World Bank Reprint Series: Number Fifty-four
Larry E. Westphal
The Republic of Korea's
Experience with Export-Led
Industrial Development
Reprinted from  W/Vorld Developmnent 6 (1978)



The most recent editions of Catalog of PubIllicationis, describing the full range of
World Bank publications, and World Bank Research Program, describing each of the
continuing research programb of the Bank, are available wvithout charge from:
The World Bank, Publications Unit, 1818 H Street, N.W., Washington, D.C. 20433
U.S.A.
WORLD BANK BOOKS ABOUT DEVELOPMENT
Research Publications
International Coinparisons of Real Product and Purchasing Power by Irving B. Kravis, Alan
Heston, and Robert Sumnmers, published by The Johns Hopkins University Press,
1978
Experiments in Fanmfilia Planniiing: Lessons from the Devclo pin7g World by Roberto Cuca and
Catherine S. Pierce, published by The Johns Hopkins University Press, 1978
Income Distril7lbtion POliCil in the Developing Countries: A Case Stuldy of Korea by Irma
Adelman and Sherman Robinsson, publish-ed by Stanford University Press (in the
Commonwealth, Oxford University Press), 1978
Interdependence in Planniying:  utilevel Programming Studies of the Ivory Coastby Louis M.
Goreux, published by The Joliim fiopkins LJniversity Press, 1977
The Mining Intdinstiy atnd the Develo ping Countries by Rex Bosson and Bension Varon,
published by Oxford University Press, 1977
Patternis in HoIiseh!old Demand and Saving by Constantine Lluch, Alan Powvell, and Ross
Williams, published by Oxford Univer-sity Press, 1977
Uniskilled Labor for Developinnei t: R,, Economn ic Cost by Orville McDiarmid, pub]lislhed by
The Johns Hopkir-ns Univeersity Press, 1977
Electricity Economics: Essays antd Case Studies by Ralph Turvey and Dennis Anderson,
published by The Johns Hopkins UTniversity Press, 1977
Housing for Lowl-Income Urban Faililes: Economics and Policy in the Developbin  World by
Orville F. Grimes, Jr., published by The Jolhns Hopkinis University Press, 1976
Village Water Supply.; Economic.c and Policy in the Developing WVorld by Robert Sauinders
and Jeremy Warford, pLublishea by The Johns Hopkins University Press, 1976
Economic Analysis of Projfeds bv Lyn Squire and Herman G. van der Tak, published by
The Johns Hopkins University PFess, 1975
The Design of Rural Dcvipret.X;t Lsons fromn Africa by Uma Lele, published by The
Johns Hopkins Universit y I'rsq, 1375
Econoinit-Wide M4odels and DeVIelOPMeint Pi'.innimmg edited by Charles R. Blitzer, Peter B.
Clark, and Lance T<elor, ypubliTihed by Oxford University Press, 1975
Patterns of Development, 1 /950 J70 1hy iloillis Clhenery and Molises Syrquin with Hazel
Elkington, published by Oxford University Press, 1975
A Systeimi of biternational oonmyai   of Grvoss Product and Purchasng Powver by Irving B.
Kravis, 7oltan KOeRnese) ,i .X it f le1e;, to nd Robert Sui miimners, publishled by 'Ihle Johns
Hopkinls University P're:;s, 1".':
Country Economic RepsLrts
CoMMovnealthl Cariblbcnli: Thc11 It. tcratiotn EApcrience by Sidney E. Chernick anid others,
published by The Johns Hclpkinsi University Press, 1978
(continted on inside back cover)



ERRATA
The Republic of Korea's Experience
with Export-Led
Industrial Development
Page 365, table in text, "Direct contributions to growth of
manufactured output:"
The contribution of import substitu-
tion for the period 1963-68 should be
3.7%, not 33.7%.
Page 370, Table 15, "Factor use in manufacturing:"
Value added in 1970 should be 1,803.0.



World Development, 1978, Vol. 6, No. 3, pp. 347-382. Pergamon Press. Printed in Great Britain.
The Republic of Korea's Experience
with Export -Led
Industrial Development
LARRY E. WESTPHAL B
Thle World Bank
Summary - This paper is a summary of findings from the author's research conducted
intermittently over the past seven years and reported elsewhere in greater detail for more
specialized audiences, It begins with a brief history of the J epublic of Korea's industrial incentive
policies, followed by an examination of historical trends in the real effective exchange rates for
exports and imports. Estimates of nominal and effective incentive rates for 19(f are
summarized, and the changes that have taken place in the structure of the Republic of Korea's
trade and production are quantified and related to its industrial incentive policies. Tlhe conclu-
sion of the paper is that the Republic of Korea's outward-looking developmer-t strategy and the
policy measures adopted to implement it have resulted in a generally efficient and equitable
process of rapid industrialization.
It is widely known that Korea has achieved         INDUSTRIAL GROWTH TRENI)S
remarkable success as an exporter of manu-
factured  products. Less widely appreciated,       The Republic of Korea, often referred to as
however, are    the  circumstances underlying   South Korea (and here simply as Korea), was
Korea's export performance and the role that    created at the end of World War II by'the
trade expansion has played in its industrial    partition of the Korean peninsula, which had
development. Several major pieces of research   been occupied by the Japanese since the early
into these questions have recently been com-    1900s. Under Japan's colonial administration,
pleted, and the purpose of this paper is to     agriculture lhad been extensively developed in
summarize their findings.' As most of the       the southern half of the peninlsula to promote
research surveyed here was conducted during     exports of foodstuffs to Japan. In turn, there
the  early  1970s, detailed  empirical results
largely pertain co the 1950s and 1960s. How-    *As may be seen in the bibliography, collaborators in
ever, where and as possible, this paper updates  the research summarized here have included Kwang
the analy.sis into the 1970s.                   Suk Kim, Charles R. Frank, Jr., Yung W. Rhee, and
The discussion is organized as follows: first  David C. Cole. The research also benefited immensely
is a short sketch of Korea's industrial growth  from having been conducted in large measure as a
performance, followed by a bri.jf history of    country study under the aegis of two large-scale
pedustrf orm nce,foed by aucb        eings      comparative  analyses: Development Strategies in
n       -   Semi-Industrial Countries, sponsored by the World
tions present time series estimates of the real  Bank and directed by Bela Balassa; and, Foreign Trade
effective exchange rate for exports and sum-    Regimes and Economic Development, sponsored
marize estimates for 1968 of effective protec-  jointly by the United States Agency for Internation;tl
tion and subsidy rates, Following a somewhat    Development and the National Blurcull of EcoIIoic
more detailed appraisal of export perfornmance,  Fesearch, and directed by Jagdish N. Bllagwati and
the paper turns to examine the role of trade    Anne 0. Krueger. Helpful comments have also been
expansion in Korea's industrial development,    received from Benjamiin B. King, Ian Little, and T. N.
H  Srinivasan. This paper is a revision of an earlier version
Here the focus is on contributions to output    which appeared as World Bank Staff Working Paper
growth as well as to improved factor utilization.  No. 249. Neither the World Bank, nor any of its
The paper then concludes with a discussion of   affiliates, nor any of the other sponsors of the
the relevance of Korea's experience to policy-  underlying research, are re:;ponsible for the views
making in other developing countries.           expressed in this paper.
347



348                             WORLD DEVELOPMENT
was little industrial development beyond tradi- output, which was nil in 1955, rose from
tional activities until 1920, when a law requir- roughly 6% in 1965 to nearly 25% in 1975
ing administrative approval for the establish-  [Bank of Korea, National Income in Korea
ment of new firms was abolished. The growth  (1976)], Within a decade, from 1965 to 1975,
of manufacturing further accelerated after 1930  the ratio of total exports to GNP more than
owing to change in Japanese policy that greatly  trebled and the share of GNP originating in the
favored imports from the Yen Bloc. None the  manufactuving  sector more than doubled.
less, by 1940 r-', South had only a small Manufactuied products constituted 42%  of
'modern' manufa- ,i,ng sector, concentrated in  total exports in 1965 and 74% in 1975.
light industries and machinery production.   Nearly  every indicator of development
Most of the heavy industry and more than 90%  performance improved dramatically after the
of the electricity generating capacity were in  mid-I 960s (see Table 1). Thus, for example, the
the North [see Kuznets (1977)].           annual growth rate of real per capita income
Economic activity in the South was domi- increased from 2.1, in the decade preceding
nated until the mid-1950s by adjustments first  1965 to 8.1% in the decade that followed. By
to partition and then to dislocations caused by  1975, Korea's population of over 34 million
the Korean War. The economy's structure in  enjoyed a per capita income in excess of $500
1955 was thus much the same as it had been  in current US dollars. The economy also per-
left at the end of the Japanese occupation. As  formed well with repect to employment and
may be seen in Table 1, manufacturing activity  income distribution. Between 1965 and 1975,
accounted for only 8% of GNP in 1955, while  total employment is estimated to have in-
nearly half of GNP originated in the primary  creased by 3.7% per annum while the popula-
sectors. Due to the disrupting effects of the  tion aged 14 and older was growing at 3.2%
Korean War, exports were but J1.4% of GNP, and  annually. Real wages in mining and manufactur-
manufactured  exports were virtually  non-  ing rose at an average annual rate of 7.8%
existent.                                  during this period.3
Industrial expansion from 1955 through the
early 1960s was largely oriented toward the
domestic market, with import substitution for  INDUSTRIAL INCENTIVE POLICIES
light manufactured and non-durable consumer
goods playing a major role. Exports in 1956  Industrial incentive policies during the last
amounted to less than half their real value in  half of the 1950s were those typically asso-
1950, the year preceding the Koiean War, but  ciated with an import-substitution, strategy
in the late 1950s they began growing at a  Large-scale purchases of won (the domestic
gradually accelerating rate. By 1960 the real  currency) at the official exchange rate by the
value of exports surpassed that in 1950 by  resident UN military establishment provided a
nearly 16%. However, in absolute terms as well  major motivation to maintain an overvalued
as relative to GNP, exports remained small: in  exchange rate, from which a complex structure
current US dollars, total exports (including  of multiple exchange rates evolved to deal with
merchandise and non-factor services) amounted  recurrent balance-of-payments problems. In
to roughly $100 million in 1960, of which only  addition, high tariffs were imposed on imports
5"c were manufactured products.2 The share of  having domestically-produced similars, mostly
total exports in GNP at current prices was only  finished consumer goods, and the government
3.3%                                      increasingly relied upon quantitative import
Manufactured exports rose rapidly in the  restrictions as an additional measure to offset
narly 1960s, albeit from a small base, but tile  the progressively greater overvaluation of the
real 'turning point' in both export and indus-  won. The principal incentive to exports during
trial growth cact; around 1965, during a period  this period came from the multiple exchange
of trade liberalization and other major policy  rate system, under which export earnings were
reforms. In the decade following 1965, manu-  converted into foreign exchange certificates
faCLured export growth coupled with rising  that were traded at a premium in a free market.
domestic demand fueled industrialization much  In addition, modest direct cash subsidies were
faster than before. The compound annual rate  also used.
of growth in the index of manufacturing output  The first half of the 1960s was a period of
was 1 l` from 1955 to 1965; it increased to  social, political and economic instability during
24, from   1965 to 1975. Underlying the    which there were a number of attempts at
acceleration of manufacturing output growth,  policy reform  and economic liberalization.
the share of exports in manufacturing (gross)  Upon taking control in 1961, the military



TIIE REPUBLIC OF KOREA'S EX PERI ENCL' WITI-I INLDUSTR IAL DEVFLOPNI FNT'      349
Table 1. Majoir economnic indicators: 1955 to 1975
Computationt of per capita inicome            1955       1960        1965       1970      1975
GNP (billion won at 1970 prices)              938.2     1,129.7     1,529.7    2,589.3   4,107.7
Population (million persons)                   21.5        24.9       28.3        31.4     34.7
GNP per capita (LIIOUSan1d won at 1970 prices)  43.6       45.3       54.0        82.4     118.4
Percentage shares in GNP at 1970 prices
Value added in:                                %                                 % %  % -7t
Primary production                            47.5        42.6       41.0        29.5     23.0
Manufacturing                                  7.9        10.8       13.9        21.6     31.9
Social overhead*                               4.2         6.0        8.5        13.3      13.7
Services                                      40.4        40.6       36.6        35.9     31.4
Gross investment                               10.0         8.6       12.9        27.2      26.3
Total exports                                   1.4         2.4        5.2        14.7      28.3
Total imports                                  11.2        10.4        9.8        24.8      27.2
Percenitage shares in GNP at current prices     % '                               ,';
Government revenue                             10.5        19.8       16.1        20.1      19.7
Government savings                              0.6         4,1        5.7         7.5       3.6
Total domestic savings                          3.7         1.6        7.7        17.1      17.7
Gross investment                               11.9        10.9       15.1        27.2      27.1
Total exports                                   1.6         3.3        8.5        14.7      30.2
Total imports                                   9.8        12.6       15.9        24.8      39.6
Comzpound anin8al growthl rates            1955 60Cr          1960-65%r 1965- 70%    1970-75.
GNP (at 1970 prices)                            3.8              6.2        11.1        9.7
GNP per capita (at 1970 prices)                 0.7              3.6         8.8       7.5
Manufacturing value added (at 1970 prices)     10.3             11.8        21.3      18.5
Index of manufaclturing output                 12.0              9.5        24.2       23.2
Total exports (at 1970 prices)                 16.3             24.0        36.5      25.0
Population 14 years and older                   -                2.5**       2.8        3.6
Economically active population                                   2.7**       2.9        3.9
Employment
Total                                          -                2.4**       3.5        3.9
Manufacturing                                                  12.2**      10.7  .   11.4
Real wages
Average in mining and manufacturing            5.1t             1.1         7.1        8.4
Agricultural labour                                             0.1         8.2        1.5
Prices
GNP deflator                                  12.0             19.3        13.7       15.3
Wlholesale price index                        10.2             17.2         7.9       18.9
Sources: Bank of Korea, Economic Statistics Yearhook (1967, 1969, 1976); National Income in Korea (1975);
Economic Planning Board, Korea Statistical Yearbook (1975); and Hong (1976).
* Includes constructioin; electricity, gas, water, and sanitary services; transportation, storage and communication.
For IQ5 7- 6f0.
** For 1960 66.
government irnmedi;ttely completed the task,    annual trade programmes, variable tariffs, and
begun under the civilian authorities who had    selective import prohibitions.
replaced  Syngman   Rhee after the student         The   liberalization  philosophy  that had
revolution in 1960, of unify ing the exchanige  einerged but re-mained largely ineffective during
rate. The transition to the unitary rate did not  the early  1960s took  firm  hold after the
appreciably affect export incenitives, since the  election of a civilian govertinment under Chung
rate established in 1961 was somnewhat lower    Hee   Park  in  early  1964. (President Park
than the pre-existing free market rate on export  continues in power today.) The ensuing two
earnings. In 1963, largely as a result of a decline  years brought major and lasting policy changes
in US grant aid, a balance-of-payments crisis led  in a number of areas. In adldition to changes
to the reintroduction of a multiple exchange    more directly related to trade policy, there
rate system, wnile import controls were again   were several key reforms aimed at increasing
tiglhtened by ieans of licensing through semi-  public and private savirgs. The administrationi



350                                WOP.LD DEVELOPMIENT
of government revenue collection was revised to  successive Five-Year Plans as well as from the
insure a high ratio of revenue to GNP, and      same price incnritive  meclhanismns that have
purposive measures were taken to hold down      applied to exportK.5 But in most sectors the
current government expenditures. An increase    only price incenitive to domiestic sales has been
in commercial bank deposit and lending rates    fhe protection potentially affordled by import
together with a price stabilization programme    -,ntrols and  tariffs. Import controls were
raised the real interest rate to apprOxiniately  gradually relaxed following the 1964 devalua-
10% per annum. By sulbstantiall) increasing the  tion as the number of items eligible for inmport
domestic savings   rate (see Table 1), these    increased along with quota amounts. However,
reforms were instrumenta! in finanicirig the   it was not until 1967 that a major step was
higher investment rates required for accelerated  taken in liberali/ing import restrictionsS through
GNP growth.4                                    a switch   from  the so-called  'positive' list
At the same time, policy-makers came firmly  system, under which only those comlnl0dities
to accept that rapid economic development       listed in the trade progiannme could be im-
depended upon an export-oriented indiustriali-  ported. to the 'negative' list systeml, nnder
zation strategy. This view was predlicated on the  which all commodities not listed could auto-
understanding that Korea's natural resource     matically be imported without rcs'riction. Im-
base was very poor and the realization that     port controls continue to this day, thougll they
further opportunities for import substitution   have gradually been liberalized. In turn, the
were only to be found in intermediate and       tariff structure has remained substantially un-
durable goods, where the limited domestic       changegd since its original creation in 1949,
market of the mnid-1960s could not justify      though there have been several minor reforms.
establishing plants large enr)Lugh to realize teclh-
nological economies of seale. A major asso-
ciated policy chiange was the lasting establish-       REfAL IFl"l(TIV\'l EXCHANGF
ment of a uniform exclhange rate in 1964, wlheni                   RArT,S
the official won - dollar exchange rate was nearly
doubled for the -,eCond time in three years.       An incentive systemn that generally f.avoured
As had been the case in the earlier devalua-  exports over import slubstitution within manu-
tion in 1960, the devaluation in 19t4 was       facturing gradually came into being over the
aimed prinmarily at simplifying the exchange    first half of the 19t)(s. In this section a
regime and offsetting domestic inflation, rather  quantitative assessment of the impact of this
than increasing the level of export incentives. In  change in policy on export incentives is made
fact, exporters had begun to benefit from an    using time series estimiates of real effettive
expanded range of explicit export incentive     exchliange rates. But first, the incentive mecha-
mechaniisms starting in 1959, when tariff ex-   nisms introduced during this period will be
emptions on imports of raw nmaterials used in   descrihdLI.
export production were first granted. Other        The most important incentives to exporters
price  incentive  nmechankrns were gradually    by 1967 included: unrestricted access to and
added, so that by 196-     exporters operated   tariff exemptions on imported intermeLliate and
under a virtual free trade regine. lbenefiting  capital goods: exemption from     payment of
from  free access to imported inputs, indirect  indlirect taxes both  on  major internmediate
tax and tariff exemptions, and reduced Aharges  inputs, whether imported or puMrchased domes-
on overhead inputs as weil as credit and direct  tically, and on export sales. generous wastage
tax preferences. (These niechlnisins are (les-  allowances in dletermining duity and indirect
cribed and analysed in greater detail in the next  tax-free raw material imports, wlich permiiitted
section.) Also, an export ptrformance criterion  the use of some of tlhese irmports in production
was ,stablislied as the basis for granting im-  for the dlom1estic market; redLucedL priLes for
polrters' licences and a ,ystem of export targets  several overhead inputs incltudiing electricity and
for illdividu1a;l "irlus in partikciil:r markets (by  railroad transport, whiclh were intended at least
plroduct andl diestination  was mitodc.l The   in part to com pemate for paynien t of indiriect
conclUding section of the paper discusses the   taxes included in the normal clharges for these
inportance of the latter.                       inputits: a 50%' reduction in direct taxes on
As regards incentives to production     for  incomni  earned in exporting, along with ac-
domestic sale, a small number of imipo,rt-      celerateLl depreciation; and, immediate access
stilubstituting  industries hiave benefited  f'rom  to b,ubsidized short- anid medlium-term cre(it to
lronm(tionnal activities (project idtelltifiCation,  finance wvorking c:apital anil fixedl investmecnt
feasilbilit) studies. etc.) uzndfer the aegis of  respectively.  In  addition,  the  so-callecl



