II Chapter 1 II

Developm ent and

Underdeve lopment

Current Interest in Development Other Dimensions of the Development

Economics 3 Gap 28
Academic Interest in Development 4 Unemployment 28
The New International Economic Order 6 The Distribution of Income 31
The Mutual Interdependence of the Growth and Distribution 34
World Economy 7 Nutrition and Health 35
The Meaning of Development and the Poverty, Famine and Entitlements 36
Challenge of Development Economics 9 Food Production 37
The Perpetuation of Underdevelopment 10 Education 46
The Measurement of Poverty and the Basic Needs 51
World Distribution of Income 11 Human Development Index 52
The Development Gap 18 The Stages of Development 54
Per Capita Income as an Index Industrialisation and Growth 56
of Development 22 Will Developing Countries Ever
The Measurement and Comparability of Catch Up? 61
Per Capita Incomes 24 Rostow's Stages of Growth 61

I Current Interest in Development

Economics
ills seem to multiply as the diagnosis deepens.
The political and public concern with the poorer
nations of the world is of equally recent origin.
Current academic interest in development econ- The majority of the national and international
omics, and the study of development economics as bodies to promote development that exist today,
a separate subject, are relatively recent phenom- such as national development banks, the World
ena. For the student today it will be difficult to Bank and its affiliates, and agencies of the United
appreciate that as recently as thirty years ago a Nations, have all been established since the Second
course in development economics was a rare feature World War. Before the war, when most of today's
of an undergraduate programme in economics, poor countries were still colonies, there was very
and that textbooks on economic development were little preoccupation with the economic and social
few and far between. Today no self-respecting problems of the developing (dependent) econ-
department of economics is without a course in omies that we are concerned with today. Perhaps
economic development; there are scores of texts; the facts were not so well known, or perhaps it
hundreds of case studies; and thousands of articles was that the attention of most people was focused
on the subject. And, as in medicine, the perceived on the depression and underemployment in the

3
4 Introduction

developed countries. Whatever the reason for areas of interest of the classical economists. Adam
neglect, the situation today is very different. The Smith, David Ricardo, Thomas Malthus, John
development of the Third World (the collective Stuart Mill and Karl Marx all dealt at some length
name for the developing countries), meaning (with divergent opinions on many issues) with the
above all the eradication of primary poverty, is causes and consequences of economic advance. It
now regarded as one of the greatest social and is entirely natural that thinkers of the day should
economic challenges facing mankind. comment on the contemporary scene, and there is
What accounts for this change in attitude and perhaps an analogy here between the preoccupa-
upsurge of interest in the economics of develop- tion of the classical economists at the time of
ment and in the economies of poor countries? A Britain's industrial revolution and the concern
number of factors can be pinpointed, which inter- of many economists today with the economics of
relate with each other. First, in the wake of the development and world poverty, the nature of
great depression and in the aftermath of war there which has been brought to the attention of the
was a renewed academic interest among pro- world so dramatically in recent decades. Develop-
fessional economists in the growth and develop- ment also represents a challenge equivalent to that
ment process and in the theory and practice of of depression and mass unemployment in the
planning. Second, the poor countries themselves 1930s which attracted so many brilliant minds to
have become increasingly aware of their own economics, Keynes among them. But the nature of
backwardness, which has led to a natural desire the challenge is of course very different. In the case
for more rapid economic progress. The absolute of unemployment in the 1930s, there was an or-
numbers of poor people are considerably greater thodox theory with which to grapple; the task was
now than in the past, which has struck a humani- to formulate a theory to fit the facts and to offer
tarian chord. Third, there has been a growing policy prescriptions. As it turned out, the solution
recognition by all concerned of the mutual inter- to the problem was to be costless: expand demand
dependence of the world economy. The political by creating credit and bring idle resources into
and military ramifications and dangers of a world play. Fancy, an economic problem solved costless-
divided into rich and poor countries are far more ly! The challenge of development is very different.
serious now than they were in the past; at the same There is no divorce between theory and the
time the old Cold War led the major developed observed facts. The mainsprings of growth and
countries to show a growing economic and politi- development are well known: increases in the
cal interest in poor and ideologically uncommitted quantity and quality of resources of all kinds.
nations. The recognition of interdependence has Countries are poor because they lack resources or
been heightened in recent years by fears of short- the willingness and ability to bring them into use.
ages of basic raw materials produced primarily in The problems posed by underdevelopment cannot
Third World countries, and by the rising price of be solved costlessly. It would be reassuring to
oil. think, however, that advances in growth theory,
coupled with more detailed knowledge of the

I Academic Interest in
Development
sources of growth, and the refinement of tech-
niques for planning and resource allocation, have
all increased the possibility of more rapid econ-
omic progress than hitherto. Certainly particular
Academic interest in the mechanics of growth ~nd theoretical models and techniques have been used
development is a renewed interest rather than a extensively in some countries, presumably in this
new preoccupation of economists. The progress belief. For example, models for calculating invest-
and material well-being of men and nations have ment requirements to achieve a target rate of
traditionally been at the centre of economic writ- growth invariably form an integral part of a devel-
ing and enquiry. It constituted one of the major opment plan, and in some countries there have
 Development and Underdevelopment 5

been experiments in recent years with such tech- stage of development one hundred years ago. The
niques as input-output analysis, for the achieve- pool of technology on which to draw, and the
ment of sectoral balance and the avoidance of scope for its assimilation, is enormous. Used with
bottlenecks, and linear programming for the discretion, it must be considered as the main
achievement of efficient resource allocation. means of increasing welfare. The role of trade,
The question is often posed as to what lessons, however, is more problematic. A lot will depend
if any, the present developing countries can draw on how rapidly the developing countries can alter
from the first-hand observations of the classical their industrial structure and on movements in the
writers, or more directly from the development terms of trade. Currently the developing countries
experience of the present advanced nations. One are probably in an inferior position compared
obvious lesson is that while development can be with the present advanced countries at a compar-
regarded as a natural phenomenon, it is also a able stage of their economic history. The dynamic
lengthy process, at least left to itself. It is easy to gains from trade are present but the static ef-
forget that it took Europe the best part of three ficiency gains are less and the terms of trade in
centuries to progress from a subsistence state to most commodities are worse. The gains from trade
economic maturity. Much of development econ- accrue mainly to the rich industrialised countries,
omics is concerned with the time scale of devel- notwithstanding the rapid increase that period-
opment, and how to speed up the process of ically takes place in some commodity prices. The
development without causing problems as acute fact that the gains from trade are unequally dis-
and worrisome as the primary poverty it is desired tributed does not destroy, of course, the potential
to alleviate. In the next millennium, when primary link between trade and growth, or constitute an
poverty in most countries will, it is hoped, have argument against trade. Rather, it represents a
been eradicated, courses in development econ- challenge for altering the structure of trade and the
omics will undoubtedly take a different form. The terms on which it takes place.
emphasis will be on inter-country comparisons, The greater knowledge and acceptance of plan-
rather than on the process of development as such ning may also mean that the development experi-
and the growth pains accompanying the transition ence of the present developing countries will be
from a primarily agrarian to an industrial economy. less protracted and painful than in the past. Plan-
As far as classical theory is concerned, the gloomy ning can potentially mobilise the prerequisites of
prognostication of Ricardo, Malthus and Mill that development more expeditiously than the market
progress will end ultimately in stagnation would mechanism, which takes time to operate, and pro-
seem to be unfounded. It has certainly been con- vided attention is paid to income distribution, the
founded by experience. Population growth and sum total of sacrifice of present generations need
diminishing returns have not been uniformly de- be no more severe per individual. Classical econ-
pressive to the extent that Ricardo and Malthus omists were generally antithetical to interference
supposed. Rising productivity and per capita in- with the market mechanism, believing that the free
comes appear quite compatible with the growth play of market forces would maximise the social
of population and the extension of agriculture. good. But fashions change in economics, and after
Classical development economics greatly under- the Second World War, there was a much greater
estimated the beneficient role of technical progress acceptance of interference with the market mech-
and international trade in the development pro- anism, and planning in developing countries was
cess. It is these two factors above all which seem to seen by many as one of the main means by which
have confounded the pessimism of much of classi- development may be accelerated. The experience
cal theory. With access to superior technology of planning in many countries, however, has not
there is hope, and some evidence, that material been favourable, and planning has come into dis-
progress in today's developing countries will be repute, not the least because of the economic
much more rapid than in countries at a similar disarray of the rigidly planned economies of the
6 Introduction

old Soviet Union and Eastern Europe. It should oped nations, and has been strengthened from
never be forgotten, however, that no country in within by rising expectations as development has
the world made such swift economic advance in proceeded. Development is wanted to provide
such a short space of time as the Soviet Union after people with the basic necessities of life, for their
1918, through the planned allocation of resources own sake, and to provide a degree of self-esteem
which favoured investment at the expense of con- and freedom for people which is precluded by
sumption. The fact that planning may be operated poverty. Wealth and material possessions may not
too rigidly, or for too long and go wrong, should provide greater happiness but they widen the
not be allowed to obscure the fact that it also has choice of individuals, which is an important aspect
merits, and that unfettered free enterprise can also of freedom and welfare. The developing countries
lead to economic disaster and social deprivation. in recent years have shown a marked determina-
What is required in most developing countries is a tion to pull themselves up by theif own bootstraps
judicious mix of public and private enterprise, of assisted, in the words of Professor Hicks, 'by such
the use of markets combined with different types crumbs of aid as the richer countries are willing to
of planning, for the maximisation of social spare, and as they themselves are willing to accept'
welfare. (1966, p. 263). They have also called for a fairer
Planning requires a certain amount of model- deal from the functioning of the world economy
building and this, too, has been inspired by econ- which they view, with some justification, as biased
omists. The most common type of model, which in favour of countries already rich.
forms the basis of much of the model-building that The official call for a New International Econ-
developing countries indulge in, is to calculate the omic Order was originated by the Sixth Special
investment requirements necessary to achieve a Session of the United Nations General Assembly in
target rate of growth of per capita income - com- 1974. The United Nations pledged itself 'to work
monly referred to as a Harrod-Damar model. urgently for the establishment of a new inter-
Neither the models of Harrod (1948) or Domar national economic order based on equity, sovereign
(1947) were designed for the purpose to which equality, common interest and cooperation among
they are now put in developing countries, but their all States, irrespective of their economic and social
growth equations have proved to be an indispens- systems, which shall correct inequalities and re-
able component of macro-economic planning. We dress existing injustices, make it possible to elimin-
shall consider later the strengths and weaknesses ate the widening gap between the developed and
of using this type of aggregate model in develop- the developing countries and ensure steadily accel-
ment planning, and the pros and cons of planning erating economic and social development and
in general. peace and justice for present and future genera-
tions.' The programme of action called for such
things as: improved terms of trade for the exports

I The New International

Economic Order
of poor countries; greater access to the markets of
developed countries for manufactured goods;
greater financial assistance and the alleviation of
past debt; reform of the International Monetary
A second major factor accounting for the upsurge Fund and a greater say in decision making on
of interest in the growth and development process international bodies concerned with trade and
has been the poor nations' own increased aware- development issues; an international food pro-
ness of their inferior economic and political status gramme, and greater technical cooperation.
in the world, and their desire for material im- The call for a New International Economic
provement and greater political recognition through Order has been reiterated several times by various
economic strength. This was precipitated by decol- UN agencies. In 1975, the United Nations Indus-
onisation and by increased contact with the devel- trial Development Organisation (UNIDO) pro-
 Development and Underdevelopment 7

duced the Lima Declaration which set a target for trade and the balance of payments. There exists an
the developing countries to secure a 25 per cent interdependence in the world economy such that
share of world manufacturing production by the the malfunctioning of one set of economies im-
year 2000 compared with the share then of 10 per pairs the functioning of others. This was no more
cent (and a present share of 15 per cent). This evident than in the world economy in the 1980s
target will not be achieved since it requires a which, owing to the rising price of energy and the
growth of manufacturing of 10 per cent per an- debt crisis, displayed mounting economic chaos.
num in developing countries compared with 5 per The 1980 Brandt Report, entitled North-South: A
cent in developed countries, which is 3 per cent Programme for Survival (1980), and its sequel
more than achieved from 1965 to 1990. On the Common Crisis (1983), stressed the mutual be-
monetary front, in 1980 there was the Arusha nefit to all countries from a sustained programme
Declaration which demanded a UN Conference on of development in the Third World, and docu-
International Money and Finance to create a new mented the current adverse trends in the world
international monetary order 'capable of achiev- economy which pointed to a sombre future if not
ing monetary stability, restoring acceptable levels tackled co-operatively: growing poverty and hunger
of employment and sustainable growth' and 'sup- in the Third World; rising unemployment with infla-
portive of a process of global development'. And tion; international monetary disorder; chronic
at the 7th Session ofUNCTAD in Geneva in 1987, balance of payments deficits and mounting debts
policy approaches were called for in four major in most Third World countries; protectionism,
areas: debt and development resources; commodi- and tensions between countries competing for
ties; international trade; and the problems of the energy, food and raw materials. Development
least developed countries. economics addresses itself to many of the issues
contributing to disarray in the world economy.
There is not only a moral case for greater efforts

I The Mutual Interdependence of

the World Economy
to raise living standards in Third World countries,
but a sheer practical case which would be in the
self-interest of the developed countries themselves.
The ability of poor countries to sustain their
The third major factor responsible for the growing growth and development means a greater demand
interest and concern with Third World develop- for the goods and services of developed countries,
ment has been the increasing awareness, particu- which generates output and employment directly
larly on the part of developed countries, that and which also helps to maintain the balance-
dependence is not one-way. The rich countries of-payments stability of these countries which is so
have been compelled out of economic and political crucial if there is to be a reciprocal demand for the
necessity to rethink their economic relations with goods of developing countries. Any constraint on
the poorer nations of the world. On the political demand in the system arising from, say, poor
front the old divide between East and West forced agricultural performance in poor countries, or a
the Western capitalist and the Communist coun- balance-of-payments constraint on demand in de-
tries to compete financially for the favours of large veloped countries, will impair the functioning of
parts of the Third World, and regretfully one of the whole system and reduce the rate of progress
the unfortunate side-effects of the urgent desire of below potential. Herein lies the importance of the
poor countries for improved living standards was transfer of resources to poor countries to maintain
that some allowed themselves to become a politi- their momentum of development (global Keyne-
cal battleground for the great powers in the Cold sianism), and of international monetary reform to
War. smooth the burden of balance-of-payments adjust-
On the economic front, the fortunes of coun- ment and to shift more of the burden of adjust-
tries, rich and poor, are locked together through ment from the deficit to the surplus countries.
8 Introduction

The Brandt Report called for a short-term emerg- economic and political harmony which cannot
ency programme as a prelude to longer-term ac- thrive in a world perpetually divided into rich and
tion, consisting of four major elements: a large poor. There has also been an affirmation by many
scale transfer of resources to developing countries; developed countries of a moral obligation towards
an international energy strategy to minimise the poorer nations. Not all aid and development as-
dislocation caused by sudden and rapid increases sistance is politically inspired. Particularly over the
in the price of oil; a global food programme; and a last three decades, the developed countries have
start on some major reforms in the international shown a genuine humanitarian concern over the
monetary system. Very little has been done. plight of Third World countries, which has
In the longer term, the Brandt Report called for: resulted in the establishment and support of sev-
a twenty-year programme to meet the basic needs eral institutions to assist developing countries, and
of poor countries, involving additional resource which led the period 1960-70 to be named the
transfers of $4 billion a year; a major effort to First Development Decade. We are now in the
improve agricultural productivity to end mass Fourth Development Decade, and the pledge to
hunger and malnutrition; commodity schemes to assist developing countries out of humanitarian
stabilise the terms of trade for primary commodi- concern has been reaffirmed. The goal of a greater
ties; easier access to world markets for the exports degree of income equality between the citizens of a
of developing countries; programmes for energy nation seems to be gaining support, albeit slowly,
conservation; the development of more appropri- as an objective among nations. Moreover, the
ate technologies for poor countries; an inter- propagation of this ideal is not confined to the
national progressive income tax, and levies on supranational institutions that have been es-
trade and arms production, to be used by a new pecially established to further it. Recent years have
World Development Fund (to fund development witnessed the spontaneous creation of several
programmes rather than projects); a link between national pressure groups, in different parts of the
the creation of new international money and aid to world, whose platform is the abolition of world
developing countries, and policies to recycle poverty; and the Church, which remained silent
balance-of-payments surpluses (as accumulated by and inactive for so long, periodically makes its
the Arab oil export countries since 1973, for voice heard. Aid from voluntary agencies to devel-
example) to deficit countries to remove balance- oping countries now amounts to over $3.5 billion
of-payments constraints on demand and to annually. But whatever the motive for concern, the
remove the risk of a slide into international protec- reality of world poverty and underdevelopment
tionism. Many of these issues we shall be discuss- cannot be ignored. Furthermore, primary poverty
ing in the course of this book. 1 Such a programme in developing countries is likely to persist for many
would be of mutual benefit to all parties, rich and years in the future. The economist has a special
poor. It would create investment confidence, responsibility to contribute to an understanding of
which is the crucial ingredient maintaining the the economic difficulties which poor countries face
dynamics of any economic system; it would and to point to possible solutions. This is a text-
stimulate trade and investment, and help the pros- book devoted to that end. Let us start by consider-
pects of sustained growth in the world economy. ing the meaning of development and establishing
It would be wrong to give the impression, how- the magnitude of poverty and of economic div-
ever, that the developed countries' concern with isions in the world as precisely as the data will
world poverty is motivated exclusively by the selfish allow. Then we shall focus on conditions within
realisation that their own survival depends on the developing countries, particularly the employ-
ment situation, the income distribution, the level
1 For a discussion and appraisal of the Brandt Report, see the
of nutrition, and other basic needs.
collection of articles in Third World Quarterly, October 1980,
and Kirkpatrick and Nixson (1981).
 Development and Underdevelopment 9

I The Meaning of Development

and the Challenge of
Development Economics
eradicate the feeling of dominance and depen-
dence which is associated with inferior economic
status.
Freedom refers to freedom from the three evils
of 'want, ignorance and squalor' so that people are
Development implies change, and this is one more able to determine their own destiny. No man
sense in which the term development is used; to is free if he cannot choose; if he is imprisoned by
describe the process of economic and social trans- living on the margin of subsistence with no educa-
formation within countries. This process often tion and no skills. The advantage of material
follows a well-ordered sequence and exhibits com- development is that it expands the range of human
mon characteristics across countries which we choice open to individuals and societies at large.
shall discuss later in the chapter. But if develop- All three of these core components are inter-
ment becomes an objective of policy, the impor- related. Lack of self-esteem and freedom result
tant question arises of developmeqt for what? Not from low levels of life sustenance, and both a lack
so long ago, the concept of development, defined of self-esteem and economic imprisonment be-
in the sense of an objective or a desired state of come links in a circular, self-perpetuating chain of
affairs, was conceived of almost exclusively in poverty by producing a sense of fatalism and
terms of growth targets, with very little regard to acceptance of the established order - the 'accom-
the beneficiaries of growth or to the composition modation to poverty' ~s Galbraith (1980) has
of output. Societies are not indifferent, however, called it.
to the distributional consequences of economic Using Goulet's concept of development, there-
policy; to the type of output that is produced, or to fore, and in answer to the question 'development
the economic environment in which it is produced. for what?', we can say that development has oc-
A concept of development is required which em- curred when there has been an improvement in
braces the major economic and social objectives basic needs, when economic progress has con-
and values that societies strive for. This is not easy. tributed to a greater sense of self-esteem for the
Perhaps the best attempt to date is that by Goulet country and individuals within it, and when ma-
(1971) who distinguishes three basic components terial advancement has expanded the range of
or core values in this wider meaning of develop- choice for individuals. The fact that many of these
ment, which he calls life-sustenance, self-esteem ingredients of development are not measurable
and freedom. does not detract from their importance: the condi-
Life-sustenance is concerned with the provision tion of being developed is as much a state of mind
of basic needs, which we discuss later in the chap- as a physical condition measurable by economic
ter. The basic needs approach to development was indices.
initiated by the World Bank in the 1970s. No The challenge of development economics lies in
country can be regarded as fully developed if it the formulation of economic theory and in the
cannot provide all its people with such basic needs application of policy in order to understand better
as housing, clothing, food and minimal education. and to meet these core components of develop-
A major objective of development must be to raise ment. Clearly the range of issues that development
people out of primary poverty and to provide economics is concerned with is quite distinctive
basic needs simultaneously. and because of this the subject has developed its
Self-esteem is concerned with the feeling of own modus vivendi (ways of doing things),
self-respect and independence. No country can be although drawing liberally on economic theory as
regarded as fully developed if it is exploited by do other branches of economics. If it is to be
others and does not have the power and influence useful, however, a great deal of conventional econ-
to conduct relations on equal terms. Developing omic theory must be adapted to suit the conditions
countries seek development for self-esteem; to prevailing in developing countries, and many of
10 Introduction

the assumptions that underly conventional econ- foremost importance. It was, first of all, settled
omic models have to be abandoned if they are to agriculture that laid the basis for the great civilisa-
yield fruitful insights into the development pro- tions in the past, and it was the increase in agricul-
cess. Static equilibrium theory, for example, is tural productivity in England in the eighteenth
ill-suited for the analysis of growth and change century that laid the basis for, and sustained, the
and of growing inequalities in the distribution of first industrial revolution. If there is one overriding
income between individuals and countries. It is single factor which explains why some countries
probably also true, as Todaro (1989) strongly developed before others, and why some countries
argues, that economics needs to be viewed in the are still backward without a significant industrial
much broader perspective of the overall social sector, it lies in the condition of agriculture which
system of a country, which includes values, beliefs, in the early stages of development is the sector
attitudes to effort and risk taking, religion and the which must provide the purchasing power over
class system, if development mistakes are to be industrial goods.
avoided which stem from implementing policy The conditior. of agriculture depends on many
based on economic theory alone. factors, institutional as well as economic, and
physical conditions are also of key importance.

