Institute of

Development Studies

The Meaning of Development

Dudley Seers

IDS Communication 44, 1969

© Institute of Development Studies 1969
 INSirnJTE OF DEVELOPMENT
STUDIES IM

THE MEANING OF DEVELOPMENT

fey

DUDLEY SEERS.

Communication Series N o . 44
 1
THE MEANING OF DEVELOPMENT

by Dudley Seers
The challenges of any period depend on the tasks that face thos
living in it, I believe we have misconceived the nature of the main
challenge of the second half of the twentieth century. This has been
s e e n as achieving an increase in the national incomes of the ’’developing"
countries, formalised in the target of 9% growth rates s e t for the first
development decade. Of course, we have all been aware that development
c o n s i s t s of much e l s e b e s i d e s economic growth. Sir Arthur Lewis made
this point in the opening pages of "The Theory of Economic Growth" in
1955 , and the World Economic Survey for 1968 emphasises i t once more.
Yet little more than lipservice i s paid to i t ; we are still setting
targets mainly or only for the national income.

I t i s now time to make the point more sharply. Surely we could

hardly say that the situation depicted by one s e t of projections was
preferable to that shown by another s e t simply because the former implied
higher p e r capita incomes. After all, in what sense i s South Africa
mere developed than Ghana, o r Kuwait than the U.A.R., or the United
S t a t e s than Sweden?

Why do we concentrate on the national income in this way? I t is

of course convenient. Politicians find a single comprehensive measure
useful, especially one that is a t l e a s t a year out-of-date. Economists
are provided with a variable which can be quantified and movements in
which can b e analysed, into changes in sectoral output, factor shares
or c a t e g o r i e s o f expenditure, making model-building f e a s i b l e . While
i t is very slipshod f o r u s to confuse development with economic
development and economic development with economic growth, i t is
nevertheless very understandable. We can after all fall back on the
supposition that i n c r e a s e s in national income, if they are faster than
the population growth, sooner or later l e a d to the solution of social
and political problems.

1. I am grateful for comments from Hans Singer on an earlier draft,

which was also d i s c u s s e d at seminars a t the Universities of Boston and
Toronto.
 2

But the experience of the p a s t decade makes this belief look

rather naive. S o c i a l problems and political upheavals have emerged in
countries at all s t a g e s of development. Moreover, we can s e e that
t h e s e afflict countries with rapidly rising per capita incomes, as well
as t h o s e with stagnant economies. In fact it looks as if economic
growth may n o t merely fail to s o l v e s o c i a l and political difficulties;
certain types of growth can actually c a u s e them.

Nov; that the complexity of development problems i s becoming

increasingly obvious, this continued addiction to the u s e o f a single
aggregative yardstick, in the face of the e v i d e n c e , takes on a rather
different appearance. I t begins to look like a preference for avoiding
the real problems of development.

The definition of development

The starting point in d i s c u s s i n g the challenges we now f a c e is
to brush aside the web of fantasy we have woven around ‘development’
and decide more p r e c i s e l y what we mean by i t . Only then will we be
able to d e v i s e meaningful targets or measures o f p r o g r e s s , to judge
the relative importance of various problems which arise in the p r o c e s s
of development, and thus to help improve policy, national or international.

The starting point i s that we cannot avoid what the p o s i t i v i s t s

often disparagingly refer to as ’’value judgments”. ’Development 1 i s
inevitably t r e a t e d as a normative concept, as almost a synonym for
improvement. To pretend otherwise is j u s t to hide o n e ’ s value judgments.

But from where are these judgments to come? The conventional

answer, which Tinbergen accepts for his system of economic planning, is
to draw our values from governments. But governments have necessarily
a rather short-term view, in some c a s e s discounting the future at a
very high r a t e . Moro seriously, some governments are themselves the
main o b s t a c l e s to development, on any plausible definition, and once
one concedes t h i s , where i s one t o obtain the yardsticks by which
government o b j e c t i v e s are to be judged?”

Another approach is to copy the development paths o f ether

c o u n t r i e s , which implicitly takes their p r e s e n t s t a t e as a desirable

1. Even supposing that governments represented faithfully, in some s e n s e ,

popular attitudes, these are endogenous to the development p r o c e s s and
therefore cannot provide a means of a s s e s s i n g i t .
 3
objective. This i s implicit in much model-building, for example when
p r o j e c t i o n s are b a s e d on international cross-section analysis. But few
if any o f the rich countries appear to the outside world as really
desirable models. Some a s p e c t s , such as their housing standards, seem
are a s s o c i a t e d
enviable, but with these / , perhaps inseparably, evils such as urban
sprawl, commercialism and chronic t e n s i o n . B e s i d e s it i s by no means
obvious or even likely that the r e s t o f the world could retrace the
history of the industrial countries even if they wanted t o .

If values are not to be found in p o l i t i c s or history, does this

mean that we are e a c h l e f t to adopt our own personal s e t of values?
T h i s would b e an alarming conclusion, if i t were treated by civil
servants as a l i c e n s e to introduce their own p o l i t i c s into their
advice (which they do often enough without l i c e n s e l ) . But surely
the values we need arc staring u s in the f a c e , if we go back, appro
priately in 1 9 6 9 , to the q u e s t i o n s r a i s e d by G a n d h i ’ s thought and
life. We must ask o u r s e l v e s : what are the necessary c o n d i t i o n s for a
universally acceptable aim, the realisation of the potential of human
personality?

If we ask what is an absolute n e c e s s i t y for t h i s , one answer is

obvious - enough f o o d . Below certain l e v e l s of nutrition, a man
lacks n o t merely bodily energy and good health but even interest in
much b e s i d e s f o o d or the capacity to concentrate. He cannot r i s e
significantly above an animal existence. If anyone h a s any doubt on
the primacy of f o o d , they should r e f l e c t on the implications of r e c e n t
r e s e a r c h showing that nutritional shortages among children can cause
1
lasting impairment n o t merely o f the body, but also of the mind.
Gandhi’s repeated u s e o f phrases like ’’the dumb, semi-starved millions”
showed the attention h e paid to nutrition.

S i n c e to b e able to buy f o o d i s a matter of income, the criterion

c a n be e x p r e s s e d in terms of income l e v e l s . T h i s enables i t to take
account also of c e r t a i n other minimum requirements. People will never
spend all their money and energy on f o o d , however poor they a r e . To be

1. S e e "Malnutrition, Learning and B e h a v i o u r " , e d . Scrimshaw and

Gordon (MIT P r e s s , 1 9 6 8 ) .
 A-
enough to f e e d a man, his income has also to c o v e r basic needs of
clothing, footwear and s h e l t e r .

