'THE CRITICAL MINIMUM EFFORT' THESIS:

A CRITICAL APPRAISAL 1
P. R. BRA.HMA.N A.NDA.

I
IDtroduction : Growth-Retarding
and Growth-Promotiog Forces
Professor Harvey Leibenstein's name is familiar to economic-demo-
graphers. In an earlier book, titled "Theory of Economic-Demographic
Development", Professor Leibenstein had tried to analyse the meaning and
significance of the population obstacle through the methodology of modem
aggregative theory. In this work, he developed the thesis that a 'minimum
critical effort' was necessary in order to overcome the population obstacle.
EHorts of a smaller magnitude would tend to be frustrated. The above
idea has been generalised and carried to its logical conclusions in "Econo-
mic Backwardness and Economic Growth...• Professor Leibenstein has here
sought to present a new and general approach towards the problem of
economic development of underdeveloped countries. This work is
characterised by an integration of the applications of the minimum critical
level thesis to a number of inter-related magnitudes of vital significance in
determining the tempo of growth. The approach is original in Conception
as well as in the methodology of treatment. The conclusions are as striking
as the route traversed by Professor Leibenstein in arriving at them II
fascinating. A number of magnitudes which economists would have dis-
missed as outside the pale of the analytical treatment have been persuasively
handled in Professor Leibenstein's methodology. From this point of view,
Professor Leibenstein has enlarged the conventional scope of treatment of
development problems. The present work is, by and large, non-mathematical
in character and depends largely upon the diagramatic method for
exposition. It will, naturally, be a boon to the normal economist pedagogue.
Professor 4Jibenstein's work is certainly a tour-de-force.
2. Let us now glance at the central idea underlying Professor Leibenstein'.
work. Professor Leibenstein distingnishes between factors which retard

1 I am grateful to Professor Howard S. Ellis for many helpful suggestions in

the preparation of this Paper; I am also thankful to Dr. R. C. Porter and the
members of the Research Students' Seminar of the Department of Economics,
University of Bombay, for many critical comments.
• 'Economic B""kwtvrdne•• Gn4 Economic G-rowth', John Wiley and Sona, (Di...
tributed by Asia Publishing House in India), 1957.

J ...... 1959 VI N. 107

 P. R. BlIAHMANANDA

growth and those which promote it. The strategic variable in the whole
approach is the level of per capita income. At varying levels of per capita
income, varying rates of income retardation and income promotion are
obtained. Whether the system is stable or not a particular level of per
capita income is determined by the relative strength of the growth retard-
ing and promoting forces. Starting from a low level of income, the growth-
retarding forces increase their strength upto a particular range; there-
after, the intensity of these forces is Constant or tends to get reduced. The
growth-promoting forces are feeble at a low level of income; but as the
level of income rises, their intensity goes on rising. 'The economic situa-
tion in an underdeveloped system is characterised by a certain degree of
quasi-stability. 'The level of per capita income will have a tendency to
stabilise itself at a low level. The state of quasi-stability is characterised
by the fact that upto a particular level of per capita income, the marginal
growth-retarding forces are greater in intensity than the marginal growth-
promoting forces. Small displacements of per capita income have a ten-
dency to be reversed; thus, the system is tending to obtain the subsistence
level of per capita income.
3. What are these growth-retarding forces and growth-promoting forces?
Professor Leibenstein brings to bear a number of inter-related factors upon
the discussion of this issue. The chief growth-retarding force is popula-
tion. If we start from a low level of income, increases in this level tend to
accelerate the growth of population; but this accelerated 'effect is absent
after a particular level of income is obtained. The population factor is an
obstacle in this range because a large amount of population diffuses capital
per head. At a low level of income, population is relatively large and any
given increase in income leads to a population upsurge, which calls forth for
increased expenditures of maintenance. "Population growth dilutes the
amount of capital with which the representati~e worker may operate....
Any event that increases income will, at first, also increase the rate of
population growth. This, in tum, implies an increase in the labour force,
and both capital and land per worker are accordingly reduced. Further-
more, this tends to reduce income per capita, which depresses the induced
rate of population growth, if not, the actual rate", (p. 56). Other
growth-retarding forces are, the deficiency of marKet incentives, unwiIling-
Dess to accept entrepreneurial risks, a high rate of preference to white-
collar jobs, encouragement of honouriflcally valued cultural education in
place of technical education, etc.
4. Professor Leibenstein argues that as the income level rises, the growth-
promoting forces gain in intenSity. What are these growth-promoting
forces? Levels of productive efficiency would go up as per capita income
levels rise., The degree of specialisation would go up. At low levels of
income, low growth rates are anticipated; hence growth agents are not

108 Art"" Vi;_7IG

 'CRITICAL MINlMUM EFFORT' THESIS

powerful; growth agents ( entrepreneurs, skilled workers, knowledge-

contributors, etc.) are most active when the anticipated rate of per capita
income changes is high and rising. At ·Iow levels of income, speculative
. and trading activities tend to dominate. It is only when income is growing
and at a high rate that positive-sum games are favoured. Professor Leiben-
stein argues that the number of entrepreneurs, as well as the-magnitude
of entrepreneurial contribution to income will rise with rising levels of
income. Similarly, in regard to activities concerning skill formation. At a
high level of per capita income, the contribution of skilled workers to
national income rises. Furthermore, as per capita income level reaches a
particular level, the rate of population growth would tend to decliDe. This
is ~use the desire for an increase in family size is reduced; other alter-
natives become more important. Conventional costs of a given increase in
the size of family grow. The utility of children as a source of security in
the parents' old age and as productive agents goes on declining. Thus, the
net impact of both these forces would be to bring in vogue smaller sized
families. Professor Leibenstein also argnes that at higher levels of income,
technological indivisibilities are overcome. Projects of delayed profitability
are taken up. External inter-dependencies are overcome, The conditions of
balanced growth are better satisfied at higher levels of per capita income
than at lower levels of income. The capital.-output ratio too will tend to
fall as income levels rise. .
n.
ImplicatioDS of Critical EfI'ort Thesis
•
5. The core of Professor Leibenstein's thesis can be demonstrated in the
following diagrams.