I The Perpetuation of
Underdevelopment
Climate, particularly, affects the conditions of pro-
duction. Heat debilitates individuals. Extremes
of heat and humidity also deteriorate the quality
of the soil and contribute to low productivity of
The study of economic development helps us to certain crops. It cannot be coincidence that almost
understand the nature and causes of poverty in all developing countries are situated in tropical or
low-income countries, and the transformation of subtropical climatic regions and that development
societies from being primarily rural to being pri- 'took off' in the temperate zones.
marily industrial, with the vast bulk of resources The condition of agriculture has not been helped
utilised in industrial activities and in service activi- by what Lipton (1977) has called urban bias
ties which serve the industrial sector. But why which has in many countries starved agriculture of
have some countries never participated in this resources. This has happened because ruling elites
process or got left behind? The first industrial generally originate from, or identify with, the
revolution gave the present developed countries an non-rural environment, and because policy-makers
initial advantage which they then sustained were led astray both by the empirical evidence
through the existence of various cumulative forces which shows a high correlation between levels of
at work against those left behind. In the last forty development and industrialisation, and by early
years there has been a second industrial revolution development models which stressed investment in
which has propelled another bloc of countries (the industry.
so-called newly industrialised countries of South Many other internal conditions have acted as
East Asia and Latin America) into a virtually barriers to progress in poor countries, which in-
industrialised state, and many others into a semi- teracted in a vicious circle. In some countries
industrialised state. But many countries are still population size presents a problem combined with
left behind in a semi-feudal state, including the low levels of human capital formation. The latter
very poorest which have now become the prime in turn perpetuates poverty which is associated
focus of concern of the Wodd Bank and other with high birth rates and large family size. This is a
development agencies. form of 'accommodation to poverty' (Galbraith,
There are many theories of the perpetuation (1980)) which then perpetuates low living stan-
of underdevelopment but none seem to have dards in a circular process. In other countries there
universal validity. The state of agriculture is of may not exist the psychological conditions re-
 Development and Underdevelopment 11

quired for modernisation built on individualism

and the competitive spirit, coupled with a strong
work ethic, rationalism and scientific thought,
which characterised the industrial revolutions of
eighteenth- and nineteenth-century Europe and
I The Measurement of Poverty
and the World Distribution of
Income
which have played a large part in the emergence of By any standard one cares to take, the evidence is
the newly industrialised South East Asian coun- unequivocal that the world's income is distributed
tries in the latter half of the twentieth century. extremely unequally between nations and people,
External relations between countries also play a and that there exists in the world a broad north-
part in the poverty perpetuation process, which south divide into rich and poor countries. The
have given rise to structuralist and dependency World Bank classifies the countries of the world
theories of underdevelopment. It seems to be the into three broad categories: low-income countries;
general lesson of history that once one set of middle-income countries, and high-income coun-
countries gains an economic advantage, the ad- tries. This classification for 1992 is given in Table
vantage will be sustained through a process of 1.2. In later discussion, it is largely the low-income
what Myrdal (1957) has called 'circular and and middle-income countries that we shall refer to
cumulative causation', working through the media as the developing countries, and the high-income
of factor mobility and trade. (For a full discussion, countries as the developed countries. There are
see Chapter 5.) Favoured regions denude the back- several dimensions of the 'development gap', but
ward regions of capital and skilled labour, and focusing for the moment on income per capita and
they trade in commodities whose characteristics ignoring measurement difficulties, we see from
guarantee that the gains from trade accrue to column 3 that 37 countries are classified as low-
them. Colonialism was an extreme form of depen- income in a state of primary poverty with a
dency, which did exploit many countries that are weighted average level of income per head of only
still poor today. On the other hand there are a US $350 per annum. At the other end of the
number of countries, such as Ethiopia and Thai- spectrum there are 24 high-income countries en-
land, that were never colonised, which are equally joying an average annual per capita income of
backward. Dependence takes more subtle forms, close to US $20,000. This gives some idea of the
however, based on the international division of range of income differences.
labour, for example, which then leads to unequal
exchange relations between rich and poor with the
poor dependent on the rich for capital and tech- D Poverty
nical progress to equip their industrial sectors. The
current indebtedness of the less developed coun- The World Bank defines poverty as the inability of
tries, the 'increasing price' that poor countries people to attain a minimum standard of living. 1
have to pay for development inputs relative to the This definition gives rise to three questions. How
price they receive for their exports and the grow- do we measure the standard of living? What is
ing number of poor people are manifestations of meant by a minimum standard of living? How can
this dependency. There are exceptions to the thesis we express the overall extent of poverty in a single
of 'circular and cumulative causation', but it re- measure?
quires in most cases a strong exogenous shock to The most obvious measure of living standards is
break out of a vicious circle of poverty and depen- an individual's (or household's) real mcome or
dency. We take up some of these issues later in
Part III of the book. Let us now turn our attention 1 The 1990 World Development Report published by the
to the magnitude of poverty in developing coun- World Bank was devoted to a consideration of the measure-
tries and to the world distribution of income. ment, magnitude and nature of poverty in the Third World.
12 Introduction

Table 1.1 The Extent of Poverty in Developing Countries

Social indicators
Extremely poor Poor (including extremely poor)
Under 5 Net primary
Headcount Headcount mortality Life enrollment
Number index Poverty Number index Poverty (per expectancy rate
Region (millions) (percent) gap (millions) (percent) gap thousand) (years) (percent)

Sub-Saharan Africa 120 30 4 180 47 11 196 50 56

East Asia 120 9 0.4 280 20 1 96 67 96
China 80 8 1 210 20 3 58 69 93
South Asia 300 29 3 520 51 10 172 56 74
India 250 33 4 420 55 12 199 57 81
Eastern Europe 3 4 0.2 6 8 0.5 23 71 90
Middle East and North
Africa 40 21 1 60 31 2 148 61 75
Latin America and the
Caribbean 50 . 12 1 70 19 1 75 66 92
All developing
countries 633 18 1 1116 33 3 121 62 83

Source: World Development Report 1990.

expenditure (with an allowance made for output regarded as an acceptable standard of living in the
produced for own consumption). The same level United Kingdom will be different (and higher)
of real income and expenditure in different coun- from that regarded as acceptable in Nigeria. In
tries, however, may be associated with different practice, however, for the measurement of poverty
levels of nutrition, life expectancy, infant mor- in the Third World, the World Bank takes just two
tality, levels of schooling etc. which must be con- figures for per capita income: one to classify the
sidered as an integral part of 'the standard of total poor, the other to measure the extremely
living'. Measures of living standards based on per poor. In 1990, the figures taken were $370 per
capita income, therefore, may need to be supple- annum and $275 per annum, respectively.
mented by other measures which include these Once the poverty line has been calculated, the
other variables. We discuss later the attempt by simplest way to measure poverty is by the head
the United Nations Development Programme count index which simply adds up the number of
(UNDP) to construct a Human Development In- people who fall below the poverty line (sometimes
dex which takes some of these factors into account. expressed as a proportion of the population). By
To separate the poor from the not so poor, an this measure, the W odd Bank calculated that the
arbitrary per capita income figure has to be taken number of poor people in the developing countries
which is sufficient to provide a minimum accept- is just over one billion, and the number of ex-
able level of consumption. In theory, a consump- tremely poor is just over 600 million. The numbers
tion-based poverty line can be thought of as by continent are shown in Table 1.1.
comprising two elements: firstly an objective One weakness of the headcount index, how-
measure of expenditure necessary to buy a mini- ever, is that it ignores the extent to which the poor
mum level of nutrition, and secondly a subjective fall below the poverty line, so that crude compari-
additional amount that varies from country to sons between countries, or over time, may be
country reflecting the cost of individuals partici- misleading. To overcome this weakness, the con-
pating in the everyday life of society. What is cept of the poverty gap may be used which
 Development and Underdevelopment 13

measures the transfer of income required to bring Figure 1.1

the income of every poor person up to the poverty
line. In Table 1.1, it is measured as the aggregate 100~--------------------------------~

income shortfall of the poor as a percentage of 90

aggregate consumption.
Both the headcount index and the poverty gap 80
index are insensitive to the extent of inequality
70
amongst the poor. For example, if income is trans- Q)

ferred from a poor person to someone who is even E

8 60
poorer, neither of the measures changes, but ·=0 50
poverty will be more evenly spread, and represent
E
a social improvement in some sense. This con- Q)

" 40 -
sideration has implications for judging the claims ~
a.
of public policy because clearly if the headcount 30
index is taken as the measure of poverty, the
20
easiest way to reduce the head count index would
be to focus all attention on those just below the 10
poverty line, but this would not be socially just.
We need measures of poverty which also take 0 10 20 30 40 50 60 70 80 90 100
account of movements in the distribution of in- Per cent of population
come between the poor.
It is interesting to note that despite the massive
numbers of people in absolute poverty, the trans- tion and receive only 6 per cent of the world's
fer needed to leave everybody above the poverty income; the middle-income countries contain 15 per
line ($370) is relatively small- only 3 per cent of cent of the world's population and receive 17 per
total consumption in developing countries. cent of world income, and other rich industrialised
The focus of the World Bank is now very much countries contain 25 per cent of the world's popu-
on poverty eradication. In May 1992, the Presi- lation yet receive 77 per cent of world income.
dent of the World Bank, Mr Lewis Preston, Income distribution data of this type (for indi-
declared that poverty reduction will be 'the bench- vidual countries as well as for groups of countries)
mark by which our performances as a develop- can be represented graphically on a so-called
ment institution will be measured'. This represents Lorenz curve diagram as shown in Figure 1.1. The
a shift of emphasis away from the Bank's tra- 45° line represents a perfectly equal distribution of
ditional emphasis on project lending for economic income across the population. The bowed curve is
efficiency to secure an economic rate of return the Lorenz curve showing graphically the degree
from projects which may have done nothing to of inequality. To draw the curve, first rank coun-
improve the lot of the poor. tries or groups of countries in ascending order
according to the ratio of the percentage of income
they receive in relation to the percentage popula-
D Distribution of World Income tion they contain; then cumulate the observations,
and plot on the diagram. Taking the data given
Turning now to the consideration of the distribu- above, our ranking is obviously low-income,
tion of world income in relation to the population, middle-income and 'high' -income. The cumulative
and using the three-fold classification of low- distribution of income is 6/60 then 23/75 when the
income, middle-income and industrial countries, middle-income country figures are added to the
we find that the low-income countries contain first observation figures for the low-income coun-
approximately 60 per cent of the world's popula- tries, and 100/100 when the 'high' -income figures
14 Introduction

are added. Plotting these distributions gives the prices. If we regard $1200 as only barely above
Lorenz curve shown. If historical data are avail- subsistence, the major part of present income
able, changes in the distribution of income disparities between developed and developing
through time can be shown. It is possible, how- countries must have arisen over the last century.
ever, that two (or more) Lorenz curves may cross, Some countries, through a combination of fortune
precluding a definite conclusion on whether the and design, have managed to grow much faster
distribution has narrowed or widened from a than others. The overriding influence has been
visual inspection of the curves alone. In this case a industrialisation and the technological progress
more precise measure of distribution is required. associated with it. The close association between
One measure is to express the area enclosed be- industrialisation and living standards spells out
tween the Lorenz curve and the 45° line as a ratio the clear policy message that to base a develop-
of the total area under the 45° line. This is the Gini ment policy on agricultural activities alone would
coefficient of distribution which varies from 0 be misguided, however attractive such aphorisms
(complete equality) to 1 (complete inequality). as 'back to the land' and 'small is beautiful' may
Calculation of this ratio for the world economy sound to those disillusioned with the recent in-
would give a Gini coefficient of approximately dustrialisation experience of the developing coun-
0.6. tries. Sutcliffe (1971) is right when.he argues:
There is little evidence that the distribution of
world income has been narrowing over time. Ac- It is understandable that vague memories of the
cording to Kuznets (1965, pp. 142-75), if Lorenz oppression of the working class in 19th century
curves are plotted for the years 1894, 1938 and Britain, the contemporary horrors of American
1949 there is no indication that the curves have machine-age society, and the Stalinist attack on
been shifting closer to the 45° line. Andie and the Russian peasantry, should arouse feelings
Peacock (1961) reach the same conclusion in a which are hostile to industrialisation. Yet to
comparison of 1949 and 1957. Their estimate of oppose machines altogether, like Gandhi, or to
the Gini ratio, taking 62 countries, is approxi- argue that a long run rise in the standard of
mately the same for both years (0.637 and 0.636, living is possible without industrialisation,
respectively). What this suggests is that while per are no more than forms of sentimentalism, es-
capita income may have been growing in the low- pecially when the condition of most of the
income countries, it must have been growing population of the non-industrialised world is
almost as fast in the high-income countries. This is now both terrible and worsening. It is not senti-
certainly true of the period since 1965 as column 4 mentalism to demand that the process of in-
in Table 1.2 indicates. Average per capita in- dustrialisation should be made as humane and
come growth in the developing countries has been as painless as possible and that the long term
2.5 per cent, and 2.4 per cent in the developed aims of equality at a higher standard of living
countries. should be constantly borne in mind as the pro-
It is easily forgotten that the rich-poor country cess goes on.
divide is a relatively recent phenomenon. All coun-
tries were once at subsistence level, and as recently The concentrated impact of industrialisation on
as 200 years ago, at the advent of the British living standards in the Western world is dramati-
industrial revolution, absolute differences in living cally emphasised by Patel's illustration (Patel
standards between countries on average cannot (1964)) that if 6000 years of man's 'civilised'
have been great. The average per capita income of existence prior to 1850 is viewed as a day, the last
the developing countries today is approximately century or so represents little more than half an
$1200 per annum and this was about the average hour; yet in this 'half-hour' more real output has
level of real per capita income in Western Europe been produced in the developed countries than in
in the mid-nineteenth century measured at current the preceding period. It is true .that living stan-
 Development and Underdevelopment 15

Table 1.2 Basic Indicators

GNP per capita

Life
Average annual Average annual expectancy
Area
growth rate rate of inflation birth
Population (thousands
Dollars (percent) (percent) (years)
(millions) of square
mid-1990 kilometers) 1990 1965-90 1965-80 1980--90 1990

Low-income economies 3 058.3 t 37 780 t 350w 2.9 w 8.0w 9.6 w 62 w

China and India 1 983.2 t 12 849 t 360w 3.7w 3.2w 6.8 w 65 w
Other low-income 1 075.1 t 24 931 t 320w 1.7 w 17.3 w 15.1 w 55 w

Mozambique 15.7 802 80 .. .. 36.6 47

Tanzania 24.5 945 110 --0.2 9.6 25.8 48
Ethiopia 51.2 1222 120 --0.2 3.4 2.1 48
Somalia 7.8 638 120 --0.1 10.2 49.7 48
Nepal 18.9 141 170 0.5 7.8 9.1 52

Chad 5.7 1284 190 -1.1 6.2 1.2 47

Bhutan 1.4 47 190 .. .. 8.4 49
Lao PDR 4.1 237 200 .. .. .. 49
Malawi 8.5 118 200 0.9 7.4 14.7 46
Bangladesh 106.7 144 210 0.7 15.9 9.6 52

Burundi 5.4 28 210 3.4 5.0 4.2 47

Zaire 37.3 2 345 220 -2.2 24.7 60.9 52
Uganda 16.3 236 220 -2.4 21.4 107.0 47
Madagascar 11.7 587 230 -1.9 7.7 17.1 51
Sierra Leone 4.1 72 240 0.0 7.9 56.1 42

Mali 8.5 1240 270 1.7 9.0 3.0 48

Nigeria 115.5 924 290 0.1 14.6 17.7 52
Niger 7.7 1267 310 -2.4 7.5 2.9 45
Rwanda 7.1 26 310 1.0 12.5 3.8 48
Burkina Faso 9.0 274 330 1.3 6.3 4.5 48

India 849.5 3 288 350 1.9 7.5 7.9 59

Benin 4.7 113 360 --0.1 7.4 1.9 50
China 1 133.7 9 561 370 5.8 --0.3 5.8 70
Haiti 6.5 28 370 0.2 7.3 7.2 54
Kenya 24.2 580 370 1.9 7.2 9.2 59

Pakistan 112.4 796 380 2.5 10.3 6.7 56

Ghana 14.9 239 390 -1.4 22.9 42.5 55
Central African Rep. 3.0 623 390 --0.5 8.2 5.4 49
Togo 3.6 57 410 --0.1 7.1 4.8 54
Zambia 8.1 753 420 -1.9 6.3 42.2 50

Guinea 5.7 246 440 .. .. .. 43

Sri Lanka 17.0 66 470 2.9 9.4 11.1 71
Mauritania 2.0 1026 500 --0.6 7.6 9.0 47
Lesotho 1.8 30 530 4.9 6.7 12.7 56
Indonesia 178.2 1 905 570 4.5 35.5 8.4 62

Honduras 5.1 112 590 0.5 5.7 5.4 65

Egypt, Arab Rep. 52.1 1 001 600 4.1 6.4 11.8 60
16 Introduction

Table 1.2 Basic Indicators (continued}

GNP per capita

Life
Average annual Average annual expectancy
Area
growth rate rate of inflation birth
Population (thousands
Dollars (percent) (percent) (years)
(millions) of square
mid-1990 kilometers) 1990 1965-90 1965-80 1980--90 1990
Middle-income economies 1 087.5 t 41139 t 2220w 2.2 w 21.1 w 85.6 w 66w
Lower-middle-income 629.1 t 22 432 t 1530 w 1.5 w 23.6w 64.8 w 65 w
Bolivia 7.2 1 099 630 -0.7 15.9 317.9 60
Zimbabwe 9.8 391 640 0.7 5.8 10.8 61
Senegal 7.4 197 710 -0.6 6.3 6.7 47
Philippines 61.5 300 730 1.3 11.4 14.9 64
Cote d'Ivoire 11.9 322 750 0.5 9.4 2.3 55
Dominican Rep. 7.1 49 830 2.3 6.7 21.8 67
Papua New Guinea 3.9 463 860 0.1 8.1 5.3 55
Guatemala 9.2 109 900 0.7 7.1 14.6 63
Morocco 25.1 447 950 2.3 7.0 7.2 62
Cameroon 11.7 475 960 3.0 9.0 5.6 57
Ecuador 10.3 284 980 2.8 10.9 36.6 66
Syrian Arab Rep. 12.4 185 1000 2.9 7.9 14.6 66
Congo 2.3 342 1010 3.1 6.8 0.5 53
El Salvador 5.2 21 1110 -0.4 7.0 17.2 64
Paraguay 4.3 407 1110 4.6 9.3 24.4 67
Peru 21.7 1 285 1160 -0.2 20.6 233.9 63
Jordan 3.2 89 1240 .. .. .. 67
Colombia 32.3 1139 1260 2.3 17.5 24.8 69
Thailand 55.8 513 1420 4.4 6.2 3.4 66
Tunisia 8.1 164 1440 3.2 6.7 7.4 67
Jamaica 2.4 11 1 500 -1.3 12.8 18.3 73
Turkey 56.1 779 1 630 2.6 20.8 43.2 67
Romania 23.2 238 1640 .. .. 1.8 70
Poland 38.2 313 1 690 .. .. 54.3 71
Panama 2.4 77 1 830 1.4 5.4 2.3 73
Costa Rica 2.8 51 1 900 1.4 11.2 23.5 75
Chile 13.2 757 1 940 0.4 129.9 20.5 72
Botswana 1.3 582 2 040 8.4 8.4 12.0 67
Algeria 25.1 2 382 2 060 2.1 10.9 6.6 65
Bulgaria 8.8 111 2 250 .. .. 2.2 73
Mauritius 1.1 2 2 250 3.2 11.8 8.8 70
Malaysia 17.9 330 2 320 4.0 4.9 1.6 70
Argentina 32.3 2 767 2 370 -0.3 78.4 395.2 71

Iran, Islamic Rep. 55.8 1 648 2 490 0.1 15.5 13.5 63

Nicaragua 3.9 130 .. -3.3 8.9 432.3 65
Upper-middle-income 458.4 t 18 706 t 3 410 w 2.8 w 19.3 w 102.1 w 68 w
Mexico 86.2 1 958 2 490 2.8 13.0 70.3 70
 Development and Underdevelopment 17

Table 1.2 Basic Indicators (continued)

GNP per capita

Life
Average annual Average annual expectancy
Area
growth rate rate of inflation birth
Population (thousands
Dollars (percent) (percent) (years)
(millions) of square
mid-1990 kilometers) 1990 1965-90 1965-80 1980-90 1990

South Africa 35.9 1221 2 530 1.3 10.3 14.4 62

Venezuela 19.7 912 2 560 -1.0 10.4 19.3 70
Uruguay 3.1 177 2 560 0.8 58.2 61.4 73
Brazil 150.4 8 512 2 680 3.3 31.3 284.3 66
Hungary 10.6 93 2 780 .. 2.6 9.0 71
Yugoslavia 23.8 256 3 060 2.9 15.2 122.9 72
Czechoslovakia 15.7 128 3 140 .. .. 1.9 72
Gabon 1.1 268 3 330 0.9 12.8 -1.7 53
Trinidad and Tobago 1.2 5 3 610 0.0 13.7 6.4 71
Portugal 10.4 92 4 900 3.0 11.7 18.1 75
Korea, Rep. 42.8 99 5 400 7.1 18.4 5.1 71
Greece 10.1 132 5 990 2.8 10.3 18.0 77
Saudi Arabia 14.9 2 150 7 050 2.6 17.9 -4.2 64
Libya 4.5 1 760 .. 3.0 15.4 -0.2 62
Oman 1.6 212 .. 6.4 19.9 .. 66
Low- and middle-income 4 158.8 t 78 919 t 840 w 2.5 w 16.7 w 61.8 w 63 w
Sub-Saharan Africa 495.2 t 23 066 t 340 w 0.2 w 11.4 w 20.0 w 51 w
East Asia & Pacific 1577.2 t 15 572 t 600w 5.3 w 9.3 w 6.0w 68 w
South Asia 1147.7 t 5 158 t 330w 1.9 w 8.3 w 8.0w 58 w
Europe 200.3 t 2171 t 2400w .. 13.9w 38.8 w 70 w
Middle East & N. Africa 256.4 t 11 334 t 1790 w 1.8 w 13.6w 7.5 w 61 w
Latin America & Caribbean 433.1 t 20 397 t 2180 w 1.8 w 31.4 w 192.1 w 68 w
Other economies 320.9 t 22 634 t .. .. .. .. 71w
Severely indebted 455.2 t 21 048 t 2140 w 2.1 w 27.4 w 173.5 w 67w
High-income economies 816.4 t 31 790 t 19 590 w 2.4 w 7.7w 4.5 w 77w
OECD members 776.8 t 31 243 t 20170 w 2.4 w 7.6 w 4.2 w 77w
tOther 39.6 t 547 t .. .. 13.8 w 26.1 w 75 w

