The Changing Composition
of Developing Country Exports
World Bank Staff Working Paper No. 314
January 1979
The views and interpretations in this document are those of the authors
and should not be attributed to the World Bank, to its affiliated
organizations or to any individual acting in their behalf.
Prepared by: Hollis B. Chenery
Donald B. Keesing
DevelopmenT Policy Staff
Copyright () 1979
The World Bank
1818 H Street, N.W.
Washington, D.C. 20433, U.S.A.






The views an.d interpretations in this document are solely those of
the authors, and should not be attributed to the World Bank, to its
affiliated organizations, or to any individual acting in their behalf
WORLD BANK
Staff Working Paper No. 314
January 1979
THE CHANGING COMPOSITION OF DEVELOPING COUNTRY EXPORTS
The role of exports from developing countries has been at the
center of postwar discussions of the world economic order. This
paper explores the influences that have shaped export growth since
1960 and the prospects for their further expansion. Particular
emphasis is given to manufactured exports, which grew more than
twice as fast as total LDC export. from 1960-75.
If the momentum of manufactured export growth can be continued
as the World Bank has recently projected, LDC export growth in the
period 1975-85 could exceed 6% per year despite slow growth (probab-
ly no better than 3-4% per year) of primary exports. Manufactured
export composition and prospects in four groups of LDCs are analyzed.
A majority of these exports come from a small number of countries
that have specialized in manufactures starting in labor-intensive,
technologically stable ("older") products, using cost-cutting stra-
tegy based on low wages, while relying heavily on foreign buyers
for marketing and product design.  These countries are now trying
to diversify into more complex and skill-intensive products. Mean-
while,increasing numbers of LDCs not in this group have shifted
their policies at least part way in an export-oriented direction.
The rapid increase in manufactures exported from LDCs inter-
acting with developed countries' own poor economic performance
have combined to trigger rising protectionism, adding to uncertain-
ties about markets. An especially serious aspect of the problem is
the effect on export prospects of those LDCs that are not yet suc-
cessful exporters of manufactures, but will need to be so to faci-
litate their growth. Special measures to give the poorer and less
successful LDCs special export opportunities may have to be
considered.
Prepared by: Hollis B. Chenery
Donald B. Keesing
Development Policy Staff
Copyright  (c   1979
The World BaEk
1818 H Street, N.W.
Washington, D.C. 20433
U.S.A.



THE CIIANGING COMPOSITION OF DEVELOPING COUNTRY EXPORTS*
Hollis B. Chenery and Donald B. Keesing
Development Policy Staff, World Bank
C O N T E N T S
Page No.
Introduction                                           1
I.   EXPORT TRENDS IN DEVELOPING COUNTRIES             3
A.  Export Trends by Region                      3
B.  Trends in Primary Exports                    5
C.  Trends in Manufactured Exports               12
II.  CHANGING COMPARATIVE ADVANTAGE AND LDC           20
A. Comparative Advantage by Country Groups    21
B. Changing Composition of Manufactured          29
Exports
III. FUTURE PROSPECTS                                 25
A.  Effects of World Growth                      35
B.  Supply vs Demand Constraints                 37
C.  The Changing Role of Manufactured Exports  46
* This is a revised version of a paper presented to a symposiur
organized by the Institute for International Economic Studies
on "The Past and Prospects of the Economic World Order",
-tCockholm, August 25-28, 1978.



ThE CHANGING COMPOSITION OF DEVELOPING COUNTRY EXPORTS
Hollis B. Chenery and Donald B. Keesing
World Bank
Ts^_ role of exports from developing countries lias been
at the center of postwar discussions of the world economic order.
Until the mid-1960s the developing countries (LDCs) had a stead-
ily declining share of world exports, and manufactured goods
provided only about 10% of their export earnings. Slow export
growth was correctly perceived as a major obstacle to accelerated
development.
This picture has changed dramatically over the past
fifteen years without any drastic change in the policies of
industrialized countries toward the LDCs. The sustained growth
of the world economy up to 1974 and the shift toward more export-
oriented strategies by a large number of developing countries
have accelerated the growth of LDC exports, primarily from
those countries in a position to export manufactures. For
developing countries as a group, the export lag appears to have
been overcome, and nearly half of the increase in earnings now
1/
derives from manufactures.
1/ Although the rise in the price of oil contributed enormously
to the increased export revenues of a few developing
countries--notably Iran, Venezuela, NJigeria, Indonesia,
Algeria and Iraq--supply as well as demand limitations make
a continuation of these increases unlikely. (Throughout this
paper, the capital surplus oil countries--Kuwait, Libya, Oman,
Qatar, Saudi Arabia and United Arab Emirates--are treated
as a separate group so as not to distort the analysis of
other developing countries.)



-2-
When we take a less aggregated view of recent perfor-
tiance, major features of the earlier picture reemerge. Export
expansion has been a major contributor to the growth of a number
of countries, whose incomes have risen substantially over the
past twenty years. However, both export expansion and GNP
growth have been concentrated in countries containing only a
third of the population of the developing coitntries.  The poorer
LDCs, concentrated in South Asia and sub-Saharan Africa, supply
only 10% of LDC exports aAd still have most of the features that
characterized the third world in the 1950s.
The purpose of this paper is to assess the main factors
causing this transformation of LDC exports and to consider the
contribution that they may make to the future growth of different
1/
groups of countries.   Since the bulk of export growth is expected
to take place in manufactured goods, we focus mainly on this cate-
gory. The first section of the paper examines the changing com-
position of exports since 1960 and its association with the
emergence of a small number of successful LDC exporters of manu-
factures. The second section relates this performance to the
development strategies and changing comparative advantage of
different groups of countries. The final section takes up growth
2/
prospects and some of the policy issues raised.
1/  For simplicity, only merchandise exports will be considered;
henceforth, exports means merchandise exports except where
specifically noted.
2/ This study is an outgrowth of continuing work in the World
Bank on the prospects for developing countries in relation
to the future evolution of the world economy. It makes use
of two recent reports, Prospects for Developing Countries,
1978-1985 (World Bank, Development Policy Staff, 1977) and
World Development Report, 1978 (WDR).



-3-
I. EXPORT TRENDS IN DEVELOPING COUNTRIES
Since at least the 1950s, LDC exports have grown in
volume more slowly than those of industrialized countries (MDCs)
and those of the world as a whole, largely because they have been
concentrated in primary products, where the growth of demand
and world trade has been relatively slow. This situation is
now changing rapidly as increasing numbers of LDCs shift toward
manufactures. The share of manufactures in exports has risen
rapidly, and they will soon comptise over half of LDC merchandise
1/
exports other than oil. If this. expansion continues as expected,
it will lad over the next few years to an export growth rate
roughly equal to that of the rest. of the world. These trends
2/
for major commodity groups are stmmarized in Table 1.
A. Export Trends by Region
Growth of exports and cutput has been generally
fastest in regions, such as East Asia, that have been most
successful in exporting manufactures, while low growth perfor-
mance has been especially characteristic in low-income countries
(those with GNP at or below $250 in 1976). If Indonesia is
excluded, the low-income countries have experienced very little
growth in their exports since 1965. GNP increased in the low-
income countries at only 3.1% per year from 1960-75 compared to
1/ Even if Southern Europe is excluded, this share is now close
to 40%.
2/  This is the "base" or middle case in which GNP in MDCs is
projected to grow at 4.2% per year. (Cf. Table 10 below.)



Table 1: PAST AND PROJECTED RATES OF EXPORT GROWTH BY BROAD PRODUCT GROUPS
(In Constant 1975 Prices)
Percent of LDC Exports       PercenL Share
World     LDCs       World      LDCs                                     of Increas
1960-75   1960-75   1975-85   1975-85       1960    1975    1985       1960-75  1975-i5
Fuel and Energy              6.3       6.2        3.6       3.4         39     40        30         42        18
Agricultural Products       4.2        2.6        4.4       3.1         43     27        20         16        12
Non-fuel Minerals            3.9       4.8        4.2       5.8          7       7        7           6        6
Kpnufactures                 8.9      12.3        7.8      12.2         11      26       43         36        64
TOTA' KERCHANDISE         7.1        5.9        6.4       6.4        100    100       100        100       100
Sources: World Bank, World Development Report, 1978, Tables 13 and 25 and unpublished projections for WDR.



