. 11
( 21)


easy ISI strategy is that there is a danger of this stage becoming virtually permanent, so
that tariff protection is not fully removed.18 And that has happened. In India and many Latin
American countries, high levels of protection continued for far longer than can be justi-
fied by the infant industry argument and by appeals to the transitional inefficiencies of new
local firms. Producing behind a highly protective tariff wall, domestic producers gained a
quasi-captive market which resulted in above-normal profits, often with minimal attention
being paid to efficiency, to improvements in technology or training, to product quality, or
to consumer preferences. Industrialists in an ISI sector where tariffs are not progressively
removed may be able to reap long-term economic rents, that is, to earn profits higher than
are necessary for calling forth the current level of production, since the tariff prevents the
full force of foreign competition from being felt by domestic producers. Consumers end up
The initial structural transformation 293
paying prices that are wholly unwarranted, as local industrialists earn unjustified profits in a
less than competitive environment.
When tariff protection is prolonged rather than phased out, this typically is the consequence
of close links forged by entrepreneurial elites in the ISI sector with administrators in
government having responsibility for the industrialization program. Such connections can
diminish the independence of state decision-makers to act from broader social and economic
interests. These links may be quite informal. Industrialists and government officials may
belong to the same clubs. They quite likely operate in the same social circles, meeting at
restaurants, the theatre, social gathering, weddings, and so on. They often are related by
blood or marriage.
The contact between protected industrialists and government bureaucrats can involve
substantial degrees of corruption and bribery. When governments lack relative autonomy
from strong vested interests, they often respond to the specific, private concerns of those
groups with the power, money, influence, and access to government to make their voice
heard. In the process, more general interests of society for more efficient, more technological,
more productive, more equitable, and higher-paying and expanding domestic industry are
sacrificed, as the ability of the majority segment of the economy to voice its interests is
circumscribed by relatively closed and exclusive political processes that respond to powerful
interests first and public concerns later, if at all.
This problem of a captured state is not, however, one inherent to countries which initiate
an easy ISI strategy. It is rather due to the nature of the political process and the privileged
access that some groups have to state decision-making which create a barrier to the desired
reduction of tariff protection on ISI industries. These elites are able to turn state economic
policy in their direction for their own profit at a cost to the domestic economy and to long-run
consumer interests. Whenever the state is to some degree “captured” by special interests
who can create and maintain tariffs, laws, and regulations that allow them to earn a larger
producer surplus and economic rents without becoming more efficient, economic policy will
be distorted in the direction of those elites™ interests at the expense of overall economic
efficiency and social welfare.19
Distortions and interference in state policy by predatory elites for their own special
interests, however, can arise regardless of the particular economic strategy an economy
pursues. It is not a necessary consequence of the easy ISI stage, as the experience of the
East Asian economies, reviewed in the next chapter, vividly demonstrates. Further, the
experiences of South Korea, Taiwan, and other LDCs do not support the conclusion that
permanent tariff protection and inefficient production are necessary outcomes of the ISI
stage of industrialization (Amsden 2001).
The problem of continued protectionism, then, is one of domestic politics and is not due
to the particular economic strategy in force.20 It is purely coincidental that such “captured”
states are most evident during the easy ISI stage of industrialization, but that is only because
ISI is most often the natural and logical first stage of industrialization and of the structural
transformation process required for higher levels of development (see Focus 9.5 on India™s
development trajectory).
In nations where an economic elite dominates or can corrupt the political process, tariff
protection is more likely to be maintained long beyond its justifiable usefulness for infant
industry protection. But it is not the easy ISI strategy that is to blame. It is not a problem of
the state providing new firms with protection from import competition so that transitional
inefficiencies can be addressed. Rather, it is the nature of the political process and its captured
status that lead to the detrimental over-extension of tariff protection, and this is more likely
294 The Process of Economic Development

Much has been written in recent years about India™s economic renaissance. Long known
for its low “Hindu rate of growth,” India™s economy has gained sufficient momentum in
recent decades, to be viewed, along with China, as an economic success story of grand
proportions. How realistic is this view, and what has happened in India?
From 1995 to 2005, average per capita income grew by 4 percent per year, a monu-
mental achievement for a nation of 1.1 billion people. From the period 1965“79 to the
twenty-year period 1980“2000, annual GDP growth doubled to an impressive 5.8 percent
rate “ unheard of in any previous period in India. The 1980s ushered in a decade of debt-
led growth (driven by both a domestic debt build-up and international borrowing) that,
finally, could not be sustained.
While private sector investment was modest in the 1980s, the government boosted
public sector investments in infrastructure and industry, where emphasis was placed on
cheapening inputs to raise the growth rate. These expanded investments were curtailed in
the late 1990s, and state investment fell from 10.3 percent of GDP in the 1980s to roughly
8 percent in the 1990s.
Popular accounts of India™s recent boom rely upon a simplistic story that has some
grains of truth. In 1991, the government, long active in guiding and promoting the economy,
reduced a number of regulations, opening the way for greater market participation, more
foreign trade, and enhanced foreign direct investment. India was nudged in this direc-
tion by a major International Monetary Fund “stabilization” program (these programs are
discussed in Chapter 17) in order to obtain some relief from its unmanageable level of
external debt. In any event, as public sector investment declined in the 1990s, private
sector investment surged as new opportunities opened up. Private sector investment rose
from 4.4 percent of GDP in the 1980s to 6 percent in the 1990“5 period and then to 8.3
percent in the 1995“8 period.
India has become famous for its service-sector-led growth in the 1990s and into the
twenty-first century. What is often conveniently forgotten is that state policy changes
from the 1970s through 1984 led to the prioritization of the information technology (IT)
Particularly favorable promotional policies included the decline in import tariffs, the
reduction in regulations “ including the elimination of industrial licensing, permitting the
entrance of foreign and domestic firms, the marketing of exports in international markets,
the setting up of an array of industrial parks, the building of a specialized communica-
tions infrastructure, and the provision of crucial computer installations. As a result of this
new cooperation between the private and public sector in the IT industry, output growth
reached spectacular levels of 50 percent per year in the 1990s.
These changes also highlight not only the importance of state policy, but that there
needs to be “good” state policy, that is, policies that reward and promote efficiency and
competition. While India cannot be said to have come as far as Taiwan or Korea or other
recent successful developers in balancing the role of the state and the private sector, there
has been some positive movement.
After 1984, the Indian government reduced some of its more ambitious schemes of
income redistribution and populism and adopted a more pro-growth perspective. The
more pro-business stance, however, should not be taken as a retreat of the state in the
economic sphere. India never had a high state capacity, such as has been the case in
Korea. India™s new posture reflects many elements, including the fact that from independ-
ence until the 1980s an industrial or business element had matured, and this new element
was increasingly able to influence the future of the policy structure of India because of its
increased effectiveness in the marketplace.
The initial structural transformation 295
Nonetheless, it must be noted that in the early twenty-first century India™s policy posture
remains statist in nature. Public-owned firms loom large in the economy, import tariffs still
remain significant (on average over 100 percent in 1990, falling to 40 percent by 1999),
foreign investment is held to low levels, and the state inhibits the movement of finan-
cial capital into and out of the economy. It would be an inaccurate statement to assert
that India™s improved economic performance is a demonstration of a major shift toward
economic liberalization.
To some degree India™s improved economic performance is the result of the historical
legacy of positive state intervention. Large public investments in technological educa-
tion and capital-intensive industries from the 1940s through 1964 were to be expected
to have a long gestation and a distant payoff. In the 1950s, the government created five
elite universities that have trained two generations of Indian engineers in the justly world-
famous Indian Institutes of Technology. Because of these institutes and a thriving univer-
sity system, India has the third-largest national scientific and technical capacity in the
world. This cadre of knowledge workers is a very important resource that provides India
with the possibility of great change (Chapter 13 discusses in detail the importance of these
knowledge workers).
India™s ISI policies also generated positive externalities in terms of a growing, well-
trained labor force and more efficient supplier networks.
Still, many commentators now argue that the chief bottleneck facing India is an infrastruc-
ture deficit of daunting proportions. This demonstrates that the slowdown in state invest-
ment in the 1990s, championed by some as a sign that the Indian state was shrinking,
was not necessarily a good developmental decision. Today India lacks adequate roads,
sea ports, airports, water, and electricity. Current plans are for massive state investment
and creative public“private partnerships designed to improve essential infrastructure. The
prime minister has advocated an ambitious effort to invest $300“500 billion spread over the
2007“12 period. This may be unattainable, as India is currently attracting only modest levels
of foreign investment, which will be needed to generate spending of this magnitude. But at
least the government is looking forward and seems to be moving in the right direction.
Sources: Hamm 2007; Kohli 2004

to occur in nations without a history of democratic experience, which is the situation many
less-developed economies face. The barrier of a weak, captured, and predatory state needs
to be overcome by appropriate institutional reform, culminating in the creation of a forward-
looking developmental state, as described in Chapter 7.

