bruton: a reconsideration of import substitution a ...faculty.nps.edu/relooney/bruton.pdf · of the...

Journal of Economic Literature Vol. XXXVI (June 1998) pp. 903–936

Bruton: A Reconsideration of Import Substitution

A Reconsideration of ImportSubstitution

HENRY J. BRUTON1

1. Introduction

From the outset of widespread con-cern with the economic development oflow-income countries in the late 1940s,much attention has been given to therole that international activities play inexplaining growth or its absence. Oneof the earliest general strategies of de-velopment that directly involves inter-national transactions is that usuallycalled import substitution. Import sub-stitution of one form or another pre-vailed in many developing countriesduring the 1950s and early 1960s. Inthe late 1960s and early 1970s a quitecontrary approach, identified as out-ward (or export) oriented, became in-creasingly common.

This paper studies the import substi-tution story—why and how it came tobe—and the rise of the outward-ori-ented approach—why and how it cameto be and—temporarily at least—to winthe battle between the two approaches.I then examine the doubts that havearisen with respect to the latter ap-

proach, and the nature of the new ten-sions that now exist with respect to anappropriate trade strategy.

The paper is organized as follows: af-ter a preliminary definition of the basicnotions, the next section reviews the ra-tionale of the import substitution ap-proach found in the 1950s and early1960s. In Section 4, I examine the im-plementation of the import-substitutionapproach as generally practiced. This isfollowed by a summary of actual devel-opments in a number of countries from1950 to 1970, when import substitutionwas a widely followed policy. In Section6 I briefly discuss the rise and earlysuccesses of outward orientation, thensummarize developments in the post–1980 period, and discuss sources of dif-ficulties with openness. There is a shortconcluding section.

The arguments of the paper are builtaround the notion that the primarysources of development are learningand knowledge accumulation. The prin-cipal reason for the failure of importsubstitution was that, as practiced, itcreated an environment that discour-aged learning. The outward-orientedstrategy, on the other hand, fails to ap-preciate that learning requires condi-tions that are essentially internal anddependent on the basic characteristics

903

1 Williams College. Acknowledgments: RalphBradburd, the referees, and John Pencavel wereexceptionally helpful with their comments onearlier drafts of this paper. I am also greatlyindebted to Larry E. Westphal for his comments,suggestions, criticisms, doubts and hopes, andquestions on previous versions.

of the society. This failure means thatoutward orientation as such needs sub-stantial qualification and redirection.

2. Preliminary Notions

Import substitution refers to a set ofideas about why mass poverty has pre-vailed and continues to prevail in manycountries while other countries havegrown rich, and about a general ap-proach to the elimination of that pov-erty. The explanation of mass poverty isgenerally found in the structure of pro-duction—mainly the dominance of agri-cultural and mineral activities—in thelow-income countries, and in their in-ability, because of their structure, toprofit from international trade. Tochange this situation, the argument con-tinues, the low-income countries (theSouth) must alter their structure—theymust industrialize. To industrialize,given the existence of already industri-alized and highly productive economies(the North), the countries of the Southmust protect their economies from im-ports from the North and concentrateon putting in place new activities thatwill produce an array of manufacturedproducts currently imported. Thuswould the structure of the economy bechanged and, at some future time, makepossible a foreign trade that contributesto the development objectives.

In the literature of development eco-nomics, import substitution is sharplycontrasted with an outward- (or export-) oriented approach to development. Inthis approach, primary attention isgiven to the advantages of foreign tradein general and of exports in particular.Whether explicit export-promotion poli-cies are required depends on the spe-cific circumstances of the individualcountries, but the basic notion is tokeep the domestic economy open to for-eign competition and foreign capital,

and to ensure that exports are in no waypenalized, if not specifically encour-aged. Structural change would then oc-cur over time according to changes inthe dictates of comparative advantagethat are assumed to occur.

Over the last 15 or so years, the out-ward-oriented approach has gaineddominance among academic economistsand those in international organizationsconcerned with development. Many na-tional aid agencies in the North havealso become convinced of the validity ofthe outward-oriented approach. A num-ber of countries have made noteworthyefforts to shift from an essentially im-port substitution approach to a moreoutward-oriented approach, other coun-tries are trying to do so, and virtually allcountries are being urged to do so byaid donors and advice givers.

In more recent years, evidence andargument have questioned this new or-thodoxy. There is increasing doubt thatgrowth is as simple as it appears in theexport-oriented arguments, and re-newed emphasis is being placed onmore basic characteristics of an econ-omy, especially entrepreneurship, insti-tutions, and knowledge accumulationand application. Outward orientation ispushed hard by influential individualsand institutions at the same time that astrong revisionist movement is under-way and gaining strength. A state oflimbo exists with respect to trade strat-egy (and development strategy in gen-eral) that qualifies the simple, universalprescription of the last decade or so.

3. The Beginning

As colonial empires collapsed and thegreat differences in per capita incomebetween North and South were recog-nized, the obvious question confrontingthe development economics communitywas how to go about raising incomes in

904 Journal of Economic Literature, Vol. XXXVI (June 1998)

the South. Something fundamental wasnecessary. Three things seemed espe-cially important in the early 1950s.

3.1 Rejection of the Market Solution

The view that a more or less freemarket would not solve the develop-ment problem was widely accepted. Theproblem was not market failure in theusual textbook sense (externalities, de-creasing costs, etc.). Rather, the notionwas that the division of labor betweenthe rich countries and the poor onesseemed to doom the latter to perma-nent poverty. The most widely cited evi-dence was data purporting to show thatthe net barter terms of trade had turnedagainst the developing countries overthe decades prior to 1940. Raul Pre-bisch, Hans Singer, and others calcu-lated many such series that seemed toshow a secular deterioration in theterms of trade of the poor countries.2Prebisch’s explanation that the gainsfrom productivity growth in the Northresulted in rising wages, not fallingprices, due to the monopoly power ofboth labor and firms in the North, waswidely accepted. In the South, depen-dent mainly on agricultural and mineralexports, there was lower productivitygrowth, and wages were held down bysurplus labor, weak unions, and compe-tition among exporters. The rewards ofproductivity growth in manufacturingactivities were thus not available toimporters of such products in theSouth.

Also cited as a source of difficultieswere Engel curve arguments that theincome elasticity of demand for agricul-tural products and raw materials in theNorth declines as incomes reach higherand higher levels. If exports lagged be-hind the growth of income in the South

for this reason, then import substitutionof some kind must take place to protectthe balance of payments, or growthwould slow or stop. Widespread protec-tion of agriculture in many rich coun-tries exacerbated this effect. Added toall this was the argument that cyclicalchanges in the North resulted in re-duced employment and income, andhence imports, rather than in fallingproduct and factor prices, while in theSouth it was wages and prices that re-sponded to downturns.3

The specific arguments about theterms of trade were buttressed by moregeneral views that the market was an in-strument that kept poor countries poorand rich countries rich. There were sev-eral reasons for such views. Men wholed independence movements had livedtheir adult lives during two world warsand a devastating worldwide depressionthat severely penalized the South. Pre-dictions in the mid to late forties weregenerally to the effect that the post-World War II world economy would re-semble that of the 1930s.4 Though theyears 1870 to 1914 were fairly satisfac-tory, that period was hardly fresh in anydecision maker’s mind,5 and at thatdate, Europe and the United States had

2 Singer (1950) and Prebisch (1950) are earlystatements of this position. There are many more.

3 This argument is most clearly worked out invarious publications of the Economic Commissionfor Latin America, where Prebisch was the domi-nant figure in the early 1950s. It was widely ac-cepted inside and outside Latin America in oneform or another. See, for example, Prebisch(1950).

4 There were articles in both professional jour-nals and newspaper accounts that predicted a re-turn to widespread depression after the huge war-time expenditures were reduced. No one foresawthat the quarter century from 1950 was to be theremarkable boom time that it turned out to be. Itwas, however, well into the 1950s before all Euro-pean countries achieved the full convertability oftheir currencies. An interesting question is whydid some policy makers seem to catch on that thepost 1950s were really different from what theywere expected to be and others did not.

5 W. Arthur Lewis (1978) is a fine study of theseprewar decades.

Bruton: A Reconsideration of Import Substitution 905

been growing steadily for 50 to 75years, while most of the rest of theworld’s population remained mired insevere and mass poverty.6 In addition,the Keynesian ideas that even a per-fectly functioning market may not en-sure full utilization of resources werebecoming widely accepted. The conclu-sion for many policy makers andprofessional economists was clear: the“structure” of the economies of the de-veloping countries had to be changed infundamental ways if they were to com-pete on equal terms in the world mar-kets, and a market mechanism could notbring about this sort of structuralchange.

The Soviet Union’s experience hadyet to be understood very well, andmany otherwise informed economistsand political leaders were impressed bywhat evidence was available. The greatevils of the Stalin era were not widelyknown. It was known, however, thatgrowth during the 1920s and 1930s hadbeen quite remarkable. The USSR’scommitment to central planning and tolarge-scale, capital intensive industriali-zation was especially appealing to thosecountries that put great weight on be-coming a world economic power.

India was such a country, and the In-dian effort was widely regarded as amodel by other developing countries inthe 1950s. The defense of the invest-ment in heavy industry rested on strongassumptions that there were economy-wide effects on productivity growth cre-ated by a domestic capital goods sector;furthermore, economic independencerequired a country to have its ownlarge-scale capital goods sector.

This view of development was mostclearly articulated by P.C. Mahalanobis

of the Indian Statistical Institute,7 whoargued that the countries must not onlychange their structure, but must changeit by creating a domestic heavy capitalgoods sector. The Indian Second Plan(1956–61) was greatly influenced by theMahalanobis view. Wilfred Malenbaum(1962, p. 87) shows that the investmentallocation for the second plan was virtu-ally equivalent to that worked out byMahalanobis in his operational researchexercises (Mahalanobis 1955). In both,about one-third of total investment wasallocated to “basic investment goods,”about 18 percent to industrial consumergoods, and 17 percent to agriculture.Equally important, there was essentiallyno effort to allocate resources optimallyin the usual sense. Other peopleshowed that the objectives could havebeen achieved with less capital than theplan called for, and that more jobscould have been created.8 Such find-ings were not looked upon as especiallyrelevant (or accurate), given the as-sumed (but not measured) externalitiesand the importance of the economic in-dependence objective.9

Mahalanobis’ argument fit well withthe structuralism of Prebisch and hisLatin American colleagues. There it waswidely assumed that factor prices, espe-cially wage rates and the exchange rate,had little effect on the quantity of suchfactors demanded or on the choice of

6 Many things were of course happening in theSouth prior to World War I, especially in LatinAmerica. The general statement in the text, how-ever, holds.

7 See in particular Mahalanobis (1955).8 Malenbaum (1962, ch. IV) is a good discussion

of India’s Second Plan and the role that Mahalano-bis’s work played. I have followed Malenbaumclosely in these paragraphs. A more general (andlater) study of the early plans is Bhagwati andPadma Desai (1970).

9 The Indian economy performed well duringIndia’s first plan, but by the end of the secondthere were severe problems. The problems in thelate 1950s were due in significant part to the diffi-culties in agriculture and in earning foreign ex-change. Anne Krueger in Theodore Morgan et al.(1963) is a convenient early reference, along withMalenbaum and Bhagwati and Desai cited in theprevious note.


production techniques; output and itscomposition were the determining fac-tors. In Latin America more than else-where, the strong structuralist view pre-vailed that wage rates could be high inorder to attack the poverty problemwith no cost in terms of employment.Similarly, the exchange rate did notmatter much for exporting, so its valuecould be set to achieve other objectives,such as inducing capital formation ordampening inflation.10

Given these arguments, many stu-dents and policy makers in much of theworld believed that the appropriatestrategy for development was to replaceimports from the rich North with theirown domestic production. Large-scalecomprehensive planning, rather thanthe market, was assumed to be the ap-propriate instrument, even though theunderstanding of how to design and im-plement a plan was as primitive as wasthe understanding of growth.

3.2 Key Role of Capital Formation

These ideas put primary emphasis oncapital formation as the source ofgrowth. The most obvious differencebetween firms in rich countries andthose in poor ones was the extent towhich physical capital was available towork with the labor. There are virtuallyno data on capital/labor ratios for theearly 1950s, but there is little doubtthat this ratio was vastly higher in theNorth than in the South in virtually allsectors, except for foreign firms en-gaged in mining of one kind or another.Thus another policy objective was to ac-celerate the rate of investment. Thisview was supported by a now-famousArthur Lewis statement: “The central

problem in the theory of economicgrowth is to understand the process bywhich a community is converted frombeing a 5 percent saver to a 12 percentsaver” (Lewis 1955, pp. 325–26).11 Thenearest thing available to a formal the-ory of growth was that of Roy F. Harrod(1939, 1949), in which the only speci-fied source of growth was capital forma-tion.12 Harrod specified a simple rela-tionship between increased output andincreased capital—the incrementalcapital/output ratio (ICOR). This ratiowas assumed to be constant due to tech-nological factors. There were numerousestimates of the ICOR, and most plansmade specific assumptions about itsvalue for sectors and for the economy asa whole. Such estimates were widelyused to determine the amount of newcapital required to achieve a givengrowth target.13 Most observers consid-ered domestic saving the primary con-straint, and the earliest arguments forforeign aid rested on the assumptionthat the savings of the poor countrieshad to be supplemented by foreign sav-ings if acceptable growth rates were tobe achieved.14

10 Richard Eckaus (1955), a widely read paper,worked through many of the implications of thisset of assumptions, particularly those that questionthe role of factor prices in the choice of tech-nique.

