aterial about the age o f slaughter. The figures thus obtained have been m odified slightly on the basis of available statistics concerning the percentage of slaughtered animals.) I have n o t m ade an estim ate for detecting the consum ption of cereals by animals fo r th e following three reasons: a) Existing figures on th e subject are fragm entary and rather unreliable, b) Animal consum ption is extrem ely small in m ost of the under-developed countries, with the exception of those in Latin America, c) I have postulated a certain com pensation between animal consum ption of cereals and the fact th a t we have not taken into consideration certain agricultural com m od ities, namely, poultry m eat, oil seeds, dairy products and cane sugar. I chose th e countries for which I m ade this correction on the basis o f the production of different crops and on the relative value of exports of agricultural products (as m uch in relation to total exports as by the value o f total agricultural production). I have to thank the librarian o f I.N.E.A.C. (In stitu t National pour lEtude Agronomique du Congo Beige) for his kind cooperation. Publication o f the I.N.E.A.C., un-num bered (Brussels, 1958). O ut o f some fifty studies and articles dealing with different commodities which I examined only the six following works contained any figures relating to labour norms: T. Eden, Tea (London, 1958) esp. p. 61; J. L. Collin, The Pineapple: Botany , Cultivation and Utilization (London, 1961) a few facts on p. 141; Ch. Surre and R. Ziller, Le palmier a huile (Paris, 1963) esp. p. I l l ; J. Champion, Le bananier (Paris, 1963) esp. p. 203; L. Burle, Le cacaoyer , Vol. II (Paris, 1962) esp. pp. 50917; F. Martin, La canne a sucre (Paris, 1935) esp. p. 183. This rate is based on m aterial relating to several tropical crops (especially in the works o f Eden and Burle m entioned above).
216 The Economic Development o f the Third World since 1900 24. Even confined to the group of countries included here, calculating gross production alone entailed research, transcrip tion and handling o f about 20,000 statistical com putations. I have to thank Monsieur H. Gelders for helping me with this p art of the study. 25. Especially for China and Cuba and for manioc and soya beans. The Chinese material was o f course the m ost difficult to assemble. For the official* figures the main sources were Govern m ent statistical publications (Dix grandes annees , Pekin, 1960) and the press com m uniques o f the New China Agency; the w estern material has been drawn from the analyses in the Far Eastern Economic Review (Hong Kong), including the annual supplements; E. F. Jones, Emerging p attern o f Chinas econo mic revolution in An Economic Profile o f Mainland China, Vol. I (Washington, 1967); R. M. Field, How m uch grain does com m unist China produce? in The China Quarterly , 33 (Jan March 1968); S. Swamy and S. J. Burki, Foodgrain ou tp u t in Peoples Republic o f China, 19581965 The China Quarterly , 41 (Jan March 1970); W. K latt, A review o f Chinas economy in 1970 The China Quarterly , 43 (July Sept. 1970); and T. G. Rawski, R ecent trends in th e Chinese econom y The China Quarterly , 53 (Jan March 1973). 26. O ther factors such as stature and climate help to differentiate this minimum according to region. Here, by way o f example, is the am ount of calories o f average food available per head for some under-developed countries for which the F.A.O. has calculated these figures (general averages 1963/65 1966/681969): Latin America Argentina 3,150 Bolivia 1,760 2,540 Brazil 2,520 Chile Equador 1,850 Mexico 2,610 2,180 Peru Asia Burma Sri Lanka India Indonesia Pakistan Philippines Thailand Middle East/Africa 1,890 Algeria 2,840 Egypt Morocco 2,130 2,160 Nigeria 2,450 Syria Tanzania 2,140 2,760 Turkey
2,010 2,160 1,940 1,750 2,290 2,000 2,020
Year Book o f Production , 1970 (F.A.O., Rome, 1971), pp. 4447. 27. In my study Niveaux de developpement 6conomique de 1810
Notes to 2, Agriculture
217
28.
29.
30.
31.
32.
33.
34.
a 1910 I have assumed 45 per cent of the population to be active during th e nineteenth century in countries which are now developed. But given a very different demographic structure (especially the much greater proportion o f children under 10) in the under-developed countries, I have n o t kept the same rate. The rate accepted here is based on figures from most of th e under-developed countries for which statistics on the subject are available (see Chapter 9). If a weighted average is made for the p art of the employed population o f under-developed countries n o t occupied in agriculture, this works o u t at about 20 per cent for Asia and Africa and more than 35 per cent for Latin America. The determ ining o f this rate is rather hazardous. It is obviously very difficult to p u t an exact figure on the extent o f these losses. It is of course a national average, and inequalities which are as m uch regional as assignable to social classes may lead to insufficiencies for a large part o f the population. In addition it is possible and even probable th a t such a food availability implies qualitative nutritional scarcities due to protein or vitamin deficiencies. In addition to stockraising we should also take account o f the possibilities o f stockpiling, b u t as in the preceding case this factor plays a negligible role in under-developed countries. Of course it is true th a t at th e m om ent foreign trade allows m ost o f the under-developed countries to keep th e famine risk in check, particularly by means o f surpluses o f cereals from agriculture in the developed west. In m y study Revolution industrielle et sous-developpement I have already expressed doubts on this subject after examining the percentage o f durable goods made locally (pp. 1723). Indias famine in 1966 illustrates tragically th e reality of this situation. It should be pointed o u t th at the first edition o f this study was com pleted in the course o f 1965 just when it seemed as if the food situation of th at country was improving, the harvest of 1965 being b etter than the rather poor one o f 1964. Thus if cereal stocks are large in relation to the harvests o f the countries w ith surplus production (U.S. 50 per cent o f the annual harvest; Canada 75 per cent) the quantity is much less when com pared w ith the volume o f world production. Wheat stocks (at 31 July o f each year) fluctuate around 55 million tons from 1960 to 1970, th a t is about 20 per cent o f world
ial com
218
The Economic Development o f the Third World since 1900 production. Stocks of all cereals averaged 130 million tons from 1960 to 1970, o r 12 per cent of world o u tp u t (figures for stocks are those o f the principal exporters and form the essential part of stocks). Sources: Foreign Agriculture Circular, U.S. Dept, o f Agriculture and F.A.O. Given the yield fluctu ation, it is quite conceivable th a t world cereal o u tp u t could be 12 per cent less than the average for tw o successive years. The level of cereals consum ed by animals in the developed countries is such th a t it is theoretically possible to make a large quantity of cereals available in case o f urgent need. To illustrate the situation here is an estim ate o f production and consum ption o f all cereals produced in the western countries (figures rounded up to millions o f tons): Situation at 1970 380 360 30 80 250 25 Forecast fo r 1980 510 455 40 90 315 65
35.
Production Consumption Seed Food Animals Surplus
Sources: F.A.O. especially Com modity Review 197071 (Rome, 1971) and Agricultural Com modity Projection 1970 8 0 (Rom e, 1971). Therefore, in an hypothetical situ ation where adverse conditions in the under-developed coun tries create an additional demand a t a m om ent when supply in the west is reduced, it would always be possible, in theory, to cover the deficit by sacrificing some livestock; since in 1970 animals in western countries consumed ten times as m uch as the surplus exported to under-developed countries (and, in part, to the eastern countries). 36. Correlations betw een the tw o series provide the following results: for all tw enty-four countries under consideration here 0.92; for African countries 0.91; for Latin Amer ic a 0.97; for Asian countries (including China at the 5.8 level) 0.25. 37. Agricultural land is taken to m ean arable land and land under perm anent crops plus m eadows and perm anent pasture. 38. It should be noted th a t the margin of error in the productivity indices of countries where plantations are im portant is probab ly greater than for other territories because the working
ial com
Notes to 2, Agriculture
219
39.
40.
41. 42.
43.
44. 45.
population em ployed in plantations has to be estim ated. This additional margin o f error is not, however, enough to explain the gaps which are so evident. Thus in France the final o u tp u t per male agricultural w orker rises from 1,894 Frs. (Frs. 1 9 0 5 -1 4 ) in 1 8 9 5 -1 9 0 4 to 4,788 Frs. in 195558, an increase of 151 per cent in 57 years (an annual rate o f 1.7 per cent); from A. Toutain, Le produit de Vagriculture franâise de 1900 a 195 8 , Vol. II (I.S.E.A. pam phlet no. 115, Paris, 1961). In the U.S. the index o f gross agricultural productivity per man hour rises from 45 in 1 9 1 0 -1 4 (base 1 0 0 = 1 9 4 7 -1 9 ) to 149 in 1 9 5 3 -5 7 , an increase of 231 per cent in 42 years (an annual rate o f 2.9 per cent); from Historical Statistics o f the United States (Wash ington, 1960), p. 602. During the period 1948/521968/72 the productivity of th e agriculture of all western developed countries increased by some 5.5 per cent per annum. The growth o f productivity in Mexico is generally attributed to agrarian reform. As I said in the Introduction, I do not seek here to provide explanations o f differences, b u t the question may well be asked w hether a p art o f the progress could n o t be attributed to the better coverage o f statistics made possible by the reforms. This reflexion is n o t m eant to cast doubt on the positive value of agrarian reform in Latin American countries, w ithout which it is hardly likely there would have been any noticeable im provem ent in the economic condition of these countries. Based on government figures the annual rate of growth would be 3.03.5 per cent. It should be noted, however, th a t I have not been able to find a significant correlation between the rates o f growth in the em ployed agricultural population and productivity. Any degree o f bias in the comparisons introduced by these differences is not large (probably less than 10 per cent) and tends to under-estimate the nineteenth-century figures in relation to the under-developed countries. See my study, Revolution industrielle et sous-developpement, pp. 7 5 - 6 , 2 2 3 - 4 and 2 9 2 -3 . When comparing the estim ates o f income per head in France, Great Britain and th e U.S. at the beginning o f the nineteenth century with those o f the under-developed countries (allowing for changes in prices and purchasing power) I had already come to the same conclusion (see my article, Le my the de la
220
The Economic Development o f the Third World since 1900 croissance econom ique rapide au XlXeme siecle in Revue de Tlnstitut de Sociologie , 2, (1962), pp. 31920). Note th at th e small decrease in land available occurring a t the start of developm ent was only tem porary and was in fact changed into an increase in cultivated land because of the decline in the practice of fallowing. As we have seen earlier, the land available today is greater due to the fall in the population employed in agriculture. Estimate based on the ratio o f the calorie consum ption from milk, eggs and vegetables compared to total consum ption in these countries; compared to the same ratio for the under developed countries. However, we should not take available land for granted, for in certain countries it is certainly possible to augm ent it. But it pre-supposes investm ent on a massive scale and there is always the risk o f damaging the ecological equilibrium for, as Prof. Harroy points o u t in La conservation des ressources naturelles du Tiers-Monde, in Tiers-Monde, 20 (O ct Dec. 1964) the Third World countries, by reason of their geographical position, suffer from the existence of ecological conditions which are much more easily damaged than those in Europe, N orth America or the U.S.S.R. The western countries are o f course those referred to, for th at is where the surpluses are. In addition the developed countries beyond the Iron Curtain seem to have entered a phase in which it is unlikely th at they will have much o f a surplus in the near future. R. D um ont, Developpement agricole africain (Paris, 1965), p. 211. For the problems concerning the famine menacing the Third World, another of his works should be consulted (D um ont and Rosier, Nous allons a la famine , Paris, 1966. English trans. The Hungry Future , New York, 1969). Mostly based on a publication of the F.A.O. The Green Revolution (lecturers guide); and The State o f Food and Agriculture (both Rome, 1972). The situation is different when it is a question of yields changing due to climatic conditions: in this case all producers are on average affected similarly and as a result there is some compensation between changes in price and the am ount produced.
46.
47.
48.
49.
50.
51.
52.
3. Extractive Industry 1. For the eighteenth century the figures are those collected for my study, Revolution industrielle et sous-developpement.
Material com direitos autor,
Notes to 4, Manufacturing Industry
221
2.
3.
4.
5.
Contem porary figures have been provided by the Services Tech niques du G roupem ent des Hauts Fourneaux de Belgique and I should like to thank them for their kindness. Annuaire Statistique de la Federation professionnelle des pro ducteurs et distrib u tees d electricite de Belgique , several years. As I shall show later, it is clear th at this widening gap is also partly due to the im port by developed countries o f products o f extractive industry originating in the under-developed coun tries. But as a whole the gap remains due to the factors set o u t in this section. Thus, for all non-communist countries (devel oped and under-developed) and for the period 1958/60 1968/70, the annual rate o f growth of m anufacturing industry was 6.0 per cent, while it was 4.7 per cent for the whole o f extractive industry, and 2.4 per cent for extractive industry excluding oil and natural gas. Most prices have been taken from Metal Bulletin and Engineer ing and Mining Journal. I have to thank here the Societe Beige des Minerais and, in particular, Monsieur Lamberts for having provided me with inform ation in this field. Ivar Hogbom, Developpement de la production mondiale de matieres premieres , appendix to the R eport o f the Commission for the study of the problem of raw materials, League of Nations, 1934; cited by Industrialisation et commerce exterieur (League of Nations, Geneva, 1945), p. 55.
4. M anufacturing Industry 1. It is clear th at there are considerable differences between one country and another, as well as betw een one period and another, which arise particularly from variations in energy consum ption for private use. The average used here (5 per cent) results from the calculations made for a large num ber of under-developed countries for which the tw o series (i.e. index numbers of m anufacturing production and energy consump tion) exist. Because the index num bers are rounded off to whole numbers, the rates of annual variations for short periods contain an additional margin o f error; the figure after the decimal point in my figures, is consequently no more than an indication. C.f. also my article Le m ythe de la croissance econom ique rapide au XIXe siecle Revue de VInstitut de Sociologie, 2, 1962. For the take o f f periods o f m ost o f the countries m entioned