6% in the other ("middle-income") LDCs. Exports per head are
12 times higher in this second group. Broad differences in the
composition of exports between these and other groups of countries
1/
are summarized in Table 2.
In coming years. much as in the recent past, export
growth is expected to be fastest in regions with a high propor-
tion of manufactures in their exports. This is illustrated by
Table 3, which shows projected growth rates of LDC exports by
region in 1980-85 (a choice of years which allows us to exclude
complications associated with the post-1975 recovery) compared
to 1965-76. Export growth is expected to be fastest in the
middle-income countries of East Asia and the Pacific followed by
Southern Europe;  these are the regions most successful (and
most specialized) in manufactures. Sub-Saharan Africa is
expected to have the slowest export growth followed by Ncrth
Africa and the Middle East; these are the regions most depen-
dent on primary commodities.
B. Trends in Primary Exports
1. Fuel. In contrast co the slow growth in mcst other
primary exports, there has been a bonanza in LDC oil exports,
thanks to the four-fold rise in price in 1973-74, following a
sustained growth of volume at rates of 6-7% per year. As a
result, even with capital-surplus oil exporters excluded, fuel
has come to outweigh all other primary exports combined.
1/ Apparent similarities between low- and middle-income LDCs in
the composition of exports disappear in a disaggregation by
regions.



Table 2: WORLD EXPORTS BY MAJOR GROUPS OF COUNTRIES, 1975
Capital
Surplus
Low Income         Other                          Centrally      Oil
Africa and Asia   Developing   Induatrialized    Planned   Exprters    World
Value of Exports (US$ bils.)          18               162              551              82          52        865
Percent Share of:
World Population                    30                 22              17              30          0.3        100
World Exports                        2                 19              64               9          6          100
Export Composition (X)
Fuel and Energy                     33                 41               3              is        100           19
Agricu'tural Products               36                 26              16              16          -           17         S
Non-tu,i Minerals                    9                  7                5              6                       5
Manufactures                        23                 26               76             60          -           59
Total                              100                100              100            100        100          100
Sources: World Bank background data for World Development Report 1978 baaed on UN. trade statistics; populatious
from World Bank Atlas, 1977.



- 7 -
Table 3: PROJECTED GROWTH RATES OF DEVELOPING COUNTRIES' EXPORTS
BY REGION, 1980-85, COMPARED TO ACTUAL RATES FOR 1965-1976
(Percent Per Annum)
Region                                     1965-76            1980-85
Sub-Saharan Africa                           4.*3/              3.4
Low-Income Asia                              3.  /              5.7
Middle-Income Asia and Pacific              12.2               10.0
North Africa and Middle East                12.*8cj             4.1
Latin America and Caribbean                  4.7                5.6
Southern Europe                              9.2                8.4
All Developing Countries                   8. 3,=            6.6
a/ With Nigeria and Gabon excluded this becomes 2.6.
b/ With Indonesia excluded this becomes 1.1
c/ This is biased upward by the deflators used (see note below), as
a result of rising oil prices.
Note: 1965-76 rates are not in constant prices but are deflated by
unit value indexes taken from UN Yearbook of International Trade
Statistics 1976, Vol. I, p. 81, and UN Monthly Bulletin of
Statistics, July 1978, p. xx. These show unit value of LDC
exports for all LDCs (1970-100) increasing from 93 in 1965 to
331 in 1976, and for importers of petroleum, from 89 in 1965
to 196 in 1976. Oil exporters' nominal export growth was deflated
by the first and that of other LDCs by the second index.
Sources:  1965-1976 estimated based on export statistics from IMF
International Financial Statistics and Direction of Trade,
UNCTAD Handbook of International Trade and Development
Statistics, June 1978; 1980-85 projections based on
projections based on background studies for WDR.



Benefits from this windfall have been concentrated in
a sr,all set of oil-exporting countries. It includes two large,
poor cDuntries--Indonesia and Nigeria--whose growth has been
substantially accelerated in the past five years. The other
OPEC members have benefitted even more, but their total popula-
tion is less than 100 million people.
Prior to the oil price rise, world demand for fuel
exports rose faster than world income, but since then it has been
expanding more slowly. The greatly increased incentives to expand
domestic fuel supplies and to economize on energy use are helping
to turn up new sources in LDCs while slowing the growth of world
trade and consumption. On balance, the prospect seems to be for
growth in export volumes at rates of 3-4% per year with a possi-
bility of future price increases in real terms.
2. Agricultural Products. Table 1 shows that agricul-
ture, livestock, forestry, fishing and food processing together
..                l~~~~~/
supplied 43% of LDC exports in 1960, but this share fell to
27% in 1975 and is trending downward. World demand for exports
of agricultural products nearly keeps pace with world income,
growing at rates over 4% per year. Expansion of export demand
reflects not only rising incomes but also growing food deficits
in ittan) centrally planned economies (CPEs) and LDCs. While the
LDC_. as a group were virtually self-sufficient in food in the
1950s, by 1975 their net imports exceeded 30 million tons.
Present ^stimates poinit to a food deficit of some 45 million tons
by  98r.
1/ or 52% in current prices.



-9-
The LDC share of world exports of agricultural products
has fallen over time; their growth rate of 2.6% from 1960-75
was less than half that achieved in MDCs. This weak performance
has been due, first, to an initial concentration in tropical
products such as coffee, cocoa, tea and bananas, in which world
demand has expanded only slowly. Second, trade and tax regimes
in many LDCs have discouraged agricultural production and exports
in favor of other activities, notably manufacturing for the home
market. Third, world efforts to raise farm output through agri-
cultural research and extension, investments and improved
inputs are largely concentrated in temperate zones and richer
countries, where LDC exports are also held back by MDC protection.
However, despite these handicaps, agricultural output in LDCs
has actually grown faster than in MDCs, but has been offset by
rising home demand due to population and income growth.
Increasing the growth rate of agricultural exports
from LDCs hinges on accelerating supply in the face of fast-growinc
domestic demand. In a number of agricultural products (grain,
meat, timber, rubber, etc.), the potential export market is
favorable, even though food products face widespread protection,
particularly from nontariff barriers in OECD countries designed
to give better prices to farmers. Policies toward agriculture
have been improving in some of the major developing countries.
As a result, both production and export growth are expected to
rise, the latter to over 3% a year through 1985. Efforts to
stabilize prices and earnings in agricultural commodities, to the
extent they are successful, could contribute to export expansion
by reducing uncertainties that discourage investment. Projected



- 10 -
grxi.fwth -ates of LDC exports in leading agricultural products,
together with those in leading minerals are shown in Table 4.
3. Non-fuel Minerals. In this category (which
includes nonferrous metals), LDCs have achieved export growth
rates close to 5%. While world demand expands more slowly than
world income, the LDC share of total supply is growing, thanks to
untapped supplies and less exploitation than in more industrial-
ized countries. These resource advantages are partly offset by
the reactions of MDC investors to increased taxation and the
political uncertainties in LDCs. Here, as in oil, the diffusion
of technology has improved the bargaining position of host coun-
tries, which increasingly have the options of national ownership
or joint ventures as alternatives to foreign-owned mining opera-
tions. (Domestic demand beyond the processing stage is usually
a negligible influence on export availability.)
Growth of LDC mineral exports is projected to accelerate
somewhat and to continue to outpace world demand, with export
growth rates approaching 6% per year. To a somewhat lesser extent
than in fuel, these exports come mainly from a small set of LDCs.
The mineral exporters rarely have high GNP per capita compared
to the general range for middle-income countries, and in sub-
Sahal-an Africa most of these would be classed as low-income or even
";east developed" countries were it not for their mineral exports.
4.  Diversification.  A notable trend in primary exports
has Leen product diversification at the country level, outside
t-ie vil-exporting countries.  Our colleague, Martin Wolf, has
comp-i»1 the concentration of prir'--' -ommodity exports in 1960



- 13. -
Table 4: HISTORICAL AND PROJECTED GROWTH RATES OF LDC EXPORTS
BY VOLUME IN LEADING AGRICULTURAL AND  MIAL PRODUCTS
Value in      a/   Volume Growth Rate of LDC Exports (% p.a.)
1974 (US$ mu.)-        1960-1976Ž/                 197,i/76-1990
Copper               5,320                  3.7                         3.3
Sugar                5,083                  2.8                         2.7
Coffee               3,984                  1. A/                       2.6
Fats and Ol.s        3,877                  57. c                       5.2
Timber Logs          2                      8.6c/                       2.2
Sawnwood          2                      8.8                        8.0
Cotton               2,291                 -0. 2-=                    -0.3
Rubber               2,212                  3.6                        3.6
Iron Ore             1,693                  6.9                        3.9
Phosphate Rack       1,692                  5 2d/                      5.1
Cocoa                1,555                  2.2-                       2.5
Tin                  1,256                 09                         1.5
Maize                1,153                  5:4/                       3- .6
Rice                 1,094                 -0 7                        0.0
Tobacco                691                  3.9c                       5.3
Tea                    672                  l.  '                      2.3
Beef                   609                 -1.7'/                      5.4
Bananas                602                  3.3                        2.1
a/ No one year is entirely representative because of price fluctuations; for
example, in 1974 prices were exceptionally high in copper and phosphate rock.
b/ Trend growth rates.
c/ 1961-1976.
d/ 1955-1976.
e/ Growth rates are for all coarse grains.
f/ 1961/63 - 1974/76.
Source:  Commodities and Export Projections Division, World Bank.