But I am not talking about consumption needs in general; I am

talking about the capacity to buy physical n e c e s s i t i e s . The deprivation
due to lack of food is n o t t o be compared with that due to lack of a
motor c a r . Y.rhat I am asserting technically is that below the level of
income at which a man can buy ( i n some s e n s e ) ’enough 1 food for his
family, the marginal utility of income is much greater than i t i s above
that l e v e l , We can argue about definition of income and specifications
of poverty l e v e l s , but a utility function must be of this general shape.

Another b a s i c n e c e s s i t y , in the sense of something without

which personality cannot develop, i s a j o b . T h i s does n o t j u s t mean
employment: i t c a n include studying, working on a family farm or keeping
house. But to play none of t h e s e accepted r o l e s , i . e . to be chronically
unemployed, to be chronically dependent on another p e r s o n ’ s productive
capacity, even f o r ' f o o d , i s incompatible with s e l f - r e s p e c t , especially
for somebody who has been spending years at s c h o o l , perhaps at univer
sity, preparing f o r an a c t i v e r o l e .

I t i s true o f c o u r s e t h a t b o t h poverty and unemployment are

a s s o c i a t e d in various ways with per c a p i t a income. If per. c a p i t a
incomes are falling, absolute poverty can hardly be reduced much, nor
c a n unemployment ( e x c e p t in the very short run and exceptional
circumstances). But certainly i n c r e a s e s in per capita income are far
from enough, as the experience o f petroleum economies show, to achieve
either of these o b j e c t i v e s . In f a c t , a r i s e in per capita income, as
we very well know, can be accompanied by, can even c a u s e , growing
un emp11 oy men 4t . 1

The direct link between p e r capita income and the numbers living
in poverty i s income distribution. I t i s a truism that poverty will
be eliminated much more rapidly if any given rate of economic growth is

1. Thus in Trinidad the growth in p e r capita, income averaged mere than

a year during the whole p e r i o d 1953 to 1?68, while overt unemployment
showed a steady increase to more then lC£o of the labour f o r c e .
 5
accompanied, by a declining concentration of incomes. Equality should
however b e c o n s i d e r e d an o b j e c t i v e in i t s own right, the third element
in development. Inequalities to b e found now in the world, especially
( b u t not only) outside the industrial countries, are objectionable by
any religious or ethical standards. The social barriers and inhibitions
of an unequal society distort the personalities of those with high
incomes no l e s s than of those v/ho are p o o r . Trivial differences of
accent, language, d r e s s , customs, e t c . acquire an absurd importance and
contempt i s engendered for those who lack social g r a c e s , especially
country dwellers. Perhaps even more important, since r a c e i s usually
highly c o r r e l a t e d with income as w e l l , economic inequality l i e s at the
h e a r t of racial tensions.

The questions to ask about a c o u n t r y ’ s development are therefore:

What has been happening to poverty? What has been happening to
unemployment? What has been happening to inequality? If all three of
these have declined from high l e v e l s , then beyond doubt this has been
a p e r i o d of development for the country concerned. If one or two of
t
t h e s e central problems have been growing worse, especially if all three
1
h a v e , i t would be strange to call the result ’development , even if per
capita income doubled. This applies o f course to the future t o o , A
’ p l a n ’ which conveys no targets for reducing poverty, unemployment and
inequality can hardly be c o n s i d e r e d a 'development plan’

Of c o u r s e , the fulfilment o f human potential requires much that

cannot b e s p e c i f i e d in purely economic t e r m s . I cannot spell out all
the other requirements, b u t i t vzould b e very unbalanced if I did not
mention them at all. They includ.e adequate educational
l e v e l s , freedom of speech, citizenship of a nation that is truly
independent, both economically and politically, in the sense that the
views o f other governments do n o t largely predetermine his own
government’s d e c i s i o n s .

1. Suppose, for example, that a p e r s p e c t i v e plan specified that the p e r

capita income of Brazil doubled in the next 30 y e a r s , but assumed no
change in distribution or in the proportion unemployed. Then at the
turn of the century, a b i g landowner in the Mato Cross o would run two
c a r s , instead of 1 , and a p e a s a n t in the North-East vzould have two p i e c e s
of meat a month instead of 1 . H i s son s t i l l might well b e o u t o f work.
Could we really call that ’development’?
 6

As undernourishment, unemployment and inequality dwindle, these

educational and political aims become increasingly important objectives
o f development. Later s t i l l , freedom from repressive sexual c o d e s , from
noise and pollution, become major a i m s . But these would not
override the b a s i c economic p r i o r i t i e s , at l e a s t for really poor countries,
with large numbers of undernourished children. A government could
hardly claim to b e ’developing* a country j u s t because its educational
system was being expended or a political order was being established, or
limits s e t on engine n o i s e , if hunger, unemployment a n d inequality were
significant and grov/ing, or even i f they were not diminishing. Indeed,
as we shall s e e , one would doubt the viability o f the political order
in these c i r c u m s t e n c e s , i f one d i d n ’ t consider the claim prima facie
somewhat s u s p e c t .

Before leaving this i s s u e I must make i t clear that the national

income i s not totally meaningless, j u s t because i t i s not an indicator
o f development. I t h a s some significance as a measure o f development
potential. Suppose that two countries start a decade with the same
per capita income and one grows faster than the other over ten y e a r s ,
but that the increase in income there goes entirely to the rich, and
that unemployment r a t e s remain unchanged in both c o u n t r i e s . Then
although the country with faster growth but greater inequality h a s , in
my view, fallen back in development compared with the other, one
could argue that i t had achieved greater development notentiaJ for
the future.

In the f i r s t p l a c e , the fiscal system could bring about develop

ment more rapidly, because of the income available for transfer to the
poor. Moreover, the country has a greater savings potential, which
could lead to a faster growth rate in the future, increasing the
possibilities of redistribution then. Indeed this country may well
already have h a d a higher level of investment per capita. I f this
investment has been in capital that i s needed for employment t o b e
increased or for s c h o o l s , this could mean that genuine development i s
already foreshadowed for the future.