11 Lewis added after a dash, “with all thechanges in attitudes, in institutions, and in tech-niques which accompany this conversion.” Thispart of the sentence is not famous, and in anyevent makes the first part of the sentence lessclear cut. Lewis also defined the subsistence sec-tor as “that part of the economy which is not usingreproducible capital” (Lewis 1954, p. 149).

12 Harrod (1939 and 1948) noted that in theoriginal Keynesian formulation of income determi-nation, investment was necessary to achieve fullutilization of resources today, but this same invest-ment added to the capital stock tomorrow, andtherefore required an increase in output for fullutilization of resources to continue.

13 Benjamin Higgins (1959) and Harvey Leiben-stein (1957) have good brief discussions of theICOR and estimates of its value. There are alsoreferences to additional sources of estimates. BelaBalassa (1971, p 38) has estimates for a number ofcountries over the 1950s.

14 Moses Abramovitz (1952) is a much widerranging study of growth, its sources, and mecha-nisms than the Harrod formulation. Abramovitz


The allocation of capital within theboundaries laid down with respect tostructural change was an important is-sue. In some instances, the ICOR esti-mates were used as a criterion; that is,invest where the ICORs are smallest.There were efforts to identify specificobjectives to be served by the new in-vestment: for example to raise the sav-ing rate, increase exports, maximize anemployment effect, meet certain re-gional objectives, etc. Evidently thisway of thinking revealed further doubtsabout an effective market answer.15

The capital goods sector was small ornonexistent in the newly independentcountries, and most capital goods had tobe imported. An obvious way to encour-age investment was to maintain an ex-change rate that kept capital’s domesticprice low. This could be most easilydone by maintaining an overvalued ex-change rate (or, less frequently, multi-ple exchange rates). Overvalued ex-change rates (relative to a free tradesituation) appeared as a means to en-courage investment.16 This producedbalance of payment pressures, and to

counter these a variety of tariffs, importlicenses, and exchange controls wereput in place. Protection in many formswas afforded currently imported con-sumer durables (and now and then sim-ple capital goods) behind which domes-tic production took place. Consumergoods (especially durables), rather thancapital goods, were protected on thegrounds that their costs of productionin the developing country would berelatively less than those of capitalgoods, because production of the lattergoods was assumed to be more capitalintensive and to employ more complextechnology. Where Mahalanobis’ viewstrongly prevailed—India, Brazil, andpossibly other large countries—domes-tic production of capital goods was en-couraged by keeping out imports and bydirect subsidies.17 Even in such coun-tries, however, attention was givenmainly to protecting the domestic mar-ket from importation of consumer dur-ables.

3.3 Replicating the North

A third issue refers to the notion ofdevelopment that seemed to prevail inthe newly independent countries.Higher per capita incomes were ofcourse sought, but more generally theidea seemed to be to replicate theNorth. The very idea of import substitu-tion implied this: keep out that which isnow imported from the North and pro-duce it at home. Lewis (1954), the mostwidely studied article of the time, im-plied this. Lewis’ dual economy con-sisted of a small modern or capitalistsector and a large traditional sector.

explicitly excludes Harrod’s model from his surveyon the grounds that it is more nearly the require-ments for continued growth at full employmentthan it is an explanation of the growth that hasoccurred. Bert Hoselitz, ed. (1960) has a numberof chapters on growth theory in the 1950s.

15 Examples of this approach are Hollis Chenery(1953), Walter Galenson and Leibenstein (1955),and Alfred Kahn (1951). One of the earliest textson development, Gerald Meier and Robert Bald-win (1957, pp. 343 ff), is a useful general discus-sion. It was not until later that benefit cost studiesusing shadow prices became popular.

16 The evidence on the overvaluation of the ex-change rate usually took the form of estimates ofdomestic resource costs of production relative toimport costs and to direct estimates of relativepurchasing power. Less formally, estimates weremade of the costs of specific items—for example,haircuts—in countries, and compared to the ex-change rate. Black markets were common. Thatthe exchange rates were overvalued in numerouscountries in the 1950s and later is clear, but (aswill be emphasized later) what the exchange ratesshould be is not.

17 It is sometimes written that import substitu-tion was adopted in order to correct balance ofpayment problems. This seems to me to be incor-rect. There were controls of many kinds, but thesedid not originate to correct existing balance ofpayment problems. They emerged as a conse-quence of the particular approach to import sub-stitution that was widely followed.


The former was similar to the econo-mies of the North in terms of technol-ogy, the use of capital, and dependenceon a market mechanism. In the tradi-tional sector, labor productivity wasmuch lower, the marginal product mayhave been zero, virtually no physicalcapital was available, growth was ab-sent, and a conventional market mecha-nism was not in place. In particular,payments to labor were likely to equalits average, rather than marginal, prod-uct. Lewis argued that all new invest-ment should be in the modern sector;labor would be pulled in from the tradi-tional sector as needed, and the modernsector would expand while the tradi-tional sector diminished. Eventually,the modern sector would encompass thewhole economy, and the traditional, un-derdeveloped economy would havebeen made over in the image of the richNorth.

Lewis placed primary emphasis oncapital formation (as noted above); sucha mechanism depended very directly onthe capacity to import physical capitaland to put it in place within a tradi-tional society. The institutions, organi-zations, and values of the traditionalsector were to be pushed aside if theyimpeded this process. It was thereforeas much a theory of displacement as itwas a theory of development, becausethe existing traditional economy was tobe replaced with an imported one.18

One might then say that this approachwas “import oriented” as well as importsubstituting. This approach may be con-

trasted with that which seeks to create asustained dynamism within the tradi-tional economy. Lewis (and especiallyJohn Fei and Gustav Ranis 1964) em-phasized the importance of agriculturaldevelopment, but this emphasis waswidely ignored.19

This way of thinking reflected consid-erable optimism: just get the invest-ment rate up, import the capital (withits built-in technology) and soon theend of underdevelopment would be athand. There was general optimism else-where as well. President Truman’s in-augural address in January 1949 re-flected the view that the North(especially the United States) could(and would) transfer physical capitaland technical and administrative knowl-edge to the new countries so that theytoo could become modern, like theNorth. The idea that development was asimple matter easily executed with for-eign aid was reflected in much of thepopular writing and in many of the re-ports and studies of international andbilateral aid agencies.

These three areas of thought and pol-icy describe what I believe to be the ba-sic way of thinking that produced im-port substitution. Details of policyvaried widely around the world, and theextent to which a particular country’sdecision makers thought along theselines depended on a great variety of ad-ditional factors.20 There were many

18 On the notion of development as displace-ment in Lewis and elsewhere, see Peter Gran(1980) and Marshall Wolfe (1976) and the litera-ture cited there. In his book on growth (Lewis1955, p. 80), Lewis notes that “backward” societies“can grow simply by modelling themselves on themore dynamic features of the more advanced.”There are many arguments and illustrations inLewis (1955) that suggest this notion of displace-ment.

19 Creating a continuing dynamism within thetraditional sector has been emphasized by severalagricultural economists, for example BruceJohnston, Vernon Ruttan, John Mellor, and others,and by sociologists and geographers interested indevelopment. See, for example, Peter Berger andMichael Hsiao (1988). Albert O. Hirschman in nu-merous places emphasizes such a notion, for exam-ple in Gerald Meier and Dudley Seers (1984).Michio Morishima (1982) and Tessa Morris-Suzuki (1989, 1994) reflect this idea, as doesBruton (1985, 1997).

20 Literature, in addition to that already cited,was generally consistent with this rationale. Therewere many guidelines for planning and for ensur-


country plans written in the 1950s,often with the help of foreign advisorsand at the behest of lending agencies.Such plans were generally rather unso-phisticated and were often mainly a listof public sector projects, but they didreveal the assumptions examined above.The United Nations, for example, pub-lished many studies of planning meth-ods and techniques. The United Na-tions Commission for Asia and the FarEast (1960) is a good example. Ian Lit-tle (1982, ch. 3) is a helpful brief surveyof the general issues involved in plan-ning and industrialization in these earlyyears of development.

One can conclude with confidencethat the import substitution strategyand its rationale, as described above,were widely accepted. There were twomajor dissenters: Gottfried Haberler(1950) and Jacob Viner (1952). Theirbasic position was essentially that freetrade and full reliance on an interna-tional market was, without a doubt, theappropriate strategy. Their views hadlittle impact in the developing countriesthemselves or in the academic develop-ment community in the 1950s as the de-velopment efforts began to take hold.

4. Approaches to Implementation

The general rationale of import sub-stitution created three broad policy is-sues: how to provide the protection thatwas deemed necessary, how to increasesaving and investment, and how to goabout the planning process. In none of

these areas was formal textbook eco-nomics of much help. Reliable dataabout the newly independent econo-mies were sorely lacking, and many off-the-cuff policy decisions had to bemade in the face of abundant igno-rance.21

There were more general problems aswell. The understanding of the variousforms of protection was quite primitive,even among academic economists. Ideasthat became important 20 or so yearslater—value added protection, advan-tages of a fairly uniform tariff rateacross the board, real exchange rates—were essentially unknown as the newcountries and new leaders began theirtasks. In many countries the planningagency had relatively little power andwas often ignored by the old mainlineministries. There were numerous for-eign advisors, many with little experi-ence in planning. The lack of confi-dence in an international marketmechanism was thus matched by wobblyunderstanding of how to plan, as well asby incomplete data and other materialsnecessary for effective planning.22

4.1 Planning

In none of the developing countriescould planning be completely control-led and implemented at the center. In-

ing “balance.” Paul Rosenstein-Rodan (1943) andRagnar Nurkse (1953) paid special attention to therole of government in ensuring balance in invest-ment allocation. The former author argued also forthe need for a “Big Push.” Gunnar Myrdal (1956)argued that cumulative processes were dominantin the world economy, and so countries with ahead start always had an advantage unless specificactions were taken. Leibenstein (1957) urged theidea of a “critical minimum effort,” similar to a“Big Push.”

21 One of the earliest comprehensive studies ofimport substitution was conducted at WilliamsCollege in the late 1960s and early 1970s. The pro-ject produced more than 50 empirical and theo-retical papers on the subject in general and on theexperiences of particular countries. Bruton (1970)and John Sheahan (1973) are reviews of the find-ings of this project funded by the United StatesAgency for International Development. More in-fluential was the study of Ian Little, Tibor Sci-tovsky, and Maurice Scott (1970). This work andthe six country studies that were done in conjunc-tion with it were extremely wide ranging and semi-nal. The project was sponsored by the OECD. Seealso Hirschman (1967).

22 Albert Waterston (1965) is a good descriptionof the problems of planning in the 1950s. Thereare many details and examples of country plans inthis book.


deed this was explicitly recognized inmost countries. Indian officials, for ex-ample, made it clear that they did notaim at eliminating the private sector,and this is more or less clear in mostother plans.23 Then the task of deter-mining where to engage in direct physi-cal planning and where to leave it to theprivate sector became evident. Bhag-wati and Desai (1970, p. 5), for exam-ple, note that major irrigation projectswould yield low returns unless feeder-channels to take water to the fieldswere also built. Could these be left tothe farmers or must they be included inthe plan? If the latter, the plan wouldbe so detailed as to make implementa-tion unlikely, and if the former, therewould be heavy dependence on exten-sion workers as well as on the farmersto get a very complex job done. AlbertHirschman (1967) found similar issuesarising in Latin America. So, along withphysical planning in which public in-vestment played a significant role, aneffective market was necessary to createsignals and inducements that wouldlead the private sector to respond. Thenotion of a plan and market coexistingadded to the complexities of policymaking.

Despite these difficulties, planningbecame widespread over the 1950s. Bythe early 1960s virtually every develop-ing country had something that wascalled a plan. The World Bank and theUnited States’ and other countries’ for-eign aid programs pushed hard to getplans in place. In some instances a planwas necessary before loans and aid ofvarious kinds were dispersed. Aid agen-cies often financed foreign advisors tohelp with the planning process.

The plans varied widely in terms ofsophistication and detail. India, with its

highly competent civil service, was per-haps the most impressive at this time,but some other plans were often quitedetailed. Almost all plans announced agrowth (of GDP and some sectors) tar-get and then allocated the anticipatedinvestment among the sectors of theeconomy believed necessary to achievethe target. The ICOR was the basic in-strument to determine the amount ofinvestment needed to meet the targets.All plans concentrated attention on in-dustrialization, some more than others,but in all the main idea was to changethe structure of the economy in orderto grow and to become more indepen-dent of other countries.24 The size ofthe plan led to many disputes betweenoptimistic planners (often including for-eign advisors) and cautious finance min-isters and central bankers. This issuehad obvious links to the macro stabilityissue and to availability and accept-ability of foreign aid.25

4.2 Other Instruments

The import substitution idea, by itsvery nature, involved protection, andfrom the beginning of the 1950s virtu-ally all developing countries began toput in place a variety of instruments toprotect their economies from a largenumber of imports.

23 See Waterston (1965) and Malenbaum (1962)for more detail on this point.

24 All plans of course had something to sayabout agriculture and about rural development ingeneral. The general thrust of the documents how-ever was clearly to industrialize. Waterston (1965,ch. III) has a number of quotations from variousplans that make this point and the “control of eco-nomic destiny” point quite unambiguously.

25 The two-gap model that distinguished be-tween a domestic saving and a foreign exchangeconstraint became popular and had implicationsfor foreign aid. The chief architect of the two-gapmodel was Chenery, who collaborated with numer-ous others. The main paper is Chenery and AlanStrout (1966). There were numerous papers priorto this one. This model added import coefficientfixity to production coefficient fixity of the Harrodmodel.