- 12 -
1/
to  . ut  n l'74 for fifty-five non-OPEC LDCs.   The proportion
-' iese b_untri,i aependent on one commodity for more than half
of r! _1ix.itary e<port earnings fell from 49% in 1960 to 18%
ir...ii4.  She mean share of the leading single primary commodity
fell from 47% to 35%, but there was no tendency for the second
and thirc. exports to decrease in relative importance. ThLis diver-
sificatio. seems to be largely a result of deliberate efforts by
LLCs to escape from dependence on a single commodity by building
up expozt±; of other products.
C. Trends in Manufactured Exports
The most basic change in the pattern of LDC exports
has been 1-he spectacular growth in manufactures. From 1960 to
1975 these exports expanded at rdtes of over 12% per year. This
growth accelerated in the late 1960s and early 1970s and has con-
tinued strongly despite the setbacks suffered by the world economy
since 1973. During the world recession of 1975 the volume of LDC
manufactured exports seems to have risen in volume by at least
6-7% even though output and consumption of manufactures fell
2/
sharply in MDCs.    In the 1976 recovery, LDC manufactured
exports leaped forward by over 20%; and in the slow growth con-
ditions of 1977, they appear to have expanded once again in real
terms by over 10%, despite a decline in clothing and textiles,
3/
although world trade increased only 4%.
1/ In unpublished research for World Development Report.
2/ This export growth was largely masked by a shift in the terms
of trade against these exports.
3,  Thiese estimates are all prelimi.-.ry.



- 13 -
1. Country Concentration. The bulk of manufactured
exports come from a small number of industrially relatively
advanced LDCs, as shown in Table 7 below. The Republic of Korea,
Taiwan (Republic of China), Spain and Hong Kong together supply
close to 45% of the total. Next come Yugoslavia, Brazil and
Mexico. However, the growth in these exports has been widely
shared, cis illustrated by Table 5. More than forty LDCsE now
have exports worth over $100 million, compared to 22 in 1970 and
fewer earlier.  Although the scale and pace of this growth have
varied from one country to another, in many the upsurge has been
persistent and powerful.
At a product level, the3 breadth of this advance is
impressive. From almost any year in the 1960s through 1976, real
growth rates have been over 10%, not only in clothing and elec-
tronic assembly (which have made striking gains), but also in
machinery, transport equipment, :extile yarn and fabrics, steel,
chemicals, and almost every other major group. Growth of 10%
per year or more seems to be the rule rather than the exception
1/
even at the level of individual products.    However, most of these
strong performances started from a low base in the 1960s, parti-
cularly in products that are technically complex, skill-intensive
and/or subject to strong economies of scale.
Looked at slightly differently, the steady advance of
manufactured exports has been accompanied by an equally persistent
diversification, with successful export breakthroughs being wide-
spread. In practically every case the more complex and
1/ A detailed analysis of LDC produce performance is given in
Donges and Riedel (1977).



- 14 -
Table 5: INCREASING NUMBERS OF LDCS EXPORTING MANUFACTURES,
1965-1975
(Cumulative, at 1975 Prices)
Exports of Manufactures Of.er:       1965    1970   175
US$2 Billion                      0       2        9
US$1 Billion                      3       6       L2
US$500 Million                    7      11       .L5
US$200 Million                  12       15       25
US$100 Million                  18       22       40
US$50 Million                   27       37      46
Source: WDR Table 15 and UN trade data; price index from UN Monthly
Bulletin of Statistics,