2.
3. 4.
222
The Economic Development o f the Third World since 1900 here, readers are referred to my study Niveaux de developpe m ent economique de 1810 a 1910. It might be claimed th a t Nazi Germany had a centralized economic regime, running parallel to a capitalist system, but this did not occur until after 1935. See particularly: J. Ahmad, Im port substitution and structural change in Indian m anufacturing industry, 19501960, in j . o f Development Studies, IV (April 1968); J. Ahm ad, Measuring the influence of foreign trade and dom estic demand on econom ic grow th, Institut International d Economie Quan titative , Publication 4 (Montreal, 1970); H. B. Chenery, S. Shishido and T. Watanabe, The pattern of Japanese growth, 19141954, Econometrica , 30, No. 1 (1962); S. Lewis and R. Soligo, Grow th and structural change in Pakistan manu facturing industry, 19541964, Pakistan Development Re view , 5, No. 1 (1965); R. Soligo and J. Stern, Tariff p rotect ion, im port substitution and investm ent efficiency, Pakistan Development Review , 5, No. 2 (1965); United Nations, A Study o f Industrial Growth (New York, 1963). The official figures up to 1958 come from the publication Dix g r a n d e s a n n e e s (Pekin, 1960). A fter 1958 figures are based on a system atic search for inform ation about China, both in official docum ents and in th e following reviews: Far Eastern Economic Review; China News Analysis; The China Quarterly , and Problemes chinois in Notes et etudes documentaires (La Documentation franqaise). For Western estimates, apart from miscellaneous articles and reviews in Far Eastern Economic Review , the following have been consulted: Cho-Ming Lis article, Chinas industrial developm ent 19581963, in In dustrial Development in Communist China, ed. Cho-Ming Li (New York, 1964), which gives the various estimates published up to 1963; and Kang Chao, The Rate and Pattern o f Industrial Growth in Communist China, (University o f Michigan Press, 1965). M. Field, Industrial production in Com munist China: 19571968, The China Quarterly , 42 (April June 1970). For the period 1949 65 see the same author, Chinese Communist industrial production, in An Economic Profile o f Mainland China, Vol. I, Part 2, pp. 269 95. Le Monde, 2 March 1971. Le Monde, 23 January 1972. For th e years 195052, according to the available western
5.
6.
7.
8.
9. 10. 11.
Notes to 4, Manufacturing Industry
223
estim ate (Kang-Chao, op. cit. p. 88) the annual rate o f growth would be 26 per cent (as against the 34 per cent given in the official index). For the period 195257 the four main western estim ates provide very similar growth rates (as annual rates, 9.6, 14.2, 13.6 and 13.3 per cent respectively). Given that growth was faster from 1950 to 1952 than from 1952 to 1957, I have retained as the average o f western estim ates a rate for the period 195057 o f 14.0 per cent. For 1958 and 1959, the rates for the four estim ates are: 1958, 32.5; 19.6; 19.5 and 30.3 per cent; for 1959, 31.6; 27.0, 32.3 and 31.5 per cent. I have kept th e average o f these estim ates (25.0 and 31.0 per cent). 12. M. Field, Industrial production in Communist China: 19571968, The China Quarterly , 42 (April June, 1970). 13. Especially W. Klatt, A review o f Chinas economy in 1970, The China Quarterly , 43 (July 1970), pp. 100 20 and analyses in Far Eastern Economic Review. 14. Where Pakistan is concerned this is partly due to the transfer of factories, especially textile factories, following the partitition of India at independence when M ahommedan industrialists emigrated. In Taiwan the situation was similar, but was also combined with considerable financial aid and a rise in domestic demand resulting from even larger m ilitary aid. In this country, and in South Korea, the opening o f western m arkets, particu larly in the United States, re-inforced as it was by the Vietnam War, played an essential role. 15. The cases of the United States and Morocco are an illustration of this situation. In th e United States, m anufacturing industry forms only 29 per cent of the Gross Domestic Product while working population so occupied forms 27 per cent of th e total, th at is a gap o f 7 per cent betw een the two rates. Morocco, on the other hand, in which m anufacturing industry employs only 9 per cent o f the working population, this sector produces 14 per cent o f th e Gross Domestic Product, th at is a gap o f 55 per cent between the tw o rates. 16. In order to have a very approxim ate yardstick o f the productivity o f m anufacturing, I have calculated the value of o u tp u t per worker in dollars for some o f th e countries examined here for which the m aterial available enables such a com putation to be made. (The figure is the value o f Gross Dom estic Product generated by m anufacturing related to workers in the sector and expressed in U.S. dollars.) The
224
The Economic Development o f the Third World since 1900 following are th e figures for 1961 63 (annual averages): $ US 550 2030 1040 1080 1100 $ 'US 190 210 560 740 940
Morocco Argentina Chile Mexico Peru 17.
India Pakistan Philippines Thailand Turkey
Comparisons are made m ore difficult because o f the differing changes recorded betw een 1930 and 1950. While India was experiencing a period o f peace in which industrial expansion was not hindered, China, with the Sino-Japanese War followed by civil war, w ent through a troubled period which resulted in a fall in industrial production. This is why, in an attem p t to reduce any statistical bias, I have chosen to com pare the highest level attained in both countries before 1950. This m ethod nonetheless favours India since the maximum levels achieved by this country in general occurred in 1949, whereas the maximum Chinese levels occurred before 1940. Here, therefore, for the four indices given in Table 25, are the m axim um levels achieved before 1950 by these two countries: Energy 120 100 Steel 3.0 3.9 Cotton 1.9 2.3 Transport 140 118
China India
18. As already noted, a m ore detailed analysis o f this subject will be found in my book, Revolution industrielle e t sousdeveloppement , especially in C hapter 5, Agriculture, facteur determ inant d amorce du developpem ent, (pp. 73 84) and in Chapter 7, Le schema explicatif du processus d am orce du developpement econom ique (pp. 98113). English-speaking readers should refer to my study Agriculture and the Industrial Revolution. 19. P. Bairoch, Niveaux de developpem ent econom ique de 1810 a 1910*, except Austria which was calculated for my study Agriculture and the Industrial Revolution. 20. The kind o f agricultural econom y is m eant in which m ost farmers sell no m ore than a small p art o f their output. In this situation a fall in yields means th at a farm er has no surplus
Notes to 5, Foreign Trade
225
o u tp u t for sale, and cannot therefore com pensate for the fall in available income by selling produce at a higher price. 5. Foreign Trade 1. In recent literature, the reports o f the two U.N.C.T.A.D. conferences Geneva 1964 and New Delhi 1968 are espe cially notew orthy: United Nations Conference on Trade and Development , Proceedings , 8 vols. (1964) and Proceedings , 5 vols. (1968). Other references are: B. A. Balassa, Trade Prospects fo r Developing Countries (Homewood, 1964); M. Z. Cutajar and A. Franks, The Less-Developed Countries in World Trade (London, 1967); G. de La Charriere, Commerce ex terieur et sous-developpement (Paris, 1964); H. G. Johnson, International Trade and Economic Growth (London, 1958); C. P. Kindleberger, Foreign Trade and the National Economy (New Haven, 1962); A. I. Macbean, Export Instability and Economic Development (London, 1966); A. Maizels, Exports and Economic Growth o f Developing Countries (Cambridge, 1968); J. Pinkus, Trade, Aid and Development (New York, 1967); Revue de Tlnstitut de Sociologie special number, Division internationale du travail et organisation du commerce m ondial (Brussels, 1965); M. Saint-Marc, Commerce exterieur de developpement (Paris, 1968); J. Wciller, Veconom ie inter nationale depuis 1950 (Paris, 1965). In addition the GATT (Geneva) has published every year since 1952 a report called International Trade . . ., in which one section is devoted to trade in the unindustrialized regions. See also Trends in International Trade , published by the GATT (Geneva, 1958). Readers are reminded th at prices f.o.b. (free on board) mean that the prices do not include freight costs, insurance and the other charges connected with transport. On the other hand, prices c.i.f. (cost, insurance and freight) do include these costs. Exports are usually expressed in f.o.b. prices while im ports are expressed c.i.f. Note, however, th a t there is a strong possibility th a t the available index num bers of export prices are significantly distorted and th at the increases in the volume o f world exports have in fact been slightly less im portant. In applying a correction based on the average o f 1960 70, the gap between export values and world im ports would be according to my own calculations 5.4 percent.
2.
3.
4.
226 5.
The Economic Development o f the Third World since 1900 According to my estimates, by including the m erchant fleets o f Panama and Liberia amongst those o f the developed countries, a figure is obtained for the m erchant tonnage o f the under developed countries as at 1 January 1965 o f slightly less than 6 per cent o f the world total, while the trade o f these countries forms about 20 per cent o f the world total. We can therefore assume that approxim ately 70 per cent, at least, o f the international maritime trade o f the under-developed countries is carried in ships belonging to the developed countries. I say at least since the international trade o f the under-developed countries comprises a much higher proportion o f sea transport than th at o f the international trade of the developed countries. A part from th e Asian Middle Eastern countries, the Middle East comprises Egypt, Ethiopia, Libya, Somalia and Sudan. These countries are consequently not included in Africa. See my book, Revolution industrielle et sous-developpement , pp. 3 3 4 -5 . International Trade 1968 (GATT, Geneva, 1969). Indeed the situation is not much better in Latin America. Thus in 1970 exports intended for Latin American countries form ed only 11 per cent of total exports from these countries, in spite of the comparative success o f tw o custom s unions. The following, expressed in per capita terms, is the situation as far as im ports coming from under-developed countries are concerned (in U.S. dollars): EFTA Common Market United States Japan Under-developed countries U.S.S.R. China 1965 56 50 34 29 5 5 0.7 1970 69 71 49 84 6 5 0.4
6.
7. 8. 9.
10.
11.
In fact I have been engaged for several years on the more general study o f the relationship between international trade and econom ic development. My research was greatly helped by a generous grant from the Canadian Council of Arts in 1970. The parts which I have been able to publish cover only the period o f nineteenth-century free trade (Commerce exterieur et developpement economique: quelques enseignements de rexperience libre-echangiste en France au XIXe siecle, in
Notes to 6, The Terms o f Trade
227
12. 13.
14.
Revue Economique XXI, no. 1 (1970); and Free trade and European econom ic developm ent in the 19th century in European Economic Review , 3, no. 3 (1972). A much larger study on trade and econom ic developm ent of 19th-century E urope will appear during 1975 in the form of a book (in French). I have n o t yet resumed the more specific problem of the foreign trade o f the Third World, which I had begun to deal with (b u t not to master) during the three years I spent in the research division o f the GATT. These are the reasons why the conclusions here m ust be considered as preliminary. Including Manchuria. Tea, which form ed about 50 per cent o f to tal exports in 1870 had by 1910 fallen to 9 per cent. Between the same dates the share of silk and silk articles fell from 40 to 25 per cent. A. Feuerwerker, The Chinese Economy c. 18701911 , Michi gan Papers in Chinese Studies, No. 5 (1969). The data relating to the foreign trade of China for the period 1950 to 1965 is based alm ost exclusively on R. C. Price, International trade o f Communist China 19501965, in An Economic Profile o f Mainland China, Vol. 2, Part 4, pp. 5 7 9 -6 0 8 .
6. The Terms of Trade 1. Ever since the notion of terms of trade was first introduced by J. Stuart Mill and A. Marshall it has been widened so th a t we now possess three major concepts, all different, o f which the m ost com m only employed is net barter terms of trade: this expression, under the name o f terms of trade, is the only one to be used here. 2. The estim ate was repeated again in another U.N. publication, Instability in Export Markets o f Under-Developed Countries (New York, 1962). 3. W. A rthur Lewis, World production, prices and trade 18701960, in The Manchester School o f Economic and Social Studies , XX (1952), pp. 10534. We-may include also the index com puted by Kindleberger for industrial Europe which is discussed below. 4. W. Schlote, Entwicklung und Struckturwandlungen des englischen Aussenhandels von 1700 bis zur Gegenwart (Jena, 1938). English translation by W. O. Henderson and W. H. Chaloner,
228
The Economic Development o f the Third World since 1900 British Overseas Trade from 1700 to the 1930's (Oxford, 1952). Annual figures are given here because in the index for industrial Europe calculated by Kindleberger data is available only for isolated years. The countries included are Great Britain, Germ any, France, Italy, Holland, Belgium, Sweden, Switzerland. C. Kindleberger, The Terms o f Trade and Economic Development, in The Review o f Economics and Statistics , special supplem ent, Problems in International Economics (Feb. 1958), pp. 72 90. For further details see also by the same author, The Terms o f Trade. A European Case Study (Cambridge, Mass., 1956). Index compiled about 1886 by A. Sauerbeck and published in Prices o f com m odities and the precious m etal, in Journal o f the Statistical Society (1886). It was continued in the review known as The Statist. Industrialization and Foreign Trade (League o f Nations, Geneva, 1945), p. 183. Thus of the 55 articles included in Sauerbecks index (45 products, 10 of which are included twice according to type and quality) prices for 36 are im port prices, to which m ust be added another 10 whose prices are a direct function of im port prices (e.g. price o f flour = a direct function of the price of American wheat). Only 1012 articles therefore may be considered as representing domestic prices, i.e. only 20 per cent of the total (all the articles in this index are given equal weight). Thus 80 per cent o f the index is based on im port prices. As will be pointed o u t below, this bias is n o t peculiar to Sauerbecks index; it is quite general and stems from the fact th at im port price statistics were usually the only ones available in the nineteenth century. Readers should note th a t although, as we have seen, the assumption o f a secular deterioration in the terms o f trade of raw materials is very widely accepted (especially by people dealing with th e problem s of the Third World), the idea that transport costs played a part in the deterioration has occasion ally been form ulated, b u t only by theorists o f international trade and w ithout any systematic effort to make an estim ate of their im pact (see particularly, G. Haberler, A Survey o f International Trade Theory (Princeton, 1961) and T. Morgan, The long run terms o f trade between agriculture and m anu facturing in Economic Development and Cultural Change,
5.
6.
7.
8. 9.
10.