Growth Promotion " Retardation

y y y
GR G.
UP
~_F''7f'''"-GR
_~E;;'"&-_G.
_~Ei«'.:;.-_GR

OL-----~E------~X

-,. F
I.cId .of Per Capita lacome

On the X axis, we indicate levels of per capita income. On the Y axis,

X

we indicate the contribution (in terms of income rise or decline) of growth-

contributing and growth-retarding factors. Three types of possibilities can.
broadly be visualised. Both the growth-retarding and growth-promoting
factors can be considered to be linear. GR is the curve indicating the

J"... 1959 VI N J 109

 P. R. BRAHMANANDA

magnitude of growth-retardation at varying levels of income. GP is the

magnitude of growth-promotion at varying levels of income. E' is the level
of income where GP intersects GR from below (Diagram 1). Levels of
income to the right of E tend to be more than seI£-maintaining. Levels of
income le~ of E tend to be unstable downwards. This picture indicates
that small displacements of equilibrium on either side would plunge the
system into cumulative growth or decline, the latter without limit. The
second picture is one in which GR intersects GP from below. The level of
income OE is stable. Any displacement towards the right or towards the
left will set into motion forces which restore stability ( Diagram 2).
Professor Leibenstein rejects both these types of states as typically charac-
teristic of underdeveloped countries. He adopts a non-linear relationship
between income and income-induced growth-retardation and growth-
promotion. At a low subsistence level of income, the slope of GR is greater
than GP. At a higher level of income, (Professor Leibenstein would term
this as the 'critical minimum level'), the slope of GP is greater than that
of GR. Points E and F, (Diagram 3), indicate both these positions. Any
displacement which lands the system at a level between E and F would lead
to forces which bring the system back to equilibrium E. A displacement
beyond the level of income F would lead to cumulative growth. Professor
Leibenstein's thesis is that small displacements from E do not land the
system into the area of cumulative growth. The displaC6!flents in terms
of stimulants must be greater than the area indicated by EF. Any shock
which lands the system from a liwel beyond F to a level between E and
F would make the economy relapse to E. •
6. Why does Professor Leibenstein assume non-linearity? As income
level goes up, the rate of population growth is accelerated. After, perhaps.
reaching a maximum level. it decelerates. Population becomes a hurdle
only if levels of income rise slowly. There is no such obstacle if there is
a large displacement. High rates of income growth can overcome the Popu-
lation obstacle. Similar is the argument in regard to other growth-promot-
ing and retardation activities. The growth-promotion activities, in general,
assert the values with greater intensity as income levels rise. The retardation
forces, however, have a maximum value. The growth-promotion forces,
perhaps, have such a maximum value but this maximum lies above that
of the growth-retardation forces.
Thus, there is a vicious circle, but this can be broken. It cannot be broken
if displacements ar~ small. The equilibrium of underdevelopment is quasi-
stable for small displacements but unstable for large displacements. The
mechanism of a slow (steady) growth, therefore, cannot take the system
out of the rut. What is required is a big push, a quick dash. a sizeable dis-
placement from a low level to high level per capita income through a fast
rate of growth of per capita come.

110 Mth4 Vii_

 7. At this stage, it is tantalizing to ask questions about the quantitative
significance of the above thesis. What is the level of income at which the
system is unstable upwards? What is the range of income within which
it is quasi-stable? What is the magnitude of the minimum effort necessary
to overcome the obstacle? How much of an increase in investment and/or
income is necessary to make entrepreneurs undertake positive-sum games?
At what level of income and rate of growth does population growth dece-
lerate? What form should the displacement take? ProfeSsor Leibenstein
does not go into the question of the institutional agencies necessary to
achieve- the displacement. He carefully avoids any quantification of his
results and appears to hold the view that quantification is not possible,
nor is it necessary.
• 8. Superficially at least, Prof. Leibenstein's thesis tends to have consider-
able appeal to students of development. It might-though not necessarily-
justify the efforts at planning for high rates of income growth. Under-
developed countries are underdeveloped because they seek to grow at a
slow rate. Previous writers used to emphasize a- number of socio-economic
deficiencies in the datum of an underdeveloped country in order to prove
why a low rate of growth was being obtained or why a state of quasi-stag-
nation existed. A large relative population size with considerable disguised
unemployffient, the economic implications of long-standing colonial status,
the prevalence of socio·economic and cultural hindrances to mobility, the
deficiency of a keen economic motive, secularly persistent adverse terms
of trade, etc. were 'among the list of basic factors that conspired to keep
a country underdeveloped. Professor Leibenstein's thesis breaks through
the vicious circle by highlighting one central fact that most of the obstacles
operate. only at low levels of income; they don't cause a low
leveL A high rate of income change depresses the significance of the above
factors and releases. powerful uplifting factors. Thus, economic stagnation
does not lead to low rate of change; a low rate of change causes stagna-
tion. Conventional opinion in this respect does not deny that growth-
promoting and retarding factors exist ·in an underdeveloped economy, or
for that matter in any economy. But it would not assert that these factors
are a unique function of the rate of change of income. Conventional theory
again would not deny that upto a particular stage rising growth rates
would lead to further rates of rise in growth rates. But it would not accept
the thesis that it is only a large displacement that can break the vicious
circle. Conventional theory would certainly maintain that a low rate of
change would delay the period of attainment of a given target in income
level and rate of income change. The conservative among the conventiona-
lists might even maintain that there are 'economies' even at a slow rate
of change as well as diseconomies at a high rate of change. The process
would proceed with less friction; the results would be more stable; the

Ju,,- 1959 V 1 N, 111

 P. R. BRAHMANANDA

necessary responses would take some time to evolve any way, and so on.
Professor Leibenstein thus puts himself at one extreme as regards deve-
lopment theory.

