The Import-Substitution Strategy of Economic Development: A Survey

Author(s): Henry J. Bruton

Source: The Pakistan Development Review , Summer 1970, Vol. 10, No. 2 (Summer 1970),
pp. 123-146
Published by: Pakistan Institute of Development Economics, Islamabad

Stable URL: https://www.jstor.org/stable/41981693

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

is collaborating with JSTOR to digitize, preserve and extend access to The Pakistan
Development Review

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
The Import-Substitution Strategy of
Economic Development: A Survey
by

Henry J. Bruton*

Over the past several years, the import-substitution strategy of develop-

ment has been examined in considerable detail by numerous economists. For
the most part, these studies have been concerned with one or another side of this
many-sided approach to development policy. It now seems useful to review
this literature in an attempt to isolate major themes and arguments and to try
to put together a cohesive and comprehensive picture of where we stand now.
This paper is not intended as a summary of the individual articles and books on
import substitution, but rather is aimed at bringing together the theoretical issues
and the empirical results that not only are of interest in themselves, but which
also seem to add up to something that might legitimately be called an approach
to development. To do this, I asked three general questions: 1) what appears
to be the essential mechanics of import substitution as it has been practised in
various countries that have been investigated; 2) what problems have
emerged, and why, as a consequence of the conventional import-substitution
(IS) strategy; 3) what has been found that suggests or leads toward an alter-
native approach to development that incorporates the good and eliminates the
bad of this conventional approach.

The general conclusions that emerge from this survey may be summarized
in the following way. Although the countries that have built their development
policies around import substitution have experienced great difficulties, there are
reasons to believe that a satisfactory approach to development can be built
around this approach. The difficulties have arisen as a consequence of the
activities selected for domestic development and of the methods employed to
provide the incentives to bring about their development. Part of these diffi-
culties arise from a view of the economic development process that now appears
misleading, and part arise from assumptions about the developing economies
that the empirical evidence shows to be unacceptable. More specifically, it
appears that the distortions and misallocations that have been imposed on the

♦The author, a former Joint Director of the Pakistan Institute of Development Econo-
mics, is at present associated with Williams College, Williamstown, U.S.A. The research for
this paper was supported by a grant from the agency for International Development to
Williams College, Mass., USA. The Agency is in no way responsible for any of the views
expressed in the paper.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
 124 The Pakistan Development Review

economy by the conventional approach to IS have them

that contributed significantly to the widespread failu
as satisfactorily as many expected. To repeat the pr
different way: to implement their IS policy, countrie
and techniques that seem, in effect, to prevent th
successful.

On the other hand, there are important aspect

appear most useful indeed. The concept of a strateg
policies designed to create readily apparent investm
the notion that new activities must be established in t
equally acceptable and equally important. What app
to achieve these advantages of IS without creating the
tions that to-date appear so ubiquitous, and that appea
The work on remodelling the IS strategy is just begin
more clearly understood than is where and how t
therefore, spends more time on the difficulties with
recommendations. The latter have, however, begu
are implicit in many of the criticisms of the IS str
date. It is important to keep in mind from the ou
of the literature does not show that IS is necessarily f
a recommendation for abandonment, but rather one f
to its implementation.

Section I examines briefly the notion of a str

general, and the origins of an IS strategy in partic
IS strategy first in very broad terms, and then a
aspects. Section IH considers some of the more pos
and Section IV suggests a range of policy issues th
positive results. As noted, these policy issues are n
as they have yet to be fully analyzed in the liter
something of a link with the past work and leads (hop

Both terms, strategy of development and import substitution, merit

attention at the outset. We will begin with strategy in general and work our
way into import-substitution strategy in particular.

A. Strategy of Development

One may contrast strategy with theory of development in several ways.

The latter term seems to refer to an explanation of the sources of growth - to
increased capital and labour, increased productivity, improved resource alloca-
tion, demand conditions - and how these sources act and interact to produce

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Brut on: Import-Substitution Strategy 125

growth. Development policies flowing from this kind of

with capital formation, saving rates, technical change
items. One looks also at rates of return on alternative in
and at using resources to maximize present value of f
Emphasis is placed on the availability of these sources of
may be used to produce an acceptable growth rate.

Strategy, a concept introduced into the development

Hirschman [10], implies something a bit different. Spe
(or both) of two things. In the first place, it suggests
torment of the economy, or by some gimmick, the un
economy can be set in motion. The primary obstacle to
absence or inadequate supply of the source of gro
absence of a catalyst that creates an environment or a circu
the already available, but unrealized, resources. Hirschm
mobile , a prime mover, necessary to activate the late
investment allocation decisions, one looks not at rates o
projects themselves and present value of income stream
One looks rather at whether or not a specific capit
circumstances which may be expected to induce furthe

Hirschman's well-known book is perhaps the most

this notion, but there are other arguments that employ es
The models of W. A. Lewis [16] and Ragnar Nurkse [18]
unemployment assumption have this strategy notion at
Rodan's 'Big Push' [22] and Leibenstein's 'Critical Minim
as well that supply limitations are not at the heart of the
economy does not grow, and imply that if the econom
catalytic effort, it will begin to grow.