Ireland 3.5 70 9 550 3.0 11.9 6.5 74

Israel 4.7 21 10 920 2.6 25.2 101.4 76
Spain 39.0 505 11 020 2.4 12.3 9.2 76
Singapore 3.0 1 11 160 6.5 5.1 1.7 74
Hong Kong 5.8 1 11490 6.2 8.1 7.2 78
New Zealand 3.4 269 12 680 1.1 10.3 10.5 75
Belgium 10.0 31 15 540 2.6 6.6 4.4 76
United Kingdom 57.4 245 16 100 2.0 11.2 5.8 76
Italy 57.7 301 16 830 3.0 11.3 9.9 77
Australia 17.1 7 687 17 000 1.9 9.5 7.4 77

Netherlands 14.9 37 17 320 1.8 7.5 1.9 77

Austria 7.7 84 19 060 2.9 5.8 3.6 76
18 Introduction

Table 1.2 Basic Indicators (continued)

GNP per capita

Life
Average annual Average annual expectancy
Area
growth rate rate of inflation birth
Population (thousands
Dollars (percent) (percent) (years)
(millions) of square
mid-1990 kilometers) 1990 1965-90 1965-80 1980-90 1990
France 56.4 552 19 490 2.4 8.4 6.1 77
United Arab Emirates 1.6 84 19 860 0 0 0 0 1.1 72
Canada 26.5 9 976 20 470 2.7 7.1 4.4 77
United States 250.0 9 373 21790 1.7 6.5 3.7 76
Denmark 5.1 43 22 080 2.1 9.3 5.6 75
Germany 79.5 357 22 320 2.4 5.2 2.7 76
Norway 4.2 324 23 120 3.4 7.7 5.5 77
Sweden 8.6 457 23 660 1.9 8.0 7.4 78
Japan 123.5 378 25 430 4.1 7.7 1.5 79
Finland 5.0 338 26 040 3.2 10.5 6.8 76
Switzerland 6.7 41 32 680 1.4 5.3 3.7 78
Kuwait 2.1 18 0 0 -4.0 15.9 -2.7 74
World 5 283.9 t 133 342 t 4200 w 1.5 w 9.2 w 14.7 w 66w

t means total.
w means weighted average.
Source: World Development Report, 1992.

dards in the developing countries have been rising biguous. Since living standards in all countries
faster since 1950 than at any time in the past; but tend to rise absolutely over time, it obviously
so, too, have the living standards in the developed refers to the comparative position of poor coun-
countries, and the gap between rich and poor tries, but is the comparative position being measured
countries continues to widen. Although develop- taking absolute or relative differences in per capita
ment consists of more than a rise in per capita income? How should the 'development gap' be
incomes, income disparities are the essence of the assessed? Unfortunately there is no easy answer to
so-called 'development gap'. Let us examine the this question, yet the answer given has a profound
nature and magnitude of this gap more closely. bearing on the growth of per capita income that
poor countries must achieve either to prevent a
deterioration of their present comparative position
• The Development Gap or for an improvement to be registered. Relative
differences will narrow as long as the per capita
The statement that 'the rich countries get richer income growth rate of the developing countries
and the poor countries get poorer' has become a exceeds that of the developed countries; and this
popular cliche in the literature on world poverty, excess of growth is a precondition for absolute
but without ro.uch discussion of the facts or the differences to narrow and disappear in the long
precise magnitude of the development task facing run. In the short run, however, a narrowing of
the developing countries if the per capita income relative differences may go hand in hand with a
gap between rich and poor nations is to be nar- widening absolute difference, given a wide ab-
rowed. Indeed, the statement itself is not unam- solute gap to start with, and thus the rate of
 Development and Underdevelopment 19

growth necessary to keep the absolute per capita tries, the top 20 per cent of the richest people in
income gap from widening is likely to be substan- the rich countries had incomes 150 times higher
tially greater than that required to keep the rela- than the incomes of the poorest 20 per cent of
tive gap the same. But suppose the relative gap people in poor countries.
does narrow, and the absolute gap widens, are the Now let us turn to the future comparative posi-
poor countries comparatively better or worse off? tion of the poor countries, and the magnitude of
There is a tendency in economics to measure the development task as far as more equitable
phenomena, especially dispersions of income, in world living standards are concerned. To avoid
relative rather than absolute terms - to compare the issue of whether strategy and assessment
differences in the rates of change of variables, as should be concerned with absolute or relative per
with Lorenz curves, rather than absolute differ- capita income differences, and to facilitate quanti-
ences. In comparing rich and poor countries, how- fication, matters will be made simple by assuming
ever, it is not difficult to argue that even if a that the desirable goal is to narrow and eliminate
relative per capita income gap is narrowed, the both the absolute and the relative gap. We shall
comparative position of the poor may have take as a target the average per capita income of
worsened because the absolute gap has widened. the industrialised countries and attempt to answer
Take for illustration the case of the average Indian four specific questions as reliably as the data will
living on the equivalent of $350 per annum com- allow:
pared with the average American living on ap-
1 Given the recent growth experience of the
proximately $22 000. Suppose the Indian's income
poor countries, how long would it take for
rises by 20 per cent and the American's income by
them to reach the current average level of per
10 per cent. The Indian is now relatively better off,
capita income in the industrialised countries?
but is he not comparatively worse off? The Ameri-
2 Given the recent growth experience of the
can's increased command over goods and services
poor countries relative to the industrialised
(i.e. 10 per cent of $22 000) far exceeds that of the
countries, how many years would it take for
Indian (i.e. 20 per cent of $350), and unless mar-
the per capita income gap to be eliminated?
ginal utilities differ radically, divergences in total
3 Given the rate of growth of the industrialised
utility and welfare will widen in favour of the
countries from now until the year 2010 (say),
American. On welfare grounds there would seem
how fast would the poor countries have to
to be a case for paying as much attention to
grow for per capita incomes to be equalised by
absolute differences in per capita income between
that date?
rich and poor countries as to rates of growth of
4 Given the rate of growth of the industrialised
per capita income.
countries, how fast would the poor countries
As it happens, both the absolute and the relative
have to grow merely to prevent the absolute
gap between rich and poor countries has widened
per capita income gap between rich and poor
in the last thirty years. Research by Dowrick
countries from being any wider in the year
(1992) shows that between 1960/64 and 1984/88,
2010 than now?
average per capita income growth in the rich coun-
tries was 2.49 per cent per annum; in the middle- By asking the first two questions some idea can
income countries 2.16 per cent per annum, and in be obtained of the time scale of the catching up
the poor countries 1.36 per cent per annum. There process by the poor countries given their recent
is no evidence of living standards converging in the growth performance. The answers to the latter
world economy. In 1960, the richest 20 per cent of two questions give some idea of the growth task
the world's population had incomes 30 times greater facing the poor countries in their struggle not only
than the poorest 20 per cent of the world's popula- for parity of living standards, but also in simply
tion. In 1990, this ratio was 60 times greater. preventing the absolute gap from widening.
Taking account of income inequality within coun- Given the basic data, the answers to the ques-
20 Introduction

tions posed involve little more than simple ma- The solution to the second question is obtained
nipulation of the formula for compound interest: from the expression:

s= p (1 + r)n

where P is the principal sum and S is the sum to from which we can find how long it would take (n)
which the principal grows at an annual rate of for the per capita income gap to be eliminated
growth, r, over n years. For illustration: between rich and poor countries, as long as the
Let YDt be the current level of per capita in- rate of per capita income growth in poor countries
come in the industrialised countries = $20 000; is greater than in the industrialised countries -
Y Dct be the current level of per capita in- otherwise, of course, the absolute gap would
come in the poor countries = $1200; widen for ever.
rDbe the per capita income growth rate in
the industrialised countries from 1990 to the year 1o gYDt
2010 = 3 per cent (say); --
YDct
rDe be the actual per capita income growth n = ------~~----
log(1 + rDc) - log(1 + rD)
rate in the poor countries = 2 per cent;
Y~ be the assumed level of per capita in-
A calculation can be made for any individual
come in the industrialised countries in the year
country whose average per capita income growth
2010 = $36 000 at today's prices (assuming 3 per
was in excess of that of the industrialised
cent growth);
countries. Korea, for example, has a per capita
and r~c be the required per capita growth
income of approximately $5000 and per capita
rate of the poor countries.
income has been growing at approximately 7 per
The solution to the first question is then obtained
cent per annum. How long would it take Korea at
from the formula:
this rate to catch up the present industrialised
countries growing at 3 per cent? The answer is:

from which 1 20 000

og 5000
n = = 36 years
log(1.07) - log(1.03)

n=------ This is a relatively short space of time, but clearly

log (1 + rDc) the lower the initial level of per capita income and
the smaller the excess of growth above 3 per cent,
Applying the assumed values above gives: the longer the time it would take to catch up. For a
country starting with an average level of per capita
1 20 000
. income of $1000 and growing at 4 per cent, it
og 1200 would take 309 years!
n= = 95 years The solution to the third question is obtained
log (1.02) from the expression:

In other words, at a growth rate of 2 per cent, it

would take the average poor country, with a per
capita income of $1200, 95 years to reach the where Y~ is the assumed level of per capita income
current living standards enjoyed in the industrial- in the industrialised countries in the year 2010.
ised countries. We can then solve for the required growth rate of
 Development and Underdevelopment 21

the poor countries between the base period (1990) the assumed 3 per cent average growth rate of the
and the year 2010(n = 20) to equalise per capita industrialised countries.
incomes. All the above calculations are sensitive to the
assumed future growth rate of the industrialised
countries, the choice of the target year in the fu-
ture, and the base year level of per capita income
Applying the assumed magnitudes of the variables taken for the poor countries. No one can possibly
gtves: know with precision what the future rate of ad-
vance of the industrialised countries will be, and 3
r';;c = 20Y(36 000/1200) - 1 per cent per capita income growth - the historical
average from 1950 to 1990- would seem to be as
= 18 per cent
reasonable an assumption as any. But the lower
The required growth rate is 18 per cent per annum the growth rate, the less formidable the growth
and hardly feasible. The magnitude of the develop- effort of the poor countries to achieve parity of
ment task is clearly colossal if defined in terms of living standards. No special significance should be
achieving roughly comparable living standards attached to the year 2010 as the choice of target
throughout the world by the beginning of the next year. Some year has to be taken to make these
century. For most of the poor countries per capita 'catching up' calculations - not too close to the
income growth would have to increase six-fold, present to give no hope and not too far away for
necessitating a ratio of investment to national in- the goal to be lost sight of. As far as the base year
come of 50 per cent or more. Investment ratios of level of per capita income in the poor countries is
this order are simply not feasible, and in any case concerned, a note of caution is in order. To the
the countries themselves could not absorb such extent that income statistics invariably understate
investment. the value of production in poor countries (see
The solution to the fourth question is obtained later), the calculations of the catching-up time and
from the expression: the growth rate required for parity of living stan-
dards will be exaggerated. The degree of over-
estimation, however, is not likely to be so great as
to invalidate the conclusion that the growth rates
where the left-hand side represents the base level required for parity of living standards by the year
per capita income gap and n is 20 years. Solving 2010 are not feasible, and that on current growth
for the growth rate that would have to be achieved performance some countries will never catch up.
to prevent the present gap from widening gives: It can be argued, of course, that world income
equality is an impracticable ideal, and that the
primary aim is not equality of living standards
throughout the world but 'tolerable' living stan-
dards in all countries, which is a very different
Applying the assumed magnitudes of the variables
matter. The problem is to define 'tolerable' living
gtves:
standards, and especially to guarantee a reason-
ably equitable distribution of that average level of
r0 c = 20Y[36 000 - (20 000 - 1200)]/1200
real income. The time scale involved to reach
- 1 = 14 per cent 'tolerable' living standards is clearly less than that
required to eliminate the gap entirely, but even so,
This again is a growth rate not feasible, with the if the average level of per capita income now en-
implication that the per capita income gap be- joyed in the industrial countries is regarded as the
tween rich and poor countries will almost cer- tolerable level, we estimate it will take almost a
tainly be wider in the year 2010 than now, given century for the average poor country on current
22 Introduction

performance to attain it. Can these countries wait different stages of development. Perhaps the most
that long? amusing set is attributable to the Brazilian econ-
omist, Roberto Campos, who distinguishes five

I Per Capita Income as an Index

of Development
categories of countries: the HICs, PICs, NICs,
MICs and DICs. These stand for hardly industrial-
ised countries, partly industrialised countries,
newly industrialised countries, mature industrial-
Having considered the world distribution of in- ised countries and decadent industrialised coun-
come, and used per capita income figures as a tries! The HICs and the PICs would certainly
measure of the 'development gap', we come now cover all the low-income countries and at least the
to the question of the use of per capita income lower half of the middle-income countries. The
figures as an index of development and for making NICs cover most of the latter half of the middle-
a distinction between developed and developing income countries - Brazil, Mexico, Hong Kong,
countries, as well as between rich and poor. While Singapore being prime examples. The MICs and
there may be an association between poverty and DICs cover most of those countries classified as
underdevelopment and riches and development, industrial market economies with the exception of
there are a number of reasons why some care must Ireland, New Zealand and Australia which have
be taken in using per capita income figures alone become rich through agriculture. The United
as a criterion of development (unless underde- Kingdom is a prime example of a DIC which has
velopment is defined as poverty and development lost over four million jobs in manufacturing indus-
as riches). Apart from the difficulties of measuring try since 1966.
income in many countries and the difficulties of Because of the arbitrariness and potential de-
making inter-country comparisons, which will be ficiencies of per capita income as a measure of
considered in the next section, a single per capita development, other criteria are often suggested.
income figure to divide developed from developing We have already mentioned the level of indus-
countries must inevitably be somewhat arbitrary, trialisation as one. Another might be an index of
ignoring such factors as the distribution of income the quality of life based on such indices as the level
within countries, differences in development po- of infant mortality; life expectancy; literacy and
tential and other physical indicators of the qual- food consumption per head. Furtado (1964, pp.
ity of life. It is not so much a question of whether 141-3) has attempted a structural definition of
or not low-income countries should be labelled underdevelopment which stresses imbalances be-
'underdeveloped' or 'developing', but what in- tween factors of production, and factor under-
come level should be used as the criterion for utilisation: 'Underdevelopment is a state of factor
separating the developed from the developing imbalance reflecting a lack of adjustment between
countries, and whether all high-income countries the availability of factors and the technology of
should be labelled 'developed'. In many ways it their use, so that it is impossible to achieve full
should be the nature and characteristics of the utilisation of both capital and labour simul-
countries that decide the income level which is taneously'. An underdeveloped structure is there-
used as the dividing line. It also makes sense to fore a situation in which 'full utilisation of
categorise separately the oil rich countries which available capital is not a sufficient condition for
have high per capita incomes but by no stretch of complete absorption of the working force at a
the imagination can be regarded as developed. level of productivity corresponding to the technol-
Within the countries outside the industrialised ogy prevailing in the dynamic sector of the sys-
bloc, the per capita income level dividing the low tem'. By this criterion, Kuwait is obviously not a
and middle-income countries is arbitrary, but developed country, and Australia and Canada
none are fully industrialised and all are developing may be classified as totally developed even though
in this sense. Acronyms abound to describe the they have underutilised natural resources and
 Development and Underdevelopment 23

great development potential. But bearing in mind stage implying any causal relation between the
the arbitrariness of per capita income, it is still characteristics of low-income countries and the
very convenient to have one readily available, and extent of their poverty or underdevelopment. Low
easily understandable, criterion for classifying per capita incomes do seem to go hand in hand
countries, and perhaps per capita income is the with such characteristics as high birth rates and an
best single index we have. It also has one positive absence of industry, but it is always dangerous to
advantage, namely that it focuses on the raison equate association with causation, and in this con-
d'etre of development, which is the raising of liv- text what is cause and what is effect cannot be
ing standards and the eradication of poverty. And deduced without adequate theorising. As Kuznets
in the last resort per capita income is not a bad (1961, p. 9) reminds us, 'it is easy to translate close
proxy for the social and economic structure of statistical association into significant causal rela-
societies. If developing countries are defined on the tionships ... [but] in view of the continuous
basis of a per capita income level so as to include interplay of income levels and these associated
most of the countries of Asia, Africa and South characteristics this simple translation is a logical
America, striking similarities are found between trap that should be avoided lest it lead to intellec-
the characteristics and development obstacles of tual sterility and to a dangerously mechanistic
many of the countries in these continents. These approach to policy implications'. A simple mecha-
characteristics include a high proportion of the nical argument ascribing poverty to such factors as
labour force engaged in agriculture and low agri- low savings and primary product production
cultural productivity, a high proportion of domes- ignores the fact that countries may be at different
tic expenditure on food and necessities; an export stages of their economic history and may differ
trade dominated by primary products and an im- radically with respect to past history and future
port trade dominated by manufactured goods; a prospects. In this respect, four main categories of
low level of technology; high birth rates coupled low-income countries need distinguishing: first,
with falling death rates; and savings undertaken those countries with low per capita incomes but
by a small percentage of the population. There which are progressing rapidly and with enormous
may, of course, be some countries which on a per future potential based on indigenous resources;
capita income basis are classified as developed and second, those countries with rising per capita in-
which possess most of the above-mentioned char- comes but with less hope of rapid self-sustaining
acteristics, but the exceptions will be few, and the growth because of resource limitations; third,
reverse of this situation is almost inconceivable. those countries rich in resources but with per capi-
Also these countries have many social problems in ta income still relatively stagnant; and last, those
common, such as growing unemployment in countries with a stationary per capita income and
urban areas; inegalitarian income distributions, with little prospect of raising living standards ow-
and poor health and standards of education - ing to a sheer lack of resources. A low-income
about which we shall say more later. In general, country may fall into any one of these four broad
therefore, we conclude that per capita income may categories and it would clearly be misleading,
be used as a starting-point for classifying levels of without further information, to ascribe low per
development, and can certainly be used for iden- capita income in a particular country at any par-
tifying the need for development. The only major ticular point in time to characteristics it shares
reservation that we shall have to make later con- with other low-income countries. Association be-
cerns the case of geographically dual economies tween low incomes and certain development char-
where an aggregate per capita income figure can acteristics is all that one is really entitled to claim.
disguise the need for the development of a sizeable There is a difference, however, between using
region as great as the need for the development of per capita income as a guideline for classifying
a country itself, e.g. southern Italy. countries into developed and underdeveloped at a
It should be emphasised that we are not at this point in time and using the growth of per capita
24 Introduction

income as an index of development over time. The other hand, development is hardly possible with-
difficulty of using per capita income for the latter out growth; but development is possible, as we
purpose is the obvious one that if, in a particular have suggested, without per capita income rising.
period, per capita income did not grow because It would be a strange, rather purposeless, type of
population growth matched the growth of a coun- development, however, which left per capita in-
try's total income, one would be forced into the come unchanged, unless the stationary per capita
odd position of denying that a country had de- income was only temporary and a strong founda-
veloped even though its national product had in- tion was being laid for progress in the future. For
creased. This is an inherent weakness of linking the ultimate rationale of development must be to
the concept of development to a measure of living improve living standards and welfare, and while
standards. an increase in measured per capita income may
This leads on to the distinction between growth not be a sufficient condition for an increase in
and development. Development without growth is individual welfare, it is a necessary condition in
hardly conceivable, but is growth possible without the absence of radical institutional innovations,
development? If per capita income is rejected as an such as the distribution of 'free' goods.
index of development over time, an answer to this An increase in income is not a sufficient condi-
question is not possible without defining terms tion for an increase in welfare, because an increase
more precisely. in income can involve costs as well as benefits. It
The difficulty is defining 'development'. The may have been generated at the expense of leisure
meaning of 'growth' is fairly unambiguous; most or by the production of goods not immediately
economists would accept the definition of a rise in consumable. If development is looked upon as a
real national income, i.e. a rise in money income means of improving the welfare of present genera-
deflated by an index of prices. But 'development' tions, probably the best index to take would be
is an elusive term meaning different things to differ- consumption per man-hour worked. This index, in
ent groups of social scientists. 1 Most would agree contrast to an index of per capita income, focuses
that development implies more than just a rise in directly on the immediate utility derivable from
real national income; that it must be a sustained, consumption goods in relation to the disutility of
secular rise in real income accompanied by work effort involved in their production.