- 15 -
difficult-to-produce of these products come initially from one
or two of the most industrialized LDCs. Among less industrialized
LDCs, one finds the same sort of process going on at a smaller
scale and in simpler products.
2.  Exports by Destination.  Strong growth        taken
place in manufactured exports in all major markets.  Especially
rapid expansion has occurred in LDC exports to MDCs, which have
outpaced the rest. Those going to other LDCs and to CPEs have
also increasedin 1960-75 at annual rates above 10%.
Table 6 illustrates the compositional differences
between manufactures exported to MDCs and to other LDCs in 1975.
In the aggregate, nearly two-thirds of LDC manufactured exports
go to MDCs, nearly one-third to other LDCs, and limited amounts
to CPEs. There is considerable vrariation by product: products
such as clothing and electronic products go mainly to MDCs, while
products such as textiles and chetmicals are traded more among
LDCs. This is also true of some types of machinexy and transport
equipment. However, trade in capital goods among LDCs i.; quite
limited: nine leading exporters reported only $587 million in
*                                        ~~~~~~~~~~~~~1/
exports of non-electrical machinery to other LDCs in 1974.
The table also shows the composition of LDC imports
from MDCs, which outweighed trade in the opposite direction by
$123 billion to $26 billion, or more than 4 to 1, along with the
1/ Based on an unpublished World Bank study by Yoriko Kawaguchi;
the nine LDC exporters are Brazil, Argentina, India, Yugo-
slavia, Malaysia, Mexico, Hong Kong, Korea and Colombia. In
machinery and transport equipment, U.N. trade statistics
appear to be inflated by reexports among LDCs and to under-
state exports to MDCs.



Table 6: COMPOSITION OF LDC TRADE IN MANUFACTURES BY DESTINATION, 1975
(Percent)
Trade Among   LDC Exports to   LDC Imports   Share of Inter-LDC Trade
LDCqJ           MDC8           from MDCs       in Total LDC Imports
Machinery and
transport equipment        31              18               55                      6
Textiles                     14             10                4                     28
Clothing                      5             19                1                     39
Chemicals                    13              7               12                     11                         C
tron and steel                6              5               10                      7
Other manufactures           31             41               19                     15
TOTAL                     100            100              100                     10
a/ Includes slgnificant reexports, notably in machinery and transport equipment.
Source: UN Yearbook of International Trade Statistics, various issues.



- 17 -
share of trade among LDCs in total LDC imports of manufactures
in each broad product group.
3. Overview. In summary, the rapid growth of manufac-
tured exports appears to have been mainly the result of a supply
breakthrough. Demand growth has played only a secondary role,
especially since LDC shares of most markets were initially quite
small. Substantial growth has taken place even in products in
which aggregate demand grew only slowly: in clothing, consump-
tion appears to have grown in MDCs at about 3% per year while
LDC exports to MDCs grew at over 20% a year.
LDC success in exporting manufactures seems to have
four main explanations, all but one of them on the supply side:
-     (1)  Rapid overall development has increased the
industrial capabilities of the leading exporters of manufactures
and contributed to their ability to move into related export
products.
(2) There has been a widespread shift in LDC policies
away from inward-looking industrialization around the home market,
toward a systematic effort to export industrial products.
(3) Improvements in transport and communications
have facilitated growth of trade and an international division
of labor even over long distances.
(4) International trade policies of the industrial
countries have maintained access to their markets and, at least
until recently, continued the postwar trend toward trade liberal-
ization in industrial products. LDC exports have benefitted
from Dillon and Kennedy round tariff cuts, offshore assembly



- 18 -
provisic.s in tariffs, preferences from the European Community,
Gentiralized Syster of Preferences (GSP) schemes in various MDCs,
-nd resistance to pressures for more restrictive nontariff
1/
barriers.
For the last two reasons, trade in manufactures has
grown faster than output in both MDCs and LDCs. The faster
export growth in developing countries is attributable to the
first two influences--rapid industrial development and a shift
2/
toward outward-looking policies.
It is not widely appreciated how many developing coun-
tries have been changing their policies to promote manufactured
exports. This shift began around 1960 and went further later in
Spain, Taiwan (Republic of China), Republic of Korea, Yugoslavia,
3/
GreectC, israel and Portugal.   In the course of the 1960s, Brazil,
Mexico and Singapore and to a lesser extent others (e.g., Pakistan)
shifted policies in the same direction. More recently export
prGrotiuLuA measures have become a significant component of indus-
trialiZaLion policies in many other countries, notably Colombia,
Argen.ira, Uruguay, Chile, Sri Lanka, Tunisia, Morocco, Haiti,
Dominican Republic, some of the Central American republics, Cyprus,
1/  hev%eve,r, trade liberalization has favored MDC industrial
e%ports over LDC specialties, which tend to be exempted from
t4riff cuts and GSP schemes.
2/  The initial composition of these exports would almost certainly
have led to a slower growth rate in LDCs than MDCs, if both
grouks of countries had maintained their market shares.
In Hong Kong export promotion effors s began b-y the late 1940s,
wAilN  various industrial countries  Ind Puerto Rice) set
eC9rmples in the 1950s.



- 19 -
l/
Malta and Mauritius.
1/ Still others have begun to experiment with export processing
zones (e.g., Indonesia, Syria), export-oriented investments
(e.g., Venezuela, Iran) or regional common markets of various
kinds (in Africa, Latin America and the Caribbean).



- 20 -
II.  CHANGING COMPARATIVE ADVANTAGE AND LDC GROWTH
A number of studies have shown a strong relation
between export expansion and the growth of GNP.   It is fairly
well established that countries cannot continue to grow rapidly
by steadily reducing the share of imports in GNP and that expan-
ding the share of imports and exports in GNP permits more effi-
cient allocationof resources and more rapid growth. At least
over an extended period, export growth seems to be a necessary
2/
though not a sufficient condition for rapid GNP growth.
The previous discussior. shows that in order to accelerate
the growtn of exports the develoFing countries have in recent years
changed tLeir export composition substantially, shifting away
from the ,lower-growing primary products toward manufactures,
and within manufactures, diversifying into new commodities as
skills and experience are acquired. We will now examine this
process in somewhat more detail for groups of countries that
differ in their initial income levels, resources and trade policies.
Recent empirical studies have had considerable success
in explaining the composition of manufactured exports from LDCs
in terms of both the Heckscher-Ohlin approach (including skills
I/  S-e, for example, Maizels (1968), Chenery (1971), Balassa (1977a)
and Michaely 1977). The last allows for the fact that exports
are a component of GNP by relating GNP growth to the increase
in the share of exports in GNP, with highly significant results.
2,/ F'- limited periods, capital inflows can replace exports as
a source of foreign exchange, but in cases of successful devel-
opment this has only deferred the need to expand exports by
a decade or so.



- 21 -
as a factor of production) and more recent technological explana-
tions based on the product cycle, technical progress and scale
economies.    There have been fewer explorations of other influences
such as learning processes in marketing and product design, that
contribute to success in exporting manufactured go6ds. These
influences appear to be part of a cumulative process that has
produced a relatively small number of very successful LDC exporters.
A. Comparative Advantage by Country Groups
To illuminate the relationships among initial econovic
structure, trade policies and export performance, we shall use
a modified version of the classification of trade and production
patterns proposed by Chenery and Syrquin (1975). *Table 7
classifies selected developing countries (including all those
having more than $150 million worth of manufactured exports in
1977) into four groups. The first three contain countries with
a per capita income (in $1976) of more than $300, which are well
along in the process of development; the fourth consists of
poorer countries.
I.  Countries that specialized relatively earlyin
exports of manufactures and have followed generally outward-
looking policies.
1/ See, for example, Hufbauer (1970), Hirsch (1974), Stern (1975),
Helleiner (1976), and Balassa (1977b). A number of regional.
applications are given in Giersch, ed. (1974).
2/ Except Jamaica, whose exports of manufactures consist almost
entirely of aluminum oxide, and New Caledonia, which exports
ferrous nickel.



Table 7: EXPORTS OF MANUFACTUIRS AND COUNTRY CHARACTERISTICS OF SEICTED Lix.a
Cou_trgd GP Per Ca pit4                                          M_nufactures as            _j   ofNanufat-d Goo4                               ___-
.i 1l; 1976)  (US$ 1976)  AyL. _Growth   of Goods Exports             Total      Per CanL i     textiles  C1othig.         geal Awer .e
ft p.eflU               J52us$)                                                                Grcwth
1960- 1976        1960    1975            197g           1976        L                        .tkWr  Lrrtuct  L. ,41
_(1 of Tctal)          196S-75
Israel                    3.6     3.920          4.3            61       83             1,850          514                10                   11.1
SingapoLe                 2.3     2,700          7.5            26       43             1,79C          7781
Greece                    9.3.    2,590          6.1              9      48             1,252          138                45                   28.7
liong Kong                4.5     2,110          6.5            80       97             6.480       1,440                 S8                    11.91'
portugal                  9. 7    1,690          6.5            55       71             1,198          124                44                     7.8
Taiwan (A,  ,e :IjDna) 16.3       1,070          6.3            14       81             6,921         425                 44                   28.
Korea, Rep.  f           36.0       670          7.3            14       82             6,675          188                50                   36.0
tT   Large Sei T,,1..., . - .  Countries
Spain                    35.7     2,920          5.5             12      70             6,025          169                19                    22.6
Yugoslavia               21.5     1,680          5.6            44       72             3,383          157                21                     9.9
Argentina                25.7     1,550          2.8              4       25              972           38                23                    16.7
Brazil                  110.0     1,140          4.8              3       27            2,332           21                29                    25.4
Mexico                   62.0     1,090          3.0             12       52            2,327k'         38                15-!                  21.2
Turkey                   41.2        990         4.2             25       36              466           11                70                    32.2