THE REPUBLIC 01V KOREA'S l:X131.R I iNCF WITHI I NDLUS I I [AL D):VI 1(ePIFNT  35 I
'export-import link' system  entitled selected  estimates do not incorporate the subsidies due
exporters to import certain popular items that  to the export import link system, the excess of
were not otherwise approved for import. This   wastage allowances over actual wastage in
system was used to subsidize exports during the  production for export, or price reductions on
late 1950s and much of the 1960s. It has since  overhead inputs. Very little is known about the
been used only intermittently on a more or less  precise magnituide over time of the first of these
ad hoc basis to offset exporter's temporary losses  subsidies. More is known about the subsidy due
due to market fluctuations or entry into new   to wastage allowances, which has been esti-
inarkets.                                       mated to have been equal to 2.4%isX of the total
Some of these incentives are not genuine     value of nmerchaln(lise exports in 1968. There is
subsidies. Except where there is over-rebating,  also evidence that this subsiLly has flutcttuated
the exemption of export sales from   indirect   over time, generally rising and falling in relative
taxes on the same products sold donnestically  value with the degree of the won's overvalua-
merely establishes tax neutrality tinder the   tioio. In turn, the value of overhead price
destination principle that exempts exports from  reductionis has always been quite small in
indirect taxes while such taxes are imposed on  relative termns; it equalled only 0.4", of exports
imports. In turn, exemption from    tariffs on  in  1968. [Westplial and   Kim  (1977), pp.
imported inputs and from indirect taxes on all  3 -39.]
inputs regardless of the source of the purchase   Table 2 shows the total value per dollar of
does favour exports vis-a-vis production for   exports of those incentives for which consistent
domestic sale; so too does permitting exporters  time series information is available. For con-
to have free access to imported inputs, while  venience, the value of incentives is slhown as a
access is restricted in the case of production for  piercentage of the official exchange rate. Some
domestic sale. V"is-a-vis world market prices,  discussion is in order regarding the estimation
however, the effect of thesc measures is sinplsN  of direct tax ancl inttrest subsidies. 'rhe total
to subject exporters to a free trade regime,   subsidy owiing to the 50'7 reductioni in direct
exporters purchase their (tradeablel inputs and  taxes lhad been  allod11.ited as tlIe differenCe
sell their output at worldn market prices.      between tax liabilities in the absence of any
The effective exchange rate for exports is   such preference andi actual direct tax paymenls.
obtained by adding export subsidies per dollar  Sinlilarly, the total interest stubsidy has been
of exports to the official exchange ratL. It   comiputed as the difference between the in-
would therefore appear that the valut of       terest that would have been paid on outstand-
indirect tax and tariff exemptions should not  ing credits of various types, including both
be included in the calculation. However, while  short and medium-term    loans, at the non-
indirect tax and tariff exemptions do not add   prefernlltial :omonmerciall bank lending rate and
to exporters' revenues, they do reduce prodUc-  the interest actually paid. The resulting esti-
tion costs below those involved in production   mate understates credit subsidies because it
for the domestic market. Thus, an increase in   does not reflect the faet that exporters benefit
indirect tax or tariff rates paid to produce a  from  immediate access to credit within very
given product for domestic sale makes its       generous limits, while credit from formal-sector
production for export using existing capacity   financial institutions has been subject, depend-
more attractive, even though it does not in-   ing upon the period, to more or less stringent
crease the profit rate earned on export sales. In  rationing for all other borrowers. Mloreover, the
this sense, inclusion of these exemptions yields  non-preferential commercial bank lentding rate
an index of the incentive to exporting vis-a-vis  is neither an equilibrium rate nora robust proxy
selling d1omestically, but not of export incen-  for the average rate on all producers' outstand-
tives as they affect profits from exporting. Two  ing debt, the latter becauise of -ener:n l recourse
effective exchange rates for exports should thus  among those frustrated by e:r t.dit rationing to
be distinguished: a 'grosss' rate, which includes  an active informal mioney  iuiarket on wlich
indirect tax aind tariff exemnptionis per dollar of  interest rates have been as high as 40 to 501')
exports; and, a 'net' rate, whlich does not     per innUmi in real terms.
include them. In turn, profits from exporting      Because of the way in which the interest
will be affected if such exemptions are grantedi  sllub.idy is nicas:tred, a clhanige in its value per
or withdrawn over time, which makes the gross   dollar of exports does not necessarily imply a
rate the only meaningful measure of incentives  change in thie cost of borrowed capital to
when there are clhanges in the scope of exemp-  exporters, it mav sometimes simiply reflect a
tions.                                         change in the non-prelerential commercial bank
For lack of consistent time series data, our  lendiilng rate. One should thus be careful to



Table 2. Effective exchange rates for exports and imports (Annual averages; won per US dollar)
Purchasing power parity adjusted
As a percent of the official exchange rate              (Real) effective exchange rates*
Per dollar of exports:    Per dollar                      For exports
Exchange                           of imports:                Excluding     Including
premia     Direct     Indirect      Actual                    indirect     indirect
Official    plus       tax         tax         tariffs     Including     tax and      tax and
exchange    direct     plus        and           and          only         tariff       tariff
rate       cash    interest     tariff        tariff      revenue     exemptions    exemptions   For
(Nominal)  subsidies  subsidies  exemptions   equivalents   incentives     (net)        (gross)   imports
%          %1          %            %,                                                      :4
1958        50.0      128.0      2.4          0.0         28.8         276.7        279.6        279.6      156.5   0
1959        50.0      169.4      2.6           -          65.6         321.8        324.9          -        197.9   t
1960        62.5      134.2      2.0           -          60.3         317.5        320.1          -        217.4   e
1961       127.5       17 3      0.8           -          15.3         287.4        289.3          -        282.2   t
1962       130.0        7.9      1.1          7.6         12.6         244.7        247.1        264.2      255.5  tz
1963       130.0      33.8       2.8          9.1         13.9         253.5        258.8        276.1      215.9   o
1964       214.3       19.9      3.2          8.2         15.3         278.7        286.1        305.3      268.0   =
1965       265.4       0.0       3.7         11.1         10.4         265.4        275.3        304.6      293.1    '
1966       271.3       0.0       4.6         14.4          9.3         256.4        268.2        305.1      280.4   i
1967       270.7       0.0       7.4         15.7          9.4         242.8        260.7        298.8      265.4
1968       276.6       0.0       6.6         21.5          9.4         233.2        248.5        298.7      255.0
1969       288.2       0.0       5.1         22.7          8.5         234.3        249.3        299.4      254.5
1970       310.7        0.0      6.7         21.7          8.3         239.9        255.9        307.9      260.0
1971       347.7       0.0       6.5         23.1          6.3         253.5        270.1        328.6      269.7
1972       391.8       0.0       3.2         23.7          6.0         275.0        283.8        348.9      290.2
1973       398.3       0.0       2.2         21.5          4.9         320.6        327.6        396.5      332.5
1974       407.0       0.0       2.1         19.1          4.5         279.2        285.1         338.4     288.1
1975       485.0       0.0       2.7         14.0          5.1         275.0        282.3         320.9     286.6
Source: Westphal and Kim (1977), Tables B.1 and B.2.
*At 1965 prices.



TIE REPUBJ_ '   -' YOI:A'S lXPIRIAT:NC  WIT11 INI)DLSTRIAL DI VI:LOPMNTI'      353
distinguiish between the effective exchange rate   There is unfortunately no inforniation on
in which the interest subsidy is as calculated  the value of tariff exemsptions between 1959
above and    the rate in whlich the interest    and 1961. Thus, using the net rate for these
component reflects only (changes in) the in-    years and the gross rate for the remaillilng years,
terest rate paid by exporters. However, only the  one may conclude only that the average level of
first of these rates will be consideredl here. In  the gross real effective exchange rate for ex-
turn, preferential direct tax treatment is a    ports was somnething in excess of 295.5 between
subsidy only insofar as there are profits to be  1958 and 1965. The 'weragc gross rate was
taxed, so that the magnitude of the subsidy     302.0 between 1966 and 1970. However, the
depends upon the profit rate per dollar of      rise in the overall indlLceenieiit to export result-
exports. All of this is to say that our net     ing from the incentive policies adopte(i over the
effective exchange rate, which includes cx-     first half of the 1 960s was certainly greater
change prerr,;a and direct cash subsidies as viell  than indicated by this comparison. In the first
as direct tax plus interest subsiidiCs, is an   place, the gross real effective excliange rate for
imperfect index of export incentives as they    exports exhibits wider fluctuations before 1965
affect profits from exporting, becatuse the latter  than for the following five years. Thus, its
subsidy  component may      fluctuate  withoLit  average value in 1962 and 1963, prior to the
there having been any change in either the      devaluation upon the return to a unified ex-
direct tax rate on income earned from export-   change rate, was only 270.1. In addition, our
ing or the interest rates paid by exporters.    estimates neglect two incentive inecliasnisms
Only real effeetive exchange rates are shown  that became   important in   the mid-1960s,
in Table 2. These have been obtained from the   namely wastage allowances and the export
corresponding nominal rates in the following    inmport link system, though the latter was
manner. Assuming that the prices of a eountry's  unimportant after about 1967. Because of the
exports move in parallel with foreign price     understatunicint note(i earlier, they also fail to
movementb, multiplying a nominal effective      .n1dicate adleqjuately the increasc in credlit sub-
exchange rate by an index of prices in overseas  sidies arising fronm the relaxation of credit
markets yields the number of won ih, current    ceilings for exporters that took place in tlle
prices received per dollar of exports, the latter  early and mid-1960s.
in constant prices.6 Thlen, deflating the figure   But most imiiportanit, the comparison ne-
so obtained by an index of domestic prices      glects the simultaneous reduction in incentives
gives the number of won in constant prices      given to import comrpeting production. includ-
received per dollar of exports also in constant  ing the gradual relaxation of import controls.
prices, i.e. a p)urchasing power parity adinLsLedl,  For, it is of course not the level of export
or real, effective exchange rate for exports.   incentives per se that shloUld concern us, but
Table 2 allows the gross andi net rates to be   ratlher thieir level in relation to the incentives to
compared with the rate including only those     produce for dLoMestiC sale. Table 2 also gives
incentives which directly affect revenues, i.e.  estimates of the real effective exchange rate for
exchange premia and direct caslh subsitdies.    irnports, the nominal value of which equals the
Exporters first obtained tariff exenmptions  official exchange rate plus average customs
on imported internmediate inputs in 1959; in-   duties and tariff equivalents (e.g., premia paid
direct tax exemptions. in 1962; and, tariff     to purchase foreign exchange certificates from
exenmptions on imlported capital goods, in 1(166.  exporters) paid per dollar of implorts.7  Un-
Thus, for appraising trends in export inLentives  fortunately, tlhoulgh., the effective exchange rate
before 1966, as well as for comparisons of      for imports is a virtually meaningless inidicator
incentives before 1966 with those after 1966,   of protection from imports due to its failuire to
the gross real effective exchianige rate for ex-  reflect the effect of import controls. F\cepting
ports is the only relevant indlicaLtor, since its  estimiiates for 1968  (discussed in the next
value alone reflects the graniting of exemptions.  section), there is no solid evitdenice concerning
The nlet rate is the super ior indtex for examining  the inmpact o(f imnport conitrols, let alone how it
trendLs after 1966. sinice there have been IIo  lhas chaUnged over time. But it is nonetlheless
suibsequ ent clhanges in the scope of exenmptiin.,  apparent that import controls have been pro-
That is, after 1966, variations in the relative  1!ressively lillerali/ed silnce 1064.
value of exemptions, which affect the gross but    Owing first to the sale of exporters' foreign
not the net rate, reflect only changes in tariff  exchange receipts in a free mlarket, and then to
and indiirect tax rates fromii which exports were  deaulaationi in 1961, export inicentives were by
already exemnpted, including the effect of silifts  no means lacking in the late 1950s anid early
in the composition of exlports.                 196 0s. In fact, it wvill be recalled that exports



35e,                               WORLD DEVELOPMIE;NTI
began growing at a modestly rapid pace in       the Korean    experience between export in-
1957. While the growth of exports up to         centives and the growth of exports. Nonethe-
sometime around    1960 may be interpreted      less, efforts to 'prove' the relationship statisti-
simply as a return to the situation prevailing  cally have not yielded notably robust results.9
before the Korean War, it is notable that this  However, based on regressions of real exports
did not take place in the absence of incentives.  against the gross real effective exclhange rate
Exports increased rapidly between 1960 and      and real non-agricultural output, Frank, Kim
1965, at an annual conmpoundi rate of '4', in   and Westplhal (1975, pp. 85 86) conclude that
real terms (see Table 1). But in the five years  '. . .  the responsiveness of [manufaCtUred]
following  1965, the real growth of exports     exports changed sharply after 1963 .... Before
accelerated to 37'r per annum.                  1963, sensitivity to [effective] exchange rate
Relative to its average over the preceeding  policy was lacking . . .'. This change appears to
five or six years, the net real effective exchange  be due to two main factors: the government's
rate for exports increased nearly 30', between  intention  to  stabilize  exporters' profits at
1970 and 1973, principally as a result of the   relatively high rates was probably only elearly
appreciation of the Japanese yen. In turn,      perceived in 1963 or shortly thereafter, while
exports in 1973 were more than two and a half   the general prodluctive capacity of the economy
times their real val.ae in 1970. In response to  did net begin expanding rapidly until after that
these and other, less transitory indications that  year.
export incentives were perhaps yielding ex-       As to the overall effectiveness of Korean
cessive profits, the government abolished a     exchange rate and incentive policies, Frank,
number of incentive mechanisms starting in      Kim  and Westphal (1975, Chiapters 8 and 9)
1973. Thus the benefits of lower direct tax   (develop and estimate a simuiltaneous equation
rates and automatic tariff exemptions on im-    moodel to demonstrate that the historical values
ported capital goods were withdrawn, while      of the of'f'icial exchainge rate, import tariffs and
wvastage allowances declined as did interest    export incenitives, taken togetlier, resulted in
subsidies. Nonetheless, the net real effective  nearly achieving the maximumn potential GNP
exchange rate in   1974 and 1975 remained       growtlh rate. Th1us, assuming that incentives to
higher than its average over the latter half of  exports and tariffs on imp0orts had remained
the 1960s. Owing not to the change in incen-    unichaniged at their historical levels, the optimal
tives, but rather to the world recession, the   official exchange rate was found to be about
value of real exports fell by 2.3'; in 1974. In  equal to the historical rate.
turn, real export growth was nearly 15", in        To conclude: The clhanges in export in-
1975 and regained its rapidl pace in 1976 8     cent ive policies that accompanied the adoption
The real effective exclhange rates for exports  of an  export-led  industrialization  strategy
shown in Table 2 are of course only very erLude  olearly increased( the incentive to export. Fur-
indicators. It would increase the precision of  tlmermore, the new measures led to the -radlual
the estimates as indicators of profitability if the  replacement of a complicated, largely ad hoC
nominal exch3nge rate were multiplied by an     system  of incei,tives based oln multiple ex-
index of export unit values or export prices    clhange rates and direct cash subsidies, and
abroad rather than by an ilnLiCX of the general  requiring frequent adjustments, with a simpli-
price level abroadl. Fqually. the overall dorne,tic  fied and nmore stable syostemi The export policy
wholesale price indlex should ideally he replaced  refornms may thus be credited with having laid
by an inLdCex of wlholesale prices weighted by  the follulLaLiols for conltillued rapid export
export volumes (to gauge the protitahilll. of   growtli once a larger base hlad been establlised.
exports relative to domestic sales) or by an    They provided assurance of stable lprofits on
illnldx of export proULuction costs. Liket% ise, tlle  ec\ports.
estiindLeS would indicate the co0111petitivenexss of  In response, export growth a1CCel0ra.teCd for
Korea's exports more precisely if illdices of   several years after 19')(5 and( hlas continiuied to
domestic and foreign    roLdUCtion costs were   be quite raLlpid LdSplite a cm,ttiniwlly expainding
used in place of price ilnldices. Rel'ilwnients bly  lbase. The growth of exports. in tuirni, has
Balassa (1977, (Clapter 8) of the estimates in  contributed to output growth, leadling to a very
the latter direction confirmii that Korea's ex-  substantial increase in the slhare nf exports in
ports were sliglhtly' more competitive in 1974  GNP andl that of manufactured exports in
and 1975 than in the late 1960s, whlile they had  output. At the same tine, sustainiing this
achieved  an  extremiiely  highi level of com-  performance has re(luire(   periodic devalua-
petitiveness between 1971 and 1973.             tiolnsI0 andL Changes in exp(rt incentivNe rates
There clearly appears to he a relationship in  (principally  wastage  allowances  and  credit