1. Even for countries at a high level o f development in any s e n s e , the

u s e of national income as an indicator is being widely challenged, e . g .
by Mishan-, on the grounds that the environmental c o s t s are ignored.
 7
But o f course this may not materialise. If the investment has
been in modern capital-intensive techniques, unemployment and the
distribution o f income may well grow worse in the years ahead. To release
the development potential of a high rate of economic growth depends
therefore on the policy being followed. A country may have little or
no economic growth but be busy reshaping i t s institutionspolitical
so
a country
that, when growth comes, i t can be turned into development; such/probably
has a greater development potential than one with fast growth where
1
political power remains very firmly in the hands o f a rich minority.

Priorities in the social sciences

In order to withstand the s t r o n g intellectual attraction of the
aggregative approach we need to go more deeply into i t s origins. The
explanation l i e s largely in the f a c t that by about 1950 the great
economic problems had been brought largely under control in the indus
trial countries. Unemployment had been reduced to historically very low
l e v e l s ; absolute poverty in the sense I u s e the word had been largely
eliminated; taxation ano educational advances had reduced economic
inequalities, and, though a good deal of racial and s o c i a l inequality
remained, this was not a source o f great political concern at that
time, and i t was largely overlooked by the social s c i e n t i s t s , especially
the e c o n o m i s t s .

We could say that t h e s e countries had managed in various ways to

m e e t , in some degree at l e a s t , the challenges they had f a c e d in the 19th
century. One r e a s o n of course was that they benefitted from world
economic leadership and political power - to this I ’ l l return later. But
another was that social s c i e n t i s t s such as Booth, Rowntree, Boyd-Orr,
the Webbs, Keynes, Beveridge and Tawney f o c u s s e d attention sharply on

1. In an interesting paper published recently by the Banca Nazionale

del Lavoro, V . V . Divatia and V . V . Bhatt put forward a different index
of development potential, b a s e d on fundamental factor inputs such as
capital and skills (though this i s misleading described in p l a c e s as a
measure of the "pace of development”). It .foreshadows what the
future pace o f economic growth could b e . JThe index for India is encouraging
because i t shows a r a t e of increase twice as f a s t as the real national
incomeu But, of c o u r s e , i t does not follow that development potential
will be r e l e a s e d even in the sense of faster economic growth, l e t
alone faster development.
 8

poverty, unemployment and inequality in the first half of this century#

( l hope I am n o t being excessively nationalistic in choosing British
examples: the names are rather significant.) Most eoonomists, even Pigou
took greater equality as an obviously desirable objective.

the big
With the easing of/problems, hovzever, economists turned their
attention primarily to innovations in professional techniques. In as
far as they retained interest in current affairs, it vzas mainly in the
progress of the nation conceived as a whole. The national income seemed
ideal f o r comparing growth rates of a country during different periods,
or of different countries - i . e . for constructing a league table o f
competitive performance. Moreover, i t had a role as an indicator of the
level of employment - i f the economy i s diversified and the labour force
i s m o b i l e , big changes in the national income in the s h o r t term are
c l o s e l y a s s o c i a t e d with changes in employment.

We now s e e that even in the industrial countries basic economic

problems had n o t really been c u r e d . Their social scientists, notably
in the United S t a t e s , have recently been rediscovering their own problems
of poverty. But the fundamental problems have never even started to
Latin America
disappear from sight in the Third World. In Africa, Asia or/development
had been very limited on any of the three economic c r i t e r i a until 1950 •

S i n c e then, there has certainly been some reduction in proportion,

*2
even if not in the absolute numbers, living in p o v e r t y . ' 1' But unemploy
ment h a s grown almost everywhere. H. A. Turner has recently calculated
that, in 14 Third World countries for which usable figures are available,
the volume of overt unemployment has been growing at an average of about
2
8-|% a year, or more than three times the rate of population growth.
I t i s probable, though data are extremely p o o r , that in most countries
inequality has not b e e n reduced: in many, i t may well have increased.
A paper by A. J . Jaffe on five Latin American countries for which
comparable studies over time are available concludes that all showed
■i ■!
■ i w.
-iiriMia W'-rnw* i ■■i ■ ■■ »ea»r.» m m bi —

1. F o r India, s e e B . S . Minhas, D i s c u s s i o n Paper Mo. 42 of the India

Statistical Institute.

2. A paper read at a conference at the Insi'.itute of Development S t u d i e s .

This moreover i s despite the absorption of many young people in higher
education.
 9
increasing inequalities, with the possible exception of Mexico (and an
unpublished study by Senora de Navarette eliminates even this exception)
I t is even possible that if the data were available we would s e e
economic growth directly a s s o c i a t e d with growing unemployment and
increasing inequality; if that has indeed been the c a s e , there has
been a negative correlation between growth and development.

I t could b e argued that, in various ways, especially through

but also
investment and education, /through technical advances in agriculture
and increased understanding of problems, the basis was
being created for development some time in the future. But if
anything, the real strategic o b j e c t i v e s are even more remote now for many,
perhaps most, c o u n t r i e s , than they were 10 years ago. I t has recently
b e e n estimated by Francis Keppel that 7 out of every 10 children in the
entire world are ’’affected by the apathy typical of chronic protein
deficiency, an apathy which translates into diminished learning potential”;
the fraction among many countries o f the Third World, such as India,
must of c o u r s e b e higher. The n o s e s of the s o c i a l s c i e n t i s t and the
statistician should be rubbed into such social r e a l i t i e s during the
decades that l i e ahead. Our r e s e a r c h efforts must be devoted above
all to understanding the c a u s e s o f poverty, and the mechanisms by
which unemployment emerges and inequalities grow, as a b a s i s for genuine
development p l a n s .

Problems of measurement
The challenges for the remainder of this century arise out of the
analysis above. The f i r s t i s how to find measures of development to

1. Different conclusions are reached by different writers for different

periods in India, but there has certainly been no massive shift towards
equality, i f one allows f o r price changes. Two r e c e n t papers, one by
A. K, C h a t t e r j e e and by M. Mukherjee, and the other by N. Bhattacharya
and G-. 3 . C h a t t e r j e e , b a s e d on National Sample Survey data, suggest some
reduction in the G i n i coefficient of inequality for India in the late
195O’s; on the other hand, both p a p e r s a l s o point to relative price
increases for the lower income groups.
2. ’’Malnutrition, Learning and Behaviour" (op. cit.).
 10

replace the national income, or more precisely t o enable the national

income to be given its true, somewhat limited, significance.