III
Per Capita Income as Strategic Variable
9. Professor Leibenstein has chosen per capita income as the chief deter-
mining influence upon the behaviour of critical variables, like population
growth, productive efficiency, degre.e of specialization, supply and levels
of activity of growth agents like entrepreneurs and skilled technicians.
Each of these magnitudes has a particular pattern of relationship with
income, such that if income could be changed, the behaviour and/or re-
action pattern also could be changed. Professor Leibenstein rejects aggre-
gate income, employment, savings and investment as the media through
which the various forces assert themselves. This is a pity inasmuch· as ·one
is not clear about the working of the mechanism which links income
change and the behaviour of the critical variables.
Consider the following possibilities :
(a) Aggregate income and population may be growing at the same
rate; yet the distribution of income and assets is changing in
favour of the growth agents.
(b) Technical change is taking place and most of the gains may
be accruing to the entrepreneurial groups.
( c) Per capita income may be falling temporarily yet investment
forces may be gaining momentum so that a discontinuous rise
is taking place in the rates of aggregative income growth.
One can visualize a number of cases in which aggregative
forces are more important.
Growth in per capita income levels does not always mean rising levels
of saving and investment; nor do rising levels of saving and investment
only follow upon a rise in per capita income levels. .
10. Professor Lebenstein is also not very careful in isolating the relation-
ship between income and each of the chief variables. The relationships
between income and productive efficiency and skill formation, degree of
specialization, population growth, etc. affect and are affected by one
another. The various .influences like the degree of specialization, the level
of productive efficil1.ncy, the supply of entrepreneurship and the amount
of its contribution, the contribution of knowledge-promotion activities, and
the tendency towards a reduction in the average family size act and
interact upon one another. Their various influences cannot be isolated
and added up. Professor Leibenstein examines the various growth-retard-
ing factors as if they were independent of growth-promoting factors.

112 Art"" Vii1l4114

Here again, growth-promotion and retardation are not independent results_
Both reactions are diHerent facets of the same force or forces. One cannot
also take each influence separately and present both promotional and
retardational aspects as if they, in tum, are not determined by the mutual
conflguration of various other influences and background conditions.
The above are some of the methodological deficiencies which might
have confronted any effort of this sort. They do not necessarily detract
from Professor Uibenstein's central thesis.

IV
~tions 1lehind tile LeibeusteiD Thesis'
11. Let us now dissect ProfessOl' Leibenstein's thesis a little more eare-
fully. How much of this thesis is based upon justifiable relationships? Does
he choose the most important and relevant variables and facets of relation-
ship? How much of his theory is logic derived from conclusions?
Professor Leibenstein's thesis of 'minimum critical effort' is based upon
the following assumptions :
(i) There is a particular level of per capita income (designated a&
'subsistence income') at which the growth-retardation curve intersects the
growth-promotion curve from below. Here the slope of the GR curve is
greater than the slope of the GP Curve. _
(ii) A given increase in income leads to a certain extent oE increase
in growth-retardation and in growth-promotion. The slope of GR. curve
increases at a decliniDg rate. The slope of the GP curve increases at an
increasing rate. There is a particular level of per capita income ( the
critical minimum level) at which the GP curve must intersect the GR curve
from below. Here the slope of the GP curve is greater than slope of the
GR curve.
(iii) The relationship between growth-retardation and growth-promo-'
tion forces is reversible, i.e., a rise ill level of per capita income indicates,
an increase in marginal growth retardation; a decline in per capita income
would indicate a decline in marginal growth retardation. A rise in the
level of per capita income would imply an increase in the value of mar~
gina! growth-promotion. A fall in- the level would imply a decline in the
value of marginal growth-promotion.
(iv) Autonomous income-depressant and promotion factors are gene-
rally neutral in character. Changes in these factors do, not afFect the shapes
or positions of the growth-retardation and promotion curves. These changes
again are not significant factors in explaining economic development. ,
12. A number of questions arise in connection with the above set of. as,.,
sumptions. There are many other sets of assumptions that could ser\lllas.'
Ii suitable basis f<lI' evolving a theory of development. Do the growth"

JII& t'6' V IN ~ 118.

 " 'I'. ':Iv BlIAHMANANDA

promotion and retardation .forces necessarily have the shapes assigned to

them in ~he Leibenstein model? Does 'the incidence of growth-retardation
increase with rising levels of income (though, may be, a~ a diminishing,
rate)? What if the growth-retardation forces are constant or if the inci-
dence may steadily fall as levels of income rise? Secondly, what about the
positions of growth-retardation and growth-promotion curves? May they
not be, and are they not inHuenced significantly by autonomous factors?
The Leibenstein model requires that changes in non-iticome-induced m.:'
fluences do not shift the growth-retardation and growth-promotion curves.
Again, what if with every increase in'the level of income the gains obtain-
ed are not lost, even if the level of income slips down somewhat? Finally,
would not differential rates at which the different factors respond affeet
the net result? Should growth-promotion forces always act with a greater
lag as compared to growth-retardation forces?
Each one of the above questions does merit a detailed examination. For,
if they have some significance, then the critical minimum effort thesis does
not appear to hold.Hence, it is necessary to examine some at least of the
Leibenstein hypotheses in greater detail.