Potential gains from the easy ISI stage of industrialization
What are some of the expected gains to countries from initiating the easy ISI stage of indus-
First, given the nature of the methods used to produce the typical easy ISI good, this stage
of industrialization tends to be relatively labor-intensive. As output in this sector grows, easy
ISI industries can provide increasing employment opportunities for an expanding proportion
of the labor force. This will be especially important if the increasing number of migrants
exiting the agricultural sector, attracted to factory jobs by the pull of higher wages in the ISI
sector, is to be absorbed in productive employment.
Second, during the easy ISI phase, the industrial labor force develops both specific and
general human capital skills as a result of “learning-by-doing” as they work with the modern
machines and technology they encounter in the factories. General human capital skills, by
definition, will be at least partially transferable to other enterprises, thus shortening the
296 The Process of Economic Development
lag-time required for those firms to become more productive. This happens as some workers
or managers leave one firm to work in another or to start their own spin-off venture in the
same or a different industry. There is thus an acceleration of the societal learning process that
can accompany the expansion of the easy ISI sector as new production linkages, both back-
ward and forward, are created. Of course, as already discussed in the previous section, such
positive externalities are more likely to appear as infant industry tariffs are lifted with some
time-certainty so that the original easy ISI firms succeed in becoming competitive and shed
their transitional inefficiencies. Then what is learned in these firms will be at least partially
applicable in other parts of the economy.
As the endogenous growth theories of Chapter 8 suggest, such positive technological and
human capital externalities can accelerate the pace of economic expansion, even with the
same level of other productive resources in use. Through their interaction with the existing
pool of knowledge or technology and the learning which accompanies the actual process of
being involved in production, the ability of an economy™s labor force to better utilize existing
knowledge and technology can be improved and the growth and development prospects of
the economy can be enhanced. Further, management, financial, marketing, accounting, and
essential entrepreneurial skills can be acquired and improved during the easy ISI stage as
a consequence of experience in producing and striving for efficiency and market share as
tariffs are reduced. These are skills that are integral to successful and sustained industrializa-
tion over the long term, especially in the absolutely essential post-easy ISI stages of industri-
alization considered in the next chapter.
Third, easy ISI acts as a training ground for entry-level local capitalists who have an
opportunity to develop their own skills in operating profit-oriented and efficiency-focused
enterprises. Over time, with successful easy ISI, the initial disparity between the manage-
ment skills and information levels of domestic managers and capitalists and those of foreign
industrialists can be narrowed. This stage of industrialization for late industrializers, which
all the less-developed nations are, facilitates the establishment, extension, and solidification
of a domestic class of private entrepreneurs who will be essential to the continuation of the
process of development into the future.
Every society has latent entrepreneurs. Prior to industrialization, many will be involved
in trade, finance, and, perhaps, illegal activities where profit-making opportunities are the
greatest. The expansion of the easy ISI sector provides these entrepreneurs with another
opportunity to maximize their earnings. The difference is that the activities of entrepreneurs
in the ISI sector are likely be more socially and economically productive than when the
opportunities for profitable pursuits were more circumscribed in a pre-industrial setting.
Providing creative space for entrepreneurs to be part of a more productive sector of the
economy, and to learn and thrive there, can be one of the more fundamental transformations
for the future growth and efficiency of an economy to be derived from a successful easy ISI
In the easy ISI stage, the rise of the modern capitalist “business ethic” is encouraged and
reinforced. The attention to detail and quality, to financial and accounting costs, to time
schedules, and to contracts that are characteristic of production in more developed nations
have a chance to begin to seep into the thinking and behavior of businesspeople. Where
corruption and cheating of the customer might have been the norm of behavior by sellers
in the past, the progressive elimination of tariff protection under an easy ISI regime can
contribute to an alteration in entrepreneurial behavior that drives them to achieve levels of
efficiency, stability, and responsibility to the consumer that modern industrialization demands
(Gerschenkron 1962: 47“8).
The initial structural transformation 297
Given these potential advantages, easy ISI would seem to be an imperative stage of
economic and social transformation, though it is not by itself sufficient to the continued
prosperity of the development process. Easy ISI is the first step, then, on a journey toward
a more complex process of industrialization and development. It is not, and cannot be, the
final stopping point.

Para-state firms and social capital21
In many instances, private sector development in less-developed economies will require
assistance from state-owned firms, perhaps especially during the initial ISI transformation.
Infrastructure investments such as heating oil and natural gas, electricity, ports, and tele-
communications are all industrial activities that require enormous start-up costs in terms of
physical and financial capital outlays. They are also industries which demand a mastery of
intermediate levels of technology that may be beyond the capacity of local private entrepre-
neurs to provide in the early stages of industrialization.22 The state, however, may be able to
supply the necessary capital resources, the organization of production, and contribute to the
shaping of a cadre of engineers and state-managers capable of providing the infrastructural
base and other key inputs into the industrialization process that are required for the eventual
efficient operation of the private sector of the economy.
There are those who argue strongly against the expansion of state-owned firms, even
in the early ISI stage. It is most commonly alleged that state enterprises are prone to
be less efficient than if they were to be privately operated. It is asserted that the prices
charged by state-owned firms all too often are set too low, that is, below what the unfet-
tered market price would be. State-owned enterprises thus often fail to recover all of their
costs of production or to make a reasonable rate of return on the initial public investment
in them. Any losses incurred by state enterprises are funded out of the national budget,
thus creating a drain on scarce government revenues that cannot then be used for other
While it is true that state enterprises often do operate at a loss, this is not always the case,
even in those instances in which a state enterprise does not recover its full costs, it is not
accurate to claim that the mere existence of an accounting loss incurred by a state enter-
prise implies a social loss from the operation of the para-statal. To the extent that a state
enterprise™s operations promote the production of positive externalities accruing to private
producers in the economy in the form of increased private returns, it is entirely possible that
a subsidized state enterprise selling its output at a price below what unsubsidized private
producers would charge will contribute to reaching the socially optimal level of production
better than if the enterprise were to be in private hands.
The theory of positive externalities is quite unambiguous in asserting that, in the presence
of high transactions costs such as are involved in any large-scale infrastructure investment,
producers of positive externalities must be subsidized if they are to reach the optimal level
of production where marginal social benefits are equated to marginal social costs. Unsub-
sidized producers of positive externalities most certainly will under-produce.23 Mustering
evidence of state enterprises with accounting losses falls short of providing evidence of
their relative inefficiency compared to private firms. Such state enterprises may, in fact, be
contributing quite effectively to social efficiency by producing the socially optimal level
of electricity, water, transportation, communications, or gas by contributing the attendant
positive externalities helping to make the private sector more productive, to increasing
incomes, and to raising employment in the private industrial sector. Is there any evidence to
298 The Process of Economic Development
support the argument supporting state provision of infrastructure to stimulate private sector
In Taiwan in the 1950s, during that nation™s phase of easy ISI, state enterprises of all types
produced well over half of all of Taiwanese industrial output. Para-state activity overshadowed
private firms in “fuels, chemicals, mining and metal working, fertilizer and food processing,
textiles, and utilities” (Wade 1990: 78). In the glass, plastics, steel, and cement industries,
state enterprises initiated production, removing the high fixed-cost investment barrier, and
then, after the new firms were up and running and had reached a level of production sufficient
to reduce per unit costs, these firms were turned over to entrepreneurs in the private sector
to be run for profit.
The para-state solution is, however, only one among many possibilities for providing
essential inputs, especially infrastructure, to an emerging industrial sector. Depending on
the circumstances, joint ventures of ownership and control shared by national entrepreneurs,
foreign owners, and, perhaps, the state in various combinations can substitute for a para-state
firm. In other situations, for example where technology is very expensive and well-trained
domestic management and maintenance personnel scarce, the most reasonable solution for
obtaining the production and positive externalities from needed infrastructure investment
may well best be via the intervention of a wholly-owned multinational firm. In any case, all
successful ISI programmes have incorporated para-statals to one degree or another, and these
enterprises were often catalytic forces in industrial transformation. Para-statals need not be,
by definition, purveyors of inefficiency.

Measuring the success of easy ISI
A successful easy ISI stage of industrialization shifts a country™s production possibilities
frontier (PPF) outward both along the agricultural axis, if agricultural productivity is raised
simultaneously, and along the manufactured goods axis, as shown in Figure 9.3.
The curve PPF1 is the production possibilities frontier of a country prior to the initiation
of easy ISI. Most of the economy™s output comes from the agricultural sector. The country
initially operates at a mix of production like that at A, which lies inside PPF1 as a result
of the misallocation of resources, particularly of labor, that characterizes less-developed,
agriculture-based economies. Often the labor force occupied in that dominant sector has low
or even zero marginal product for at least part of the year.
With the start of easy ISI and the transfer of labor from agriculture to industry and
the introduction of new technology into both agriculture and industry, a new production
possibilities curve, PPF2, emerges that lies outside the former PPF1. The new mix of
production at, say, B, involves a movement toward both the industry axis and toward
the new PPF2 frontier as a result of positive technical efficiency change, as resources
are better utilized, especially labor, which is shifted from low marginal productivity
uses in agriculture to higher productivity occupations in manufacturing. Both the share
of total production and of total employment in manufacturing rise over the transition.24
It is precisely these sorts of transformations that lead to higher levels of income and

Summary and conclusions
Easy ISI is only a first step on the path toward a higher level of industrialization and
development. Let us repeat this so there can be no misunderstanding:
The initial structural transformation 299
Agricultural goods

PPF after easy ISI



PPF before
easy ISI


Manufactured goods

Figure 9.3 Impact of easy ISI on the productive possibilities frontier (PPF).