IIT   Emerging f'.    .-y Specialization
Veneruela                12.4     2,570          2.6              0        1               I S' 81'                        2s/                  20.1
Iran                     34.3     1,930          8.2             3         1              208Z/          65/              71Sf                   t.9       s
Chile                    10.5     l,050          0.9              4        a8109_/                      110/               0                    n.c. 4
Malaysia                 12.7        860         3.9              6       18               667S/        54Z/              145                  18.2
Tunisia                   5.7       840          4.1             10       20               203          36                49                    15.0
Coloumbi                 24.2       630          2.8              2       21               306-'        135'              3,5/                  17.3
Ivory Cost                7.0       610          3.4              1       12               117          17                27                    17.4
Morocco                  17.2        540         2.1              8       13               1975'         110             61-J                  16.6
Philippines              43.3       410          2.4              7       17               255-C         651              275!                    7.8
Thailand                 43.0        380         4.5              2       23               305E           7t/              286'                 30.0
[V.  LArge Pool       ,i*s
Egypt                    38.1        280         1,9             10       34               386           10                73                     7.6
Indonesia               135.2       240          3.4              0        1               119           1                (1owe                  14.1
Pakistan                 71.3        170         3.1             22       55               s89g/          9E/              7910                    a/
India                   620.4        150         1.3             44       45             1,961./         3E/               52-/                   2.8
Bangladesh               80.4        110        -0.4             ..       63               220            3                97                      !/
a/ Including reexports.
b/ Estimated including border zone, with help of U.s. as well as Mlxicen data.
c/ 1975
d/ 1974
S/ For Pakistan and Ieanladeeh together, axporta of snaufacturea grew at 8.5 percent.
qm,rcea:  Firat five colunas from World Bank,   eorld              ent   AMort *statistical annex; other trade data from UN Yearbook of Internatlonal
T'ade Stactitica and other TIJN sources, plus national statistics for Taiwan(Rapublic of China) end lexico; populatiots for 1975
A*,d 1974 from Uorld  Bank Atlaa; real export growth 1965-75 'owuted using UN unit value index for SITC 5-8 for deloping
(0-"C ries as a deflator.



- 23 -
II. Large semi-industrial countries, with relatively
low export shares in GNP, that have achieved considerable success
in industrialization based mainly on the home market, but in
recent years have also tried to promote exports of manufactures.
III.  Countries now shifting part way from  -ecial-
ization in primary exports in an effort to diversify their
exports and accelerate development.
IV. Large poor countries with significant exports of
manufactures.
These groupings are meant to illustrate the main policies
1/
and countries involved, without being exhaustive. Brief
comments follow on each group.
Group I: Early Specialization in Manufactures
This group contains four East Asian countries (hong
Kong, Singapore, Taiwan and Korea) and three Mediterranean
countries (Israel, Portugal and Greece). Although waqe levels
were somewhat higher in the Mediterranean group in 1960, they
are all characterized by limited natural resources, relatively
educated labor, and the need to export manufactures (or services)
,in_order to develop. The composition of their exports generally
corresponds to expectations based on trade theories of both
factor-proportions and product-cycle types. With local varia-
tions, their exports have been built initially around
1/ A classification of all transitional countries is given in
Chenery (1977), which also traces the past growth of each
of the main groups. In addition to the countries shown in
Table 7, there are many smaller exporters of manufactures.
/



- 24 -
labor-intensive, technologically stable ("older") products
such as textiles, clothing, footwear, assembled electronic com-
ponents, toys, etc. 1/
In these footloose industries, the role of policy is
substantial in determining the location of particular products.
For example, over 90% of LDC clothing exports and almost all the
electronic products come from locations where imported inputs
are given virtual free-trade treatment by one means or another.
The electronics exports come from places with favorable treat-
ment of "multinational" corporations (MNCs) as well. 2!
More than trade theory has emphasized to date, LDCs'
limited capabilities in marketing and related aspects of design
appear to restrict their capacity to export even labor-intensive
consumer and capital goods. -/ LDCs depend heavily on outsiders
to market exports of these products even where production is in
local hands (as is typical, for instance, in footwear and
1/ Natural resource processing has also played a role in several
cases.
2/ Although a few products such as radios are largely produced
by locally-owned firms.
3/ Marketing and related aspects of design are defined here to
include almost all nonproduction activities required to adjust
a product to the tastes of the customer, to profit from
product differentiation, to find buyers and "sell" them on
the product, and to test, package, ship, distribute and
service the product so as to meet the special needs of the
customer. We would exclude from "design", and treat as a
separate activity, as well as technology transfer, adaptation,
and research having to do with production.



- 25 -
clothing). Tailoring production to what the customer needs,
and marketing suitably differentiated products, involves exper-
ience, skills, information and organization that are difficult
for newcomers to achieve. As a rule LDC firms compete mainly
in markets for consumer goods where a cost-cutting strategy can
be effective, for example, in selling to wholesalers for distri-
bution to people (in LDCs or MDCs) too poor to be concerned with
brand names. They may also furnish production capacity to MDC
firms that know how to compete in markets characterized by brands.
and administered prices. Even marketing consumer goods exports
with help from MDC buyers involves a learning process that is not
easy. It is striking that a large majority of LDC exports of
clothing come from just three countries in this group--Hong Kong,
Republic of Korea and Taiwan--perhaps because they have accumulated
the necessary information links and experience in getting the
product together and delivering reliably on time, even though
several of the other LDCs that are trying to export clothing have
lower wages and all have fewer problems with quotas.
Group II: Large Semi-Industrial Countries
The countries in Group II have developed a substantial
industrial structure based on import substitution and protection
and have gradually dismantled at least some of these policies
over the past ten to twenty years. Spain and Yugoslavia were the
earliest to undertake this shift and have reached per capita
levels of manufactured exports that are comparable to Group I.
However, the share of exports in GNP for these large countries
is considerably below that of most of the smaller, more specialized
economies in Group I.



- 26 -
Most of these countries began to expand manufactured
exports when afflicted with shortages of foreign exchange: in
some cases the benefits of increasing imports may have outweighed
considerations of efficiency in the choice of export sectors.
Over time these countries have managed to develop export markets
for the capital goods, chemicals and other intermediate goods
for which their domestic markets, augmented by exports, provide
sufficient economies of scale. As a result, Spain, Yugoslavia
and Brazil have become the leading LDC exporters of capital
1/
goods.    This is a logical direction for the evolution of com-
parative advantage in large middle-income countries, but success
depends to an even greater degree than in consumer goods on
design, marketing and technical services.
Although the present diversified export pattern of
the Group II Countries (which is discussed further in the next
section) is largely the result of protection and import substi-
tution, it is still a matter of dispute as to whether the earlier,
inefficient policy was a precondition for the present degree of
success in building up exports based on scale economies. This
issue--which is discussed for the Latin American countries by
Diaz-Alejandro (1974)--is of mainly historical interest for the
present members of this group, and for India, which now has a
lony history of inefficient import substitution. However, it is
centrai to the future policies of prospective members of the Group,
such as Egypt, Pakistan and Indonesia.
1/  Fo'Llowed by Taiwan, Republic of  e-La, Argentinat and India.
The role of market size is evident.



- 27 -
The continued success of the Group II countries
depends on a set of cumulative processes somewhat different
from those in the smaller, more specialized economies of Group
I. The potential advantages of large countries lie in their
ability to develop economies of scale based on their home
markets, leading over time to production and exports of machinery
and transport equipment along with standardized intermediate
goods such as chemicals and basic metals. This sequence has
been followed most successfully by Spain, Yugoslavia and Brazil.
Another reason for this pattern of specialization is
that several of the Group II countries have considerable export
earnings from natural resources, contributing to relatively high
wage levels.    This may mean that like other resource-rich
countries before them (the United States, Canada, Sweden, and
Australia) they will be forced to build their industrial exports
around relatively skill-, capital- and natural-resource-intensive
products, where their wage levels will not be a serious handicap.
Group III: Countries Emerging from Primary Specialization
The countries shown in Group III all exported relatively
insignificant amounts of manufactures in 1960 but have achieved
significant increases in the past fifteen years. This has helped
to diversify their exports and as a rule has improved their
prospects for more rapid export growth. The optimal timing of
such a shift into manufactured exports depends in part on the
prospects of (and effects on) primary exports. To accomplish
1/  This is particularly true in Argentina, Spain and Brazil, and
is becominq true of Mexico, as a result of new oil discoveries.



- 28 -
this shift through the market mechanism has proved difficult
for countries rich in natural resources whose current exchange
rates are based on this pattern of specialization. Conversely,
countries in this group that still have rather low wage levels
by comparative standards, such as Colombia, the Philippines and
Thailand, find this shift easier.
Within Group III, Malaysia, Colombia and Thailand
have managed to stimulate the growth of manufactured exports
without excessive interference with markets, through subsidies
or otherwise. The Philippines and Chile have shifted policies
sharply in recent years in pursuit of this goal.
In Iran and Venezuela industrial exports are almost
inevitably energy- or capital-intensive, while the poorest
Group III countries are pushed more in a labor-intensive direc-