v
The Population, Hump
13. Do the forces that retard growth have the characteristtc that Profes-
sor Leibenstein postulates they have? Let us take each of the retardatory
factors. •
, Professor Leibenstein distinguishes between autonomous and income-
induced inHuences upon population growth. It is difficult to isolate the
two influences. The problem is complicated inasmuch as through time
some of these forces would any way be behaving in a particular manner.
The effects which accrue as a result of the passage of time and the large
responses of previous efforts as well as changes due to variations in levels
of income are all mixed up in the final outcome." Professor Leibenstein
argues that. the aggregate utility ofa child of a given birth order would
be a declining function of the level of per capita income. The aggregate
utility is compounded of utility of a child as a consumption item, asa
source of security in the parents' old age and as a' productive agent. The
direct costs of child maintenance and the indirect opportunity costs attri-
butable to a child (of a given birth order) would be an' increasing func-
tion of per capita income. As the level of per capita income rises, the death
rates would tend to fall, creating an increasing gap between death-rates and
birth rates. The initial impact of a fall in mortality rates will be an increas«)
in tHIl age of survival; the ratio of productive years to non-productive years
of a unit increment in populatioh will at first increase. The survival effect

114' ;. Art"" Vij~

 'CRITICAL ·MINJMUM EFFORT' THESIS

hen~ increases the value of a child as a source of the family inCOme and
as a source of old age security. As a· result, motivations mnducive to fe,rti-
lity reduction will not tend to develop at the early stage of mortality decline,
"Indeed, quite the reverse motivations might be established at first".
(~l~). ". .
The occupational effects tend to add to the direct and indirect cos~
of children. .
Thus, Professor Leibenstein argues that upto a particu.lar stage in the
level of per capita income the rates of population gro~h will tend to n;.~.
Thus, small increases in the level of per capita income' will induce a hump
in population growth, whose consequences will be to retard the growth
in per capita income.
14. Professor Leibenstein's analysis of economic aspects of the demo-
graphic problem is somewhat unusual. At some stage; he refers to the
lag between mortality decline oind feitility decline. The purely demo-
graphic viewpoint here would be that as mortality rates decline
unexpectedly and at a fast rate the decline in fertility rates would tend
to take some time to be adjusted. There is, however, a reason to believe
that fertility rates would decline. Thus the steep decline in mortality. rates
does create a temporal or rather a temporary problem. The population
depressant factor would tend to be reduced in intensity with the passage
of time. Professor Leibenstein's argument, however, is that the hump bears
a functional relationship to levels of per capita income. If per capita level
of incoine 'rises to the critical level, there would be no harm. It is this
functional relationship between levels of per capita income and popula-
tion growth that is perhaps most questionable in the Leibenstein thesis:
This linIc, however, is vital to his analysis. If the population-hump is due
essentially to the delayed adjustment of fertility rates,. then . the problem
of development is a question of the length of the period one has to wait
before the' decline in fertility rates sets in. There is here no question of
a critical minimum displacement of per capita income. It is rather a ques-
tion of a critical minimum period of waiting, (autonomous forces may,
however, reduce the lag). The Leibenstein model requires the desired
family size to increase with rising income .levels upto a stage. The causa-
tion is as follows :' An increase in per capita: iIicome leads to' a decline in
death rates. Given the birth rate, the decline 'in death rate implies an in-
crease in the size of the families. An increase in the levei ..of per capita
income implies a rise in the desired .size of the family. .As levels of income
rise, however, the desired size of the. family declines. Leibenstein's.·thesis
necessarily implies' either that the subsistence .level of inQOllle tn~ .be
characterised by maximum poSsible fertility mes or that initially·: ,un-
provements in .levels of income 'should 'leaq ~o ,:riSin~. fertility rlltes, given

J"... 195' VI N.
 P. II. BIIMIMANANDA

constant mortality rates or declining mortality rates, given fertility rates.

"Upto a particular stage the desired family size adjusts to the changing com-
plex of death and birth rates.
One can raise here a number of questions : Whether the initial adjust-
meht of individuals to changing death rates is voluntary or involuntary?
How far is the steep decline in death" rates attributahle to autonomous
factors? Why should not an increase in the ratio of productive to non-
productive years of a child lead to a reduction in the number of children
that are required ? Why should not th~ costs of maintaining children grow
more than proportionately? Is income determined by the family size or
"is the desired family size determined by income? What exactly does an
individual forego when he reduces the family size by a small unit? Doesn't
he forego a higher level of income, consumption and/or wealth? If this
is so, then the opportunity costs of an increase in the size of family should
tend to rise more than proportionately. Professor Leibenstein refers to
the utility of children as consumption items. Isn't it reasonable to assume
that parents consider the possible level of status of children as important
items in their decisions? Suppose parents plan their families with the idea
that children should maintain at least as high a standard of living as the
parents have obtained. What if parents should like to bequeath to their
children a higher level of living status than what they have got? In the
Leibenstein thesis, children are considered solely as items of materialisti\l
utility to their parents. Suppose parents do consider the well-being of their
children as a thing to be desired on its own. The consumption utility of
children may not be constant. It is also a function of the levels of living
'status which parents visualize for their children. Professor Leibenstein
also leaves unexplained as why the social capillarity thesis should not
be applied at an early stage.
There is, however, no clear reason to believe that the population-hump
is due essentially to attempted displacements to levels of income below
the critical level. If the population-hump is due to lags then there is DO
such thing as a minimum critical amount of displacement.
Again, as will be pointed out later, not all increases in population" size
are detrimental to the system as such. As the income level rises, the level
of consumption would tend to go up and other conditions equal, a good
proportioD of-underemployed population can DOW be used in productive
employment and capital formation. Hence, the retardation effect indicated
by population growth may DOt go on increasing. It may actually. tend
"to decline. As levels of per capita income rise, it is possible that the signi-
ficance of both the income-indu~ population factor as well as the auto-
nomous growth-induced pop\,latiOJl factor would tend to decline.