Easy ISI is only a first step on the path toward a higher level of industrialization and
But it is an essential first step if countries are to move to a higher growth and income
path. It is true that some nations have become stuck at this stage of industrialization because
of internal barriers to transforming the productive and social structures of the nation,
particularly where there are powerful vested interests. It is also the case that for those nations
that have successfully transitioned through easy ISI, this stage of industrialization has been
fundamental in creating the conditions for future progress.
In the next chapter we shall consider the subsequent stages of structural transformation
and industrialization that different economies have followed after the easy ISI stage. We
shall see that there appears to be a bifurcation in possible transitions: one path “ the optimal
path “ seems to lead toward greater opportunities to realize even more of the dynamic
effects of industrialization and structural transformation, as seen in the discussion of
endogenous growth theories; the other path “ the sub-optimal path “ reduces dramatically
the possibilities for continued progress for the economy as a whole, though some sectors
do benefit.
What is not seriously in question, though, is the need for initiating industrialization
if poverty is to have a chance to be overcome, and easy ISI is the natural start of this
300 The Process of Economic Development
Questions and exercises
1 Table 9.1 suggests a strong positive correlation between the rate of growth of the industrial
sector and the rate of growth of total gross domestic product (GDP), but this is shown
only for regions and three countries. In this exercise, you will further consider the rela-
tionship between these growth rates.
Select four less-developed countries other than China, India, or Korea, and record
for each the rate of growth of industry and the rate of growth of GDP for at least two
periods. You can typically find the data in Table 4.1 in the World Development Indi-
cators available online at http://worldbank.org. If this source is not available, check
elsewhere on the World Bank website or another source for the data. Using this data,
draw a graph with the “industry growth rate” on the horizontal axis and the “growth rate
of GDP” on the vertical axis. For each country, show its specific industry growth rate
and its growth rate of GDP as one point on your graph for each of the time periods for
which you have data.
Do you see any relationship between the two variables? Draw a straight line through
the points you have graphed. Does it have a positive or a negative slope? What does that
positive or negative slope tell you? Are the industry growth rate and the GDP growth
rate systematically related to one another in your data? (If you have access to Excel or
some simple regression program, you can input the data in two columns, one for the
growth rate of industry and the other for the growth rate of GDP; use a scatter diagram
to show the data. You can have Excel draw a trend line for you.)
2 This exercise will demonstrate the gains to total national output from a better distribu-
tion of an economy™s labor force to higher productivity activities.
Choose a less-developed country that interests you or one you are assigned. If you
wish, you can use one of the countries for which you collected data in the previous
problem. Find the following:
a the share (i.e. percentage) of total output (GDP) produced in agriculture, industry,
and services for a recent year. You can typically find the data in Table 4.1 in the
World Development Indicators available online at http://worldbank.org. Go to the
“Country at a Glance” section in the middle of the page; select your country to find
the data in the “Structure of the Economy” section.
b Then find the share (i.e. percentage) of the labor force working in agriculture,
industry, and services from Table 2.3 in the same World Development Indicators
online. Choose data for the period closest to the year you selected for your data in
part a of this question. You will have to average the male and female rates if both
are given. If your country does not have data for both GDP and labor shares, you
will need to pick another country.
c Now, calculate the productivity of each 1 percent of the labor force in these three
sectors, i.e., what percent of total GDP is produced by 1 percent of the workers in
agriculture, industry, and services (hint: divide the output share in each sector by its
labor share);
d Is the labor force in this country distributed optimally, that is, could total GDP be
increased by shifting labor from one sector to another? Explain.
e Is there “surplus” labor in agriculture in your country? Or some other sector?
Explain what is meant by “surplus” labor (hint: there is a discussion of surplus
labor in the section of this chapter on the Lewis model).
The initial structural transformation 301
3 How does shifting labor from a low-productivity sector to a higher-productivity sector
affect an economy™s GDP? This exercise will show you how and why that happens.
Utilizing the data you calculated from part c of question 2 above, determine the net
impact on total GDP for your country from transferring out 10 percent of the labor force
in your country™s surplus labor sector and adding that to the workforce in the sector
with the highest per worker productivity (e.g., if your surplus labor sector is agriculture,
reduce the agricultural labor force by 10 percent, say from 45 percent to 35 percent;
then add that 10 percent of workers to the high productivity sector, increasing the labor
force by 10 percent in that sector; for example, in industry, increase the labor force from
17 percent to 27 percent).
You should be able to determine the net gain to GDP from such a movement of
workers as: the loss in output from a decrease in 10 percent of the workforce in the
surplus sector plus the increase in output resulting from an increase in the labor force
by 10 percent in the high productivity sector. The result you get should be a specific,
concrete numerical percentage change in GDP resulting from a negative change in output
in the low-productivity sector and a positive change in output in the high-productivity
sector as a result of the shifting of labor.
4 Table 9.2 in the text provides evidence for a close relation between a decrease in the
share of the labor force engaged in agriculture (the primary sector) and an increase in
the share of the labor force employed in industry (the secondary sector) and services (the
tertiary sector) and a higher level of development for a country.
You are going to draw two graphs. In the first, put the “level of development” on the
vertical axis and the “share of the labor force employed in agriculture” on the horizontal.
Using the same four LDCs for which you collected data in problem 1 above, find the
HDI value (this is the “level of development” measure) and the share of the labor force
employed in agriculture for the most recent year possible, and plot the data for these two
variables (if you have access to Excel or some other spreadsheet program, input the data
there and use it to draw a scatter diagram). You can find the data at the UN Development
Programme website, http://www.undp.org.
On a second graph, again put the “level of development” on the vertical axis but
now put the “share of the labor force employed in industry” on the horizontal. Find and
plot this data for your selected countries using the same source as above (of course you
already have the HDI data from the first graph).
In both graphs, draw a straight line through the points which best fits the data (if you
are using Excel, have the program draw a trend line for you).
What relationship do you find in each graph as determined by the slope (positive or
negative) of the “regression” line you have drawn? In your own words, explain what
“story” the data tell you about the relationship between the share of labor at work in
agriculture and industry and the level of development of the four countries as measured
by their HDI values.
5 Carefully explain the economic rationale for imposing an infant industry tariff to
protect new producers from import competition. If you can use a graph to illustrate
your argument, please do so. Why are these called “infant industry” tariffs and not just
6 Carefully explain what is meant by transitional inefficiencies? A few possible sources of
transitional inefficiencies were noted in the text. Can you list two or three other sources
of such inefficiencies in less-developed economies? Why are these called “transitional”
inefficiencies and not just “inefficiencies”?
302 The Process of Economic Development
7 Discuss alternative ways other than tariffs that governments in a less-developed country
might encourage and support easy ISI firms to produce in competition with imports.
Are there things government must do to help the private sector be more competitive if
industrialization is to be successful? Explain.
8 As noted in the text, many economists worry about the wisdom of an infant industry
tariff strategy as a means to promote industrialization in LDCs. The following exercise
gives you a chance to consider the effect of such tariffs.


45 Sw + tariff

Quantity of sheets
168,000 186,667 280,000 466,667

a Calculate the consumer surplus prior to imposing an infant industry tariff. What,
exactly, does this consumer surplus value measure?
b How much was actually spent by consumers on sheets before the tariff? After the
c Calculate the consumer surplus after the infant industry tariff is imposed. (How
much is the tariff?)
d What is the total loss in consumer surplus due to the imposition of the infant industry
e What are the government™s tariff revenues after imposing the tariff?
f What is the producer surplus after the imposition of the tariff?
g How much deadweight loss is there as a result of the imposition of the infant
industry tariff?
h Add the totals you obtained from d, e and f. How does this total compare to what
you calculated for part c? Explain what you have found and what it means.
i If domestic producers succeed in becoming as efficient as foreign producers by
overcoming their transitional inefficiencies, what would be the value of consumers™
surplus, assuming the infant industry tariff is removed? Show on the graph where
this consumer surplus is.
j How can this infant industry tariff be justified?
9 Explain how the accounting profits (or losses) of an unsubsidized public enterprise,
the output of which creates positive externalities for other firms, are likely to overstate
(understate) the true level of profits (losses) of that firm. Is it possible for a para-state firm
which creates positive externalities to private sector enterprises to have an accounting
loss but still be “socially profitable”? Explain.
The initial structural transformation 303
10 Focus 9.1 shows the share of primary exports as a percentage of total exports for a
limited range of countries. Choose three low-income less-developed nations not listed
in the table. Does the primary product export pattern dominate in each of these as well?
What potential problems do countries face when primary products are a large part of
their total exports? Why is it “better” to have manufactured goods exports as a larger
share of total exports?