tion. In between, there is not a compelling need to stress
labor-intensive exports.
Group IV: Large Poor Countries
This group includes the principal countries with
income levels below $300 per capita that have achieved a signi-
ficant volume of manufactured exports or have the industrial base
1/
to do so in the near future.    They also include 75% of the
population of all countries with per capita incomes of $280 or less
in 1976, and the possibility of their benefitting more fully from
1/ Haiti, Kenya, Sri Lanka and Vietnam have some potential but
their actual exports of manufactures are still very small.



- 29 -
international trade is therefore of great concern to world
development. This is a heterogeneous group. At one extreme,
India has a long history of industrial development and a diver-
sified production capability; at the other, Bangladesh has litLle
industrial capability beyond processing of jute, and its manu-
factured exports are almost wholly jute textiles. Limited
supplies of foreign exchange have hampered the growth of India,
Pakistan and Egypt over most of the past fifteen years, while
Bangladesh is heavily dependent on external aid.
The industrial base for exportinc- a variety of manu-
factured goods has already been   'ablished in India,   ypt and
Pakistan, but these countries have only followed policies favor-
ing manufactured exports for brief    iods (Pakistan in the late
1960s, India in the last several years). In addition to th
comparative advantage deriving from low wages, all three (and
particularly India) are in a position to benefit from the economies
of scale inherent in a large domestic market. As shown ir the next
section, the composition of Indian exports shows some changes in
this direction in recent years.
B. Changing Composition of Manufactured Exports
Compared to manufactured exports from industrial
countries, those from LDCs contain a much smaller share of capital
goods and consumer engineering products, and a larger share of
clothing, footwear and textile.   This is a familiar reflection
of differences in skilled vs unskilled labor requirements and the
technical complexity of tile goods involved. However, only a
small number of LDCs ir. Groups I and II have been particularly



30 -
successful in exporting finished consumer goods, including
clothing and footwear, let alone capital goods.
Table 8 shows the product composition of manufactures
in seven main categories, chosen on the basis of marketing
characteristics as well as technological and factor requirements.
I                                    ~~~~~~~~~~~~1/
Only in the principal countries in Group II --Spain, Yugoslavia,
Brazil and Argentina--do capital goods comprise more than 12% of
manufactured exports, compared to the 32% average in MDCs. Capital
2/
and consumer goods together comprise over 40% of exports only
in Hong Kong, Taiwan, Korea, Spain, Yugoslavia and Brazil, which
are also the six largest LDC exporters of manufactures. In
almost all leading MDCs, this share is over 40%.
In countries with limited exporting experience--
particularly those of Groups III and IV--exports of manufactures
consist largely of textile yarn and fabrics and other standardized
intermediate goods, such as leather, plywood, cement, steel and
assorted chemicals.  These products have widely recognized
objective standards and descriptions and can be marketed through
existing channels of trade-not unlike standardized primary
1/ Apart from Ivory Coast, which assembles trucks and heavy con-
struction equipment for partners in a regional customs union.
In this table, capital goods are defined as Standard Interna-
tional Trade Classification (SITC) categories 71 (less 714-9),
722.1, 723.2, 724.9, 726, 731, 732.3, 734.1 and 735; consumer
engineering goods include SITC 724.1, 724.2, 725, 732.1, 864
and 891; clothing and footwear, SITC 84 and 85; textiles,
SITC 65; other clearcut consumer goods, SITC 696, 82, 83
an,' S9 (less 891); standardized intermediate goods, SITC 5
(less 541, 553, 554 and 571), 611, 631, 641, 661, 662, 664
and 67.
2/ Consumer goods here are the sum of three categories in the
table:  consumer engineering, clothing and footwear, and other
clearcut consumer goods.



Table 8: PERCENT COMPOSITION OF MANUFACTURES EXPORTED FROM SELECTED LDCS
AND DEVELOPED COUNTRIES IN 1975
Capital    Consumer       Clothing &   Other Clearcut    Textiles         Standardized       Other and
Country                    Goods    Etineering    Footwear    Consumer Goods   incl. rugs    Intermediate    Miscellaneous
excl. Textiles
Developed Countries          31.8         9.4            2.7              4.0            4.6               24.1             23.3
Developing Countries-'       12.5         5.8           21.8              9.8           14.9               16.2             19.0
Group I
Israel                    8.9         1.7            6.9             4.1             3.2                6.4            68.8
Greece                    5.2         1.3           17.8              3.1           17.3               40.1             15.2
Hong Kong                 2.8        11.3           45.7             19.7            9.7                0.7             10.0
Portugal                  9.0         5.9           18.4              2.5           23.0               15.2             25.9
Taiwan                    9.5         9.8           27.8             14.9           15.1                8.5             14.4
Korea                     7.0         5.2           32.4             12.3           15.7               14.7             12.7
Group II
Spain                    23.5         5.6           11.4             8.4             4.6               22.3             24.2
Yugoslavia               25.4         3.1           13.2              5.5            6.1               21.1             25.5
Argentina                18.0         7.8            2.8              4.3            0.3               24.9             41.9
Brazil                   25.4         6.1           12.2              5.0           12.4               21.1             17.8
Turkey                    2.8         0.5           25.2              2.0           33.6               22.6             13.4
Group III
Venezuela                 0.3          -              -               0.7            2.1               40.3             56.6
Iran                      0.7         0.5           10.9              1.0           60.1                6.5             20.3
Malaysia                 11.4         4.5            8.5              3.4            5.1                16.6            50.5
Tunisia                   1.1         0.4           26.7              1.4           11.5               51.8              7.1
Colombia                  6.7         1.2           11.4             12.1           21.9               25.7             20.9
Ivory Coast             20.2          0.9            2.1               -            20.5               26.4             29.9
Morocco                   2.4         0.2           26.6              6.4           29.2                20.9            14.4
Philippines                -          0.4           14.1             25.5            8.7               22.2             29.1
Thailand                  0.9         2.9           16.2              7.4           24.5                16.5            31.6
Group IV
Egypt                     1.0         0.3           22.0              6.5           47.7                13.6              8.9
India                     9.3         1.2           11.2              4.4           30.6                21.0             22.3
Pakistan                  2.0          -             7.2              6.4            66.1               13.1              5.2
Bangladesh                 -          0.4              -              0.3            88.2                9.5              1.6
a/ Countries listed only.
Sources:  Computed from data in UN Yearbook of International Trade Statistics, 1976, based on classifications
explained in text footnote.



- 32 -
cc:-c;'1ities.  Many resource-rich LDCs export relatively capital-
.nltnsive standardized goods.  One also finds exports of less
standardized intermediate goods--for example, electronic compo-
nents--and trade in household articles and equipment, building
materials, etc., in which transport costs and proximity play an
1/
evident role.
Table 9 shows changes in the composition of exports
in leading countries from 1965 to 1975, illustrating the rising
share of capital goods and consumer goods in almost all cases.
This trend helps to confirm that export success has been based
in part on learning special skills involved in marketing and
2/
producing for customer specifications.
Two other points deserve some comment. First, a few
leading Group I and II countries are now exporting technically
complex products, such as supertankers, aircraft and heavy
machinery. Second, in regard to the product cycle, there seems
to have been, if not a shortening of this cycle, a moving up in
1/ LDCs close to MDC markets, such as Mexico or Yugoslavia,
emphasize labor-intensive products in which transport costs
and delivery time afford them an advantage. Transport costs
are also important in trade among LDCs.
2/ The requisite management, marketing and design capabilities
appear to be areas where LDCs are weak compared to MDCs and
MNCs, and may be profitable areas for LDC development efforts.



- 33 -
Table 9: EXPORTS OF CAPITAL GOODS AND CLEARCUT CONSMUER GOODS
AS A PERCENT OF MANUFACTURES EXPORTED FROM LEADING DEVELOPING
COUNTRIES IN 1975 COMPARED TO 1965
Country          Capital Goods          Consumer Goods         Combined Total
1965     1975         1965      1975         1965       1975
Group I
Israel          2.0       8.9         8.5       12.7         10.5       21.6
Greece   a/     7.1       5.2         11.5      22.2         16.6       27.4
Hong Kong-      1.5       2.8         57.3      76.7         58.8       79.5
Portugal        3.5       9.0         12.8      26.8         16.3       35.8
Taiwan          3.4       9.5        19.4       52.5         22.8       62.0
Korea           2.5       7.0         34.0      49.9         36.5       56.9
Group II
Spain     b/   20.4      23.5        21.2       25.4         41.6       48.9
Yugoslavia-    29.9      25.4         19.9      21.8         49.8       47.2
Argentina      14.7      18.0        13.3       14.9         28.0       32.9
Brazil         16.8      25.4           -       23.3         16.8(+)   48.7
Group III
Malaysia       12.9      11.4         15.8      16.4         28.7       27.8
Colombia        5.3       6.7         4.5       24.5          9.8       31.2
Group IV
EgyptL/         0.3       1.0          5.8      28.8          6.1       29.8
India           1.3       9.3         5.5       16.8          6.8       26.1
a/ Reexports are included in 1965 but not 1975.
b/ Exports to centrally planned economies (where marketing is quite different)
were important in Yugoslavia in both years and in Egypt's consumer goods
exports in 1975.
Source: Computed from data in UN Yearbook of Industrial Statistics.



- 34 -
it on the part of some LDCs, as reflected in exports by some
of the Group I countries of such products as "citizens' band"
radios, TV games and pocket calculators, and the latest types
of digital watches, as well as "knockoff" copies of fashion
designers' latest popular clothing styles.-/ One may also view
color television, steel, petrochemicals and even large ships
as "older products" in which leading LDCs have started to com-
pete successfully; and they have already taken over much of
the export market for radios and monochrome TV sets. Large-
scale exports of passenger cars from Brazil and Korea may soon
materialize on the same basis.
1/ Regarding the prcduct cycle as a general phenomenon, see
especially Vernon (1966) and Wells (1972).



- 3S -
III. FUTURE PROSPECTS