I Notes to 6, The Terms o f Trade 229
11.
VIII, 1 (Oct. 1959), and Les tendances des term es de rechange et leurs repercussions in Les Minutes de la Conference de Brissago (Cahiers de 1T.S.E.A., 167 P. 11 (Nov. 1965), pp. 99142). According to the calculations of world values o f trade m ade by the statistical office of the League o f Nations (Industrialization and Foreign Trade) the difference betw een the two aggregates is as follows for the period under consideration: 1 8 8 1 -8 5 1 8 8 6 -9 0 1 9 2 1 -2 5 1 9 2 6 -2 9 1 9 3 1 -3 4 17.5% 17.1% 4.3% 10.9% 12.4%
12.
13.
14.
15.
16.
17.
18. 19.
A measure o f the drop in insurance costs can be found in the percentage o f ships totally lost or missing. For all registered ships these losses were 2.8 per cent annually betw een 1871 and 1880. By 1926, in spite o f the much greater distances involved, the percentage fell to 0.7. (From Shipping wreck statistics, Encyclopaedia Britannica, 19th ed.) Ignoring the effects o f transport costs, I have concluded th at the probability was an im provem ent o f 10 to 25 per cent, which by merely adding to the 10 to 20 per cent due to freight costs gives us a figure o f 21 to 50 per cent. In this case all under-developed countries including the com m unist ones, as well as Australia, New Zealand, South Africa and Canada. B. M. Bhatia, Terms o f trade and economic development: a case study o f India, 18611939, in the Indian Economic Journal , XIV (April June, 1969), pp. 41433. The basic figures were extracted by S. Bambrick (Australias long-run terms o f trade in Economic Development and Cultural Change, 19, Oct. 1970) from R. Wilsons study, Capital Imports and the Terms o f Trade (Melbourne, 1931). Calculated by the Canadian Bureau o f Statistics and repeated in M. C. U rquhart (ed.), Historical Statistics o f Canada (Toronto, 1965), p. 184. From J. Bjerke, Langtidslinjer i norsk Qkonomi , 18651960 (Oslo, 1966), pp. 1 4 2 -4 . From E. Pihkala, Suomen Ulkomaankauppa , 18601917: Finland's Foreign Trade (Helsinki, 1969).
230 The Economic Development o f the Third World since 1900 20. 21. From A. 01gaard, Growth , Productivity and Relative Prices (Amsterdam, 1968), p. 242. From M. Levy-Leboyer, Lh6ritage de Simiand: prix, profits et term es des echanges au XIXe Siecle, in Revue Historique , 493 (Jan March, 1970), pp. 11011. As the dom estic price series stops at 1921, I have com pleted it by the wholesale price index calculated by the Institut National de Statistiques (see Annuaire Statistique de la France). From W. G. Hoffmann, Das Wachstum der Deutschen Wirtschaft Seit der Mitte des 19. Jahrhunderts (Berlin, 1965), pp. 5 6 2 -6 3 and 5 6 9 -7 0 . A. Montesano, * 1 1 m ovimento dei prezzi in Giappone dal 1878 al 1958 in Giornale degli Economesti e Annali Economia (Nov Dec. 1967), p. 968. Taken from Etudes sur Teconomie mondiale , I, Les pays en voie de developpement dans le commerce m ondial (U.N., 1963), p. 20. As, for instance, for fuel prices, U.S. export prices have been taken. It is regrettable th a t the average for 195052 should have been chosen as a reference point for the fifties since it includes the boom year o f 1951 which was due to the Korean War. Note th at for the period after 1950 the statistical office of the U.N. have calculated an index o f export prices which is m ore elaborate than the one used here for long-term changes. Le commerce international en 1952 (Geneva, 1953), table on p. 130. The weighting of the index is based on o u tp u t in 1929. Thus, between 1929 and 1952, for example, th e index of wholesale prices grew by 2,500 per cent in France, 230 per cent in Belgium, 174 per cent in Holland, about 5,000 in Italy, 165 in Sweden, and 68 even in Switzerland. Statistical Yearbook 1969 (New York, 1970), p. 54; details about the m ethod o f construction and the statistical coverage appear in a docum ent which, so far as I know, has n o t y e t been published. Thus, between 1954 and 1962 the index o f wholesale prices of prim ary products in the United States fell by 3.5 per cent, while th a t o f m anufactures rose by 10.2 per cent and th a t of capital equipm ent by 23.0 per cent. The index o f average prices o f exports o f capital equipm ent from the developed countries to the Third World is used; also the index of average prices o f prim ary product exports from Third World countries to developed countries. The trade in both cases concerns non-comm unist countries.
22.
23.
24.
25.
26. 27.
28.
29.
30.
Notes to 7, Level o f Education 31. 32. 33.
231
34.
For a discussion of this problem the works referred to in fo o tnote 1 o f Chapter 5 should be consulted. 1951 is excluded in order to eliminate the effects of the boom caused by the Korean War. World production (excluding China) of plastics and resins rose from a figure o f about 2 million tons in 1953 to 10.6 million tons in 1963. In 1970 production was 30.7 million tons, i.e. a larger tonnage than th at o f world production o f all non-ferrous metals, including aluminium. The following variations in this index occurred between 1952/54 and 1964/66. Developed countries 2.7% -7 .1 % 16.3% 0.3% 37.9% Undeveloped countries -1 5 .4 % -1 0 .8 %
Food products O ther agricultural products Minerals Total o f above Non-ferrous metals 35.
-1 0 .5 % 40.3%
36.
Based on the average o f the growth rate, 1934/381963/65 for world o u tp u t (excluding U.S.S.R.) o f th e following products: citrus fruits (excluding U.S.) 159 per cent; ground nuts 69 per cent; bananas 175 per cent; cocoa 79 per cent; coffee 55 per cent; rubber 126 per cent; copra 36 per cent; cotton fibre (excluding U.S.) 109 per cent; palm and palmoil 157 per cent; ju te 60 per cent; tea 123 per cent; cane sugar 123 per cent. The average is 105.9 per cent. Thus for the United States, while betw een 1889 and 1937 the annual average rate o f growth o f production per un it o f total input was 0.5 per cent in the farm sector, in non-farm sectors it was 1.7 per cent. Between 1937 and 1957 it was 3.4 and 2.1 per cent respectively. The differences for labour productivity are even more striking. (From Historical Statistics o f the United States , pp. 599 602.)
7. The Level of Education 1. For the long-term developments in differing technological conditions, readers are referred to my study Revolution industrielle et sous-developpement , 4 th ed. especially Chapter 12, Les obstacles decoulant des m odifications du contenu de la technique, pp. 16475. 2. Thus, to quote only a few examples, here are the rates of
232
The Economic Development o f the Third World since 1900 growth of total population compared to th a t o f ages 514: Total population +23.0 per cent +57.2 per cent +67.4 per cent +24.3 per cent +73.0 per cent Population ages 514 +39.6 per cent +58.6 per cent +69.2 per cent +30.4 per cent +99.0 per cent
India Philippines Venezuela Chile Morocco
(1 9 5 1 -6 1 ) (1 9 4 8 -6 3 ) (1 9 5 0 -6 4 ) (1 9 5 2 -6 0 ) (1 9 5 2 -6 3 )
5. 6.
8.
9.
10.
11,
Rates prior to 1950 have been estim ated on the basis of scattered data relating to the age groups in 1950. Figures for 195070 are based on UNESCO com putations. Rates are calculated from statistics on the subject in m any European countries and from various sources, especially the historical sections o f statistical yearbooks. N ote th a t the rate mainly applies to the period 18301900 since statistics earlier than 1830 are usually lacking. See especially C. M. Cipolla, Literacy and Development in the West , (London, 1969). A rate o f 95 98 per cent can be considered as the practical lim it which has n o t up to now been exceeded in the developed countries. Facts used in this chapter about education in China after 1958 have m ostly been taken from : S. E. Fraser (ed.), Education and Communism in China. An Anthology o f Commentary and Documents (Hong Kong, 1969); L. A. Orleans, Communist China's E ducation, An Economic Profile o f Mainland China (Joint Econom ic C om m ittee o f U.S. Congress, Washington, 1967), pp. 4 9 9 518; Chu-Yuan Cheng, Scientific and engine ering m anpower in Communist China, An Economic Profile o f Mainland China, pp. 519 47; China News Analysis , esp. 792 (20 Feb 1970), 816 (2 O ct 1970) and 868 (21 Jan 1972). See Le Monde, 8 July 1970. Statistics o f Students Abroad 19621968 (UNESCO, Paris, 1972). Including students registered to study abroad. From Statistical Yearbook (UNESCO) and Statistics o f Students Abroad 1962 -1968 . China News Analysis , 792 (20 Feb 1970).
Such co him quyei
Notes to 8, Urbanization 8. Urbanization 1.
233
2.
3.
Small countries founded on international trade are excluded from this statem ent: such countries could achieve very high levels o f urban concentration by reason o f their role as intermediaries. Venice, the other Italian city republics, and the Low Countries spring to mind as the m ost obvious examples. Even such factors as policies governing transport prices, especially reduced w orkers fares, can have profound effects on the localization o f population, and underline th e am biguity of even the m ost straightforw ard definitions of urban population. These are the pairs o f countries: Urban population percentage (1960) 38 28 60 52 38 25 Per capita (195 8 6 3) GNP (US $) 805 1570 930 1180 830(*) 1200(*)
Austria Switzerland Holland Belgium Hungary Czechoslovakia
*Net material product per capita. The dollar conversion has been made on the basis o f the 1960 basic ra te . On the basis of a non-commercial rate the figures would be $470 for Hungary and $830 for Czechoslovakia. 4. B. Ward, The Poor W orlds Cities; the cities th at came too soon, in The Economist , 233 (Dec 6 1969), pp. 56 62. 5. G. Jenkins, Africa as it urbanizes in Urban Affairs Quarterly , II (March 1967), pp. 6 8 -8 0 . 6. P. Verhaegen, Vurbanisation de VAfrique noire: son cadre , ses causes et consequences economiques , sociales et culturelles (Bibliographical study), (Brussels, 1962). 7. P. Bairoch, Urban Unemployment in Developing Countries. The nature o f the problem and proposals fo r its solution (I.L.O., Geneva, 1973). 8. Growth o f the World's Urban and Rural Population 1 9 2 0 -2 0 0 0 (U.N., New York, 1969). 9. The definition is agglomerations of 20,000 inhabitants and more, which is clearly as arbitrary as any other definition. Selecting a single definition for all types o f economy at all periods, although certainly the least objectionable solution, suffers from obvious disadvantages.
234 10.
The Economic Development o f the Third World since 1900 United Nations projection. For reservations about this see below. According to my own estim ate the figure should be 4.9 per cent. The notion o f hyper- or over-urbanization came into use about tw enty years ago. In particular it was a t the heart of the discussions a t the jo in t U.N. and UNESCO seminar in Bangkok in 1956 (see Ph. Hauser, ed., Le phenomene d }urbanisation en Asie et en Extreme-Orient (UNESCO, Calcutta, 1959). The notion had already been present in K. Davis and H. H ertzs study, Urbanization and the developm ent o f pre-industrial areas in Economic Development and Cultural Change III (Oct 1954), pp. 626. The authors com m ent th a t they took the notion o f over-urbanization from a com m unication by P. Parke Jr. at the annual meeting of the Eastern Sociological Society in 1954. I am, of course, aware o f the unreliable nature o f simple indicators o f growth levels. But as work in progress on this problem has made no substantial progress, the indicators used here remain th e least objectionable o f those available. I have used here figures from my study Les ecarts des niveaux de
d e v e lo p p e m e n t e c o n o m iq u e en tre p a y s d e v e lo p p e s e t sou s-
11.
12.
13.
14.
15.
developpes de 1770 2000 in Tiers-Monde, XII (July Sept. 1971). Making fairly crude corrections to account for differences in purchasing power produced a lower difference about 65 per cent. There is a vast literature on this problem. The following recent studies can be selected from it: E. J. Berg, Wages and unem ploym ent in less-developed countries in The Challenge o f Unemployment to Development and the role o f Training and Research Institutes in Development (O.E.C.D. Development Centre, Paris, 1970), pp. I l l 41; J. C. Caldwell, African Rural-Urban Migration: The Movement to Ghana's Towns (London, 1969); M. P. Todaro, A m odel of labor migration and urban unem ploym ent in less-developed countries, in American Economic Review , 59 (March, 1969), pp. 138 48; C. R. Frank Jr., Urban unem ploym ent and economic growth in A frica in Oxford Economic Papers, 20 (July, 1968), pp. 2 5 0 -7 4 . One m ethod of testing the increase in the gap which is sometimes used is to compare agricultural and industrial prices.
Notes to 8, Urbanization
235
16.
17.
18.
19.
20.
21. 22. 23.
But this is only valid in certain circumstances and when the rates of growth in productivity are similar. R. D um ont, False Start in Africa (London, 1966), p. 88 (translated from the French, L Afrique Noire est mal partie , Paris, 1966, new ed.). L. Roussel, Measuring rural-urban drift in developing coun tries: a suggested m ethod in International Labour Review , 101 (March, 1970), pp. 229 46. It goes w ithout saying th a t the relationship level o f education = propensity to em igrate is not only applicable to lessdeveloped countries and th a t it also exists in industrialized societies. If they were not actually urban centres created de novo , then they were centres whose rapid growth arose from the choice of the site as a focus o f trade penetration and therefore, because of this, they were alm ost always ports. Such were, in Asia: Bombay, Calcutta, Madras, Manila, Djakarta, Singapore, etc; in Africa: Niamey, Brazzaville, Dakar, Yaounde, Pointe Noire, etc. Because of European settlem ent Latin-American, and some N orth African, cities do not come into this category. Articles and reviews on these problems will be found in the special number of Anna les, E.S.C. on Histoire et Urbanization (No. 4, July Aug. 1970). See also G. Hamdam, Capitals o f New Africa, in Economic Geography (July, 1964), reprinted in G. Breeze (ed.), The City in Newly Developing Countries , (Engle wood, N.J., 1969), pp. 146 61. See in this connexion A. Landry (under the direction of), Traite de demographie (Paris, 1945), p. 194; M. G. Mulhall, Dictionary o f Statistics (London, 1899), pp. 181 9. The industrial character o f the tow n is not by itself the cause o f the higher m ortality, because, as E. A. Wrigley notes in Population and the Industrial Revolution , m ortality was higher than in rural areas even when cities were administrative centres. W. Petersen, Population (London, 1969), p. 551. B. Benjamin, Social and Economic Factors affecting Mortality (The Hague and Paris, 1965), p. 35. The figures presented here are m ostly based on the following sources; E. Ni, Distribution o f the Urban and Rural Population o f Mainland China, International Population Reports, U.S. Dept, o f Commerce, No. 56 (Washington, Oct. 1960, in roneotype); J. P. Emerson, Manpower training in J. W. Lewis
236
The Economic Development o f the Third World since 1900 (ed.), The City in Communist China (Stanford, 1971); J. S. Aird, Population growth and distribution in mainland China, in An Economic Profile o f Mainland China, Vol. 2, Part III (Washington, 1967). It should be noted, however, th at according to forecasts by the population departm ent at the U.N. (see Monthly Bulletin o f Statistics for November, 1971), the percentage o f urban population measured according to national definition (which in China begins at 1000 persons insofar as 75 per cent o f these have no agricultural em ploym ent) moved from 18.3 per cent in 1960 to 24.8 per cent in 1970.
24.