TIIE REPUBLIC OF KOREA'S EXPlERIE.NCT WITH INDUSTRIAL D)FVl LOPMEINT'       355
preferences) between devaluations in order to  pressed in ternis of the nominal rate of protec-
maintain the real effective exchange rate for  tion, which is the percentage excess of the
exports at a relatively constant level in the face  domestic over the world market price, with the
of more rapid inflation    domestically than   latter converted into its domestic currency
abroad.                                        eqguivalent at the prevailing exchange rate.12,
I Average rates of legal and nominal protec-
tion are compared in Tables 4 and 5 below
EFFECTIVE RATES OF 'PROTECTION              accord-ing to several classifications of sectors.
AND SUBSIDY 'IN 1968                (The basis for the classification used in Table 5
is explained further below.)
A more detailed analysis of incentive policies  Korean legal tariff rates in 1968 were quite
may be obtained from an effective protection   low  by  comparison   with  other developing
study, which makes possible a quantitative     countries.'4  Furthlermore,  the   protection
comparison of incentives granted to exports    poteiitially afforde(d by tariffs to domestic sales
versus  those  granted  to import-competing    (i.e. of domestically-produced  output) was
production. Such a study conducted for 1968    generally realized only in very small measure.
sheds valuable light on the impact of incentives  The average tariff rate across all commodities in
under the policy regime adopted in the mid-    1968 was 54%'i' whereas the average nominal
1960s. Its principal results are summarized in  protection rate was only 14%.' IThere was
this section.' 1                               thus a good deal of 'water' in Korean tariff
As indicated previously, legal tariff rates  rates, with this being most evident for indul-vial
have never been a good measure of the diver-   rather tharn for primary products.
gence between domestic and world market           In view  of these results, it is tempting
prices in Korea. First, many imports are exempt  immedjiately to conclude that import controls
from  duties, including intermediate goods im-  added little to the protection provided by the
ported for use to produce exports and capital  tariff structure. Ilo%v".'er, it is necessary to
goods imported for various uses. Second, tariffs  analyse import restricted conmnmodities separa-
are virtually prohibitive for a number of com-  tely to verify this conclusion, because legal
modities with the resuLlt that domestic produc-  tariff rates in  1968 were the sum  of two
tion is sufficient to satisfy local demand at or  elements: a regular rate that was legislated and
below the world market price plus tariff. In   a special rate thiat was administered and used to
these two cases, the legal tariff overstates the  'mop-up' the scarcity premiums resulting from
nominal (i.e. actual) degree of protection.    import controls.
Third, many imports are subject to controls.      Final judcigerulenit on the importance of im-
The domestic price of such commodities can be  port controls thus rests on a comparison of
higher than  the world price plus tariff if    nonminal protection witlh regular tariff rates
demand at the tariff-inclusive price exceeds   alone. 'I'able 3 gives estimates which are weight-
domestic production plus the permitted volume  ed averages over all commodities for which
of imports.                                    nominal protection exceeded the regular tariff
For the study of incentives in 1968 it was  rate. The definitions chosen for assigning sec-
thus necessary to compare domestic and world   tors to trade categories classify a sector as
market prices direct!y. The divergence for a   'exporting' if more than 10% of its output is
particular commodity is conventionally ex-     exported, as 'import competing' if more than
Table 3. Protection due to import cLoltrols inl 1968
Trade                Number of conimodity   Nominal   Regular
category.                    groups         protection  tariff
Export                         5              64.9      56.5
Import Competing              22              41.5      18.0
Non-import Competing          46              66.2      26.9
XIC                            4              98.6      38.7
All                           77              62.6      26.6
Source: Westphal and Kim (1977), p. 2-19.
Note:   Includes only those commodity groups for which nominal protec-
tion exceeded the regular tariff rate; see the discussion in the text.



356                                WORLD DEVELOPMENT
Table 4. A verage incentive rates for major inidustry groups in 1968
Primary sectors      total        All
Agri. Mining  Total manufacturing  industries
%  /.      %        %C,
On domemric sales
Legal protection    36.5   12.2    35.1     67.6        54.3
Nominal protection  17.0    8.9    16.5     12.2        14.0
Oni total sales
Effective protection  18.1  2.9    17.1     -0.9         9.9
Effective subsidy   22.1    4.7    20.9     -6.5        10.0
Source: Westphal and Kim (1977), Tables 2.A and 2.B.
Note:   Agriculture includes forestry and fishing; 'all' induMtries refers only
to commodity producing sectors.
10%  of domestic supply is imported; and, as       We now    turn   to  estimates of effective
'non-import competing' if neither the export    protection and subsidy rates for 1968. These
nor the import share exceed 10%. XIC (export    estimates are based on nominal protection
and import competing) sectors comprise those    rates, rather than   tariff rates, and further
in which both shares exceed 10%. The classifi-  incorporate the effects of all incentive policies
cation is based on export and import shares in  operating in 1968, including the export incen-
1968.                                           tives discussed in the previous section as well as
Except for commodity groups in the export    subsidies granted to key import-substituting
category, import controls did afford some       industries.1 7 Averages based on estimates made
commodities significant adaitional protection.  at the level of 150 commodity produicing
However, out of a total of 365 commodity        sectors are given in Tables 4 and 5. In turn,
groups included within the sample for the price  Annex Table 1 gives estimates averaged accord-
comparison survey, there were only 77 for       ing to industrial classification,
which nominal protection exceeded the regular      Effectivc protection calculationis express the
tariff rate, and these 77 accounted for only    inmpact of protection measures on value added
11% of total domestic sales within the sample.  per unit of production instead of the gross
Thus, in total effect, import controls were     price.' 8 The effective protection rate of -0.9%,
relatively uinimportant. This holds even though  for manufacturing output as a whole (see Table
import controls were ostensibly imposed on      4) thus indicates that protection measures on
competing imports in the markets for com-       balance provided no protection to value added
modities represt.nting 76% of all domestic sales  in the manufacturing sectors, the protection on
in the sample.' 6                               outputs being slightly more than offset by the
Table 5. Average inicentive rates in niianzefructlwr itng by trade category in 1968
Import    Non-import
Exporting  competing   competing      XIC         All
sectors    sectors     sectors     sectors  manufacturing
vo          %'          'v         Cf
Oni domestic sales
Legal protection       83.4      56.4         65.5        75.7        67.6
Noniinal Protection     8.1      32.2          5.1        37.8        12.2
Effective protection
Domestic sales       -18.0       93.1         16.4        72.8         1.4
Exports                 4.6     - 8.6        - 0.8       - 2.1         3.1
Total sales          -10.7       91.7       --16.1        45.2         0.9
Effecctrie suibsidv
Domestic sales        26.2       91.4       -24.3         55.0      - 8.9
Exports                13.5      35.3          6.1          8.7       12.4
Total sales          -13.4       90.7        -23.7         37.9     - 6.5
Sources: Westphal and Kim (1977), Tables 3.A and 3.B.
Note:    The basis for classifying individual sectors is the same as that employ5ed in Table 3.



THE REPUBLIC 0OF KOREA'S EXPERIENCE WITII INDUSTRIAL DEVELOPMENT               357
cost-raising effects of protection on inputs. It  centive mechanisms is strengthened by adding
may further be seen that protection measures    the effects of direct tax and credit policy.
afforded an implicit subsidy to primary pro-       Of course, these results obscure differences
ducers, particularly in the agricultural sectors.  in the incidence of incentives among particular
Both of these results - no effective protection  industries and between exports and domestic
to manufacturing, and higher effective protec-  sales. As we are here most interested in the
tion to primary producers - are nearly unique   manufacturing sectors, Table 5 presents esti-
to Korea, as most other developing countries    mates for these sectors averaged by      trade
protect industry at high levels and at the      category, with separate figures given for exports
expense of primary production.                  and  domestic sales. Starting first with the
Effective  subsidy calculations further in-  results for all manufacturing, one finds that
clude the impact of credit and direct tax       greater effective incentives were afforded to
preferences. These incentive mechanisms do not  exports than to domestic sales. By virtue of
change unit value added in world prices, how-   being able to import their inputs duty free,
ever they do affect the composition of value    exporters faced world market prices both for
added as well as profits after taxes. These     their outputs and for their tradeable inputs.
subsidies were incorporated in the following    The positive effective protection rate of 3.1%
manner: the actual total direct tax liabilities of  on export sales reflects the implicit subsidy to
all firms were reapportioned to each sector on  exports through generous wastage allowances
the basis of its share in the total tax base, i.e. it  and preferential rates on electricity and trans-
was assumed that each firmn would have paid an  portation. In turn, export sales benefited more
identical average tax rate on its net income    than did sales on the domestic market from
under a neutral tax policy. The difference      direct tax reductions and credit preferences, so
between the reapportionied tax liability and a  that all the incentives together yielded an
sector's actual tax liability is the estimated tax  effective subsidy rate on exports of more than
subsidy, which could thus be negative as well as  I '2. By contrast, the effective subsidy rate on
positive depending upon whether the sector      domestic sales was - 9e ;,2
actually paid a higher or lower tax rate than the  As among manufacturing sectors classified
average. Interest subsidies were determined in  by trade category, the sectors exhiibiting the
an analogous fashion, assuming that under a     lowest average effective subsidy rates on total
neutral policy all sectors would have paid an   sales are the export and non-import competing
identical average interest rate on outstanding  sectors. These are also the sectors in which the
loans, that rate being determined as the ratio of  effective subsidy on export sales on average
total interest payments by all sectors to total  exceeded that on domestic sales. These sectors
loans outstanding. Total direct tax and interest  togetlher accounited for 84%ivo of Korea's manu-
subsidies were then added to each sector's value  factured  exports (at producers' prices) in
added in domestic prices.'9 The effective sub-  1968.22 The share of exports in the output of
sidy  rate is the percentage by which this      the export industries was 36%; that for the
adjusted value added exceeds value added at     non-import competing industries was roughly
world market prices. Since the sum of all direct  2X. In turn, effective subsidy rates were highest
tax and credit subsidies over all sectors is zero,  on average in the import-competing sectors,
the weighted (by value added at wvorld market   where domestic sales were highly protected.
prices) average of all effective subsidy rates is  The share of exports in the output of these
equal to the weighted average of all effective  industries was less than 2%7".
protection rates.                                  The results for the XIC sectors are somewhat
Under the incentive system    prevailing in  puzzling on first glance. (Recall that for these
Korea, preferential credit goes largely to the  sectors both the ratio of exports to output and
mainuifactutring sectors while the primary sectors  the ratio of imports to diomiiestic supply exceed
benefit from lower direct tax rates. As may be  10 .'.) Though the effective subsidy rate on
seen in Table 4, the net result of credit and   domestic sales was over six times tllat on
direct tax preferences in 1968 was implicitly to  exports for these sectors, the share of exports
tax manufacturing further at the expense of     in their output was nearly 40%. Correspond-
primary   activity. Adding  these preferences   ingly, ain exanmination of individual export (as
reduces the effective incentive to manufactur-  distinct from  XIC) industries reveals that total
ing from  -0.9 to -6.5%  and increases that to  incentives to domestic sales were also higher
primary production from    17 to 21%.2o Thus    than total incentives to exports in a number of
the overall result that manufacturing receives  these [Westphal and Kim (1977), Sectioni 3.5].
no preferential inducement under Korea's in-    There is strong, independent evidence of car-



358                                 WORLD DEVELOPMENT
telization among producers in most if not every  of the Korean won. Such an adjustment is,
one of these export and all of the XIC sectors,  however, necessary as the actual exchange rate
which together accounted for something less      is affected by protective measures against im-
than 20%   of Korea's manufactured exports in    ports and by export incentives. The adjustment
1968. Nearly all of these sectors benefited from  required to obtain 'net' effective incentive rates
above   average  nominal protection    on   the  is the replacement of the existing exchange rate
domestic market. Cartels in these sectors thus   with an estimate of the free trade exchange rate
appear to have operated in the classic mold of   in the calculation of value added at world
discrimination monopoly, covering their 'ixed    market prices. The free trade exchange rate is
costs through   high  prices in the domestic     the rate that would have prevailed in the
market and selling at or above marginal cost      absence of all protection and incentive measu-
abroad.2 3  This would   explain the apparent    res, except optimal tariffs. Net effective rates
inconsistency between high export shares and      thuts indicate  the difference between value
greater effective subsidy rates on    domestic    added at domestic prices under the observed
sales.                                            policy regime and world market price value
The measures of incentives presented to this   added at the free trade exchange rate.
point are gross estimates, in the sense that no     Westphal and Kim (1977, Section 3.6) pre-
adjustment has b. -n made for the overvaluation   sent a full set of estimates of net effective
Table 6. Exports of goods and services
1960     1965     1970    1975
A.    In billion weon at current prices
Exports of goods and non-factor services      8.2    68.6     381.2  2,730.3
(Gross) factor income from res. of world      1.9     8.3      47.1     99.2
B.    Percentage shares in exports ofJgoods and
noni-factor services at current prices
Merchandise                                 44.0     69.7      75.2     90.5
Primary products                          38.0    27.3      17.0     16.7
Manufactured products                      6.0    42.4      58.2     73.8
Services                                     56.0    30.3      24.8      9.5
C.    In million dollars at currenit prices
Foreign exchange receipts fromn current
transactions                                111.8   298.0   1,306.7  5,909.7
D.    Percenztage shlares in foreign exchange
receipts from current transactions, at
currenzt prices
Visible exports                              29.0    57.8      62.4     83.2
Invisible exports                            71.0    42.2      37.6     16.8
Receipts from euvernment transactions*   56.3     25.3      20.0      2.7
Military prucuremllentT                36.9     13.4     10.6
WVon sales to UN militirv personnel    19.1     11.5      4.0
Source: A ard B are based on national income accounlting data fouind in Bank of Korea, iVational
Incomtie in Korea (1975), and Economic Siatirtics Yearbook (1 976). C and 1) are based oln foreign
exchange settlements data found in various issues of Bank of Korea, Economic Stativtitjs Yearhook.
* Incluides nol-,nuttary transactions as well as military procurement and won sales to UN military
personinel.
t Includes US offshore procurement related to the Vietnam War; also includes sales of electricity,
water, and transport and comiiiniuniication services to the UN military conmmand.
Note: There are differences in statistical sources. coverage, and the times at which various
transactions are recorded between the national incomne and foreign exchange settlements data; also,
the former are in domestic currency values while the latter are in US dollir values. Thus, at the
official exclhange rate, merchandise exports do not exactly correspond in current prices to visible
exports, nor does the sum of non-factor service exports plus factor income from abroad equal
invisible exports.



THE REPUBLIC OF KOREA'S EXPERIENCE WITHf INDUSTRIAL DEVELOPMENT         359
incentive rates. Here we need only note their  whereas the share of exports in manufacturing
principal finding in regard to export incentives.  output in 1970 had been only 13%. 25
Relative to the free trade exchange rate, the  The increasing diversification of Korea's
degree of overvaluation of the won in 1968 is  merchandise exports over time may be seen in
estimated to have been roughly 9%. In turn, the  Table 7, while the following discussion is based
average net effective subsidy rate on manu-  on somewhat more detailed statistics. The most
factured exports was virtually zero. Thus, value  important items in Korea's merchandise exports
added on the given volume of exports was   in 1960 were primary products, including (in
about the same as under a free trade regime,  order of their importanlice) metallic ores, rice,
with excess wastage allov,ances plus credit and  crude animal and vegetable materials, fish, and
direct tax subsidies roughly compensating for  dried seaweed. Woven cotton  fabrics con-
the modest overvaluation of the won.        stituted more than half of exported manu-
factures, which in turn accounted for less than
14% of total commodity exports. The share of
A CLOSER LOOK AT EXPORT              manufactures reached slightly more thar 50%
PERFORMANCE                    by 1965. Clothing, plywood, woven cotton
fabrics, and plates and sheets of iron and steel
Korea's export performance has been re-  each accounted for more than 5 o of com-
markable not merely for the rapid growth of  modity exports in that year. By 1970, clothing
exports. Other dimensions touched on in the  had grown to more than a quarter of Korea's
following include the diversification of exports,  exports of goods, plywood and textile fabrics
by composition and destination, and the pos-  were each more than 10, and electronics were
sible importance of unique factors in explaining  approaching 5 I. In turn, the ability of Korean
the rapid growth of Korea's exports.        entrepreneuirs to respond aggressively to world
market trends is exenmprlified in the rapid rise of
human hair and wig (consisting almost ex-
Export comnposition             clusively of the latter) exports from nil in 1960
to about 12% of commodity exports in 1970.
As shown in Table 6, more than half of   These exports later declined in value, falling tc
Korea's foreign exchange receipts from current  1.5%e of exports in 1975, as the spotlight of
transactions in 1 960 were earned on govern-  fashion  turned away from  wigs and false
ment transactions, of which military procure-  eyelashes.
ment was the single largest item. Military     Korea's exports in 1975 were well diversified
procurement consists largely of services, with  in comparisoln with those of other developing
construction being particularly important. Sales  economies. Fxports exceeded one hundred
of won (for personal use) to resident UN    millionl US dollars for each of the following
military personnel are also classified under  items in that year (listed order of importance):
government transactions and make up the     woven textile fabrics (of which cotton fabrics
second largest item under this heading. Mer-  were less than 10%), electrical machinery and
chandise exports, which accounted for only  appliances  (including  electronics),  miscel-
44% of total exports in 1960 were dominated  laneous manufactures, fish, plates and sheets of
by primary products.                        iron and steel, veneer sheets and plywood,
By 1975, governmkkit transactions were the  footwear, transport equipment (largely ship-
source of less than 3% of foreign exchange  building, plus rolling stock), cIotlohing, manu-
earnings. In turn, the share of merchandise in  factures of metal, and non-metallic mineral
total exports had reached 91%-n, while that of  manufactures. Together, exports of primary
manufactured products grew to 74%. Though   products included in SITC categories 0 through
foreign exchange earnings in current prices  3, which had represented more than 80',, of
expanded more than 50 times over these 15   commnodity exports in 1960, aLcounted for
years, manufactured exports expanded over   only 18'%o of total exports. Nonetheless, the
900 times, a rate of growth averaging more than  value of these exports increase(d from less than
55% per annum.74 Admittedly, the growth of  $30 million in 1960 to $925 million in 1975.
manufactured exports started from a very small  As is shown in Table 8, the destination of
base, but it continued at a rapid pace after  Korea's exports also  broadelled  as manu-
manufactured exports had reached a substantial  factured exports grew. Japan's share fell from
percentage of output. Thus more than 25% of  nearly two-thirds in 1960 to a little less than
the increase in manufactured output between  one-fourth in 1975, while the share of the
1970 and 3975 was due to export growth,     United States rose from nearly 7%' in 1960 to