Tariffs. Tariffs were levied on arather ad hoc basis. In many instances,nominal tariff levels appeared to be de-termined simply by what was deemednecessary to allow an activity into exist-ence. The consequence was that a greathodgepodge of rates appeared, with vir-tually no evidence of any considerationof costs or efficiency. Effective rates ofprotection (ERPs) began to be calcu-lated in the late 1960s, and these weregenerally higher, sometimes muchhigher, than nominal rates. That ERPsusually exceed nominal rates reflectsthe escalation of nominal rates fromlower to higher levels of fabrication inmanufacturing activities. There werenumerous negative rates, and the vari-ance of rates in a given country amongactivities was high.26 Negative rateswere found most frequently on tradi-tional exports, further dampening theinducement to export.27

Though the estimated values of theERPs applied mainly to the 1960s, theirimplications were not clear until theybecame widely known around 1970 orso. Even then, most authors were care-ful to point out the grave difficultiesthat had to be faced in interpreting theresults. There were ubiquitous dataproblems, several possible formulae, as-sumptions about substitution possibili-ties, indirect taxes, and so on. Despitethese difficulties, it was widely believed

that ERPs had important effects onwhat was imported and what was not,and hence on the allocation of invest-ment.28 Since the array of rates showedvery little evidence of an economic ra-tionale, but were deemed important inallocation decisions, the conclusion wasobvious: new investments were beingmade in a haphazard way from the stand-point of customary economic arguments.

A particularly interesting point aboutthe ERP picture as it evolved is that anumber of countries, later achievingoutstanding success, showed the samesort of protection picture as did laterfailures. An obvious example is Taiwan.As late as 1966, Taiwan’s ERP for con-sumer goods was higher than that of thePhilippines and vastly higher than thatof Mexico. For the whole of the manu-facturing sector, Mexico’s rate (in 1960)was also much lower than that for Tai-wan, and that for the Philippines onlyslightly higher. Average values for theentire manufacturing sector are some-times misleading, as often there will benegative values for certain categorieswhich offset very high values elsewherein the sector. Little (1982, p. 136 ff) ar-gues that estimates of ERP do conformto what one would expect: lower levelsand variance of ERPs for the North andfor most of the more strongly perform-ing countries. Evidently the role ofERPs is still ambiguous.

Protection provided by nominal andeffective tariff rates was supplementedby numerous quantitative and directcontrols on the availability of foreignexchange. These were often imposed asa quick fix to the balance of paymentsproblems that emerged as the importsubstitution (IS) policies continued,rather than as a means of implementingthe basic objectives of the import sub-

26 There are now many estimates of effectiverates of protection. The first ones to attract wide-spread attention were those in Little, Scitovskyand Scott (1970) and Balassa and Associates(1971). These volumes made clear the high valuesof the rates of protection and their great varianceamong the sectors of a country. I think that it isfair to say that most members of the professionwere surprised at these results.

27 Negative value added at world prices for someactivities was found in a number of countries.When they were first found, they attracted consid-erable attention, even excitement. This excitementdied down rather quickly. Bhagwati and Desai(1970, pp. 363 ff) discuss the many reasons whynegative value added at world prices might appear.

28 Max Corden (1971) studies the various issuesassociated with ERPs, and Little (1982, pp. 136 ff)is a good, brief summary.


stitution strategy. Import licenses werewidely used to ensure that productsdeemed essential for consumption orfor creating new productive investmentwere available. To some extent these di-rect controls followed from the plans,and yet in most instances their designand execution added to the distortions.

Exchange Rates. The other major pol-icy instrument was the exchange rate.Attention has been called to the wide-spread practice of overvaluation of do-mestic currency (relative to a free tradesituation) in order to keep the domesticprice of capital low and, in some in-stances, to dampen inflation. The otherfactor affecting the exchange rate wasthe frequent evidence of Dutch Dis-ease. Many countries had one or twomajor export items, and when the worldmarkets for these products were favor-able, an exchange rate that protectedthe balance of payments made it diffi-cult to export other (than the major)items and especially to develop new,nontraditional exports. An overvaluedexchange rate added to the difficulty ofeffecting the Hirschman kind of link-ages, as it was so cheap to import thatproducers spent their time and energytrying to get foreign exchange ratherthan searching for ways to produce withdomestic resources.

The overvalued exchange rate was de-fended, not only as a subsidy to capitalformation, but also on the grounds thatthe economy would grow into thehigher rate as the productivity of capitaland labor increased. A failure of pro-ductivity to grow meant, among otherthings, that the exchange rate remainedovervalued, and exporting continued toface an extra high hurdle.

4.3 Further Sources of Distortions

There were numerous sources of mis-leading price signals. Interest rate sub-sidies were prevalent and negative real

rates common. They too were defendedas an encouragement to investment.

A source of distortion not directlyconcerned with trade, but highly rele-vant to the form that investment took,had to do with wage rates. In a laborsurplus economy, most models keepreal wage rates constant until full em-ployment is approached. In fact, wagestended to rise, especially in the formalsectors. In most instances the increaseswere attributed to the usual array of in-stitutional factors that can push wagesup beyond what the market alone woulddictate.29 Especially in Latin America,wage rates were raised or allowed torise as a matter of government policyaimed to affect the distribution of in-come. In some countries, wages in thedomestic economy were affected by thestrong possibility of external migration.The Middle East in the 1970s becamethe prime example of this kind of effecton wages. In Egypt, for example, skilledlabor—plumbers, electricians, carpen-ters, etc.—could find jobs in the Gulfcountries at 10 to 20 times their wage inEgypt. In such an environment, theright wage for such labor in Egypt is notimmediately evident. There were someexamples as well in parts of Latin Amer-ica and South Asia.

There were two main indicators ofdistortions in the use of capital: under-utilization of the capital stock, and thefailure of favorable investment rates tomake a significant dent in widespreadunderemployment. The latter result wasexplained in large part by the capital in-tensity of the capital imported from theNorth, combined with the general in-ability of people in the South (and thelack of incentives) to adapt the im-ported technology to the domestic fac-

29 Lewis (1979) notes that his original model ismost in need of modification in the assumptionswith respect to wages and the movement of laborfrom the traditional to the capitalist sector.


tor supply situation. Some prevailingideas—the Harrod model, the input-output models, the two-gap notion—im-plied that adaptation was not feasible.This failure of employment in produc-tive sectors to grow was a prime factorin the continued prevalence of large-scale severe poverty.

The emergence of underutilization ofcapital was more perplexing. The smallcapital stock relative to labor and otherresources was widely deemed to be theprincipal explanation of the low laborproductivity in the South. And, as justreviewed, a principal objective of policywas to increase the rate of capital for-mation, yet underutilization of existingcapital appeared increasingly common.30

Explanations were in terms of the un-availability of intermediate goods and ofappropriately qualified labor; limited de-mand for specific outputs; hoarding inanticipation of changed government poli-cies; the practice of importing capitalwhen one received permission or foreignexchange rather than when one genu-inely wanted the capital; the practice ofgranting import licenses based on in-stalled capacity; protection and subsi-dies for capital goods that could makeinvestment profitable at low utilizationrates; and foreign aid that often sup-plied capital that had few links with theneeds of the economy. A situation inwhich capital formation was being pushedhard while widespread underutilizationof existing capital prevailed clearly rep-resented a severe and costly distortion.

4.4 Agriculture

There was particular difficulty in ap-preciating a role for agriculture in the

general IS package. Agriculture, muchthe largest sector in most countries, wasexpected to decline in relative terms. Itdid not seem generally appreciated thata sluggish agricultural sector would al-most inevitably penalize the growth ofnonagricultural sectors. New manufac-turing activities could not export, in partbecause of the policies just reviewed, andso had to depend on domestic demandwhich in turn was heavily dampened bythe slow growth of the largest sector.There was widespread “squeezing” ofagriculture in order to finance the newmanufacturing activities. Squeezing wasdone by price controls on many basic fooditems and by other forms of taxation.Part of the rationale of controlling foodprices was to help prevent living costs(and hence wages) from rising in urbancenters where the new activities wereconcentrated. The confusion about arole for agriculture was especially dam-aging, and I return to it later.31

A Brief Summary. The picture de-scribed above became known as the im-port substitution syndrome. That syn-drome included reliance on a centralplanning effort of greatly varying effi-cacy; a set of nominal tariffs and ERPsthat generally showed little economicrationale; quotas; exchange controls;overvalued exchange rates that contrib-uted to unemployment and underutili-zation of capital in capital-scarce econo-mies, and penalized exporting; and, inmany countries, a difficult wage-settingsituation. In most countries, agriculturewas also penalized in one way or another.The justification for all this seemed tobe the assumption that once the struc-ture of the economy was changed,learning would occur automatically andresolve the difficulties. Learning, how-ever, proved more difficult.

30 Gordon C. Winston was perhaps the first tonote that capital was in fact widely underutilizedand that such underutilization had important im-plications for growth and growth policy. Winston(1974) is his major paper on the subject, and therewere a number of papers previous to it.

31 It is useful to note again that Lewis (1954,1955) and Fei and Ranis (1964) gave a great dealof attention to agriculture.


5. Development in the Fifties and Sixties

5.1 Evidence of Success

Despite increasing distortions in theeconomies, there were important devel-opments in the two decades after 1950.The stagnation that characterized theinterwar period was broken. Data fromAngus Maddison (1970), shown in Table1, permit a comparison of the 1950s and1960s with prewar years for a numberof variables. The rates of growth ofGDP are markedly higher after 1950,except for Argentina and Malaya, com-

pared to the interwar decades. In manycases the differences are substantial andare not obviously associated with theextent of commitment to import substi-tution. Investment rates are higher,except in Argentina. Even agriculture,generally penalized in one way oranother, grew faster in all countriesexcept Argentina and Colombia, al-though often the differences weresmall.

Exports are more mixed. There wereeight countries in which the dollarvalue of exports grew faster—some-times much faster—than during the

TABLE 1GROWTH RATES BEFORE AND AFTER 1950

Average Annual Rates of Growth

GDPPhysical Volume of

AgricultureDollar Value

of ExportsNonresidential Gross

Investment Rates

1913–50

1950–60

1950–68

Prewarto

1945–50

49/51–64/66

1913–37

1950–67 1950

Average1950–66

Argentina 3.0 3.1 3.0 .1 .1 1.6 .4 14.7 14.3Brazil 4.6 5.8 5.3 1.3 4.0 .4 1.2 11.3 12.5Ceylon n.a. 3.6 3.6 2.0 2.5 2.1 .3 7.1 10.0Chile 2.1 3.5 4.0 1.8 2.0 1.1 7.1 9.6 11.7Colombia 3.7 4.6 4.6 4.1 2.4 4.1 1.5 11.4 12.9Egypt 1.6 5.4 5.2 1.0 3.1 1.0 .6 10.2 11.5Ghana 3.8 4.8 3.9 n.a. 4.8 3.6 2.2 7.4 13.1India 1.2 3.7 3.8 –.1 2.5 –.1 1.9 7.4 11.2Malaya 4.3 2.3 4.1 2.3 2.6 4.2 .7 5.0 10.3Mexico 2.6 6.1 6.2 3.6 5.6 1.6 4.7 10.8 13.2Pakistan 1.2 2.7 4.1 .4 1.9 –.1 1.2 4.8 9.9Peru 3.2 5.1 5.4 2.1 2.9 3.2 8.6 11.0 15.3Philippines 2.2 4.8 5.2 1.1 3.1 4.9 5.4 7.4 10.7South Korea n.a. 6.1 7.1 –2.1 5.0 4.3 16.8 5.7 8.8Taiwan 2.7 7.7 8.7 –.4 5.3 6.8 13.4 9.8 12.2Thailand n.a. 6.0 6.5 2.7 4.5 2.4 4.9 11.9 13.9Turkey n.a. 5.8 5.5 1.7 4.4 .7 4.1 6.9 10.0Venezuela n.a. 7.6 6.3 n.a. 5.4 8.0 5.5 20.0 21.1

Source: Maddison (1970).


interwar years, while in other countriesgrowth was slower. This picture is aboutwhat was anticipated by most observersat the outset of the import substitutioneffort. Maddison (1995) provides datafor the years 1950–73 for more coun-tries, but the picture is about the same.Nine African countries averaged 4.5percent per year over the period, andseven Latin American countries aver-aged 5.2 percent.

Life expectancy at birth rose in mostcountries, and infant mortality fell. Lit-eracy rates rose, infrastructures (roads,irrigation facilities, schools) were devel-oped. In some countries the GreenRevolution took a firm hold. Manufac-turing increased as a proportion of GDPand imports began to change to reflectthe import substitution objective of im-porting capital goods while producingconsumer and simple capital goods athome. An index in Little, Scitovsky, andScott, (1970, p. 245) shows that manu-factured exports from all developingcountries increased from a base year of100 in 1953 to 283 in 1965. The experi-ence of these years, often identified asthe easy stage of import substitution,created considerable hope amongeconomists and country leaders as well.

This array of achievements was sig-nificantly higher than anyone antici-pated in the early 1950s.32 The firstWorld Development Report (1978, p. 3)stated that, on the basis of modest evi-dence, growth in the 1950s and 1960swas “a substantial improvement of thehistorical record.” Indeed, one can gofurther and say that if the rates ofgrowth of GDP and levels of investment

that were in fact achieved had been an-ticipated in the early 1950s, mosteconomists at that time would have be-lieved that the major developmentproblems of most countries would besolved by 1975.

During the 1960s, however, price dis-tortions became increasingly evident,and the publication of the multi-countrystudies of Bela Balassa and Associates(1971) and Little, Scitovsky, and Scott(1970) were especially clear on the ex-tent and nature of these distortions.Both sets of studies gave attention tothe exchange rate, and both allowedsome protection (or subsidy) for thedevelopment of domestic industry. Bothnoted that agriculture was beingpenalized, although they were cautiousabout attributing the penalty to thebroad strategy of import substitu-tion.