In the second place, strategy is frequently used in

tinguish between marginalism and structuralism, and m
with respect to the investment decision. This year's in
to the existing capital stock, may be allocated in accord
marginal-productivity criteria. This small investme
within the existing structure, but will not alter it. Yet
difficulty is a structural one. In general terms, structure
the composition of output, techniques employed, impor
arrangements, educational system, class mobility,

policy objective, then, is not to so allocate investi

of income streams are maximized in the narrow s
tible resources to create a new structure whic
quantity of resources, is presumed to yield a h
within the existing system, and, what is more im

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
126 The Pakistan Development Review

growth. In this context, strategy refers to an approac

seeks to modify the nature of the economy in qui
assumption that such modification is necessary before

This strategy of development is most clearly i

works of Raul Prebisch and other Latin American
various arguments of Maurice Dobb [8]. The notio
circles as well. In discussing India's developme
"Moreover it is very early for India to be able to
advantage will lie" [17]. A bit later in the same ar
as large as India with a wide complement of natu
surprising if it did not turn out to be economical for
almost everything". In both these statements, Li
future after India has changed her structure into
state, then she will allocate resources in a manner to
fashion. In the meantime, presumably some (unst
allocation is to be used. Similarly, in many of th
that a country needs or requires specific capital it
dictates of conventional maximizing procedure), t
on some notion of a changed or new structure to b

These two bases of strategies are not mutuall

to change the structure can be used as a means of
And the objective of a policy or a strategy may in
At the same time there are advantages in keepin
ference will be made to both origins, a prime-m
changing origin, of strategies.

Consider now more specifically import substi

development. To a very large extent it was arrived at
the most common (frequently observed) strategy
perhaps the only one. This is true partly because
possible strategies and partly because it is easily
clear that most developing countries tend to slip in
ment. In this case, then, IS is not a strategy in the
approach, but rather a situation into which a count
development effort is made.

It is this ease of initiation that seems to be th

source of the prevalence of IS's popularity. To
create investment opportunities or to change the stru
matter1. If a country decided on an export-promo

iNote that it is the initiation of IS that is stated to be "eas

tion. A principal aim of later sections of this paper is
question.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton : Import-Substitution Strategy 127

not so evident how to begin. Similarly, to build a developme

industrialization without IS does not lend itself to an obvious se
same may be said of a strategy built around increasing agricult

One may, then, visualize the origins of IS in somewhat the

policy-makers become aware (or convinced) that conventional op
cannot (do not) lead to policies that produce as rapid a rate
able either because the need for an explicit prime mover or
structure of the system is (or is thought to be) alien to growth
then searches for an alternative guide to policy-making - o
the rate of investment and create a larger number and more o
opportunities. Even if it is assumed that a number of strategie
IS strategy is likely to be selected because it apparently calls m
out imports, a task most governments can be expected to a
that the country is likely to be experiencing balance-of-payme
to the attractiveness of this approach. Also IS will create g
where very obvious investment opportunities exist, thereby cr
situation that will generate Hirschman's primum mobile. F
will keep out goods previously imported, thereby encourag
production, and thereby (again) contributing to struct
strategy would then seem to meet all, or at least a wide ra
believed to be blocking development.

That few policy-makers have, in fact, systematically surv

policy opportunities and concluded that IS is the best hope
portant. That they have in fact most often backed into suc
balance-of-payments (or some other) crisis is indeed not in
preceding arguments. It merely means that the situation that
is a crisis, not that the crisis is the rationale of the action tak

B. Import Substitution

Import substitution has a variety of meanings in the

frontal attack on a simple definition is no small under
attention is limited to a single product, there is little difficul
a policy that reduces or eliminates entirely the importatio
and, hence, leaves the domestic market exclusively for d
Measures of IS that are based on changes in the ratio of i
products to their total domestic absorption are of course c
definition. Difficulties emerge when we seek to aggre
reduces the proportion of the quantity of a product that is im
same time, increase that proportion for another product. Whe
this case should be called IS becomes ambiguous. Simila

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
128 The Pakistan Development Review

aggregate import-GNP ratio may hide the impact of IS p

It is a bit misleading, therefore, to rely on changes
identify the effectiveness of IS policies aimed at changin
absorption of a good that is imported.

The issue can be put into an understandable fram

the objective of IS policies that are aimed at reducing
modities. Few policy-makers actually wish to shut ou
already noted, the objective is structural change or s
incentive-creating idea. Part of the rationale of the s
is that the existing structure makes the economy u
matters outside its own control. One of the guides t
the structure be changed so that the economy is less at
trade activities? The extent of the achievement of
easily measured at all, and certainly not by means of
import-GNP ratio.

The picture then seems to be this. Developin

achieved sustained growth because of their structure
mover. The IS strategy to change this structure a
mobile is to replace imports by domestic production
Thus, IS in the narrow sense is limited to specific a
by increases in the ratio of domestic production to t
In the broader sense, import substituting within individ
the more far-reaching objectives just stated. These l
the long-run success of IS rests heavily on the specif
substitution is carried out, and the methods chosen to b
expansion. An analysis of the IS strategy of developmen
examination of the rationale of the selection of thes
sequences of the methods and policies used to bring abou
On this latter issue, interest is primarily directed to
policies on the growth of the economy.
II

In this section, we first examine the broad patter

over the last several years or so. Then, we look at a n
pattern that stand out in the broader, overall picture
we try to establish a series of positive conclusions th
cussion of the main characteristics of the IS model.