I
changes in social attitudes and customs which
have in the past impeded economic advance (see
our earlier discussion). But at this point agreement The Measurement and
on what constitutes development would probably
end. But whatever definition of development is Comparability of Per Capita
given, growth is clearly possible without the Incomes
broader societal changes referred to. The upswing
of the trade cycle is the most obvious example of We turn now to the difficulties of measuring real
the possibility of growth without development, per capita income and comparing living standards
and examples of abortive 'take-offs' are not hard between countries. These difficulties cannot be
to find where countries have grown rapidly for a ignored any longer and must be continually borne
short time and then reverted to relative stagnation. in mind in using per capita income figures both for
Historically, Argentina is a case in point. On the classification purposes and for comparing the rate
of development in different countries over time.
1 For semantic entertainment on the meaning of 'development'
The difficulties of obtaining meaningful and accu-
and 'underdevelopment', see Machlup (1967). Machlup him- rate measures of real per capita income relate
self defines economic development as 'those changes in the use more to the measurement of real income than to
of resources that result in potentially continuing growth of
national income per head in a society with increasing or stable
population, and we shall thus concentrate briefly
population'. on some of the problems of national income
 Development and Underdevelopment 25

accounting and the uses of national income stat- illiteracy rife, and in which many goods produced
istics in developing countries. and consumed do not exchange for money. Differ-
The first point to bear in mind is that if no ences in the extent of the subsistence economy
allowance is made for the non-monetary sector in between developing countries, and differences in
the national income accounts of a developing the ease and difficulty of collecting data, may
country, any long-term growth estimates are markedly influence estimates of national income,
bound to have an upward bias owing to the grad- and therefore of per capita income differences,
ual extension of the money economy and the shift between these countries and the rest of the world.
of economic activities from the household to Some testimony to the role that the subsistence
the market-place. Furthermore, if no allowance is sector must play in the economies of most de-
made for the subsistence sector in some countries, veloping countries is illustrated by the inconceiv-
it may be misleading to compare periods in these ability that 60 per cent of the world's population
countries' history and to compare growth rates could remain alive on the equivalent of $1200 per
between countries, especially between the de- annum. But this is not the whole story.
veloped and the developing countries. The other part of the story, and probably the
Growth rates may also be biased upwards by major part, concerns the understatement of living
using prices as weights in compiling national in- standards in developing countries when their
come totals from the output statistics of different national incomes measured in local currencies are
sectors of the economy (unless the weights are converted into US dollars (as the common unit of
revised frequently), since goods with high prices, account) at the official rate of exchange. If the US
which subsequently fall, are usually the fastest dollar is used as the unit of account the national
growing. This is more of a danger in developing income per head of country X in US dollars 1s
countries than in developed countries because given by
of less sophisticated accounting techniques, the
greater difficulty in revising price weights, and the GNPx
more widespread introduction of new goods with . -;- Exchange rate
Popu1atwn
high initial prices.
A consideration of prices is also necessary in For example, if the GNP of country X is 100
deciding what price index to use as a deflator of million rupees, its population is 5 million, and
money national income in order to obtain an in- there are 10 rupees to the dollar, then per capita
dex of real income. The task of converting money income of country X in dollars is:
income statistics into real income raises all the
difficulties, not peculiar to developing countries, 100 -;- 10 = $2.
connected with the use of index numbers, such as 5
which base year should be taken, how to take
account of changes in the quality of products, But if living standards between two countries are
which weighting system to employ, and so on. to be compared by this method, it must be
These are conceptual issues to be sorted out by the assumed that 10 rupees in country X can buy the
national income statistician rather than by the de- same living standard as $1 in the United States. It
velopment economist, but it is important for the is well known, however, that official exchange
economist to know how figures for real national rates between two countries' currencies are not
income, or per capita income, have been arrived at very good measures of purchasing-power parity
prior to analysis. between countries, especially between countries at
But apart from the problem of bias and the different levels of development. The reason is this:
choice of price deflator there is the sheer practical exchange rates are largely determined by the
difficulty of measuring money national income in a supply and demand for currencies based on goods
rural economy where communications are bad, which are traded, the prices of which tend to be
26 Introduction

equalised internationally. Purchasing-power par- where Q; is the geometric mean of the quantities of
ity, however, depends not only on the price of each good consumed in the two countries.
traded goods, but also on the price of non-traded The purchasing-power-equivalent ratio can then
goods which is largely determined by unit labour be used to convert one country's national income
costs, which tend to be lower the poorer the country. measured in local currency into another country's
In general, therefore, the lower the level of de- currency as the unit of account (e.g. the US dollar).
velopment the lower the ratio of the price of non- For example, suppose that the official exchange
traded goods to traded goods and the more the rate between the Indian rupee and the US dollar is
use of the official exchange rate will understate 10 : 1, while the purchasing-power-equivalent
the living standards of the developing country ratio is 5 : 1. This means that converting the
measured in US dollars. The ratio of the price of Indian national income into US dollars at the
non-traded goods to traded-goods tends to rise purchasing-power parity rate will give double
with development as wage levels in the non-traded the income than the conversion at the official ex-
goods sector rise while productivity growth is change rate.
relatively slow, and slower than in the traded The method described above is only one poss-
goods sector. To make meaningful international ible approach to the construction of a binary
comparisons of income and living standards, purchasing-power parity ratio. Instead of the rela-
therefore, what is required is a measure of tive prices being applied to the geometric average
purchasing-power parity, or a real exchange rate, of the quantities of each good consumed in the
between countries. two countries, the quantity weights could be the
There are several methods of constructing quantities consumed in either the one country
purchasing-power parity ratios in order to make or the other. If the quantity weights of the poor
binary comparisons (one country with another) or country are used, the dollar valuation of the poor
'multilateral' comparisons in which the currency country's income would probably be even higher
of any one of a group of countries can act as the because relatively large weights would be given to
unit of account without altering the ratios of living those items which are widely consumed in those
standards between countries. countries with low relative prices, therefore im-
One approach to the construction of a proving the purchasing-power parity ratio.
purchasing-power parity ratio between two coun- Alternatively comparisons can be made from
tries is to revalue the national incomes of the two the output side by valuing output in both countries
countries by selecting a comparable basket of goods either at one country's prices or at the other's
and services in each country and estimating the pnces.
purchasing-power equivalent of each item in coun- The results of constructing and using pur-
try A relative to country B. Thus if P;a is the price of chasing-power parity ratios for making binary
item i in country A and P;b is the price of item i in comparisons have been quite dramatic. In an early
country B, the purchasing-power equivalent of study, Millikan suggested that the real incomes of
item i in country A relative to country B is P;aiP;b· many African and Asian countries in 1950 were of
By extending this calculation to all goods, and the order of 350 per cent larger than indicated by
applying the price ratios to the average quantities UN statistics in US dollars per capita (Kindleber-
consumed of each item in the two countries, we ger (1965), p. 9). Likewise, Usher (1966, pp. 10-
obtain a formula for the over-all purchasing- 17) estimated the ratio of British to Thai national
power equivalent in country A relative to country income per head to be 13.06 : 1 converting Thai
B: income in local currency into pounds at the
foreign exchange rate. However, by-passing the
exchange rate and valuing British and Thai income
directly at Thai prices the ratio fell to 6.27: 1, and
valuing both incomes at British prices the ratio fell
 Development and Underdevelopment 27

Table 1.3 Per Capita GDP in 1970 in $US Using Official Exchange Rates and in International $s Using
International Prices

Per capita GDP (1970) in Per capita GDP (1970) in

$US at official exchange rates I$ at international prices

Colombia 329 763

France 2902 3599
West Germany 3080 3585
Hungary 1037 1935
India 98 342
Italy 1699 2198
Japan 2003 2952
Kenya 144 275
United Kingdom 2143 2895
United States 4801 4801

to 2.76 : 1. In making binary international com- rates increases the estimate of per capita income
parisons of per capita incomes, therefore, it makes more than two-fold (e.g. in the case of Colombia
a great deal of difference whether the official and India).
foreign-exchange rate is used as a measure of Purchasing power parities can be derived from
purchasing-power parity, or, if the use of ex- the table in the following manner. The purchasing
change rates is discarded, which quantity weights power parity rate (PPPR) is equal to the official
are applied to relative price indices or which price exchange rate divided by the extent to which the
weights are applied to relative quantities. official exchange rate conversion of per capita
More recently, Irving Kravis and collaborators income understates the true level of per capita
have developed a method of making multilateral income when measured at international prices.
comparisons of real per capita incomes across Suppose, for example, there are 7 Kenyan shillings
countries by constructing world price ratios based to one US dollar, and that the official exchange
on price and quantity data for over 100 com- rate conversion understates Kenyan per capita in-
modity categories in over 100 countries. The inter- come by 90 per cent (which it does according to
national prices are then used to value quantities in Table 1.3). The PPPR of Kenyan shillings to $US is
each of the countries. The international prices and therefore PPPR = 7/1.9 = 3.7. That is, to compare
product values are expressed in international dol- living standards between Kenya and the US, the
lars (1$). An international dollar has the same real exchange rate of 3.7 ought to be used, not the
overall purchasing power as a US dollar for official rate of 7 shillings to the dollar.
national income as a whole, but relative prices for The pioneer work of Kravis is now regularly
each country are relative to average world prices extended and updated by his collaborators, Sum-
rather than relative to US prices. This multilateral mers and Heston, who produce international com-
approach allows a direct comparison between any parisons of price levels and real per capita incomes
two countries using any country's currency as the at international prices for all major countries in
unit of account. The results of this exercise for ten the world since 1950, which can be compared
countries originally taken by Kravis et al. ((1975); with the World Bank estimates of per capita in-
see also Kravis et al. (1978)) are summarised in comes based on official rates of exchange with the
Table 1.3. It can be readily seen that converting US dollar (see, for example, Summers and Heston
some of the poor countries' incomes at inter- (1988)).
national prices rather than at official exchange Before ending this section it should be stressed
28 Introduction

that purchasing-power parity ratios, however de- head; annual number of domestic letters sent per
rived, should not be interpreted as equilibrium head; stock of radio receivers per head; stock of
exchange rates to be used as a standard for telephones per head; stock of road vehicles per
measuring the extent to which the official ex- head; and annual meat consumption per head.
change rate may be either under- or over-valued. Predictions are made by Beckerman and Bacon for
As we have said before, exchange rates are deter- eighty countries, and the countries ranked. Unfor-
mined by the supply of and demand for traded tunately there is no indication given of the extent
goods, while many goods not traded figure in the to which the ranking of countries by this method
purchasing-power parity ratios. differs from the ranking by published per capita
Apart from the construction of purchasing- income statistics. If the rankings differ substan-
power parity ratios there have been several other tially, it may be that the Beckerman-Bacon approach
attempts to overcome the problems of using sus- is the most appropriate to adopt for obtaining a
pect income statistics to compare living standards measure of comparative levels of welfare between
between countries and over time. One consists of countries as measured by consumption per head. If
the use of non-monetary indicators, such as levels the rankings do not differ markedly, income per head
of health and education, the number of cars per may still be regarded as a good proxy for the com-
head, steel production, etc., and to rank countries parison of living standards. (This is the conclusion of
according to an index of each indicator (United Hagen and Hawrylyshyn (1969).)
States = 100). The indices of all indicators are
then averaged. However, this approach suffers
from the drawback that there is no satisfactory
weighting system that can be used for combining
I Other Dimensions of the
Development Gap
the different indices. An alternative approach has
been suggested by Beckerman and Bacon (1966) The human problems of developing countries are
which they claim to be 'theoretically valid, poten- not confined to low levels of per capita income.
tially very accurate, and at the same time, almost Developing countries generally experience much
costless'. The approach is based on correlation higher levels of unemployment - open and dis-
and regression analysis. Their procedure is as fol- guised - than do developed countries; their
lows. First, they take reliable estimates of relative income distributions tend to be much more in-
(to the United States) real consumption per head egalitarian, and the levels of health, nutrition and
for thirteen countries. Then they take several non- education are often abysmally low. Policy in
monetary indicators and pick out those that corre- developing countries is becoming increasingly
late best with the reliable observations of real concerned with these other dimensions of the de-
consumption per head, experimenting with differ- velopment gap. The growth of per capita income
ent functional forms. Finally, with the aid of a has never been the sole objective of development
computer they find which combination of indica- policy but more attention is now being paid to
tors, and which forms of the equations linking other objectives which in some instances may con-
them to real consumption per head, give the best flict with growth of per capita income.
results in terms of the multiple correlation coef-
ficients and the standard errors of the regression
equations. From the best results, predictions can • Unemployment
be made of relative consumption per head for
countries, where data on income per head are Open unemployment in the urban areas of de-
thought to be unreliable, from the non-monetary veloping countries is another dimension of the
independent variables. The variables used as pre- development problem, and an increasingly serious
dictors were as follows: annual crude steel con- one. The reduction of unemployment has become
sumption per head; annual cement production per a major policy priority of governments and inter-
 Development and Underdevelopment 29

Table 1.4 Unemployment Rates in Selected Developing Countries

Percentage of labour force

1965 1970 1975 1980 1981 1982

Africa:
Egypt 1.9a 2.4 2.5 5.2
Middle East:
Syria 7.4 6.4 4.8 3.8b
Asia:
S. Korea 7.4 4.5 4.1 5.2 4.6 4.4
Philippines 8.2 6.0 4.4 4.8 5.1 5.3
Latin America and Caribbean:
Argentina 5.3 4.8 2.3 2.3 4.5 5.7
Chilec 5.4 4.1 15.0 10.4 11.3 20.0
Colombiad 8.9 8.2 10.5 9.9 8.1
Panama 7.6 7.1 6.4 8.8e
Peruf 4.7 4.9 7.0 6.8 7.0
Trinidad and Tobago 14.0 12.5 15.0 10.0 12.2
Uruguayg 7.2h 7.5 8.1i 7.3 6.6
Venezuelai 7.7k 6.3 7.6 1 6.0 6.2 7.1
• 1964, b 1979, c Gran Santiago, d Bogota, e 1979, 1 Lima Callao, 8 Montevideo, h 1967, ' 1974, i urban, k 1967, 1 1976.
Source: M. Godfrey, Global Unemployment: The New Challenge to Economic Theory (Brighton: Wheatsheaf Books, 1986).

national agencies concerned with developing lack of co-operating factors of production to work
countries. Poor countries for a long time, and par- with labour: capital in particular. As far as migra-
ticularly since the population explosion, have been tion is concerned, there are both push and pull
characterised by underemployment, or disguised factors at work. The push factors have to do with
unemployment (see Chapter 3), on the land. What the limited job opportunities in rural areas, and
has happened in recent years is that disguised un- the greater willingness and desire to move fostered
employment on the land has transferred itself into by education and the improvements in com-
open unemployment in the towns. The reasons for munications. The pull factors relate to the de-
this, and the rationale for migration, will be consi- velopment of industrial activities in the towns
dered later, but first let us outline the facts. Some offering jobs at a higher real wage than can be
idea of the seriousness of unemployment is given earned in rural areas, so that even if a migrant is
by the figures in Table 1.4 for a selection of de- unemployed for part of the year, he may still be
veloping countries. The figures are not so different better off migrating to the town than working in
from unemployment rates recorded in some de- the rural sector. If there is no work at all in the
veloped countries in the 1980s, but the recorded rural sector, the migrant loses nothing, except
figures measure only the tip of the iceberg. perhaps the security of the extended family sys-
There are a number of contributory factors to tem. The rate of growth of job opportunities in the
the emergence of unemployment in the towns on rural sector depends on the rate of growth of
an increasing scale. The problem is not so much demand for the output of the rural sector and
one of a deficiency of demand for labour in an the rate at which jobs are being 'destroyed' by pro-
aggregate demand sense. The causal factors relate ductivity growth. 1 The growth of demand for the
to the incentives to labour migration from the 1 i.e. since 0 = L(OIL), then dLIL = dOlO- d(OIL)I(OIL),
rural to the urban areas, and the incapacity of the where L is employment, 0 is output and OIL is labour pro-
urban areas to provide employment owing to a ductivity.
30 Introduction

output of the rural sector will be equal to the rate necessarily solved by a faster rate of capital ac-
of growth of total population plus the rate of cumulation in the urban sector, because migration
growth of per capita income multiplied by the is not simply a function of the actual difference in
income·elasticity of demand for rural output. Sup- real remuneration between the two sectors, but
pose population is growing at 2 per cent, that per also of the level of job opportunities in the urban
capita income is growing at 4 per cent and that the sector. If the rate of job creation increases, this
income elasticity of demand is 0.5, then the rate of may merely increase the flow of migrants with no
growth of demand for the rural sector's output reduction in unemployment. The solution would
will be 2.0 + 0.5 (4.0) = 4.0 per cent. Now seem to be to create more job opportunities in the
suppose that the rural sector's productivity is rural sector. This will require, however, not only
growing at 2 per cent. The rate of growth of the redirection of investment but also the exten-
demand for labour will then be the difference be- sion of education and transport facilities, which in
tween 4 per cent and 2 per cent, i.e. 2 per cent. If the past few years have themselves become power-
the rural-sector labour force is growing at 3 per ful push factors in the migration process. Whereas
cent, then 1 per cent of the labour force will be formerly redundant labour might have remained
becoming redundant annually. If the level of dis- underemployed on the family farm, nowadays
guised unemployment in the rural sector were not education and easy transportation provide the
to increase, this figure would constitute the poten- incentive and the means to seek alternative
tial volume of migrants. If the urban labour force employment opportunities. While education and
is only one-fifth of the size of the rural labour improved communications are desirable in them-
force, a 1 per cent migration of rural labour would selves, and facilitate development, their provision
represent a 5 per cent increase in the urban labour has augmented the flow of migrants from rural to
force due to migration. On average, this is about urban areas.
the extent of the influx from the rural sector into The pull factors behind migration are not hard
the urban areas of developing countries. On top of to identify. The opportunities for work and leisure
this there is the natural increase in the work-force provided by the industrial, urban environment
in the urban area to consider; this is of the order of contrast sharply with the conservatism and stul-
between 2 and 3 per cent. If job opportunities in tifying atmosphere of rural village life and act
the urban areas are only increasing at 5 per cent, naturally as a magnet for those on low incomes or
then 2 to 3 per cent of the urban labour force will without work, especially the young. Given the
become unemployed annually, thus raising the un- much higher wage in the urban sector, even the
employment percentage year by year. prospect of long spells of unemployment in
Historically, the process of development has the urban area does not detract from the incentive
always been associated with, and characterised by, to migrate. Moreover, the choice is not necessarily
an exodus from the land, continuing over centuries. between remaining in the rural sector and migrat-
The uniqueness of the present situation is not ing to the urban sector with the prospect of long
the migration itself but its magnitude and speed. periods of unemployment. The unemployed in the
And the problem is that the urban sector cannot urban sector can often find work, or create work
absorb the numbers involved. For any given tech- for themselves, on the fringes of the industrial
nology the rate at which the urban (industrial) sector - in particular in the informal service sector
sector can absorb migrants largely depends on the of the industrial economy. The wages may be low,
rate of capital formation. If labour and capital but some income is better than no income. In other
must be combined in fixed proportions, and the words, unemployment in urban areas may take the
rate of capital accumulation is only 5 per cent, form of underemployment, or become disguised,
then the rate of increase in job opportunities can just as in the case of the rural sector - its mani-
be only 5 per cent also. Unfortunately, however, as festation being low income. This has led to the
we shall show in Chapter 3, the problem is not notion of an income measure of unemployment
 Development and Underdevelopment 31

which needs to be added to registered unemploy- gramme sponsoring missions to several countries
ment to obtain a true measure of unemployment to undertake detailed diagnosis of the employment
and the availability of labour supply. problem. Some of the major findings and conclu-
One way of measuring the extent of unemploy- sions have been surveyed by Thorbecke (1973).
ment disguised in the form of low-productivity/ The report on Colombia estimated that an em-
low-income jobs is to take the difference between ployment growth rate of 7 per cent per annum
the actual labour employed at the sub-standard would be required to absorb all increases in the
income and the labour that would be required to labour force. Since the projected rate of growth of
produce a given level of output or service at an labour productivity would make this impossible,
acceptable level of income per head. Before the report recommended a move to more labour-
measurement took place, of course, the acceptable intensive techniques, at least in the industrial sec-
(standard) level of income would have to be de- tor. The report on Kenya emphasised the income
fined. It could be that level set as the 'poverty line', measure of unemployment discussed above. It
below which health and welfare become seriously drew attention to the large number of people
impaired. The income measure of unemployment forced to work in the informal sector of the indus-
would thus be: trial sector with extremely low productivity, such
that income falls below an acceptable minimum.
0 0 Visible (open) unemployment was estimated at be-
U = L- L* tween 8 and 14 per cent. Adding low-income
OIL OIL*
workers in the manner described above, the per-
centage rises to 20 per cent for males and 50 per
where L is the actual labour employed, L * is the
cent for females. The report on Iran gave priority
labour consistent with an acceptable level of in-
to population control (which we discuss in Chap-
come per person employed, OIL is the actual level
ter 6) but, as it reputably said in the draft of the
of productivity, OIL* is the acceptable level of
report, the· population problem cannot be solved
income per employed person, and 0 is output.
overnight! The report on Sri Lanka is distinctive
Let us work an example. Suppose that the
for its thorough analysis of structural unemploy-
annual flow of output of an activity or service is £1
ment. There are three main types of structural
million and that the existing number employed is
imbalance in most developing countries which
10 000, giving a level of productivity of £100.
affect employment. One is between skills, attitudes
Now suppose that the acceptable level of pro-
and expectations on the one hand and oppor-
ductivity to produce an acceptable level of income
tunities on the other. This affects mainly educated
per person employed is £200. The income measure
people. Second, there is imbalance between re-
of unemployment is then:
gions. Third, there is the sheer lack of co-operating
factors for labour to work with, particularly capit-
u= 1 000 000 1 000 000 = 5000 al and skilled managerial enterprise. We take up
100 200 some of these difficulties in later chapters.