116 Arl1l4 Vi;""""

 VI
Inteasity of Retardation Factors
15. Let us now examine the other retardation factors.·
Professor Leibenstein refers to zero-sum entrepreneurial activities
directed towards the maintenance of existing economic privilege through
inhibition and curtailment of potentially expanding economic opportu-
nities. At low levels of income and with zero or low rates of. growth, the
economic system would be characterised by circular flow. Most of· the
activities involve transfers of gains and losses. Now there is no reason
to visua1ize that as income level rises, the growth-retardation influences
exerted by these activities will increase in intensity. On the other hand,
they would tend to decrease. One can reasonably say that the higher the
level of income, the lower. would tend to be the incidence of activities
described above. The resistance to change will not increase when income
levels are rising. They will tend to be constant or possibly decrease. In
fact, the lUgher the level of per capita income, the greater the possibility
of forced savings, through price increases. Hence,. the greater the scope
of credit expansion and the Schumpeterian type of innovation activity. We
need not,· therefore, agree with Professor Leibenstein's postuIation that
the intensity of the growth-retardation influence on these counts will in'-
crease with rising levels of income.
16. Professor Leibenstein refers also to .the conservative activities both
of organised and unorganised labour directed against change (such as
security regulations) and for the maintenance of existing techniques. He
argues that the higher the level of income, the lower the incidence of the
above activities. In fact, one might reasonably argue that at low level. of
income with considerable unemployment, the working classes would be
mostly unorganised; expansion can take place without countering direct
and indirect demands for wage increases. Some' of the obstacles to which
Professor Leibenstein refers, are characteristic of the state of full employ-
ment in developed societies. They might be true in underdeveloped ~un­
tries in the case of those industries which. are well organised and well
developed. It is no doubt true that the concept of welfare state does imply
some pressure exerted by the state to improve the levels of living condi-
tions of working class. But these ideas are not always induced by Income
changes. They are autonomous influences. One is also not clear as· to
whether the net impact of state legislation may not be beneficlal in tenns
of work efficiency and dependability. , .
11.· Professor Leibenstein then refers to forces which resist new know--
ledge and ideas as also to the attraction that classical lmowledge and
old ideas tend to have in a stagnant system. One is not ,very clear about
the logical links in Professor Leibenstein's thesis here. It is true that ~me
 underdeveloped societies have very old cultures and civilization. It is not
clt;,ar that such states should automatically imply r-esistance to chauge.
There is, however, no reason why the resistance to new knowledge should
increase in intensity 'with risiDg levels of income howsoever small such
i'ises are. The type of state characterised by Professor Leibenstein would
be more true' of developed countries which are undergoing stagnation
because of I:esource-limitations.
18. ProfessOl: Leibenstein refers to essentially non-productive conspi-
cuous public or private consuinption expenditures that use resources that
could otherwise be used" for capital accumulation. How does it follow that
ruing levels 'of' income would increase such expenditure? How far is it
Correct to argtie that" these characteristics are true only of underdeveloped
Countries? The Tai Mahal and the Pyramids were constructed at a time
when 'most countries, were underdeveloped and 'before the idea of mate-
rial . advance made possible' by technological' revolutions had becoiIie
known. Some· of the relatively undesirable expenditures in some of the
underdeveloped countries are due to the fact that they have just become
free and there is'a certain social overhead expenditure associated with the
maintenance of organization. and administration in a free country. Conspi-
"euaus private. or public expenditures will not increase in intensity when
incomes rise, nor are they in any way associated with income changes either.
Many of Professor Leibenstein's observations are perhaps true of some
Latln American countries. Even here many of the demonstrative forms of
resource mal-allocation are perhaps due to basic. weaknesses in the dis-
tribution system. Granted again the existence of vigorous public opinion-
as is the case in many underdeveloped countries with a democratic struc-
ture-gross wastages of resources need not and do not generally occur.
I!i..Professor Leibenstein argues also that the incremental capital-output
ratio is likely to be higher in the less developed stages than in the more
developed stages: The viewpoint is again questionable. There is no reason
why low levels of income should always imply a high incremental capital-
output rntio.. Average as well as marginal capital-output ratios are deter-
mined by a number of factors like the pattern of investment, the state
of technology, qualitative factors affecting labour efficiency, the level at
which external ecOnomies and internal economies are obtained, the degree
bfutilization of capacity etc. There may be a case for a rising capital-
output ratio depending' upon the types of projects that are chosen. Any-
waY, there is no logical relationship between the rate of income change
and the value of capital coefficient. For example, iIi the case of the U.S.S.R.,
ili its' developing stages, the capital-output ratio Was lower than that in
the developed 'countrieS. In cases of India and China the capital-output
ratiOs do not appear to have been higher than those in the developed coun-
tries, It may be' title that in the 'initial stages. the 'ratio toay be high' but

lis
 'CRITICAL. MINIMUM· EFFollT' THESIS

this high ratio, however, is not a function of the level of income. We can,
in fact, say that the ratio of capital to output may be high at a subsistence.
level of income. It may tend to fall as income increases; but will certainly
not tend to rise, unless other conditions are not equal (for example, if there
are initial weaknesses in the agricultural or the transport system).
Professor Leibenstein draws attention to a number of other factors like:
~evels of physical efficiency, degree of specialization, the skill component
in labour supply etc. But these factors would not tend to hinder change.
20. . Professor Leibenstein refers next to indivisibilities and external and
internal .diseconomies. Existence of indivisibilities might imply the pre-
valence of less capital-using as well as less-overhead-requiring forms of,
organization, methods of production, techniques of education and so on.
It may be that inferior quality products and services are supplied; bu~
any way, there is no reason to postulate that the growth-retardation in-
fluences will tend to rise in intensity or that they will even maintain
a constant intensity. Nor will external and internal diseconomies increase
in intensity with rising leVels of inCome.
It is obvious that the larger the size of the market, the lower would
tend to be the unit cost of production, but Professor Leibenstein himself
recognizes that this is not a problem for large countries. The same argu-
ment applies to a number of other discontinuities. Then again, in most
cases, the choice is not between all or none. A number of adjustments
are possible in products as well as in technique. Professor Leibenstein,
for example, ignores the importance of the Wicksell effect. Professot
Leibenstein also argues that Ilrojects of delayed profitability are not
favoured at low levels of per capita income. It is true that some projects
mature after a long time though, on the net, their total gains may be larger
than the other projects which mature early. ~ the market calculus the
rate of interest would be used as a standard in order to ascertain which
of these are more worthwhile. Professor Leibenstein does not prove that
low level of per capita income must imply automatically high levels of
interest rates.
The net result of the above discussion is that the income-induced
grOwth-retardation influence need not, and probably will not at all hav,e
an upward slope.