I t has always been argued on behalf of national income as a

development indicator that i t could at l e a s t be quantified. But what
are all the voluminous tables of national income accounts really worth?
So far as the Third World i s concerned, large sections of what they ought
to cover are virtually outside the scope o f official stastistics. This
applies above all to . agricultural output for domestic consumption, even
the s t a p l e s , l e t alone subsidiary crops which come under the general
heading of "market gardening" (Amer, "truck farming"), not to speak of
fishes, f o r e s t products, e t c . Extremely rough methods o f estimation
are often u s e d , much o f th- output being assumed to r i s e in proportion
t o the rate o f increase in rural population, which i s also assumed, in
the absence o f registration of births and deaths, or data on migration.
Secondly, we know very little about the capital investment in the rural
sector which i s undertaken by the farming community i t s e l f ; this amounts
to a good deal if one takes account not only of building h o u s e s , but
wells and fences and
also clearing land, digging /ditches, cons true ting /ledges, e t c . Thirdly,
we know very little indeed about domestic service and other personal
service industries which are economically significant in such countries,
though much o f i t i s unpaid. Which of the non-marketed activities are
covered varies from country t o country.

If anyone estimates a national income increase o f a year, we

should ask him: How many fishes were caught in Province A in the years
concerned? How many huts were constructed in Province B? How many
barbers worked in Province C? And how do you know? How did you
decide whether activities a farm family does for itself without payment,
such as haircutting for example, should be included in the national
income?

We must also ask those who quote the national income, for example
in a planning office or a university, how much time they have spent with
the estimators? I t is unsafe and therefore unprofessional to u s e
national income data until one has personally s a t i s f i e d oneself on how
such questions were handled.
 11

I have examined the worksheets in about 20 countries; the blunt

truth of the matter i s that when one takes into account also the diffi
culties o f allowing for inventory changes and depreciation, and o f
deflating current-price data\ the published national income series for
a large number of countries have almost no relevance at all to economic
reality. In many c o u n t r i e s , any reasonably competent statistician
could produce s e r i e s showing the real p e r capita income either rising
or falling. Decimal p l a c e s are pure fantasy. It would of course be
very convenient if the national data published in such quantities had
objective meaning. Unfortunately this does not make them meaningful.
Some national income s e r i e s are in f a c t in a way more misleading than
s e t s of random numbers would b e , because they appear to have some
S1. . I„I. c ano c • 2

There i s one other major conceptual i s s u e about income s e r i e s ;

how appropriate are the prices which are u s e d for weighting physical
outputs? Although the national income i s often treated as an objective,
value-free indicator, i t is in f a c t by far the most value-loaded: every
type of product and service i s assigned i t s own particular weight (many
being z e r o ) . This weight i s determined by market f o r c e s , which reflect
the country* s income, distribution. A familiar question in economics is
how adequately disposable income measures demand, when i t s distribution
i s unequal. The question g e t s additional point when the distribution
i s as highly concentrated as i t i s in the countries of the Third World,
and when tastes are to some extent imported. B u t , in addition, official
policies, e . g . fostering import substitution by c o n t r o l s , often increase
the prices of luxuries much more than o f n e c e s s i t i e s . There are good
reasons for such policies, b u t , - although one of their
main motives may be egalitarian, the outcome is paradoxically that

1. Interesting work on the quality of price deflation which confirms

one*s doubts is being done by Mr. Makherjee at the Indi an Statistical
Institute.
2. There is an upward bias as w e l l . The share o f output covered by
official statistics, and included in the national income, tends to r i s e ,
partly because a growing proportion of output p a s s e s through the
hands of organised b u s i n e s s , which i s more adequately covered by
official statistics, but also partly because of the general improvement
in data collection.
 12

increases in production o f luxuries count very much more highly in the

e s t i m a t i o n of rates of economic growth than they do in industrial
countries. While p r i c e s of staple foods and clothing may be comparable
between poor countries and rich, perhaps lower in the former, prices of
c a r s , refrigerators, e t c . are several times as h i g h . I t is absurd if,
in a country where there i s s e r i o u s poverty, a car counts for more than
ten tons of r i c e . In addition, indirect taxes on luxuries are relatively
heavy and consumer goods are often imported at exchange rates which
make them expensive; s o such biases are particularly severe when market
p r i c e s are u s e d as w e i g h t s .

Still, i t might b e argued that national income s e r i e s are at

l e a s t available, whereas t h o s e on poverty, unemployment and inequality
are very scrappy. T h i s iowever the result not s o much of basic diffi
culties in estimation as of attitudes to development. The type of data
c o l l e c t e d reflects priorities; i f governments become more interested in
social problems than in the national income, statistical offices will
prepare the appropriate statistics.

I do not deny that there are practical problems here t o o . These

were f a c e d by Rowntree (whom I have already mentioned) at the turn of
the century; he attempted to establish what he c a l l e d a ’human needs
standard* for the population of York, worked out the c o s t s for different
family s i z e , and then estimated what proportion o f the town’s popula
tion could not afford i t . I t i s difficult t o exclude conventional
expenditure entirely from a poverty l i n e . Even a minimum diet must
reflect traditional eating h a b i t s . Moreover, there are always other
conventional claims on expenditure. Rowntree made another calculation
showing how many people in f a c t failed to reach the nutritional
standard s e t by the poverty l i n e , because,, f o r conventional or other
r e a s o n s , they ”mis-spent” their income. The official minimum standard
in the United S t a t e s , e s t a b l i s h e d by the Council of Economic Advisers
allows about $250 a year per head for food alone, and another $500

1, The S t a t i s t i c a l Office of the United Nations h a s prepared ”A

complementary system o f s t a t i s t i c s o f the distribution of income,
expenditure and wealth” which is a useful starting point.
 13

for other i t e m s , which certainly gives a margin above strict physical

needs. But we should n o t l e t ourselves be diverted by such problems
from the attempt to establish official poverty l i n e s , as was done in
India by a working group in 1962 (which established 240 rupees p e r head
a year as a basic minimum). The resultant estimates o f the proportion
living in poverty, however rough, have some significance and this is a
usable yardstick for measuring development. Indeed comparisons o f the
proportions of families, or even income r e c e i v e r s , with l e s s than any
level of real income are n o t without meaning, certainly they convey
more than the per capita national income,

A general measure which has i t s u s e s i s the infant mortality

rate, and s e r i e s showing the incidence o f d i s e a s e s of undernourishment,
such as r i c k e t s , are further c l u e s on development, as are data on the
height and weight o f schoolchildren. All these can be combined to give
u s an indicative profile of the prevalence of poverty in a nation, such
as the U.N. Research Institute for S o c i a l Development in Geneva has
been experimenting with. In f a c t they have produced a tentative
2
’’development indicator”, a weighted average, of various s e r i e s . The
Institute’s investigations of multiple a s s o c i a t i o n s are very interesting
and worthwhile, but we should n o t fall into the trap - as we c o u l d ,
although the I n s t i t u t e ’ s D i r e c t o r warns us against i t - of treating this
indicator as " n o r m a t i v e ” . I t simply measures the extent t o which a
country has advanced along a path indicated by data from countries at
different s t a g e s of p r o g r e s s .