That exports were penalized was alsorecognized, and arguments were pre-sented to the effect that foreign tradehad dynamic as well as the textbookstatic allocation effects. These argu-ments were quite vague and cautious,little advanced beyond those of Haber-ler and Viner previously referred to.Little, Scitovsky, and Scott (1970,p. 345) state that “it is not suggested, ofcourse, that a country should export forexports’ sake,” and add (p. 346) thatthey believed that export pessimism“has been greatly overdone.”

These studies are also cautious aboutgeneralizing about policy. In referringto agriculture, for example, Little, Sci-tovsky, and Scott (p. 349) state that“somehow a balance has to be struckbetween drawing some resources fromagriculture, and not discouraging out-put too much.” They emphasize that ag-riculture development is not simply amatter of price incentives. These arequite qualified conclusions, and in lightof the developments over the previous

32 In an article that attracted considerable atten-tion, Rosenstein-Rodan (1961) made projectionsfor 66 developing countries for the following 15years. Of these, 17 were deemed correct and 43proved too low. See David Morawetz (1977, pp. 21and 85). Morawetz also cites other projections thatmore often than not proved below the rates actu-ally achieved.


two decades, not the all-out attack onimport substitution that came later. Norwas there a strong condemnation of asignificant role for government at thistime. The World Development Reportfor 1978 was not really critical of thestrategy and the policies, and in severalinstances expressed considerable sym-pathy for the approach.

5.2 Emerging Problems

The collections of Balassa and Littleet al. made abundantly clear that importsubstitution policies had created majordistortions and had thereby resulted ina misuse of resources.33 So it was in-creasingly evident that somethingneeded fixing. Two other sources ofconcern appeared in the early 1970sand suggested that there were otherthings that needed fixing as well.

The first, already referred to, wasthat the demand for labor in the newactivities was growing more slowly thanthe rates of growth of output and in-vestment had led most observers to ex-pect. As a consequence of the slowgrowth of employment (and otherthings), poverty was alleviated onlymodestly, or, in some instances, wors-ened.34

The second concern had to do withproductivity growth. At the end of the1960s, the role of productivity growthhad begun to be appreciated more fully,as the Solow residual became more

widely understood and estimates of itsvalue increasingly available. AngusMaddison (1970) provides a number ofestimates of total factor productivity(TFP) growth for 18 non-European de-veloping countries for the decade 1950–60. The average rate was .27 percentper year. If Taiwan’s 3.30 percent is re-moved from the list, the average falls toless than .10 percent per year (Mad-dison 1970, p. 53). There were manynegative rates as well. Later estimatesof TFP growth (e.g., Chenery, ShermanRobinson, and Moshe Syrquin 1986, pp.20–22) showed rates somewhat more fa-vorable for the developing countries forthese decades, but well below those ofthe North. Understanding of the role ofTFP growth was only beginning in1970, but the low estimates were recog-nized by many as evidence that importsubstitution was not proceeding as itwas originally expected to do (Bruton1967). Economists for the most part,however, could say very little at thattime about the origins of productivitygrowth.

5.3 Africa

African countries need a special com-ment. The physical and human capitalavailable to the Sub–Saharan Africancountries at their independence was, ingeneral, much less than that available todeveloping countries elsewhere. Liter-acy rates were much lower, the laborforce was less experienced and lesstrained, saving and investment rateswere lower, infrastructure—roads,power facilities, institutional develop-ments—were much less extensive andmarkets less complete. The new stateswere often ill-defined as to geographicboundaries and extent of governance.Ethnic, language, and tribal diversitywas (and remains) rich with opportuni-ties and dangers. Since most Africancountries did not achieve their political

33 This feature of import substitution was appre-ciated much earlier by Prebisch. Prebisch (1964)emphasized the pressing need for import substitu-tion in Latin America, and favored an attempt toproduce internally all imports that were notdeemed essential, but then stated that “the crite-rion by which choice was determined was basednot on considerations of economic expediency, buton immediate feasibility, whatever the cost of pro-duction” (p 71). Santiago Macario (1964, p 61)spoke of the “extemporaneousness” of protectionin Latin America.

34 Hans Singer (1970) was among the first ob-servers to call attention to these features of thegrowth that was taking place.


independence until the late 1950s andearly 1960s, less attention was given tothem than to Latin America and Southand East Asia in the literature of the1950s and 1960s.35

Most of the smaller African states didnot actively pursue an import substitu-tion (or any other) strategy. An array ofimport substitution measures did ap-pear in some of the larger African coun-tries (Kenya, Zimbabwe, Ivory Coast,Nigeria) in the 1960s, but such mea-sures seemed even more ad hoc than inother parts of the developing world.The anticipated structural changesshowed few signs of occurring. RogerRiddell (1990, p. 38 ff) argues that im-port substitution did not fail in Africa,rather it was never really tried.36 In the1960s and 1970s, there was substantialpublic sector investment in manufactur-ing in many African countries, oftenwith the direct involvement of foreignfirms and the World Bank. Such invest-ments later became sources of difficul-ties of many kinds—balance of pay-ments problems, budget deficits, andproductivity growth in particular. (Jef-frey James 1995 has a good summary ofthese issues and additional literature ci-tations.)

Professional economics did not offerthe Africans much help as they becameindependent. The first two World Devel-opment Reports (1978 and 1979) for ex-

ample, supported short-lived protection(exact form not specified), infrastruc-ture investment, and labor-intensiveproduction. The 1979 World Develop-ment Report (p. 69) added: “Techno-logically more sophisticated productsmay need to await the development ofadequate supplies of skilled labor andtechnological capacity.” A brief sen-tence urges incentives to keep agricul-ture growing in order to provide de-mand for locally produced manufacturedproducts. Other conventional sourcesof economic analysis and prescriptionwere equally sweeping and truistic.

In the late 1950s, the 1960s, and the1970s, for the most part, African coun-tries grew more slowly than most otherdeveloping countries. There were im-portant exceptions however. Kenya,Nigeria, Tanzania, Ivory Coast, andMalawi were countries that did reason-ably well over these years. It is difficult,however, to attribute their success totheir foreign trade strategies. Botswana,a small country in terms of popula-tion, achieved extra-high growth rates,largely, but not entirely, as a conse-quence of its diamonds. Some othercountries, however, more or less col-lapsed in the 1970s, and toward the endof that decade difficulties for almost allAfrican countries were severe indeed.Many countries experienced fallingGDP per capita.

One could point to external condi-tions for the difficulties of the 1970s:the oil price increases, the breakdownof the Bretton Woods agreements, theslowdown in the growth of the West(especially in the last half of the dec-ade), and the decline in foreign aid. Allof these events affected all developingcountries, but African countries wereleast equipped to deal with them. Itdid seem that African countries weredifferent in some fundamental wayfrom Latin American and Asian

35 Neither Little et al. (6 countries) nor Balassa(7 countries) included an African country. Thelater Bhagwati/Krueger studies included Ghanaamong its 10 countries, and the still laterKrueger/Bhagwati studies on employment had 2African nations among their 10. Juergen Donges(1976) study of 15 semi-industrialized countriesincluded no African countries.

36 The World Bank’s World Development Reportfor 1987 shows Cameroon and the Ivory Coast as“moderately outward oriented” during 1963–73,but no African countries were so classified during1973–85. Kenya, Madagascar, Nigeria, and Senegalwere classified as “moderately inward oriented” inthe later period and five other African countries as“strongly inward oriented.”


countries.37 In particular, simply get-ting prices right and opening the econ-omy did not appear adequate, even tomany who believed that was all that wasneeded in other developing countries.Africa above all needed to find an effec-tive way to get agriculture growingsharply.

5.4 What Was Learned During the Fifties and Sixties

Despite evidence of some success,the profession had begun, by 1970 orso, to have doubts about the develop-ment process that seemed underway.Something was wrong, and certain ideasand assumptions, accepted uncriticallyin the 1950s, began to be questioned.Severe distortions, modest headway inreducing underemployment, poverty,and inequality, and little evidence ofproductivity growth were incompatiblewith long-run development. A numberof specific changes in prevailing viewssurfaced. The following are the mostimportant.

1. Much of the world was experienc-ing a boom of remarkable proportionsin growth of output and especially of in-ternational trade. The 1950s and 1960sturned out to be not remotely like the1930s, and inflation became much moreof a concern than recession or deep de-pression. The main effect of these boomdecades was to undermine the argu-ment that developing countries couldnot export.

2. Economic agents at all levelsturned out to be more responsive toprice incentives than was thought to bethe case at the beginning of the 1950s.

3. The assumption of fixed produc-tion coefficients (ICORs, input–outputratios, the Harrod model, the two-gap

idea, etc.) was proving exceedinglydamaging. This development was due toaccumulated empirical evidence on in-put combinations,38 and to the increas-ing awareness of the neoclassicalgrowth models of Solow, Swan, andMeade that provided convincing alter-natives to the fixed coefficient models.

4. Physical planning was not prevent-ing bottlenecks and misallocations. Thefailures of the Soviet system were alsobecoming appreciated in many places.

5. Numerous studies convinced mostobservers that the strong views of the1950s that the terms of trade had dete-riorated for the developing countrieshad to be significantly qualified.

6. Imports rose faster than expected,and hence balance of payment problemswere widespread, and “economic inde-pendence” was even lower than before1940. The faster growth of imports wasdue mainly to the demand for capitalgoods and intermediate goods to sup-port the new industries.39

7. The work of Abramowitz, Kuznets,Denison, and Solow on the sources ofgrowth made it increasingly clear thatsimply more physical capital was notsufficient for sustained growth. Theproductivity of resources had to in-crease if growth was to be maintained.40

8. The transfer of technological, ad-ministrative, and marketing knowledgewas proving to be much more complexthan was expected in the early 1950s.With fixed production coefficients andimported physical capital, it was diffi-

37 Tyler Biggs, Manju Shah, and Pradeep Srivas-tava (1995) is an especially useful study of the dif-ficulties that African firms have in increasing theproductivity of their resources.

38 The Constant Elasticity of Substitution pro-duction function made it easy to estimate the elas-ticity of substitution.

39 The import intensity of import substitutionwas first made clear in Carlos Diaz-Alejandro(1965). This rapid growth of imports led to an em-phasis on saving foreign exchange, a practice thatcaused more problems than it solved.

40 The conclusion that productivity growth waskey to the rapid growth in East Asia has recentlybeen questioned by a number of economists. I dis-cuss this in a later footnote.


cult to understand why productivitycould be lower in one country than an-other, and why it would not growequally fast in all countries. The term“infant industry” implied that simplygetting older and larger would result inincreased productivity, but this was nothappening.

9. Items 7 and 8 provided strong evi-dence that indigenous learning pro-cesses generally were not emerging inthe import substituting countries. The(implicit) assumption that simply chang-ing the structure of an economy wouldalso change its capacity to learn and toaccumulate knowledge was evidently in-correct. The task was much more com-plex.

This set of ideas and circumstancescreated a different theoretical and pol-icy-making environment from that pre-vailing in the 1950s and early 1960s,and implied a strategy quite differentfrom that which had led to the earlierpolicies. Thus a new approach had to bedetermined. It became quickly evidentthat changing basic policies from thosein effect for some time is more difficultthan making a policy on a virtually cleanslate (as had been possible at the timeof independence for many countries),simply because new policies may wellharm existing interests, including thoseof some government leaders, who will,where possible, fight any change. Thetask for the policy makers was thus tomove away from import substitution to-ward an unclear alternative in the con-text of pressure groups that often sup-ported the status quo.

6. Korea and Taiwan and the Rise ofOutward Orientation

That the import substitution syn-drome could not produce sustainedgrowth was generally recognized by1970, but the only alternative was to

eliminate distortions and maintain ahigh investment rate. It was the increas-ing awareness of the Taiwanese andSouth Korean successes that led to theemergence of a new development strat-egy built around exporting.

6.1 The Role of Taiwan and South Korea41

The exceptional performance of theRepublic of Korea and Taiwan in the1960s had, by 1970 or so, attracted theattention of the development commu-nity. Their growth rates of GDP in-creased markedly in the 1960s relativeto the 1950s: from about 6.5 percent toover 10 percent in Taiwan, and from 4.4to 9.1 in Korea. These increases weremuch larger than those for any othercountry. Employment growth and pov-erty alleviation were also proceedingmuch better than in most other coun-tries. Taiwan and Korea had mademarked policy changes in the late 1950sand early 1960s that reduced distortionsand encouraged firms to export. Exportshad responded remarkably well. In bothcountries the rate of growth of exportsexceeded that of GDP during the 1960s,and this strong growth was largelyamong nontraditional products and ser-vices. By the early 1970s or so, Taiwanand Korea were widely thought to bemarket-driven economies with minimalgovernment, though this view began tobe disputed early on.

The remarkable growth and exportperformance of Taiwan and Korea inthe 1960s after their changed policy,

41 Hong Kong and Singapore are generallyadded to Korea and Taiwan in discussing the riseof an outward orientation strategy. These two citystates were, in my view, different enough from theother two that it is misleading to include them inthe present kind of study. The main way that Sin-gapore and Hong Kong were different is that, assmall city states, they had essentially no alternativeto openness and trade. Some brief references aremade, but in general I will not consider them.


combined with evidence from othercountries (e.g., Brazil and Colombia atcertain times) that manufactured ex-ports were generally feasible, producedthe great interest in outward orienta-tion and export promotion. It was not,so far as I can ascertain, any sort of newtheoretical developments or insightsthat emerged independently of events.42

The view that both were essentiallymarket economies lent additional sup-port to the view that markets do work indeveloping countries and that planningdoes not.