Table 7. The composition of merchandise exports
1960                1965               1970                1975
Value    Percent    Value    Percent    Value    Percent     Value    Percent
Food and live animals (0)                  9.7      29.6      28.2      16.1      65.6       7.9       602.3     11.9
Beverages and tobacco (1)                  0.5       1.5       0.9       0.5      14.2       1.7        67.6      1.3
Inedible crude materials (2)              15.8      48.2      37.0      21.1     100.0      12.0       150.5      3.0
Mineral fuels (3)                          1.1        3.3      1.9       1.1       8.7       1.0       104.5      2.1
Animal and vegetable oils and fats (4)     0.2       0.6       0.1       0.1       0.1       0.0+        0.9      0.0+
Chemicals (5)                              0.4       1.2       0.4       0.2      11.4       1.4        74.8      1.5
Manufactured goods by material (6)         3.9       11.9     66.4      37.9     220.9      26.4     1,484.6     29.2
Wood and cork products (63)               -         -        18.2      10.4      93.5      11.2       227.6      4.5
Textiles (65)                             -         -        10.5       6.0      84.9      10.2       648.9     12.8             o
Non-metallic mineral manufactures (66)    -         -         2.8       1.6       6.5       0.8       106.8      2.1             ;
Iron and steel (67)                       -         -        12.7       7.3      13.4       1.6       231.5      4.6
Manufactures of metal (69)                -         -         2.2       1.3      12.2       1.5       124.1      2.4
Machinery and transport equipment (7)      0.1        0.3      5.5       3.1      61.5       7.4       702.1     13.8
Electrical machinery and appliances (72)  -         -         1.9       1.1      43.9       5.3       441.6      8.7              O
Transport equipment (73)                  -         -          1.1      0.6       9.2       1.1       183.7      3.6
Miscellaneous manufactured articles (8)    0.1        0.3     34.5      19.7     352.5      42.2      1,882.6    37.1              z
Clothing (84)                             -         -        20.7      11.8     213.6      25.6      1,148.2    22.6
Footwear (85)                             -         -         4.1       2.3      17.3       2.1       191.2      3.8
Miscellaneous (89)                        -         -         8.9       5.1     114.1      13.7       383.6      7.5
Human hair and wigs (89995)              -         -         6.8       3.9     101.1      12.1        75.3      1.5
Unclassified (9)                           1.0        3.0      0.2       0.1       0.4       0.0+       11.1      0.2
TOTAL                                     32.8     100.0     175 1     100.0     835.2     100.0      5,081.0   100.0
Source: Bank of Korea, Economic Statistics Yearbook, various issues.
Note: The figures shown here are based on customs clearance data and thus exclude exports not cleared through customs, such as sales of
goods to military forces overseas and offshore sales of fish.
Values are in millions of current US dollars. rotals may not reconcile due to round-off error. Numbers in parentheses are SITC codes.



THE REPUBLIC OF KOREA'S EXPERIENCE WITH INDUSTRIAL DEVELOPMENT           361
Table 8. Destination of merchandise exports
1960            1965            1970           1975
Value   Percent Value   Percent Value   Percent Value   Percent
United States  2.2      6.7   61.6     35.2  395.0    47.3  1,536.3   30.2
Japan        20.8      63.4   43.9     25.1  233.9     28.0  1,292.9   25.4
Europe        4.6      14.1   21.4     12.2   76,0      9.1   936.7    18.4
Other Asia    3.8      11.6   41.8     23.9   81.8      9.8   760.0    14.9
Rest of World  1.4      4.3    6.3      3.6   47.6      5.7   555.2   10.9
TOTAL        32.8     100.0  175.1    100.0  835.2    100.0  5,081.0  100.0
Source: Bank of Korea, Economic Statistics Yearbook, various issues.
Notes:  See notes to Table 7.
almost 50% in 1970 before falling to roughly  tary procurement. Since exports under this
30%  in 1975. In turn, Korea's exports to    scheme are included within the militarily re-
Europe and the rest of the world increased   lated expenditures just discussed, we need not
more than proportionately, with those to the  consider them further. In turn, Korean export-
Middle East rising particularly fast after 1973.  ers may have benefited from  more subtle
Uniquie factors possibly          forms of preferential treatment, for example in
explainingexfa ort grossiy        the allocation of import quotas on textiles by
explainin1g export growth          the United States. It is not known whether this
It is appropriate here to confront the com-  is so. In any event Korea's rapid diversification
mon tendency among those unfamiliar with its  away from textile exports reduces the relevance
details to ascribe Korea's success as an exporter  of this possibility in explaining the growth of
to its special relationship with the United States  its exports.
and Japan. First, the data in Table 8 show that  It has also been suggested that Korea's
the share of Korea's merchandise exports to  exports have benefited from commercial rela-
countries other than United States and Japan  tionships (e.g., direct foreign investment, sub;
increased from 30 to 44% between 1960 and    contracting) fostered by its close ties to the
1975. This observation is not consistent with  United States and Japan. The most solid evi-
such an explanation.                         dence concerning this possibility is in regard to
Korea's relationship with the United States  direct foreign investment.2 7 While Korea has
obviously increased its foreign exchange earn-  relied extensively on foreign savings to finance
ings through expenditures stemming from the  investment, most of the private capital inflow
stationing of UN forces in Korea and, during  has been in the form of commercial loans.28
the war in Vietnam, from offshore procurement  Between 1966 and 1971, direct foreign invest-
by the United States. As indicated in Table 6  ment averaged less than 4% of total foreign
under 'receipts from government transactions',  capital inflows. Cumulative direct foreign in-
militarily related expenditures (the sum of the  vestment in Korea prior to 1970 was appre-
two components shown) in the past accounted  ciably less than $100 million in current prices.
for a sizable fraction of Korea's foreign ex-  Assuming that all of this investment went into
change earnings. However, the share of these  the manufacturing sector and combining this
expenditures in total earnings has fallen steadily  figure with the capital stock data given in Table
over time, and was less than 3% by 1975. Thus,  15 below indicates that no more than 5% of the
only a small fraction of the growth in total  capital stock in manufacturing in 1970 was
export earnings between 1960 and 1975 was    financed by direct foreign investment. In turn,
contributed  by  militarily related  expendi-  Cohen (1973) estimates that 'foreign' firms (i.e.
26
tures. . Furthermore, the merchandise com-   those with anzy foreign equity participation)
ponent of these expenditures has always been  were responsible for only a small fraction of
well below one-half of the total.            Korea's exports: 11% in 1970 and 14% in 1971.
With but one exception, neither the United   The situation has changed somewhat since
States nor Japan have granted Korean exporters  1971, for the share of direct foreign investment
visible, special preferences unavailable to other  in total foreign capital inflows rose to an
developing countries' exporters. The exception,  average of nearly 20%,in the following four
alluded to above, occurred during the Vietnam  years. This was largely due to Japanese invest-
War, when Korean exporters were granted      ment permitted by    the relaxation  of the
eligibility as suppliers under US offshore mili-  Japanese government's control over foreign



362                               WORLD DEVELOPMENT'
capital outflows and stimulated by rising wages    TIHE ROLE OF TRADE EXPANSION
and  environmental controls at home. The            IN INDUSTRIAL DEVELOP.NIENT
cumulative direct foreign capital inflow from
1970 to 1975 totals in the neighborhood of       We at last turn to examine at a broad level
$700 million, or roughly 17% of total manu-   how the growth of exports has contributed to
facturing investment over this period. I have  industrialization. It is first instructive to com-
been unable to obtain estimates of the share of  pare Korean experience with that of other
foreign firms in Korea's total exports for more  developing countries. The atypically low share
recent years.29                                of C Xports in Korea's GNP olbservedL in the
It was not until 1970 that Korea established  1950s suiggests that the growtlh of exports can
its first free trade zone explicitly desigiled to  be interpreted simply in ternis of catch:1ing-up to
attract direct foreign participation in exports.  the international 'norm'. If true, this would
Up to mid-1973, total exports from     firms   make the Korean case sornewhat less interest-
located in this zone cuinmuilated to only $20  ing, as it could not be claimed to typify an
million [in 1973 dollars, from Chloe (1975)].  export-led inidustrialization strategy.
More than two-thirds of these exports canme      What is needed is a conmlarisoni of Korea's
from  electrical machinery and appliance pro-  econiomic structure with that of other countries
ducers, most of whom   were involved in off-  of similar size and level of overall development.
shore electronics assembly. In fact, foreign  It is not very useful to do this directly,
firms have dominated Korea's electronics ex-  however, for there are very few coLluitr'ics in the
ports from  the start. In 1972, for example,   world that are similar to   Korea in these
wholly-owned foreign subsidiaries accounted   respects. A more useful, albeit indirect, comn-
for 34%, of total electronics produiction in   parison  can be made using the results of
Korea and 54"' of the industry's exports: the  crwss-c'ountry regressioni analysis aimed at dieter-
respective figures for joint ventures were 20 anid  miniing the average, or 'norm', economiiic struLc-
18%, [S. C. Suh (1975)]. With foreign firmls  ture at different levels of (leveloplmlelnt and
accounting for roughly 15'r of Korea's total   according to country size. The cross-cLu)ntry
manufacturing  exports, however, electronics  regressions used are froni ('helner) and Syrquin
represents an atypical case.30                (1975).
There has been very little research into the  Table 9 comnipares structural norrms applic-
marlketinig clianniels employed or the entre-  able to Korea for 1955 and 1972 with the
preneurial activity  involved  in  developing  historically observed structural shares. ( 1972 is
Korea's exports. Thus, for example. one cannot  used rather than a later year to avoid the
say  ' hat proportion of Korea's exports has   possible distorting effects of the world boom in
takeni  place   through   initernational  suh-  1973 andl the subsequent world recession.) Two
contracting arrangements not linked to direct  sets of norm estimates are given  the first uses
foreign investment. Nor can one do more tLhan  Ko-'an values of per capita income, popuilation,
speculate  whether Koreans derived    uniique  and .ne ratio of the curreint account balanlce-
benefits from  their extensiv'e contact with  of-ymrnents deficit to total domestic resources
foreigners (during and after the Korean War. In  to  calculate the norms from  cross-coluntry,
some cases this seenms likely. For example, an  regressions in which these variables appear as
explanationi of the current success of Korean  explanatory  factors: the  secon(d also  uses
construictionl firms in beinig awarded nmajor  Korean values of per capita incone and popula-
construction  contracts in the Middle East     tion but arbitrarily assumnies that the capital
would be incomplete without reference to the  inflow ratio is zero. [)ifferences between these
learning-by-doing  gained   through    earlier  sets of norms reflect the effects of foreign
contracts under mzilitary procurement in botlh  capital intlows at the unusually high rates
Korea and Vietnam. But such factors should     observedl in Korea.
not detract from what is evident to anv''ne whio  In 1955, thle sha.re of indlustry ( mantuifacltur-
lhas come into conltact with Korean exm)ortcrs.  ing plus :omistrucL ion) in (;NIl was somewhat
that they are both aggrcssivv an(i successful in  below that which would be expected wvlhilc the
seeking  and   ex;ploiting  opportunities  for  slhare of exports was very far below the average
profitable export. In this respect', (onC Lel'tlilyl&'  for a cou..try of Korea's size and pet'r capitt]
caninlot explain Korea's success as an exporter  ineomiie. BNy 1 172, iuusuAlly rapid industrializa-
of nmanufactuires wvithout reference  to the  tium lhadl reversed the pattern: industry's share
efforts of its entrepreneurs or the incenitive  was sonmewhaL above the norm while the share
policies under which they operate.             of ex-,ports was nearly twice the nornm when



THE REPUBLIC OF KOREA'S EXPERIENCE WITH INDUSTRIAL DEVELOPMENT               363
Table 9. Observed and norm industrial structure
Observed structural shares in Korea     1955           1972
Per capita GNP (1965 prices)        $79            $179
Capital inflow ratio                  7.7%            4.9%
Share of investment in GDP           12.0            20.8
Share of exports in GDP               1.7            21.0
Share of mfg. exports in GDP          0.4            17.8
Imports as percent of GDP            10.0            26.1
Primary share of GDP                 48.0            32.0
Industry share of GDP                13.0            26.0
Utilities share of GDP                3.5             7.5
Services share of GDP                35.5            34.5
Actual capital   Zero capital
Norm structural shares                 inflow         inflowv
Per capita income                 $79     $179     $79   $179
Share of investment in GDP       14.4     20.2    12.8   19.2
Share of exports in GDP           9.       10.8  16.0   14.8
Share of mfg. exports in GDP      1.4      2.9     0.5    2.3
Imports as percent of GDP        17.6     15.8    16.1   14.8
Primary share of GDP             52.8     33.5    55.4   35.3
Industry share of GDP            14.4     24.9    11.7   23.1
Utilities shire of GDP            5.2      7.1     5.6    7.4
Source: Westphal and Kim (1977), Table J. The norms are those estimated
from the Chenery-Syrquin large country sample.
adjusted for the inflow of foreign capital. From  GNP into these three sources, where import
the  figures shown in Table 9, it may be       substitution is defined in terms of the changing
concluded  that: (1) the share of primary      share, for each industrial sub-sector, of imports
production has probably been atypicai!y lew    in total supply.32
over the past 20 years; (2) the pace?.            The sources of Korea's industrialization
industrialization has been faster than in mainy  from  1960 to 1968 are compared in Table 10
other countries; (3) the growth of manu-       with crudely estimated cross-country norms for
facturing exports has been unusually rapid and  the growth of per capita income from $100 to
reflects more than simply catching up to the   $200. (The estimates for Korea have unfor-
norm after the dislocations caused by Japanese  tunately not been carried beyond 1968.) While
colonial policy and two wars. It is also interest-  import substitution contributed very little to
ing to observe that the growth of investment   Korea's industrialization, the growth of exports
was abnormally great and too quick to be       contributed  more   than  twice the relative
attributed merely to high foreign capital in-  amount that is typically associated with the
flows.3'                                       doubling of per capita income from $100 to
$200. What makes this comparison all the more
striking is that during this period Korea's per
Sources of growth                 capita income increased by only 55% in con-
trast to the 100%o increase underlying the norm
While it is apparent from the comparisons    estimates.
above that Korea's industrial development has     One may also use input--output data to
clearly been export-led, this stands out even  calculate the contribution   of each  source,
more dramatica"y when the sources of Korea's   sector by sector, to changes in output quan-
industrialization are compared with interna-   tities, rather than to changes in sectoral shares
tional norms. Using input-output data, it is   in GNP as above. Seven input-output tables
possible to measure the contributions of domes-  spanning the period from  1955 to 1973 at
tic demand expansion, export expansion, and    roughly three-year intervals are available for
import    substitution  to   industrialization.  Korea. But to insure that the analysis is
Roughly   speaking, the measure used     here  meaningful, it is necessary to convert data for
decomposes the increase of industry's share in  different years to constant (domestic or world)



364                                WORLD DEVE'LOPMENT
Table 10. Sources of industrialization
Domestic demand      Export         Import
expansion        expansion    substitution
%                %.11          %
Korea (1960-68)            60               38             2
Large Country Norm         55               24            21
All Country Norm           50               18            32
Source: Frank, Kim, and Westphal (1975), p. 95.
prices. This has so far been done only for the  column for each source include the inidirect
tables covering  1955 to   1968, so that the    contribution due to induced changes in inter-
analysis which follows is necessarily restricted  mediate demand.33  Thus the expansion of
to this period. In turn, data are not available at  exports, not just in manufacturing but through-
the same level of detail for 1955, so that parts  out the  economy, contributed    24%'  (12.9
of the analysis cover only 1960 to 1968.        divided by 53.2) to the growth of manu-
The top half of Table 11 shows the sources   factured output, which means that there were
of the growth of aggregate output decomposed    strong  backward linkages from    exports; in
to indicate separately the relative contributions  contrast. import substitution, by  which  is
of the primary, manufacturing, social overhead  meant a fall in the share of imports in total
and service sectors, Bet%veen 1955 and 1968,    domestic supply, accounted directly for little
the growth of the manufacturing sectors was     more than 2% of the growth of manufacturing.
responsible for more than half of the growth of  Its total contribution was even less, indlicatiing
aggregate gross output. In turn, the growth of  that import substitution   generated indirect
manufactured exports directly accounted for     demands on sectors having higher than average
16%  (8.4  divided  by 53.2 expressed as a      requirencn its for imported iLitermcediate im-
percent) of the increase in manufactured out-   ports.
put. The estimates shown under the total
Table 1. Direct and total growth contributions: 1 955- 68
Domestic demand      Export           Import
expansion        expansion       substitution
Direct   Total   Direct   Total   Direct   Total  Total
1955--1 968: All Goods & SerLices'
Primary sectors                    18.8%,  16.3/o    0.7%    3.3%   -2.3%   -2.4%    17.2%
Manufacturingt                    42.6     38.5      8.4    12.9     2.2      1.8    53.2
Social overhead                    14.4    13.8      1.4     1.9    -0.0      0.1    15.8
Services                           13.2    11.9      0.7     2.0    - 0.2   -0.2     13.7
AU sectors                         89.0    80.5     11.2    20.2    -0.3    -0.6    100.0
I960-1968: Mfanufactured (i:,-ds**
Exporting sectors                  10.2     8.7     11.5    13.0     0.5      0.5    22.2
Import competing sectors           23.6    22.4      0.3     2.4    _1.8    -2.7     22.1
Non-import competing sectors       44.1    39.7      1.1     6.0     6.2      5.7    51.4
XIC sectors                         3.9     3.2      2.1     2.8    -1.7    -1.7      4.3
All manufactures                   81.7    74.0     15.1    24.3     3.2     1.7    100.0
Source: Westphal and Kim (1977), Table N.
* Based on data at constant domestic prices.
t Includes 'other'.
**Based on data at constant wQrld prices.
Note: Figures may not reconcile due to round-off error. Growth contribLutions between 1955 and 1960
(1960 and 1968) were first calculated at the 29 (117) sector level and then aggregated before converting to
percentages.