1 The industrial sector includes the manufacturing sector. Within the industrial sector, but not classi-
fied as manufacturing industries, are electrical power generation, communications, gas and water
generation, mining, transportation, and other non-manufacturing enterprises.
2 It is perhaps worth remembering the insight of an early development economist, Alexander
Gerschenkron, that there exists the possibility of “substitution” in the way any particular country
becomes developed. Becoming developed is not a process of simply copying what other successful
countries have done before. There are patterns, of course or there would be no purpose in teaching
about development, but each country must forge its own particular path within the patterns of struc-
tural change that have been identified as being crucial.
Nobel Prize-winning economist Simon Kuznets wrote in this regard: “there is a connection
between the high rate of growth associated with modern economic development and a variety of
structural changes, not only economic but also social; not only in institutions but also in ideology.
This does not mean that all the historically associated shifts in economic and social structure and
ideology are requirements, and that none of them could be avoided or substituted for. It does mean
that some structural changes, not only in economic but also in social institutions and beliefs, are
required, without which modern economic growth would be impossible” (Kuznets 1971: 348).
Further, as the endogenous growth analysis examined in Chapter 8 suggests, it is not just struc-
tural change that is the key to successful growth. Efforts at structural change alone “ for example,
the transfer of labor from agriculture to industry “ that lack the required changes in the quality
of inputs to production via human capital accumulation, funding for R&D and technology and
new knowledge acquisition, and application and additions of more productive physical capital are
likely to be substantially less effective. Further, as we shall consider in subsequent chapters, there
are fundamental macroeconomic concerns and constraints that must also be a part of a successful
development strategy.
3 Industrialization involves the transfer of what is often called surplus labor from agriculture to the
emerging industrial, factory system, as is reviewed in the section on the Lewis model below. If that
movement of labor is not to result in a decrease in basic food production, in higher food prices,
and for the need for expanding food imports, domestic agricultural production must become more
efficient at the same time, as already noted.
One important additional and non-economic reason to transform agriculture is to remove a
potential political obstacle, that is, an Ayresian ceremonial barrier, to further industrialization and
development, particularly during the export-substitution stage of industrialization discussed in
Chapter 10. If large and politically powerful landowners derive a portion of their wealth from
their primary product exports, they are more likely to oppose the evolution of the industrialization
process into diversified exporting, particularly when the aim of such a transformation is to change
the export profile of the nation by replacing agricultural and other primary exports with the manu-
factured outputs of an emerging domestic industrial sector.
4 It is typically the case that the benefits of primary product specialization, that is, of an economic
structure dominated by agricultural and natural resource extraction, tend to accrue to only a rela-
tively small elite in the country. Industrialization opens the door, then, to a wider distribution of the
gains of production and an increase in total social welfare not as easily attained if a nation remains
in the primary product producer category (Chenery 1979: 35). Thus, industrialization increases
the opportunities for greater income dispersion, as well as higher incomes per capita, that can
contribute to improvements along the HDI dimension as well, particularly as universal primary
and secondary education become the norm. To the extent that income distribution is improved as
a consequence of such an evolution in the productive structure and in the accumulation of human
304 The Process of Economic Development
capital, and to the extent that an improved income distribution contributes to a higher level of
development, this transformation is “growth-enhancing.”
5 This can be demonstrated relatively easily by determining the productivity of 1 percent of the labor
force in producing GDP in agriculture, industry, and services. In many less-developed economies,
the productivity of the agricultural sector is less than 1 and substantially less than the productivity
of labor in industry or services. Shifting workers from lower-productivity agriculture to higher-
productivity industry will thus increase total GDP. exercise 2 at the end of the chapter asks you to
show this.
6 It is “easy” ISI in two senses. First, since the demand for a particular manufactured good already is
known from the quantity of imports, the potential size of the market for that good also is known; all
local producers need do is produce to service that demand. And second, this stage is “easy” because
the technology used in production is relatively simple; often standardized, off-the-shelf machinery
can be purchased on the world market. The term “primary ISI” recognizes this stage as a first phase
of ISI, not its endpoint. The term “horizontal” ISI recognizes that the first stage of ISI is taking
place within particular industries; during this stage, industrialization does not extend backward into
the other, supplier industries, which is called vertical ISI, and which involves a deepening of the
industrialization process.
7 As one of the poorest regions among the LDCs, Sub-Saharan Africa™s exports “ more than 80
percent “ remain dominated by unprocessed primary products; Wood and Mayer 2001: 376.
8 This may be a further reason why too much income and wealth inequality can limit the pace of
economic growth, as Rodrik™s study suggested in Chapter 8. Excessive inequality limits the size of
the domestic market, thus impeding the pace of domestic industrialization and growth and develop-
ment by restricting total demand. Efficient levels of production cannot be reached if the market is
“too small.”
9 Bruton (1989) prefers the term “infant economy” protection, which perhaps better conveys that, for
the less-developed economies, the transformations in production, organization, education, public
policy, and so on that are required to overcome the transitional inefficiencies go far beyond any
particular industry.
10 Evenson and Westphal (1995: 2284“5) argue that protection is only a second-best policy, however.
The first-best policy would be to have efficiently functioning capital and financial markets, as
discussed in Focus 9.2, such that protection would be unnecessary. However, in poor nations, the
option of forgoing tariffs may not be open. It is not just because waiting for the maturation of finan-
cial markets may be costly in terms of long-run growth which is sacrificed. It is also because, for
poor countries lacking government revenues, tariff revenues obtained from infant industry protec-
tion contribute a share of total government revenues. And given, as Evenson and Westphal note,
that no program of protection of infant industry which has lacked needed investments in techno-
logical capacity has been successful, those tariff revenues provide a potential source of funding for
precisely the technological development required, from research and development to the training
of human capital, if they are put to good use. These issues are considered in detail in Chapters 12
and 13.
11 This quantity could be zero if the domestic supply curve begins at a reservation price, or minimum
supply price, which is greater than PF.
Adding a fixed tariff amount to each unit of a good imported, say £3 per unit, is to impose a specific
tariff. Alternatively, if a tariff is imposed as a percentage of the value of the imported good, say, a
30 percent tax, then this is an ad valorem tariff. In this example, with a fixed international price,
either an ad valorem or specific tax will have the same effect of causing a parallel shift upward of
the world supply curve, SW.
However, if the world supply curve were to be drawn as upward sloping, then an ad valorem
tax would result in a non-parallel upward shift of the world supply curve for the country imposing
the tariff. As the price of the imported good rose, the gap between the original world supply curve
and the world supply curve plus the ad valorem tariff would widen, since the price to which the
percentage tariff was being applied would be greater.
13 Producer surplus is the extra revenue domestic firms receive above and beyond what is necessary
for them to be willing to supply a particular level of output. It is equal to the area above the domestic
supply curve and below the prevailing market price. In this example, total producer surplus at price
PT is equal to the triangle PTP0b. However, there was producer surplus even when there was free
trade at price PF equal to the triangle P0aPF. Therefore the additional producer surplus from the
The initial structural transformation 305
infant industry tariff is the difference between the two areas, i.e. PT P0b’ P0aPF which is the area
14 The deadweight loss equals the loss of consumer surplus minus the increase in producer surplus minus
the increase in government tariff revenues. Graphically it is: PFfdPT ’ PFabPT ’ bced = abc + def.
15 Prior to the imposition of the infant industry tariff, domestic firms produced output C0 at price PF.
Domestic producer surplus was equal to area P0aPF with domestic supply curve SD, the position
of which reflected the transitional inefficiencies of production for domestic enterprises. With the
elimination of the transitional inefficiencies, shown by the shifting of the domestic supply curve to
SD2, the new producer surplus, assuming a price of PF, would be triangle PEfPE.
16 Most countries levy some minimum tariff on a large array of imports for reasons other than infant
industry protection. Such tariffs, particularly when they are uniform (say, 4 percent or 6 percent)
and do not discriminate against any particular good, are often assessed as a means to raise revenues
to help cover the costs of customs and other border operations, such as immigration. Thus, with the
end of ISI protection, tariffs do not necessarily, and most likely will not, decline to zero.
17 There may be some exceptions where tariff protection is not fully removed, for example, in the
case of certain industries deemed to be critical for national security. However, even in such indus-
tries (motor vehicle production may be one such industry), there are likely to be better means to
guarantee the survival of firms when full tariff protection is withdrawn, e.g., low-cost loans and
government procurement programmes, that help to push enterprises to be more competitive and
18 For a critique of ISI policies in general, see Balassa 1982.
19 This effort to capture economic rents is referred to as DUP or “directly unproductive profit-seeking”
activity in the economics literature. In the process of attempting to retain tariff protection, ISI firms
expend resources for lobbying and perhaps graft for dishonest politicians and other public officials
to convince them to maintain high tariffs or other favorable treatment. Such expenditures are an
unproductive uses of society™s resources, as they do not contribute to greater output or efficiency,
thus increasing the net loss of a tariff to society (see Colander 1984). For a recent analysis of state
structures, see Evans (1995: Chapter 7), who describes what we call a “captured” state as a “predatory”
state. Perhaps Evans™s language is even more evocative of what occurs when the state is dominated
by special interests.
20 One need only think back to David Ricardo™s analysis of the advantages of free trade and of the
struggles over the Corn Laws in England discussed in Chapter 4 to realize that the search for
economic rents by those with privileged access to political power is neither new nor confined to
countries making use of ISI development strategies. Large english landowners struggled hard to
keep european grains from entering British shores through the implementation of restrictive trade
measures which kept their “rents,” or incomes, artificially and unnecessarily high, at the expense of
others, particularly low-income British consumers.
21 Para-state firms, or para-statals, are government-owned and -operated enterprises.
22 The standard market failure argument for government provision of such goods and services applies
here. Infrastructure tends to generate substantial positive externalities that cannot be captured by
the provider of the service or good and often have substantial public goods characteristics.
State production tends to lower the cost of essential inputs, like electricity, to third-party private
industrialists, thus increasing their profits without any private investment. Further, the mere exist-
ence of sufficient electrical generation capacity, as one example, creates opportunities for new
production to emerge.
In other words, the social profitability of infrastructural investment by government will be larger
than its accounting profitability, and this is why state firms often seem to be unprofitable. That is
true only when looking at the revenues and costs of the firm using a standard accounting approach.
However, if an economic accounting approach is taken then the additional revenues of private
firms using the state-provided input that can be imputed to the existence of the infrastructure must
be added to the actual revenues received by the state firm for its services. Taxes on private profits
are one means to capture at least part of the third-party benefits that result from state-generated
inputs, such as electricity, water, roads, ports, telephone and other communications services, and so
on. Taxing to pay for the provision of such public goods and to cover accounting losses of the state
easily can be justified by the theories of positive externalities and public goods.
23 Unsubsidized private producers of positive externalities have no way of appropriating any of the
additional profits that their production process creates for other firms. Concerned only about their
306 The Process of Economic Development
own private benefits of production, which, when there are positive externalities, are less than the
social benefits of production, the private level of (unsubsidized) production will be less than the
social optimum since the marginal private benefit curve intersects the marginal social cost curve
at a lower level of output than the marginal social benefit curve intersects the marginal social cost
curve. Thus, the private prices of firms affected by the positive externality will be higher than the
socially optimum price and their output levels will be lower than would be desirable because of
under-provision of the positive externality by private firms.
24 For those who argue that ISI breeds inefficiencies, the issue still remains whether the level of
production with ISI, even if taking place inside PPF2, results in a higher growth path than the mix
of production associated with the original PPF1 with the output mix at A. It is not transitional, static
inefficiency of ISI versus an ideal growth path that is crucial but rather the impact on growth and
developed associated with pursuing the ISI strategy versus staying on the status quo path. Of course
continued ISI inefficiency is not desirable but that is contained in the concept of “transitional” inef-
ficiencies, which can be reduced as tariff protection (or subsidization) is removed, as was argued

Amsden, Alice. 2001. The Rise of “the Rest”: Challenges to the West from Late-Industrializing
Economies. Oxford: Oxford University Press.
Balassa, Bela. 1982. Development Strategies in Semi-Industrializing Economies. Baltimore, MD: The
Johns Hopkins University Press.
Blair, Calvin. 1964. “Nacional Financiera,” pp. 193“238 in Raymond Vernon (ed.), Public Policy and
Private Enterprise in Mexico. Cambridge, MA: Harvard University Press.
Bruton, Harry. 1989. “Import Substitution,” Chapter 30 in Hollis Chenery and T.N. Srinivasan (eds.),
Handbook of Development Economics, vol. ii. Amsterdam: North Holland Publishers.
Chenery, Hollis. 1979. Structural Change and Development Policy. New York: Oxford University
Colander, David C. (ed.). 1984. Neoclassical Political Economy: The Analysis of Rent-Seeking and
DUP Activities. Cambridge, MA: Ballinger.
Evans, Peter. 1995. Embedded Autonomy. Princeton: Princeton University Press.
Evenson, Robert E. and Larry E. Westphal. 1995. “Technological Change and Technology Strategy,”
Chapter 37 in Jere Behrman and T.N. Srinivasan (eds.), Handbook of Development Economics, vol.
iiia. Amsterdam: Elsevier Science.
Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge,
MA: Harvard University Press.
Hamm, Steve. 2007. “The Trouble with India,” Business Week (March 19).
Higgins, Benjamin and Jean Downing Higgins. 1979. Economic Development of a Small Planet. New
York: W.W. Norton & Co.
Kohli, Atul. 2004. State-Directed Development. Cambridge: Cambridge University Press.
Krieckhaus, Jonathan. 2002. “Reconceptualizing the Developmental State: Public Savings and
Economic Growth,” World Development 30 (October): 1697“712.
Kuznets, Simon. 1971. The Economic Growth of Nations. Cambridge, MA: Harvard University Press.
Lewis, W. Arthur. 1954. “Economic Development with Unlimited Supplies of Labour,” Manchester
School of Economic and Social Studies 22 (May 1954): 139“91.
Stiglitz, Joseph. 1992. “Alternative Tactics and Strategies for Economic Development,” in A.K. Dutt and
Kenneth P. Jameson (eds.), New Directions in Development Economics. Notre Dame, IN: University
of Notre Dame Press.
Syrquin, Moshe. 1988. “Patterns of Structural Change,” Chapter 7 in Hollis B. Chenery and T.N.
Srinivasan (eds.), Handbook of Development Economics, vol. i. Amsterdam: North Holland
UNDP (United Nations Development Programme). 1995. Human Development Report 1995. Oxford:
Oxford University Press.
The initial structural transformation 307
””. 2001. Human Development Report 2001. Oxford: Oxford University Press.
Wade, Robert. 1990. Governing the Market: Economic Theory and the Role of Government in East
Asian Industrialization. Princeton: Princeton University Press.
Wood, Adrian and Jörg Mayer. 2001. “Africa™s Export Structure in Comparative Perspective,”
Cambridge Journal of Economics 25 (May): 369“94.
World Bank. 1993. World Development Report 1993. Oxford: Oxford University Press.
””. 1995. World Development Report 1995. Oxford: Oxford University Press.
””. 2000. World Development Report 2000/2001. Oxford: Oxford University Press.
””. 2001. World Development Indicators. Washington, D.C.: World Bank.
””. 2007. World Development Indicators. Washington, D.C.: World Bank.
10 Strategy switching and industrial

after reading and studying this chapter, you should better understand:
• the limits to easy import substitution as the motor of industrialization and growth
and development;
• the importance of the proper sequencing of phases of industrialization and timely
strategy switches to successful development programs;
• the importance of export substitution as a fundamental stage in industrialization;
• the costs associated with premature difficult import substitution;
• the importance of trade as a promoter of increased productivity and of more effi-
cient technology use;
• the importance of public policy to augmenting resource endowments appropriate
for future progress;
• the need to search for and promote dynamic comparative advantage and the role
of “contests” in achieving this objective;
• the role of appropriate institutions in supporting warranted strategy switches.