that is, one-half of the ex1stmg labour force is

disguisedly unemployed in the sense that the level • The Distribution of Income
of output is not sufficient to maintain all those
who currently work at an adequate standard of The average level of per capita income in the de-
living. veloping countries has increased fairly substan-
Such has been the concern in recent years with tially over the last two decades, as Table 1.2
employment and unemployment in developing shows, yet the evidence suggests that growth and
countries that the International Labour Office development in many countries has left the vast
launched in 1969 a World Employment Pro- mass of people untouched. The growth that has
32 Introduction

taken place has served largely to benefit the few - income. The data show fairly conclusively that
the richest 20 per cent of the population. Rural inequality increases up to a certain stage of de-
and urban poverty are still widespread, and if velopment and then declines, tracing out an in-
anything the degree of income inequality within verted U-shape similar to the pioneer work of
the developing countries has increased. It should Kuznets for the now developed countries. The
not come as a surprise, however, that the trans- average Gini coefficient for forty-three developing
formation of economies from a primitive subsist- countries is 0.467 compared with 0.392 for thir-
ence state into industrial societies, within a basically teen developed countries. The maximum equalisa-
capitalist framework, should be accompanied in the tion percentage is 35.8 for the developing countries
early stages by widening disparities in the personal compared with 28.4 for the developed countries.
distribution of income. Some people are more in- The greater degree of income inequality in the
dustrious than others, and more adept at accumu- developing countries appears largely due to the
lating wealth than others. Opportunities cannot, higher share of income received by the richest 5
in the very nature of things, be equal for all. In the per cent of income recipients. In developing coun-
absence of strong redistributive taxation, income tries this share is 28.7 per cent, compared with
inequality will inevitably accompany industrialisa- 19.9 per cent in developed countries. The share
tion because of the inequality of skills and wealth going to the poorest 20 per cent in developing
that differences in individual ability and initiative, countries is slightly higher than in the developed
and industrialisation, produce. countries, but there can be no comfort in this fact.
The evidence for individual countries suggests, The focus must be on absolute, as well as relative,
however, that income inequality ceases to increase poverty; and the poorest in countries with average
at quite low levels of per capita income - at about per capita incomes less than $1000 per annum
$300 per annum at 1965 prices, or at about $1500 must be very poor indeed. On average, the per
at 1992 prices. Beyond this level, income in- capita income of the poorest 20 per cent of the
equality tends to decrease as industrialisation pro- population in the typical developing country is
ceeds. Kuznets's work (1955, 1963) shows that in about 30 per cent of the national average. As we
many of the present developed countries, the ex- said earlier in the chapter, there are about 600
tent of inequality decreased in the later stages of million people living on per capita incomes of less
industrialisation, and certainly the degree of in- than $275 per annum at 1985 prices: 50 million in
equality in the developing countries is greater than Latin America; 450 million in Asia, and 100 mil-
in the present developed countries, largely as a lion in Africa.
result of the heavy concentration of income among Several formidable barriers exist to raising the
the top 5 to 10 per cent of income recipients. The living standards of the poorest, and to narrowing
work of Kravis (1960) also shows that the degree the income distribution overall. There is the dualistic
of inequality first increases within countries and nature of many economies (see Chapter 5), perpetu-
then declines. ated by feudal land-tenure systems and urban bias
The most comprehensive data assembled to in the allocation of investment resources. There is
date are those by Adelman and Morris (1971), and inequality in the distribution of education facilities
extended by Paukert (1973), which show the size to contend with, particularly the lack of facilities
distribution for fifty-six countries, developed and in rural areas where the poorest are concentrated.
developing (see Table 1.5). The table also shows Third, there is disguised unemployment on the
two measures of income concentration calculated land and underemployment and open unemploy-
from these figures - the Gini coefficient and the ment in urban areas created by rural-urban migra-
maximum equalisation percentage, which indi- tion, a shortage of investment resources, and
cates what percentage of total income would have inappropriate technological choices. Until de-
to be shifted between the quintiles of income reci- velopment policy comes to grips with these prob-
pients in order to achieve an equal distribution of lems, there will continue to be large pockets of
 Development and Underdevelopment 33

Table 1.5 Size Distribution of Personal Income Before Tax in Fifty-Six Countries: Income Shares Received by
Quintiles of Recipients in the Neighbourhood of 1965

Country and level Percentiles of recipients Gini Maximum GDP per

of GDP per head coefficient equalisation head in 1965
Below 20 21--40 41-60 61-80 81-9595-100 percentage ($US)

Under $100
Chad (1958) 8.0 11.6 15.4 22.0 20.0 23.0 0.35 25.0 68
Dahomey (1959) 8.0 10.0 12.0 20.0 18.0 32.0 0.42 30.0 73
Niger (1960) 7.8 11.6 15.6 23.0 19.0 23.0 0.34 25.0 81
Nigeria (1969) 7.0 7.0 9.0 16.1 22.5 38.4 0.51 40.9 74
Sudan (1969) 5.6 9.4 14.3 22.6 31.0 17.1 0.40 30.7 97
Tanzania (1964) 4.8 7.8 11.0 15.4 18.1 42.9 0.54 41.0 61
Burma (1958) 10.0 13.0 13.0 15.5 20.3 28.2 0.35 28.5 64
India (1956-7) 8.0 12.0 16.0 22.0 22.0 20.0 0.33 24.0 95
Madagascar (1960) 3.9 7.8 11.3 18.0 22.0 37.0 0.53 39.0 92
Group average 7.0 10.0 13.1 19.4 21.4 29.1 0.419 31.6 78.3

$101-200
Morocco (1965) 7.1 7.4 7.7 12.4 44.5 20.6 0.50 45.4 180
Senegal (1960) 3.0 7.0 10.0 16.0 28.0 36.0 0.56 44.0 192
Sierra Leone (1968) 3.8 6.3 9.1 16.7 30.3 33.8 0.56 44.1 142
Tunisia (1971) 5.0 5.7 10.0 14.4 42.6 22.4 0.53 44.9 187
Bolivia (1968) 3.5 8.0 12.0 15.5 25.3 35.7 0.53 41.0 132
Ceylon (Sri Lanka) (1963) 4.5 9.2 13.8 20.2 33.9 18.4 0.44 32.5 140
Pakistan (1963-4) 6.5 11.0 15.5 22.0 25.0 20.0 0.37 27.0 101
South Korea (1966) 9.0 14.0 18.0 23.0 23.5 12.5 0.26 19.0 107
Group average 5.3 8.6 12.0 17.5 31.6 24.9 0.468 37.2 147.6

$201-300
Malaya (1957-8) 6.5 11.2 15.7 22.6 26.2 17.8 0.36 26.6 278
Fiji (1968) 4.0 8.0 13.3 22.4 30.9 21.4 0.46 34.7 295
Ivory Coast (1959) 8.0 10.0 12.0 15.0 26.0 29.0 0.43 35.0 213
Zambia (1959) 6.3 9.6 11.1 15.9 19.6 37.5 0.48 37.1 207
Brazil (1960) 3.5 9.0 10.2 15.8 23.1 38.4 0.54 41.5 207
Ecuador (1968) 6.3 10.1 16.1 23.2 19.6 24.6 0.38 27.5 202
El Salvador (1965) 5.5 6.5 8.8 17.8 28.4 33.0 0.53 41.4 249
Peru (1961) 4.0 4.3 8.3 15.2 19.3 48.3 0.61 48.2 237
Iraq (1956) 2.0 6.0 8.0 16.0 34.0 34.0 0.60 48.0 285
Philippines (1961) 4.3 8.4 12.0 19.5 28.3 27.5 0.48 35.8 240
Colombia (1964) 2.2 4.7 9.0 16.1 27.7 40.4 0.62 48.0 275
Group average 4.8 8.0 11.3 18.1 25.7 32.0 0.499 38.5 244.4

$301-500
Gabon (1960) 2.0 6.0 7.0 14.0 24.0 47.0 0.64 51.0 368
Costa Rica (1969) 5.5 8.1 11.2 15.2 25.0 35.0 0.50 40.0 360
Jamaica (1958) 2.2 6.0 10.8 19.5 31.3 30.2 0.56 41.5 465
Surinam (1962) 10.7 11.6 14.7 20.6 27.0 15.4 0.30 23.0 424
Lebanon (1955-60) 3.0 4.2 15.8 16.0 27.0 34.0 0.55 41.0 440
Barbados (1951-2) 3.6 9.3 14.2 21.3 29.3 22.3 0.45 32.9 368
Chile (1968) 5.4 9.6 12.0 20.7 29.7 22.6 0.44 33.0 486
Mexico (1963) 3.5 6.6 11.1 19.3 30.7 28.8 0.53 39.5 441
Panama (1969) 4.9 9.4 13.8 15.2 22.2 34.5 0.48 36.7 490
Group average 4.5 7.9 12.3 18.0 27.4 30.0 0.494 37.6 426.9
34 Introduction

Table 1.5 Size Distribution of Personal Income Before Tax in Fifty-Six Countries: Income Shares Received by
Quintiles of Recipients in the Neighbourhood of 1965 (continued)

Country and level Percentiles of recipients Gini Maximum GDP per

of GDP per head coefficient equalisation head in 1965
Below 20 21-4041-60 61-80 81-9595-100 percentage ($US)

$501-1000
Republic of South Africa (1965) 1.9 4.2 10.2 26.4 18.0 39.4 0.58 43.7 521
Argentina 7.0 10.4 13.2 17.9 22.2 29.3 0.42 31.5 782
Trinidad and Tobago (1957-8) 3.4 9.1 14.6 24.3 26.1 22.5 0.44 32.9 704
Venezuela (1962) 4.4 9.0 16.0 22.9 23.9 23.2 0.42 30.6 904
Greece (1957) 9.0 10.3 13.3 17.9 26.5 23.0 0.38 29.5 591
Japan (1962) 4.7 10.6 15.8 22.9 31.2 14.8 0.39 28.9 838
Group average 5.1 8.9 13.9 22.1 24.7 25.4 0.438 32.9 723.3

$1001-2000
Israel (1957) 6.8 13.4 18.6 21.8 28.2 11.2 0.30 21.2 1243
United Kingdom (1964) 5.1 10.2 16.6 23.9 25.0 19.0 0.38 28.1 1590
Netherlands (1962) 4.0 10.0 16.0 21.6 24.8 23.6 0.42 30.0 1400
Federal Republic of
Germany (1964) 5.3 10.1 13.7 18.0 19.2 33.7 0.45 32.9 1667
France (1962) 1.9 7.6 14.0 22.8 28.7 25.0 0.50 36.5 1732
Finland (1962) 2.4 8.7 15.4 24.2 28.3 21.0 0.46 33.5 1568
Italy (1948) 6.1 10.5 14.6 20.4 24.3 24.1 0.40 28.8 1011
Puerto Rico (1963) 4.5 9.2 14.2 21.5 28.6 22.0 0.44 32.1 1101
Norway (1963) 4.5 12.1 18.5 24.4 25.1 15.4 0.35 24.9 1717
Australia (1966-7) 6.6 13.4 17.8 23.4 24.4 14.4 2.30 22.2 1823
Group average 4.7 10.5 15.9 22.2 25.7 20.9 0.401 29.0 1485.2

$2001 and above

Denmark (1963) 5.0 10.8 18.8 24.2 26.3 16.9 0.37 25.4 2078
Sweden (1963) 4.4 9.6 17.4 24.6 26.4 17.6 0.39 28.6 2406
United States (1969) 5.6 12.3 17.6 23.4 26.3 14.8 0.34 24.5 3233
Group average 5.0 10.9 17.9 24.1 26.3 16.4 0.365 26.5 2572.3
Source: F. Paukert, International Labour Review (August 1973), based on data compiled by I. Adelman and C. Morris, 'An
Anatomy of Income Distribution Patterns in Developing Countries', AID Development Digest (October 1971).

absolute poverty and a marked degree of inequal- countries have a comparatively equal income dis-
ity in income distribution. In deciding on the tribution while many of the slowest growing coun-
allocation of investment resources and the choice tries have a high degree of income inequality. It
of projects, a high weight needs to be given to appears that income inequality is not necessary for
projects which raise the income of the poorest in high levels of saving and investment or other fac-
the income distribution (see Chapter 8). Fortu- tors that contribute to fast growth.
nately, evidence from the World Bank does not
suggest that growth and equity necessarily con-
flict. If anything, countries with a greater degree of • Growth and Distribution
equality have grown fastest. This is shown in Fig-
ure 1.2 opposite where the growth of income per Progress towards achieving the twin objectives of
head is measured on the vertical axis, and income faster growth and a more equal distribution of
inequality is measured on the horizontal axis, and income can be examined simultaneously by con-
it can be seen that many of the fastest growing structing poverty-weighted indices of growth.
 Development and Underdevelopment 35

Figure 1.2 Income Inequality and Economic Growth The idea of constructing poverty weighted in-
dices of growth is to give at least equal weight to
Percentage GOP growth per head, 1965-89
all income groups in society, if not a greater
8
weight to the poor, in order to obtain a better
7 measure of the growth of overall welfare combin-
ing the growth of income with its distribution.
6 In the above example, for instance, if each
group is given an equal weight of one-third, the
5 measured growth of welfare becomes:
4
% growth of 'welfare' = 1(0.33) + 1(0.33)
+ 10(0.33) = 4%
2 which is much less than the rate of growth shown
by the conventional measure of GNP growth when
distributional considerations are taken into
account.
0
A society could go further and say it places no
-1 value or weight on income growth for the richest
0 5 10 15 20 25 30 35 40 45 third of the population, and places all the weight
Income inequality on the lower income groups with, say, a 60 per
The mulliple of the income of the richest 20% cent weight to the bottom third and a 40 per cent
to the income of the poorest 20% weight to the middle third. The growth of 'wel-
Source: World Development Report 1991. fare' would then look derisory:

GNP growth as conventionally measured is a % growth of 'welfare' = 1(0.6) + 1(0.4)

weighted average of the growth of income of differ- + 10(0) = 1%
ent groups of people, where the relevant weights
are each group's share of total income. The This approach has been experimented with by
measured growth rate pays no regard to the dis- economists from the World Bank (see Ahluwalia,
tribution of income. A high growth rate may be Carter and Chenery (1979)) to compare countries,
recorded, which has benefited only the rich. For giving a 60 per cent weight to the lowest 40 per
example, suppose the bottom one-third of the cent of the population; a 40 per cent weight to the
population receive 10 per cent of income; the mid- middle 40 per cent and no weight to the top 20 per
dle one-third receive 30 per cent of income, and cent. In countries where the distribution of income
the top one-third receive 60 per cent of income. deteriorated, the poverty-weighted measure of the
GNP growth would be measured as: growth of welfare shows less improvement than
GNP growth, and where the distribution of in-
% growth of GNP = r (0.1) + r
1 2 (0.3) come improved, the poverty-weighted growth rate
+ r 3 (0.6) shows more improvement than GNP growth.

where ru r 2 and r 3 are the respective rates of growth

of income of the three groups. Suppose r 1 = 1 per • Nutrition and Health
cent; r2 = 1 per cent and r 3 = 10 per cent. A growth
rate of GNP of 6.4 per cent would then be recorded Low absolute levels of income can have serious
which looks very respectable but the position of the consequences for the nutrition and health of indi-
poorest would hardly have changed. viduals. It has been estimated by the UN Food and
36 Introduction

Agriculture Organisation that there are at least portant for brain development in the first three
one billion people in the world suffering from years of life, during which the brain grows to 90
various degrees of malnutrition, including two- per cent of its full size. Brain damage due to pro-
thirds of the world's one billion children. Mal- tein deficiency is irreversible.
nutrition among children is particularly serious
because it stunts growth and mental development,
and adds another element to the vicious circle of
poverty. Malnutrition is also a major cause of
I Poverty, Famine and
Entitlements
infant mortality, the rates of which are more than
twice as high in developing countries as in de- Malnutrition is caused by a lack of access to food,
veloped countries. The costs of treating various but this does not only depend on the availability of
forms of malnutrition are trivial relative to the food; it also depends on people's entitlement to
tangible benefits, and to the costs of treating the food. Vast sections of the population may go short
consequences. To prevent malnutrition in children of food, and experience famine conditions, not
from the age of six months to three years, which is primarily because food has become scarce, but
a child's most vulnerable period, can cost as little because their entitlement to food has been im-
as $50 at current prices. The annual cost of pre- paired. This is the powerful thesis put forward by
venting malnutrition is about the same as the daily Professor Sen (1984), who argues that to under-
cost of treating its effects. Vitamin A deficiency is a stand poverty and starvation, or malnutrition
cause of blindness. The annual cost of supporting associated with it, it is necessary to understand
a blind person is at least 1000 times the annual both ownership patterns and exchange entitle-
ingredient cost of the vitamin A needed for preven- ments, which requires in turn an understanding of
tion. Iodate deficiency is a cause of goitre, which modes of production and the class· structure. He
leads to cretinism and deaf-mutism. The cost of attempts to document the theory drawing on the
iodate to prevent goitre is one-sixth of a cent per experience of major famines such as the Great
person per year. And so one could go on. Preven- Bengal famine of 1943; the Ethiopian famine of
tion is better than cure not only for the individual 1973-5; the famine in the Sahel region of Africa in
but also in a very real economic sense for the the early 1970s, and the Bangladesh famine of
welfare of society as a whole. 1974. It seems that some of the worst famines
The relation between low-income and food in- occurred with no significant fall in food avail-
take is two-way. Low income is a cause of mal- ability per head.
nutrition. Malnutrition in turn is a cause of low What does the entitlement to food depend on?
income by impairing working efficiency and pro- Above all, it depends on the ability of individuals
ductivity. All too little is known about the second to exchange productive resources and goods for
relationship. But we do know that the food re- food. This depends in turn on such factors as the
quirements considered by nutritionists to be neces- ownership and employment status of individuals
sary for efficient working and healthy living are far (for example, whether they are owners of land,
greater than the levels achieved by a large minority labourers, peasant farmers, sharecroppers, etc.);
of the population in developing countries. Calorie productivity; non-working income in the form of
deficiency causes loss of body weight, tiredness, subsidies and transfer payments; and on the terms
listlessness, and a deterioration of mental faculties. of trade between food and other goods. Exchange
Calories are also required for the absorption of entitlements may deteriorate independently of a
protein which is otherwise used up for energy. general decline in the supply of food which raises
Protein deficiency causes such conditions as its price and worsens the terms of trade for other
kwashiorkor (the bloated stomachs and staring goods. Job opportunities may diminish; real wages
eyes we see on our television screens), which may or productivity may fall, and other people may
cause death in children. Protein is particularly im- become better off and demand more food. During
 Development and Underdevelopment 3 7

the Great Bengal famine of 1943, 3 million people kept pace with population growth and there has
died, yet in terms of the total availability of food been no margin of safety or very little provision
grains, 1943 was not a subnormal year. Starvation for improving the distribution of food supplies. A
occurred because food entitlements shrunk as a dangerous dependence has grown up on North
result of first, a rise in the price of food due to America and the EEC. In 1972, for the first time
military procurement, secondly, because the price since the war, world food production actually fell.
of other commodities fell as more monetary de- In 1973 the world was threatened with a food
mand was switched to food, and thirdly, because crisis. Stocks of wheat amounted to only four
of a fall in the output of other goods. Because of weeks of world consumption. A crop failure in just
their ownership position, however, those in rural one major producing area would have spelt disas-
areas suffered less than the urban poor. The ab- ter. Disaster was averted by reasonably good
sence of famine in China and other socialist coun- weather and harvests. But the situation is still pre-
tries is not so much the result of increases in food carious and malnutrition persists. It is an intoler-
production per head, but the result of a shift in the able situation that year in and year out, at the end
entitlement system through guaranteed employ- of the twentieth century, millions of people should
ment and social security provisions. have their very lives threatened by the vagaries of
The policy message is that the alleviation of the weather, or by sudden political upheavals
famine requires the establishment and preserva- which impair their ability to import food. How
tion of adequate entitlements to food, not simply has this precarious situation arisen? The answer is
the provision of more food - important as that simple enough: through the neglect of agriculture.
may be. Public action requires programmes of If blame is to be apportioned it must lie jointly
food security which guarantee that people have with the international development agencies and
access to enough food at all times, and nutrition the developing countries themselves. In the early
programmes working through clinics targeted par- 1950s the developing countries saw industrialisa-
ticularly at children and pregnant women. 1 tion as the road to development and starved agri-
culture of resources. International development
agencies supported industrialisation plans and
• Food Production rarely lent money for agricultural projects. Tech-
nically, it is well within the world's grasp to in-
At the global level the problem of nutrition, and of crease agricultural production on an immense
food supply to those who need it, is also essentially scale. What is required is the initiative and politi-
a distribution problem. It is not a capacity prob- cal will, both within the international community
lem in the sense that the world is physically incap- and the developing countries, to make the radical
able of producing enough food to feed its changes necessary. As far as the international com-
inhabitants adequately. Since the Second World War munity is concerned, it could increase its foreign
global food production has generally kept pace aid for agricultural projects, especially where this
with population growth, and the world is prob- could lead to a significant breakthrough in agri-
ably capable of feeding itself ten times over if need cultural production. For example, the investment
be. The worrying factor that needs to be stressed, of $3 billion to eliminate the tsetse fly in infested
however, is that most of the increase in world food areas of tropical Africa could open up seven mil-
production that has taken place in recent years has lion square kilometres to livestock and crop pro-
been in the granary of North America. Food pro- duction. There must also be genuine international
duction in the developing countries has barely co-operation and agreement to guarantee world
food supplies. One possibility would be to have a
1 See the work of Dreze and Sen (1989) for the hunger and
system of granaries strategically placed across the
poverty project of the World Institute for Development Econ- world under international supervision, which
omics Research (WIDER). could store the food surpluses of the 'north' and
38 Introduction

release them at time of need. This in no way need to low life expectancy and a high incidence of
preclude or hinder the fundamental agricultural infant and child mortality. Tables 1.6 and 1.7
reforms that everyone recognises are necessary in present some selected statistics showing changes in
many of the poorer countries in Africa and else- the level of food production per head; daily per
where if there is to be self-sustaining growth. The capita calorie intake as a percentage of require-
desirability of a development policy based on ments, and various health-related indicators in-
a healthy agricultural sector is emphasised in cluding the number of people that share a doctor,
Chapter 3. the percentage of the population with access to
Lack of adequate food supply and nutrition, safe water, life expectancy at birth and rates of
combined with rudimentary health facilities, leads infant and child mortality.

Table 1.6 Health Related Indicators: Food Production, Calorie Intake and Doctors

Average index of
food production Daily calorie supply Population per
per capita (per capita) doctor
(1979-81 = 100)
1988-90 1965 1989 1984

Low-income economies 119w 1975 w 2406 w 5 800 w

China and India 127 w 1966 w 2464 w 1650 w
Other low-income 105 w 1994 w 2298 w 14160 w

Mozambique 81 1712 1 680 ..

Tanzania 88 1 831 2206 24 970
Ethiopia 84 1 853 1 667 78 780
Somalia 94 1 718 1 906 19 950
Nepal 115 1 889 2 077 30 220

Chad 85 2 395 1 743 38 390

Bhutan 93 .. .. 9 730
Lao PDR 114 2135 2 630 1 360
Malawi 83 2 259 2 139 11 340
Bangladesh 96 1970 2 021 6 390
Burundi 92 2 131 1 932 21 020
Zaire 97 2 187 1991 13 540
Uganda 95 2 361 2 153 ..
Madagascar 88 2 447 2 158 9 780
Sierra Leone 89 2 014 1799 13 620

Mali 97 1938 2 314 25 390

Nigeria 106 2 185 2 312 6 410
Niger 71 1996 2 308 39 670
Rwanda 77 1 856 1 971 35 090
Burkina Faso 114 1 882 2 288 57183

India 119 2 021 2229 2 520

Benin 112 2 019 2 305 15 940
China 133 1 929 2 639 1 010
 Development and Underdevelopment 39

Table 1.6 Health Related Indicators: Food Production, Calorie Intake and Doctors (continued)

Average index of
food production Daily calorie supply Population per
per capita (per capita) doctor
(1979-81 = 100)
1988-90 1965 1989 1984

Haiti 94 2 045 2 013 7140

Kenya 106 2 208 2 163 10 050

Pakistan 101 1 773 2 219 2 900

Ghana 97 1 937 2 248 20390
Central African Rep. 91 2 055 2 036 ..
Togo 88 2454 2 214 8 700
Zambia 103 2 072 2 077 7150

Guinea 87 2 187 2132 ..