A. The General Nature of Import Substitution

As noted in Section I, the immediate approach of

imports of specific activities with domestic product
simplest way to do this in a market economy is to impo

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton : Import-Substitution Strategy 129

This limitation on imports has a variety of impacts [19].

economy that make for obvious investment opportunit
activities of the economy, usually manufacturing. Resou
into new industrial channels (structural change) and th
enlarged. In the latter sector, expected new profits may
the saving rate and to further increases in investment, as ca
to be high savers and accumulators. The new capital goo
are paid for with the foreign exchange released by the redu
of the commodities whose domestic production is being
simple picture, two not-so-simple questions arise: what m
off from foreign competition and what method should be em
off. Most of the literature under review has something to say
on these two issues. Consider first which products to pro

In practice the answer is consumer goods in almost a

certainly the case at the outset of the IS process. There
this concentration on consumer goods.

The simplest reason is that the cost disadvantage bet

produced and imported consumer goods is less than for c
termediate goods. Thus, it appears to policy-makers tha
importation of consumer goods, the advantages of IS can
mum costs. The cost argument is supplemented by the e
demand, i.e., the consumer goods are being imported, w
capital goods, intermediate goods, or raw materials dep
of an investment programme. Finally, consumer goods
are universally deemed inessential to development, and
costs and in their prices assumed to be less harmful than
of capital goods. The latter goods are, thus, imported w
and frequently at exchange rates that greatly understate
Thus, not only is the domestic production of capital good
by the tariffs and exchange-control policies, but also by
that keep their imported costs below real costs. The ra
policies favouring the importing of capital goods rests in ge
as to the essentiality or to the strategic role of physical cap
process. The relatively low tariff rates on raw materials
mean further that the protection afforded the value add
goods producing activities is markedly higher than the rate
good itself would indicate. Hence, both the extent of protec
activity and the extent of the cost disadvantage of that
understated by an examination of nominal tariff schedules.

This first stage of the IS process ends when the expansi

sumer-goods capacity hits the limit of the domestic m

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
130 The Pakistan Development Review

the economy has a number of new activities wh

form of protection and whose expansion cannot
then, must take place in activities which are now im
lished IS activities must enter the export marke
reached the hurdles the developing country must
number of reasons.

1) Since this second stage is difficult (and the fir

is extended to its maximum extent. Thus, prote
a range of consumer goods as possible. One gets,
has called "premature widening" of the productiv
into a large number of relative small-scale activiti
on a few. Advantages accruing from scale effects are
of activities, and may well be absent entirely.

2) If IS is to continue, the economy must mov

mediate- or capital-goods production. This shift n
of growth (the cost of the domestic production of
goods is larger relative to their import costs than is
consumer goods for which domestic supply has
but it also affects the costs of production and the le
the new consumer-goods industries. Evidently, if
goods had been using imported intermediate (or capi
and is now forced into using higher cost (and pro
cally produced inputs, his production costs will ri
tariff rates, protection on the value added of co
decline. To maintain the same level of protec
sumer-goods activities, therefore, requires an increa
This pushes up the cost in these activities, and (
lessens their chance of exporting.

3) At the same time that investment costs are

developments are occurring that tend to reduce t
As the economy expands its domestic production of
may fall relative to those of domestically pro
income probably shifts away from government a
In the second place as the domestic output of con
demand for foreign exchange for the importati
material inputs, not all of which can be produced
are not forthcoming, underutilization in the con
The evidence of this underutilization creates obvious
sumption simply because, if the capacity to prod
would appear nonsense not to use it. To use it m
saving rate below that which might be achieved

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton: Import-Substitution Strategy 131

rate in these circumstances will not produce investm

way to use the resources released from producing cons
capital goods, i.e., capacity built to produce consumer g
produce capital goods [26] and exporting is not possib
Added to these two reasons is the possibility that the m
the economy by the IS policies will reduce total output
have reached in the absence of such policies, and hence
to fall even if the saving-income rate remained constant.

The outcome appears rather clear: at the very poi

must save more in order to maintain its rate of growth of
incentives and pressures to reduce that rate. Essentia
consumer goods - or the investment criteria of least
items to import substitute for leads to developments
eventually halt the growth that the original protection
dampening effects have to do with investment moving
costs - consequent to the failure of the new activities to e
- at the same time that saving declines.