Continuing structural change
In the previous chapter, we saw that the initial impulse toward a higher level of development
comes with the movement of labor from agriculture toward industrial production. This is
typically achieved with easy import substitution industrialization (ISI), as relatively simple,
non-durable manufactured goods are produced for the domestic market by replacing at least
some of the imports of these goods. This shift in the productive structure and of labor usage
via import substitution is common to virtually all successful, and to many unsuccessful,
development experiences.
easy ISI is only the first step along a complex path of structural change. easy ISI makes
further progress possible but without guaranteeing success. No single development strategy
is likely to be sufficient over time if economic growth and progress are to be sustained. Deci-
sion-makers, especially those in government, need to be prepared to make changes in the
prevailing strategy when it no longer provides the base for continued growth. The marks of
good policy-making and of relatively long and stable periods of sustained expansion are the
ability to recognize the need for and to quickly and effectively make what we call strategy
Strategy switching and industrial transformation 309
Industrial sequencing: beyond easy ISI
The stage of easy import substitution industrialization (ISI) examined in Chapter 9 has a
natural limit as a contributor to sustained growth and development. When easy ISI begins,
there is typically a fairly substantial increase in economic growth and the level of GDP
per person. This occurs as a result of the shift of low-productivity labor from agriculture
to higher-productivity activities in industry and services. However, once all the potentially
viable non-durable consumer good imports have been replaced by domestic production, then
further industrial growth in the domestic market will be limited to population growth plus
any aggregate income expansion and changes in tastes and preferences.1
The pace of economic growth will slow as the ability of the manufacturing sector to lead
the growth of the rest of the economy is reduced as the opportunities for shifting labor from
agriculture to industry are exhausted. In fact, diminishing returns to the easy ISI strategy will
be apparent long before all import substitution possibilities for non-durable consumption
goods have been realized.
Countries thus face a dilemma. What can be done to ensure continuing economic growth
and the attainment of higher levels of development? What measures might facilitate strategy
switches in the design of a country™s economic strategy to further the desired structural trans-

The changing composition of imports
Complicating matters, during the easy ISI stage there is a change in the composition of
imports. Non-durable consumer imports as a percentage of total imports obviously decline
as these are replaced by domestic production via the easy ISI strategy. In their place, the new
ISI industries begin to import more expensive and more complex manufactured goods. As a
consequence there will be a rising share of total imports that are intermediate inputs, such as,
needles, bobbins, dyes, and thread for, say, clothing production, and physical capital goods,
such as sewing machines, destined for these new firms in the easy ISI sector.
To the extent that easy ISI is fairly generalized and not limited to a restricted range of
goods, the percentage of total imports accounted for by non-durable consumer goods will
continue to shrink as easy ISI proceeds.2 Intermediate imports for the easy ISI industries and
durable consumer goods (like motor cars, refrigerators, and computers) will loom larger as
a share of total imports. Of course, these imports often are more expensive than the former
non-durable consumer imports.
It was at one time thought that an easy ISI strategy could lessen the dependence of less-
developed countries on the world market. By reducing the demand for non-durable consumer
imports by replacing them with domestic production, it was thought that countries would
produce more of the goods they consumed, which was partly true, and hence would be
more self-reliant, which turned out to not always be true. easy ISI replaced the demand for
imported consumer non-durables with the need to now import many of the inputs for the
expanding ISI firms.
The urgency to secure sufficient export earnings thus remained as compelling after
starting domestic industrialization as before, perhaps even more so, if the inputs for the
ISI industries were to be imported. Dependence on the external market simply changed its
form. Certain intermediate imports became indispensable to the domestic industrial produc-
tion process, to economic growth, and to continued employment within the expanding easy
ISI sector.
310 The Process of Economic Development
In an important sense, then, dependence on the world market is intensified after initiating
easy ISI, even though the overall import coefficient declined for most countries at first.3
However, the composition of imports changed during easy ISI. Imported machinery and
transport equipment and other capital goods increased during the easy ISI phase of expan-
sion, as shown in Table 10.1. The shift toward intermediate and capital goods imports was
most pronounced in the regions where industrialization has been most rapid and longest-
lived, particularly East Asia and Latin America. It was this “import imperative” that ulti-
mately forced economies to make changes to their development strategies as the easy ISI
state of industrialization begins to reach its natural limits of expansion.
We have not shown the composition of imports for more recent years, since for many of
the economies of East Asia and Latin America the easy ISI stage of industrialization had
been completed by the 1980s. Further, the change in imports ultimately is less important
over the long run than the change in the mix of exports for the development process. For
countries in Africa, however, when they do begin more rapid industrialization, the prob-
lems of the import mix will show itself in the foreign exchange shortages discussed in the
next section that will, at some point, require a strategy switch to resolve. And that is the
importance of looking at the changing composition of imports, for the pressure it puts on
the existing development strategy.

Foreign exchange shortages
Most countries that initiate easy ISI find that recurrent balance of payments problems
threaten economic stability at some point.4 Very simply what this means is that earning the
foreign exchange needed to pay for the imports of manufactured inputs to keep the easy ISI
industries operating and the economy growing becomes a constant struggle. This is because
the bulk of foreign exchange earnings continue to be derived from the same limited array of
primary product exports the country has long been selling to the international market, often
since colonial days, even though the economy is becoming more industrialized internally.
easy ISI does not relieve this constraint on the balance of payments. easy ISI changes
what is produced internally and makes an economy more industrial appearing, that is true.
However, what is exported does not change much during this stage in most economies. They
continue to export to world markets the same primary products that were being exported
before industrialization began.

Table 10.1 Composition of imports (as a percentage of total imports)
Primary goodsa Machinery and Other manufactured
transport equipment goods

1970 1992 1970 1992 1970 1992

East Asia and Pacific 23 15 33 39 37 36
Latin America and Caribbean 18 16 35 40 36 35
Middle East and North Africa 26 21 32 35 39 39
South Asia 33 21 24 31 39
Sub-Saharan Africa 15 16 38 39 42 37
32 16 25 35 33 41
High-income economies

Source: World Bank 1994: 189.
a Includes food but excludes fuels, so components do not sum to 100 percent.
Strategy switching and industrial transformation 311
Whether because of instability in commodity prices or declining terms of trade or as
a consequence of inefficiencies in primary product production or a combination of these
effects, the income derived from primary exports tends to fall short of what is needed to
pay for the imported inputs required by the easy ISI industries and to meet the demand for
imported durable consumer goods. The lack of dynamism in traditional primary product
export markets limits the growth of foreign exchange earnings at the same time the need for
such income to finance the purchase of industrial inputs becomes ever more pressing.
We can express this foreign exchange shortage, when export income (X) is insufficient to
pay for the desired imports (M), by the inequality, M > X. Simplifying greatly, the gap between
import expenditures and export earnings creates the need to borrow foreign exchange or take
other steps to finance the expenditure excess. In fact, import spending cannot exceed export
income without the borrowing of foreign exchange to pay for the excess spending.5 We thus
say there is a foreign exchange shortage that must be resolved if an economy is to have
import expenditures, M, greater than export earnings, X.
A severe imbalance between export income and import expenditures thus requires external
borrowing of foreign currency from foreign banks or economies. If the magnitudes of needed
financing are large and persistent, they cannot be sustained by countries forever, just as indi-
viduals or families cannot spend more than their income forever by adding without limit to
their debt. Countries are no different. There is a limit to how much debt is reasonable.6
When external debt becomes unsustainable or economies find that it has become difficult
to borrow foreign exchange, then what was a foreign exchange shortage becomes a foreign
exchange crisis. A foreign exchange crisis forces economies to achieve a balance between
export earnings and import spending when it is no longer possible to borrow. More than
likely it will be necessary to actually have X > M, that is, to earn more foreign exchange
than is spent, so as to be able to make payments on the debt of foreign exchange already

Resolving a foreign exchange crisis via a strategy switch
How does an economy go from having X < M to a situation where X > M? How can a foreign
exchange crisis be overcome over the longer term?
There are two strategies that have been followed in an attempt to ease the foreign exchange
constraint that results from the easy import substitution stage of industrialization: easy export
substitution or difficult ISI. each takes an economy in a different direction in terms of its
industrial structure, export and import patterns, and the impact on the balance of payments.
Further, each strategy has distinctive effects on long-term economic growth and human
development possibilities.
From examining the historical experience, there would seem to be a “superior” and a
“sub-optimal” strategy switch for ordering industrial growth and for furthering the structural
Ultimately, though, the issue is not one of choosing between import substitution or
exporting as overall strategies, as neoliberal economists often frame the issue. The policy
choice is rather between the sequencing of import and export substitution stages and the
nature of what is exported and at what stage in the cycle of industrialization. Proper indus-
trial “strategy sequencing” involves the appropriate combination of exporting and ISI; it is
not a choice between one or the other. No less-developed country has successfully made the
transition to a higher stage of economic growth and development without passing through
both import substitution industrialization and the stage of exporting manufactures to the
312 The Process of Economic Development
world market. Industrialization and higher levels of development begin with easy ISI, but
they must not end there.