Through most of the 1950s, Taiwanused trade and exchange rate policies tolimit external competition. There weretariffs, a fairly high set of ERPs, andmultiple exchange rates, most of whichrepresented an overvaluation. All for-eign exchange had to be turned over tothe Central Bank, and demand for for-eign exchange greatly exceeded supplyat the prevailing exchange rates. Publicsector imports were given a preferencerelative to private sector requests. Theimport substitution syndrome appearedin full regalia.43 Then in the late 1950sTaiwan began to dismantle this array ofcontrols, to establish a single exchangerate at a realistic level, and to encour-age exporting in a number of otherways.

The Korean conversion story is simi-lar.44 The 1950s were characterized bymarked inflation and balance of pay-ment problems. An objective of the

overvalued exchange rate was to helpbring inflation under control, as well asto subsidize investment. The high tar-iffs, quotas, and other interventions inforeign trade were then necessary toprotect the balance of payments. Therewere some efforts to encourage exportsin the 1950s, but these were modest.

The big change for Korea began inthe early to mid–1960s. A unified ex-change rate was put in place at a deval-ued level compared to the 1950s, andfurther export incentives were intro-duced. These included preferential ac-cess to funds; tariff exemptions onintermediate and raw materials, andcapital goods; reduction of direct taxeson profits from exporting; and other in-ducements that made firms eager to tryto export. The export incentives seemedstronger in Korea than in Taiwan, but itis difficult to be sure, since the ex-change rate matters so much.45

Such policy changes in the late 1950sand early 1960s convinced many ob-servers that the two countries were re-ducing distortions, moving toward get-ting prices right, and—most evident ofall—were exporting. In order for ex-porting to take off the way it did, therehad to be a strong response from do-mestic output. That response was forth-coming to a remarkable degree. A majorquestion is simply: how were these twocountries able to emerge from a decadeof import substitution policies to sud-denly become able to expand exports atsuch a high rate? Several factors were atwork.

The rapid rate of growth of worldtrade was generally acknowledged to be

42 The notion of trade as an engine of growthwas of course an old topic, but not directly relatedto the strategy controversy.

43 There are many books and articles on thisearly stage of Taiwan’s development. Robert Wade(1990); Fei, Ranis, and Shirley Kuo (1979); Ranisin Walter Galenson (1985); and Kuo and Fei inGalenson (1985) are convenient sources that re-view the earlier decades.

44 Chong-Hyun Nam in Takatoshi Ito andKrueger (1995), Westphal in Balassa and Associ-ates (1982), and Westphal (1978) are good generaldiscussions.

45 It may be that export incentives simply offsetcontrary incentives already in existence in Korea,and did not really result in favoring exports overproducing for the domestic market. It does seemhowever, as Nam (in Ito and Krueger 1995) ar-gues, that the array of inducements did more thanoffset other biases in the economy and constituteda net positive inducement to export.


a necessary, but insufficient, conditionfor Korea and Taiwan’s success in ex-porting. During the first half of thetwentieth century, Japan ruled Koreaand Taiwan very firmly. However un-pleasant the experience was for the peo-ple of these countries, it left both coun-tries with a major accumulation ofhuman and physical capital. A long listof observers (Ho, Barclay, Ishikawa, Ra-nis, Pack and Westphal, Wade, Kohli)has emphasized that the Japanese cre-ated infrastructure of all kinds, ex-panded irrigation facilities, spread theuse of fertilizers, and organized farmersinto co-ops to facilitate the spread ofknowledge and partly to keep the farm-ers under control (Wade 1990, p. 77).Manufacturing was strongly supported,and Koreans and Taiwanese weretrained as industrial workers at mostlayers of management. A small numberof Taiwanese and Koreans became man-agers, technicians, professionals, andentrepreneurs. As Japan prepared forwar, it used these areas as sources ofsupply, not as markets as was conven-tional practice with most colonial pow-ers.

A great deal of collective learningwas passed on to successive generationsover this period, which made the peopleof these two areas exceptionally wellequipped to respond to opportunitiescreated by the changed policies and thebooming world market.46 Thus thequality of the labor force was very dif-ferent from that in other developingcountries at the time, and, in particular,unskilled labor was much more effec-tive than that in other parts of Asia, inAfrica, and in most of Latin America.47

This well-developed human capital was,it now seems, an important factor in thecapacity of Korea and Taiwan to create,borrow and adapt from the North in-creasingly productive technologies.

Human capital was not Japan’s onlylegacy. There were institutional devel-opments (e.g., the Bank of Korea, someR&D establishments in agriculture, or-ganizational techniques, etc.) that sig-nificantly enhanced the capacity of Ko-rea and Taiwan to respond quickly toopportunities. Park Chung Hee, thepresident of Korea at the outset of itsgreat growth sprint, attended Japanesemilitary academies in Manchuria and inTokyo prior to World War II. Especiallyin Manchuria, he learned about theJapanese state’s involvement in eco-nomic development.48 The statementsometimes made that Korea and Taiwanbecame newly industrialized countriesin a matter of three or four decades isquite misleading.

The Japanese had broken up or pre-vented the emergence of powerfulvested interests among landowners andbusiness leaders. Land reform in par-ticular in the late 1940s and early 1950sin both Korea and Taiwan made it eas-ier for governments to act indepen-dently, and so a development statecould proceed very much as such.49

46 Collective learning is a term used by Hayekand developed with considerable illumination inthe work of Douglass North. See especially North(1994).

47 Krueger (1995, p. 23) has the following foot-note: “It is widely recognized that an outer- oriented trade strategy cannot succeed unless

development of infrastructure (ports, roads, rail-roads, electric power, communications), increasingeducational attainments, and a number of otherpolicies are conducive to growth.” This statementwould mean that as of 1960 very few developingcountries could have succeeded with an outer- oriented set of policies. It may be noted that Kore-ans and Chinese have long put great emphasis onliteracy. The achievement of widespread literacywas not the main creation of Japanese occupationsnor of the determination to grow after 1950.

48 Mark L. Clifford (1994) and Atul Kohli (1994)discuss the impact of Park’s Japanese military expe-rience on his ideas of development.

49 Mancur Olson (1982, ch. 4) makes a similarargument in explaining the rapidity with which Ja-pan and West Germany recovered from their de-feats in World War II.


From 1950 in Taiwan and 1960 in Ko-rea, a political leadership fully commit-ted to strong economic performancewas firmly in place. This was probablynot the case in any other developingcountry. Foreign aid was substantial inboth places. Koreans in particularlearned from U.S. army engineers inthe 1950s, and this learning was highlyrelevant in the success of the Koreansin bidding for big construction jobs inthe oil-rich Middle East later on. TheTaiwanese were able to exploit the sig-nificant “Overseas Chinese Network” inmany respects.50

6.2 Minimal Government

That distortions result in a level ofoutput below that technologically possi-ble was textbook stuff in the 1960s. Inthe textbooks, distortions were cor-rected by government regulations andother policies. It was generally believedthat the costs of such distortions—Har-berger triangles—were modest, andthere appeared no compelling evidencethat they penalized growth. Still, thedistortions had become so severe thatmany observers were convinced thatgrowth was in fact being penalized.

The governments of developing coun-tries were inexperienced and staffedlargely by ill-trained, ill-prepared peo-ple. By 1970 or so it was also clear thatgovernment policies were themselvesthe sources of many of the distortionsand other market failures. The idea ofthe government correcting market fail-ures then seemed illogical. There wasincreasing evidence that import li-

censes, investment permits, governmentcontracts, and similar devices createdlucrative rents for those fortunateenough to obtain them. It was poten-tially profitable for a firm to allocate re-sources to try to obtain the rights thatwould in turn permit the capture ofthese rents. The cost of the resourcesallocated to rent seeking must then beadded to the traditionally recognizedcosts of monopoly and other distortions.Krueger (1974) made a convincing casethat the costs of rent seeking were sig-nificant enough to matter. Costs of rentseeking plus the costs of distortionsthemselves, plus the limited ability ofmost bureaucracies to design, adminis-ter, and implement sensible plans andcontrols, offered strong evidence thatthe government was part of the problemrather than part of the solution. What-ever market failures were, they wereminor compared to government fail-ures. Thus the idea of minimal govern-ment began to attract widespread favor.Such a view was quite different fromthat which prevailed in the 1950s and1960s.

The rent seeking notion elicited aconsiderable literature, and (amongother things not linked directly to de-velopment) led to a broader notion thatcame to be identified as the New Politi-cal Economy. This notion disputed thetraditional idea that the governmentwas an entity independent of the rest ofthe economy and could thereby betrusted to act in the national interest bycorrecting market failures. Rather, thegovernment must be seen as having itsown agenda, seeking to maximize itsown welfare, and unable and unwillingto take a disinterested and informedstand on economic matters. Rent seek-ing and the New Political Economy ledto an obvious policy: minimize the role(and size) of government and make ex-tensive use of the market. This policy

50 The Overseas Chinese now constitute a majorplayer in East Asian (and world) economic mat-ters, and did so to a lesser extent in the early1960s. This role is hardly clear, and is worth moreattention than it has received. See papers by Gor-don Redding and by Gary Hamilton in Tu Wei-Ming (1996) for helpful summaries of this role. Amore complete analysis of this role is Redding(1993).


was supported by the belief in manyplaces that Korea and Taiwan were es-sentially free-market economies. Theobvious question about minimizing gov-ernment was, simply, who would bringit about. If the government profitedfrom being the government, then it wasunlikely that it would minimize itself.

6.3 Lessons from Taiwan and Korea

The remarkable success of Taiwanand Korea over several decades raisedthe question of the extent to whichtheir experiences offered a blueprintfor development in other countries. At-tention was focused on the role of ex-ports, the package of policies actuallyemployed, and the role of government.

The numerous advantages of a strongexport performance are generally ap-preciated: exports facilitate employ-ment of a country’s most plentiful re-sources and the exploitation of anyeconomies of scale, they help preventbalance of payment problems and thestop/go situation that such problemsoften create, subject the exporting firmto international competition, and givean unambiguous focus to policy making.For the development objective, themain role of exports is its possible con-tribution to the acquisition of new tech-nical knowledge and the consequent in-crease in productivity through contactwith foreign importers combined withthe pressures of strong competition.These latter arguments apply largely tonontraditional exports. Korea and Tai-wan had shown that such exports werepossible and apparently had such conse-quences in their economies—and couldhave similar effects in other countries.51

The transferability of aspects of Ko-

rea and Taiwan’s policy packages, otherthan encouraging exports, was lessclear. Dani Rodrik (in Jere Behrmanand T. N. Srinivasan, Vol 3B, 1995),Alice Amsden (1987 and 1989), LarryWestphal (1990), Howard Pack andWestphal (1986), and John Helliwell(1992) review the disputes with manyadditional citations. Two points standout. The policy instruments used in Ko-rea and Taiwan have been generally thesame as those used in other developingcountries—import quotas and licenses,export subsidies, public ownership, taxholidays.52 The second point follows:the main difference seems to be themanner of implementation and monitor-ing, rather than the policies themselves.Effectiveness of application seemed tobe such that they worked well in somecountries, and not in others. The WorldBank (1993) recognizes this point, andconcludes from it that other countriestherefore should rely primarily on themarket.53 The basic conclusion of thisvolume (and numerous others) was thatgovernment activity that interfered withthe market should be abandoned.

The third issue refers to the role ofgovernment more generally. The signifi-cant role of government in developmentin Taiwan and Korea since the 1950swas emphasized in a number of earlystudies. (See Charles R. Frank et al.1975; Westphal 1978; and Leroy Jonesand Il Sakong 1980). Even so, the pro-fession at large seemed reluctant toacknowledge such a role. Why did ittake western economists so long to ap-

51 A question arose whether or not the worldcould (or would) absorb the exports of all develop-ing countries were they to achieve rates of growthequal to those of Korea and Taiwan. See WilliamCline (1982) and the comment by Ranis (1985).

52 Ann Harrison (1991) surveys a great numberof regressions involving growth and openness(measured in a variety of ways) and finds thatcountry dummies are generally significant overtime even after accounting for policies, labor, edu-cation, etc.

53 The Bank’s position seems to be that if thegovernment withdraws from economic activity, awell-functioning market will emerge more or lessautomatically.


preciate the governments’ roles? Tworeasons stand out. First, the rise ofrent-seeking and the New PoliticalEconomy made it difficult for westerneconomists to see the complex way inwhich these governments did intervenein a way that facilitated the achieve-ment of development objectives ratherthan penalizing them.54 Little’s state-ment (in Galenson, ed. 1979, p. 480)seemed to reflect a common view:“apart from the creation of [these neo-classical conditions] it is hard to findany good explanation for the sustainedindustrial ‘boom’ in Taiwan.”

The second reason refers to the na-ture of the role of government in thetwo countries. The idea that the Tai-wanese and Korean governments simplycorrected market failures of the text-book sort is quite inadequate. The gov-ernment and the private sector aremore intertwined, more intermeshedthan fits well with western textbookanalysis. In particular, it was difficult toappreciate how government could en-force the discipline and induce the ef-fort in private sector firms that is usu-ally assumed to be the consequence ofcompetitive markets. Imports did notoffer competition, but discipline was af-forded by the state. Such conventionalterms as mixed economy or neoclassicalsynthesis do not capture the basic fea-tures of the role that the governmenthas played in these countries. The formof the intermeshing has its roots deepin institutional and historical argumentthat also fits poorly into conventionalformulations. Michio Morishima (1982),Shigeto Tsuru (1993), Tessa Morris-Suzuki (1989), and Christopher Howe

(1996) are especially helpful in definingthe role of the state in Japan.