This is not, however, the end. There is the fur

complications arising from the nature of the protect
created activities. The most common protection pa
aspects: 7) an exchange rate that undervalues foreign e
burdens to the effort to enter the export markets. As
rationale of such a policy is that it keeps the price of imp
relatively low and so encourages investment, plus the
demand for most export items is assumed to be inelasti
is undervalued, then of course equilibrium in the balan
achieved by means of tariffs or the direct licensing of imp

The overvalued local currency may encourage investm

characteristics much less likely to facilitate growth.
intensity in production higher (due either to choice of
product) than would be the case if foreign exchang
thereby dampens the employment effect of the IS ac
a level of import intensity in production that is inco
exchange-earning power, i.e., it penalizes export comp
need for imports. Further distortions are imposed by t
necessary to keep the balance of payments under con
and exchange controls become more rigid, for the re
distortions increase. As the latter occurs, supply bott
economy's capacity to transform its resources into cap
services is increasingly impaired. These developments a
sustained growth.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
132 The Pakistan Development Review

The picture just described is perhaps a bit ext

one country, but the basic pattern seems applicabl
in Latin America and Asia. To summarize briefly
necessity of selecting activities to protect and
procedures have left a residue of high cost, nong
countries. Added to the problem of the selection
their protection. Again, conventional procedures
of distortions which seem to increase as IS is purs
tortions as simply resulting in a lower level of outpu
no distortion. This it surely does, but there is ev
portant consequence, namely a major reduction in
to maintain its growth of output in the manner
successful IS policy must find ways to select the r
must protect them in ways that will not so disto
effectively thwarted.

B. Some Specific Aspects of Import Substitut

In order to consider how these latter objectives m
is helpful to look more closely at some specific r
practices foster and which seem to be especially d
to emphasize as well a few hints as to how such effec

These points may be considered under the foll

1) A structure of production emerges in which
available capacity without large-scale capital inflow
if one assumes that prior to the construction (or p
there exist several options as to techniques of
product to produce, but after construction both tech
At the time of investment (of the construction of
of import-substitution policies are reflected in impor
goods) that are below the costs of producing the ex
the underpricing of domestically produced goods,
tion from outside competition. In such a situatio
incentives to choose a product and a technique to
patible with relative supplies of inputs. Thus, a str
tion efficiently only with a rising flow of impo
however, is discriminated against by the very same f
structure. Unless then a rising flow of unrequired
eventually see underutilization appear. Given an ex
sort, correcting the source of the distortions (by
of capital, labour subsidies, etc.) will not immediat
unless one is willing to assume that technique and
on existing physical capital in a rapid and pain
correct the distortions in these ways can exacerbat

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton : Import-Substitution Strategy 133

tion may lead to inflation, raising interest rates reduce in

events do occur they may drive policy-makers to more dir
to the distortions. An important by-product of the sam
the underutilizations is that they also result in a rate
employment that, independently of the level of ut
that which might be achieved with different policies.

2) The above developments lead to the position th

are brakes on development. Thus, Gordon C. Winston
capital-goods sector can be too small given the potentia
capacity. In [26], he offers evidence to support the view
of the Pakistan development effort, the size of the capital
to be the limiting factor in that country's growth. More g
leads to the development of an economy whose variou
and are feasible only by fluke as the mechanisms for c
structural consistency have been destroyed by the IS po

External aid may then be claimed on the grounds

shortage is relieved by aid then the capacity made idle
brought into use, and the economy will in consequenc
rapidly [5]. This result does hold for a given period, i.e
distorted. But aid that simply provides the missing inp
harm than good. It can enable the economy to continu
the existence of the distortions, i.e., aid can be used only t
to continue to live with the policies that produced the spec
with. Aid used in this way may hinder the correction of t
knowledge of their existence [4]. Aid used to permit correc
distortions will have a much more permanent effect o
In order to use aid in this way, however, the distortion
origins understood, and their method of elimination know

3) A further complication arising from most appro

with the extent and sources of the distortions imposed
issue is seen most clearly in terms of tariffs and indire
imported product plus any sales tax on the imported p
on any domestically produced competitive product pro
mation to the extent of "nominal protection" [13]. The ext
(in contrast to a product) is protected depends as well o
tion" afforded the inputs used by the particular proces
on these inputs. Similarly, the not uncommon overva
currency adds a further element that must be introduced
the consequence of (e.g.) tariffs and indirect taxes. Le
important, is the fact that the several markets in an econo
the extent of their competitiveness, in their pricing po

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
134 The Pakistan Development Review

that they absorb or pass on taxes. There are t

this. In the first place, consideration is almost a
target actually aimed at with the given instru
evidence of an examination of the many policy instr
target, and that act on the efficacy of the specific i
tion.

The second place follows from the first plac

consequences of a single policy - e.g., understandi
tariff - can be extremely difficult, even practi
particularly relevant in a situation where policies (ta
circumstances, distort are instituted to correct a
that also distorts. In developing countries where
is assumed not to) work effectively, and hence di
this point has considerable practical relevance.

The studies of the structure of protection by

Paul G. Clark [6] shed considerable light on these
empirically the extent and nature of protection by c
for indirect taxes, input taxes, and overvaluatio
out for Pakistan [13] and for Brazil [6] and the d
substantial differences between the nominal tariff r
to take into account overvaluation, indirect taxe
the Lewis study shows that one sector may also be p
minatory government policies against another se
the agricultural sector is estimated to have (in th
35 per cent less when trading domestically than it c
in the international market. This disadvantage o
chief supplier of wage goods to the manufacturing s
real wages for manufacturing workers favourabl
cost pressure in the manufacturing sector modest.

Evidently just looking at the tariff schedule on

one to predict its consequences in terms of inve
correcting distortions due to tariffs requires more i
tariff schedule. It also means that a wide range o
context of existing policies, and examined in the con
policies are institutional constraints or are thems
The unravelling of all this in an economy with wi
controls, import and investment licensing, and an
usly a task of considerable magnitude, and one that
formula.