Easy export substitution industrialization: the optimal strategy
switch after easy ISI
Assuming infant industry tariffs have been progressively eliminated during the easy ISI
phase, as is indispensable to future economic progress (remember our discussion at the end
of Chapter 9), then imports will be able to re-enter the domestic market. Domestic producers
will need to be competitive with these goods, or consumers will substitute toward imports
and away from less desirable domestic goods as tariffs fall. Those easy ISI firms that have
learned to compete with imported goods somewhere on the price-quality continuum will
not only be able to survive the competition; they will then have a further option beyond the
domestic market for expanding their production, sales, and profits. They now can begin to
export the same non-durable ISI consumer goods to the international market.
The price-quality spectrum permits substantial substitution possibilities from “lower quali-
ty-lower price” to “higher quality-higher price” combinations compared to the average world
price and average quality of any particular good. For example, some of the non-durable
consumer goods China exports may be of slightly lower quality than comparable products
from other countries. If those lower-quality Chinese commodities are to be competitive on
the international market, they will also need to be priced lower. Goods of equal quality can be
priced at the average international price and still be competitive. There is thus not one single
price for a good, such as plastic outdoor chairs. There can be a variety of price and quality
combinations that can be competitive on the international market.
This is why it is so important that infant industry tariffs be but a temporary means to allow
domestic producers to overcome transitional inefficiencies common to new enterprises. At
some point, and sooner rather than later is best, these tariffs have to be eliminated. It is
this fundamental policy step that provides the “stick” of competition from imported goods
that pushes domestic easy ISI firms to learn to be efficient or face the possibility of being
eliminated from the market. Governments can help firms become more competitive and
able to export beyond reducing tariffs. Besides the “stick” of competition, they can provide
“carrots” that help to reduce costs via: international marketing assistance; an exchange rate
that is not artificially over-valued and is, perhaps, mildly under-valued; low but positive,
real interest rates for productive investment (“financial repression”); education subsidies;
and so on.8 The point is that for those easy ISI firms able to meet foreign competition, their
potential market extends far beyond domestic consumers to regional and world markets. For
smaller nations with relatively limited domestic markets and few firms in each industry, the
ability to compete with foreign firms may require the early intervention of government to
assist in pushing production into export markets so that economies of scale and technological
maturity can be realized more rapidly than would be possible from depending on domestic
market size alone.9
The advantages for countries entering export markets are many, and not solely and most
obviously for the individual firms which expect to earn higher profits. From the national or
social viewpoint, an expansion of total exports by selling easy ISI manufactured goods in
international markets provides for the possibility of financing a higher quantity of imported
goods, a faster pace of economic growth and development, higher levels of income, and
more employment for rural migrants entering into the relatively labor-intensive easy ISI
industrial sector. The exports of non-durable manufactured goods contribute to the foreign
Strategy switching and industrial transformation 313
earnings needed to import the necessary intermediate and capital inputs required by the ISI
industry itself, thus helping to self-finance the export expansion.
Further, and of great importance, to the extent that the exports of ISI firms begin to replace
and substitute for traditional primary exports, the economy™s export structure will include
a greater share of manufactured good exports. This can reduce the pressures arising from
the Prebisch-Singer declining terms of trade dilemma (considered in Chapter 6) that can
contribute to export earning problems and to recurring foreign exchange problems.10 To the
extent that manufactured and other non-traditional exports increase in significance compared
to primary product exports, then we can say that export substitution is taking place. The
significance of primary exports declines in both relative and absolute terms within the export
profile with easy export substitution.
In other words, for an economy pursuing easy export substitution industrialization, manu-
factured exports are more than just an addition on top of the primary product export base.
Over time, easy ISI manufactured exports take the place of primary product exports, hence
the name, easy export substitution. Old exports (primary goods) are replaced by new exports
(secondary goods).
Gustav Ranis (1981) has suggested that export substitution is the logical next stage of
evolution of the industrialization strategy after easy ISI. It is this stage which was followed
by the East Asian nations of Korea, Taiwan, Hong Kong, Japan, and other HPAEs (“high
performance Asian economies”). And it is one that economic policy-makers in nations still
in the easy ISI phase of transformation, or taking steps to deepen ISI, should be planning to
implement in the not-too-distant future. Sachs (2005: 18“20) refers to this as the “develop-
ment ladder,” a natural progression in production which China is now climbing, too.
exporting simple consumer non-durables allows less-developed countries to penetrate the
international market at a low level in the so-called “life cycle” of manufactured good exports.
It is a natural, evolutionary niche in that production tends to be labor-intensive and hence
relatively low-cost because of low wages and the continuing labor surplus in agriculture in
the domestic economy. Further, by continuing the expansion of the original easy ISI firms
into world markets, the technology is still relatively low-level and international levels of
efficiency can be more easily maintained.

The gains from easy export substitution industrialization
During the easy export substitution stage, labor-intensive production methods continue to be
used, since it is simple non-durable consumer goods that are being produced for both domestic
and, now, for foreign consumption. Easy export substitution deepens the industrialization
process and allows for the continued growth of employment in the domestic manufacturing
sector. Rising exports can help to maintain a higher rate of economic growth for a longer
period of time.11
By increasing total production to meet domestic and export demand, scale economies in
production may be more easily attained so that per unit costs of production continue to fall
with higher levels of output. Of particular importance too will be the management, finan-
cial, marketing, technological, and other essential capacities that can be learned from oper-
ating successfully in the international marketplace. Such skills often are transferable to other
domestic industries as spillover effects multiply and as the nation™s pool of domestic talent is
enlarged and extended to new arenas. As will be remembered from Chapter 8, these external
effects can help an economy not only reach a higher level of income, but also to maintain
high growth rates over the longer term.
314 The Process of Economic Development
It is the ability to export manufactured goods, however, not the exporting per se that
provides the benefits to the economy, because the ability to export secondary goods requires
a level of world-class efficiency having been reached somewhere on the price-quality
continuum. The ability to export these goods is a signal that domestic firms have become
efficient producers, and it is efficiency in production and the ability to use knowledge effec-
tively that marks successful development over the long haul.
This prospect of backward and forward linkages in new areas of production emerging
through a process of entrepreneurial-deepening is another of the potential benefits of moving
into the easy export substitution phase of industrialization. This stage provides continuity
in the industrialization process and helps to augment the training of the domestic entrepre-
neurial and professional class, as well as upgrading the skills of the labor force as they work
and learn on the job. Local entrepreneurs are allowed to flourish in the international market-
place and possibly to spin off new domestic industries related to the original easy ISI firms.
Easy export substitution industrialization thus permits the local entrepreneurial class to
come to greater maturity by being forced to remain competitive on price and quality and
via the continuous upgrading of technological skills and training that maintain productive
efficiency. It is precisely these sorts of positive externalities that endogenous growth theories
envisage as fundamental to sustained growth over time, and it seems that the capacity to
export manufactured goods helps to endogenize such behavior within the domestic economy
by imbedding the search for efficiency within the productive culture of the economy.
In a study of thirty developing economies over the period 1970“82, Sebastian Edwards
(1992) confirmed that trade and openness to the international economy are important not
solely because exports are a contributor to growth, but because, combined with the appro-
priate human capital base and with supporting government policies, openness is a signifi-
cant transmission mechanism for technological learning by domestic firms operating in the
global economy.
Foreign capital and foreign technology may have a role to play in this stage of the indus-
trial transformation process, but at least in the East Asian economies, foreign interests and
foreign capital were subordinated to local interests, local capital, and local entrepreneurs
(see Focus 10.1). We shall see that this was not the case in Latin America and other regions.
It has become an article of faith in recent years among many economists that an export
orientation, as opposed to an inward-oriented ISI policy, is essential for economic success in
less-developed nations. exports are perceived as an engine of growth for the economy as a
whole. Sometimes the successes of the HPAEs and Japan before them have been put forward
as exemplars of the efficacy of this strategy. A greater volume of exporting per se, however,
is no panacea for less-developed nations or the former colonies would have become devel-
oped long ago! Exporting is not a new phenomenon; the LDCs have always been exporters.
Recent research, such as that of edwards noted above, suggests that it is not exporting,
but export substitution and before that ISI, that were essential to the gains from exporting
of a certain type. And even easy export substitution is not sufficient for successful economic
development, even if it would appear to be a necessary stage of industrial transformation.
As we know from the endogenous growth models of Chapter 8, higher and sustained rates of
economic growth also require improvements in the stock of human capital and in the adapta-
tion and use of new technologies and knowledge in production if economic growth is to be
Countries which fail to accumulate human capital at a sufficiently high rate and of the
right type and to maintain knowledge acquisition ultimately will face lower growth rates than
comparable countries with more human capital accumulation, even if both have identical
Strategy switching and industrial transformation 315

Korea utilized foreign investment as a strategic means to gain access to technology and
skills at a lower cost than might otherwise have been possible through domestic chan-
nels alone. The Korean state™s approach to investment from abroad has been anything
but laissez-faire, taking what has been called the “eye of the needle” approach, to make
certain that any foreign investment met South Korea™s needs.
One study found that only 29.7 percent of foreign direct investment (FDI) in South Korea
took the form of wholly-owned subsidiaries of multinational companies. This compares
with 33.1 percent for Japan at the same time (1976), and an average 69.1 percent ratio
for the sixty-six countries in the study. In fact, Korea™s ratio of wholly-owned FDI was the
lowest in the sample.

Foreign investors were expected by their partners and by the Korean government to
make a continuing contribution to Korean development, one which was complemen-
tary to, rather than at the expense of domestic manufacturing interests.

In terms of technology transfers, the government usually approved these through
the Economic Planning Board or the Ministry of Finance, with input from the Ministry of
Science and Technology. Technical assistance contracts were typically limited to no more
than three years, except in complex processes, with the intention of forcing domestic firms
to learn how to do technology themselves rather than depending on foreign consultants.
Foreign technology in Korea was viewed as a teaching tool to encourage local adapta-
tion; it was not a fundamental cog in the development project meant to stand alone and
apart from the domestic economy. FDI was one means used to facilitate such learning, but
the focus was on how to upgrade the skills and efficiency levels of local factors of produc-
tion, including entrepreneurs, engineers, skilled workers, and so on.
Source: Luedde-Neurath 1988: 84“5, 90“3

levels of manufactured exports. Further, they will forgo some of the potential technological
gains available from the world pool of “best practice” production knowledge and sacrifice
economic efficiency, even if they are exporting manufactured goods on a relatively large,
or increasing, scale. Accumulating the proper human resources is a complementary cog to
industrial phasing and to exporting. 12