There is another more positive aspectof the government role in development.Ranis (in African Development 1991,pp. 128–29) emphasizes that “what hap-pened in Taiwan was not Mandarins sit-ting around saying this is what we haveto do now. There was a lot of bumblingand stumbling and going back andforth.” Morishima (1982, p. 71) statesthat in Japan “there was never an accu-rate blueprint for the Meiji Revolution;the revolutionaries learnt what theissues and solutions were by repeatingthe process of trial and error andapproached the correct ones step bystep.” Pack and Westphal (1986, p. 99)emphasize the flexibility and respon-siveness with which Korean decisionmakers approached their tasks; decisionmaking “has generally been able toelicit, digest, and act on information un-covered in the process of implementingprevious decisions.” When a policyworks, pursue it; when it fails, changeit. That a government makes mistakes isinevitable. That it does not learn fromthose mistakes means that it needs tofind ways to learn. Government learn-ing, not government minimizing, is theobject.55

A detailed comparison of Korea andTaiwan with other developing countriesis not appropriate here, but a few obser-vations may identify some key differ-ences. At the outset of India’s indepen-dence, its bureaucracy was firmly inplace and strongly based on its Britishheritage. The government was run by awell-established party “imbued with

54 Ya-Hwei Yang (in Ito and Krueger 1993) pro-vides some empirical evidence that the prefer-ential policy for strategic industries has not signifi-cantly improved the investment, financialsituation, and operational performance of firms inTaiwan in the early 1980s. Yang is careful to em-phasize the limitations of the study.

55 Gustav Ranis (in William Beranik Jr. and Ra-nis 1978, pp. 15 ff) cites evidence from severalsources that the Meiji government made substan-tial mistakes in its early attempts to imitate West-ern-style agriculture and manufacturing technolo-gies. Turnkey projects in the public sector alsoproved largely failures. See Kuame Jomo (1994)for further elaboration.


Gandhian austerity and devotion to thecause of a united and prosperous India”(Vijay Joshi and Ian Little 1994, p. 8).Nehru was a Fabian socialist with deeplinks to British ideas, institutions, andethos, all quite alien to the real India,the India of villages, of illiteracy, ofcastes, of hopelessness. He thus had astrong suspicion of markets and foreigntrade, and great confidence in the abil-ity of an elite to plan and govern. Stum-bling and bumbling in search of theright strategy was not an approach thatmade sense to Nehru and his elite bu-reaucracy because they knew the rightanswers from the outset of indepen-dence.

In Africa, states were just beginningto be states, and there was little or noheritage of producing, marketing, ortechnological learning, and essentiallynone of economic management and pol-icy making. In Latin America there wasa heavy dose of foreign capital, and adomestic economic organization domi-nated by a powerful landowning classthat shaped national policy to a signifi-cant degree. It is easy to appreciate thegreat appeal of structuralism there.

There were thus huge differences be-tween Korea, Taiwan, and Japan, andthe rest of the developing world in theyears 1950–70 with respect to a greatvariety of characteristics with implica-tions for the kind of policies that couldbe implemented, or even perceived,and that would be effective. That a sin-gle policy package would fit all thecountries is unlikely—except that oflearning from one’s mistakes.

7. The New Orthodoxy: OutwardOrientation and Minimal Government

The lessons learned from the importsubstitution experience and from thestories of Taiwan and Korea pointed toa new strategy. The central notions

were a return of confidence in themarket combined with a strong com-mitment to exporting nontraditionalproducts and to liberalizing imports.Emphasis was placed on elimination ofprice distortions, recognition of thepower of comparative advantage, priva-tization of virtually all public firms,acceptance—even encouragement—ofprivate foreign investment, mainte-nance of price level and balance of pay-ments stability, and becoming interna-tionally competitive. The view that aneffective market mechanism would ap-pear if the government simply removeditself from the economy was implicit inmany formulations even though evi-dence to support the view was rarely of-fered.

Some of the learning, however, wasnot included in the new orthodoxy. Rec-ognition of the deep-seated difficultiesof the international transfer of technicaland other knowledge and of the funda-mental role of searching and learningby firms and governments, of a neces-sary role for agriculture, of the role ofinitial conditions and hence of historyand institutions, and of the fact that ef-fective implementation of policies is asimportant as the choice of policies—these are all missing. Ideas of economicindependence (including ambivalencetoward foreign direct investment) to ac-company political independence, so im-portant in the import substitution ap-proach, are given no attention by thenew orthodoxy.

Along with the World Bank, the In-ternational Monetary Fund and theUnited States Agency for InternationalDevelopment became strong advocatesof outward orientation. The influentialweekly, The Economist, pushed (andpushes) hard for the approach, as do asignificant number of university andthink tank figures. It is much less clearhow committed the government policy


makers in developing countries were(and are) to the new ideas. Leaders arehesitant to reduce their chances of re-ceiving aid from international agencies,but there is considerable evidence thatmany leaders find outwardness argu-ments unconvincing, and drag their feetwhen urged to implement opening andliberalizing policies.56 This state of af-fairs has led to outward orientation be-ing identified as the “Washington Con-sensus,” suggesting a position held inWashington D.C., but much less soelsewhere—except here and there.

7.1. The Development of the WashingtonConsensus

By the 1970s the distortions and anti-export biases were beginning to have aneffect, yet growth rates on the average

held up reasonably well in this decade.Merchandise exports for low-incomecountries, an exception, did declinesharply in this decade relative to the1960s. Table 2 shows the well-knownfact that many countries suffered sharpdeclines in rates of growth and even intheir levels of income during the 1980s.A number of factors contributed tothese unfortunate circumstances. Thesecond oil price increase, the ensuingdebt crisis in many countries, and therecession and high interest rates in theUnited States all had adverse effects onmost developing countries, especiallythe lowest-income ones. The lattercountries—minus China and India—av-eraged 2.9 percent annual growth in the1980s compared to 3.6 in the previousdecade, and growth rates in the middle-income countries fell by over 50 per-cent. Agriculture and industry growthrates also fell sharply, while exports forthe low-income group picked up a bit.

TABLE 2GROWTH RATES AFTER 1970

GDP Agriculture

1970–81 1980–90 1990–94 1970–81 1980–90 1990–94

Low-Income Countries 4.5 5.8 6.2 2.3 3.5 2.8 Excluding India and China 3.6 2.9 1.4 2.3 2.0 1.5

Middle-Income Countries 5.6 2.2 .2 3.0 .9 Lower Middle 5.6 2.2 –2.3 3.2 Upper Middle 5.6 2.2 3.4 2.6 2.5 .9

Industry Merchandise Exports

1970–81 1980–90 1990–94 1970–81 1980–90 1990–94

Low-Income Countries 3.6 7.4 11.4 –.7 5.7 9.1 Excluding India and China 3.2 2.7 –.7 –.8 1.0 2.6

Middle-Income Countries 6.8 1.3 4.1 3.5 7.0 Lower Middle 7.4 3.0 Upper Middle 4.5 2.1 2.6 7.0 3.5 7.8

Sources: 1970/81 World Development Report, 1983. 1980/90 and 1990/94 World Development Report, 1996.

56 Some older policy makers recall being di-rected to develop a formal plan in order to qualifyfor assistance and subsequently being directed torely on the market to qualify.


This general picture continued into the1990s.

There were, however, important andilluminating exceptions. China and In-dia, long committed to a severe form ofimport substitution, achieved sharp in-creases in their growth rates of GDP,agriculture, industry, and exports. Bothcountries put in place a number of poli-cies that significantly opened theireconomies in the late 1970s and early1980s. Chile and Turkey also openedsignificantly in the 1970s and 1980s,and manufacturing exports and GDP (toa lesser extent) in both countries re-sponded. Other countries that did littletoward opening up did less well. TheIvory Coast, Kenya, Nigeria and otherAfrican countries, and Latin America(save Chile) showed sharply reducedrates of growth in almost all aspects oftheir economies. The Korean govern-ment’s experiment with heavy chemi-cals and industry—and its subsequentbacking away—also suggested to manythe advantages of a market driven, openeconomy.57 There was the emergenceof Malaysia and Thailand as newly in-dustrialized countries, and their greatsuccess seemed to be a consequence oftheir openness, especially their unquali-fied acceptance of foreign direct invest-ment. The success in these two coun-tries, and in Korea and Taiwan, inmeeting employment and poverty alle-viation objectives added further supportfor the effectiveness of the new ortho-doxy. Arguments were offered that theopen economies rode the oil price in-creases and other international disloca-tions more effectively than did thecountries dominated by the import sub-stitution approach.

During the 1980s, the outward orien-tation arguments became more unquali-fied and less hesitant compared to ear-lier, more cautious statements. TheWorld Development Report of 1987 wasespecially fervent and aggressive in de-fending outwardness and in condemn-ing import substitution, while the 1991report was only a bit less so. The 1987report, for example, classified 41 coun-tries into strongly and moderately out-ward oriented and strongly and moder-ately inward oriented. There were onlythree countries—Korea, Singapore, andHong Kong—classified as strongly out-ward oriented. (Taiwan was not in-cluded in the 41–country sample.) Inthese three countries, real manufactur-ing value added averaged 15.6 percentgrowth from 1963 to 1973. Moderatelyinward and moderately outward ori-ented each averaged 9.5 percent andstrongly inward, 5.3. The rates for theyears 1973 to 1985 were all lower, butshowed similar differences among thecategories of countries. The implicationthat (presumably) the Bank wished toconvey was that if a country shifted intothe strongly outward-oriented category,it too would achieve manufactures andother growth rates similar to those ofKorea, Singapore, and Hong Kong.58

A large number of books and articleswith carefully constructed and empiri-cally supported arguments appeared toshow that outwardness produced unam-biguously superior economic perfor-mance to its counterpart.59 The World

57 Korea’s experience with heavy chemicals andindustry is open to much dispute. See JosephStern et al. (1995) for a general review, and thecontributions of Rodrik and Wade in Fishlow et al.(1994) for more specific questions and discussion.

58 This exercise by the World Bank has beencriticized often, but the evidence and accompany-ing arguments have been widely cited.

59 In Ito and Krueger (1995, p. 24), Kruegerwrites that “having agreed that outward orienta-tion is a necessary condition for economic growth,analysts have then considered the extent to whichit was sufficient.” Krueger adds immediately thatit was not sufficient and that a “high rate of in-vestment, provision of infrastructure, a well- functioning labor market, and the overall policyframework conducive to efficient production were


Bank’s East Asian Miracle (World Bank1993) may be looked upon as a culmi-nating statement of the great power ofan outward-looking approach to gener-ate and maintain growth. Many ob-servers, and the World Bank as an insti-tution, became convinced that exportingand minimal government explained thesuccesses of Taiwan and Korea and thenthose of Malaysia and Thailand, andthat such a strategy would perform thesame “miracle” for all other developingcountries.60 These later writings placedheavier emphasis than did earlier litera-ture on exporting as the great source ofdynamism, and “export led growth” be-came a key phrase. The problem withimport substitution seemed now to beless with distortions as such than that itcreated biases against exporting. Thisshift of emphasis away from the costs ofdistortions to the gains from exports be-gan in the mid–1970s. See for exampleBalassa’s contribution in Balassa andAssociates (1982) and Westphal (1978,1990). The shift seemed to be a re-sponse to overwhelming evidence thatKorea and Taiwan pursued policies thathad significant distorting effects (in theconventional sense), but had achievedremarkable rates of growth of exports,as well as employment, distribution,poverty alleviation, and other objectivesthat Washington economists empha-sized.

7.2. Problems with the New Orthodoxy

A literature of doubt and questioningof the New Orthodoxy gained strengthfrom the mid–1980s. This literature israrely a defense of import substitutionas originally practiced. Rather, it is acritical evaluation of the arguments ofthe outward-oriented approach and ofthe analyses of what had actually takenplace in Korea, Taiwan, and other suc-cessful countries. Almost all compo-nents of the outward-looking packageare under attack.

The Role of Exporting. There is gen-eral agreement that nontraditional ex-ports have numerous advantages, andfew observers recommend policies thatdampen inducements to export.61 Themore complex issue is to understand theway in which exporting and domesticlearning interact. Studies of knowledgeaccumulation—especially the ideas oftacit knowledge, on-the-job learning,learning by doing and by using—com-bined with studies of technologicalchange in individual firms and indus-tries offer strong evidence that simplyexporting is not sufficient to result in orto substitute for the creation of a strongindigenous learning process. 62

clearly major contributing factors.” That anyonecould think that outward orientation was a suffi-cient condition is puzzling, and once the otheritems are added to the list, the role of outwardnessbecomes very blurred.

60 The East Asian Miracle also emphasized therole of human capital in the East Asian successstories. As already noted, this volume agreed thatthe governments of the various East Asian coun-tries played important roles, but recommendedagainst significant government intervention inother developing countries. Extended reviews ofThe East Asian Miracle by several authors are Al-bert Fishlow et al. (1994), Amsden, ed. (1994),and Ranis (1995).

61 Rodrik (in Fishlow et al. 1994) speaks of “ex-port fetishism,” and argues strongly that the WorldBank’s The East Asian Miracle fails to make a con-vincing case that exporting has played as strategica role in East Asian development as the studyclaims.

62 There is a great literature on these matters.Howard Pack (in Chenery and Srinivasan, Vol. 1,1986) and Evenson and Westphal (in Behrmanand Srinivasan, Vol. 3A, 1995) are excellent sur-veys with many references.