THE RIPUBLIC OF KOREA'S LXPERIENCEI WITH [NI)NISTRIAL DEVIP'.IPNIIKNT          365
A comparison of the relative imllportance of  in  Annex  Table 2, which presents various
export   expansion  and   import substitution   indices relevant to gauging the structural change
within su6-period3  between 1955 and 1968       that has occurred witlhin Korea's manufacturing
clearly  demonstrates the marked change in      sector.
strategy that occurred during the first half of    At an even more disaggregated level, bet-
the 1960s, as may be seen below:                ween   1960   and  1968, import substitution
contributed more than 20% to the growtlh of 12
Direct contributionis to gicottlh of mainifizctured Outplut  out of the 80 manufacturing sectors dis-
1Q55 60   1960) 63  1963 68   tinguiished in the analysis from  whichl these
Exprortexpansion     Sl':,    6¼       17.4%'   estimates are drLwn.3     .\nmong these were
Import substitution  24. ;    69.2     3317w.    sectors prod4ing fertilizers, petroleum  pro-
ducts, sewing machines, electrical equiipiiienlt
To,al contributions to growthi of aggregate ottprt  and products, drugs, steel ingots, paper and
paperboard, basic inorganic clieni icals, andl cast
1955 -60  1960 63   1963 68   and forged steel (listed in ordler of the relative
Export expansion   12.9¶'.    6.3,     25,1'     contribution of import substitution). But, over
Import substitution  I 0.'   - 6.9     -0:Xr;   the eight-year period import shares acttually
increased, leadinig to n'l,w ive import substitui-
Import substitution   in  manufacturing   and   tion, in 39 of the mlanuit;facturilng sectors (and
primary export growth characterize the period   eight out of 12 primary sectors). In turn,
prior to the policy reforms, which began in     export expansion was tlle source of more than
1960. By contrast, export e.xpansion became     '20%  of output growth for 20 manufacturing
the dominant trade-relatedl source of indutistrial  sectors. Inicltuded among these were sectors
growth   after the adoption of an outward-      producing various textile products at dlifferent
looking set of policies. Thus, between 1963 and  levels  of  fabrication, miscellaneous  nainu-
1968, export -rowth dtirectly contributed IT '  factured  pro(dlucts,  hlinie)r  andl plywood,
to the growth of manullfaetuiriig, it was directly  apparel and accessories, electronics andl elect ri-
and inidirectly respo,nsible for rouglhll' one-  cal equipment. At the same time, the contrihi-
quarter of the expanisioni of .iggregate output.  tion of domestic demand expansioon exceeetld
In the bottom lhalf of Table 11, the growthi  80' 0  in 53 out of the 80 nianiifacturing sectors.
of nmanufactuiredL output is decomposeti accord-  Thus the importance of dlomestic dleinand
ing to the classification of sectors by trade   growth observed in the aiggregate carries over to
category employed in Tables 3 and 5 above.      the individual sectors as well.
Unfortunately, the data required for this do not   The pattern of manufacturing growtlh fro
extend back to    1955. More than half the       1960 to 1968 is clearly one of wvidespread
growth of output in the exporting sectors was   export expanisioni, concentratedL in the 1ab1ou0r-
due to export expansion, while these sectors    intenisive sectors, coupled withi selective inmporL
were the soLurce of more than 75'      of the   substitution, primarily in sectors prodlucing
growth of manlufaLtured exports and 20; of      basic intermnedliate products. A more recent
the growth of `anu;lllfactUring. Nonetleless, the  study, S. T. Suh (1975), analyses direct growtli
expansion of donmestic demn:and played a major  contributions through 1973. 'IThe resuilts of this
part, as may be seen either by its contributtion  study are uinfortunately not comillpwale wvitl
to the growth of all mianuifacturing output,    the estiniates presente(l above for thle lperiod
which is nearly 75", or by the total imiagnitudle  through 1968, plartly because a dlirferenlL sector
of the contribution of the non-import com-      classificationi and level of aggregation are uisetld
puting sectors. in whichi exports were less than  but more iniport:intl) IbecaIse thle tlitinlatLs are
I 0; of oultput in 1968. Only amiong1 the sectors  all based on dlata in current prices' They IIoIIe
that had acliievved non-import coin peti ig status  the less documLentL ttIe c inua ion of tlle same
by 1968 was in puort soubsitution the 0lalVCe of  pattern: wvidespread  export ex pansion  aildt1
more than 10'. of outptut growth.               selective  import substiwtion. Sulh's resul s
Hi(ddeni in the aggregate figures given in   equally indicate an increase since 1)068 in tbl
Table 1 1 is the important role played by import  ielative importance of the heavy indlustriu
substitution in some sectors. Nwhiich-i was offset  sectors in the growtli of manufacturing. boti
in the aggregate by net'gatirt, import suibstittitioni  with  respect to the domestic inarket and
(i.e. a rising ratio of imports to domelvstic   exports.)l5 This is brought out in Table 12.
supply) in other sectors. This may be seen from    The sectors exlleriencinlg relat'vel) substan-
the somewhfiat more dfisaggregated tigmires given  tial import substitLItion  between  19t08 and



366                               WORLD DEVELOPMENT
Table 12. Direct growth conztributions: 1960-73
Domestic demand  Export      Import
expansion    expansion  substitution  lTotal
70           %0          %          %0
1960-68
Light industry       56.9          10.8       --1.0       66.7
Hleavy industry       25.8          1.2        6.3        33.3
rotal industry        82.7         12.0        5.3       100.0
1968- 73
Light industry        40.1         21.8       --0.8       61t1
Heavy industry       30.3           9.1       -0.6        38.9
Total industry        70.4         30.9       -1.4       100.0
1960--73
Total industry        73.1         26.8        0.1       100.0
Source: S. T. Suh (1975), Table 5.5.
Note:   Based on data at current prices. Totals may not reconcile due to round-
off error.
1973 included fibre spinning and textile fabrics,  they  also  operate  by  increasing allocative
rubber products, chemicals, iron and steel,   efficiency whenever the domestic resource cost,
finished  metal products, and non-electrical  at shadow prices, of exports is less than the
machinery. With the exception of the last two  shadow-priced value of foreign exchange re-
industries, import substitution  was concen-  ceipts. Note that either current or future
trated in the production of intermediate goods,  consumption may be increased as a result, the
and took place largely between 1968 and 1970.  latter Lhrough increasing the rate of investment.
Much of this import substitution was due to the
production of basic steel products and petro-
chemical derivatives in medium-scale plants.
Among the sectors classified as heavy industry,      Import substitution and export
those contributing most to the expansion of                backward linlkages
exports over 1968 to 1973 were chemicals,
steel products, finished metal products, and     The role of import substitution in Korea's
electrical machinery and appliances.         industrial development deserves further com-
The figures presented above amply demon-   ment. A useful starting point is the apparent
strate that the growth of exports has played a  anomaly in the relationship between the shares
dominant role in Korea's development accouint-  of manufactured exports and industrial output
ing for roughly 20% of the growth of aggregate  in GDP. Returning to Table 9, one finds that
output between 1955 and 1968 when backward    Korea's share of manufactured exnorts in GDP
linkage effects through the demand for inter-  in 1972 exceeded the norm by a far greater
mediate inputs are included (Suh does not     spread than did the share of industry in GDP;
present estimates of total contributions). More-  the difference is about 15 percentage points in
over, these estimates understate the full con-  the first case and only two or three percentage
tribution of export growth, since the only    points in the second. This is largely explained
indirect contribuition included is that due to the  by the absence of large-scale import substitu-
derived  demand   for  domestically-produced  tion, which in turn is related to the pattern of
intermediate inputs. (For the same reason, the  realized backward linkages from exports to the
estimates understate the full contribution of  domestic production of intermediate inpiuts.
import substitution.) Two additionial macro-     The share of domestic value added (direct
economic indirect effects may be distin.guished:  and indirect) in Korea's manufactured exports
a multiplier effect due to expenditure out of  has been roughly 50% over most of the past
the additional income generated by the rise in  decade.36 This is not a consequence of the
production; and, a foreign exchange effect due  export incentive system, for the system does
to increased production made possible by a rise  not discriminate between the use of imported
in foreign exchange receipts. These effects are  and domestically-produced intermediate inputs
most dramatic where factors of production     (of the same quality) in the production of
would have otherwise been unemployed, but     exports, except in so far as wastage allowances



THE RI'TUBLIC' OF KOREA'S EXPERIEN('IC WITI INDJUSIR IA L. I):EVLOPM I N'I'    367
on imported inputs provide a bias in favour of  from   exports have been exploited through
the latter. 37 Withl protection in the domestic  import substitution, the rapid change in the
market and currency overvaluation, it is of     composition of exports has offset the import
course true that unrestricted access to alid tariff  substitution that has occuLrred, with the result
exenptions on imported inputs lead to a bias in  that the share of domestic value added in
favour of using them   unless somehow offset.   aggregate exports has remained roughly conl-
However, protection and currency overvalua-     stant.
tion are both low in Korea. Additionally, the     (Governmnent policy with re'ipect to imports
government has used the device of theo? 'domnes-  of consismer durables and other 'Iluxuiry' coni-
tic L/C', which gives the full range of export  sumption goods has also been a contributing
incentives to producers of intermediate goods   lactor to the minllscule (:contribUtionl of import
supplied to exporters, to niegate the remaining  substitution to overall growth. Imports of such
bias. Suppliers of exporters are thus permitted  goods - notable examples bi-ing auitomnobiles,
access to intermietliate inputs at world market  refrigerators, television sets, and the like  have
prices, the same as exporters.38 By avoiding    been for the most part prolhibited, so that
excessive subsidies to  backward integration    nearly all of tlie latent dLonmestic demand for
from  exports, this system   appears to lhave   these goods has been satisfied only after the
induced  only those backward linkages that      start of domnestic production. Thus, the govern-
permitted efficient production.39               ment has managed to (lelav their purchase in
The high import content of Korea's exports   significant quantities until they are produced
stems in the first instance from the types of   locally, wlich gives rise to a pattern to which
products being exported. There are a few        the term 'import suibstitution' does not, strictly
manufactured exports, incluiding silk textiles,  speaking. apply. The mathenmatics of calcuilatinig
cement and cLrainics, for which Korea does      growth contributions imp'icitly re`ognizcs this
have the requisite natural resource base. Most  distinction by assigning a very low weight to
others, such as cotton and wool textile pro-    these sectors in thie a!gg1Cegeite import substitu-
ducts, Plywood and steel pro(ducts, require raw  tioIn estimiates as a result of their initiilly small
materials - cotton, wool, roundwood, iron ore,  share in domestic demand.
anthracite coal - local supplies of whchl are      The rapid growtti of foreign exchalnge re-
either insufficient or prohibitively expensive. In  ceipts from exports has also been a key faetor
several of these cases, import substitution for  that has allowec import suLstitution to be
intermediate inputs has been carried back to    selective, which in turn is reflecte(d in its
the stage where only the unprocessed natural    relatively  srmall contrib,ution  to  aggregate
resource  product   is imported. Thus, for      growth. Not to be owerlooked is the facit that
example, natural-fibre  yarns are produced      the pursuit of an economiy'S coiimparative advan-
domestically, as is steel. Howveer, many ex-    tage is a matter not only of the comnposition of
ports, such as petrochemical based products     exports and imports but also of their size
(until recenitly) or electronic assernbly (even  relative to domestic production and couisLmnpli-
now), recquire intermediate inputs for which the  tion. The shares of both exports and imiiports in
Korean market has been too small for efficient  GDP were abnormallya1v low at the end of the
scale production or for wlich sophisticated and  Korean W'ar, and to pursue its coumpardtive
costly  technological  knlow-hiow  is needed.   advanitage has required that Korea increase the
Overall, for most export products, but subiject  sliares of both to atypically high levels. Import
to its natural resoLurce endowmient, Korea has  substitution is naturally small in the aggregate
exported the output of the labour-intensive     when the shiare of imports is rising. Moreover,
processinig stages. However, as noted above,    selective import sUbstitution  has perm1ittedi
there has also been a considerable degree of    scarce investnment resources to be conCentrated
backwardl integrationi in a number of product   in one or a few sectors at a time and therel))
lines, with iniport substitution for intermediate  enabllledi greater exploitationi of eCon11n11mieS of
products having been, by and large, carried     scale and of tlle linkages am111on1g closely inter-
back as far as economnical given Korea's poor   related activties.40  Thus, certaiilly in most
natural resource endowment.                     areas, imlport substitutioni has been delayed
Taklen togetlier, the large contribution made  until demlnand was suifficienlt to support efficient
by export growth to Korea's industrialization   scale planls. This is not to deny, however, that
andl the highly selective exploitation of back-  import -.ubstitution  in  other areas, most
ward link-ages from exports help explaiin why   notably petrochienmica:s and aLutonlobiles, was
inmport suibstituLtiorn has played such a sub-  probably preniature, as it certainly appears to
ordinate part. While some backward linkages     hiave been if learning-by-doing and otlher ex-



368                              WORLD DEVELOPMENT
ternal economy' plhenomena are neglected.    presumptive rather than conclu.ive, there is
ample evidence to indicate that emphasis on
exports has resulted in a generally efficient
Comparative advantage             allocation of resources. This is discussed below
in terms of, first, the factor intensity of Korea's
Simply to demonstrate that exports con- exports and imports, and then, the increase in
tributed substantially to Korea's industrializa-  its factor utilization over time.
tion does not establish that its strategy has been  The determinants of comparative advantage
efficient in terms of factor use. Though the  are many  and   complex. They include the
immediately foregoing discussion suggests that  effects of natural resource endowment, labour
this conclusion would be generally valid, it is  and management skills, and their accumulation
obviously desirable to seek further evidence. through learning-by-doing, risk and uncertainty,
The issue of paramount concerii is how Korea's  and a host of other factors which are either
industrial strategy has affected the growth of  difficult to quantify or have not been quanti-
GNP and the attainment of other development fied with sufficient precision in Korea's case.
objectives in comparison to alternative strate-  None the less, to the degree that Korea's
gies that might have been pursued. But to    comparative advantage during the 1960s may
examine it fully would require conducting    be said to have been in labour-intensive as
counter-factual experiments using a sophisti-  opposed to capital-intensive activities, a partial
cated general equilibrium model that has been  assessment is possible. This assessment is based
validated against the obseived history, some-  on a simple two-factor model of static compara-
thing which has not yet been attempted for   tive advantage, which is insufficient to illumi-
Korea, Furthermore, there are undoubtedly    nate the full range of issues bearing on a
additional indirect effects from exporting that  country's comparative advantage. Thus, merely
would be hard to capture in such a model - for  finding that Korea exported labour-intensive
example, the exploitation of economies of scale  conmnmodities and  imnported  capital-intensive
through increased market size and the motiva-  ones does not 'prove' that resource alloc,tion
tion that competition in foreign markets pro-  was generally efficient, but it does place the
vides for techlnological change.             burden of proof on those who would argue
Additional research would be required to   otherw.se.
gauge whether Korea has benefited in the latter  Table 13 gives average labour--capital ratios
directions as a result of following an export-led  for exports and for the replacement of imports
strategy. On the other hand, though it is    by   domestic  production. These ratios are
Table 13. The factor intensity of trade
Labour-capital ratios
(Man-years per million won)
1960    1963    1966    1968
Direct factor requirements
Manufactured products
Domestic output    2.97    2.89    2.67    2.64
Exports            2.72    3.02    3.24    3.55
Imports            2.09    1.93    1.98    2.33
All goods and services
Domestic output    4.39    4.59    4.46    4.12
Exports            3.25    2.52    2.41    2.10
Imports            4.53    4.87    4.05    4.29
Total fictor requirements
Manufactured products
Domestic output    5.43    5.41    5.03    5.14
Exports            3.74    3.71    4.09    4.29
Imports            2.77    2.40    2.40    2.74
All goods and services
Domestic output    4.39    4.59    4.46    4.12
Exports            3.42    3.05    3.25    3.15
Imports            3.78    3.66    3.26    3.48
Source: Westphal and Kim (1977), Table P.