Sri Lanka 87 2 171 2 277 5 520
Mauritania 85 1903 2 685 11 900
Lesotho 86 2 049 2 299 18 610
Indonesia 123 1 791 2 750 9 410

Honduras 83 1967 2 247 1 510

Egypt, Arab Rep. 118 2 399 3 336 770
Afghanistan 85 2 304 .. ..
Cambodia 165 2 292 2 166 ..
Liberia 84 2 158 2 382 9340

Myanmar 93 1 897 2440 3 740

Sudan 71 1 938 1 974 10190
Vietnam 127 2 041 2 233 950

Middle-income economies 102 w 2 489 w 2 860 w 2 250 w

Lower-middle-income 98 w 2 415 w 2 768 w 3 000 w

Bolivia 109 1 868 1 916 1 530

Zimbabwe 94 2 075 2 299 6 700
Senegal 102 2 372 2 369 ..
Philippines 84 1 875 2 375 6 570
Cote d'Ivoire 101 2 352 2 577 ..
Dominican Rep. 90 1 834 2 359 1 770
Papua New Guinea 103 1996 2 403 6 070
Guatemala 91 2 026 2 235 2 180
Morocco 128 2112 3 020 4 730
Cameroon 89 2 011 2 217 ..
Ecuador 100 2 191 2 531 810
Syrian Arab Rep. 80 2177 3 003 1 250
Congo 94 2 260 2 590 ..
El Salvador 97 1 853 2 317 2 830
Paraguay 116 2 586 2 757 1 460
40 Introduction

Table 1.6 Health Related Indicators: Food Production, Calorie Intake and Doctors (continued)

Average index of
food production Daily calorie supply Population per
per capita (per capita) doctor
(1979-81 = 100)
1988-90 1965 1989 1984

Peru 100 2 323 2 186 1 040

Jordan 100 2277 2 634 860
Colombia 104 2179 2 598 1 230
Thailand 106 2138 2 316 6 290
Tunisia 87 2 217 3 121 2150

Jamaica 95 2 232 2 609 2 040

Turkey 97 2 698 3 236 1390
Romania 92 2 988 3 155 570
Poland 109 3 292 3 505 490
Panama 90 2 241 2 539 1 000
Costa Rica 91 2 367 2 808 960
Chile 113 2 581 2 581 1 230
Botswana 75 2 045 2 375 6 900

Algeria 96 1 701 2 866 2 340

Bulgaria 96 3 443 3 707 280
Mauritius 100 2 269 2 887 1 900
Malaysia 147 2 353 2 774 1 930
Argentina 93 3 163 3 113 370

Iran, Islamic Rep. 104 2 060 3 181 2 840

Albania 92 2 374 2 761 ..
Angola 81 1 907 1 807 17 750
Lebanon 135 2 485 .. ..
Mongolia 86 2 364 2 479 ..
Namibia 93 1900 1 946 ..
Nicaragua 58 2 305 2265 1 500

Upper-middle-income 109 w 2 584 w 2987w 940 w

Mexico 102 2 570 3 052 ..

South Africa 87 2 759 3 122 ..
Venezuela 96 2266 2 582 700
Uruguay 109 2 812 2 653 510
Brazil 115 2 417 2 751 1 080

Hungary 113 3 134 3 644 310

Yugoslavia 95 3 243 3 634 550
Czechoslovakia 119 3 397 3 632 280
Gabon 84 1 950 2 383 2 790
Trinidad and Tobago 87 2 496 2 853 940
 Development and Underdevelopment 41

Table 1.6 Health Related Indicators: Food Production, Calorie Intake and Doctors (continued)

Average index of
food production Daily calorie supply Population per
per capita (per capita) doctor
(1979-81 = 100)
1988-90 1965 1989 1984
Portugal 106 2 647 3 495 140
Korea, Rep. 106 2 178 2 852 1160
Greece 103 3 019 3 825 350
Saudi Arabia 189 1 850 2 874 730
Iraq 92 2150 2 887 1 740
Libya 78 1 875 3 324 690
Oman .. .. .. 1700
High-income economies 100 w 3 091 w 3 409w 470w
OECD members 101 w 3 099 w 3 417 w 460w
Other SOw 2546 w 3 072 w 880 w
Ireland 109 3 605 3 778 680
Israel 95 2 799 3 174 350
Spain 112 2 770 3 572 320
Singapore 69 2 285 3 198 1410
Hong Kong 80 2 486 2 853 1 070
New Zealand 102 3 238 3 362 580
Belgium 108 .. .. 330
United Kingdom 105 3 304 3 149 ..
Italy 94 3 097 3 504 230
Australia 95 3 053 3 216 440
Netherlands 111 3 024 3 151 450
Austria 106 3 244 3 495 390
France 103 3 355 3 465 320
United Arab Emirates .. 2 639 3 309 1 020
Canada 108 3 127 3 482 510
United States 92 3 234 3 671 470
Denmark 126 3 420 3 628 400
Germany 112 3 088 3 443 380
Norway 100 3 036 3 326 450
Sweden 99 2 930 2 960 390

Japan 101 2 668 2 956 660

Finland 105 3 126 3 253 440
Switzerland 101 3 471 3 562 700
Kuwait .. 2 766 3 195 640
World 112w 2 383 w 2 711 w 4200w
104 w 2093 w 2642 w 4480 w

w means weighted average.

Source: World Development Report 1992.
42 Introduction

Table 1.7 Life Expectancy

Under-5
mortality rate Life expectancy at birth (years)
(per 1000 live births)
Female Male
Female Male
1990 1990 1965 1990 1965 1990

Low-income economies 91 w 98w SOw 62w 48 w 61 w

China and India 69w 72w 52w 66w SOw 65 w
Other low-income 131w 145 w 45 w 56w 44w 54w

Mozambique 194 215 39 48 36 45

Tanzania 182 203 45 49 41 46
Ethiopia 185 205 43 50 42 46
Somalia 200 223 40 50 37 47
Nepal 183 175 40 51 41 53

Chad 198 221 38 49 35 45

Bhutan 183 179 40 47 41 50
Lao PDR 159 179 42 51 39 48
Malawi 242 255 40 47 38 46
Bangladesh 160 142 44 51 45 52
Burundi 167 187 44 48 41 45
Zaire 143 162 45 54 42 50
Uganda 185 206 48 47 46 46
Madagascar 160 178 45 52 42 50
Sierra Leone 236 261 34 44 31 40
Mali 209 238 39 50 37 46
Nigeria 152 171 43 54 40 49
Niger 204 227 38 47 35 44
Rwanda 192 213 45 50 42 47
Burkina Faso 190 210 40 49 37 46
India 121 116 44 58 46 60
Benin 155 173 43 52 41 49
China 29 40 57 71 53 69
Haiti 126 144 47 56 44 53
Kenya 97 112 50 61 46 57
Pakistan 151 145 45 55 47 56
Ghana 127 144 49 57 46 53
Central African Rep. 156 176 41 51 40 48
Togo 133 151 44 55 40 52
Zambia 123 140 46 52 43 48
Guinea 221 245 36 43 34 43
Sri Lanka 21 26 64 73 63 69
Mauritania 193 215 39 48 36 45
 Development and Underdevelopment 43

Table 1.7 Life Expectancy (continued)

Under-5
mortality rate Life expectancy at birth (years)
(per 1000 live births)
Female Male
Female Male
1990 1990 1965 1990 1965 1990

Lesotho 125 142 50 57 47 55

Indonesia 75 90 45 64 43 60
Honduras 70 85 51 67 48 63
Egypt, Arab Rep. 95 110 50 62 48 59
Afghanistan 241 .. .. .. .. ..
Cambodia 161 180 46 52 43 49
Liberia 168 193 46 56 43 53
Myanmar 78 94 49 64 46 59
Sudan 159 178 41 52 39 49
VietNam 46 59 51 69 48 64
Middle-income economies 57w 68 w 60w 69w 56w 64 w
Lower-middle-income 62w 73 w 58 w 67w 55 w 63 w
Bolivia 109 127 47 62 42 58
Zimbabwe 66 78 50 63 46 59
Senegal 120 137 42 49 40 46
Philippines 45 57 57 66 54 62
Cote d'Ivoire 126 144 44 57 40 54

Dominican Rep. 68 75 57 69 54 65
Papua New Guinea 70 84 44 56 44 54
Guatemala 76 91 50 66 48 61
Morocco 84 99 51 64 48 60
Cameroon 117 134 47 59 44 55
Ecuador 58 72 57 68 55 64
Syrian Arab Rep. 55 67 54 68 51 64
Congo 172 185 47 56 41 50
El Salvador 63 76 56 68 53 60
Paraguay 33 44 67 69 63 65
Peru 78 93 52 65 49 61
Jordan 62 68 52 69 49 66
Colombia 40 49 61 72 57 66
Thailand 28 38 58 68 54 63
Tunisia 50 63 52 68 51 66

Jamaica 16 22 67 75 64 71
Turkey 73 80 55 69 52 64
Romania 23 32 70 73 66 67
44 Introduction

Table 1.7 Life Expectancy (continued)

Under-5
mortality rate Life expectancy at birth (years)
(per 1000 live births)
Female Male
Female Male
1990 1990 1965 1990 1965 1990
Poland 18 23 72 75 66 67
Panama 21 29 65 75 62 71
Costa Rica 18 22 66 78 63 73
Chile 18 23 63 76 57 69
Botswana 41 53 49 69 46 65
Algeria 83 91 51 66 49 65
Bulgaria 14 19 73 76 68 70
Mauritius 21 28 63 73 59 67
Malaysia 17 22 60 72 56 68
Argentina 30 40 69 75 63 68
Iran, Islamic Rep. 103 122 52 63 52 63
Albania 28 33 67 75 65 70
Angola 207 230 37 48 34 44
Lebanon 0 0 0 0 64 0 0 60 0 0

Mongolia 76 91 51 64 49 61
Namibia 119 140 47 59 44 56
Nicaragua 66 80 52 66 49 63
Yemen, Rep. 172 191 41 49 39 48
Upper-middle-income 49w 60w 62w 71w 58 w 65 w
Mexico 41 51 61 73 58 66
South Africa 81 98 54 65 49 59
Venezuela 36 45 65 73 61 67
Uruguay 22 28 72 77 65 70
Brazil 62 75 59 69 55 63
Hungary 16 22 72 75 67 67
Yugoslavia 25 30 68 76 64 69
Czechoslovakia 13 17 73 75 67 68
Gabon 148 167 44 55 41 52
Trinidad and Tobago 25 34 67 74 63 69
Portugal 14 17 68 78 62 72
Korea, Rep. 17 24 58 73 55 67
Greece 13 15 72 80 69 74
Saudi Arabia 72 87 50 66 47 63
Iraq 81 89 53 66 51 61
Libya 84 100 51 64 48 60
Oman 36 46 45 68 43 64
 Development and Underdevelopment 45

Table 1.7 Life Expectancy (continued)

Under-5
mortality rate Life expectancy at birth (years)
(per 1000 live births)
Female Male
Female Male
1990 1990 1965 1990 1965 1990

High-income economies 9w 12w 74 w SOw 68w 74w

OECD members 9w 11w 74 w SOw 68w 74w
Other 14 w 18 w 70w 77w 65 w 73 w
Ireland 8 10 73 77 69 72
Israel 11 15 74 78 71 74
Spain 9 12 74 79 69 73
Singapore 7 10 68 77 64 71
Hong Kong 7 10 71 80 64 75
New Zealand 10 15 74 79 68 72
Belgium 10 12 74 80 68 73
United Kingdom 9 12 74 78 68 73
Italy 10 12 73 80 68 75
Australia 8 11 74 80 68 74
Netherlands 8 10 76 80 71 74
Austria 9 13 73. 80 66 73
France 8 10 75 81 68 73
United Arab Emirates 23 32 59 74 56 69
Canada 7 9 75 81 69 74
United States 10 13 74 80 67 73
Denmark 9 11 75 78 70 73
Germany 8 11 73 80 67 73
Norway 9 11 76 81 71 74
Sweden 6 8 76 81 72 75
Japan 5 7 73 82 68 76
Finland 7 9 73 79 66 73
Switzerland 7 9 75 82 69 75
Kuwait 14 20 65 76 61 72

World 64w ?Ow 58 w 67w 55 w 64w

w stands for weighted average.
Source: World Development Report 1992.
46 Introduction

As far as food production per capita is con- poorest countries life expectancy is little more
cerned, it has risen on average by just over 1 per than 40 years. The infant and child mortality rates
cent per annum in the last ten years in the low- in poor countries are particularly high relative to
income countries, but hardly at all if China and the rich countries. Infant mortality is higher than
India are excluded from the sample. In the middle- 100 per 1000 in some poor countries compared
income countries progress has also been very slow with 8 per 1000 in rich countries. And while child
and in some low and middle-income countries per mortality averages ten per 1000 in the rich coun-
capita food production actually fell. tries, it averages 70 per 1000 in the low-income
Daily per capita calorie intake shows most of countries (excluding China and India).
the poorest countries below the required level of
2500, while the 'obese' industrial countries have an
average intake of 40 per cent above requirements. • Education
The average people to doctor ratio in developed
countries is 500 : 1, while in low-income and There has been an enormous growth in public
middle-income countries over 5000 people share expenditure on education in developing countries
one doctor. in recent years - some would say too much - but
In the light of the above facts, it is no surprise expenditure per capita is still only one-twelfth of
that life expectancy is low in developing countries that in developed countries. The statistics in Table
compared to the rich industrialised countries. In 1.8 indicate the relative underprovision of facili-
low- and middle-income countries we find that life ties and opportunities in poor countries, and the
expectancy at birth averages 63 years compared to still low rate of literacy in the poorest countries.
77 years in developed countries. In many of the The adult literacy rate in the low-income countries

Table 1.8 Education Statistics

Percentage of age group enrolled in education

Primt;try Secondary

Total Total Tertiary (total)

1965 1989 1965 1989 1965 1989

Low-income economies 73 w 105 w 20 w 38 w 2w ..

China and India 83 w 119w 25 w 44 w 2w ..
Other low-income 50 w 77w 10 w 28 w 1w 4w

Mozambique 37 68 3 5 0 0
Tanzania 32 63 2 4 0 0
Ethiopia 11 38 2 15 0 1
Somalia 10 .. 2 .. 0 ..
Nepal 20 86 5 30 1 6

Chad 34 57 1 7 .. 1
Bhutan 7 26 0 5 .. ..
Lao PDR 40 111 2 27 0 2
Malawi 44 67 2 4 0 1
Bangladesh 49 70 13 17 1 4
 Development and Underdevelopment 4 7

Table 1.8 Education Statistics (continued)

Percentage of age group enrolled in education

Primary Secondary

Total Total Tertiary (total)

1965 1989 1965 1989 1965 1989

Burundi 26 71 1 4 0 1
Zaire 70 78 5 24 0 2
Uganda 67 77 4 13 0 1
Madagascar 65 92 8 19 1 4
Sierra Leone 29 53 5 18 0 1

Mali 24 23 4 6 0 ..
Nigeria 32 70 5 19 0 3
Niger 11 28 1 6 .. 1
Rwanda 53 69 2 7 0 1
Burkina Faso 12 35 1 7 0 1

India 74 98 27 43 5 ..
Benin 34 65 3 .. 0 2
China 89 135 24 44 0 2
Haiti so 84 5 19 0 ..
Kenya 54 94 4 23 0 2

Pakistan 40 38 12 20 2 5
Ghana 69 75 13 39 1 2
Central African Rep. 56 64 2 11 .. 1
Togo 55 103 5 22 0 3
Zambia 53 95 7 20 .. 2

Guinea 31 34 5 9 0 1
Sri Lanka 93 107 35 74 2 4
Mauritania 13 51 1 16 .. 3
Lesotho 94 110 4 26 0 4
Indonesia 72 118 12 47 1 ..

Honduras 80 108 10 .. 1 10
Egypt, Arab Rep. 75 97 26 81 7 20
Afghanistan 16 24 2 8 0 1
Cambodia 77 .. 9 .. 1 ..
Liberia 41 .. 5 .. 1 3

Myanmar 71 103 15 24 1 5
Sudan 29 .. 4 .. 1 3
48 Introduction

Table 1.8 Education Statistics (continued)

Percentage of age group enrolled in education

Primary Secondary

Total Total Tertiary (total)

1965 1989 1965 1989 1965 1989

Middle-income economies 93 w 102 w 26 w 55 w 7w 17 w

Lower-middle-income 88 w 101 w 26 w 54 w 7w 17 w

Bolivia 73 81 18 34 5 23
Zimbabwe 110 125 6 52 0 6
Senegal 40 58 7 16 1 3
Philippines 113 111 41 73 19 28
Cote d'Ivoire 60 .. 6 20 0 ..
Dominican Rep. 87 95 12 .. 2 ..
Papua New Guinea 44 73 4 13 .. ..
Guatemala 50 79 8 21 2 ..
Morocco 57 68 11 36 1 11
Cameroon 94 101 5 26 0 3

Ecuador 91 118 17 56 3 25
Syrian Arab Rep. 78 108 28 54 8 20
Congo 114 .. 10 .. 1 6
El Salvador 82 78 17 26 2 17
Paraguay 102 106 13 29 4 8

Peru 99 123 25 67 8 32
Jordan 95 .. 38 .. 2 ..
Colombia 84 107 17 52 3 14
Thailand 78 86 14 28 2 16
Tunisia 91 115 16 44 2 8

Jamaica 109 105 51 61 3 5

Turkey 101 112 16 51 4 13
Romania 101 95 39 88 10 9

Poland 104 99 69 81 18 20
Panama 102 107 34 59 7 22
Costa Rica 106 100 24 41 6 27
Chile 124 100 34 75 6 19
Botswana 65 111 3 37 .. 3

Algeria 68 94 7 61 1 11
Bulgaria 103 97 54 75 17 26
Mauritius 101 103 26 53 3 2
Malaysia 90 96 28 59 2 7
Argentina 101 111 28 74 14 41
 Development and Underdevelopment 49

Table 1.8 Education Statistics (continued)

Percentage of age group enrolled in education

Primary Secondary

Total Total Tertiary (total)

1965 1989 1965 1989 1965 1989

Iran, Islamic Rep. 63 109 18 53 2 7

Albania 92 99 33 80 8 9
Angola 39 94 5 11 0 ..
Lebanon 106 .. 26 .. 14 ..
Mongolia 98 98 66 .. 8 ..
Namibia .. .. .. .. .. ..
Nicaragua 69 99 14 43 2 8
Yemen, Rep. 13 .. 3 .. .. ..