Although the recent trends in LDC exports have been
quite encouraging, the question remains whether world markets
for manufactures will support the export needs of the LDCs in
the future. This question leads to such matters as economic
management and protectionism in MDCs, LDC policies and supply
potentials, and the country distribution of increases in exports.
We will comment selectively on these issues in the general context
of the World DeveloEment Report.
A. Effects of World Growth
To illustrate the sensitivity of LDC exports to growth
in the advanced countries, the World Development Report makes
alternative assumptions about the rate of the advanced (MDC)
countries' growth varying from 3.7% to 4.7% a year for the period
1975-1985, with the midpoint (4.2%) as the base case. The
corresponding variations in world trade and in LDC exports are
shown in Table 10. In the aggregate, the latter vary from 5.4%
to 7.4%, proportionately more than the variation in world growth.
The breakdown indicated for major categories of exports
depends largely on the supply response of producers, which is
taken up below. Overall it is assumed that manufactured exports
are highly responsive to demand conditions while primary exports
are assumed to be limited more by supply conditions over this
period.



- 36 -
Table 10: IDLICATIONS OF ALTERNATIVE ECONONC GROWTE ASSUWTIONS
FOR LDC EXPORTS, IIORTS AND OUTPuT CR0W
(Averag5 Anuual Percentage Growth Rates, 1975-85)
LOW Growth   Hh Growth
B_ _e Scario      Scenrio      Scenario
GDP in MDCs                   4.2             3.7          4.7
GDP in LDCa                   5.7             5.2          6.1
World Trade                   6.4             5.7          7.4
LDC Imports                   s.6             4.8          6.5
LDC fxportsr&/                6.3             5.4          7.4
Yual                        3.4             3.1          3.4
Agricultural Product        3.1             2.7          3.2
Noufual Minerals            5.8             5.3          6.6
Nanufactures               12.2            10.2         14.2
.1/ Including s*vices.
Source: WDR Table 32 ad 33 and background projections.



- 37 -
B. Supply vs Demand Constraints
As a result of LDC successes, market penetration and
employment effects have reached a sufficient level in many
products so that protectionist responses have been triggered in
MDCs. MDC demand growth has therefore become a major concern,
and demand limitations along with supply will shape LDC perform-
ance in the future. The seriousness of protective pressures and
of MDC demand constraints will depend to some extent on how many
LDCs are highly successful at the same time. Nevertheless, growth
of LDC exports of manufactures will continue to depend heavily
on supply performance almost regardless of the treatment of these
exports or the economic situation in MDCs.
1. Supply Potential. In the long run, policies will
be decisive in shaping LDC export performance: not just trade
and exchange rate policies but how well development is managed
on all fronts, how fast industrialization moves forward, and
whether countries become internationally competitive. Since it
takes time for better policies to produce these effects, there
are only a few candidates to add to the successful manufacturing
exporters in Groups I and II within the next decade.
In terms of our previous classification, only one or
two Group III countries (notably the Philippines at this moment)
and perhaps India in Group IV now pursue policies that might
increase their manufactured goods exports to the higher levels
of Groups I and II.



- 38 -
There is al¶ uncertainty in regard to what products
LDCs can develop next. The Republic of Korea and Taiwan are
making an intensive effort to build up exports of capital goods
and other complex prod.cts, spurred by restrictions imposed on
their simpler exports.  As our earlier comments suggest, this
will be a difficult shift. Another qiestion is, how fast can
Brazil, India, Mexico, Argentina, etc., expand their manufactured
exports, if they tilt policies further in this direction, given
the domestic orientation of their production capacity, infra-
structure, and experience? A third question is, how fast can
exporting and marketing know-how be built up in LDCs that are
now pnly minor exporters?
More generally, LDC exports are no longer expanding
from such a low base, s6 high growth rates may be difficult to
maintain in the countries that have already reached relatively
high levels.
2. erket Penetration. LpCs now have a substantial
share in MDC markets for manufactured imports, as Table 11
illustrates. In telation to MDCs' total consumption of manu-
factures, the LDC market sh4re remains small except in a few
narrow product categories. Nevertheless, at the margin LDCs
are making a signifiqant impact on MDC demand for manufactures.
Table 12 shows that from 1970-75--partly as a result of the
recession at the end of the period--increased imports from LDCs
supplied about 7% of the increase in MDC consumption of manu-
factures. This marginal share is projected to remain over 5%



Table 11: IMPORTS FROM LDCS AS A PERCENT OF DEVELOPED COUWTRY IMPORTS IN
SELECTED MANUFACTURES, 1975
U.S.    Japan    EEC (exclu4in&        All Developed
inter-trade)    (in*l. Inter-trade
Clothing                   81.0     65.3          74.1                 44.6
Footwear                   61.5     53.1          61.7                  34.5
Textiles                   43.3     42.0          50.5                  18.9
Electrical Machinery       42.4     23.8          15.8                  12.0
Nonelectrical Machinery    6.5        7.1          7.9                   3.0
Transport Equipment         1.7       2.4         12.3                   2.3
All Manufactures      19.9      20.1          21.4                  9.4                      1
Source; OECD, Trade by Cornmodities, Hark*t Suaries;  Imports  Vol. I, 1975.



Table 12: LDC EXPORTS OF MAMUFACTURED PRO1WCTS AS A SHARE OF MARKETS
IN INDUSTRIALIZED COUNTRIES IN SeLEcTED YAS FROM 1960 to 1985
(Percent)
Share In Market Growth
1960   1970    1975   1985   1960-70   1970-75   1975-45
In laports          5.9     5.8    8.9   13.6        5.8       18.6       17.5
In Consuaption      0.4    0.7      1.2    2.7       1.0        7.1        5.4
Source: WDIR Table 26.                                                                       o



-   -~41-
through 1985. The total share in consumption is projected to
rise from 1.2% in 1975 to 2.7% in 1985.
In particular products, market penetration is
considerably higher. For example, in textiles and clothing
together, by 1975 the share of consumption reached 5%, varying
in the largest MDCs from 8% in the Federal Republic of Germany
to 2% in France. In the European Community as a whole, imports
from LDCs supplied 55% of the market for men's woven shirts
and 45% for women's blouses, as well as 22-26% in five other
major products.   In the U.S. in the same year, imports from
all sources (mainly from LDCs) supplied 56% of the market in
women's sweaters and over one-fourth in a number of other types
2/
of clothing.    Imports from LDCs also supplied roughly half the
radios and monochrome television sets purchased by U.S. consumers.
3. Effects of Protection. This successful market
penetration has contributed to the steady decline of some of the
industries most affected, particularly those in which demand is
growing slowly (clothing and footwear).  Repercussions on prices,
profits, and employment have hurt all the more because of depressed
business conditions and high unemployment in MDCs. Imports from
1/ Based on unpublished EEC data.
2/ Gloves and mittens, men's woven shirts, and knit blouses,
woven shirts and blouses, trousers and slacks, coats, and
body support garments for women. Based on data from U.S.
Department of Commerce.



- 42 -
LDCs (and Japan and CPEs) have become a lightning rod for
feelings caused by other factors--automation, technical progress,
e.i.petition among MDC8, etc..
One result has been greatly intensified pressure for
protection. In some industries--clothing, textiles, footwear,
steel--these pressures have succeeded in creating new protective
measures across a wide front. In others the pressures have been
generally resisted so far, but scattered inroads have been made.
The overall outcome still hangs in the balance, with efforts
being made to stem the tide and add new international agreements,
as a result of the Tokyo Round.
Alongside protection, subsidies to industry have reached
unprecedented proportions in MDC8 and are now a major obstacle
to LDC exports, especially in Western Europe. Shipbuilding,
steel and textile products are among the sectors most affected.
The direct impact of recent increases in protection
on LDC exports of manufactures will be greatest in clothing, textiles
and footwear. These products comprised over one-fourth of LDC
manufactured exports to MDCs in 1975, and (as Table 7 shows) they
are especially important for the poorer and less industrialized
LDCs.
The most severe new restrictions have come in textile
products, with the European Community leading the way. Principal
1/ Actual labor displacement by LDC imports appears small compared
to that resulting from these other influences, based for
example on estimates in Wolter (1977), Frank (1977) and other
objective studies.



- 4? -
suppliers have been forced to accept quotas for 1978 in some
products that are below 1976 levels, and most of the new quotas
grow through 1982 at low rates in the range from 0.5 to 4% per
year. Not only have all the significant LDC suppliers in each
product been hit by quotas but very low "trigger levels" have
been set for adding quotas against other suppliers if they start
1/
to be successful.    New, more severe import restrictions have
also been created in Canada, Australia, Sweden and Norway. The
United States, which has had an effective system of quotas up to
now, has imposed a one-year standstill in its quotas on Hong Kong,
Taiwan and Korea, holding them in 1978 at 1977 levels; their
aggregate quotas through 1982 will allow for a combined growth
rate of just 4.5% per year over 1971 levels in their textile
2/
product exports to the U.S.    The combined result will be to
slow sharply the growth of LDC textile and clothing exports,
which had kept pace with the overall growth of LDC manufactured
exports up to 1976.
To determine the net effect of these changes, the
World Development Report makes illustrative projections of what
1/ For example, in each of eight "sensitive" products of Group I,
which account for over 60% of EEC imports from LDCs by weight,
a quota will be created against any supplier whose exports
come to exceed 0.2% of EEC imports from LDCs and CPEs combined.
In Group II (most other major products) this threshold share
is 1.2% to 1.5%.
2/ Based on recent agreements and data in U.S. International Trade
Commission (1978).



- 44 -
1975-1985 growth of manufactured exports might be like. This
growth pattern is compared to 1970-1975 in Table 13. Growth
of clothing exports is projected conservatively at 5.5% past 1976
and textiles at 4.5%, as a result of recent increases in protection
in MDCs.
Using these projections as a starting point, we get a
sense of the quantitative importance of protection, at least in
clothing and textiles. If trade barriers in textile products were
eased to the point where exports of clothing grew at 15% and
those of textiles at 12% from 1975-1985, the overall growth rate
of LDC exports would increase from 6.4% to 6.9% and that for
exports of manufactures, from 12.2% to 13.5%.
4. Prospects for Trade Among LDCs. The upsurge of
protectionist pressures in MDCs intensifies interest in the
possibilities for accelerated growth of trade among LDCs.
This trade is already growing at well over 10% per
year, and the politically realistic possibilities for regional