VII
Autonomous Factors
21. Lei: us now take up the question as to how far Professor Leibenstein
is justified in giving sole attention to income-induced influences as again~t
autonomous inlIuences. The positions of the CP and ·CR curves and
·shapes may be considerably influenced by the maguitude of the autono-

-J..,," US9 V 1 N t .11.9

 :Po II. IIlIAHMANANDA

mous income-promotion or depressant factors. One can easily visualize

a Case iIi which shifts in these autonomous factors can yield a configura-
tion of positions in which the slope of the growth-promotion curve is above
the slope of the growth-retardation curve even at a given low level of per
capita income. The autonomous forces can impede also development even
if the level of per capita income is high. A viewpoint somewhat opposite
to that of Professor Leibenstein would be that it is the configuration of
autonomous forces that is reSponsible for the continuance of the state of
underdevelopment. One of the chief income-lifting autonomous factors
is innovation:' New products, new factors, new methods of production and
organization are not essentially dependent upon the rate of income change.
The policy measures to 'be undertaken to strengthen the autonomout
income-rising forces need not necessarily imply a prior shift in the levels of
per capita income. Schumpeter's theory of economic development, empha-
sizing, as it does, innovation as the chief determining influence upon the rate
of economic development, would run' contrary to that of Professor Leiben-
stein. One can argue that even at a low level of per capita income itself;
innovations can shift these ~es. Such shifts brought through forces
other than income change may represent big forward steps. Despite the
considerable attention which Professor Leibenstein has given to entrepre-
neurship, skill "formation, degree of specialization, etc. one is left with the
impression that the model is too mechanical in character: Much of eco-
nomic backwardness 'Can be explained by the relative weakness of non-
income-induced factors. .
22. There is another important aspect to the whole story. Professor
Leibensteio assumes that the basic functional relationships are reversible.
The gains that arise as Ii result of an improvement in the level of income
will disappear when the level of income falls. Similar is the case in regard
to retardation influences. Thus, for each particular level of income, there
corresponds a. unique amount of growth-retardation and promotion. Now,
a number of gains' that arise .as a result of income change would remain
in th.e system even when there is an adverse change in income movement.
An improvement in the level of wages, for example, would improve the
extent of· skill. Suppose the workers' wages are now reduced; the skill
that is obtained will not be lost. An improvement in the level of per
capita income may lead to an increase in supply of entrepreneurs as well
as an increase in the level of their activity. It might lead to an increase
in the number of growth agents. It might increase the amo.unt of useful
knowledge. These influences would not disappear when the per capita
income level falls. Improvements in business . organization, level of
techniques, degree o~. specialization, etc. would not be lost because pet'
capita income declines. Even In regard to population growth, the bene-
-ficial influences caused by a reduction in the desired size of a family would

120
not undergo a change because inco~e level is reduced. On the other hand,
If the income receivers have decided that a reduced size of family should
correspond to a particular level of income, a decline in that level would
actually mean a greater reduction in the size of family than visualized
earlier. Thus, a number of the beneficial influences arising as a result of
positive income improvement will yet remain when there are income
declines. To the extent that this is so, the Leibenstein thesis will have to
be drastically modified. Even a small displacement would lead to some
improvement, which would stand. For" every shift towards the right in
regard to the level of income there corresponds varying positions and
shapes of retardation and promotion factors. It may be that as a result,
even small displacements may lead to some cumulation. It is somewhat
puzzling that Professor Leibenstein makes no use of the ratchet principle.
This principle is in evidence even in Marshall's analysis of increasing re-
turns. Many of the observations which professor Leibenstein makes in
regard to growth-promotion can be modified in the light of the ratchet
principle.

VIII
The Otbel'" VeniOD
23. How does Professor Leibenstein's thesis stand quantitative verifica-
tioni' As seen earlier, it is difIlcUlt to find out how much of net growth or
net decline has been due to shifts in autonomous factors or to income-
" induced changes." The experience of most developed countries indicates
that they started from a low level of per capita income and grew - even
though "at a slow rate. There is no doubt the case of recently developed
countries like the U.S.S.R., wherein the rate of growth was accelerated
through deliberate policy measures. But, there is no reason as to why the
U.S.S.R. should not have gro~ at a steady rate even otherwise, nor is it
"proved that it was the high rate of growth of income that made the ob-
stacles disappear. Nevertheless, it is true that the growth rates are low
"initially and tend to accelerate and then perhaps decelerate. There are some
advantages in that a high rate of growth may lead to the perpetuation of
a high rate of growth. A high rate of growth is no doubt desirable; but
. Professor Leibenstein's thesis is that, if a high rate of growth is Dot obtain-
"ed, the system will relapse into low income subsistence equilibrium; hence,
countries which cannot grow at a high rate because of technical reasons
can never break the vicious circle. It is this statement which it is difIlcUlt
. to prove.
24. Professor Leibenstein does not give due attention to the mechanism
" of how diHerent factors affect each other. It is possible to argue that as
the level of income rises, even though steadily. the population growth may