Are some exports better than others?
It is important that countries identify exports that have a reasonable prospect for expanding
demand over time and which also promise substantial value-added to the domestic economy.
Value-added in production is the income created at each stage of production of a good, so
the greater the value-added in production of a product the more income that is created for the
domestic economy. Goods with high value-added are those which require more labor time
and more machinery, technology, and knowledge to produce. Countries want to increase the
production of goods with higher value-added accruing to domestically owned resources,
since it is the production of these commodities for the domestic market and for export which
will add most to a country™s GDP.
How do countries determine which higher value-added goods to export? To create compar-
ative advantage in goods that are dynamic and adaptive dictates an export mix dominated by
commodities with income elasticities that are not just positive, but preferably greater than
one and which promise significant opportunities for growth in export earnings over time.13
316 The Process of Economic Development
Many manufactured goods, such as most electronics products, computer software, recre-
ational and sports equipment, plumbing supplies, and so on, fall into this category. Some of
these can begin to be produced during the easy ISI stage.
The success of not only Taiwan and Korea in recent decades, but of Japan before them,
would seem to strongly suggest that the sequencing of industrialization from easy ISI and
then transitioning to easy export substitution of higher valued-added commodities is a design
that not only builds upon the created comparative advantage forged during the easy ISI stage
but also furthers the goal of augmenting and revealing new dynamic comparative advantage
in the future.14
Comparative advantage changes, sometimes quite rapidly, as the result of increases in
education and in the skill levels of workers and managers, with the pace of technological
change and its domestic adaptation, and as a consequence of rising incomes and the changing
structure of domestic demand that accompanies an expanding urban, industrial society. The
economy™s domestic productive base, as well as its export structure, must be evolving in
concert with, or perhaps even leading, these changes.
Table 10.2 provides some information on the evolution of export structures over time for
the major regions and for various countries.
East Asia, dominated by the HPAEs, such as Korea and Taiwan, and South Asia, led by
India™s economy, show the greatest movement toward manufactured exports among the
different regions. Latin America and the Caribbean, despite a relatively high level of indus-
trialization by some measures (for example, the share of the labor force working in industry),

Table 10.2 export structure
Manufactured good exports (percentage of merchandise exports)

1970 1990 2000 2004

East Asia and Pacific 32 60 80 80
Latin America and the Caribbean 16 36 58 56
Middle East and North Africa 5 4 4
South Asia 48 71a 76
Sub-Saharan Africa 19 “ 24 31
79 81
High-income economies 72 80
Argentina 14 29 32 29
65 81a 91c 90
Brazil 13 52 59 54
14 9 4 5
“ 91
China 72 88
Côte d™Ivoire 6 11 14 20
52 73
India 70 77
Kenya 12 29 21 21
Korea 94 91 92
54 76
Malaysia 7 80
32 43 83
Mexico 80

Sources: World Bank 2002, 2006.
a Predominately textiles and clothing.
b 1999.
c 1998.
Strategy switching and industrial transformation 317
has not shown as strong a movement toward manufactured exports. In 1990, little more than
one-third of merchandise exports were manufactured goods.15 By 2004, there had been a
noticeable shift toward manufactured exports in Latin America, though the region still lags
behind the progress made in East Asia and the Pacific.
Looking at individual countries, Korea™s large share of manufactured exports in merchan-
dise exports reflects the effects of the government™s conscious efforts to promote export
substitution as ISI™s gains began to diminish. Among Latin American countries, Mexico has
shifted its export profile more in the direction of manufactured goods than have Argentina
and Brazil. India and, especially, Bangladesh have an export profile more similar to East Asia
and the Pacific, but these exports tend to be concentrated in low value-added commodities.
This is also the case with China, for the moment at least. For countries in Sub-Saharan Africa
like Cameroon, Côte d™Ivoire, and Kenya, manufactured exports remain very low. This is
due to the low level of industrialization in these economies; most are doing limited easy ISI
and have not switched to easy export substitution.
The growth in manufactured goods exports of the East Asian countries was obvious in
the evolution of the region™s share of total manufactured goods exports. Considering only
Hong Kong, Taiwan, Korea and Singapore, their combined share of manufactured goods
exports as a proportion of the manufactured goods exports of all less-developed countries
grew steadily, from 13.2 percent in 1965 to an astounding 61.5 percent of the total in 1990
(World Bank 1993: 38).
By 2004, the share of these economies had fallen to just over 50 percent, as other econo-
mies have been transitioning toward greater manufactured exports via export substitution
or other policies. There is learning taking place as the benefits to being able to export
manufactured goods as the East Asian economies have done is being replicated in other
Latin America™s share of less-developed country manufactured goods exports has declined
over the same period. This shift is not surprising when it is realized that East Asia™s exports,
led by manufactured export expansion, grew at an annual rate of 10.5 percent over the 1980s,
11.0 percent in the 1990s and 15.3 percent from 2000“4. Latin America™s total exports grew
by a much smaller 2.9 percent per annum in the 1980s, increasing to 8.5 percent in the 1990s
before falling back to a 4.4 percent growth rate for total exports from 2000“4. As a result of
these different trends in export expansion, by 2004 East Asia and the Pacific™s total exports
were more than double Latin America™s exports, having been slightly less in 1990.
What accounts for the relatively poor performance of Latin America™s manufactured
goods exports relative to the HPAEs and for the marked difference in the pace of develop-
ment? How have the HPAEs been able to “leap over” the level of development of the larger
Latin American countries, like Brazil and Mexico, and begin to approach the threshold of
developed country status at a more rapid pace than the Latin American countries, which had
a head start? And, by extension, what can other late-developing nations in Africa and South
Asia perhaps learn from the two experiences?

Difficult ISI: a sub-optimal strategy switch immediately after easy ISI
One explanation for the growing gap between the HPAEs and some other economies in the
less-developed world is found partially in the fact that an alternative sequencing of the path
of industrialization was followed after exhausting the possibilities of easy ISI other than easy
export substitution.16
The larger Latin American nations transitioned from easy ISI directly to difficult ISI
318 The Process of Economic Development
thus forgoing and skipping the easy export-substitution phase followed by the East Asian
economies.17 It will be recalled from the discussion of foreign exchange crises earlier in
this chapter that the structure of imports is transformed during easy ISI toward a growing
proportion of intermediate and capital goods, as non-durable consumer good imports are
progressively replaced by domestic production of those commodities. Difficult ISI deepens
the domestic industrialization process by extending import substitution backward into the
domestic production of durable consumption goods (like motor cars), intermediate goods
(such as tires and batteries), and capital goods (like car body stamping machines).
The shift toward difficult ISI is motivated by the same two parallel concerns that led the
East Asian economies to switch to easy export substitution. First, the exhaustion of easy
ISI growth possibilities meant that continued economic expansion fueled by manufacturing
growth is impossible without production for new markets. One possibility is to create these
markets through domestic vertical integration by pushing the domestic production structure
backward into intermediate and capital goods and consumer durables instead of continuing
to import them. This creates the possibility for continued economic growth based on local
demand via the usual macroeconomic income and employment multiplier effects of new
investments and spending and the crowding-in of tertiary production that emerges to service
the new difficult ISI industries that emerge.18
Second, the foreign exchange shortages fomented by persistent trade deficits due to having
import spending, M, greater than export earnings, X, can motivate a decision to “strategy
switch” toward difficult ISI. Thus, the decision to reduce the import bill even further via
additional ISI beyond easy ISI does seem a logical alternative to increasing foreign exchange
export earnings via export substitution, which is the alternative means for attempting to
resolve the foreign exchange disequilibrium that accompanies early industrialization.
The thinking behind such a decision is based on the assumption that an additional unit of
foreign exchange saved by a deepening of ISI is equivalent to an additional unit of foreign
exchange earned via export substitution. After all, if a country has a foreign exchange
shortage because M > X, and there is a need to turn this around so that X > M, then one way
to do that is obviously to reduce import expenditures, M, further so that they are less than
export earnings, X.
However, this static and purely mathematical view obscures longer-term effects that
suggest that a one-unit saving in foreign exchange from reducing M is actually worth less
over time than a one-unit increase in foreign exchange earnings that results from switching to
an export substituting strategy that increases exports, X. The easy export substitution indus-
trialization stage followed by the East Asian economies also had the goal of relieving the
foreign exchange shortages of the easy ISI stage. However, the export substitution strategy
resolves the M > X inequality by increasing export earnings, X, rather than reducing imports,
M. It can be seen that an industrialization strategy that increases export earnings not only
can result in X > M, but it also opens the possibility of a higher level of imports too. Further,
imports can only be reduced so much; they cannot go to zero. Exports, however, have no
such limit. So, in a real sense, increasing export earnings, X, is a superior strategy for over-
coming a foreign exchange crisis, because it can be virtually limitless.
The Latin American economies (and India) prematurely entered the difficult ISI phase of
industrialization, however, for reasons to be explored, and they thereby short-circuited the
potential for more dynamic growth and structural transformation. As a result, the period of
ISI-led industrialization in Latin America was distorted and over-extended, lasting in some
larger countries from the 1890s, and in others from the 1930s, to the 1950s in the easy ISI
phase, followed by the stage of difficult ISI from the 1950s to the late 1970s or so. The reason
Strategy switching and industrial transformation 319
why we call Latin America™s “ and India™s “ shift to difficult ISI immediately following easy
ISI premature is that they were not ready to produce the more sophisticated commodities
characteristic of that stage of industrialization, as will be explained more fully below.
By contrast, easy ISI as a central component of the structural transformation in the east
Asian economies lasted only about a decade, from roughly 1953 to 1963 (though there was
some isolated ISI before), and then the strategy switch was to export substitution and a
more dynamic export pattern, dominated by manufactured goods as primary product exports
declined in importance. We will see that when the East Asian economies eventually shifted
to difficult ISI, they did so after the easy export substitution stage had transformed their
economic structures. The economies were ready to produce the difficult ISI goods them-
selves, unlike in Latin America and India.

Premature difficult ISI in Latin America and India and its costs
What were the costs of skipping easy export substitution and instead strategy switching from
easy ISI to difficult ISI, as happened in the larger Latin American economies and India? Why
is it that transitioning to the production of more sophisticated manufactured goods, such as
automobiles, steel, and other heavy industries was a sub-optimal choice after easy ISI?