common markets, shared industrial plants, etc., are largely
being exploited. The only way that countries that are far apart
are likely to open their markets to one another further is
through reductions of protection in major LDCs on a most-favored-
nation basis.  This could come about through unilateral shifts
in policies or through reciprocal concessions in future
negotiations with MDCs, for example in exchange for reduced NTBs.
Moderated protection in major LDCs and reduced NTBs in MDCs both



- 45 -
Table 13; PROJECTED GROWTH A£ND CHANGING COHPOSITION OF LDC EXPORTS
OF MANUFACTURES By BROAD PRODUCT GROWS, 1975-85
Averts* Growth Rates      Percent of Total
(Z p.-. in 1975 prices)
Product Group               1970-75    1975-85       1975       1985
Clothing                      20.3         8.3        13          9
Textiles                      17.8         6.2        12          7
Chmicals                      16.5        13.0        10        11
Iron & StAl,                  10.7        14.5         6         7
Machinery & Transport Eqt.    20.3        17.3        23         36
Other                         10.2        10.0        36        30
All Manufactures          14.9        12.2       100       100
Source: WDR Table 27 and background estimtes.



- 46 -
appear highly desirable for LDC growth over the long run, and
would help MDCs as well.
Along with trade liberalization, expansion of trade
among LDCs depends also on continuing progress in particular
markets, for example in making LDCs more competitive in capital
goods. For the latter purpose, better financing of supplier
credits and improved information and marketing are examples of
matters deserving attention. However, it would be unreasonable
to expect a rapid acceleration of the growth of trade among
LDCs except as a result of a general liberalization of their
import restrictions.
C. The Changing Role of Manufactured Exports
In the postwar period the possibility of exporting
manufactured goods to the more advanced countries has done much
to make possible the emergence of a number of transitional
countries that are now approaching the status of mature industrial
economies. Although many of them are small, the group also
includes several large countries--Spain, Yugoslavia, Brazil,
Mexico, Korea. Their graduation into the ranks of the MDCs
could potentially make a substantial change in the world trade
outlook for the remaining LDCs, yet most of them will still
have incomes and wages below the MDC average for a long time
to come. In this concluding section we will speculate on the
likelihood that manufactured exports can facilitate the growth
of poor countries and the changes in international policy that
may be required if this is to take place.



- 47 -
Success in exporting manufactures--unlike minerals--
has come largely from a skilled and hard working labor force and
efficient development policies. There appears to be a strong
element of learning by doing, which underlies the concentration
of manufactured exports in a small number of countries. t ce
countries have acquired this ability, it seems to offset rising
wages for a considerable period and makes it possible to retain
their shares of markets in which they would otherwise be losing
their comparative advantage.
This cumulative aspect of export performance and the
increasing number of successful competitors may make it increas-
ingly difficult for newcombers to get established in the sectors
in which they have a comparative advantage. Even if transitional
countries make room, expansion of exports from a few successful
LDCs could swallow up most of the opportunities, leaving too
little for the rest of the LDCs.
If we assume a second-best world in which the advanced
countries are only willing to move away from unprofitable types
of manufacturing rather slowly, there is a strong argument for
favoring the poorest countries in the limited number of sectors
where they are likely to become efficient exporters. Where the
export market is already being parceled out by quotas--which
cannot be avoided in textiles and some agricultural products--
it may be necessary to redesign quoltas to discriminate systamaticall
in favor of the least developed and poorest LDCs, if not others



- 48 -
that need exports most.
However, there are also potential dangers for LDCs in
a systematic attempt to regulate the pattern cf trade:
(1) Competition to export successfully, under the
present system, stimulates LDCs to higher standards in their
policies and economic management, both at the national level
and at the level of the firm. Would a less competitive approach
do this and lead to desirable learning effects?
(2) Any change away from a system of buying goods
abroad wherever they are most satisfactory seems certain to
diminish the benefits from trade and make buyers less satisfied
with imports from LDCs. Is this cost worth the gain to poor
countries?
Since the existing trading system has managed to
accommodate the rapid growth of LDC manufactured exports up to
now, it may not be easy to strike a balance of these issues.
However, these are questions that should be of great concern
to the developing countries in discussing a New International
Economic Order.
1/ To some extent "textile" quotas already operate in the
direction suggested, but they strongly aiscourage new
exporters from getting started, no matter how badly the
exports are needed. In the aid field, donor policy has
already been moving in this direction, but there has
_ee- less evolution of trade criteria appropriate to
the transitional status of the middle income countries.



REFERENCES
Bela Balassa (1977a), "Export Incentives and Export Pi.-
in Developing Countries:  A Comparative Analysis", WorLd E.
Staff Working Paper No. 248, January 1977.
Bela Balassa (1977b), "A 'Stages' Approach to Comparative
Advantage", World Bank Staff Working Paper No. 256, May 197".
Hollis Chenery, "Targets for Development", in Barbara Ward, ed.,
The Widening Gap, New York, 1971, pp. 27-47.
Hollis Chenery, 'Transitional Growth and World Industrialization
in Bertil Ohlin, Per-Ove Hesselborn and Per Magnus Wijknan, ed
The International Allocation of Economic Activity, London,
1977, pp. 457-490.
Hollis Chenery and Moises Syrquin, Patterns of Development,
1950-1970, London, 1975.
Carlos F. Diaz-Alejandro, "Some Characteristics of Recent Export
Expansion in Latin America", in Herbert Giersch, ed., The
International Division of Labour: Problems and PersoeFtlves,
Tubingen, 1974, pp. 215-235.
Juergen B. Donges and James Riedel, "The Expansion of Manufactur
Exports in Developing Countries: An Empirical Assessment of
Supply and Demand Issues", Weltwirtschaftliches Archiv, 1977
Band 113, Heft 4, pp. 58-87.
Charles R. Frank, Jr., Foreign Trade and Domestic Aid,
Washington, D. C., 1977.
Herbert Giersch, ed., The International Division of Labcur,
Problems and Perspectives, Tubingen, 1974.
Gerald K. Helleiner, "Industry Characteristics and the
Competitiveness of Manufactures from Less-Developed Coun.
Weltwirtschaftliches Archiv, 1976, Band 112, Heft 3, FP :
Seev Hirsch, "Hypotheses Regarding Trade between Developing
and Industrialized Countries", in Herbert Giersch, ed., T-.
International Divisicn of Labour:  Problems and Pers:etc
Tubingen, 1974.
Gary C. Hufbauer, "The Impact of National Characterist-`-s _nf.
Technology on the Commodity Composition of Trade i;- -. .f  a.
Goods", in Raymond Vernon, ed., The Technology Fact:;, ;,
International Trade, New York and London, 1970, pp. 145- :



-2-
Alfred Maizels, Exports and Economic Growth of Developing
Countries, Cambridge, England, 1968.
Michael Michaely, "Exports and Growth:  An Empirical
Investigation", Journal of Development Economics, 1977,
Vol. 4, pp. 49-53.
Robert M. Stern, "Testing Trade Theories", in Peter B. Kenen, ed.,
International Trade and Finance: Frontiers for Research,
Cambridge, England, 1975.
U.S. International Trade Commission, The History and Current
Status of the Multifiber Arrangement, Wassington, D. C., 1978.
Raymond Vernon, "International Investment and International
Trade in the Product Cycle", Quarterly Journal of Economics,
May 1966, Vol. 80, pp. 190-207.
Louis T. Wells, Jr., ed., The Product Life CyEle and
International 'rrade~, Boston, 1972.
Frank Wolter, 'Adjusting to Tmports from Developing Countrios",
in Herbert Giersch, ed., Reshaping the World Economic Order,
Tubingen, 1977, pp. 97-130.



RECENT PAPERS IN THIS SERIES
No.                       TITLE OF PAPER                               AUOR
284     Pakistan:  Forestry Sector Survey                       S. Draper, A. Ewing.
J. Burley, G. Grayum
(consultants)
285     The Leisure Cost of Electric Power Failures             M. Munainghe
286     Shadow Pricing and Power Tariff Policy                  M. Munasinghe, J. Warfo.
287     Wages Capital Rental Values and Relative                S. Guisinger (consultan,
Factor Prices in Pakistan
288     Educational Reform in the Soviet Union:                 I. Blumenthal,
Implications for Developing Countries                  C. Benson (consultants)
289     Petroleum and Gas in Non-OPEC Developing                R. Vedavalli
Countries: 1976-1985
290     Major Reforms of the Swedish Education System           A. Heidenheimer (consult
291     Industrialization, Technology and Employment -         T. Rawski (consultant)
China
292     Development and Income Distribution - Zambia            C. Blitzer
293     World Potash Survey                                    W. Sheldrick, H. Stier
294     The Economic Dimensions of Malnutrition in             M. Selowsky
Young Children
295     The Technology of Rural Development                    J,P. McInerney (consultar
296     The Financial Cost of Agricultural Credit:             C.D. Datey (consultant)
A Case Study of Indian Experience
297     Agricultural Sector Planning Models:                   A.C. Egbert
A Selected Sum-ary and Critique
298     Textbooks and Achievement:  What We Know               S.P. Heyneman.
J. Farrell,
A. Sepulveda-Stuardo
(consultants)
299     An Economic and Social Analysis of the                 C. Bruce
Chao Phya Irrigation Improvement Project II            Y. Kimaro



.ITLE OF PAPER                              AUTHOR
300     Two-Studies of Development in Sub-Saharan Africa    S. Acharya
B. Johnston (consultant)
301     The Int.rmadiate Sector, Unemployment, and The        W.F. Steel
Employment-Output Conflict:  A Multi-Sector Model   Y. Takagi (consultant)
302     The Economic Theory of the HOusehold and Impact       D. Chernichovsky
KsAuuruseat of Nutrition and Related Health
Programs
303     Traae Rtstrictions and International Price            M. Bale
Instability                                           E. Lutz
304     Intergvernamital Fiscal Relations in Developing    R. Bird  (consultant)
Countries
305     4 Programming Approach to Fertilizer Sector           A. Choksi
Planning                                              A. Meeraus
306     The Foreign Exchange Gap, Growth and Industrial       K. Dervis
Strategy in Turkey:  1973-1983                        S. Robinson
307     The Importance of Risk in Agricultural Planning       P. Hazell, R. Norton
Nhdel-                                                M. Parthasarathy.
C. Pomareda (consultant)
308     Guidelines for School Location Planning               W. Gould (consultant)
309     Growth and Poverty in Developing Countries            M. Ahluvalia, N. Carter
H. Chenery
310     Teacher Training and Student Achievement in           T. Husen, L. Saha
Less Developed Countries                              R. Noonan (consultants)
311     Optimum Economic Power Supply Reliability             M. Munasinghe
M. Gellerson (consultant)
312     Intra-Industry Trade and the Integration of           B. Balassa
Developing Countries in the World Economy
313     Export Promotion Policies                             B. de Vries



I