121
rise somewhat; but the,propwtionof"population that is employed might
rise and despite the, rise in, population growth rate the marginal income-
depressant effect of a change in the level of income as a result of induced
population growth may' dec!i~e. Rising population levels 'associated with
rising income levels do not, indicate that the income depressant factors
released by population torces, are getting intensified. One may have rising
population and yet a weakening of the, income depressant factors. '
25. Professor Leibenstein again does not indicate how changes in income
levels after a stagll lead to reduced population growth rates, improvement
in the degree of specialization, etc. A rise in income growth may and does
create conditions under which more resources can be spent upon invest- .
ment in knowledge :and. in qualitative' improvement of human beings. It
may lead to conditions under ,which levels of consumption standards may
rise and forces of ,social capillarity work. These, however, are not in-
trinsically related -to income changes....As such, ,many of these forces could
occur' otherwise. Special efforts ,\jave to be directed even if income changes,
in order that these forces are sb;engthened. Not all economic poliCies will
lead to the same result. From this point of view, Professor Leibenstein's
thesis can be summed up as follows f
(i) A rise in the level of per capita income tends to, increase per
capita saving potential.
(ii) An increase in per capita saving potential, other conditions
equal, can make for a: higher level of per capita investment.
(ill) A higher level of per capita investment, other conditions equal,
can lead to a higher rate of income growth. '
These links, however. are not as direct and as close as are pointed out
.in the above chain of reasoning. High levels of per capita income may be
dissipated in high levels of consumption;, or in particular forms of con-
sumption" which ,do not contribute to income change. The pattern of
investment need also not be' of that type that leads to high income growth
rates; for example, at some stage Professor Leibenstein argues that invest-
'ment in agriculture need not necessarily be justified; This appears to con-
Hict with what he says'in the chapter on "Theory of Underemployment".
A 'high level of per' capita consumption of wage-goods can increase the
level of productive efficiency and 'thus enable high levels of investment
in the system, In a sense, Jpto a particular stage, improvements in the
'levelS of conSumption 'have a multiplier effect upon levels of incomes.
Many of the dillicultiesassoeiated ~th countries in subsistence equilibrium
,Can be explaiiledas arising on accOunt of a'deficiency of aggregate supply
in regard to consumption goods. In fact, the crucial variable affecting
population growth appears 10' be the'relationship between current levels
" of cOnsumption arid the desired level of constimption rather than the level
. of income. In' the -extreme, 'case' where income and' consumption move, at

122 ArtkG VijnaluJ.

 the same rate the distinction may not have much significance; but from
a practical point of view, as Professor Leibenstein himself recognizes, this
distinction is analytically important.
· 26. The question that may be asked is as to whether Professor Leiben-
stein's thesis can hold true in the case of some countries at least. In the
East at least, the chief cause of population upsurge is the unexpected and
· swift reduction in death rates. There is no clear case that the birth rates
are rising. Logically, the birth rates must fall if death rates are falling,
· unless the concept of a desired family size undergoes a change. It is true
·that in some of the Latin American countries, birth rates appear to have
risen. It is reasonable to visualize that under conditions under which levels
of income are rising fast and aspiration levels in regard to consumption
· standards are being reasonably satisfied and there is a continuous ten-
·dency for full employment, the above result may occur: Rising rates of
· inCQme growth may be accompanied by rising rates of induced population
:growth. There is no reason why this tendency should reverse itself just
because the rate of growth of income has reached a particular figure. They
do and might reverse if aspiration levels in regard to consumption standards
move up very fast. A high birth rate may persist because of the relative
impotence pf the demonstration effect in the Latin American countries I
'There might, therefore, be a conllict in development policies. Permitting
high levels of consumption standards might imply low levels of investment,
though when equilibrium adjustments are made, the levels of savings
'would lJ.e higher than whatever they are because of the effect of high con-
sumption standards upon population growth. The best way to bring about
a sustained rise in the level of investment in the long period may be to
permit a. rise in the levels of consumption.

IX
· Conclusion: Need of Theory of Mechanism
2:1. The above strictures on the .critical minimum effort thesis might per-
haps appear to be somewhat too drastic. Professor Leibenstein himself
,does not derive any policy conclusions from his thesis. It does appear ob-
vious that only a large displacement of income by means of a large
stimulant can provide a solution. This is hardly an edifying conclusion for
those underdeveloped countries which cannot alford a large displacement
as a result of internal forces. Leibenstein's thesis poses before them an
all-or-none choice. Either it is to be a big leap or no forward step at aI).
The picture that is dangled in front of the underdeveloped countries is
that ptoVided a large displacement is accomplished, there are no basic
"problems of development thereafter. We have seen above that even small
displacements can help. Many of the forces which induce economic deve-

J ...... 1959 V t N II -128

 P. R. BIlAHMA.l'~ANDA

lopment as well as retard it can be aHected other than through the mecha-
nism of income changes. Every improvement, however small, makes it
possible to obtain an easier and perhaps a bigger forwar!I step. Nor is it
obvious that when once a particular stage is obtained, the process would
',tend to be cumulative. Alfred Marshall used to emphasize the fact that
progress is not a matter of course. Perhaps the initial steps in economic
development are qualitatively different from the later steps: but there will
be some displacement possible through shifts in autonomous factors at
all critical stages. These autonomous forces, which elude 'manipulation
',from the me'chanism of income change, might be easily susceptible to other
, types of influences. Even with a given savings potential, much can be done
,through appropriate allocation of resources; better patterns of consump-
_tion and investment might make a higher rate of growth possible. Changes
in education, sOdia-cultural mores, improvements in the degree of mobility,
establishment of appropriate institutional agencies, etc. might go some
way towards smoothening the obstacles to growth. Hence, the practical
significance of Leibenstein's thesis may not appear as striking as the theore-
tical novelty of the analytical framework. The difficulty in the type of
-theorising which characterises the present work is that in the underdeve-
loped countries there exists no analytical system explaining the relation-
ships between different variables. We have no theory of employment and
income determination 'for an underdeveloped system; 'We know very
little about how levels of investment and savings are determined. We know
also very little about the factors affecting the choice of techniques and
the social and economic implications of various types of choices. It might
appear, that no theory might fit in this region; at the same time that some
part of every theory might appear to fit in. One extreme view may be that
it is neither possible nor desirable to evolve a system of analytical relation-
ship parallel to that of the Neo-classical or the Keynesian system in the
context of the economic situation in the underdeveloped countries. One
chief !Derit of the Keynesian method of approach is that the basic relation-
-ships can, be quantified. Hence its importance for economic policy. :rhus,
the limitations of Professor Leibenstein's approach are partIy due to lack
of a'theory explaining the mechanism of the working of an underdeveloped
system. Jt is only perhaps when some such theory is obtained, that one
. ,can argue 'about the significance of a critical minimum eHort.
. ,. . , . "