There are several empirical studies that showthat total factor productivity (TFP) growth inKorea, Taiwan, and Singapore has not been higheror even as high as in other developing countries.See Alwyn Young (1995) and Yuan Tsao (1985) asexamples. Estimates of TFP growth are open tomany doubts, and other investigators, Pack inHelleiner, ed. (1992) and World Bank (1993) showmuch higher rates for Taiwan and Korea. Morefundamentally, the rapid rates of growth of capitalrequired the accumulation and use of much new


An idea, implicit in many outward-oriented arguments, is that firms in adeveloping country can “leapfrog” froma technology vintage that is well withinthe technology frontier to one that is onthe frontier, and thereby become imme-diately internationally competitive, sothat conventional Heckscher–Ohlin ar-guments apply. The evidence againstsuch an argument is strong. After de-tailed studies of numerous manufactur-ing activities in South Korea, Taiwan,Hong Kong, and Singapore, MichaelHobday (1995, p. 200) concludes that“TNCs [transnational corporations] andlocal East Asian firms engaged in apainstaking and cumulative process oftechnological learning: a hard slog,rather than a leapfrog.”63 The achieve-ment of the hard slog in turn requiresconsideration of more fundamental as-pects of a society—entrepreneurship,institutions, values, social incentives,commitment to growth, and a variety ofother factors that define a society.These characteristics vary hugely amongthe countries of the developing world.

Similarly, the basic question aboutthe contribution of TNCs to the sus-tained, independent growth of theirhost countries is the extent to whichtheir presence contributes to the in-

digenous learning efforts of nationalfirms. Several observers (e.g., Morris–Suzuki in Shojiro Tokunaga ed. 1992)emphasize that the dependence on for-eign investment in Malaysia and Thai-land over the last decade or so has re-sulted in favorable growth rates, butquestion whether this foreign invest-ment is leading to strong indigenouslearning capacity by national firms. For-eign investment with few technologicaland other spillovers onto national pro-ducers may help short-run problems,especially unemployment, while creat-ing longer run, more intractable ones.(See Kunio Yoshihara 1988, and Lall1984 and 1996, for examples and fur-ther discussion.) The basic objective isnot to attract foreign investment assuch, but to create an internal socialand economic environment within whichthe national knowledge-accumulatingprocess profits from the presence offoreign firms.64

Importing. These issues have directrelevance for import policies. TheWashington Consensus seems to oppose(arguments are not always completelyclear and there are differences amongadvocates) any import controls or otherforms of protection. Reform packagesalmost always include “import liberali-zation” as a significant component. Pro-tection under import substitution cre-ated a great number of problems, butonce the leapfrog process is recognizedas not possible, then some form of pro-tection for learning is necessary. Koreaand Taiwan (and Japan) have always hadnumerous forms of protection that(along with government influence) in-duced learning. The major policy issue

knowledge. Firms do not move along a well- established isoquant as the capital–labor ratiorises. They must find their way by developing andapplying new knowledge that enables the full ex-ploitation of the new resources. Whether TFPgrowth took place is one thing, whether learningand knowledge accumulation did is another. Thelatter must have occurred in the East Asian coun-tries. See Richard Nelson and Pack (1996) forfurther development of this point.

63 Hobday’s study illuminates well the link be-tween exporting and domestic learning capacity.There are numerous other studies: Rodrik (inFishlow et al. 1994, and Helleiner 1992), SanjayaLall (1984), Amsden (1989), Wade (1990), andothers. There are many different nuances and in-sights among these authors, but the underlyingtheme in all of them is the importance of domesticlearning and knowledge accumulation.

64 The differences between development in Ko-rea and Taiwan and that in Malaysia and Thailandare especially revealing of the issues reviewedhere. In addition to the citations in the text, seeDanny M. Leipziger, ed. (1997) and Bruton et al.(1992).


then is to design protection measuresthat induce learning rather than theeasy life.65

Right Prices. To define right prices ina context where learning and produc-tivity growth are the basic objectives(rather than the allocation of given re-sources with given technology) requiresnew considerations (Helleiner 1995;Stiglitz 1987; Chenery and Srinivasan1988 Vol. 1; and Amsden 1989). Espe-cially important are the exchange rateand the wage rate. The overvalued ex-change rate is condemned, but the mostfrequently identified alternative is a“realistic” exchange rate. In the pres-ence of large-scale capital movements,worker remittances, Dutch disease, andproductivity growth, an exchange ratethat simply protects the balance of pay-ments is almost certainly wrong fromthe standpoint of more fundamental ob-jectives.66 Similar difficulties arise withrespect to wage rates where factorssuch as labor migration, internationalcapital movements, and institutionalpressures complicate the task of findingthe right wage rate.67

Employment and Poverty Alleviation.Studies of the effect of openness on em-ployment and poverty alleviation, onceit is fully established, are less available.The evidence that Korea and Taiwanhave been highly successful in these re-

gards, largely because productive em-ployment has grown so strongly, iswidely accepted. This success, for rea-sons examined above, is often explainedby their initial conditions—long com-mitment to education, social cohesion,capacity to learn in general and espe-cially to adapt imported technologies tolocal conditions, and a strong govern-ment pushing economic growth. Othercountries with different conditions areexpected to face different hurdles insolving these problems in an open econ-omy. The transition to an open economyhas also proved costly in terms of em-ployment and poverty relief in severalcountries.68

The Role of Government. “Minimalgovernment” and “market friendly” arenot illuminating notions. Similarly,“rent seeking” is much more complexthan the Washington Consensus im-plies. (See, for example, John Toye1987, and Paul Mosely et al. 1991). Therole that government can play de-pends—more than is the case for mostissues—on the institutions, the history,and the culture of the community. Thatsome governments are deadly, thatmany are inept and uncaring is widelyrecognized, but there is much that onlya government can do. So learning is ascrucial a notion for governments as it isfor firms and households. The questionis how can a government learn, not howcan it be minimized.69 One specificrole for government is the design andimplementation of an effective agricul-ture policy. A strongly performing agri-culture sector is essential, and there arefew examples of success stories in agri-

65 Staffan Linder (1961) argues that a countrycannot export new products successfully until ithas produced them for the home market. This ar-gument, frequently found in the literature, is stillgenerally applicable, although there are excep-tions, especially where TNCs are involved. Firmsmust produce before they can export, and theyhave to learn to produce.

66 The exchange rate may be used as an instru-ment of protection and export promotion. MaxCorden (1985) argues that Japan did this after1953. See Bruton (1997, ch. 8) for more discussionof the advantages of an “undervalued” exchangerate.

67 Freeman (1993) is a valuable discussion ofwage issues and their relevance to developmentstrategy.

68 See, for example, Hans Singer and SumitRoy (1993), Kunibert Raffer and Singer (1996),Giovanni Cornia et al. (1987), Morrison (1994), andearlier discussions in Paul Streeten (1973).

69 The World Development Report for 1997treats the role of government in detail, and in amuch more cautious and probing way than its pre-vious reports have done.


culture that do not involve a strong rolefor the government.

These theoretical complexities haveadded to the problems for empirical re-search. A review of the empirical workrelevant to the role of trade orientationand growth is not part of the presenteffort. There are many such surveys. Inaddition to Sebastian Edward’s JEL sur-vey (Edwards 1993), papers by Levineand Renelt (1992), Levine and Zervos(1993), Harrison (1991), Rodrik (inBehrman and Srinivasan Vol. 3B, 1995),Pack (in Chenery and Srinivasan Vol. 1,1988), Tybout (1992), Havrylyshyn(1990), and Helliwell (in Pasinetti andSolow, eds. 1994) are especially helpfulin surveying the existing empiricalwork.70 The final paragraph in Edwards(1993, p. 1390) is, in my view, a legiti-mate evaluation of the state of play withrespect to empirical work on exportsand distortions and economic perfor-mance. Edwards argues that economistsoften ask too much of the availabledata, and try to extract information thatsimply is not there. He argues thatcross-country aggregate data sets havelittle information about trade policy andgrowth. Cross-country regressions onaggregate data cannot reveal muchabout historical factors at work norabout the microeconomics of innova-tion, trade, and growth. Studies of his-torical episodes look more promising,and there, as noted, the role of tradeseems less strategic and less powerful.71

Finally, one may note that empiricalwork has not been able to solve the di-

rection of causation question: Doesgrowth cause exports or do exportscause growth?

8. Conclusion

It remains correct to say that outwardorientation is still urged strongly on alldeveloping countries by powerful insti-tutions and well-placed individuals.Why? One may speak of an easy stage ofoutward orientation as well as of importsubstitution. In an economy in whichthere are great price distortions, under-utilized capacity, rampant inflation, andlarge balance of payment problems, cor-recting these problems may well pro-duce quick increases in output andmeasured factor productivity. Such aneasy stage can continue for some time,especially if world trade is booming.But what happens after this easy stageof outward-orientation reform has ex-hausted its momentum, and somethingmore fundamental has to come intoplay? What is it that is more fundamen-tal?72 If, in all countries, distortions areeliminated—if all prices are right—thegovernment’s role is minimal, no infla-tion exists anywhere, saving and invest-ment are at least 15 percent of GDP,and world trade is booming, would allcountries grow at the same rate as Ko-rea and Taiwan have in recent decades?The answer surely must be no.73

70 Levine and Zervos (1993) is valuable becauseit is a good clear statement of the dangers of usingcross-country regressions as if they met the neces-sary assumptions of formal statistical and econo-metric theory.

71 I think it is fair to say that those who are con-vinced of the power of the outward-oriented strat-egy find the empirical evidence more convincingthan indicated in the text. This is not meant to beunkind, but simply to reflect the inadequacy ofboth theory and measurement on the subject.

72 The following statement from Krueger (1995,p. 13) is of interest: “It was not until the 1980sthat experience was to demonstrate that the prob-lems confronting developing countries were fun-damentally more deep-seated than that of stop-gocycles and that foreign exchange shortage wasmore a symptom than a cause of the problem.”Krueger, one of the strongest proponents of out-ward orientation, does not expand on what shemeans by “fundamentally more deep-seated.” Lan-des (1990) and Olson (1996) probe more deeply insearch of the more fundamental source of successand failure.

73 It is doubtful that many people would arguethat all countries would grow as fast as Korea andTaiwan. At the same time, there is, in much of the


This second question leads to a moregeneral question: Why did per capitaGDP not begin to grow in Latin Amer-ica, Africa, and Asia 150 years ago whenit began to grow in Western Europe,Northern America, and Australia? Orwhy did economy-wide growth not takehold in periods of commodity booms inpresently less-developed countries? Sofar as I know, no one explains these fail-ures in terms of import substitution.74

Import substitution and outward ori-entation offered easy solutions to thedevelopment problems. Import substi-tution as implemented failed, and thejustifications for outward orientation (asusually presented) are being increas-ingly undermined.75 The findings re-viewed in this paper suggest stronglythat no quick and easy fixes to develop-ment problems are available. To acceptlearning and knowledge accumulationboth as the bottom line of growth andas having roots deep in the ethos andhistory of a society requires that expla-nation and policy prescription probethese precincts that are so alien tomainstream thinking. This is the greatmessage of the histories of technicalknowledge accumulation and of the sto-ries of the failures of import substitu-tion and outward orientation reportedherein.

REFERENCES

ABRAMOVITZ, MOSES. “Economics of Growth” inBERNARD F. HALEY, ed., A survey of contem-porary economics. Vol. II. Homewood, IL:Richard D. Irwin, 1952, pp. 132–81.

African development: lessons from Asia. Proceed-ings of a Seminar on Strategies for the Future ofAfrica. Washington, DC: Winrock InternationalInstitute for Agricultural Development, 1991.

AMSDEN, ALICE. Growth and stabilization in Ko-rea. Helsinki: Wider Institute for DevelopmentEconomics Research, 1987.

———. Asia’s next giant: South Korea and lateindustrialization. New York: Oxford U. Press,1989.

———. ed. “The World Bank’s ‘The East AsianMiracle’: Economic Growth and Public Policy,”Special section of World Devel. Apr. 1994,22(4), pp. 614–70.

BALASSA, BELA AND ASSOCIATES. The structure ofprotection in developing countries. Baltimore:Johns Hopkins U. Press for the World Bank andthe Inter-American Development Bank, 1971.

———. Development strategies in semi-industrialeconomies. Baltimore: Johns Hopkins U. Pressfor the World Bank, 1982.

BEHRMAN, JERE AND SRINIVASAN, T. N. , eds.Handbook of development economics. Vols. 3Aand 3B. Amsterdam: Elsevier Science Publish-ers, 1995.

BERANIK, WILLIAM, JR. AND RANIS, GUSTAV, eds.Science, technology, and economic develop-ment. New York: Praeger Publishers, 1978.

BERGER, PETER L. AND HAIN-HUANG, MICHAELHSIAO, eds. In search of an East Asian develop-ment model. New Brunswick and Oxford, Trans-action Books, 1988.

BHAGWATI, JAGDISH N. AND DESAI, PADMA. In-dia, planning for industrialization. Oxford: Ox-ford U. Press for the OECD, 1970.

BIGGS, TYLER; SHAH, MANJU AND SRIVASTAVA,PRADEEP. Technological capabilities and learn-ing in African enterprises. World Bank Techni-cal Paper Number 288. Washington, D.C:World Bank, 1995.

BRUTON, HENRY J. “Productivity Growth in LatinAmerica,” Amer. Econ. Rev., Dec. 1967, 57(5),pp. 1099–116.

———. “The Import Substitution Strategy of De-velopment: A Survey,” The Pakistan Devel.Rev., Summer 1970, 10(2), pp. 123–46.

———. “The Search for a Development Econom-ics,” World Devel., Oct./Nov. 1985, 13 (1O/11),pp. 1099–124.