Unemployment i s of c o u r s e notoriously difficult to measure in

non-industrial societies. An urban unemployed p e r s o n can be roughly
identified by the usual, t e s t questions designed t o reveal the l a s t
o c c a s i o n when work was sought (though this means excluding from the
unemployed t h o s e who would only look for a job if they thought there
was any chance of finding it, and on tho other hand including those

1. D r . liinhas e s t i m a t e s that nearly half the Indian population were

recently living at l e v e l s below this line ( o p . c i t . ) . Ronald F .
Henderson describe poverty line r e s e a r c h in Melbourne in an article
in "Poverty in the Affluent S o c i e t y ” ( 7 . E . A . 1 9 6 8 ) .
2. S e e Research Note No. 2 , July 1969*
who would in fact only a c c e p t particular types of work). The measurement
of disguised unemployment which exists in the town as well as the
country, i s well-known to be a very difficult b u s i n e s s , especially
because of its sharp seasonal variation; but attempts are being made and
further research should enable them to be improved. What i s needed i s
much more detail by s e c t o r , by region, by s e x , by age, by educational
qualification, to throw light on the nature o f unemployment in any
country. 1

Inequality can be measured in many dimensions - by factor shares,

by s i z e , or by r e g i o n . Regional c o n t r a s t s are significant, and can
usefully be supplemented by data on social s e r v i c e s , educational levels,
e t c . , which are already sometimes available on a regional basis. But
s i z e i s the most meaningful classification for income, especially since
the r e s u l t s can b e linked to measures of poverty. Of comprehensive
coefficients, the one devised by G-ini, derived from the Lorenz curve,
i s probably still the most u s e f u l , for either income (post-tax) or
wealth. But, i f we are really concerned with poverty, a more meaningful
measure may b e to e x p r e s s ( s a y ) the lowest decile as a fraction o f the
median (following the general approach in a r e c e n t study by Lydall )•
We arc after all not greatly interested in changes within the top half
of the income ladder.

The conceptual problems o f t h e s e measures do not seem to b e more

formidable than those of the national income. We have j u s t grown
accustomed to ignoring the l a t t e r . But there are also practical problems.

1. S e e B . Hansen "Employment and Wages in Rural Egypt" in The American

Economic Review, V o l . LIX, No. 3 , June 19&9»
2. See H . Lydall "The Structure of Earnings" (Clarendon P r e s s ) .
3 . Of course all t h e s e measures r a i s e the same problem as national
income measurement, namely what allowances should be made for activities
which are n o t marketed.
 15
All t h e s e measures require information about supplementary incomes,
age-and-sex composition of receiving units, e t c . , additional to what
i s obtained from statistics which are prepared as a by-product o f
administration, at l e a s t in countries where only a small proportion of
income r e c e i v e r s pay direct taxes, I t is also hard to measure even
overt unemployment where unemployment registration does not exist or
covers only part of the labour f o r c e . But again we must n o t be diverted
by such technical problems from attempting the r e a s s e s s m e n t which
really matters.

There i s one p o s s i b l e source for all o f these measures, as indeed

for the national income i t s e l f - regular sample surveys of households.
The development o f the sample information required to study trends in
poverty, unemployment and income distribution requires a permanent
sampling organisation, such as India has in i t s National Sample Survey.

I cannot explore here the measurement of the educational and

political elements in development. In as far as the former i s covered
by the formal educational system, a technique for showing the changing
2
profile over time has been developed by Richard Jolly. Measurement
o f the extent to which the political aims have been achieved i s of
c o u r s e much more difficult; possible clues include the number of
prisoners h e l d for political or quasi-political r e a s o n s ; the s o c i a l and
racial composition of parliaments, business b o a r d s , senior public
administrative grades, e t c . and also of those enjoying secondary and
university education: the incidence of petty theft; rates of suicide
and alcoholism.

C l u e s on the degree of national independence include the extent

to which the country v o t e s in the same way an a great power at the
United N a t i o n s , the e x i s t e n c e o f foreign military b a s e s and overflying
rights, the ratio of aid from the l a r g e s t donor to total foreign

1. Technical problems o f measuring distribution are d i s c u s s e d in an

unpublished paper "On the possibility of measuring personal distribution
of income" by Professor Dich of Aarhaus.
2. S e e A. R . Jolly "Planning Education for African Development",
( E a s t African Publishing House, Nairobi, 1969) .
 16
exchange receipts, etc. Indirect indicators are the proportion of
a s s e t s , especially subsoil a s s e t s , owned by foreigners, the extent
to viiich one trading partner dominates the pattern of trade, and the
pr-. portion of the supply of capital ( o r intermediate) goods which is
imported.

The internal consistency of the development p r o c e s s

The second s e t of challenges to the s o c i a l scientists , poli
ticians and administrators in t h e decades ahead is to find paths o f
development which enable p r o g r e s s to b e made on all these criteria.
S i n c e development i s far from being achieved at p r e s e n t , the need i s
n o t a s i s generally imagined, to accelerate economic growth - which
could even be dangerous - but to change the nature of development
precesses.

A major question i s whether the criteria are mutually consistent.

Th? answer i s that, in many r e s p e c t s development on one o f the criteria
implies, or h e l p s bring about, o r i s even a necessary condition f o r ,
development on one or more of the o t h e r s .

To reduce unemployment i s to remove one of the main c a u s e s of

poverty and inequality. Moreover, a reduction in inequality will of
c o u r s e reduce poverty, c e t e r i s paribus.