Suml1llJry
'The paper seeks to examine critically the thesis propounded by Pro-
'£essor Harvey Leibenstein in his book 'Economic Backwardness and
Economic Growth' (John Wiley and Sons, 1957) to the. effect that an

1.24
 'CllITICAL MINIMUM EFFORT' THESIS

under-developed cowitry embarking on a career of economic development

must reach upto a critical minimum effort in order to escape a low-income
Hagnation. -
The fundamental concepts on the basis of which Professor Leibenstein's
growth-promoting and growth-retarding curves have been derived are
examined in detail and on that hasis it .is argued that there is no reason
why these curves should necessarily have the particular shapes on which
the critical minimum hypothesis essentially depends. The general. refuta-
tion of Professor Leibenstein is sought to be based further on three major
lines of argnments, as follows: ( a) The growth-promoting and growth-
retarding curves are not independent of each other and it is, therefore,
illogical to derive equilibrium situations from their interaction, (b) the
autonomous factors influencing economic development are much more
important than assumed, and (c) the basic assumption of reversibility
involved in the process of slipping back, towards stagnation is inadm~sible
in the light of the 'ratchet effect' that is known to operate.
On the basis of these considerations, the all'IJr-none element in the
economic take-off of under-developed economies, implied in Profes-
sor LeibenHein's thesis, is denied and the possibility of continuous slow
growth reaffirmed.

mm
, A"'dqf.tu~4cp F«iI" <PI fli'&ICd 3I'h ~ 31lelij'1l
oft. am:. qrTIk
.:..
)II. tR• l"Siliil*I«...
- =>. 0.0fi o. , •
,Cfil.,t t:tifi ilCfl46.,ij
0 0fi
-"-" U''''
SR' '40..,1.,. +iifi 'MI~
v;r t~ ..." i mrr. m ~ f;m;u ~ Slt'IiNd f.t;lrr ~ f.!; arf.r-!;fim
~ "" fil4>Itt41 ...,1 i ~ 'f.!..,ail f"Uli"4> lI"I<'1" ~ dll4q44><11 ~ ~ r.mt
II"IfR if ~ r.r.... m: 'I<: ijift 1;~. w ft«rRr 'I>T lR¥ q i <rtmur ~.
sit. .,14ii"t'eI\.,· it ~ ~ ill"4dlat"f 'til' ~ f'<i4>rn~\ am:
Wmr-~\ Simian ~ ~~ m t.
'li'f ~ ill"4dlat"f 'I>T ~ f"qe"ul
if" It f.t;lrr ~ am: ;rnt4T ~ f.!; ~ 4mlatl 'til' ~ d!l'IiR'f it ~ 'IR 'til'
8114444><11 '1{f~, f....- 'I<: 'fiIl:'fd1£ f.. uli44> sm:r'
'I>T m lJ.Wd': dll'lIf<a
f.t;lrr 'T4T t. arrit "'" 'Ii'1; sit. "I"ii"t'el\'" t ft«rRr 'I<: f.,..,f.,f<ild aml<r
~ it fiI>t( t· (81") f<r;t;m~.. am: f'<i4>rn-~ 4"'«t1~ 1% ~ t
lI"ITJ1f tt liffir '1{f t am: ~ ~ <it'1"f 'li'f 1ffi'I1:-flio4Taff tt ~ tqt

\. flI;rnr -' ~.- growthpromoting.

\ flI;rnr - ~.. - growthretardiog.

Iv.. 116' V IN. 126

 P. R. BRAHMANANDA

~ ~ ftqfu dI'lltlFOld ~ 'IiITif'I;~. (diT) dI'Tf'I<I; f'A;rn 'I>'t 1I'1llf<rn ifi>A'Ile

~ ifiroIffl 'liT ~ F;;rwrr ifTiIT q'lU ~ ~« 'If@' dif'l'ifi ~ aft~ (~) f.p:;r m
~ <1"('Ii ftI;m;r ;;rf.f ~ 6111'1'11 dOl' ~ ~ f.!; <lit' f.lifiHI'fi1{ ~ fqq
mrr if ~ ~ srfl!m Ifl'it ;;rr~. \lJf.r;;r ~ ifr.lIm <I~1f<I' '1&:1 ~ ~~
f.!; ~ mrr if '!fir ilifi~'Ile'l 'fu .y.rr srfl!m ~ if amft ~.,
l •

lIT. "I<l~*,,~ ~ ftnrt;a' « itm arpr!'1' FOlifi"dl ~ f.!; dif~ ~.,.;r

dloi<"mqlai'f if '~ lIT ~." f'A;rn ~ ~ ~ ~~. ~ 'l<'I<l' i/'dT.
'Ii.: ~ ;f ifi~ ~ f.!; ~ fiR·fiR am etrnm: f'A;rn ~ if ~ an<rfu 'Ity t.
. .

~ ffi'I'<! ifiR1'I - autonomous factors.

y ~-rachel-11;ifi ~it \ijit",Sl ~ ~ ~ ~~rit 'f.lf.t lr {)~V;'Ii ijifllifi m"11
~ ''{.U "Ir ~- • all or none'.
126 . ArtTuJ Vii_