9. The Labour Force and Em ploym ent 1. 2. Labour Force Projections 19651985, Part V (World Sum m ary), I.L.O., Geneva, 1971). For th e m ethods used to m ake th e estim ates and the im portance of the margins o f error, see P. Bairoch and J-M. Limbor, Changes in the industrial distribution o f the world labour force by regions, 18801960 in International Labour Review , 98, (Oct. 1968), pp. 31136. For further details, see P. Bairoch, La structure de la population active du Tiers-Monde, 19001970 in TiersM onde , X (April June, 1969), pp. 393 403. Part I of this chapter is largely based on this article. For further details and regional changes, see P. Bairoch and J-M. Limbor, op cit. Also S. Baum, The w orlds labour force and its industrial distribution, 1950 and I9 6 0 , in International Labour Review , 95, (JanFeb. 1967), pp. 96112. It should be noted, however, th at as the process o f economic development tends tow ards specialization, rural industrial labour (especially textiles and construction) gradually dis appears, which leads to an increase in the statistical im portance of the secondary sector o f em ploym ent. Austria, Belgium, Czechoslovakia, Denm ark, France, Germany, Holland, Italy, Luxemburg, Norway, Sweden, Switzerland. It should be noted that, according to projections m ade b y W. Schulte, L. Naiken and A. Bruni in Projections o f world agricultural population in Monthly Bulletin o f Agricultural Economics and Statistics , 21, (F.A.O., Rome, Jan. 1972) the fall in the relative im portance o f the agricultural labour force will be greater than the one I have projected. According to
3.
4.
5.
6. 7.
Notes to 9, The Labour Force and Employment
237
8. 9. 10. 11.
12.
13. 14. 15. 16.
17.
these authors, the percentage o f the agricultural labour force compared to total labour force would fall from 71.3 per cent in 1960 to 65.2 per cent in 1970 and from 57.7 per cent in 1980 to 53.7 per cent in 1985. These rates appear to me to assume much too great a fall in the relative size of the agricultural sector. In fact from 1950 to 1960 the latter fell by an average rate o f 0.6 per cent per annum; from 1960 70 by 0.8 per cent; and the rate expected betw een 1970 and 1985 would be 1.1 per cent. J. Berque, Le village (mimeographed), (Paris, 1959), p. 16. Shorter Oxford English Dictionary. O. Bloch and W. U. Wartburg, Dictionnaire etymologique de la langue franpaise 5th ed. (Paris, 1968). This was due to the combined effects o f a rise in the daily hours o f w ork and a fall in the num ber of days em ploym ent on saints and holy days. The num ber o f hours worked daily on industrial activities in the seventeenth and early eighteenth centuries seems to have been on average nine to ten. J. E. Van Dierendonck, Historisch Overzicht in P. J. Verdoorn, Arbeidsduur en Wewaartspeil (Leiden, 1947). In addition to Sundays the num ber of holy days was from 60 to 100 or even more, to which were added periods o f enforced seasonal inactivity which reduced still more the total num ber o f days worked. One of the m ost detailed surveys o f the average num ber of days worked during the year in pre-industrial societies is V aubans, in which he estim ated th a t a t the end of the seventeenth century day-labourers and craftsm en worked no more than 180 days during the course o f the year. D. Turnham assisted by I. Jaeger, The Employment Problem in Less Developed Countries. A Review o f Evidence , (O.E.C.D., Paris, 1971). According to D. Turnham, op cit.f p. 57, and Yearbook o f Labour Statistics 1971 (I.L.O., Geneva, 1972). See D. Turnham, op. cit. and The World Employment Programme, (I.L.O. Geneva, 1969), pp. 4 2 5. G. A rdant, Le plan de lutte contre la faim (Paris, 1964). This section is mainly based on my study (especially Chapter 3) entitled Urban Unemployment in Developing Countries: The nature o f the problem and proposals fo r its solution. This study may be consulted for further details. The definition used here is the same as was used in the last chapter, i.e. agglomerations of 20,000 inhabitants and more.
238 18.
The Economic Development o f the Third World since 1900 W ithout detailed studies to explore this aspect o f the problem it is obviously difficult to offer any reasons. But it is at least probable th at the concentration o f government offices with lower accidental and cyclical unem ploym ent at least partly explains why the difference should exist. General consider ations of public order or simply prestige may also help to p u t a brake on the immigration o f workless people. Instead of afigure for urban population of 19.7 per cent o f the total which is th e one forecast by the United Nations projections, I have adopted a figure o f 21 per cent which assumes a pace of urbanization slightly below w hat was achieved between 1950 and 1960. D. Turnham, op. cit.t pp. 4 8 50. G. W. Jones, U nder-utilization o f m anpower and demographic trends in Latin Am erica in International Labour Review , 98 (Nov. 1968), pp. 4 5 1 -6 9 . I do not make any distinction here between overt under employment defined as the percentage o f workers w ith a job in which the working tim e is less than normal or who are seeking or would accept a supplem entary job, and disguised under-employment defined as a situation characterized by a poor distribution o f m anpower resulting in especially low returns and under-utilization of skills or a low level o f productivity (see in this connexion Eleventh International Conference on Labour Statistics (I.L.O., Geneva, 1966). See especially, The World Employment Programme.
19.
20. 21.
22.
23.
10. Macro-Economic Data 1. A. Eckstein, The National Income o f Communist China (New York, 1961), p. 84. Also the same author, The strategy of econom ic developm ent in Com m unist China, in The American Economic Review (May 1961), p. 509. W. W. Hollister, Capital form ation in Com m unist China in Industrial Development in Communist China, ed. Choh-Ming Li (New York, 1964), p. 40. P. Bairoch, Revolution industrielle et sous-developpementy pp. 1 3 2 -4 . Readers should consult my book quoted in note 3; for the mechanisms thus set in m otion see especially Chapters 6 and 7; for the causes o f the difference betw een the western and the under-developed countries see Chapters 12 and 13.
2.
3. 4.
Notes to 10, Macro-Economic Data 5.
239
6.
7.
8.
T. Mende, From Aid to Recolonisation. Lessons o f a Failure (London, 1973). The book is, as the author says, 'an ambitious attem pt to reconsider an immense subject, namely, the relations o f the industrial powers of the northern hemisphere with the ex-colonial countries in the southern hem isphere. The attem pt is not only m ost successful b u t the book itself reads like a novel. Development Assistance Com mittee of the O.E.C.D. which in 1972 included sixteen o f the developed countries o f the West (Australia, Austria, Belgium, Canada, Denmark, France, Ger many, Italy, Japan, Netherlands, Norway, Portugal, Sweden, Switzerland, United Kingdom and United States). The national product o f these countries forms 98 per cent o f the total o f all western developed countries. Resources fo r the Developing World. The Flow o f Financial Resources to Less-Developed Countries 19621968 (O.E.C.D., Paris, 1970). Development Assistance (O.E.C.D., Paris, 1971). The following is the relative im portance of the aid for the principal developing countries in 1960/62 and 1969/71 (as a percentage of Gross National Product at m arket prices).
1960/62 1969/71 1960/62 1969/71
Belgium Canada France Germany Italy Japan
1.3 0.3 2.0 0.9 0.8 0.6
1.1 0.7 1.2 1.0 0.9 0.9
Netherlands Sweden Switzerland United Kingdom United States
1.5 0.3 1.8 1.1 0.8
1.5 0.7 0.7 1.1 0.6
9. 10. 11.
(According to Development Assistance 1972 , O.E.C.D., Paris, 1972.) The relative size of aid from the developed countries of eastern Europe is even lower about 0.2 to 0.3 per cent of G.N.P. including a rough estim ate for aid to the communist under-developed countries. See below section B; also General Conclusions. These figures and the following ones come from the annual reports of B.l.R.D. and A.I.D. Figures assembled for the non-communist under-developed countries as defined for this study (U.N. definition; see above Introduction). Changes in the public debt (end o f December) and servicing the debt (in billions o f dollars) for the eighty or so under-developed countries for which precise inform ation is available were as follows:
240
The Economic Development o f the Third World since 1900 Debt 33.4 38.4 43.4 48.7 53.1 59.7 Service 3.0 3.4 3.6 4.0 4.5 5.2
1965 1966 1967 1968 1969 1970
12.
13.
14.
15.
16.
17. 18.
19.
(According to World Bank and l.D.A. Annual Report 1972, Washington, 1972.) This certainly does not exclude, as we shall shortly see, the possibility or even the probability o f an over-estimation of recent rates o f growth. L. J. Zimmerman, The distribution o f World Income, 18601960, in Essays on Unbalanced Growth , (Institute of Social Studies, s Grave nh age, 1962). Frequent revisions are m ade in this series, as in others, which often result in considerable m odifications to the figures. Thus, the figures available to the end o f 1966 were used for growth rates for th e period 1957 63, b u t the figures published after 1968 modified these rates. To quote only one example, the growth rate o f G.D.P. for the non-communist developing countries was changed from 4.2 to 4.7 per cent. S. Kuznets, Problems in comparing recent growth rates for developed and less-developed countries in Economic Develop ment and Cultural Change, 20, No. 2 (Jan. 1972), pp. 1 8 5 -2 0 9 . For a recent study on international differences of income structure see Income Distribution in Latin America (E.C.L.A., U.N., New York, 1971). Calculated in my article Le m ythe de la croissance rapide au XI Xe siecle. It should be pointed o u t here that rates for the developed com m unist countries are higher still; thus, for the period 1950 70 growth of Gross Product at constant prices for the U.S.S.R. and other eastern countries was 8 per cent per annum. It should be remembered of course th a t the m ethods o f calculating th e Gross Product are different in the com m unist countries (e.g. services are excluded); however the differences do not result in much variation in growth rates. W. W. Hollister ( China's Gross National Product and Social Accounts 19501957 , M.I.T., 1958) gives an estim ate o f
Notes to 10, Macro-Economic Data
241
20.
21.
G.N.P. at constant prices (pp. 1323) which gives us for the period 195057 an annual growth rate o f 9.3 per cent. The author thinks th a t this perhaps is an over-estimate o f 1.5 points, which would leave us with the figure o f 7.8 per cent (p. xviii). Later th e same author revised his estimates (Estimates o f th e Gross National Product of China' in The Realities o f Communist China, Yuan-Li Wu (ed.), 1960) and advanced the rate to about 8.5 per cent. Ta-Chung Liu and Kung-Chia Yen ( Preliminary estimates o f the national income of the Chinese mainland 19521959 in American Economic Review, May 1961) give five estimates of national income which indicate an annual growth rate o f 5.9 per cent from 1952 to 1957. But growth was faster between 1950 and 1952 than betw een 1952 and 1957 (according to official figures 20 per cent and 9 per cent respectively); on the other hand growth in 1957 was slower. Choh-Ming Li (Economic Development o f Communist China, Los Angeles, 1959), by converting the figures o f Chinese accounting according to western m ethods, obtains a growth rate of 8.9 per cent between 1952 and 1957 (the same trend as quoted above: a peak in 1952 and a trough in 1957). But Choh-Ming Li thinks these rates are too high due to an over-estimate of industrial prices. But he does n o t give revised estim ates to allow for this. See also Ta-Chung Lius synoptic study The Tem po o f Economic Development o f the Chinese Mainland, 19491965 in An Economic Profile o f Mainland China, pp. 47 75. Ta-Chung Liu and Kung-Chia Yen, op. cit. (13 per cent for 1958 and 16 per cent for 1959); W. W. Hollister, Estimates of the Gross National Product of China (20 per cent in 1958 and 8 per cent in 1959). Y. L. Wu, F. P. Hoeber and M. M. Rockwell {The Economic Potential o f Communist China, Stanford, 1963, quoted by Cho-Ming Li in Industrial Develop ment in Communist China, New York, 1964) show an increase of 16 per cent in G.N.P. between 1957 and 1959 and a fall of 25 per cent between 1959 and 1962, which would place the 1962 level below the 1957 level. This seems to me extrem ely unlikely and smacks more of wishful thinking than o f scientific com putation. Ta-Chung Liu {op. cit.) places the total product for 1965 at the same level as 1958. Thus long-term credits (excluding repaym ents) rose to a total of 1720 million roubles from 1950 to 1957 (Choh-Ming Li, op. cit., p. 174), i.e. 430 million dollars at the then rate of exchange
ial com
or
242 The Econo?nic Development o f the Third World since 1900 which is $0.1 per capita per annum. This is a very low figure, because net public aid per capita in all non-com m unist countries taken together alone rose to $4.4 per annum in 1968 70. It should be rem em bered, moreover, th a t after 1960 aid from Russia and the other countries o f the com munist block was reduced to practically nothing and th a t since then China herself has supplied aid to other under-developed countries. L. J. Zimmerman, The D istribution of World Income, 18601960, op. cit. A. Eckstein, op. cit. p. 69. W. W. Hollister, China's Gross National Product and Social Accounts 19501957 , p. 3. For further discussion of this problem see p. 192. A brief b u t com plete historical survey o f the various attem pts made in this field will be found in the study o f W. Beckerman, Comparaison intemationale du revenu reel (O.E.C.D., Paris, 1966); he also gives an account o f his own m ethod of evaluating real levels o f per capita consum ption based on seven non-m onetary indicators (per capita volume: steel and meat consum ption; cement o u tp u t; letters; radio receivers; tele phones; and m otor vehicles). J. P. Delahaut and E. S. Kirschen, Les revenus nationaux du monde non com m uniste in Cahiers Economiques de Bruxelles, 10 (April, 1961). For 1957 the coefficients used to correct income expressed in dollars on the basis of exchange rates into income based on the real structure o f prices are: 2.11 for per capita incomes lower than $50: 1.98 for incomes from $50 99; 1.82 for incomes from $100144; 1.71 for incomes from $150199; and 1.61 for incomes from $2002 9 9 .1 have not used th e m ost interesting figures given by W. Beckerman {op. cit.) since it is impossible to calculate from them a valid estim ate for the under-developed w orld taken as a whole. This is because the countries for which he provides both private consum ption calculated with the help o f official rates of exchange and real private consum ption calculated by means o f his own m ethod form (population-wise) only 19 per cent of all non-comm unist under-developed countries. For a discussion o f factors bearing on this question my article Les ecarts des niveaux de developpement econom ique entre pays developpes et pays sous-developpes de 1770 a 2000 in Tiers Monde , XII (July Sept. 1971), pp. 497514, should be
22. 23. 24. 25. 26.
27.
28.
Notes to 11y General Conclusions
243
29.
consulted. The other articles in this number, which is entirely devoted to the Third World in the year 2000, are o f great interest. For a recent survey o f the Chinese economy see T. G. Rawski, Recent trends in the Chinese Econom y in The China Quarterly , 53 (Jan March, 1973), pp. 133.
General Conclusions 1. The problem, however, is far from straightforward, since, insofar as the use o f synthetic materials is rapidly spreading, it is more than likely th at many natural raw materials will lose much o f their commercial value in the future.
Appendix: Synoptic Table