These propositions bog many questions, however. The reduction

o f unemployment means in part finding techniques which are labour-
intensive, with the l e a s t damage to the expansion of production.
This i s of c o u r s e a d i s c u s s i o n to which many have contributed, notably
A. K. S e n .

There i s a well-known, indeed c l a s s i c a l , argument, that inequality

i s necessary to generate savings and incentives and thus to promote
economic growth - which, as we have s e e n , can b e taken as an indicator
of some typos of development potential. I find the argument that the
need for savings j u s t i f i e s inequality unconvincing in the Third World
todr-y,: Savings propensities are after all very low precisely in
wive highly unequal distributions; the industrial countries
Ccuniries/with l e s s concentration of income h a v e , by c o n t r a s t , much
higher savings p r o p e n s i t i e s . Savings are o f c o u r s e also affected by
 17

the absolute level of incomes, but the explanation must also lie in
the high consumption l e v e l s o f the rich, designed to maintain the
standards s o important in an unequal society. Moreover, the rich in
most countries tend to have extremely high p r o p e n s i t i e s , not merely to
spend, but to spend on goods and s e r v i c e s with a high foreign exchange
content, and, for countries suffering from an acute foreign exchange
2
bottleneck, this i s a major o b s t a c l e to development. I t is true that
import demand can be held in check ( a s in India) by administrative
controls, but this l e a d s to the elaboration of a bureaucratic apparatus
which i s e x p e n s i v e , especially in terms of valuable organising ability,
and which in some countries becomes riddled with corruption. In any
c a s e , in a highly unequal s o c i e t y , personal savings often flow abroad
or go into luxury housing and o t h e r investment p r o j e c t s of low or z e r o
priority for development, even when this i s defined as j u s t a rising
national income.

The argument that only inequality can provide the incentives that
are necessary is also obviously of limited validity in a country where
there are barriers o f r a c e o r c l a s s to advancement. Still, we cannot
dismiss i t out of hand. The needs fox’ private entrepreneurial talent
vary according to the circumstances of different economies, b u t there
are very few where this need i s small. Countries relying on growing
exports of manufacturers, as many a r e , depend heavily on the emergence
of businessmen with the drive to penetrate foreign markets. All
countries depend in some degree on the appearance of progressive far
mers. Will these emerge without financial rewards on a scale that will
make nonsense of an egalitarian policy? Are r i s i n g p r o f i t s of companies
e s p e c i a l l y foreign companies, an inevitable feature of growth in many

1. In a paper given to this y e a r ’ s Kandy conference, A. K. Bagchi,

reviewing the e x p e r i e n c e s o f the 1 9 5 0 ’ s and 196O’s, drew attention to
the failure o f private savings t o . grow, despite r e g r e s s i v e tax
policies.
2. An implicit assumption here is that Engel lines are non-linear,
but this seems n o t to need specifying. Consumption of imported
luxuries is z e r o over a considerable income range.
 18

countries? Or are we exaggerating the importance of financial incentives?

Can other non-financial rewards partially take their place? Can social
incentives be developed to a point where people will t a k e on such
tasks with little or no individual reward ( a s the governments o f China
and Cuba arc trying t o procure)? This one of the great issues to be
decided, and the 1970* s will throw a good deal of light on the answer.

The compatibility of equality and rising output and employment

h a s recently become doubtful for an additional s e t o f reasons. Can
the people who are professionally n e c e s s a r y be kept in the country if
they earn only a small fraction o f which they c o u l d earn elsewhere?
Yet what are the c o s t s in terms o f human welfare and even efficiency
if they are prevented from leaving?

On the other hand, there are equally serious r e a s o n s for question

ing the compatibility of inequality and economic growth. Can a structure
of local industry b e c r e a t e d to correspond to the structure o f demand
that arises in a highly inequitable s o c i e t y (leaving aside the q u e s t i o n
of whether i t should b e c r e a t e d ) ? Will production r i s e rapidly if the
proportion of the labour force too badly nourished for full manual and
mental work i& only sinking slowly? Can the government obtain the
co-operation of the population in wage restraint, and in many other
ways that are n e c e s s a r y for development, i f there is visible evidence
of grea,t wealth which is being transmitted from generation to generation,
so that the wage earner s e e s his children and his c h i l d r e n ’ s children
doomed indefinitely to subordinate p o s i t i o n s ? Can it mobilise the
energies of the total population and break down s o c i a l customs which
obstruct development especially in rural areas?

I do not pretend to know the answers to this complex of questions,

which point t o a s e t o f ’’internal contradictions" in the development
p r o c e s s e s far more s e v e r e than those to which Marx drew attention.
The economic and political objectives are linked c l o s e l y together.
An economic system with large numbers o f undernourished and unemployed
at the bottom end of a long social ladder, especially i f they are
racially distinguishable, can never provide a firm basis for political
rights, o r for civic order. Those with high incomes from profits or
salaries are not merely slow to tackle the great social problems of
 19 -
poverty and. unemployment; they will inevitably try to find ways of
maintaining privilege, resorting ( a s dozens o f historical examples show)
to political violence rather than give i t u p . Conversely, those without
jobs or adequate incomes will sooner or later try to obtain them
through a regime which would not allow organised opposition. Judging
from p r e s e n t trends in the climate o f opinion, especially among the
young, i t i s very doubtful whether inequalities on anything like the
present scale could co-exist with political, liberties in the 1 9 7 0 ‘ s
or 1980’ s .

Y e t i t i s hard to envisage how inequality can be reduced without

setting in motion, from one direction or another, f o r c e s that reduce
political liberty. There are administrative limits to the main weapon,
direct taxation, apart altogether from any conflict with incentives,
which I shall consider l a t e r . Inequality cannot really be reduced s o
long as property ownership is heavily concentrated. So conversion
of incomes from large holdings of property into life pensions ( a s in
Cuba) or bond i n t e r e s t ( a s in C h i l e ) or their reduction through death
duties ( a s in Britain) are likely t o be more effective than taxation,
though they may be beyond the bounds of the politically feasible in
in many countries.