4) Underlying the IS approach to developm

economy that is, as noted in Section I, to be alter

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruíon : Import-Substitution Strategy 135

ment is aimed at creating a new and different structure th

achievement of this objective. Most of the new activit
domestically at rates competitive with similar activitie
world, and hence protection of some sort is required. I
policy-making of IS is the assumption that costs in the new
and a new structure will emerge whose routine functio
GNP. The argument may be put this way: the products
niques employed in the developing countries are becomi
This fact is reflected most clearly, but not exclusively, in
trade. The country then has a choice with respect t
investments: it may put its new resources into traditio
facing increasingly unfavourable terms of trade, or it m
activities that are currently operated at costs well abo
abroad. If it chooses the latter, it must be with the ex
fall secularly, and at some point the new structure
activities that can meet world competition [2]. Evident
has to do with the rate of fall in costs, i.e., the rate of
in the IS activities.

The evidence developed in [1 ;3] suggests not only that

has been generally low relative to that achieved in advan
that the very policies designed to implement the IS app
ficantly responsible for their poor showing in the prod
repeat a point made earlier, to implement their IS polic
instruments and techniques that seem, in effect, to preven
being successful. Evidence on this latter point is not w
the kind of distortions noted under item 1) above app
explanation. Added to this is the existence of a situation
impossible to exploit economies of scale, in which comp
which the imported technology is unsuitable to fac
generally, one may say that the failure of the conventi
take into consideration the strategic role of productivit
it being not only ignored but discriminated against in the
and a domestic capital-goods activity are penalized.

The neglect of productivity growth can in part be tra

acceptance of a growth model in which physical capital for
of the growth process. In paying primary, if not exclusive
capital formation, the policy-maker neglects not only that
which is crucial to the success of an IS strategy, but, as m

Productivity growth here, and elsewhere, refers to labou

(labour and capital) productivity.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
 136 The Pakistan Development Review

that which is responsible for well over half of ob

currently rich countries.

5) Finally, it is useful to mention a few rather mo

of most IS policies. Most developing countries depe
duties as a source of government finance, and as co
carry a high tariff relative to capital goods, the
former can contribute to a government finance p
do more. Since income taxes are difficult to enfor
the government into greater reliance on excise du
duties, as a means of acquiring finance. A second
note has to do with the effect of IS on the trade balance. There is little evidence
that even hints that the trade balance improves as a consequence of IS policies.
This is surely the case in the short run, but seems true as well in the longer run.
This is partly explained by the widespread tendency of IS policies to discri-
minate against exports, but also by the failure of IS to reduce imports or import-
GNP ratios, i.e., IS policies do not reduce the demand for imports. This point
comes through in many papers, (see, especially [23]).

Following on the heels of this evidence is the point that as the composition
of imports changes from heavy concentration on consumer goods to emphasis
on capital and intermediate goods, the economy becomes increasingly vulnerable
(as to the maintenance of the level and the growth of output) to interruptions in
the flow of imports. For example, curtailing the imports of 1000 dollars of
shirts will have less impact on an economy than will the curtailing of 1000 dollars
of chemical cellulose used in the manufacture of shirt material [25]. A corollary
of this point and indeed all of the points mentioned here is that the economic
structure that the IS approach to development spawns is no more flexible and
adaptable than the one that is sought to be replaced. Flexibility, adaptability,
responsiveness are attributes the advantages of which are disputed by no one,
and their acquisition is part of the goal sought after. Again, however, the
evidence shows that policies aimed at this goal have the opposite impact and
tend to solidify the structure rather than to grease the wheels on which it operates.
Although little evidence exists that helps our understanding of why some econo-
mies appear much more flexible and responsive than other economies, the evi-
dence does suggest that the very alienness of the imposed new structure does
play a relevant role [3].

A simple example may illustrate this point. A machine imported by a

low-income, nongrowing country not only creates a required flow of imported
spare parts and raw materials, but it can also impose new skill demands, supply
and marketing demands, work attitude and organizational demands. This
latter group of demands is such that the importing economy cannot respond
quickly and easily to meet effectively a situation other than that called for by the

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton: Import-Substitution Strategy 137

exact formula envisaged in the operation of the new m

somewhat similar to a firmly established, well-groomed
splendidly as long as the routine is uninterrupted, but
a file goes astray. In the same sense, an economic str
ventional IS policies is so incompatible with the rest of
ments to changes in demands, in technology, in any
accomplished painfully, if at all.

The five preceding categories of characteristics seem

IS approach. Obviously, this package does not apply e
countries that have been researched. It would be corr
that the range of issues just reviewed reveal accurately
sequences of a widely employed approach to developm
how and why these consequences can slow up the gro
productivity growth, can increase the need for rising l
indeed reduce, the extent to which the economy can explo
ces, can in fact create a situation that finds continuin
unachieveable. Also, evident in the above list are idea
lead to reformulations and modifications that suggest
development and alternative sets of policies. These po
considered a bit more explicitly now.

Our positive results may be considered under three broad headings:

distortions, productivity growth, and what earlier was called alienness.