Synoptic table summarizing the recent state of econom ic and social developm ent in selected Third World countries The table set out on the four following pages shows by means of nineteen indices the recent state o f the economic and social development in ninety developing countries, each of which has a population o f over 500,000 people. Taken together, these countries account for over 99 per cent o f total population in the T hird World. It would have been possible to give m ore up-to-date statistics in some cases, but in the interests of hom ogeneity I have decided to give either figures for 1970 or else figures for the yearly growth rate between 1960 and 1970. The advantage o f this is th at the reader can, if he wishes, use the data in the table as the basis for deducing further inform ation (e.g. by m ultiplying per capita G.D.P. by Total population it is possible to obtain Total G.D.P.). Sources: F.A.O. docum ents (mainly Production Yearbook) I.M.F. docum ents (mainly International Financial Statis tics and Direction o f Trade) I.L.O. docum ents (mainly Yearbook o f Labour Statistics) U.N.C.T.A.D. docum ents (mainly Handbook o f Interna tional Trade and Development Statistics) U.N. docum ents (mainly Statistical Yearbook , Monthly Bulletin o f Statistics , Population and Vital Statistics Report , and Demographic Yearbook) WORLD BANK docum ents (mainly World Bank Atlas , Trends in Developing Countries , and Annual Report) And statistical yearbooks of individual countries.
E X P L A N A T I O N O F H E A D I N G S IN T H E T A B L E
The sign Population Urban population Infant mortality Active populations 244
Means th a t no recent valid data is available. The total figures represent estim ates o f popu lation at mid-year. The ratio o f urban population (according to local definition) to total population. The num ber o f deaths o f infants under one year per 1000 corresponding live births. The percentage o f each sector has been calculated in relation to total economically
245
active population, excluding (whenever such categories are reported) activities not ade quately described, persons seeking work for the first time, and unem ployed persons who are not counted in the sector in which they were previously employed. Agriculture in cludes mining, manufacturing, construction and electricity. Services include commerce, transport and services. Gross Domestic Product For a large num ber o f countries the data contains a wide margin o f error. The figures have not been corrected for differences in the purchasing power o f currencies (see Chap ter 10, section C). This represents the daily net per capita food supply (average for a num ber o f years around 1970). Gives the yearly growth rate (in this case for the period 1961 65 to 1968 72) of the F.A.O. index num ber o f total agricultural production. This is the estimated apparent inland con sum ption o f m odern forms o f energy (exclud ing wood) expressed in kgs o f coal equivalent. In cases where there are two indices o f consumer prices (for European and nonEuropean) the figures reported are based on the index for non-European. Relates generally to special trade in which im ports are expressed in c.i.f. values and exports in f.o.b. values. The reason why there is not data for a large num ber of non-comm unist countries is that some of them are part o f a m onetary union. The figures include gold, foreign exchange, reserve position in the I.M.F., and special drawing rights. The figures relate to the situation at the end o f 1970. Represents the situation o f the external pub lic debt outstanding (including undisbursed) at the end of 1970.
Calories supply per capita Total agricultural production Consumption o f energy per capita Consumer prices
Foreign trade
I n te r n a tio n a l
liquidity
External public debt
246
The Economic Development o f the Third World since 1900 1970 situation or 1960
DEM OGRAPHY Total p o p u la tio n 1,000 * Urban Infant popula m o tality tio n % %o A C T IV E P O P U LA TIO N %
Year
A griculture
In d u stry
Services
L A T IN A M E R IC A A rg e n tin a B o liv ia * ' B razil C h ile C o lo m b ia C o s ta R ic a Cuba D o m in ic a n R e p . E cuador El S a lv a d o r G u a te m a la G uyana H aiti H o n d u ra s J a m a ic a M e x ic o N ic a ra g u a Panam a P a ra g u a y P e ru P u e r to R ic o T r in id a d a n d T o b a g o U ru g u a y V e n e z u e la A S IA
2 3 ,2 1 0 4 ,9 3 0 9 3 ,3 9 0 8 ,8 6 0 2 1 ,1 2 0 1 ,7 4 0 8 ,4 7 0 4 ,0 6 0 6 ,0 9 0 3 ,5 3 0 5 ,2 8 0 720 4 ,8 7 0 2 ,5 8 0 1 ,8 7 0 4 9 ,0 9 0 1 ,9 8 0 1 ,4 3 0 2 ,3 9 0 1 3 ,5 9 0 * 2 ,7 2 0 1 ,0 3 0 2 ,8 9 0 1 0 ,4 0 0
1.5 2 .6 2 .9 2 .3 3 .2 3.3 2.1 3 .0 3 .4 3 .7 3.1 2 .4 2 .0 3.3 1.2 3 .5 3.5 3 .3 3.1 3.1 1 .4 2.1 1.3 3.5
80 34 56 76 60 36 60 40 38 38 31 29 18 32 37 59 46 48 36 52 58 12 78 76
58 77 79 70 56 35 50 77 52 83 40
60 70 71 64 63
20
36
39 26 61 45 38 36 72 27 35 40 49
70 62 71 70 65 61 60 70 71 70 62 67 73 71 63 71
47 20 49 50 52 59 58 66 33 70 40 42 47 43 56 49 7 21 20 20
17 34 20 19
14 18 13 14 28
44 37 47 31 31 33 23 29 20 39
11 25 24 16 16 19 21 34 37 31 27
19 36 34 37 41 25 30 59 42 49 53
A fghanistan Burm a China (official fig.) China (w estern est.) H ong Kong India Indonesia
17,090 27,580
2.0 2.1
7 19
66
773,660 3,960 539,860 115,460 Khmer Rep. 6,840 K orea Dem. Pop. Rep. 13,890 K orea Rep. of 31,020 Laos 2,960 Malaysia 1,040 M ongolia 1,250 Pakistan 114,190 Philippines 36,850 Singapore 2,070 Sri Lanka 12,510 Taiwan 14,040 T hailand 34,380 Viet-Nam Dem. Rep. 21,150 Viet-Nam R ep. o f 18,330
1.8 2.6 2.3 2.0 3.2 2.7 2.6 2.4 3.1 3.0 2.7 3.0 2.3 2.4 2.9 3.1 2.8 2.6
25 92 20 17 12 38 41 15 28 45 13 32 81
18
75