But a great deal can be done even without attacking property

ownership. Practically every decision taken by government officials has
implications for the degree of equality - to lend to big farmers or
small, to s e t prices o f public corporations at levels that tax or
subsidise rich consumers, t o build roads for private motor cars or for
goods v e h i c l e s , to put the b e s t equipment in rural or urban schools.
I t would n o t be a bad thing t o put up in every civil service office
a sign: "’Till i t reduce inequality?” Secondly, if the administrative
and political organisation i s motivated and trained to report tax
e v a s i o n , corruption, e t c . all s o r t s of egalitarian policies, including
capital taxation, become more feasible. (Where such a spirit i s weak,
one can hardly expect rapid development in the sense I have u s e d the
word h e r e . )

Lastly, a reduction in inequality i s very hard, if not impossible,

s o long as a country is dependent on a major power. So i s a reduction
 20

in unemployment, because one of the marks of dependence is reliance

on the technology of the countries which play a dominant role in the
national life, and this may well b e inappropriate for local factor
costs. On tlie other hand, a country that leaves its social problems
unsolved i s unlikely to be strong enough to achieve or maintain
genuine independence - it may not even survive as a political unit.

Political. liberty must be considered as an e n d , the importance

o f which though high i s secondary, so long as a substantial fraction
of the people are undernourished and unemployed. But i t appears in
quite a different light as a means* S o c i e t i e s lacking open opposition
have shown themselves extremely inflexible in meeting the challenge of
changing circumstances, whether one considers the continuation by
Germany of a war for months after i t was already l o s t , o r the stubborn
n e s s of Communist regimes i n clinging to unsuccessful agricultural
policies. Moreover, as S o v i e t experience shows, there is no guarantee
that political liberties will reappear as economic problems are e a s e d .
In some r e s p e c t s , t o o , the o b j e c t i v e s are, from certain points of
view, means to achieving the other o b j e c t i v e s .

I have already treated higher educational l e v e l s as ends in

themselves. B u t education i s a l s o a means. Economists have, somewhat
belatedly, come t o s e e it as a s o u r c e of development; but they treat
it narrowly as a factor in the growth of national income, relating stocks
of manpower with certain qualifications ( e . g . university degrees) to
national income l e v e l s . Even in this r e s p e c t a l o t of work remains
to be done - wh-t skills p r e c i s e l y are needed f o r various types o f
work?.

But if development i s not just or even mainly an increase in the

national income, education takes on an entirely different aspect. We
need to go a long- way beyond the Harbison approach, valuable though
this has b e e n . What are important are n o t the "man years” at school
but the methods of s e l e c t i o n and the c o n t e n t of education. Inequality
can be r e d u c e d (and also economic efficiency increased) if secondary
and higher education are made genuinely available to those with the
l o w e s t incomes (and belonging t o minority r a c e s ) , which means of
c o u r s e that special methods o f s e l e c t i o n must, be found. Secondly, by
 21

easing shortages of high-level manpower, education can reduce the need

for high salaries for t h o s e with s c a r c e professional skills. In that
c a s e , however, the whole structure of education needs reconsideration;
education will hardly fulfil this function, if it produces a few
distinguished academic scholars, who, as in Britain, think of themselves
as superior to the r e s t of the population, and a mass o f people with
the wrong qualifications. The third function of education, and this
applies also t o adult education and the content of radio and TV
programmes, i s to prepare professional c l a s s e s conscious of the
realities o f development, both the internal realities and the realities
of the world s c e n e , with an understanding o f their historical origins,
so that they s e e what needs to b e done and voluntarily accept the
sacrifices implied. Many of the o b s t a c l e s to p o l i c i e s which would
reduce poverty and unemployment have their origin in attitudes to
manual work, especially in the countryside, to imported consumer goods,
to foreign technologies, e t c . Since each generation i s in some degree
a copy of the preceding o n e , through parental influences, such attitudes
can hardly be changed e x c e p t by c o n s c i o u s educational policy broadly
defined.

Finally, one policy area which looks different i f one discards

the aggregative approach i s population. Conventionally, population
growth i s s e e n simply as a subtraction from the increase in the national
income; the c l o s e r the growth r a t e s o f population and income, the slower
the rise in £cr_j?a£ita income. This line of argument is fundamentally
somewhat s u s p e c t because i t assumes that population and income are
independent o f each other. But the real c a s e for an active population
policy i s simply that, s o long as the labour f o r c e is growing f a s t , it
i s almost impossible to relieve unemployment and poverty - the plentiful
supply o f labour keeps the wages of the unskilled, apart perhaps
from a privileged modern s e c t o r , near levels of b a r e s t s u b s i s t e n c e .
Moreover, the growing p r e s s u r e of population on the budget makes it
very difficult to expand educational and other s o c i a l s e r v i c e s . An
additional argument, on tho above criteria, i s that this growing
p r e s s u r e i n c r e a s e s the need for foreign aid, and thus postpones the
attainment of genuine independence.
 22

Implications for planning

In many countries, i t will be decades before development strategies
can finally bo successful in eliminating poverty and unemployment, Both
educational and population policy require a great deal o f time to work.

I t was the manpower problem that f i r s t turned our attention to

• p e r s p e c t i v e planning’. I t takes nearly twenty years to turn a child
of five into a doctor; i t will take much longer b e f o r e the main shortages
of professional manpower in Africa or A s i a , however measured, are made
good. But f i r s t one has to c r e a t e the educational facilities. This i s
partly a matter of buildings, partly a matter of producing teachers o f
the right types, but a l s o a matter o f finance ( i . e . of achieving a
structure o f income flows such that producers will f e e d large numbers
of teachers and students who are not immediately p r o d u c t i v e ) . Even
when t h e s e problems are s o l v e d , there is the very great one of getting
s c h o o l s and universities to change their own structures - granted that
they, especially universities, have considerable autonomy, and that the
systems o f recruitment and promotion tend to perpetuate existing balances
between s u b j e c t s , and even existing syllabuses within s u b j e c t s .

Time in fact i s not j u s t a lag before input becomes output,

but the dimension in which psychological inertia i s overcome. It
takes, say, eight y e a r s , b e f o r e the conceptions cf this year become
school entrants, and another eight before they become entrants into
the labour f o r c e , i . e . before population policy can e f f e c t the needs for
educational expenditure and the levels .of unemployment respectively.
Moreover, the population profile may b e such that there i s a .rapidly
growing number in the reproductive age-groups, s o that even changes in
fertility are slow i n affecting the volume o f births. But in addition
i t takes time before views on family s i z e and attitudes to contraceptive
devices change, especially among the p e a s a n t s and manual workers -
this p e r i o d depends, in part at I v a s t , on the provision o f education for
women, which has its own time l a g s , as has been indicated above.