TII RETPUBLIC' 0F KOREA'S l:XPl:Rll:N('It. WI I NI)NDSTRIlA Ml Vl I ()P0INl'N']  369
weighted averages, in which the weights are    examines the total factor intensity estimates.
proportional to each sector's share in exports  They do not imply, however, that resources
and  imports respectively. For comparative     were being allocated inefficiently within the
purposes, the weighted average factor intensity  primary sector, for one must also consider
of domestic output is also shown in the table.  Korea's natural resource endowment. The rela-
Furthermore, since these estimates are based on  tive abundance of some minerals led to the
input -output statistics. both direct and total  export of these, and it was probably efficient to
factor intensity  measuires can be presented.  import foodgrahin  given Korea's poor climate
However, for teclhnical reasons having to do   and land.
with the estimation procedure as well as for      Another study, lhong (1976), investigaltes
substantive reasons, the direct factor estimnates  the factor intensity of trade ulp to 1973. O(nce
are both the most reliable as well as the most  allowance is made for differences in estimiation
relevant, the latter owing to the fact that inany  methods and in the presentation of resuilts.
important intermediate inputs are tradeable.4   Hong's estimates for manuifaicLuredi products
On exaininiilg these estimates, one finds that  over the period 1960 to 1968 are consistent
mnanufactured exports were substantially more   with those given above. However, his investiga-
labour-intensive than imports of manufactutres  tion indicates an appreciable increase in the
in every year. On the other hand, total exports  capital-intensity of manuifactture(d exports since
were more capital-intensive than the bundle of  1968, both in absolute terms and relative to
total imports. The contrast between the manu-  factor proportions in n -nufacturing as a wholo,
facturing and other sectors is to be explained  and equally with respect to the direct andl total
by the complenientarity of capital and natural  factor intensity measures. In part, this is due to
resources, where the latter are excluded as a   capital dieeplening in the manufacturing sector
factor of production in the analysis. The effect  and the rapid increase in labour productivity, as
of Korea's natural reso.;rce endowment on its   Honig documiients by uisinig capital labour ratios
composition of trade in primary produets has    by sector, specific to eacih year. Paixrtictularl
led it to   export relatively capital-intensive  after 1970, this trend may also be traced to
minerals and to import highly labour-intensive  increased exports of cement, steel, fertilizer and
agricultural products.,                         textiles and various itemns based on p etro-
The labour-"capital ratios, by sector, on    chemical derivatives. Either dlirectly, or in-
which these estimates are based are those for a  directly, all of these products require capital-
single year -- 1968. Thus, changes over time in  intensive production methods in plants subject
the estimated factor intensity of a particuilar  to severe economies of scale. In the latter
aggregate  are due soley   to changes in its    connection, given an arbitrarily imposed con-
conmposition. Comparing direct factor intensity  straint to meet the (Idioestic demand for these
estimates across years, one may conclude that   commodities   tlirougli domestic plroduction,
Korea's  manufactured   exports became    in-   temporary exports can be efficient as it permits
creasingly  more labour-intensive over time     the construction of large plarnts without leading
while its manufactured    imports tendedL to   to initial excess capacity and thereby reduces
become more capital-intensive. In fact, wl1wras  the cost of realizing greater econonmies of scale.
Korea's manufactUredL exports were less labour-  Even without such a constraint, exports of
intensive than average nmanuifactuirinig in 1960,  cement, steel and fertilizer during the first half
they  were more labour-intensive   by   1968.   of the 1970s may well have been in Korca's
Mvlanufactuired  imports  were  more  capital-  dynamic comparative advantage. It is not so
intensive than averiage manuafadCtUring through-  obvious that it was in Korea's comparative
out the period.                                 advantige to establish the pro(Luction of petro-
At least between 1960 and 1968, the shifts   chemicals to supply its textiles and plastics
over time in thle Composition of anIM1ufaLtured  indLutstries.
exports% and  imports, when translated into
changes in direct labour capital ratios, thus            Trenids intJa(tor utilhatioi
suggest that Korea was following its coni-
parative advantage withini the manu11Lfactutrinig  Anmonig the outstandillg features of Korea's
s-ector, where unskilled labour was abundant    overall development perfornmance is the increase
and capital scarce. However, largely because of  that has taken place in the rat-, of factor
the growth of agricultural imports, it cannriot be  utilization. As noted at the outset of this essay,
said that total exports were becoming more      Korea has been relatively suCCessful in fin(ling
labour-intensive relative to total imlports. These  employm-ent for its labour force, wliclh has
conclusions hold   almiiost equally when one    grown at an average rate of 3.2"' pet annum



370                               WORLD DEVELOPMENT
since 1960. This is broadly indicated by the fall able statistics showing the trend of capacity
in the unemployment rate from a peak of 8.3%  utilization rates are based on electricity usage
in 1962 to a level of 4.1% in 1975. It is very  data and define 100% capacity utilization to be
likely that there would have been serious     equal to plant operation 24 hours a day, 365
unemployment, certainly in urban areas, had   days a year. On this basis, the aggregate
not exports of light, labour-intensive manu- capacity utilization rate within manufacturing
factures grown so rapidly. At least, this can  as a whole is estimated to have increased at an
reasonably be inferred from  estimates of the  annual compound rate of 7.2%, from 18% in
share of employment due to exports.           1962 to 32% in 1971 (the survey from which
Table 14 presents estimates taken from a   these results are taken does not extend beyond
recent study based on input-output data. The  1971).42 In terms of the measure used, 32%
indirect employment included in the 'total' capacity utilization (on average, roughly the
estimates accounts only for that required in the  same as single shift operation throughout all of
domestic production   of intermediate goods   industry) is quite high by international stan-
used in exports, so that multiplier and foreign  dards, even including developed economies in
exchange effects on employment are neglected.  the comparison. While it is not possible to state
Ev/en so, exports are seen to have accounted for  the degree to which the expansion of exports
more than one-quarter of manufacturing em-    contributed to the increase in capacity utiliza-
ployment and close to 10% of total employ-    tion, there is little doubt that it played a
ment in 1970. The contribution of export      significant role.
expansion to increased employment is even        Trends in the aggregate labour-capital ratio
more impressive. The same study estimates that  and factor productivities within manufacturing
between 1960 and 1970, the growth of exports  are also noteworthy. Using data from Hong's
was (directly and indirectly) responsible for  (1976) study, these trends are summarized in
38%  of the growth of employment in manu-     Table 15. The labour -capital ratio in manu-
facturing and 33%   of the growtlh of total   facturing rose almost continuously during the
employment.                                   first half of the 1960s, and then fell almost
Korea has also been successful in increasing  continuously through 1972, again rising in
the degree of capacity utilization within the  1973. By 1973 the labour-caital ratio ex-
manufacturing sector. The most reliable avail-  ceeded that in 1960 by more than 1 5%. Both
Table 14. Percent of employment due to exports
1960    1966    1970
Manufacturing sectors
Direct employment in production for export  2.9  13.6  18.9
Total employment due to all exports    5.8     19.0    25.9
All sectors
Direct employment in production for export  2.4  3.4    5.1
Total employment in exports            3.7      6.7     9.1
Source: Cole and Westphal (1975), Table 1.
Note:   The 'A' employment estimates given by Cole and Westphal for
1960 have been converted to correspond to the 'B' estimates for
1966 and 1970 by applying the ratio of the 'B' to the 'A' estimate
for 1966 to the corresponding figure for 1960.
Table 15. Factor use in MaunUfaCturinzg
1960    1966    1970     1973
Value added (million 1970 US$)    392.8    804.5  1,803   3,215.6
Employment (thousand persons)     477      958    1,448   2,020
Capital stock (million 1970 US$)  772.0  1,273.2  2,137.8  2,808.5
Labour-capital ratio                0.62     0.75    0.68    0.72
Value added-capital ratio           0.51     0.63    0.84    1.14
Value added-labour ratio            0.82     0.84    1.25    1.59
Source: Hong (1976), capital stock, Table A.22; employment, Table 7.6; value
added, Table A.27.



TIIE REPUBLIC OF KOREA'S -XPIRIENCI' WITH INI)USTRIAL DEVELOPMENT4            371
the output-capital and output labour ratios     the effective incentive study for 1968 reported
were rising continuously during this entire     upon in    a previous section. There it was
period: the output--capital ratio rose by 1973  observed that nominal and effective protection
to more than twice its value in 1960, while the  rates were very low by international standards,
output-labour ratio increased by nearly 100%.   while effective subsidly rates were also low.
(Output is here measured by real value added.)  Equally imlportant, effective incentive rates
Total factor productivity thus roughly doubled  exhibit a relatively small dispersion among
over 13 years, giving an annual increase of more  sectors in Korea, witlh those on export sales
than 5%. Here the estimated changes in total    being somewhat less variable than those on
factor productivity reflect both changes in the  domestic sales. 43 As is widely known, effecLive
composition of the aggregate and increased      incentive rates measured e. post reflect both
factor productivity at the micro level. The     differential factor remuneration rates and corn-
former constitutes increased economic effi-     parative efficiency differenices across sectors, so
cency through   changes in the allouation of    that it cannot be ascertained without furth er
resources while the latter reflects increased   evidence whether, for example, a high effective
factor utilization as well as 'pure' productivity  incentive rate is associated on the margin with
change, which in turn may be due to changes in  higher than average profits or inefficient pro-
technique and the scale of production, or to    duction. But, low and relatively uniform rates
technological progress at the micro level. There  obviously leave very little room for much of
have unfortunately been no sufficiently de-     either excess profits or inefficiency.
tailed  investigations to  distinguish  between    In turn, one can examine the correlation
these sources in the Korean case.               across sectors between effective incentive rates
and corresponding measures of resource alloca-
tion to obtain some insiglht into the proper
interpretation of the former. Westphlal and Kim
flCeUtiPtLs antd efficieneY          (1977, Section 4.5) argue that, in Korea,
estimates of effective incentives to total sales
Additional evidence regarding the efficiency  are indicative of relative efficiency (with higlher
of Korean resource allocation may be found by   rates implying lesser efficiency), fur there is
examining   the magnitude and structure of      little evidence of factor market distortions. This
industrial incentives, including both protection  proposition  receives further support if one
measures which distort the structure of domes-  considers the speed of structural change in
tic prices vis-a-vis world market prices and    Korea's economy, reflecting the mobillty an(i
subsidy  measures which increase factor re-     adaptability of its labour force and the fast
mnuneration without directly affecting product  pace of capital accumul1ation. However, esti-
prices. Much of the literature on trade and     mates of effective incentives to sales in the
development argues that any substantial devia-  domestic and export markets additionally re-
tion  of the exchange rate from     a unified   flect the differential profitability of selling in
equilibrium  (read, 'free trade') rate and large  these markets. Indeed, the rank correlation
deviations in protection and suibsidies among   between effective subsidy rates on export sales
industries cause resources to be allocated in-  and those on total sales is only 0.26 and the
efficientlI. There are reasons to quiestion this  simple correlation only  -0.09, which suggests
view: the protection of infant induistries, the  that incentives strongly affected effective sub-
need to raise revenule from   tariffs, and the  sidies to expo)rts. By ctontrast, conmparative
ability to achieve social and political goals   efficiency would appear to domlinlate the esti-
through inanipulation of the price mechanism    mates of effective subsidy rates on domestic
argue in favour of some divergence between      sales, for the rank correlation between these
world   market and   donmestic  prices. World   and effective subsidy rates on total sales is O."mz
nmarket prices, however, provide a standard     while the simple correlation is 0.48.
against which the effects of incentive policies    Table 16 gives rank correlation .oefficient(s
can be appraised. Large divergencies from world  between various measures of tefftctive incenitive
market prices suggest the possibility, when     rates and resource allocation. Across sectors,
other jus;fications are lacking, that allocation  neitlher the share of exports in otutput nor the
of resources is inefficient,                    growth contributions of export expansion are
A  detailed quantitative assessment of in-    significantly  related  to effective protection
centive policies is required to reach ineaningflII  rates on export sales. Ifhowever, both are signifi-
conclusions regarding   their impact on    ef-  cantly (at the 0.05 level or better) ail(l positive-
ficiency. Fortunately, here we may rely upon    ly related to eff'1eCLive sIhsebidl raIte  on ex-



372                                 WORLD DEVELOPMENT
Table 16. Rank correlation coefficients between effective incentive rates
and resource allocation
Percentage
Slhare of exports  growth contribution
in output in 1968  of exports: 1960--68
Effective protection to exports        -0.16               -0.15
Effective subsidy to exports            0.29                 0.26
Effective subsidy to total sales        0.01                 0.04
Percentage
Slhare of imports  growth contribution of
in domestic supply  import substitution;
in 1968             1960-68
Legal protectioni on domestic          -0.27                 0.03
market
Nominal protection on domestic          0.30               -0.19
market
Effective protection to domestic        0.32               -0.14
sales
Effective subsidy to domestic           0.40               --0.14
sales
Effective subsidy to total sales        0.38               -0.16
Source: Westph'al and Kim (1977), Table S.
Note: The correlation coefficients were calculated at the 92 commodity producing
sector level, where time series data on resource allocation are available. Coefficients
equal to or greater than 0.16, 0.20, and 0.27 (absolute value) are significant at the
0.10, 0.05, and 0.01 levels respectively under a two-tailed test.
ports. The results are striking, for they demon-  gressed least in those sectors where the level of
strate the importance of tax and credit prefer-  effective  incentives  was highest. Moreover,
ences within the package of export incentives    though insignificai:t, the correlations between
and suggest that export incentives had a posi-   effective incentive rates on domestic sales and
tive influence on the expansion of exports.      the growth contributions of import substitution
Neither measure of resource allocation is corre-  are negative, which is what one would expect if
lated with the effective subsidy to total sales;  import substitution had   progressed least in
however, this may be explained by observing      sectors where incentives were greatest.
that the   production   of a few    of Korea's       The results suggest that imrport substitution
exports, in particular those from    cartelized  had progressed furthest in the more efficiernt
sectors (such as plywood) that seem to have      sectors. Correspondingly, there is a significant
behaved as discriminating monopolists, appears   positive relationship between effective subsidy
to have been inefficient.                        rates on total sales and import shares, while the
The   relationship  between  incentives and   relationship between the former and growth
imports prompts a different set of conclusions.   contributions of import substitution is negative.
Shares of imports in     domestic supply are     However, these results are equally consistent
inversely related to legal tariff rates. However,  with  the alternative hypothesis that higher
legal protection tends to be redundant where     protection was given to the less efficient sec-
domestic production is well established, while it  tors, for high import shares and the absence of
is sometimes reinforced by quantitative res-     import substitution may constitute presumptive
trictions where domestic production does not     evidence of comparative inefficiency.
yet constitlute a relatively high share of supply.  To    summarize: effective    incentives  to
Indeed, the relationship between nominal pro-    domestic sales seem to reflect relative efficiency
tection and import shares is just the reverse,   while effective subsidies to exports seem    to
and the correlation is significant at the 0.01   reflect profit inceniives.44  Nonetheless, the
level. The correlations between effective in-    results do not 'prove' that the allocation of
centive rates on   domestic sales and import     resources to export expansion was affected by
shares are also positive and equally significant,  export incentives, any more than they 'prove'
indicating that import substitution had pro-     that import substitution has been generally



TIIE REPUBLIC OF KOREA'S [XPERIEN('E WITII INDLJS'I RIAL D)EVJ: LOPI.NY    373
efficient. They merely demonstrate that the   instances for interpreting high effective incen-
available evidence is reasonably consistent with  tives to domestic sales, at least in part, as a
these contentions.                            subsidy to exports. Linking the ability to sell in
a profitable domestic market to satisfactory
export performance appears to have been the
CONCLUSION                      means whereby many non-traditionial indus-
tries, whose initiation was fostered by the
Until the early 1960s, Korea followed a    government, were encouraged to produce for
protectionist strategy of import substitution for  export rather than sell excluisively in the domes-
what might be termed traditional consumer     tic market. 'rhus newly established import-
goods. Once import substitution could go no   substituting  industries have been  generally
further in these areas, the government faced the  encouraged to begin exporting almost at once.
choice of continuing with an 'inward-looking'    Incentives were given to import substitution
strategy favouring import substitution for inter-  in those industries that received positive net
mediate and durable goods, or adopting an     effective subsidies on total sales. Though overall
'outward-looking' strategy providing equal in-  import restrictions were gradually liberalized,
centives to exports and to import substitution.  quotas along with high tariffs continued in
On the whole, it opted for the latter. This is  force to protect a number of small and ineffi-
reflected both in the incentive policies that  cient industries. Moreover, through the instru-
were firmly rooted by the mid-1960s and in the  ments of its planning apparatus, the govern-
structural changes that have occurred since   ment selectively promoted import substitution
1960.                                        in a few non-traditional areas, most importantly
Exporters have benefited from unrestricted  producers goods and, of late,consumner dulruLbles
access to imported inputs and have paid neither  and atitonobl)il2s. Thus many of the explicit
tariffs nor ind:rect taxes on inputs. But,    subsidy mechanisms used to promote exports,
through increasing the relative price of compet-  particularly preferenltial credit allocation, have
ing imports and hence reducing the demand for  also been applied to engender import stub-
them, protection of domestic sales has led to an  stitution.
over-valued exchange rate, the degree of over-   In  short, thouglh  outward-looking, the
valuation in 1968 having been around 9',. To  government's strategy has not been one of
offset the consequent bias against exports, the  purely neutral free trade. Incentive policies,
government has provided several explicit export  particularly the instruments of protection, have
subsidies, includi,-a, among others, direct tax  discriminated in favour of agriculture and,
(until 1973) and credit subsidies and excessive  within malnLIfactuirinig, in favour of those sec-
wastage allowances on imported inputs. As a   tors whlere there remained opportuinities for
result, the overall (including both primary and  substantial import substitution. But, with only
manufacturing production) effective subsidy   a few exceptionis, protection in the domestic
rates on domestic sales and exports were nearly  market has been quite low by international
equal in 1968. When corrected for the cur-    standards. And, by maintaining the exchange
rency's over-valuation, bcth rates were close to  rate near the free trade level and granting
zero.                                         exporters free access to imported inputs, the
Nonetheless, incentives were not uniform   government has been able overall to provide, on
across industries (see Annex Table 1). lhere  the average, roughly equal incentives to produc-
was some discrirniai.ltion against primary ex-  tion for domestic sales and for export, More-
ports, while within manufacturing the variation  over, thIougll subsidies are not provided on a
benefited less efficient exports. The less effi-  completely uniform basis with respect to value
cient manufacturing sectors received higher   added at world prices, traditional and more
than average effective subsidies to exports; but,  efficiently produced exports have not been
effective subsidies to domestic sales in these  discrim-ini:ate(d against. Additionally, tlhroughl
industries were still higher, giving rise to an  exteniding the complete package of export
apparent bias of incentives againist exporting. In  incentives to prodLucers supplying intermediate
contrast, low nominal protection in the doomes-  inpuits to  exporters,  the government has
tic market, leading to negative effective sub-  avoided  an undue stimulus to the use of
sidies to domestic sales, gave rise to an incentive  imported inputs by exporters.
bias in favour of exports in the more efficient  Incentive policies have conitribuited to the
industries.                                   rapid expansion of exports, with annual in-
As noted below in the discussion of the roie  creases averaging nearly 30% between 1960 and
of export targets, there is justification in some  1975 in terms of constant prices. Korean