71 71 71 62
4 72
66
82
54 12 10 4
41 16 24 15
71
-
48

18

34

39
20
64 15 18 24
140 62 20 50
68 70 70 63
70
55 3
56
12 17 30 14
18
29
66
30
25
70

83
_
12
-
37
Y e a r ly g r o w th r a te ( p e r c e n ta g e ) f o r t h e p e r io d 1 9 6 0 t o 1 9 7 0 , e x c e p t in th e c a se o f to ta l a g ric u ltu ra l p r o d u c tio n
Appendix: Synoptic Table L970 yearly increase

G R O SS D O M ESTIC PRODUCT Calories Per su p p ly capita in per S U. S. capita F O R E IG N T R A D E Total agricul tural p ro d u c tio n * C onsum p tion o f energy p er capita
247
V olum e to tal*
Retail prices*
Im ports $ m illions
E xports S m illions
Inter national liq u id ity $ m illions
E xternal public d eb t S m illions
4.2 5.7
6 .0
4.4 5.2
6.1
4.5 5.3 5.6 5.5 4.4

0 .8
5.0 5.1 7.0 7.3

8 .0
4 .7 5.2
8.1
4.5 1.4 5.6
1055 195 365 795 315 530 530 345 265 295 350 330 90 250 590 660 435 710 245 435 1650 800 820 920
3200 1800 2800 2600

2100
1.2
2400 2500
2100 2000
3.9 3.7 1.9 4.2 6.7 1.5 3.3 2.7

2.1
1900
2000 2100
4.0 0.9
1 .2
1900
2200
1687 193 478 1275 653 423 1076 253 293 205 239 1071 27
222
21.4 5.6 44.0 26.5

1 1 .2
2.4
2 .0
5.5
1.8
2300 2700 2300 2400 2600

2200
3.1 4.1 5.6 2.7

2 .0
2500 2400 2700 2400
-3 .4 3.9 - 0 .2 4.8
1271 1236 380 700 149 629 3242 46 0 4 909 2598
4.3 0.7 0.9 2.3 2.7 2.3 4.1 2.7
1685 159 2849 931 843 317 1300 278 248 214 284 134 55
2 21
1773 227 2739 1247 736 231 1043

2 21
1.2
1.7 9.6 2.7 3.0 41.0 1.4
522 2461 198 353 64 619 542 233 1713
218 228 299 134 40 171 343 1402 175

111
5.1 16.1 13.6 9.8 4.7 10.4 5.3 2.4 4.1 6.9 9.8
6 .0 2 .0
673 46 1187 388 206 16 32 83 63 78

20
2457 529 3809 2503 1720 217 282 337

120
176 113
4
20
10.4 7.5
6 .2 12.1
64 1044
480 233 2691
14.7 9.0 9.2 5.3

6 .0 1.0
139 744 49 304 18 329
43 175
1021
139 180 3791 224 234 142 1184 90 318 806
2.5 3.7 3.5 1.7 12.9 3.7 3.5 3.3 5.1 8.7 4.3 6.3 4.6 5.0 9.4 4.9 7.1 7.9 7.9 3.6
90 75 350 160 885 95 105

110
2100 2000
1.7
1.1
26 58
75 169
86
108
5.6 - 7 .1
2200
2400
2000
2.7 2.5 2.3

2 .8
530
1010
1900
2200
330 260 75 355 460 125 265 960 170 390 175
100 200
2300 2500
2200
2500 2500 2400

2000
2500 2300 2600 2300

2100 2200
1.7 3.7 5.6 5.4 - 0.3 5.1 3.0 8.7

2.1
3.8
1 .1
182 109 48 2185 785 93 468 885 94 292 808 149 925 256 172 302
2.9 6.3 237.0 4.4 13.9 20.7 0.9 3.6 4.1

1 .0
2800 2905 2124 893 54 1983 114 1490
2200
1151
1210
3.0 3.4
2 .0
2461 389 1524 1293
2514 2026 1009 39 835 7 1762 723 1119 1554 339 1428 710
1.5 13.8 4.3

1.8
- 5 .6 38.0
2 2 .6
4.0
6 .2
7.1 3.1 - 1 .3 24.0 5.7
46 94 1006 610 733 182 251 294 43 624 906
689
9235 3463 2637 549 4302 822 283 516 985 456
21.4
325
- 2 1 .6
242
w here th e period under consideration is 1961/65 to 1968/72 (see also the notes o f explanation given on p . 244).
248
The Economic Development o f the Third World since 1900 1970 situation or 196<
DEM OGRAPHY T otal p o p u la tio n 1,000 * Urban p o p id atio n % In fa n t m o rta lity %o A C T IV E P O P U L A T IO N %
Year
A griculture
In d u stry
Servit
A FRICA Algeria Angola Burundi C am eroon C entral A frican Rep. C had Congo Peoples Rep. D ahom ey E thiopia Gabon Ghana Guinea Ivory Coast Kenya Liberia M adagascar Malawi Mali M auritania M auritius M orocco M ozam bique Niger Nigeria Rw anda Senegal Sierra Leone Somalia Southern R hodesia Sudan Togo Tunisia Uganda Un. Rep. o f Tanzania U pper V olta Zaire Zam bia M IDDLE EAST Cyprus Egypt Iran Iraq Israel Jo rd o n Lebanon Libyan Arab Rep. Saudi Arabia Syrian A rab Rep. T urkey Y em en
14,330 5,570 3,540 5,840 1,610 3,640 936 2,690 24,630 500 8,640 3,920 4 ,310 11,230 1,520 6,750 4 ,4 4 0 5,050 1,170 810 15,520 8,150 4,020 55,070 3,680 3,930 2,550 2,790 5,310 15,700 1,970 5,140 9,810 13,270 5,380 21,570 4,180
3.1 1.3
2 .0 2 .1 2 .6 1.8
40 14 3
20
86 21
66
60

59 69

14
12

150 137
27 19
1.9 2.9
2 .2 1.1 2 .6 2 .6
27 7 30 13 9 32 31
11 21 10
55 159 55 148 187 65 92 16 136 __ 18 78 160 182
3.0 3.1 3.0

2 .6 2 .6 2.1
26 14
6 12
63 60 62
85 61 81
1.9 2.3 2.9 1.9 2.9 2.9 3.0

2.1
7 44 34
6 8
1.4 2.4 3.3 2.9 2.9 3.0 2.7 2.5

2.1 2 .8
23 3 26 14
20
18
12
13 43
10
62 71 60 63 61
66
38 56 94 77
7 15 9 27 18
1
23
10

12
35 25 5
11
11
2.5
7 4 15 31
69
46

27 19 24
62 35
32
43
633 33,330 28,660 9 ,4 4 0 2,910 2,310 2,790 1,940 7,740 6 ,2 5 0 35,230 5,770
1 .0
2.5 2.9 3.5 3.2 3.5

2 .8
3.7 1.7 2.9 2.5 2.3
39 42 41 58 82 47 41 27 24 43 39
6
25 116
72
66 66
38 56 47
28 16 27
20 20
36 14
71 61 70
23 153
70 65
9 43 19 51 74
33 26 25
20 11
34 28 26 59 30 56
28 15
Y early grow th rate (percentage) for th e period 1960 to 1970, except in the case o f to ta l agricultural pro d u ctic
,>70 yearly increase

G R O SS D O M E STIC PRO D U C T Calories Per su p p ly capita in p e r capita $U.S. 305 300 65 170 135 70 250 95 70 630 270 80 340 140 230
120

F O R E IG N T R A D E T otal agricultural p ro d u c tion * 0.9 2.7 4.7 4.0 1.7
0 .1
249
V olum e total*
2 .0
C onsum ptio n o f energy p er capita 4 60 156

8
Retail prices * _
Im p o rts S m illions 1257 369

22
E xp o rts $ m illions 1009 423 24 232 31 30 31 33

122 121
*
6 .0
International liq u id ity % m illions 339
E xternal public deb t $ m illions _
5.3
2 .8
4.7 1.3
1.2
1900 1900 2300 2300

2200 2200 2100 2200 2000 2200 2100 2100

2.4 5.0 4.9 4.1
91 62 23
222
5.0 4.7 4.6 6.4 2.7
0.9
8.9 7.6 3.7 3.6 4.8
2500
2200
80
100
2300 2300 2400

2100 2000
- 1 .4 4.0 3.1 3.2 2.9 2.3 4.6 3.4 2.7 2.3 4 .4

1 .6
36 33 887 170 97 231 148 296 67 46
5.2 3.5 7.7
3.8
1 .8
2 .6
242 34 61 57 64 172 80 411 78 388 397 150 170

86
13.0 18.0 9.1 7.1 8.4 5.6

6 .0
15
10

71
0.5
7.7 3.1 3.9 5.8 4.7 3.7 3.4 1.3 5.1 1.5 4.7 2.9 8.3 5.1 4.7 6.7 3.0 4.9
6 .1
20
133 168 198 151 19 46
10
155 225
210
2400
2200 2200
240 80 105 55 180 160 65 270

120
1.7 1.4 4.9 2.5

2.1

1 .8
433 56 469 217 213 145 60
5.3 9.9 3.9 0.7

1 2 .0 6 .8
58

220
9.9
6 .8

29
1
8.9
2300 2300
2000 2200 2200
0 .4 5.0
2.3 2.9 3.5 4 .4
2.5 4.1 4.3

2.1
-0.8
3.0
2 .6
133
1800 2600
2200
114 38
578 119 60 258 72
145 235 135

100
2300
2200
0.9 5.6 3.8

2 .0
2300 1700
2100 2100
3.0
3.3 1.7 2.9 4.5

2.1
47 56 76 684 324 58 1059 29 193 116 45 329 311 65 305

121
33
89 69
9.9
51.0 5.8 3.2 7.8 9.7
1 0 .0
46 140
488
156 32 1240 25 152
101

224
218 29 51 173 55 270 115 572 378 413 394 177 138 156 288 39 43 855
63 683
12
32.0
2 .8
8
35
21
5.5
2 .8
42
13 83 494
60
110
4.5
2.9
21.0
4.2
405
2300
272 47 533 477
31 355 293 55 183 246 238 18 781

1001
1.9 3.2 7.4 4.8 13.9 4.3 7.4 5.6 15.6 4.5 10.7
22
60 57 65
186 514
114 85 105 248 339 40 787 184 576 31 579 616
6.3 4.6 9 .0 5.8 8.9 6.4 4.4

2 .6 0 .0 6 .6
845
200
2500 2800
2100 2100
5.6 4.2
355 300 1485 285 530 1670 585 260 340 80
3000 2400 2500 2600

2200
2500 2800 1900
7.1 3.1 4 .0 5.2 6.4 - 5 .7 4.2 4.8 3.2 0.3 3.3

0 .8
1320 263 895 617 2524 295 719 578 827 457 481 13
1.1
3.7 1.7
2.1
5.6

2.5 6.4
238 787 1658 509 1422 184 568 554 692 357 894 143
108 762 2355 1093 734 34 198 2366 2423 203 589 15
7.2 3.0
1 1 .2
5.3 13.2 11.9 17.3 71.0 10.4 6.3