Upper-middle-income 99 w 104 w 26 w 56 w Sw 17 w

Mexico 92 114 17 53 4 15
South Africa 90 .. 15 .. 4 ..
Venezuela 94 105 27 56 7 28
Uruguay 106 106 44 77 8 50
Brazil 108 105 16 39 2 11

Hungary 101 94 .. 76 13 15
Yugoslavia 106 95 65 80 13 19
Czechoslovakia 99 92 29 87 14 18
Gabon 134 .. 11 .. .. 4
Trinidad and Tobago 93 97 36 83 2 6

Portugal 84 111 42 53 5 18
Korea, Rep. 101 108 35 86 6 38
Greece 110 102 49 97 10 28
Saudi Arabia 24 76 4 46 1 12
Iraq 74 96 28 47 4 14

Libya 78 .. 14 .. 1 ..
Oman .. 102 .. 48 .. 4

High-income economies 104 w 105 w 61 w 95 w 21 w 42 w

OECD members 104 w 105 w 63 w 95 w 21 w 43 w
Other 99 w 103 w 39 w 77w 11w 24 w

Ireland 108 101 51 97 12 26

Israel 95 93 48 83 20 33
50 Introduction

Table 1.8 Education Statistics (continued)

Percentage of age group enrolled in education

Primary Secondary

Total Total Tertiary (total)

1965 1989 1965 1989 1965 1989

Spain 115 111 38 105 6 32

Singapore 105 110 45 69 10 ..
Hong Kong 103 105 29 73 5 ..
New Zealand 106 106 75 88 15 41
Belgium 109 101 75 99 15 34
United Kingdom 92 107 66 82 12 24
Italy 112 96 47 78 11 29
Australia 99 106 62 82 16 32

Netherlands 104 116 61 103 17 32

Austria 106 104 52 82 9 31
France 134 113 56 97 18 37
United Arab Emirates .. 111 .. 64 0 9
Canada 105 105 56 105 26 66

United States 100 .. .. .. 40 63

Denmark 98 98 83 109 14 32
Germany .. 103 .. 97 .. 32
Norway 97 98 64 98 11 36
Sweden 95 104 62 91 13 31

Japan 100 102 82 96 13 31

Finland 92 99 76 112 11 43
Switzerland 87 .. 37 .. 8 26
Kuwait 116 100 52 90 .. 18

World 85 w 105 w 31 w 52 w 9w 16 w
w: means weighted average.
Source: World Development Report, 1992.

averages just 60 per cent compared to 78 per cent education, but the drop-out rate even from prim-
in the middle-income countries and 99 per cent in ary schools is often high. Secondary education is
rich countries. While there has been a major im- still a luxury in most developing countries, as is
provement in the literacy rate in most countries higher education. This is not to say the more edu-
over the last two decades, the absolute number of cational provision, the better. Education confers
those illiterate has actually increased. Except in undoubted benefits on individuals and societies,
the poorest countries, most children now have but it can also confer costs by exacerbating certain
access to some form of rudimentary primary development difficulties. Education inculcates
 Development and Underdevelopment 51

knowledge and skills which raise productivity, but way as to make the structural transformation of
it can also perpetuate inequalities in societies and their economies in the direction of industrial de-
impart values, attitudes and aspirations which are velopment more difficult. There may be a genuine
inimical to development; for example, the adop- dilemma here, and a trade-off between growth and
tion of practices and institutional structures in- basic needs. It depends on whether the scale of
appropriate to the environment of the country. resource transfer increases as the allocation ·is
Societies need to look carefully at the nature and changed and on the degree of complementarity
allocation of educational provision in relation to between the two strategies. On the one hand it can
the needs and aspirations of the society. It would be argued that the provision of basic needs is a
be hard to argue, however, that a higher rate of form of consumption transfer- away from invest-
literacy and numeracy and more training in voca- ment - so that growth will be retarded and the
tional skills would not be in the private and social basic needs strategy will then not be sustainable in
interest. The problem again is one of competing the long run. On the other hand it can be argued
claims on limited resources. that the provision of basic needs is a form of
investment in human capital, which may be as
productive as investment in industry.
• Basic Needs Whatever the conflict here, it can also be
argued, as Singh (1979) does convincingly, that
The provision of health services, education, hous- there must be an important complementarity and
ing, sanitation, water supply and adequate nutri- interrelationship between meeting basic needs, in-
tion, came to be known in development circles in dustrialisation and accelerated structural change,
the 1970s (and supported by the World Bank) as and both strategies need to be, and can be, pur-
the basic needs approach to economic develop- sued side by side. Specifically, that to meet basic
ment. The rationale of the approach was that the needs on a sustainable basis it is necessary for
direct provision of such goods and services is likely productive structures to be transformed in favour
to relieve absolute poverty more immediately than of industry, and that a properly conceived basic
alternative strategies which simply attempt to needs strategy which achieves a more equitable
accelerate growth or which rely on raising the distribution of income should help industrialisa-
incomes and productivity of the poor. Arguments tion.
used in support of this change in strategy were as The International Labour Organisation (ILO)
follows: growth strategies usually fail to benefit has emphasised in its various publications (see, for
those intended; the productivity and incomes of example, Meeting Basic Needs) that the satisfac-
the poor depend in the first place on the direct tion of basic needs depends crucially on the estab-
provision of health and education facilities; it rna¥ lishment of a New International Economic Order,
take a long time to increase the incomes of the which according to the Lima Declaration of 1975
poor so that they can afford basic needs; in any has as its aim to increase the developing countries'
case, the poor tend not to spend their income share of world manufacturing output to 25 per
wisely and certain facilities such as water supply cent by the year 2000 (compared to 15 per cent
and sanitation can only be provided publicly; today). Singh shows that the provision of basic
lastly, it is difficult to help all the poor in a uniform needs on a sustainable basis requires accelerated
way in the absence of the provision of basic needs. growth of 7-8 per cent per annum, which in turn
While these arguments undoubtedly contained implies a growth of manufacturing output of 10 to
an element of truth, there was some suspicion 11 per cent. This is the rate of growth of manufac-
within the developing countries themselves that turing needed to reach the Lima target of a 25 per
the international propagation of this new doctrine cent share of world manufacturing output by the
was an attack on their sovereignty and would alter year 2000. Thus the basic needs strategy and
the nature of international assistance in such a structural transformation in the direction of indus-
52 Introduction

try lead to the same conclusion as far as industrial Development Report which gives alternative mea-
growth is concerned. Whether both objectives are sures of the economic well-being or progress of
achieved largely depends on the overall economic nations which do not necessarily accord with the
strategy pursued. The experience of several coun- usual measure of the level or growth of income per
tries shows that the rapid development of industry head. As the UNDP Report says 'although GNP
and the provision of basic needs are quite compat- growth is absolutely necessary to meet all essential
ible. China has achieved remarkable progress in human objectives, countries differ in the way that
both directions by recognising the importance of they translate growth into human development'.
gearing industrial development to the needs of The UNDP defines human development as a pro-
agriculture and the rural sector, by encouraging cess of 'enlarging people's choices'. This depends
the use of modern inputs into agriculture and by not only on income but also on social indicators
encouraging the development of small scale rural such as health provision, education, leisure time
industries. Other, non-socialist, countries have a and so on. The UNDP thus constructs a Human
good record in the provision of basic needs judged Development Index which combines a measure of
by the level of life-expectancy in relation to per income per head with measures of life expectancy
capita income. These tend to be societies where and adult literacy. Countries are then ranked by
income is more equitably distributed or which the index and compared with their income per
make special provision, such as Sri Lanka. Coun- head ranking. The results are shown in Table 1.9,
tries with a poor performance in the provision of ranking from lowest to highest. It can be seen that
basic needs illustrate that a basic needs strategy some countries that rank low by per capita income
needs to be comprehensive, otherwise failure in rank high by the human development index, and
one respect will nullify progress in others. 1 vice versa. In the former category of countries are:
World Bank lending for poverty alleviation Costa Rica, Cuba, Mexico, Venezuela, Jamaica,
programmes rose quite dramatically during the China, Chile, Uruguay and Sri-Lanka. In the latter
1970s and 1980s. The amount spent on poverty category of countries, note that the United States
alleviation programmes rose from US $500 mil- ranks only 19th on the human development index,
lion in 1970 to $3300 million in 1989, or from 8 but second according to the level of per capita
to 20 per cent of the total lending programme. Of income. Other rich countries which score low on
the total in 1989, $1800 million was spent on human development are Kuwait, United Arab
basic needs including education, health and water Emirates, Libya and Saudi Arabia.
supply compared to $340 million in 1970. The The Human Development Index (HOI) is con-
rest was spent on rural development and small- structed as follows. First a deprivation index is
scale industrial projects. Industry's share of total constructed for each of the three variables - per
lending remained at just over 30 per cent; while capita income, life expectancy and adult literacy.
the share of total lending going to infrastructure The deprivation index is measured as the differ-
fell from 58 per cent to 37 per cent between 1970 ence between the desirable (maximum) value of
and 1989. the index minus the actual value of the index
divided by the difference between the desirable
(maximum) and minimum values of the index
• Human Development Index (those actually observed across countries). The
bigger the difference between desirable and actual,
In 1990, the United Nations Development Pro- the higher the degree of deprivation. The depriva-
gramme (UNDP) produced its first annual Human tion indices are then averaged, and the human
1 For an evaluation of country experience, and of the com-
development index is taken as 1 -average depriva-
plementarities and trade-offs in the provision of basic needs, see tion index. To give an example: if the maximum
Streeten eta/. (1981), and Stewart (1985). (desirable) adult literacy index is 100 per cent, the
 Development and Underdevelopment 53

Table 1.9 Human Development Index

"
~
..
Jl"" """
l......
.
c~
" c., t~ c=...
...
~
t" ~
c.,
~
~ ~
~
~
~.,c ~~ "'-•'ij' ~
3 ~ ""'~
""·
'ij'
~ ,_
~
~
~-b ~~
~~ ~~ .£~ E ~~ Q ·- .£~ E
"~ ""'~
"~
., ."!;; .:::~
~..., -"-
..e~ ~""
~l; "'( l! o.::!::.
......
~ """'""
~z

~" l@ ~ ."!;; .:::~

~..., ..@ ~
~t; "'( l!
-"-
~""
o.::!::.
......
~ """'""
~z

~" l@
Niger 45 14 452 0.116 20 1 China 70 69 2 124 0.716 22 66
Mali 45 17 543 0.143 15 2 Libya 62 66 7 250 0.719 103 67
Burkina Faso 48 14 500 0.150 13 3 Sourh Africa 61 70 4 981 0.731 82 68
Sierra Leone 42 30 480 0.150 27 4 Lebanon 68 78 2 250 0.735 78 69
Chad 46 26 400 0.157 4 5 Mongolia 64 90 2 000 0.737 57 70
Guinea 43 29 500 0.162 31 6 Nicaragua 64 88 2 209 0.743 54 71
Somalia 46 12 1 000 0.200 23 7 Turkey 65 74 3 781 0.751 71 72
Mauritania 47 17 840 0.208 40 8 Jordan 67 75 3 161 0.752 76 73
Afghanistan 42 24 1 000 0.212 17 9 Peru 63 85 3 129 0.753 74 74
Benin 47 27 665 0.224 28 10 Ecuador 66 83 2 687 0.758 68 75
Burundi 50 35 450 0.235 18 11 Iraq 65 89 2 400 0.759 96 76
Bhutan 49 25 700 0.236 3 12 United Arab Emirates 71 60 12 191 0.782 127 77
Mozambique 47 39 500 0.239 10 13 Thailand 66 91 2 576 0.783 55 78
Malawi 48 42 476 0.250 7 14 Paraguay 67 88 2 603 0.784 65 79
Sudan 51 23 750 0.255 32 15 Brazil 65 78 4 307 0.784 85 80
Central African Republic 46 41 591 0.258 29 16 Mauritius 69 83 2 617 0.788 75 81
Nepal 52 26 722 0.273 8 17 Norrh Korea 70 90 2 000 0.789 67 82
Senegal 47 28 1 068 0.274 43 18 Sri Lanka 71 87 2 053 0.789 38 83
Ethiopia 42 66 454 0.282 1 19 Albania 72 85 2 000 0.790 61 84
Zaire 53 62 220 0.294 5 20 Malaysia 70 74 3 849 0.800 80 85
Rwanda 49 47 571 0.304 26 21 Colombia 65 88 3 524 0.801 72 86
Angola 45 41 1 000 0.304 58 22 Jamaica 74 82 2 506 0.824 62 87
Bangladesh 52 33 883 0.318 6 23 Kuwait 73 70 13 843 0.839 122 88
Nigeria 51 43 "668 0.322 36 24 Venezuela 70 87 4 306 0.861 95 89
Yemen Arab Rep. 52 25 1 250 0.328 47 25 Romania 71 96 3 000 0.863 84 90
Liberia 55 35 696 0.333 42 26 Mexico 69 90 4 624 0.876 81 91
Togo 54 41 670 0.337 24 27 Cuba 74 96 2 500 0.877 66 92
Uganda 52 58 511 0.354 21 28 Panama 72 89 4 009 0.883 88 93
Haiti 55 38 775 0.356 34 29 Trinidad and Tobago 71 96 3 664 0.885 100 94
Ghana 55 54 481 0.360 37 30 Portugal 74 85 5 597 0.899 94 95
Yemen, PDR 52 42 1 000 0.369 39 31 Singapore 73 86 12 790 0.899 110 96
Cote d'Ivoire 53 42 1 123 0.393 52 32 South Korea 70 95 4 832 0.903 92 97
Congo 49 63 756 0.395 59 33 Poland 72 98 4 000 0.910 83 98
Namibia 56 30 1500 0.404 60 34 Argentina 71 96 4 647 0.910 89 99
Tanzania 54 75 405 0.413 12 35 Yugoslavia 72 92 5 000 0.913 90 100
Pakistan 58 30 1 585 0.423 33 36 Hungary 71 98 4 500 0.915 87 101
India 59 43 1 053 0.439 25 37 Uruguay 71 95 5 063 0.916 86 102
Madagascar 54 68 634 0.440 14 38 Costa Rica 75 93 3 760 0.916 77 103
Papua New Guinea 55 45 1 843 0.471 50 39 Bulgaria 72 93 4 750 0.918 99 104
Kampuchea, Dem. 49 75 1 000 0.471 2 40 USSR 70 99 6 000 0.920 101 105
Cameroon 52 61 1 381 0.474 64 41 Czechoslovakia 72 98 7 750 0.931 102 106
Kenya 59 60 794 0.481 30 42 Chile 72 98 4 862 0.931 73 107
Zambia 54 76 717 0.481 19 43 Hongkong 76 88 13 906 0.936 111 108
Morocco 62 34 1 761 0.489 48 44 Greece 76 93 5 500 0.949 98 109
Egypt 62 45 1 357 0.501 49 45 East Germany 74 99 8 000 0.953 115 110
Laos 49 84 1 000 0.506 9 46 Israel 76 95 9 182 0.957 108 111
Gabon 52 62 2 068 0.525 93 47 USA 76 96 17 615 0.961 129 112
Oman 57 30 7 750 0.535 104 48 Austria 74 99 12 386 0.961 118 113
Bolivia 54 75 1 380 0.548 44 49 Ireland 74 99 8 566 0.961 106 114
54 Introduction

Table 1.9 Human Development Index (continued)

"
,.._!?0
"""'
'~..., "
,.._!?0
'"""'
~...,

~~ ... "" .,'-'Vi'... G'

... r--:'
.,._, '"~"'
!:! ~
G' "oo
"'-".:;;- '"~"'
..., 3 ~
"OO
.:;;- ...,
~-C
~~ ~ ,_ ""~"' ~~ ~'-
"'- ""
~~ ...,.,._
"'t '"' ~~ '-'l £~ '"'
tl~ ..:!::~ '-'l
-Q... ...... ""'Q..,
~
..ooc_ ~~ .:!::L -Q... ...... ""'Q...
~
..ooc_
~~ "Z ~§
~...,
-6~ "Z ~§
~
~...,
~Q.., ~Q..,
;.:J ~ "( ~ ~~ ~ ~'-' ~~ "( ~ ~~ ~'-'

Burma 61 79 752 0.561 11 50 Spain 77 95 8 989 0.965 105 115

Honduras 65 59 1119 0.563 53 51 Belgium 75 99 13 140 0.966 116 116
Zimbabwe 59 74 1184 0.576 45 52 Italy 76 97 10 682 0.966 112 117
Lesotho 57 73 1 585 0.580 35 53 New Zealand 75 99 10 541 0.966 109 118
Indonesia 57 74 1 660 0.591 41 54 West Germany 75 99 14 730 0.967 120 119
Guatemala 63 55 1 957 0.592 63 55 Finland 75 99 12 795 0.967 121 120
Vietnam 62 80 1 000 0.608 16 56 Britain 76 99 12 270 0.970 113 121
Algeria 63 50 2 633 0.609 91 57 Denmark 76 99 15 119 0.971 123 122
Botswana 59 71 2 496 0.646 69 58 France 76 99 13 961 0.974 119 123
El Salvador 64 72 1 733 0.651 56 59 Australia 76 99 11 782 0.978 114 124
Tunisia 66 55 2 741 0.657 70 60 Norway 77 99 15 940 0.983 128 125
Iran 66 51 3 300 0.660 97 61 Canada 77 99 16 375 0.983 124 126
Syria 66 60 3 250 0.691 79 62 Holland 77 99 12 661 0.984 117 127
Dominican Rep. 67 78 1 750 0.699 51 63 Switzerland 77 99 15 403 0.986 130 128
Saudi Arabia 64 55 8 320 0.702 107 64 Sweden 77 99 13 780 0.987 125 129
Philippines 64 86 1 878 0.714 46 65 Japan 78 99 13 135 0.996 126 130

minimum is 50 per cent, and the actual is 70 per modities' with a high income elasticity of demand.
cent, then literacy deprivation is measured as Naturally enough in this schema, the develop-
(100 - 70) I (100 - 50) = 30/50 = 0.6. If this was ing countries get identified with primary produc-
the only index, the HDI would be 1 - 0.6 = 0.4. tion, the more developed countries with the
production of manufactured goods, and the ma-
ture developed economies with a high percentage
• The Stages of Development of their resources in the service sector. There can
be no dispute that resource shifts are an integral
It is often argued that countries pass through part of the development process, and that one of
phases during the course of development and that the main determinants of these shifts is a difference
by identifying these stages, according to certain in the income elasticity of demand for commod-
characteristics, a country can be deemed to have ities and changes in elasticity as development pro-
reached a certain stage of development. The sim- ceeds. But just as care must be taken to equate
plest stage theory is the sector thesis of Fisher (without qualification) development and welfare
((1933) and (1939)) and Clark (1940), who em- with the level of per capita income, so, too, cau-
ploy the distinction between primary, secondary tion must be exercised in identifying different de-
and tertiary production as a basis of a theory of grees of underdevelopment, industrialisation and
development. Countries are assumed to start as maturity with some fairly rigid proportion of re-
primary producers and then, as the basic necessi- sources engaged in different types of activity. Such
ties of life are met, resources shift into manufac- an association would ignore the doctrine of com-
turing or secondary activities. Finally, with rising parative advantage which holds that countries will
income, more leisure and an increasingly saturated specialise in the production of those commodities
market for manufactured goods, resources move in which they have a relative advantage as deter-
into service or tertiary activities producing 'com- mined by natural or acquired resource endow-
 Development and Underdevelopment 55

ments. The fact that one country produces pre- very different levels of development.
dominantly agricultural products while another Ideally a criterion of development stages is re-
produces mainly manufactured goods need not quired which leaves the proportion of resources
imply that they are at different stages of develop- employed in different activities out of account.
ment on any of the conventional definitions of One possibility is to argue that a country has
development we gave earlier. Such an association reached a developed state when productivity in the
would also ignore the different types of service agricultural sector matches productivity in the in-
activities which may exist at different stages of a dustrial sector, and that it has reached a state of
country's history. There are three broad categories maturity when productivity in all sectors, includ-
of service activities, and the determinants of re- ing services, is approximately equal provided the
source allocation to service activities accompany- level is reasonably high. The alternative is simply
ing development may operate differently on each to classify countries as industrial, semi-industrial
in an offsetting manner. Newer service activities and non-industrial, using as a criterion for division
linked with the growth of leisure and high mass some level of the net value of manufacturing pro-
consumption tend to have a very high income duction per head of the total population combined
elasticity of demand; services linked to the growth perhaps with an indicator of the degree of indus-
of manufacturing also grow but at a declining rate, trialisation of exports (see Maizels, 1963).
and traditional services of pre-industrial times de- Having said all this, however, the fact remains
cline (Katouzian (1970)). In short, tertiary produc- that there is a good deal of empirical support for
tion is an aggregation of many dissimilar service the Fisher-Clark view that the pattern of develop-
activities some of which are related to low per ment across countries evidences many common
capita incomes and some to high per capita in- characteristics, especially the shift of resources
comes. Thus the same proportion of total re- from agriculture to industry. Figure 1.3 shows the
sources devoted to services may be associated with proportions of the labour force engaged in agricul-

Figure 1.3 Labor Force Distribution

Percentage
of labor
force
80

70

Services
60 /1980

50 Services
..--·'"1965
/
/
Industry
40 /
/
1965
Industry
1980
30

20
Ag ricu ltu re
1965
10 Agriculture
1980
QL-------~--~--------~~~--------~~---
Low income Middle income High income
$300 $1000 $1 0 000
Per capita income
56 Introduction

ture, industry and services between low-income, to make estimates of the income elasticity of de-
middle-income and industrialised countries, and mand for different commodities (given by b). An
over time. One sees the broad thesis of Fisher and income elasticity of demand for a good less than
Clark confirmed. In the low-income countries over unity would imply that its proportional import-
70 per cent of the labour force on average is em- ance in total output would decline as income
ployed in agriculture, while only 10 per cent is grows. Conversely, an income elasticity greater
employed in industry. By contrast in the industrial than unity means that its relative importance in
market economies only 7 per cent on average is total output will increase. Taking this basic equa-
employed in agriculture and 35 per cent in indus- tion (with some modifications), and applying it to
try. The proportion of the labour force in services a cross section of 51 countries, Chenery found that
is also relatively low in low-income countries com- the growth elasticity of the agricultural sector is
pared to more mature industrial countries, less than 0.5, while for industry it is over 1.3 and
although there is quite a wide variation between for services it is approximately unity. Within the
low-income countries. Over time, there is a notice- industrial sector, there will also be differences in
able reduction in the proportion of the labour the income elasticity of demand for products
force in agriculture in most countries, but particu- which will cause the pattern of industry to change
larly in the industrial market economies. There is a as development proceeds. The most notable de-
slight increase in the importance of industry in the mand shift is the relative switch from basic neces-
low- and middle-income countries, but not in the sities .like food, beverages and clothes to capital
industrial market economies where on average the and consumer durable goods.
shift of resources has been away from both agri-
culture and industry towards services, which em-
ploy on average over one half of the labour force. • Industrialisation and Growth
Generally speaking the lower the per capita in-
come, the higher the proportion of the labour The importance attached to industrialisation by
force in agriculture, and the higher the level of per developing countries lies in the close association
capita income, the higher the proportion in ser- that appears to exist between industrialisation and
vices. real income per head, and between the growth of
What is true of the sectoral distribution of the industry and the growth of output as a whole. This
labour force is also true of the sectoral distribution latter observed relationship is summed up in the
of output, although the magnitude of the propor- maxim 'manufacturing as the engine of growth'. If
tions differ because productivity differs markedly we relate the average growth of gross domestic
between sectors (see Table 1.10). Because pro- product (gcDP) to the average growth of manufac-
ductivity tends to be lower in agriculture than in turing industry (g1) over the period 1970-77 for 81
industry, except in some special cases such as Aus- countries, the following regression equation is
tralia and Canada, the proportion of total output obtained:
generated by agriculture tends to be lower than its
share of the labour force, and the proportion gGDP = 1.414 + 0.569 g1 r 2 = 0.610
generated by industry tends to be higher than its (0.051)
share of the labour force. (See also Cody, Hughes
and Wall (1980).) This is a highly significant relationship and is
Chenery and others (Chenery (1960) and (1979), confirmed by many other studies. 1 Since the re-
Chenery and Syrquin (1975)) have documented the gression coefficient is significantly less than unity,
changing sectoral share of output using regression
analysis. Using an estimating equation of the form 1 See the Symposium on Kaldor's growth laws edited by the
log V = log a + b log Y, where V is value-added present author in journal of Post-Keynesian Economics, Spring
per capita and Y is per capita income, it is possible 1983.
 Development and Underdevelopment 57

Table 1.10 Structure of Production

Distribution of gross domestic product (per cent)

Agriculture Industry Manufacturing Services, etc.