1 ) Distortions of IS Policies

The most pervasive impact of IS policies has been that it distorts the
economy, and these distortions do not, except in rare instances, correct them-
selves. Indeed, as shown above, economies become increasingly distorted as
the IS process continues. Three assumptions seem to underlie much of the
analysis that leads to the policies that in turn produce the distortion. One has
to do with the assumptions about the responsiveness of the various sectors of
the system to relative price changes, the second to the existence and impact of
external and internal economies that accompany the establishment of new
activities, and the third to the role of capital formation in development. If one
assumes that relative prices have no effect on anything then they can be safely
ignored, and distortion loses much of its meaning. Consequently, the policy-
maker can ignore the impact of misleading price signals on his economy.
Similarly, if expected external (or internal) economies are assumed great enough,
then any activity can be justified, and there appears great, and unsupported,
optimism implied by many policies as to the existence of such economies in the
newly established activities. And, an assumption seems to prevail that amounts

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
 138 The Pakistan Development Review

essentially to the assertion that a high rate of ph

overcome any obstacle. Evidence from several of st
none of these assumptions can support the weight th

The most specific evidence on the role prices

foreign trade. In [24], for example, John Sheaha
leave little doubt that maintaining reasonably correct
significant effect on export earnings from Colombia
and oil. In [7], Paul Clark's equations explainin
price elasticities ranging from - .35 for capital equipm
struction material industry. While elasticities with
stand-in for GNP) were higher than the price elas
groups for which both investment and price elas
evidence indicates again that sufficient substituta
imported goods exists in Brazil, that to ignore it
extent to which the Colombian and Brazilian equ
supply parameters can not be told from either s
some extent both are effective so that the computed
estimates of a structural parameter. Nevertheles
that a price effect does exist and which in turn i
responsiveness in the economies of significant magni

There is evidence on this issue on a more gen

Mexico [21], Clark Reynolds suggests that postwa
different from those of many other countries. No
much lower than in most other developing count
moderately, if at all, since the 1930's. Also the m
to supply products of reasonable quality and at a r
years of their initiation or their protection is red
also indicate that considerable substitution exists between domestic skilled
labour inputs and imported intermediate goods. Finally, Mexico's IS activities
appeared more heavily concentrated in labour-intensive industries than has
been the case elsewhere. (Though this latter is not necessarily a desirable
condition, it appears to be so in this case.) Evidently, Mexico's relatively
strong showing since 1950 is explained by a great number of factors and cannot
be attributed exclusively to the points just enumerated. It is, however, appro-
priate to emphasize that these attributes of Mexican policy did prevent the
marked distortions from appearing that would impede the other advantages of
Mexico from having their full effect.

A last example has to do with Pakistan. In Chapter VII of [13], S. R.

Lewis' discussion of the effect of economic policy on the growth rate and
composition of output shows that policies that tended to correct the distortions
in the system did produce changes that resulted in a more efficient use of both

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton : Import-Substitution Strategy 139

domestic and aid-provided resources. Both the Expo

import-liberalization policies seemed to affect alloc
facilitated a more rational use of her resources. Oth
Lewis to some extent offset these effects, but the point

On the question of external effects, the evidence

there is, is of a more negative kind than that just m
that structural change itself does not automatically cont
nal economies. There is virtually no empirical eviden
to, seldom specifically identified, "external" effects
training, technological improvement, economies of s
establishment of import-replacing activities. Indeed
result in activities in the developing economies being les
advanced countries also have negative effects on th
externalities. More positively, one may say that th
sidered explicitly with respect to individual activitie
as to their sources are not acceptable. This is not to
change as an end in itself may not be important. It
The point here is to emphasize that our studies do no
do, in themselves, bring about major external econom

The third implicit assumption underlying much

do with the role of capital formation in developmen
which our studies do not directly consider. The evide
that merely more investment does not mean a highe
not mean of course that capital formation is not import
It does mean that merely generating a high rate of inve
assure growth. More specifically it means that a high
not in itself solve - or prevent from arising - other,
A still more important conclusion, noted in several stud
attention on capital formation irrespective of other
defeating developments.

These arguments have two broad categories of im

place they mean that distortions are very real and m
in devising and appraising policies. For example, if th
lity between labour and capital (and among products)
labour and capital would not be relevant to the dist
affect the distribution of income via wage policies (m
severance pay, etc.) will have no effect on choice of tec
Where there is such substitutability, however, these
tortions. Similarly, if the exchange rate's role as an
gible, then to undervalue foreign exchange to encourage
policy. If, as now seems clear, the exchange rate do

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
140 The Pakistan Development Review

function, then to keep it below its equilibrium level d

damage of which cannot be overcome by the mere f
formation. In general then, we can say that the
adaptability suggested by recent studies shows th
specific targets do impose distortions on the syste
impossible to achieve the specified target.

The second implication is even broader than the

it. The results suggest that the misallocations pro
policies not only reduce total output below the level th
reached, but it also reduces the growth rate, princi
productivity growth and the flexibility of the ec
that the distortion issue is a most strategic concept
to ignore it is to court failure. This we believe to be a
be said on its specific policy implications later.

2) IS-Created Alienness

We have emphasized that IS seems to create act

alien to the economic and social environment of t
applies not only to the factor endowment as such,
range of characteristics and attributes of a society
produce goods and services and to respond to unp
setbacks). Alienness is not independent of, but also
the distortion issue just discussed.