The consequences of inequality can p e r s i s t a very long time

i n d e e d , even after the income distribution has changed. T h i s applies
to preferences for clerical j o b s , for living in c i t i e s , for existing
social and racial privileges, as well as to uncritical admiration of
 23

foreign techniques and styles o f living. A key interval here i s -the

time between adolescence and death, say half-a-century. One could
almost say that it takes some fifty years before a population with
completely now attitudes can be formed. I t i s true that p e o p l e ’ s
a.ttitudes do change even after they have p a s s e d through adolescence
- if attitudes seem constant, i t i s because the pace of technical change
i s s o much faster. Moreover, while a rapidly growing population p o s t
pones the solution of problems it does have one compensation - in
a tapering population profile, changes in attitude can take place more
rapidly.

Recognition of negative p r o g r e s s in the p a s t decade and of the

time-lags, i . e . of the medium-term insolubility o f many problems,
changes o n e ’ s attitude to development strategy. Formerly the basic
technique c o n s i s t e d in extrapolating p a s t trends and choosing invest
ment patterns that would produce an acceptable increase in national
income in a five-year p e r i o d , tacitly assuming many constraints as
a iven - thus consumption patterns were p r o j e c t e d in a way that a s s u r e d
little or no change in income distribution o r t a s t e s or ©.ttitudes. Now
we must try to envisage what might b e -a plausible pattern one day, in
terms not only of production and employment structures, but of the
patterns o f consumer demand and job p r e f e r e n c e s , and then work backwards,
to s e e if there i s any plausible path for getting t h e r e .

This means a fundamental rethinking of planning. The medium

term ( e . g . five-year) plans now published need to be reconstructed to
form s t e p s towards long-term targets for the reduction of poverty,
unemployment and inequality i n s t e a d o f by moans of achieving national
income t a r g e t s .

One final point. Development strategy-making in this sense

does not involve ignoring the pattern of development in richer
countries. The major tasks on which the richer countries are engaged,
of constructing educational and political systems that satisfy indi
vidual needs and c r e a t e a safe framework for life, and of halting the
r a p i d deterioration in the environment seem somewhat remote from the
realities of the Third World - something rather for the 21st century. But
they are nonetheless relevant today. One of the challenges of develop
ment i s not to copy their experiences but to draw the right l e s s o n s from
this and avoid repeating their errors.
 24

International policy
Finally, we must look at the challenges to international
policy. The world, is a unit to which the same criteria can b e applied,
and a conference like this is precisely the occasion for doing s o .

Looking at the world as a whole, we can bring together the earlier

remarks on development inside industrial and non-industrial countries.
There has been p r o g r e s s on the poverty criterion; the proportion of
the whole human population living below any subsistence line must
have fallen.

But s i n c e the middle of the l a s t century enormous gaps have

opened between r i c h , countries and poor: inequality on the present
scale is an entirely new phenomenon, as papers by Simon Kuznets and
Surendra Patel have brought out. Inequality between nations, like
inequality within them, means differences in status and power, affecting
the attitudes of men towards each other. T h i s , again as on the- national
level, means growing tensions between r a c e s , broadly again between
the whites and the remainder. H e r e , moreover, l i e s the r o o t of the
snobbism o f foreign techniques and consumption standards.

Lastly, total overt world unemployment must have grown over

the past century, since the emergence o f unemployment in the Third
World must numerically outweigh the, decline of unemployment in the
industrial countries.

When we consider the world s c e n e , i t i s wrong to talk about

’development*, on the criteria suggested above. One cannot really
say that there has been development for the world as a whole, when the
benefits o f technical progress have accrued, to minorities which were
already relatively rich. To me, this word is particularly misleading
for the period since the war, especially the ’development decade’
when the growth of economic inequality and unemployment must have
actually accelerated. ( l am alarmed at the phrase a ’’second develop
ment decade”: another ’’development decade" like the 1 % 0 ’ s with
unemployment r a t e s anc inequality rising by further large s t e p s , would
be politically and economically disastrous, whatever the p a c e of economic
growthl)
 25

I t i s true that in some r e s p e c t s , as I have said, a basis has

been laid in many countries for p o s s i b l e development in the future.
But there han not been any basic improvement in international insti
tutions. I t is true that there are now opportunities for poor countries
at l e a s t to talk to the rich, but one could not speak of international
order; the international institutions lack the power to impose solu
tions. There is not much to show for the tremendous efforts which
went into the New Delhi UNCTAD. Virtually no safeguards have been set
up against a world r e c e s s i o n - certainly the creation o f SDR’s is
by no means adequate in i t s e l f . Nor has much been done t o open the
markets of industrial countries to imports of manufactures, the only
real possibility of export expansion for the Third VTorld as a whole.

There i s no fiscal system for the world. This may perhaps be

foreshadowed by 1% aid targets, but these targets are in fact ignored
and aid programmes remain at very low l e v e l s . As Gunnar Myrdal
warned us many years a g o , the establishment o f the national welfare
state has turned the attention o f the public in the rich countries
inwards, making them l e s s interested in the welfare of the world as
a whole. The aid that does exist often plays an important economic
r o l e , but, like immigration and trade p o l i c i e s , i t i s very largely
motivated by the self-interest of donors, sometimes by very short-term
commercial and political interests. This often in effect leads them
to support, or even help install, governments which oppose the redistri
bution of income and in other ways block development.

Many countries have in fact slipped further under the influe?«;e

of one or another of the b i g powers. This i t s e l f hinders development.
Independence is not merely one of the aims of development; it i s a l s o
one of the means. It is a force for mobilising popular support and
the force is blunted if a government is obviously far from independent.
 26

These are the challenges we now face. The role of the practi
tioner, the politician or the civil servant, is the extremely difficult
one of finding politically and administratively feasible ways of
developing in a grossly unequal world. They can b e helped by the
theorist if.h e refrains from trying to adapt uncritically models
and measures designed in and for industrial countries, where priorities
are very different, but help develop policies, national and international,
.to mitigate the great social problems of the Third World.

By so doing, indeed, ho may incidentally provide the social

scientists in the rich countries with food for thought. After all,
though poverty and unemployment are not so starkly obvious in the North
Atlantic area, economic and racial inequality are evident for all to
see. But above all, the aim must be to change international attitudes
so that it becomes impossible for the political leaders and social
scientists of Europe and North America to continue overlooking and
aggravating, often inadvertently, the obscene inequalities that
disfigure the world.