1965 1990 1965 1990 1965 1990 1965 1990

Low-income economies 41 w 31 w 26 w 36 w 19 w 27 w 32 w 35 w
China and India 41 w 29 w 29 w 36 w 22 w 30 w 30 w 35 w
Other low-income 42 w 30 w 20 w 34 w Sw .. 38 w 38 w

Mozambique .. 65 .. 15 .. .. .. 21
Tanzania 46 59 14 12 8 10 40 29
Ethiopia 58 41 14 17 7 11 28 42
Somalia 71 65 6 9 3 5 24 26
Nepal 65 60 11 14 3 5 23 26

Chad 42 38 15 17 12 14 43 45
Bhutan .. 43 .. 27 .. 10 .. 29
Malawi 50 33 13 20 .. 14 37 46
Bangladesh 53 38 11 15 5 9 36 46

Burundi .. 56 .. 15 .. 10 .. 29
Zaire 20 30 32 33 .. 13 48 36
Uganda 52 67 13 7 8 4 35 26
Madagascar 25 33 14 13 .. 12 61 54
Sierra Leone 34 32 28 13 6 6 38 55

Mali 65 46 9 13 5 8 25 41
Nigeria 55 36 12 38 5 7 33 25
Niger 68 36 3 13 2 5 29 51
Rwanda 75 38 7 22 2 15 18 40
Burkina Faso 37 32 24 24 11 14 39 44

India 44 31 22 29 16 19 34 40
Benin 59 37 8 15 .. 7 33 48
China 38 27 35 42 28 38 27 31
Kenya 35 28 18 21 11 11 47 51

Pakistan 40 26 20 25 14 17 40 49
Ghana 44 48 19 16 10 9 38 37
Central African Rep. 46 42 16 17 4 .. 38 41
Togo 45 33 21 22 10 9 34 46
Zambia 14 17 54 55 6 43 32 29

Guinea .. 28 .. 33 .. 4 .. 39
Sri Lanka 28 26 21 26 17 15 51 48
Mauritania 32 26 36 29 4 .. 32 44
Lesotho 65 24 5 30 1 14 30 46
Indonesia 51 22 13 40 8 20 36 38
58 Introduction

Table 1.10 Structure of Production (continued)

Distribution of gross domestic product (per cent)

Agriculture Industry Manufacturing Services, etc.

1965 1990 1965 1990 1965 1990 1965 1990

Honduras 40 23 19 24 12 16 41 53
Egypt, Arab Rep. 29 17 27 29 .. 16 45 53

Middle-income economies 19 w 12w 34 w 37 w 20 w .. 46 w 50 w

Lower-middle-income 22 w 17 w 32 w 31 w 20 w .. 44 w 50 w

Bolivia 23 24 31 32 15 13 46 44
Zimbabwe 18 13 35 40 20 26 47 47
Senegal 25 21 18 18 14 13 56 61
Philippines 26 22 27 35 20 25 47 43
Cote d'Ivoire 47 47 19 27 11 .. 33 26

Dominican Rep. 23 17 22 27 16 13 55 56
Papua New Guinea 42 29 18 31 .. 12 41 40
Guatemala .. 26 .. 19 .. .. .. 55
Morocco 23 16 28 33 16 18 49 51
Cameroon 33 27 20 28 10 13 47 46

Ecuador 27 13 22 42 18 23 50 45
Syrian Arab. Rep. 29 28 22 22 .. .. 49 50
Congo 19 13 19 39 .. 7 62 48
El Salvador 29 11 22 21 18 19 49 67
Paraguay 37 28 19 23 16 23 45 49
Peru 18 7 30 37 17 27 53 57
Jordan .. 8 .. 26 .. 12 .. 66
Colombia 27 17 27 32 19 21 47 51
Thailand 32 12 23 39 14 26 45 48
Tunisia 22 16 24 32 9 17 54 52

Jamaica 10 5 37 46 17 20 53 49
Turkey 34 18 25 33 16 24 41 49
Romania .. 18 .. 48 .. .. .. 34

Poland .. 14 .. 36 .. .. .. 50
Panama 18 10 19 9 12 7 63 80
Costa Rica 24 16 23 26 .. 19 53 58
Chile 9 .. 40 .. 24 .. 52 ..
Botswana 34 3 19 57 12 6 47 40

Algeria .. 13 .. 47 .. 12 .. 41
Bulgaria .. 18 .. 52 .. .. .. 31
Mauritius 16 12 23 33 14 24 61 55
Malaysia 28 .. 25 .. 9 .. 47 ..
Argentina 17 13 42 41 33 .. 42 45
 Development and Underdevelopment 59

Table 1.10 Structure of Production (continued)

Distribution of gross domestic product (per cent)

Agriculture Industry Manufacturing Services, etc.

1965 1990 1965 1990 1965 1990 1965 1990

Iran, Islamic Rep. 26 21 36 21 12 8 38 58

Angola .. 13 .. 44 .. 4 .. 43
Lebanon 12 .. 21 .. .. .. 67 ..
Mongolia .. 17 .. 34 .. .. .. 49

Namibia .. 11 .. 38 .. 5 .. 50
Nicaragua 25 .. 24 .. 18 .. 51 ..
Yemen, Rep. .. 20 .. 28 .. 8 .. 47

Upper-middle-income 16 w 9w 36 w 40 w 19 w 25 w 47 w 51 w

Mexico 14 9 27 30 20 23 59 61
South Africa 10 5 41 44 24 26 48 51
Venezuela 6 6 40 50 .. 20 55 45
Uruguay 18 11 35 34 .. 28 47 55
Brazil 19 10 33 39 26 26 48 51

Hungary .. 12 .. 32 .. 27 .. 56
Yugoslavia 23 12 42 48 .. .. 35 40
Czechoslovakia .. 8 .. 56 .. .. .. 36
Gabon 26 9 34 49 7 7 40 42
Trinidad and Tobago 8 3 48 48 .. 13 44 49

Portugal .. .. .. .. .. .. .. ..
Korea, Rep. 38 9 25 45 18 31 37 46
Greece 24 17 26 27 16 14 49 56
Saudi Arabia 8 8 60 45 9 9 31 48
Iraq 18 .. 46 .. 8 .. 36 ..

Libya 5 .. 63 .. 3 .. 33 ..
Oman 61 3 23 80 0 4 16 18

High-income economies 5w .. 43 w .. 32 w .. 54 w ..
OECD members 5w .. 43 w .. 32 w .. 54 w ..
Other .. .. .. .. .. .. .. ..
Ireland .. .. .. .. .. .. .. ..
Israel .. .. .. .. .. .. .. ..
Spain .. .. .. .. .. .. .. ..
Singapore 3 0 24 37 15 29 74 63
Hong Kong 2 0 40 26 24 18 58 73
60 Introduction

Table 1.10 Structure of Production (continued)

Distribution of gross domestic product (per cent)

Agriculture Industry Manufacturing Services, etc.

1965 1990 1965 1990 1965 1990 1965 1990

New Zealand .. 9 .. 27 .. 19 .. 65
Belgium .. 2 .. 31 .. 23 .. 67
United Kingdom 3 .. 46 .. 34 .. 51 ..
Italy .. 4 .. 33 .. 23 .. 63
Australia 9 4 39 31 26 15 51 64

Netherlands .. 4 .. 31 .. 20 .. 65
Austria 9 3 46 37 33 27 45 60
France .. 4 .. 29 .. 21 .. 67
United Arab Emirates .. 2 .. 55 .. 9 .. 43
Canada 6 .. 40 .. 26 .. 54 ..

United States 3 .. 38 .. 28 .. 59 ..
Denmark 9 5 36 28 23 19 55 67
Germany 4 2 53 39 40 31 43 59
Norway .. .. .. .. .. .. .. ..
Sweden .. 3 .. 35 .. 24 .. 62

Japan 10 3 44 42 34 29 46 56
Finland 16 6 37 36 23 23 47 58
Switzerland .. .. .. .. .. .. .. ..
Kuwait 0 1 70 56 3 9 29 43

World 10 w .. 41 w .. 30 w .. 51 w ..
w means weighted averages.
Source: World Development Report 1992.

the equation also implies that the greater the ex- facturing output itself owing to the existence of
cess of manufacturing output growth over the rate increasing returns, both static and dynamic. Static
of growth of the economy as a whole, the faster returns relate to the size and scale of production
the overall growth rate will be. Setting gGDP = g 1 units and are a characteristic largely of manufac-
gives the growth rate which divides those coun- turing where in the process of doubling the linear
tries where industry is growing faster than overall dimensions of equipment, the surface increases by
output and those countries where industry is the square and the volume by the cube. Dynamic
growing slower. In the above sample, that growth economies refer to increasing returns brought
rate is 3.3 per cent (i.e. 1.414/(1 - 0.569)). about by 'induced' technical progress, learning by
There are two good reasons for expecting a doing, external economies in production and so
strong relation between the growth of manufac- on. Manufacturing seems to be the sector where
turing industry and the growth of the overall econ- major cost saving, technical advances take place.
omy. The first is that productivity growth in The relationship between the growth of pro-
industry is closely related to the growth of manu- ductivity and the growth of industry is sometimes
 Development and Underdevelopment 61

referred to in the literature as Verdoorn's Law. strong negative correlation would be evidence of
This relationship can provide the basis of models convergence.
of geographic dualism (see Chapter 5). The second The results of testing the catch-up hypothesis
induced effect that manufacturing growth has on are mixed. A study by Baumol (1986) shows a
overall growth is that the faster manufacturing strong inverse correlation between a country's
grows, the faster the rate of transference of labour productivity level and its average growth of pro-
from other sectors of the economy where there is ductivity among industrialised countries and those
either diminishing returns or where no rela- at an intermediate stage of development, but there
tionship exists between employment growth and is no evidence of convergence as far as the poorer
output growth because there is already surplus countries are concerned. Another study of 113
labour. In both cases, a reduction in the amount of countries by Dowrick (1992) shows that while
labour in these sectors will raise productivity there is some evidence of catch-up in the sense that
growth outside manufacturing. growth rates are negatively related to initial levels
The question then is: what determines the rate of productivity, other differences have caused per
of growth of manufacturing? In the early stages of capita income growth to be faster the higher the
development the impetus to industrialisation must level of per capita income, producing a divergence
come from the agricultural sector which provides in living standards across the world (as we saw
the main source of autonomous demand for in- earlier in the chapter). Countries with higher levels
dustrial goods. In the later stages of development, of income have had higher investment ratios and a
the demand for industrial goods from outside the faster growth of the workforce which has con-
country becomes of prime importance for main- tributed to a faster growth of output.
taining the momentum of industrial growth. We Clearly a productivity gap itself is not a suf-
will develop and elaborate these points later when ficient condition for catch up. There have to be the
we consider the role of agriculture in development cooperating factors to enable poor countries to
in Chapter 3 and the idea of export-led growth in take advantage of the more advanced technology
Chapters 5 and 15. available, and they have to innovate as well.

I Will Developing Countries Ever

Catch Up?
• Rostow's Stages of Growth
Interest in stage theories of development was given
If living standards are largely determined by the new impetus with the publication of Rostow's
level and growth of productivity in industry, the book The Stages of Economic Growth (1960),
interesting question is whether the developing which represents an ambitious attempt to provide
countries will ever catch up with the performance an alternative to the Marxist interpretation of his-
of the advanced industrialised countries. It is tory - hence its subtitle 'A Non-Communist Mani-
sometimes argued that the larger the gap between festo'. Rostow presents a political theory as well
a country's technology, productivity and per capi- as a descriptive economic study of the pattern of
ta income on the one hand and the level of produc- the growth and development of nations. A brief
tivity in the advanced countries on the other, the summary of his main points will provide a useful
greater the scope for a poor country to absorb introduction to the next few chapters on the
existing technology and to 'catch up' with richer sources of growth and development.
countries. In other words, a process of converg- The essence of the Rostow thesis is that it is
ence is predicted. One test is to do a simple cor- logically and practically possible to identify stages
relation between the level of productivity or per of development and to classify societies according
capita income and the growth rate of countries. A to those stages. He distinguishes five such stages:
62 Introduction

traditional, transitional, take-off, maturity and the factory and the principle of the division of
high mass consumption. labour.
All we need say about traditional societies is On the social front a new elite must emerge to
that for Rostow the whole of the pre-Newtonian fabricate the industrial society and it must super-
world consisted of such societies; for example, the sede in authority the land-based elite of the
dynasties of China, the civilisations of the Middle traditional society. Surplus product must be chan-
East, the Mediterranean and medieval Europe, etc. nelled by the new elite from agriculture to indus-
Traditional societies are characterised by a ceiling try, and there must be a willingness to take risks
on productivity imposed by the limitations of and to respond to material incentives. And be-
science. Traditional societies are thus recognisable cause of the enormity of the task of transition, the
by a very high proportion of the work-force in establishment of an effective modern government
agriculture (greater than 75 per cent), coupled is vital. The length of the transition phase depends
with very little mobility or social change, great on the speed with which local talent, energy and
divisions of wealth and decentralised political resources are devoted to modernisation and the
power. Today there are very few, if any, societies overthrow of the old order, and in this respect
that one would class as traditional. Most societies political leadership will have an important part to
emerged from the traditional stage, as described play.
by Rostow, some time ago, mainly under the im- Then there is the stage of take-off. The charac-
pact of external challenge and aggression or national- teristics of take-off are sometimes difficult to dis-
ism. The exceptions to the pattern of emergence tinguish from the characteristics of the transition
from the traditional state are those countries stage, and this has been one bone of contention
which Rostow describes as having been 'born between Rostow and critics. None the less, let us
free', such as the United States and certain British describe the take-off stage as Rostow sees it - a
dominions. Here the preconditions of 'take-off' 'stage' to which reference is constantly made in the
were laid in more simple fashion by the construc- development literature. Since the pre-conditions of
tion of social overhead capital and the introduc- take-off have been met in the transitional stage,
tion of industry from abroad. But for the rest of the take-off stage is a short stage of development
the world, change was much more basic and fun- during which growth becomes self-sustaining. In-
damental, consisting not only of economic trans- vestment must rise to a level in excess of 10 per
formation but also a political and social transition cent of national income in order for per capita
from feudalism. income to rise sufficiently to guarantee adequate
The stage between feudalism and take-off Ros- future levels of saving and investment. Also im-
tow calls the transitional stage. The main econ- portant is the establishment of what Rostow calls
omic requirement in the transition phase is that the 'leading growth sectors'. Historically, domestic
level of investment should be raised to at least 10 finance for take-off seems to have come from two
per cent of national income to ensure self- main sources. The first has been from a diversion
sustaining growth. (On this particular point, as we of part of the product of agriculture by land re-
shall see, there seems to be very little difference form and other means. The examples of Tsarist
between the transition stage and the later stage of Russia and Meiji Japan are quoted, where govern-
take-off.) The main direction of investment must ment bonds were substituted for the landowner's
be in transport and other social overhead capital claim to the flow of rent payments. A second
to build up society's infrastructure. The precondi- source has been from enterprising landlords volun-
tions of a rise in the investment ratio consist of a tarily ploughing back rents into commerce and
willingness of people to lend risk capital, the avail- industry.
ability of men willing and able to be entrepreneurs In practice the development of major export
and to innovate, and the willingness of society at industries has sometimes led to take-off permitting
large to operate an economic system geared to substantial capital imports. Grain in the United
 Development and Underdevelopment 63

States, Russia and Canada; timber in Sweden, and, ditional society had not been made. The dates
to a lesser extent textiles in Great Britain are cited of take-off for some of the present developed
as examples. Countries such as the United States, countries are given as follows: Great Britain,
Russia, Sweden and Canada also benefited during 1783-1802; France, 1840-60; the United States,
take-off from substantial inflows of foreign capi- 1843-60; Germany, 1850-73; Sweden, 1868-90;
tal. The sector or sectors which led to the take-off Japan, 1878-1900; Russia, 1890-1914.
seem to have varied from country to country, but Then there is the stage of maturity which Ros-
in many countries railway building seems to have tow defines as the period when society has effec-
been prominent. Certainly improvement of the tively applied the range of modern technology to
internal means of communication is crucial for an the bulk of its resources. During the period of
expansion of markets and to facilitate exports, maturity new leading sectors replace the old, and
apart from any direct impact on such industries as Rostow sees the development of the steel industry
coal, iron and engineering. But Rostow argues that as one of the symbols of maturity. In this respect
any industry can play the role of leading sector in America, Germany, France and Great Britain en-
the take-off stage provided four conditions are tered the stage of maturity roughly together.
met: first, that the market for the product should Accompanying changes in the industrial struc-
be expanding fast to provide a firm basis for the ture will be structural changes in society such as
growth of output; second, that the leading sector changes in the distribution of the work-force; the
generates secondary expansion; third, that the growth of an urban population; an increase in the
sector has an adequate and continual supply of proportion of white-collar workers; and a switch
capital from ploughed-back profits; and last, that in industrial leadership from the entrepreneur to
new production functions can be continually in- the manager.
troduced into the sector, meaning scope for in- Maturity also has important political features.
creased productivity. This is the period when nations grow confident
Rostow contends that the beginnings of take-off and exert themselves - witness Germany under
in most countries can be traced to a particular Bismarck and Russia under Stalin. This is also the
sharp stimulus which has taken many different period when fundamental political choices have to
forms, such as a technological innovation or more be made by society on the use to which greater
obviously a political revolution, e.g. Germany in wealth should be put. Should it be devoted to high
1848, the Meiji restoration in Japan in 1868, mass consumption, the building of a welfare state,
China in 1949 and Indian independence in 1947. or to imperialist ends? The balance between these
Rostow is at pains to emphasise, however, that possibilities has varied over time within countries,
there is no one single pattern or sequence for as well as varying between countries. Ultimately,
take-off. Thus there is no need for the developing however, every nation will presumably reach the
countries today to recapitulate the course of events stage of high mass consumption whatever the
in, say, Great Britain, Russia or America. The balance of choices at the stage of maturity. Since
crucial requirement is that the preconditions of the developing countries have no likelihood of
take-off are met, otherwise take-off, whatever reaching this stage in the foreseeable future, how-
form it takes, will be abortive. Investment must ever, and only a handful of countries have reached
rise to over 10 per cent of national income; one or it already, we shall not consider this fifth stage
more leading sectors must emerge; and there must here.
exist or emerge a political, social and institutional Instead, let us evaluate Rostow's thesis, and
framework which exploits the impulse to expan- consider the usefulness of this type of stage theory
sion. The examples are given of extensive railway apart from it providing a valuable description of
building in Argentina before 1914, and in India, the development process and pinpointing some of
China and Canada before 1895, failing to initiate the key growth variables. Most criticisms have
take-off because the full transition from a tra- hinged on whether a valid and operationally mean-
64 Introduction

ingful distinction can be made between stages of (1965), p. 227). And on the concept of the take-off
development, especially between the so-called stage in general Kuznets concludes that lack of
transitional phase and take-off, and between common experience typifying countries in the
take-off and maturity. Critics have attempted to take~off stage, in relation to investment, etc., 'casts
argue that the characteristics that Rostow dis- serious doubt on the validity of the definition of
tinguishes for his different stages are not unique to the take-off as a general stage of modern economic
those stages. Thus the demarcation between take- growth, distinct from what Professor Rostow calls
off and transition is blurred because the changes the precondition, or transition, stage preceding it
that take place in the transition phase also seem to and the self-sustaining growth stage following it'
take place in the take-off phase, and similarly with (Kuznets (1965)). Cairncross (1961) echoes these
the demarcation between take-off and maturity. remarks of Kuznets and appears to deliver a decis-
One of the most outspoken of Rostow's critics ive blow when he asks what, if the various stages
is Kuznets, and some of his criticisms may be overlap, is the meaning of a 'stage'?
quoted as representative of the criticisms that Ros- Are we to conclude from all this that Rostow's
tow has received in general. First, there is the contribution is of little value? The answer to this
difficulty of empirically testing the theory, which must be in the negative; at least, much can be
Rostow himself makes no attempt to do. For one salvaged. While growth stage theories may be
thing there is a general lack of quantitative evi- lacking in analytical power, the purpose of stage
dence for assertions made, and for another Ros- theory is not that the stages distinguished should
tow's description of the characteristics of some of necessarily have parallels in history, or be rigidly
the stages are not sufficiently specific to define the distinct, but to distinguish the situations in which
relevant empirical evidence even if data were avail- an economy may find itself- situations which may
able. With respect to the take-off stage, for merge into one another. While the concept of a
example, what is a 'political, social and insti- 'stage' may be quibbled with, and stage theory
tutional framework which exploits the impulses to dismissed as a blue-print for development, Rostow
expansion in the modern sector'? Kuznets argues: offers many extremely valuable insights into the
'it seems to me that Rostow . . . defines these development process. As we have seen from the
social phenomena as a complex that produces the work of Chenery and Maizels, development is not
effect he wishes to explain and then treats this entirely haphazard, and there are certain features
identification as if it were a meaningful identifica- of the development process which do follow a
tion' (Kuznets (1963), reprinted in Kuznets well-ordered sequence. Moreover, there are cer-
(1965), p. 219). Kuznets seems to be calling into tain development priorities which countries plan-
question the whole of Rostow's scientific method ning development may neglect at their peril. The
and is claiming as unscientific the practice of ob- importance of agriculture and the role of invest-
serving phenomena, developing hypotheses on the ment in raising the rate of growth are particularly
basis of the phenomena, and then using the stressed, as are certain political and sociological
phenomena to support the hypotheses! preconditions for development which economists
As regards quantitative evidence that is avail- are prone to forget, and which are ignored here.
able for testing hypotheses, Kuznets questions While emphasis on investment appears to be an
Rostow's figures of investment and the incremen- unfashionable doctrine in the developed countries
tal capital-output ratio during the take-off period at present, there exists no satisfactory counter-
in the countries studied. He says: 'Unless I have argument to the doctrine in developing countries if
completely misunderstood Professor Rostow's defi- capital is properly defined. If Rostow fails to pro-
nition of take-off, and its statistical characteristics, vide an analytical breakthrough, he has aroused
I can only conclude that the available evidence once again theoretical interest in the history and
lends no support to his suggestions' (Kuznets causes of the growth of the wealth of nations.
 Development and Underdevelopment 65

I Questions for Discussion and

Review
7. What are the developing countries asking for
by way of a New International Economic Order?
8. Outline and discuss the feasibility of the main
1. What are the major reasons why some coun- recommendations of the Brandt Commission
tries are rich and others poor? Report?
2. How would you measure the 'development 9. What major structural changes take place in
gap'? the course of development?
3. Why is the distribution of income within de- 10. Is any useful purpose served by defining stages
veloping countries more unequal than in de- of economic growth?
veloped countries? 11. What lessons, if any, can poor countries learn
4. What have been the causes of growing urban from the development experience of today's
unemployment in developing countries? industrialised countries?
5. What is meant by 'income measure' of unem- 12. Is there any evidence that the developing
ployment? countries are 'catching up' with the developed
6. What do you understand by the 'basic needs countries?
approach' to development?