1 The sacrifice of viable entrepreneurial skills
The production of intermediate and capital goods and consumer non-durables characteristic
of difficult ISI tends to be more capital- and knowledge-intensive than is true of easy ISI
production. The technology for producing the difficult ISI goods is not only more difficult
to master, it often is based upon proprietary knowledge lodged in multinational corporations
(MNCs). Countries, and this was true in Latin America and India that entered the difficult
phase of ISI directly following the easy ISI stage, have done so primarily by promoting
multinational investment within their borders. It is often only MNCs which control the
technology and expertise, from management to engineering to quality control, required to
produce the more complex products involved in producing difficult ISI goods.
The dependence on MNCs resulted in a shift in the locus of power away from the still-
emerging class of domestic entrepreneurs who had been nurtured by the easy ISI strategy.
easy ISI had helped to create, protect, and promote the growth of an indigenous capitalist
class with the potential to be world-class competitors. A premature strategy switch in the
direction of difficult ISI stunted and even reversed the growth of the local capitalist class,
particularly as it occurred when these domestic entrepreneurs were not fully prepared to
produce the more complex array of products characteristic of the difficult ISI phase. They
typically had had but limited contact with the international economy and lacked the higher
levels of technological proficiency to produce even the easy ISI commodities since they had
not been forced to become efficient and competitive vis-à-vis foreign competition in export
markets as a result of the failure to reduce infant industry tariffs that would have provided
the stick of competition.19 This clearly demonstrates the importance of ending the regime of
infant industry tariffs if the efficiency gains that are desired from initiating industrialization
are to be realized.
This cutting-short of the maturation of the local entrepreneurial class was a crucial cost
of shifting prematurely toward difficult ISI strategy rather than transitioning to easy export
substitution strategy after the easy ISI stage of industrialization. Easy export substitution
allows the local entrepreneurial class to continue to evolve by becoming more efficient
320 The Process of Economic Development
producers, better able to weather international competition by being actively involved in
international export markets. In Latin America and India, the need to depend upon multi-
national investment as the principal agents in producing during the difficult ISI stage made
the technological learning process more difficult for local producers, who found themselves
increasingly closed out of the productive process.
The dependence on MNCs to produce the difficult ISI goods was necessitated by the
continued inefficiency of most local capitalists in the local and international markets. Govern-
ment policy had neither pushed nor helped local firms to become internationally competi-
tive, because infant industry tariffs had not been progressively eliminated as they should
have been. What should have been a transitional policy to temporarily shield new domestic
producers from foreign imports became permanent. Of course, many local capitalists thrived
and became wealthy without having to learn to compete, but at a cost to society as a whole,
where inefficiency was rampant.
When these economies faced the need to resolve their foreign exchange constraint so that
X > M rather than M > X, the shift to an export substitution industrialization simply was not
possible, even if it might have been contemplated. Unlike in East Asia, local firms in Latin
America and India had not yet become internationally competitive and thus could not export
the easy ISI goods they were producing for the domestic market. Given the impossibility
of shifting to a manufactured goods export strategy, Latin America and India could only
do more import substitution to try to balance out export earnings (still from primary prod-
ucts) and import spending. Since local entrepreneurs were not even efficiently producing
non-durable consumer goods, they were unable to move up and produce more sophisticated
consumer durables, intermediate inputs, and capital goods characteristic of the difficult ISI
stage. Thus, it was the MNCs that became the way to produce these goods, bypassing the
domestic entrepreneurial class. It is in this sense that we say these economies entered diffi-
cult ISI prematurely; local entrepreneurs, workers and the range of infrastructure were not
ready to produce difficult IS goods, having failed to become more efficient at producing even
lower-level, lower-technology goods during the easy ISI stage of industrialization.

2 Insufficient labor absorption
Since production in difficult ISI enterprises is more capital-intensive than in easy ISI or easy
export substitution, this strategy tends to slow the rate of labor absorption in the industrial
sector, unless the level of overall investment is extraordinarily high, which it has not been. As
workers continue to migrate from the rural areas to the cities, attracted by the lure of higher
wages, the labor force is increasingly to be found in urban rather than rural areas.
With the premature implementation of difficult ISI, however, many migrants fail to find
work in these more capital-intensive factories. Instead they often are forced to enter the
informal urban sector, where productivity and incomes are extremely low. There they toil
as artisans, petty traders, taxi drivers, day laborers, domestics, and so on, with the hope of
formal sector employment in manufacturing at best a distant possibility.
The relatively few workers employed in the difficult ISI firms often earn higher incomes
than those employed in domestically-owned easy ISI firms.20 The growth of employment
in the difficult ISI sector tends to be relatively limited, however, and thus a bifurcation of
income classes tends to emerge in the urban areas, with an adverse impact on the economy™s
overall income distribution (look back at Focus 5.3 on the Harris-Todaro model™s predictions
of precisely this sort of evolution if the wage wedge between industry and agriculture was
not closed over time). Urban slums ringing large cities became common throughout Latin
Strategy switching and industrial transformation 321
America and India as informal sector workers struggled to survive by building precarious
housing with limited access to sanitation, water, schools, and health facilities.
Just the opposite happened in Korea and Taiwan and other HPAEs. By further expanding
the labor-intensive production of consumer non-durables by pushing production outward
into the international market via export substitution policies, the domestic capitalist class
continued to be able to thrive and hone their skills, knowledge, and their management and
production expertise in an ever more open environment in which not only domestic entrepre-
neurs but also the domestic labor force learned to be more efficient.
The flow of workers from the countryside was more easily absorbed by the labor-intensive
nature of non-durable consumer goods production in East Asia. And with firms exporting
to world markets providing increased demand for these goods beyond the local economy,
employment in the industrial sector expanded rapidly. Urban unemployment rates remained
low, and the informal sector employment characteristic of economies skipping export substi-
tution was minimal. Workers and entrepreneurs upgraded their productivity and efficiency
levels, though for some time this did not appear as any substantial income growth for workers,
as gains from productivity growth were plowed back into further investment. The stage was
being set in the East Asian economies, however, for a wider sharing of the gains of income
and productivity growth among the population.
In Latin America, India, and a few other areas, premature difficult ISI resulted in continued
and expanded infant industry policies to block foreign imports from entering the domestic
market, as protection was now expanded to encompass intermediate and capital goods and
consumer durables, in addition to the easy ISI firms already being protected. Operating
behind relatively high tariff walls, domestic producers in easy ISI firms remained shielded
from the external pressures that had faced East Asian entrepreneurs and forced them to
become more technologically knowledgeable. Worker education and training on the job
was neglected too, as these investments in higher productivity were not necessary given the
high tariff levels that reduced competition from the world market.21 The result was negative
technical efficiency change, such that many economies have been moving away from “best
practice” technology and away from their potential production possibilities frontier (refer
back to Table 8.4).

3 High social costs
By jumping directly and prematurely to difficult ISI without fully gaining the benefits of
the easy ISI stage of industrialization, the Latin American and Indian economies imposed
high social costs on their populations. Governments often have been less efficient and
more prone to corruption and rent-seeking behavior. The benefits of infant industry tariff
policies in helping firms have time to learn to be efficient have not materialized because of
the over-extension of tariffs long beyond what can be justified. Thus the deadweight losses
in consumer surplus that infant industry tariffs imposed (examined in the previous chapter)
were not counterbalanced by gains in efficiency and lower costs and prices that could justify
their imposition in the first place.
Income distribution worsened when economies prematurely switched strategies to diffi-
cult ISI after easy ISI, as the informal sector with its low average productivity and low
incomes grew in size. And human capital accumulation lagged in these economies, as raising
the skills of workers was not necessary in the protected domestic environment. This lowered
average worker skills and resulted in lower incomes and has made knowledge acquisition
from the world that much more difficult for these economies.
322 The Process of Economic Development
Endowments and policies: explaining strategy switches
It is not enough to know that the East Asian economies followed a more optimal sequencing
of their industrial path that contributed to further economic expansion, while the Latin
American and a few other economies, such as India, failed to initiate an export substitution
strategy following easy ISI and “chose” a sub-optimal path of industrialization. It is necessary
to try to determine the underlying reasons for the strategy switches that led some countries
to pursue export substitution and others to shift prematurely to difficult ISI after easy ISI
if other countries are to learn from these experiences. What might explain these different
choices as to the stage of industrialization after easy ISI?

Differing endowments: the resource curse explanation
One might think that having relatively abundant natural resources would be a blessing.
However, Ranis (1981: 180“3) argues that for the Latin American countries, and by extension
for other large countries with ample natural resources, the apparent resource base “cushion”
of this initial endowment actually allowed those economies to continue along the same path
of exporting primary commodities. It was relatively easy to earn foreign exchange with
a large and elastic natural resource base and to continue their highly protected industrial
policies at the same time.
exports could be increased by simply producing more sugar or coffee or beef by using
more of the abundant land via extensive production methods, rather than having to resort to
becoming more efficient, as would have been required if land had been less abundant, i.e.,
by employing more intensive production methods, to use the language of Chapter 3. Even
with declining terms of trade per unit of these exports, total export income could be bolstered
by sufficiently increasing the physical quantity of exports if owners of resources found their
export income shrinking too much.
Further, there was the continuing influence on economic and political decisions of an
agricultural elite who profited from the existing primary export structure, and given that the
benefits of the protected ISI sector provided both profits and relatively high wages for a small
urban elite, the “many decades of import substitution growth have led to encrusted habits and
strong vested interest groups able to resist reforms or even marginal policy changes” (Ranis
1981: 180). Thus a large natural resource base, given the particular institutional barriers to
change that prevailed in Latin America and India, actually hindered the transition to the more
optimal path of export substitution. An apparent resource blessing turned into its opposite,
the so-called “resource curse.”

The availability of ample natural resources and/or foreign capital can thus be viewed
as permitting the system to continue on its old tracks, thus avoiding the political and,
at least short-term, economic pain of having to move to a different policy package.
Growth rates can in this way be maintained “ just by adding more fuel to the engine “
and difficult decisions postponed ¦ While additional resources, in theory, should be
able to ease the actual and psychological adjustment pains, they can be used, and in
the real world are often used, to put off “ or entirely avoid “ difficult decisions.
(Ranis 1981: 180)

In the East Asian economies, by comparison, land was at a premium. It was not possible to
simply increase the quantity of primary product exports to compensate for declining terms of
Strategy switching and industrial transformation 323
trade and to continue to do easy import substitution. ISI industries required a steady inflow of
imported inputs and capital equipment that could be paid for only by expanding the quantity
of exports. Sufficient foreign exchange earnings simply could not continue to be generated by
the primary product export sector if recurring foreign exchange crises were to be avoided.
Lacking a natural resource base, the East Asian economies were forced to confront the
urgency of maintaining and increasing foreign exchange earnings if the growth of domestic
industry was to be sustained, and it seemed that the only means for doing so was to find new
exports, specifically manufactured exports, given the limited natural resource base.
There is another factor, however, that helps to explain the different responses in East Asia
and Latin America and India to the given natural resource endowment. While land distribu-
tion and ownership patterns have changed very little in this century in most Latin American
countries and in much of South Asia, in East Asia, particularly in Korea and Taiwan, funda-
mental agrarian reforms were imposed after the Second World War (see Chapter 11). Thus,
East Asian policy-makers did not have to concern themselves with the “encrusted habits
and strong vested interest groups” that Ranis suggested might oppose the replacement and
substitution of primary product commodity exports with manufactured good exports from
the easy ISI industries.
The former landed class in Taiwan and Korea, as in Japan before them, had been forcibly


. 11
( 21)