The clearest case of the point appears in [3] wh

that Latin American countries exploited the opport
World War II in a more effective way than in later ye
was no inflow of physical or financial capital. Th
use what they had, to build with what they had, and
own saving. They did all this in such a manner th
much higher rate than in the postwar period when
on to a much greater extent. Also domestic resour
utilized in the War period than after, despite the
accumulation was much less. In being forced to u
nomic structure created was much more consisten
later years when IS policies became so paramount.

In Lewis* discussion of changing relative prices an

costs of new activities in Pakistan decreased show
activities where the inputs fit well with what Pakista
relative prices - hence presumably costs - dec
suitable, were also less successful. Natural resource
but certainly the whole story is not simply one of na

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruîon : Import-Substitution Strategy 141

Sheahan's model of the Colombia economy show

became more rigid, more dependent on imports as th
structure evolved [23]. This may be contrasted som
structure of which evolved in a manner much more consi
cultural heritage, entrepreneurial ability, organizing
skills.

These matters are difficult to separate from the distortion issue already
emphasized. The key point seems to be this : that if a country's economy evolves
in a manner consistent with its resource endowment, then that endowment can
be near fully exploited.

3) Productivity-Growth Dampening Effects of IS

The argument above indicated that the conventional approach to IS

creates conditions that dampens productivity growth, and rapid productivity
growth is essential to a successful IS. So an important positive notion emerging
from our studies is that countries must pursue policies that lead to a high pro-
ductivity growth or IS must fail. The question of course is what are such poli-
cies? The first point to emphasise is simply that productivity growth does not
just happen. It is not a time trend to be appended. It is rather a phenomenon
to be explained, and despite the absence of data and argument it seems clear
that it (productivity growth) is and can be affected by economic policies. In
many studies where this issue is treated most explicitly, evidence seems to indicate
that productivity growth is more likely to reach acceptable rates in economies
where distortions and alienness are at a minimum. Certainly it is correct to
say that there is nothing incompatible between distortion free economies and the
achievement of an acceptable growth of productivity.

Stronger positive statements are more difficult. In general it does seem

clear that most developing countries are not paying sufficient attention to the
possibilities of adapting and modifying imported physical capital in such a way
that it more nearly fits their domestic economies. To do this probably means
the allocation of more resources explicitly to this task. It certainly means a
conscious recognition of the necessity of this sort of activity by both government
and the private sector. General considerations indicate that this can best be
done at the firm or plant level. Hence, policies that put pressure on firms or
that provide an incentive to them to seek ways to increase the productivity of
their resources are an essential ingredient of development policy.

With respect to the effect of education on the productivity of labour, two

general points emerge from the literature on education. In the first place, the
shortage of the kind of skills learned in the classroom do not appear to be a
specific bottleneck. The educational establishment does not turn out individuals
who are immediately employed to break a bottleneck. Rather, the productivity

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
142 The Pakistan Development Review

effect of education is to equip a person to adapt and to

as they appear. Perhaps most directly they equip a per
job training. The second point has to do with th
economies where productivity is generally low, it is un
difficult tasks of educating. This means that many
large sums on education may be doing it on rather
The not-very-startling idea emerges then that the dev
need more education of the kind currently being prov
productivity of the educational activities.

Productivity growth probably is not completely

by technological improvements, and by raising th
however, are surely a significant part of the key
evidence that productivity growth is a strategic elemen
ful IS policy means, therefore, that these matters are
the key to the devising of effective development polic

In the preceding sections we have discussed the

usual approach to IS, have tried to identify the specifi
that seem to account for the difficulties that so m
policies find themselves in, and finally we sought to o
what appeared as the more positive findings of our stu
we seek to consider these positive issues in terms
policies and strategies that have emerged.

Few studies suggest a simple laissez-faire , free

arly, few deny the advantages of industrialization i
erally, the importance of changes in the structure o
and in the composition of their output. More i
strategy, as defined earlier in this paper, has great me
that many governments attack the development pr
The consequence of this is not only that demands are p
machinery which cannot be met, but also that policies
and full consequences of all aspects of the attack ar
item 2, Part B of Section II). Therefore, the notio
development objective should consist of a small numbe
has great merit. The question of which areas the go
attention depends of course on the assumptions th
specific economy functions and as to the nature of
emphasized above, the IS strategy as conventionally
tions which recent studies have seriously questioned.

What then can we say about strategic variab

growth on the basis of the preceding discussion o
kind of policies exploit these variables and these me

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton: Import-Substitution Strategy 143

that they induce industrialization and structural chang

necessary for development?

Our analysis suggests that three items, distortion, alie

tivity growth, are more demanding of attention, i.e., mor
variables which have occupied attention in the conven
governments cannot do everything, they should then con
Specific policies would then be appraised as to the extent t
or undistort the system, ii) encourage projects consistent
teristics of the economy, and iii) encourage productivit
successful in these areas, then other conditions necessar
(e.g. y capital formation) may be expected to be induced o
out major policy measures. To repeat our well-worn t
these other, these "nonstrategic," growth-producing v
strategic variables and cause the difficulties which we hav
IS strategy.

What kinds of policies make sense in this general con

are submitted with brief comments as to policies that
above-stated approach to development policy-making.