6 .2
- 2 .6
209 167 208 462 449 256 386 1590 662 55 431 59
49 1768 3022 276

2022
170

244 2626
vhcrc th e period under consideration is 1961/65 to 1968/72 (see also th e notes o f explanation given on p. 244).
aterial com direitos autor,
Bibliography
This list is not intended to be a complete bibliography on the problems o f the less-developed countries, b u t merely to give some indications for further research. For this reason it is divided into four sections: the first gives a short list o f some general works; the second collects together the principal international statistics; the third is devoted to bibliographical works (or parts o f works) dealing with the subject; and the fourth is a select list of the specialized journals devoted to the problems o f the Third World.
A. General works A U S T R U Y , J. Le Scandale du developpement . Paris, Marcel Riviere, 1965. B A I R O C H , P. Revolution industrielle et sous-developpement. Paris, 1963; 4th edition, Paris and The Hague, M outon, 1974. B A L D W I N , R . E. Economic Development and Growth. New York, J. Wiley, 1966. B A L O G H, T. The Economics o f Poverty. New York, Macmillan, 1966. B A R A N , P. A . The Political Economy o f Growth. New York, M onthly Review Press, 1957. B A U R , P. T. (ed.) Dissent on Development. Studies and Debates in Development Economics. Cambridge (Mass.), Harvard University Press, 1972. B R O O K I N G S I N S T I T U T E , Development o f the Emerging Countries. Washington, D.C., 1962. B R O W N , L. R. Seeds o f Change. The Green Revolution and the Development in the 1970s. New York, Praeger, 1970. C A I R N C R O S S , A . K . Factors in Economic Development. London, Allen and Unwin, 1962. C L A R K , C. The Conditions o f Economic Progress. 3rd ed., London, Macmillan, 1957. C O T T A , A. Analyse quantitative de la croissance des pays sousdeveloppes, Paris, Presses Universitaires de France, 1967. D O M A R , E. D . Essays in the Theory o f Economic Growth. N e w York, Oxford University Press, 1957. D U M O N T , R. and R O S I E R , B. The Hungry Future. New York, Praeger, 1969. 250
Bibliography
F A L K O W S K I , M.
251
Les problemes de la croissance du tiers-monde vus par des economistes des pays socialistes , Paris, Payot, 1968. F R A N K , G . Capitalism and Under-Development in Latin America , revised ed., New York, Modern Reader Paperback, 1969. F U R T A D O , C. Development and Underdevelopment. University of California Press, 1964. G A L B R A I T H , J. K. Economic Development. Cambridge (Mass.), Harvard University Press, 1964. G A R Z O U Z I , E. Economic Growth and Development. The Less Developed Countries. New York, Vintage Press, 1972. G E N D A R M E , R. La pauvrete des Nations. Paris, Cujas, 1963. H A G E N , E. E. The Economics o f Development. Homewood, 111., Irwin, 1968. H I G G I N S , B. H. Economic Development ; Principles, Problems, Policies, revised ed., New York, W. W. N orton, 1968. H I R S C H M A N , A . O. The Strategy o f Economic Development. New Haven, Conn., Yale University Press, 1958. H O R O W I T Z , D . The Abolition o f Poverty. London, Pall Mall Press, 1970. J A L E E , P. Le tiers-monde dans Veconomic mondiale. Paris, Maspero, 1968. J O H N S O N , H. G. Economic Policies Towards Less-Developed Coun tries. Washington, The Brookings Institute, 1967. KINDLEBERGER, C. P. Economic Development . Cambridge (Mass.), M.I.T. Press, 1965. KUZNETS, S. Economic Growth and Structure. London, Heinemann, 1966. L A C O S T E , Y. Geographie du sous-developpement. Paris, Presses Universitaires de France, 1965. L E B R E T , L. J. The Last Revolution: the Destiny o f Over- and Under-developed Nations. Dublin, LOGDS, 1965. L E WI S , w . A . The Theory o f Economic Growth. London, Allen and Unwin, 1955. M E I E R , G . M. a n d B A L D W I N , R. E. Economic Development: Theory, History, Policy. New York, J. Wiley, 1957. M E N D E , T. From Aid to Recolonisation, Lessons o f a Failure. London, Harrap, 1973. M Y I N T , H. The Economics o f the Developing Countries. 3rd ed., London, Hutchinson, 1965. M Y R D A L , G . Economic Theory and Under-Developed Regions. London, Duckw orth, 1957. M Y R D A L , G . The Asian Drama: An Inquiry into the Poverty o f Nations. New York, Panthon, 1968.
252