374                              WORLD DEVELOPMENT
entrepreneurs have shown themselves to be    has taken place has been selective, and its
highly responsive to opportunities to export,  structure has undergone continual change.
with notable examples being the rapid rise of   The structural changes induced by the shift
wig exports in the 1 960s and later the fast  to the outward-looking policy regime thus
expansion of electronics and footwear exports.  resulted in a more efficient allocation of re-
Also, miscellaneous manufactures, the composi-  sources as exports of labour-intensive manu-
tion of which is continually changing, have long  factured products expanded to finance rapidly
been a dynamic export sector.                growing imports of foodgrains and capital and
The growth of manufactured exports over   skill-intensive manufactured products, Begin-
the past 15 years has contributed to Korea's  ning in the late 1960s and continuing into the
industrial expansion in various ways. Export  1970s, a smooth transition appears to be under
expansion was directly responsible for more  way toward exploiting Korea's emerging com-
than  one-quarter of the growth of manu-     parative advantage in skill-intensive products.
factured output and for an even larger fraction  Textile exports, for example, have shifted
of the increase in manufacturing employment.  toward quality articles of finished clothing and
In turn, the manufacturing sector has accoun-  specialty items such as camping equipment and
ted for nearly 40% of both GNP and employ-   away from   cloth and lower quality apparel.
ment growth. But these figures understate the  Simultaneously, import substitution in the
contribution of export growth; among other   producer goods sectors is proceeding selectively
factors, they  do not reflect the backward   where economies of scale can be achieved,
linkage to domestically-produced intermediate  though there may be some cases of premature
inputs, the multiplier effect due to increased  domestic production.
consumption and investment out of the addi-     Part of the explanation for Korea's rapid and
tional income earned, or the increase in econo-  sustained growth thus lies in the efficiency with
mic efficiency that results from exporting in  whiclh available resources have been allocated,
line with a country's comparative advantage.  and hence in the policies adopted with the
All of the available evidence points to   change-over to an outward-looking strategy. In
increased factor utilization and allocative effi-  addition to changes in trade policy, comple-
ciency as a result of export growth. In Korea's  mentary monetary and fiscal reforms played an
case, resource allocation along the lines of  important role. During the first half of the
comparative advantage has meant not only that  1960s, the government relaxed controls on
labour-intensive  exports  pay  for  capital- interest rates, increased tax revenues, rationa-
intensive imports, but also that there be an  lized its internal budgeting operations, and
increase in the share of trade in total economic  started on a continuous campaign of varying
activity. In part this can be traced to its very  success to stabilize inflation at a low rate, These
poor natural resouirce endowment, which has  policies led to greatly increased savings and a
led to an increasing dependence upon imports  better directed allocationi of investment funds.
to supply foodgrains and especially energy   Without them, the resources required for rapid
sources. But the rise in the share of imports in  growth would have been lacking. In addition,
GNP also reflects the evolution of the com-  wages and the cost of capital have been kept
position of demand, particularly for producer  reasonably in line with relative factor scarcities.
goods, toward products for which imports are    Korea's experience with export-led industrial
temporarily, if not of physical necessity, a large  development has been striking not merely on
share of supply. AIn important contributing  the narrow grounds of GNP growth rates alone.
factor has been the continued high import    As a result of also being labour-intensive,
content of exports, which in turn results from  industrial development has produced rapid em-
Korea's export specialization in the labour-  ployment growth at the same time that labour
intensive processing stages and (in most cases)  productivity has been rising. Thus, Korea does
the delayed exploitation of backward liriakages  not at present have a serious unernployment
until demand is sufficient to support eff,.1ent  problem. In turn, the distribution of income is
scale plants. At the same time, protection in the  not markedly unequal. In fact, surveys reveal
domestic market has been quite low by inter-  that the distribution of consumption expendi-
national standards and has offered few oppor-  ture among Korean households tends to be
tunities for profitable investment in inefficient  more nearly equal than it is elsewhere in the
import substituting activities. The adoption of  world and that this distribution does not appear
an outward-looking strategy has thus resulted in  to have deteriorated over time.45 Exports have
negligible overall import substitution within the  thus proven to be a powerful engine of econo-
industrial sector. The import substitution that  mic growth and, due to their labour-intensity,



TIlE REPUBLIC OF KOREA'S EXPERIENCE WITH INDUSTRIAL DEVELOPMENT          375
have cor-L:ibuted greatly to rapid employment  land reform was initiated by the United States
growth, which in turn has doubt]iss been a   occupation government after World War II and
major factor in maintaining a relatively even  completed by the Korean government in 1952.
distribution of consumption.                 This, along with the economic dislocation
While   Korea's  successful  development  caused by the wars and the departure of the
derives from its outward-looking strategy and  Japanese, produced a relatively egalitarian dis-
the resulting process of export-led, labour-  tribution of assets throughout the society.
intensive industrialization, other factors have  Additionally, Korea inherited from its Con-
helped. The most obvious of these is the higlh fucian past a culture in which a very high value
level of foreign assistance, during the 1950s and  is placed on education. Through parents' efforts,
early 1 960s, which contributed to building the  and despite the fact that public expenditures on
infrastructure for subsequent growth. And,   education in Korea have been low by inter-
foreign capital inflows from all sources have  national standards, it has one of the highest
continued to be substantial thereafter. Between  literacy rates in the world and a very high
1960 and 1975, roughly 40%o of total invest-  proportion of secondary and university gradu-
ment was financed from abroad. At the same   ates. The large investment in human capital
time it should be emphasized that foreign    has yielded  a highly skilled labour force
capital inflows have been used efficiently:  obtained without great expenditure of public
Korea's gross incremental capital-output ratio  resources.
(around 2.4) is very low compared with those   While a variety of factors have contributed
in other developing countries. Moreover, the  to Korea's successful development, the key fact
availability of private foreign capital to Korea  none the less remains that economic policies
has largely been in response to its favourable  have made a large contribution to fostering
export performance, so that one must be      what appears to be a reasonably efficient and
careful not to conceive of capital inflows since  equitable process of industrialization. In short,
the mid-1960s as being 'exogenous'.          given Korea's poor natural resource endowment
Political factors made it possible to change  and assuming that its comparative advantage
policy in the early 1960s and to maintain sound  lies in labour-intensive activities, Korea pro-
policies thereafter. Since shortly after the over-  vides an almost classic example of an economy
throw of the Syngman Rhee regime, which was  following its comparative advantage and reaping
partly stimulated by its economic mismanage-  the gains predicted by conventional economic
ment, Korea has had a strong government      theory.
motivated and able to impose far-reaching       The most important lessons from Korea's
economic policies, including frequent devalua-  experience appear to be that exports respond to
tions of the currency and tax measures which  incentives while efficiency in the resource
have kept government savings at high levels  allocation can be assured by operating close to
except immediately after the rise in energy  a free trade regime. Both of these conclusions
prices. In turn, where the government has    require further scrutiny. Nothing has so far
intervened in labour markets it has generally  been said about 'institutional' incentives to
been to counter organized labour which as a  exports, while much deserves to be said. This
result is not a powerful interest group. But, at  concluding section will end with a brief dis-
the same time, in the framework of free labour  cussion of the balance between price and
markets, wages have risen in response to labour  institutionial incentives. In turn, the evidence
market conditions, so that in the fifteen years  presented here regarding the efficiency of
since 1960 the average real wage in mining and  Korean resource allocation is only presumptive,
manufacturing increased 5.5% per annum.      not conclusive. And, the Korean government
Also important were the 'initial conditions' has relied upon infant industry protection as
at the start of Korea's rapid industrialization,  well as subsidies to foster the selective develop-
Thus, part of the growth of industry and     ment of import-substitutilng industries, But
exports may be interpreted as a return to    what is unique in this regard is the pressure
'normalcy' after the removal of foreign domina-  brought to bear by the government for infant
tion and the disruption of two wars. But this ildutistries to begin exporting, either directly or
interpretation is valid only through the early  indirectly  as suppliers of exporters, very
1960s. In turn, with respect to the deter-   quickly after their establishment. 'Infant in-
minants  of income    distribution. Japanese  dustry' in a less apt characterization than
colonial occupation led to the virtual destruction  'infant exporter'.
of the landed aristocracy wvhichi had ruled Korea  It is not easy to gauge the importance of
prior to the Japanese takover. A far-reaching  institutional incentives to the rapid growth of



376                              WORLD DEVELOPMENT
Korea's exports. Perhaps the most difficult of  export promotion  policy, which makes it
these to appraise are the government-assisted  impossible  unambiguously  to  separate the
export marketing   efforts. The government-   effects of each, it is very clear that failure to
subsidized Korea Trade Promotion Corporation  provide   adequate  incentives  would   have
was founded   in 1964 to promote Korea's      severcly retarded export growth. In this sense,
exports and do market research. It has since  export incentives were a necessary condition
established offices in a number of important  for Korea's phenomenal export performance.
trading centres abroad. The government also   Were expurt targets also a necessary condition?
authorized the Korean Traders' Association (a  This is a question about which there is con-
private organization of exporters) to collect 1%  siderable debate among knowledgeable special-
of the value of total c.i.f. imports as an export  ists.
promotion fund. And, it requires its embassies   Possible evidence that export targets may
abroad to participate actively in export promo-  have had a part in stimulating exports in some
tion and has sent special trade missions to a  sectors comes from   the effective incentive
number of foreign countries as well. How to   study in 1968. It will be recalled that several
market exports effectively is an important     exporting sectors appear to have operated as
question, which makes the absence of any       cartels and, through protection, received higher
serious research on the part played by the    incentives to sell on the domestic market than
Korean   government's  marketing   assistance  to export. Under the circumstances, exporting
particularly unfortunate.                      can of course be explained in terms of dis-
In a complementdry direction, the govern-   criminating monopoly, but equally the govern-
ment established annual export targets broken  ment may have encouraged the formation of
down quarterly and in considerable detail by  cartels in these industries to promote statically
commodity, market, and domestic exporter      inefficient, non-traditional exports through
starting in 1962. Export targets have generally  using the export targeting system. In fact,
been met, and indeed exceeded, for the export  export subsidies such as the wastage allowance
promotion crmipaign has had high priority, as  rate sometimes appear to have been jointly
may be seen in the following. Adjoining the    negotiated between government and business
Minister of Commerce and Industry's office is  simultaneously  with  export targets. (It is
an 'export situation' room  laid out so that   notable in this regard that there is no evidence
potential target shortfalls may be found at a  of excess profits in these industries). But,
glance. A large staff maintains almost daily   whether these sectors would have exported less
contact with the major exporters and it is not  without export targets is a matter of specula-
uncommon for the Minister to intercede in the  tion. And, in any event, these industries to-
event of possible difficulties in meeting targets.  gether accotunted for no more than 20,o of
In turn, Korea's President receives a monthly  Korea's manufactured exports in 1968.
briefing regarding current export performance    With respect to most exports, the targets
and prospects. The emphasis on export promo-   complemented incentives without which rapid
tion was given further expression in 1969 when  export growth would not have been possible.
exporters began to be graded into four classes  One important function filled by the export
on the basis of export performance, with the   targeting system was to keep the government
highest  export  achievements  bringing  the   well informed regarding export performalnce so
national medal of honour, public Presidential  that timely changes could be made in incen-
commendation, and a number of more material   tives, often inciluding ad hoc assistance to
benefits includinig relaxation of tax surveillance.  individual exporters. The other function of the
It would be naive to conclude from this that  export targeting system and its trappings was to
targets independently set by the government    publicize  the importance attaclhed  by  the
determined   actuial  export  levels  via  a  governiiment to exports: export incentives were
command-type systen, for the targets were      well advertised and access to them was im-
meant to be indicative and as such were set    mediate. An atmosphiere was thus created in
jointly by the government and the various      which businessmen could be certain that the
exporters' associations. Furthermiiore, as noted,  incentive system  would reward efforts to ex-
export targets were by-and-large exceeded, even  port. Businessmen responded by taking the
when revised upwards during the course of a    substantial risks of expanding production and
year. Though both adequate export incentives   capacity for export.
and export targets were part of Korea's overall



Annex Table 1. Effective incentive rates for industry groups in 1968
-
On domestic sales           Effective protection           Effective subsidy
Legal        Nominal     Domestic               Total   Domestic              Total              °
protection    protection     sales     Exports    sales     sales     Exports   sales
Agriculture, forestry, &        %             %            o                    %                    %        S
fishing                      36.5          17.0         18.5      -16.1      18.1      22.5      - 9.9     22.1
Mining & energy                12.2           8.9          4.0      - 1.0       2.9       5.1        3.0      4.7
Total primary                  35.1          16.5         17.8      - 7.6      17.1      21.6      - 2.7     20.9
Beverages & tobacco           140.7           2.2        -19.3      - 1.9     --18.6    -25.8,      14.5    -24.2               z
Processed food                 61.5           2.9        -18.2      - 2.7     -17.0     -25.2        2.3    -23.0
Construction materials         32.2           3.9        -11.5      - 5.2     -11.3     -16.9        5.9    -15.9
Intermediate products 1        36.6           2.8        -25.5       31.0     -19.5     -29.7       43.4    -21.9
Intermediate products II       58.7          21.0         26.1      - 0.2      24.2      19.6       17.5     19.5
Non-durable consumer goods     92.3          11.7        -10.5      - 1.9     - 8.5     -20.6        5.4    -14.7
Consumer durables              98.3          38.5         64.4      - 4.7      51.0      38.2        2.4     31.3
Machinery                      52.6          29.9         44.2      -12.7      42.9      31.5        5.2     30.9
Transport equipment            62.4          54.9        163.5      -53.1     163.9     158.7      -22.8    159.1              ;O
Total inanufacturing           67.6          12.2        - 1.4         3.1    - 0.9     - 8.9        124    - 6.5               >
Source: Westphal and Kim (1977), Tables 2.A and 2.B.                                                                           ,<
0
w



Annex Table 2. Structural change: 1960 - 1968
Domestic                                                      IDirect contribution to
demand                    Shares in total commodity:             sector's growth*
composition       Output          Exports          Imports       Export     Import
1960    1968    1960     1968    1960    1968    1960     1968   expansion  substitution
%  %  %       %        %       %       %       %         %%
Agriculture, forestry & fishing  43.1    31.5    46.8    32.8    32.9      4.5    21.4    16.3      -0.1      - 7.9
Mining and energy                 1.8     1.8     2.7     2.4    25.5      6.2     1.0     1.1      15.9      - 3.9
Total primary                   44.9     33.3    49.5    35.2     58.4    10.7    22.4    17.4       1.2      - 7.6
Processed food                   12.0    10.1    12.5    11.4    10.4      7.7     9.4     3.8       7.8        1.8
Beverages and tobacco             5.4     5.4     6.1     6.4     2.2      2.3     1.1     0.0       4.0        2.2
Construction materials           0.8      1.9     0.9     2.2     0.9      1.4     0.4     0.4       5.6        8.0
Intermediate products It        10.2     14.5     8.9    15.0     4.9     24.6    16.1    16.1      17.0       14.8
Intermediate products Ilt       14.3     15.8    11.9    13.6    12.5     13.2    27.3    24.3      10.4        5.0
Non-durable consumer goods       9.0      7.4     8.0    10.8     7.3     34.9    13.8     2.7      36.2        6.8
Consumer durables                0.8      1.7     0.5     1.4     0.1      3.3     2.2     3.7      23.2      - 4.4             0
Machinery                         1.8     5.6     1.0     1.7      1.7     1.5     6.4    20.8       8.0      -49.0
Transport equipment               0.6     4.2     0.6     2.4      1.7     0.3     0.8    10.7       0.2      -41.9             t
Total manufacturing              55.1    66.7    50.5    64.8     41.6    89.3    77.6    82.6      15.1        3.2
Source: Westphal and Kim (1977), Tables H, J, and K.
Note: Based on data at constant world prices.
Totals may not reconcile due to round-off error.
The classification of industries used here is that developed by Balassa and Associates (1971).
* Growth contributions were first calculated at the 118 sector level and then aggregated before converting to percentages.
tIntermediate products I are those at the lowest level of fabrication, for example, yarn rather than cloth or steel ingots rather than steel sheet
and bars. They are also referred to in the text as basic intermediate products.