1) Uniform tariffs. Such tariffs do provide protect

minimize the distortion effect of protection, and leave la
producers to decide what is to be produced and what
established. Uniform tariffs do provide incentives for new
the primum-mobile function.

2) A possible alternative to the preceding that mig

some countries is a strong devaluation plus a tax on certai
which foreign demand is inelastic. This approach brings t
costs down relative to world costs without sacrificing for
from major traditional exports.

3) Subsidies to firms doing research. Research is

best, and there are surely external economies in such a
if firms will engage in any kind of research without som
the subsidy is to firms. The question of government
another matter.

4) Tax advantages based on productivity grow

course is to encourage productivity growth by mean
profit expectations. Note that it is productivity grow
a high productivity. Indeed, a high level of labour pro
be due to the use of a technique that, from the economy'
uneconomic.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
144 The Pakistan Development Review

5) Tax advantages or subsidies based on labou

capital employed). This is the simplest way to intro
the common existence of a labour market in which
opportunity cost. One commonly observes, of cours
est subsidies to firms on the basis of investment.

6) Indirect taxes or subsidies aimed at correctin

Such taxes can help compensate for monopoly pra
they can be used to reduce the difference between
returns. Examples have to do with electricity use,
capital goods.

7) Establishment of activities that provide ma

wide scale. Such information can help with invest
with inducement to invest, and with the firm's pla
ventory, etc., all of which add to the efficiency of its

8) Foreign aid to be used to implement poli

often be implemented only if there is give in the s
aimed at facilitating the dismantling of controls, o
inflation-producing deficit spending may serve muc
ment domestic saving or aid to cover all an alleged imp

9) A greater emphasis on direct taxes. Direc

difficult to enforce, but they also are the most power
Countries will not learn to enforce such taxes unles
and the mere fact that they are now difficult to levy
for ignoring them and relying heavily on indirect tax
purposes.

10) Technical aid aimed at helping people to go about understanding

what can be done in the context of their own environment. The tendency in
technical aid programmes has been to teach people in poor countries how things
are done in rich countries. Our various arguments above have shown that this
contributes to the alienness that we have found so damaging and to the mis-
allocations which we have found equally harmful to the development effort.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
Bruton: Import-Substitution Strategy 145
REFERENCES

1 . Bruton, H. J., Export Growth and Import Substitution. Research

randum No. 22. (Williamstown, Massachusetts: Williams College,

2.

Journal of Development Studies, Ap

3.

Economic Review, December 1967.

4.

American Economic Review, June 1969

5. Chenery H. B., and A. M. Strout,

Development", American Economic Rev

6. Clark, P. G., Brazilian Import Libe

No. 14. (Williamstown, Massachusetts

7. Clark, P. G. , and R. Weisskoff, Import

Research Memorandum No. 8. (Willia
College, 1967).

8. Dobb, M., On Economic Theory and

9. Felix, D., "Monetarists, Structurali

trializing: A Critical Appraisal", in Inf
edited by W. Baer and I. Kerstenetzk

10. Hirschman, A., The Strategy of E

Yale University Press, 1958).

11. Khan, A. R., "Import Substitution,

Liberalization: A Preliminary Report
Vol. Ill, No 2, Summer 1963.

12. Leibenstein, H., Economic Backwa

York: Wiley, 1960).

13. Lewis, S. R., Jr., Economic Policy

(London: Allen & Unwin, 1969).

14.

An Empirical Stu

15.

ing Country: The Case of Pakistan"

October 1968.

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms
146 The Pakistan Development Review

16. Lewis, W. A., "Economic Development w

Labour", Manchester School of Economics an

17. Little, "The Strategy of Indian Development", National Institute

Economic Review , May 1960.

18. Nurkse, R., Problems of Capital Formation in Underdeveloped Countries.

(New York: Oxford University Press, 1958).

19. Power, J. H., "Import Substitution as an Industrialization Strategy",

Philippine Economic Journal, Second Semester, 1966.

20. Prebisch, R., "The Economic Development of Latin America and Its
Principal Problems", Economic Bulletin for Latin America , February 1962.

21. Reynolds, C. W., Changing Trade Patterns and Trade Policy in Mexico .
Research Memorandum No. 17. (Williamstown, Mass.: Williams College,
1967).

22. Rosenstein-Rodan, P., "Problems of Industrialization of Eastern and

Southeastern Europe", Economic Journal , June-September 1943.

23. Sheahan, J. B., "Imports, Investment and Growth: Colombia Experience

Since 1950" in Development Policy - Theory and Practice edited by
Gustav F. Papanek. (Cambridge: Harvard University Press, 1968).

24. Sheahan J., and S. Clark, The Response of Colombia Exports to Variations
in Effective Exchange Rates . Research Memorandum No. 11. (Wil-
liamstown, Mass.: Williams College, June 1967).

25. Winston, G. C., "A Preliminary Survey of Import Substitution", Pakistan

Development Review , Vol, VII, No. 1, Spring 1967.

26.

sistency in Development Planni

Autumn 1967.

27.

Research Memorandum No. 24. (Willi

1968).

This content downloaded from

111.68.96.40 on Sun, 28 Apr 2024 16:24:40 +00:00
All use subject to https://about.jstor.org/terms