R.
Problems o f Capital Formation in Under-Developed Countries. New York, O xford University Press, 1960. P E A R S O N , L. B. (ed.) Partners in Development , Report o f the Commission on International Development. New York, Praeger, 1969. P E R R O U X , F. L economie des jeunes nations. Paris, Presses Universitaires de France, 1963. P I A T I E R , A. Equilibre entre developpement economique et develop pement social. Paris, Genin, 1962. S A C H S , I. La decouverte du Tiers-Monde. Paris, Flammarion, 1 9 7 1 . S A U V Y , A . Malthus et les deux Marx. Le probleme de la faim et de la guerre dans le monde. Paris, Denoel, 1963. SEERS, O. (ed.) Development in a Divided World. Harm ondsworth, Penguin, 1971. S I N G E R , H. W. International Development: Growth and Challenge. New York, McGraw-Hill, 1964. T I N B E R G E N , J. The Design o f Development (International Bank for Reconstruction and Development), Baltimore, Johns Hopkins Press, 1958. U . N . C . T . A . D . Towards a Global Strategy for Development. R eport by the Secretary-General of the United Nations. Conference on Trade and Development Docum ent TD/3/Rev. 1., United Nations, New York, 1968.
NURKSE,
B. Principal international statistical sources Im portant sources o f inform ation will of course be found in statistical yearbooks and periodicals of the different countries. Only works including statistics for a great num ber of less-developed countries or
for a group o f th o s e c o u n tr ie s are given here.
F.A.O. Food and Agricultural Organization o f the United Nations (Rome) Production Yearbook Trade Yearbook Monthly Bulletin o f Agricultural Economics and Statistics The State o f Food and Agriculture F.A.O. Commodity Review GA TT. General Agreement on Tariffs and Trade (Geneva) International Trade. . .
Bibliography
253
I.L.O. International Labour Office (Geneva) Yearbook o f Labour Statistics Bulletin o f Labour Statistics I.M.F. International M onetary Fund (Washington) Annual Report Direction o f Trade International Financial Statistics The Balance o f Payments Yearbook League o f Nations (Geneva, until 1940) Statistical Yearbook Monthly Bulletin o f Statistics World Production and Prices International Trade Statistics O.E.C.D. Organization for Economic Co-operation and Develop m ent (Paris) Development Assistance Review U.N.C.T.A.D. United Nations Conference on Trade and Develop m ent (Geneva) Handbook o f International Trade and Development Statistics Review o f International Trade and D evelopm ent. . . UNESCO United Nations Educational Scientific and Cultural Organization (Paris) Statistical Yearbook UNITED NATIONS (New York) Publications issued several times per year: Commodity Trade Statistics Economic Bulletin for Asia and the Far East Economic Bulletin for Latin America Monthly Bulletin o f Statistics Population and Vital Statistics, Reports Annual Publications: Demographic Yearbook Statistical Yearbook Statistical Yearbook fo r Asia and the Far East Studies on the Economic Situation o f Asia and the Far East World Economic Survey Yearbook o f International Trade Statistics Yearbook o f National Accounts Statistics WORLD BANK (Washington) Annual Report Trends in Developing Countries World Bank Atlas
254
C. Bibliographical studies CAIRE, G. Esquisse d un etat des travaux en matiere de developpe m en t, in A U S T R U Y , j . Le scandale du developpement , op. cit. pp. 3 2 7 -5 1 2 .
CANADIAN COUNCIL FOR INTERNATIONAL COOPERATION,
Bibliography o f International Development , O ttaw a, 1969. D E G R E E F , G. Elements de bibliographic critique du sousdeveloppement econom ique, in La Belgique et Taide economique aux pays sous-developpes , Institut Royal des Relations Inter nationales, The Hague, 1959, pp. 453512. E C O N O M I C D E V E L O P M E N T I N S T I T U T E , Selection Readings and Source Materials on Economic Development. A list of books, articles and reports recom m ended as reading m aterial for general development course of the Economic Development Institute, 1965 66. International Bank for R econstruction and Develop m ent, Washington, 1966. F . A . O . Selected Bibliography on Cooperation , 1957. G E I G E R , H. National Development 17761966. A selected and annotated guide to the m ost im portant articles in English; Scarecrow, 1969. H A Z L E W O O D , A. The Economic Development. An annotated list of books and articles, 19591962; Oxford, 1964. K A B I R , A . K. M. Social Change and National Building in the Developing Areas. A selected annotated bibliography. National Institute of Public A dm inistration, Dacca, 1965. K A T Z , S. M. and M c G O W A N , F . Selected list o f U.S. Readings on Development, prepared for the United Nations conference on the application o f science and technology for the benefit o f the less developed areas. U.S. Governm ent Printing Office, Washington, 1963. L A M B E R G , v . Literatur uber Entwicklungslander, Hanover, 1966. L A W R E N C E , M. M. Decade o f Development. Compendium o f U.S. sponsored Center of Economic Publications, 19591969. Washington (?), 1970. O.E.C.D. Economic Growth. Special annotated bibliography no. 1718. Paris, 1968. O . N . U . Bibliographic des ouvrages relatifs aux methodes et aux problemes de Tindustrialisation dans les pays sous-developpes , 1956. P I A T I E R , A. Bibliography in A. Piatier, Equilibre entre developpe ment economique et developpement social , pp. 10581. R E Q U A , E. S. a n d S T A T H A M , J. The Developing Nations, A Guide to Information Sources , D etroit, 1965.
Bibliography
STRACK, H. B.
255
Bibliographic uber Entiwcklungsldnder nach Sachgebieten , Hanover, 1965. T A R R , R. Bibliographic com m entee sur les problem es du sousdeveloppem ent in Problemes de VEurope , No. 8 and 9, pp. 2 2 8 -4 2 , 2 5 8 -7 4 . TRAGER, F. N. A Selected and A nnotated Bibliography on Economic Development 19531957% in Economic Development and Cultural Change, July 1958, pp. 257329. D. Specialized journals in the problems of developm ent In addition to the journals m entioned here, studies dealing with the problem s o f less-developed countries will often be found in journals devoted to economics, sociology and geography as well as to countries or regions of the Third World. Community Development Journal (Manchester) Croissance des Jeunes Nations (Paris) Cultures et Developpement (Louvain) Desarrollo Economico (Buenos Aires) Development Digest (Washington) Developpement et Civilisations (Paris) Economic Development and Cultural Change (Chicago) Finance and Development (Washington) France d Outre-Mer (Paris) International Development Review (Washington) Journal o f Development Economics (Amsterdam) Journal o f Development Studies (London) Journal o f Developing Areas (Macomb, 111.) Journal o f Developing Economies (Berkeley) Monde en developpement (Paris) Presence Africaine (Paris) Terzo Mondo (Milano) The Developing Economies (Tokyo) The Journal o f Development Studies (London) The Pakistan Development Review (Karachi) Tiers-Monde (Paris) World Development (Oxford)
Index
aggregate wage outlay, 60-1 agricultural revolution, 196J3 agriculture: export crops, 144L6 generation of purchasing power, 20 levels of development, 82 output: growth rates, 17-20, 196, 201: index, 16-17, 19-21: per capita, 17-18; totals, by country, 247, 249 productivity: average levels, 29-31, 35-6: changes since 1900, 35-9: China, 32i comparison with advanced countries during take-off, 1 2 4 5 ; effect on terms of trade, 134j European influence, 34; future estimates and requirements, 195, 203; index, 13-14, 21-6, 35-6, 38j and land availability, 32-4; and plantation agriculture, 34; thresholds, 26-9. relationship with manufacturing, 88-9.1 research, 45-8 share of labour force, LL 160-2, 246, 248 subsistence, 1 ^ 14 yields, 28-9 aid: financial flow, 177-181 future distribution and requirements, 2Q7-8 geographical distribution, 182 internal distribution, 209 private, 177, 179-80 proportion of G.D.P., L80 public, 177, 129 sources, 208 Aird, J. S., 6 balance o f trade, 93* 95-7 birth control, campaigns and policies, 9-10, Z ilM Borlaug, N. E., 45 capital formation: as an indicator of economic development, 122 China, 1234i European experience, 176-7 rates, 1 2 1 4 share of imports, 175-7 capital-output ratio as an indicator of economic development, 172 centralized economies: Chinese experience, 202 manufacturing growth, 62 see also state intervention China: agriculture, 37, 201 birth rate, 2 capital formation, 173-5 education, 1 39, 143 exports, LQ9_il_Q foreign trade, 109-10 Gross Domestic Product, 188-90, 123 growth rates, 188-90 manufacturing industry, growth of
Z24
per capita income, 191-3, 201-2 population, 5-7 productivity, 32 urbanization, 1 56-7 Clark, C 28 colonialism; effect on development in India, 202 effect on urbanization, 153-4 provision of infrastructure, 201 construction, employment, 160 cotton, consumption as an indicator of development, 84^ 16 Delahaut, J. P., 192 demand, domestic, ZO elasticity of, in manufactures, 129 demographic inflation, IS . Dumont, R., 45_, 203 Eckstein, A., 121
257
258
Index
China, IQ.9ilQ effect on urban growth, 149-50 fuel, 99 geographical pattern, 101-4 manufactured goods, 98-9. 105 role in development, 198-9, 205 share of developing countries, 94* 95* l-Q-4-8 Fourastie, J., Z fuel production, 53 exports, 9 see also oil General Agreement on Tariffs and Trade, 123 Gilbert, M., 192 grants, 177 see also aid Green Revolution, 45-8, 197 Gross Domestic Product: agriculture, 13 China, 123 exports, 28 growth rates:
a d v a n c e d c o u n tr i e s , 1 8 6 -7 ;
education: China, 139, 143 effect on urban growth, 152-3 impact on development, 199-2QQ primary attendance rates, 1 37-40 secondary and tertiary, 14Q-3 UNESCO conference, L40 employment; export sectors, LOS oil production, LLL8 see also labour force, individual sectors energy consumption: as an indicator of development, 83-4. 86 growth, 65 per capita, 247. 249 European Economic Community, effect on world trade, 24* 105 European Free Trade Area, effect on world trade, 94* 105 exports: by commodity, JL Q Q capital equipment, LZ6 China, LQSilQ. from developed countries, 106 and GNP, LOB. value, 93. 95, 97. 247. 242 extractive industry: annual growth rates, advanced countries, 511 employment in, 160-3 expansion, consequences of, 58-62, 198 exports, 54* 205-6 production index, 51-4 relationship with manufacturing, 49-51, 17-6.1 share of developing countries, example commodities, 5.5 transport costs, 54-7, 512 fertility rates, 9* I I Field, R. M., 6* 73_, 1A food production: calorific values, 22-4 supply, dependence on imports,
China, 188-90; developing countries, 183-6 manufacturing industry, 81-3 per capita, 193, 247, L49 proportion attributed to aid, 180 guaranteed export credits, 177, 180 Henry, L., 10 Hogbom, L. 56 Hollister, W. W., 191 imports: by commodity, 100 capital equipment, 175-7 China, 110 foodstuffs, 43 from developed countries, 105 from developing to advanced countries, 106z8 value, 93, 95-7, 247. 249 import substitution, 70* 105, 198 income: changes per capita, 190-3 China, 191-3, 201.-2 and urban growth, 1 52
43i5
Food and Agricultural Organisation, agricultural production index, 16.-2.1 foreign public debt, 181, 247, 2.4.9 foreign trade: agricultural products, 13 capital goods, 105 changes, 92-7
Index
Industrialization and Foreign Trade , 112 infant mortality, 246, 248 International Monetary Fund, aid,
17f t
259
m arket economies, m anufacturing o u tp u t, 62 Mende, T., 178 mineral production, 5JL 61-2 see also extractive industry natural gas, expansion in developed regions, 62 oil: as an irreplaceable asset, 61-2 developing countries share of production, 55 effect on world trade, 94* 96, 210 expansion in advance areas 62 Organization for Economic co-operation and Development, aid, 178 Development Assistance Committee.
122i8Q.
investment, foreign, 12 Kindleberger C., 114, 123 Kirschen, J. S., 192 Klatt, W., 6 Kravis, L B., 122 Kuznets, S., 185-6 labour force: absorption o f rural surplus, 163-5 agricultural, 11, 24 Europe, during industrialization, 18z2 growth of, 118-2 productivity: agricultural, 42; manufacturing, 63 structure, 159-61, 246, 248 land, availability for agriculture, 3 3-4, 41 Lewis, W. A., 114, 124 literacy: effect on econom ic development, 111 levels, 13-5-7and technical progress, 47 loans, 177 Macbean, A. L, 251 m anufactured goods: effect o f demand elasticity on price, 122 indices, 113-16, 120-1 international price changes, 112 m anufacturing industry, em ploym ent, 78-81. 151, 160-1. 163. 246. 248 growth rates, advanced countries, 50. 66-71: China. 72, 76; developing countries, 65-77. levels o f development, 77-89 production index, 64-5. 71. 75 relationship with agriculture, 88-91 relationship with extractive industry, 49-51. 57-61 sectoral changes, 71-2
overpopulation, rural, effect on urbanization, 151 Petroleum, see oil population: growth: advanced countries before industrialization, 8_j annual rates, 6* 67, 69; China, 5-7; effect on agricultural productivity, 17, 38-9; forecasts, 8-9. 11-12, 195; less-developed countries, 6-8; need for restriction, 203-4; urban, 145-8. 155. 200, 246, 248 size, by country, 246, 248 Prebisch, R., 132 Pressat, R., 10 primary products: prices, effect of demand elasticity, 129 effect o f technical progress, 129 e xport index, 123-5, 130 historical comparisons 112, 120-1 indices, 113-16, 120 relationship with transport costs,
112
and supply increases, 13 3 protectionism, 204-5 Raw materials, see primary products
260
Index
Rockefeller Foundation, 45 rural-urban migration, 150-5. 200
Sauerbeck, A., UL5 Schlote, W., 114 services, em ploym ent, 160-1. 246 Snow, E., 22 social structure: and agricultural improvements, 42 restraints, effect on urban growth, 152 state intervention, and economic development, 205-7 steel production, as an indicator of development, 83-4, 86 synthetics: effect on raw material prices, 129-31 share o f market, 121
terms o f trade: causes of deterioration, 128^24 definition, 111 effect o f changes on world trade, 94* 1^9 effect o f production increases, 121 historical comparisons, 112, 120-34 post-war changes, 125-8 transport, as an indicator of development, 84* 86 costs, and primary p roduct prices, 115-19; and m anufactured goods, 118; em ploym ent, 160 underem ploym ent: definition, 166-7 rural, 167-8 urban, 168* 170-1. 200 unem ploym ent: definition, 165-6 rural, 162 urban, 168-70, 200 United Nations Conference on Trade and Development, 127, 132, 180,
202
take off: and agricultural productivity, 13_, 1_8* 39* 41_* agricultural and industrial development, 88* 91 availability o f land, 24 China, 2QQ4 foreign trade, 177, 198-9 G.D.P., 187* 206 level o f literacy, 125 population growth, 8* L95 unem ploym ent, 120 technological progress, and extractive industry, 49-50 and primary p roduct prices, 122 Teheran Agreement, 60
United Nations Organization: aid, 178 index o f export prices, 124 index o f per capita income, 190 indices o f GDP, 183-6 urban inflation, causes, 150-6 urbanization: China, 1 56-7 growth rates, developing countries, 145-7, 149; Europe, 147-9; relationship with industrialization, 151 India, 156-7 Zimmerman, L. J., 18 3-4
First published in 1967, Professor Bairoch's Diagnostic de L'Evolution Economique du Tiers-Monde has gone into four editions, and has brought the author an international reputation. This English translation is, in effect, another edition based on the latest French text but incorporating much w h ich is not to be found there. The statistical tables have been revised and expanded wherever possible to include figures up to the end o f 1 9 7 2 ; the bibliography has been specially adapted to include the literature on the subject in the English language; and tw o new chapters have been w ritte n ; Chapter 8 on 'U rbanization' and Chapter 9 on 'The labour force and em ploym ent'. It has been Professor Bairoch's aim in this book to examine the developm ent of under-developed countries (including China) during the present century and through the use o f comparative statistics to form ulate a diagnosis o f their grow th. His analysis includes, whenever relevant, a comparison between the present econom ic progress of Third W orld countries and that of the developed countries at the time of their 'take off'. Special attention has also been given to China's unique path o f develop ment. In the course of his research the author has elaborated several new series. The production o f these new series and their integration w ith existing data make this book a valuable quantitative econom ic history o f the Third W orld. Paul Bairoch is Professor of Economic History at the University of Geneva.
S o m e r e v i e w s o f t he F r e n c h edi t ion A f u n d a m e n t a l b o o k . . . d e m o g r a p h y , a g ri c u l t u r e , i n d us t ry , f o r e i g n t rade, c a p i t a l f o r m a t i o n , f i n a n c i a l a id a n d e c o n o m i c g r o w t h h a v e b e e n t h e o b j e c t of a n a n a l y s i s t hat w i ll b e i m p o s s i b l e to i gn o r e . Revue Economique H a s filled a g a p . s i n c e t he re h a s b e e n till n o w n o s i n g l e s y n t h e s i s o n t h e g e n e r a l e c o n o m i c d e v e l o p m e n t o f t he Third W o r ld .
Revue des Sciences Economiques

A l t h o u g h t he b o o k is f u n d a m e n t a l l y a g e n e r a l s t u d y t h e a u t h o r s u c c e e d s in b r i n g i n g o u t t h e s p e c i f i c s i t u a t i o n s e x i s t i n g in p ar t i c u l a r c o u n t r i e s . . . . T h i s e x t r e m e l y i n t e r es t i ng b o o k c o n c l u d e s w i t h a n u m b e r o f r e c o m m e n d a t i o n s r e g a r d i n g p o l i c i e s to b e f o l l o w e d t o f ac il i ta te t h e s o l u t i o n o f t he vital p r o b l e m of eco no m ic underdevelopment. A nnals of Public and C ooperative E con om y
University of California Press
Berkeley, California 94720
isb n : 0-520-03554-2

The Economic Development of The Third World Since 1900 - Paul Bairoch

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Economic Development of The Third World Since 1900 - Paul Bairoch

Uploaded by

Copyright:

Available Formats

>ngan hak cipta

THE ECONOMIC DEVELOPMENT OF THE THIRD WORLD SINCE 1900

DK3W -2QH -YDXB

M aterial com direitos autc

The Economic Development o f The Third World Since 1900

Paul Bairoch translated by Cynthia Postan

Bahan dengan hak cipta

Bahan dengan hak cipta

C Recent changes: A green revolution?__________________ 45 3 EXTRAGI1YE.I_ND_ USTRY______________________________ 42

116 123 123 126

Bahan dengan hak cipta

Bahan dengan hak cipta

Material com direitos autorais

List o f Statistical Tables

179 184 191 193 246

Note to the English Edition

China Sri Lanka (Ceylon) India Pakistan Indonesia Philippines Thailand

Egypt Iraq Turkey

Bahan dengan hak cipta

Bahan dengan hak cipta

The Economic Development o f the Third World since 1900

Bahan dengan hak cipt

Bahan dengan hak cipta

Total inc. Oceania

Bahan dengan hak cipta

Bahan dengan hak cipta

The Economic Development o f the Third World since 1900

Bahan dengan hak cipta

ial com direitos autor

The Economic Development o f the Third World since 1900

1763 324 1152 283

2329 430 1517 377

2.8 2.9 2.8 2.9

3036 5 79 1950 500

2.7 3.0 2.5 2.8

3828 770 2397 652

2.3 2.9 2.1 2.7

943 3272 445 7

1.6 2.5 2.1

1091 4127 5438

1.5 2.4 2.0

1231 5059 6494

1.2 2.1 1.8

1763 1763 1763 1763

2272 2329 2375 2370

2.6 2.8 3.0 3.0

2863 3036 3213 3280

2.3 2.7 3.1 3.3

3475 3828 4243 4655

2.0 2.3 2.8 3.6

The Economic Development o f the Third World since 1900

The Economic Development o f the Third World since 1900

1910 1,8004 227 1,090 87 1,7704 1,560 2,700 6,320 360

30 60 550 11 950 600

The Economic Development o f the Third World since 1900

1.7 0.3 1.2 1.0 0.9 - 0 .2 - 1 .0 - 1 .0 * - 0 .9

2.5 2.5 2.5 3.1 2.7 * - 0 .2 - 0 .4 0.3 0.1

2.4 1.9 2.2 2.5 2.1 0.1 - 0 .1 - 0 .3 0.5 *

The Economic Development o f the Third World since 1900

99 783 74 73 75 58 76 71 88 563 51 623 70 91 623

2.0 0.2 0.1 1.2 1.6 3.6 3.2

-1 .0 2.0 0.6 0.1 0.4 0.1 2.3 1.5 3.0