TIIE REPUBLIC OF KOREA'S EXPl:RIlNCFE WITII IND)lSTRIAL, DEVI*L.0PNL:NT             379
NOTES
1. Based on the then available material, Balassa  summarized in Frank, Kim, and Westphal (1975),
(1971) provides a useful survey of Korea's develop-  Chapter 10.
ment through the 1960s.
12. In the case of commodities that are not exported,
2. Manufactured exports here include SITC cate-    world price refers to the c.i.f. import price; for
gories 5 through 8 except 68.                       exports, world price refers to the f.o.b. export price.
3. The slow   growthl of real agrictultural wages  13. Note that this convention implies that nonminal
between 1970 and 1975 is both inexplicable and a    protection is equivalent, from a resource allocation
misleading indicator of the change in rural versus  point of view, to an actual tariff rate, were it imposed
urban incomes. Other (reliable) data indicate that, if  at the same level. As Bhagwati (1965) has shown, this
anything, average rural income increased relative to  is not always true where markets are imperfect. The
average urban income over this period. Before 1970,  nominal rate tends to be greater than the equivalent
however, average rural income grew substantially less  tariff when  domestic production  or quotas are
rapidly than average urban income.                  monopolized, which means that nominal protection
rates in the Korean context may ovtr-estimate the
4. For comprehensive discussions of these policy   protective effects of import controls in an equivalent
reforms, see Cole and Lyman (1971) and Brown        tariff sense.
(1973).
14. The basis for this and the following comparisons
5. A system of short- and medium-term indicative   with other developing countries is found in Little,
planning was instituted with the First Five-Year Plan  Scitovsky and Scott (1970) and Balassa and Associates
running from  1962 to 1967. For a description and   (1971).
evaluation of Korean planning, see Adelman (1969)
and Westphal and Adelman (1972).                    15. In all cases, legal and nominal protection rates are
ziveraged using doiiicstic sales in world prices as
6. The index used combines wholesale pulce move-   weights; efft-tive incentive rates (see below) are
ments in Korea's major overseas markets with changes  averaged using value added at world prices as wLeights.
in exchange rate parities among these countries. See
Westphal and Kim (1917), Sectionl 2.                16. The last figure cited, however, represents a biased
estimate of the imposition of import controls relative
7. The almost continuous fall in the actual tariff rate  to total domestic sales, for a commodity group's
paid on imports is due to the rising share of tariff  inclusion in the price comparison sample was based, in
exempt imports for use in export production and not  part, on the imposition of import restrictions.
to a general lowering of legal tariff rates.
17. The subsidy due to the export -import link system
8. Exports in current dollars rose by 52% in 1976  is not included, but it is known to have been very
over their value in 1975.                          small. For details regarding the estimation procedure,
see Westphal and Kim (1977), Section 3. The esti-
9. Most of these are crude efforts focused on the  mates presented here are based on the so-called
correlation between real exports and the real effective  'Balassa method' for dealing with non-tradeables.
exchange rates for exports. The correlation may be nil  Westphal and Kim also provide estimates according to
even though the two variables are signiificantly related.  the 'Corden method'. The differences between esti-
for the implicit model is not theoretically sound.  mates under the two methods are very small. For the
Thus, for an economy in dynamic equilibrium, with   distinction between these nietlhods, see Balassa and
the real excliangc rate maintained at its appropriate  Associates (1971), p. 321 ff.
level, real exports would continue growing even if the
appropriate real exchange rate were constant over   18. Formally, the effective rate of protection is
time. This factor is partially reflected in the test by  defined as thie percentage excess of domestic over
Frank, Kim, and Westphal (see the text), since they  world price value added, where the latter equals the
include real non-agricultural output in the regression.  difference between the world mnarket price of the
In turn, one would expect lagged responses to the   product and the cost of its inputs at world prices.
removal of trade distortions to undermine the efficacy
of even this formulation, particuilarl) if the real  19. Value added so adjusted is an estimate of what
exchange rate remains conistanit after the removal of  value added in the sector would have been if there
trade distortions.                                  were no tax and credit preferences and net factor
returns were unaltered from their actual value under
10. By giving annual averages of the official exchange  the incentive policies followed.
rate, Table 2 makes it appear that devaluations
occurred more frequently than they in fact did.     20. MIanufacturing is here defined to include beve-
rages and tobacco and processed food. Removing these
11. See lVestphal and Kim (1977), Sections 2 and 3,  sectors from  manufacturing  changes the average



380                                  WORLD DEVELOPMENT
effective protection and subsidy rates to 5.7 and 0.3%  30. Cohen (1973) gives an interesting statistic in this
respe(ctively.                                    regard. He surveyed exporters of transistors, radios,
cotton cloth, cotton yarn, baseball gloves, and wigs
21. These figures are respectively 14 and - 1.8%o when  and found  that Korean companies lhad initiated
beverages and tobacco and processed food are re-  exports prior to the arrival of foreign firms in five out
moved from the average for total manufacturing.   of these six products.
22. This information and thle following export shares  31. In relative terms, 1972 was a recession year in
based on Table 3.C, Westphal and Kim (1977).      Korea, whiclh explains the investment rate's proximity
to the norm in that year; investment was more than
23. The most careful documentation of this is to be  28% of GDP from 1969 to 1971.
found in Rhiee and Westplhal (1977).
32. Thit; deconiposition is that developed by Chenery
.24. The expansion of manufactured exports is cal-  (1969). See Frank, Kim, and Westphal (1975), pp.
culated from data in Table 7.                     86-96 for additional details.
25. These estimates are made on the basis of the data  33. Using the inverse Leontief matrix, the growth of
underlying Table 1.                               demand    for  domestically-produced  intermediate
inputs has been separated into that due to domestic
26. Receipts from government transactions peaked in  final demand expansion, export expansion, and import
1969, at slightly over $275 million, when militarily  substitution; each of these components has then been
related expenditures were nearly $210 million, the  added to the corresponding direct contribution to
largest figure they ever reached. Offshore procurement  obtain the total contributions. (In the estimates of
in Korea for the Vietnam War achieved its lhiglhest  total contributions, the effect of input-output co-
level,-$64 million, in 1971.                      efficient changes is included with that of domestic
demand expansion.) All of the growtth of demand for
27. Legislation controlling non-grant foreign capital  domestically produced intermediate inputs is counted
inflows was first passed in 1960. Preferences, including  under domestic demand expansion in the estimates of
a rather standard package of tax concessions for direct  direct contributions; thus, the total contribution of
foreign investment, were simultaneously established to  domestic demand is less than its direct cointribtiuion.
stimulate the inflowv of foreign capital and techlnology,
while limits on equity investment and profit remit-  34. For the source of this and the following state-
tances by foreign investcas were completely removed  ments, see Westphal and Kim (1977), pp. 4-37 and
in 1966. The relatively low volume of direct foreign  38, and Tables T and U.
investment in Korea during the 1960s is commonly
explained either by the uncertainties of its political  35. In Korean parlance, the heavy industrial sectors
situation or by purposeful administrative tactics that  comprise non-clectrical and electrical machinery as
were apparcntly relaxed in the early 1970s; it was not  well as transport equipmncit and basic intermediate
due to the lack of legislated inducements.        products. Thiey tlhus include such products as precision
instruments and electronics.
28. The ratio of inports minus exports to investment
averaged 56% between 1955 and 1975; the ratio of  36. See Frank, Kim, and Westphal (1975), page 81 ff.
this difference to GNP averaged 9.6%.o. During the  Data limitations make it virtually impossible to assess
1950s and early 1960s, Korea's continuing trade   trends in the domestic value added content of exports;
deficit was almost wholly financed by grant aid from  on this point, see Cole and Westphal (1975).
the UN and the United States. It was not until the
mid-1960s that foreign borrowing became important,  37l Until recently, incentives granted to export and
but onice started it assumed large proportions. In  import-stibstituting activities alike have discriminated
1966, official grants were roughly equal in volume to  against purchases of domestically-produced machinery
foreign loans; by 1970, the volume of the latter was  and equipment through tariff exempti3ns and easy
more thani five times the size of the former, wlile less  access to low interest rate credits tied to the purclhase
thani half of the loan arrivals were from government  of imported capital goods. This has been tlhoroughly
and mutiltilateral sources. Most of the private foreign  documented by Rhee and Westphal (1977) for textile
loans 1lowing into Korea have come from Japan, the  machinery. In turn, Frank, Kim, and Westphal (1975),
United States, and Western Europe; many, but by no  pp. 117-119 estim:itc that in the late 1960s the
means all, have been suppliers' credits of one form or  difference between the hiighi interest rate on domestic
another.                                          commlercial credit and the interest rate chlarged on
suppliers' credits to purchase imported capital goods
29. The foregoing information on direct foreign   increased the real purchase cost (i.e. including tlle
capital inflows is based oni Frank, Kim and Westplial  discounted value of interest payments) of indigenous
(1975) and Hasan (1976), as well as various statistical  capital goods by roughly 20%. Recently, having
publications of the Bank of Korea and the Economnic  recognized the retarding effects of its policies on the
Planning Board.                                   domestic capital goods sector, the government has
abolished tariff exemptions for exporters on their



THE REPUBLIC OF KOREA'S EXPERIENCE WITH INDUSTRIAL DEVELOPMENT                    381
capital goods imports and has established a relatively  40. In the context of a planning model, WVestplial
large fund to provide medium-term domestic credit at  (1971) documents the potential gains from following
an interest rate near the world market level.     such an investment pattern.
38. This feature of the incentive system is incor-  41. See Westphal and Kim (1977), pages 4-31, 32,
porated in the effective incentive estimates presented  and 46.
previously. Effective incentive rates on domestic sales
are a weighted average of those on sales to exporters  42. See Kim and Kwon (1977), Table 2, U, series.
and on all other sales; nominal protection rates pertain
to sales other than to exporters. In turn, the estimates  43. See Westphal and Kim (1977), pages 3- 24 througl
of nominal export incentives in Table 2 include   26.
incentives both to exporters and to suppliers of
exporters.                                        44. All of the correlations reported in this section
remain virtually unchanged if estimates of effective
39. There are some instances, for example petro-  incentives under the 'Corden method' are used in place
chemicals, where a case can be made that inefficient  of those obtained under the 'Balassa method'.
production has been established in part to, supply
exporters. However, additional import-substitution in-  45, See Frank, Kim, and Westphal (1975), Chapter
centives have been granted in such cases.         11, for details regarding this and the following poinits.
Adelman (1974) and Renaud (1976) survey the
evidence regarding Korea's income distribution in
detail.
REFERENCES
Adelman, Irina, Practical Approaches to Development  my', Review of Economics anid Statistics, Vol. 60
Planning: Korea's Second Five-Year Plan (Balti-   (1973), pp. 190-197.
more: Johns. Hopkins Press, 1969).             Cole, David C., and Princeton N. Lyman Koreant
Adelman, Irma, 'South Korea', in Hollis Chenery, et  Development: The Interplay of Politics and Econo-
al., Redistribution with Growth (London: Oxford   mics (Cambridge: Harvard University Press, 1971).
University Press (for the World Bank and the   Cole, David C., and Larry E. Westphal, 'Tlhe con-
Institute of Development Studies, University of   tribution of exports to employment in Korea', in
Sussex), 1974).                                   Hong and Krueger (1975).
Balassa, Bela, 'Industrial policies in Taiwan and  Frank, Charles R., Jr., Kwang Suk Kim and Larry E.
Korea', WVeltwirtschaftliches Archiv, Band 105,   Westphal, Foreign Trade Regimes anid Economii
Heft 1 (1971), pp. 55-77.                         Development: South Korea (New York: National
Balassa, Bela, and Associates, The Structure of Protec-  Bureau of Economic Research, 1975),
tion in Developing Countries (Baltimore: Johns  Hasan, Parvez, Korea: Problems and Issues in a
Hopkins Press (for the World Bank), 1971).        Rapidly  Growing   Economy   (A  World   Bank
Balassa, Bela, Policy Reforrn in Developing Countries  country  economic report), (Baltimore: Johns
(Oxford: Pergamon Press, 1977).                   Hopkins Press (for the World Bank), 1976).
Bhagwati, Jaglish, 'On the equivalence of tariffs and  Hong, Wontak and Anne 0. Krueger (eds.), Trade and
quotas', in Robert E. Baldwin, et al (eds.), Trade,  Development in Korea (Seoul: Korea Development
Growth and Balance of Payments: Essays in Honor   Institute, 1975).
of Gottfried Haberler (Chicago: Rand McNally,  Hong, Wontak, Factor Supply and Factor Inu'ensiij of
1965).                                            Trade in Korea (Seoul: Korea Development Insti-
Brown, Gilbert T. Korean Pricing Policies and Econo-  tute, 1976).
mic Development in the 1960s (Baltimore: Johns  Kim, Young Chin, and Jene K. Kwon, 'The utilization
Hopkins Press, 1973).                             of capital and the growth of output in a developing
Chenery, Hollis B., 'The process of industrialization',  economy: the case of South Korean manu-
Economic    Development   Report   No.   146      facturing', Journal of Devclopment EconVZomics,
(Cambridge:  Harvard  University  Project  for    No. 4 (1977), pp. 265-278.
Quantitative Research in Economic Development,  Kuznets, Paul W., Economic Growthl and Structutre in
1969).                                            the Republic of Korea (New Haven: Yale Univer-
Chenery, Hollis and Moises Syrquin, with the assis-  sity Press, 1977).
tance of Hazel Elkington, Patterns of Develop-  Little, Ian M., Tibor Scitovsky, and Maurice Scott,
ment, 1950-1970 (London: Oxford University         Industry and Trade in Some Developing Coountries:
Press (for the World Bank), 1975).                A Comparative Study (London: Oxford University
Choe, Boum Jong, 'An economic study of the Masan      Press (for the Organization for Economic Coopera-
Free Trade Zone', in Hong and Krueger (1975).      tion and Development), 1970).
Cohen, Benjamin I., 'Comparative behaviour of foreign  Rhee, Yung W., and Larry E. Westphal, 'A micro
and domestic export firms in a developing econo-  econometric investigation of the impact of indus-



382                                WORLD DEVELOPMENT
trial policy on technology choice', Journai of  featuring economies of scale', in Chenery, Hollis
Development Economics, No. 4 (1977), pp.        B., et al., Studies in Development Planning (Cam-
205-238.                                        bridge: Harvard University Press, 1971).
Renaud, Bertrand, 'Economic growth and income   Westphal, Larry E., and Irma Adelmnan, 'Reflections
inequality in Korea', World Bank Staff Working  on the political economy of planning: the case of
Paper No. 240 (Washington: The World Bank       Korea', in Jo, S. H., and S. Y. Park (eds.), Basic
1976).                                          Documents and Selected Papers of Korea's Third
Suh, Sang Chul, 'Development of a new industry     Five- Year Economic Development Plan (1972- 76)
through  exports: the electronics industry in   (Seoul: Sogang University, 1972).
Korea', in Hong and Krueger (1975).          Westphal, Larry E. and Kwang Suk Kim, 'Industrial
Suh, Suk Tai, 'Import substitution and economic    policy and development in Korea'; World Bank
development !m Korea', Working Paper No. 7519   Staff Working Paper No. 263 (Washington: The
(Seoul: Korea Development Institute, 1975).     World Bank, 1977).
Westphal, Larry E., tAn intertemporal planning model



Ivory Coast: The Cliallenige of Success by Pastiaan den Tuinrder and others, published by
The Johns Hopkins University Press, 1978
Kenya: Into the Second Decade by John Burrows and others, published by The Johns
Hopkins University Press, 1975
Korea: Problemns anid Issues in a Rapidly Growing Economy by Parvez Hasan, published by
The Johns Hopkins University Press, 1976
Lesolho: A Develoupment Chlallenige by Willem Maane, distributed by The Johns Hopkins
University Press, 1975
Nigeria: Options for Long-Term Dezvelopnwent by Wouter Timns and others, published by
The Johns Hopkins University Press, 1974
Papua New G-.inea: Its Economic Sitiuationri and Prospects for Developient by George
Baldwin and others, distributed by The Johns Hopkins University Press, 1978
The Ph1ilippines: Priorities and Prospects for Dcvelopinent by Russell Cheetham, Edward
Hawkins, and others, distributed by The Johns Hopkins University Press, 1976
Turkey: Prospects and Problems of an Expanding Economny by Edmond Asfour and others,
distributed by The Johns Hopkins University Press, 1975
Yuigoslazia ,Y"- topient with Decentralization by Vinod Dubey and others, published by
The TIh      ,mkins University Press, 1975
World ribu AS.aff Occasional Papers
A Model for Ineome Distribution, Employment, and Growth: A Case Study of Indonesia by
Syamaprasad Gupta, published by The Johns Hopkins University Press, 1977
Coffee, Tea, and Cocoa: Market Prnospects and Developnment Lenzdinkg by Shamsher Singh and
others, published by The Johns Hopkins University Press, 1977
Malnutrition and Poverty: Mangnitufde and PolicyJ Options by Shlomo Reutlinger and
Marcelo Selowsky, published by The Johns Hopkins University Press, 1976
Economic Evaluation of Vocational Training Programs by Manuel Zymelman, published
by The Johns Hopkins University Press, 1976
A Developtnewnt Model for the Agriciultuiral Sector of Portugal by Alvin C. Egbert and Hyung
M. Kim, published by The Johns Hopkins University Press, 1975
Other Publications
Agrarian Reform as Unifiniishied Butsiniess: The Selectedl Papers of Wolf Ladejinisky edited by
Louis J. Walinsky, published by Oxford University Press, 1977
Twent-five Years of Economnic Development: 1950-1975 by David Morawetz, distributed
by The Johns Hopkins University Press, 1977
World Tab)les 1976, published by The Johns J-Iopkins University Press, 1976
The Troipics and1t Economic Devclopment: A Pro1v0ative Inquiry into the Poverty of Nations by
Andrew Kamarck, published by The Johns Hopkins University Pre;,s, 1976
Size Distriblution of Incomie: A Compilation of Data by Shail Jain, distributed by The Jolhns
Hopkins University Press, 1975
Redfistribution -withi Grow0th by Hollis Chenery, Montek S. Ahluvalia, C. L. G. Bell, John
H. Duloy, and Richard Jolly, published by Oxford University Press, 1974



THE WORLD BANK
Headquarters
1818 H Street, N.W.
Washington, D.C. 20433 U.S.A.
European Office
66, avenue d'Iena
75116 Paris, France
Tokyo Office
Kokusai Building
1-1 Marunouchi 3-chome
Chiyoda-ku, Tokyo 100, Japan
World Bank reprinits
No. 39. Shamsher Singh, "The International Dialogue on Commodities," Resources Policy
No. 40. Gary Kutcher and P. L. Scandizzo, "A Partial Analysis of the Sharetenancy Rela-
tionships of Northeast Brazil," Joturnal of Deuelopnienit Economics
No. 41. Bela Balassa, "The Income Distributional Parameter in Project Appraisal," Eco-
nomic Progress, Private Values atd Ptublic Policy (North-Holland)
No. 42. Dipak Mazumdar, "The Rural-Urban Wage Gap, Migration, and the Shadow
Wage," Ox fford Economic Papers
No. 43. Dipak Mazumdar, "The Urban Informal Sector," World Development
No. 44. Carmel Ullman Chiswick, "On Estimating Earnings Functions for LDCs," Journal
cf Developmcnt Economics
No. 45. Clive Bell and Pinhas Zusman, "A Bargaining Theoretic Approach to Cropshar-
ing Contracts," T7e A incrican Economic Reviewv
No. 46. Kenji Takeuchi, Gerhard E. Thiebach, and Joseph Hilmy, "Investment Require-
ments in the Non-fuel Miineral Sector in the Developing Countries," Natural
Resources Foruim
No. 47. Shlomo Reutlinger, "Malnutrition: A Poverty or a Food Problem?" World Develop-
ment
No. 48. Clive Bell, "Alternative Theories of Sharecropping: Some Tests Using Evidence
from Northeast India," TeIC Joturnial of DevelopIlm1et Studies
No. 49. H. N. Barnum and R. H. Sabot, "Education, Employment Probabilities and Rural-
Urban. ligration in Tanzania," Oxford Bulletin of Econoinics and Stiatistics
No. 50. Yung W. Rhee and Larry E. Westphal, "A vl\icro, Econometric Investigation of
Choice of Technology," Jolurnal of DetvelopInt'nt Econoinics
No. 51. Bela Balassa and Michael 5harpston, "Export Subsidies by Developing Countries
Issues of Policy," ConimLert il Policy Issues
No. 52. D.C. Rao, "Economic Growth and Equity in the Republic of Korea," World Devel-
No. 53. Paul Streeten and Shahidl Javed Bu rki, "Basic Needs. Some Issues," World Develop-
menit
No. 54. Larry E. Westphal, "The Republic of Korea's Experience with Export-Led In-
dustrial Development," IWVorld Develolmient