———. On the search for well-being. Ann Arbor:U. of Michigan Press, 1997.

BRUTON, HENRY J. ET AL. The political economyof poverty, equity, and growth: Sri Lanka andMalaysia. New York: Oxford U. Press for theWorld Bank, 1992.

CHENERY, HOLLIS B. “The Application of Invest-ment Criteria”, Quart. J. Econ., Feb. 1953,67(1), pp. 76–96.

literature, strong criticism of countries that havenot grown as fast as these two countries. Such aposition tends to defeat efforts to understand ex-actly what makes a particular economy grow.

74 There are many arguments of course as towhy the rest of the world did not begin to growwhen the West did, but an import substitution pol-icy does not seem to be one of them. As notedearlier growth picked up in almost all countriesduring the heyday of import substitution.

75 Robert Solow (in Fuhrer and Little 1996,p. 298) quotes Charles Schultz as saying that therewas nothing wrong with supply-side economicsthat dividing by 10 could not cure. A similar obser-vation holds with respect to any simple once-and-for-all solution of the development problem— except here one should probably divide by 25.


CHENERY, HOLLIS B. AND STROUT, ALAN M.“Foreign Assistance and Economic Develop-ment,” Amer. Econ. Rev., Sept. 1966, 56(4). pp.674–733.

CHENERY, HOLLIS; ROBINSON, SHERMAN ANDSYRQUIN, MOSHE, eds. Industrialization andgrowth. New York: Oxford U. Press for theWorld Bank, 1986.

CHENERY, HOLLIS B. AND SRINIVASAN, T. N.,eds. Handbook of development economics. Vols.I and II. Amsterdam: North Holland, 1988,1989.

CLIFFORD, MARK L. Troubled tiger. Armonk,New York: M. E. Sharpe, 1994.

CLINE, WILLIAM. “Can the East Asian Model BeGeneralized?” World Devel., Feb. 1982, 10(2),pp. 81–90.

CORDEN, W. MAX. The theory of protection. Ox-ford: Clarendon Press, 1971.

———. Protection, growth and trade. Oxford:Basil Blackwell, 1985.

CORNIA, GIOVANNI; JOLLY, RICHARD AND STEW-ART, FRANCES, eds. Adjustment with a humanface. Vol. I. Oxford: Clarendon Press, 1987.

DIAZ-ALEJANDRO, CARLOS F. “On the Intensityof Import Substitution,” Kyklos, 1965, 18(3),pp. 495–511.

DONGES, JUERGEN B. “A Comparative Survey ofIndustrialization Policies in Fifteen Semi-Industrialized Countries,” Weltwirtsch. Arch.,1976, 112(4) pp. 626–59.

ECKAUS, RICHARD S. “The Factor ProportionsProblem in Underdeveloped Areas,” Amer.Econ. Rev., Sept. 1955, 45(4) pp. 539–65.

EDWARDS, SEBASTIAN. “Openness, Trade Liber-alization, and Growth in Developing Coun-tries,” J. Econ. Lit., Sept. 1993, 31(4) pp. 1358–93.

FEI, JOHN C. H. AND RANIS, GUSTAV. Develop-ment of the labor surplus economy. Homewood,IL: Richard Irwin, 1964.

FEI, JOHN; RANIS, GUSTAV AND KUO, SHIRLEY.Growth with equity. New York: Oxford U. Pressfor the World Bank, 1979.

FISHLOW, ALBERT ET AL. Miracle or design? Les-sons from the East Asian experience, Washing-ton, DC: Overseas Development Council, 1994.

FRANK, CHARLES R., JR.; KIM, KUANG ANDWESTPHAL, LARRY. Foreign trade regimes andeconomic development: South Korea. New York:Columbia U. Press for the NBER, 1975.

FREEMAN, RICHARD B. “Labor Market Institu-tions and Policies: Help or Hindrance to Eco-nomic Development,” Proceedings of the WorldBank Annual Conference on Development Eco-nomics, 1992, pp. 117–45.

FUHRER, JEFFREY C. AND LITTLE, JANE S., eds.Technology and growth. Boston: Federal Re-serve Bank of Boston, 1996.

GALENSON, WALTER, ed. Economic growth andstructural change in Taiwan. Ithaca: Cornell U.Press, 1979.

———. ed. Foreign trade and investment: Eco-nomic development in industrializing Asian

countries. Madison: U. of Wisconsin Press,1985.

GALENSON, WALTER AND LEIBENSTEIN, HAR-VEY. “Investment Criteria, Productivity, andEconomic Development,” Quart. J. Econ. Aug.1955, 69(3), pp. 343–67.

GRAN, PETER. “Islamic Roots of Capitalism,” In-ternational Journal of Middle East Studies, July1980, 11(2), pp. 511–26.

HABERLER, GOTTFRIED. International trade andeconomic development. Cairo: National Bank ofEgypt, 1950.

HAMILTON, GARY G. “Overseas Chinese Capital-ism,” in WEI-MING TU, ed., 1996, pp. 328–42.

HARRISON, ANN. “Openness and Growth.” Work-ing Paper 809, Office of the Vice President, De-velopment Economics, World Bank, Washing-ton, DC, 1991.

HARROD, ROY F. “An Essay in Dynamic Theory,”Econ. J., Mar. 1939, 49(1), pp. 14–33.

———. Towards a dynamic economics. London:Macmillan and Company, 1949.

HAVRYLYSHYN, OLI. “Trade Policy and Produc-tivity Gains in Developing Countries,” WorldBank Res. Observer, Jan. 1990, 5(1), pp. 1–24.

HELLEINER, GERALD K., ed. Trade policy, indus-trialization, and development. Oxford: Claren-don Press, 1992.

———. Trade, trade policy and industrializationreconsidered. Helsinki: World Institute for De-velopment Economics Research, World Devel-opment Studies 6, 1995.

HELLIWELL, JOHN F. “Trade and Technical Pro-gress.” NBER Working Paper 4226, 1992.

HIGGINS, BENJAMIN. Economic development.New York: W.W. Norton & Company, 1959.

HIRSCHMAN, ALBERT O. Development projectsobserved. Washington, DC: Brookings Institu-tion, 1967.

———. “The Political Economy of Import Substi-tuting Industrialization in Latin America,”Quart. J. Econ. Feb. 1967, 82(1) pp. 1–32.

HOBDAY, MICHAEL. Innovation in East Asia,Aldershot: Edward Elgar, 1995.

HOSELITZ, BERT F., ed. Theories of economicgrowth. Glencoe, IL: Free Press, 1960.

HOWE, CHRISTOPHER. The origins of Japanesetrade supremacy. Chicago: U. of Chicago Press.1996.

ITO, TAKATOSHI AND KRUEGER, ANNE O. , eds.Trade and protection. Chicago: U. of ChicagoPress, 1993.

———. eds. Growth theories in light of East Asianexperience. Chicago: U. of Chicago Press, 1995.

JAMES, JEFFREY. The state, technology and indus-trialization in Africa. New York: St. Martin’sPress, 1996.

JOMO, KUAME S., ed. Japan and Malaysian devel-opment. London: Routledge, 1994.

JONES, LEROY P. AND SAKONG, IL. Government,business, and entrepreneurship in economic de-velopment. Cambridge: Council on East AsiaStudies, Harvard U., 1980.

JOSHI, VIJAY AND LITTLE, IAN M. D. India: Mac-


roeconomics and political economy 1964–1991.Washington, DC: World Bank, 1994.

KAHN, ALFRED E. “Investment Criteria in Eco-nomic Development,” Quart. J. Econ. Feb.1951, 65(1). pp 38–61.

KOHLI, ATUL. “Where Do High Growth PoliticalEconomies Come From? The Japanese Lineageof Korea’s Development State, World Devel.,Sept. 1994, 22(9), pp. 1269–93.

KRUEGER, ANNE O. “The Political Economy ofthe Rent-Seeking Society,” Amer. Econ. Rev.June 1974, 64(3), pp. 291–303.

———. Trade policies and developing nations.Washington, DC: Brookings Institution, 1995.

LALL, SANJAYA. Learning from the Asian tigers.New York: St. Martin’s Press, 1996.

LALL, SANJAYA ET AL. “Exports of Technology byNewly-Industrializing Countries,” Special Issueof World Devel., Dec. 1984, 12(5/6), pp. 471–660.

LANDES, DAVID S. “Why Are We So Rich andThey So Poor,” Amer. Econ. Rev., May 1990,80(2), pp. 1–13.

LEIBENSTEIN, HARVEY. Economic backwardnessand economic growth. New York: John Wileyand Sons, 1957.

LEIPZIGER, DANNY M., ed. Lessons from EastAsia. Ann Arbor: U. of Michigan Press, 1997.

LEVINE, ROSS AND RENELT, DAVID. “A Sensitiv-ity Analysis of Cross-Country Growth Regres-sions,” Amer. Econ. Rev., Sept. 1992, 82(4), pp.942–63.

LEVINE, ROSS AND ZERVOS, SARA J. “What WeHave Learned About Policy and Growth fromCross-Country Regressions,” Amer. Econ. Rev.May 1993, 83(2), pp. 426–30.

LEWIS, W. ARTHUR. “Economic Developmentwith Unlimited Supplies of Labor,” ManchesterSchool, May 1954, 22(2) pp. 139–91.

———. The theory of economic growth. London:George Allen and Unwin, 1955.

———. Growth and fluctuations 1870–1913. Lon-don: George Allen and Unwin, 1978.

———. “The Dual Economy Revisited,”Manches-ter School, Sept. 1979, 47(3), pp 211–29.

LINDER, STAFFAN B. An essay on trade and trans-formation. New York: John Wiley, 1961.

LITTLE, IAN M. D. Economic development. NewYork: Basic Books, 1982.

LITTLE, IAN; SCITOVSKY, TIBOR AND SCOTT,MAURICE. Industry and trade in some develop-ing countries. London: Oxford U. Press for theOECD, 1970.

MACARIO, SANTIAGO. “Protectionism and Indus-trialization in Latin America,” Econ. Bull. forLatin America, Mar. 1964, 9(1), pp. 61–102.

MADDISON, ANGUS. Economic progress and pol-icy in developing countries. London: George Al-len and Unwin, 1970.

———. Monitoring the world economy. Paris: De-velopment Centre, OECD, 1995.

MAHALANOBIS, P. C. “The Approach of Opera-tional Research to Planning in India, Sankahya,Dec. 1955, 16(1, 2), pp. 3–120.

MALENBAUM, WILFRED. Prospects for Indian de-velopment. London: George Allen and Unwin, 1962.

MEIER, GERALD M. AND BALDWIN, ROBERT L.Economic development. New York: John Wileyand Sons, 1957.

MEIER, GERALD M. AND SEERS, DUDLEY, eds.Pioneers in development. New York: Oxford U.Press, 1984.

MELLOR, JOHN W. Agriculture on the road to in-dustrialization. Baltimore, Johns Hopkins U.Press, 1995.

MORAWETZ, DAVID. Twenty-five years of eco-nomic development. Baltimore: Johns HopkinsU. Press for the World Bank, 1977.

MORGAN, THEODORE; BETZ, GEORGE W ANDCHOUDHRY, NANKA K., eds. Readings in devel-opment economics. Belmont, CA: WadsworthPub. Co. 1963.

MORISHIMA, MICHIO. Why has Japan succeeded?Cambridge: Cambridge U. Press, 1982.

MORRISON, CHRISTIAN, ed. “Adjustment withGrowth and Equity,” World Devel., Nov. 1994,19(11), pp. 1483–651.

MORRIS-SUZUKI, TESSA. A history of Japanesethought. London: Routledge, 1989.

———. The technological transformation of Ja-pan. Cambridge: Cambridge U. Press, 1994.

MOSLEY, PAUL J; HARRIGAN, JOHN AND TOYE,JOHN. Aid and power: The World Bank and pol-icy based lending. London: Routledge, 1991.

MYRDAL, GUNNAR. Economic theory and the un-derdeveloped regions. London: G. Duckworth& Co., 1956.

NELSON, RICHARD R. AND PACK, HOWARD.“Firm Competencies, Technological Catchup,and the Asian Miracle.” Paper presented at con-ference in honor of Gustav Ranis, Yale U., NewHaven, 1996.

NORTH, DOUGLASS C. “Economic PerformanceThrough Time,” Amer. Econ. Rev., June 1994,84(3), pp. 359–68.

NURKSE, RAGNAR. Problems of capital formationin underdeveloped countries. New York and Ox-ford: Oxford U. Press, 1953.

OLSON, MANCUR. The rise and decline of nations.New Haven: Yale U. Press, 1982.

———. “Big Bills Left on the Sidewalk: WhySome Nations are Rich and Others Poor,” J.Econ. Perspectives, Spring 1996, 18(2), pp. 3–24.

PACK, HOWARD AND WESTPHAL, LARRY E. “In-dustrial Strategy and Technological Change,” J.Devel. Econ. Mar. 1986 22(1), pp. 87–128.

PASINETTI, LUIGI L. AND SOLOW, ROBERT M.,eds. Economic growth and the structure oflong-term development. London: St. Martin’sPress for the International Economic Associa-tion, 1994.

PREBISCH, RAUL. The economic development ofLatin America and its principal problems. LakeSuccess, New York: United Nations Depart-ment of Social Affairs, 1950.

———. Towards a dynamic development policyfor Latin America. New York: United Nations,1964.


RAFFER, KUNIBERT AND SINGER, HANS W. Theforeign aid business. Cheltenham, UK: EdwardElgar, 1996.

RANIS, GUSTAV. “A Comment,” World Develop.,Apr. 1985, 13(4) pp. 543–45.

———. “Taiwan as a Classic Model of Asian De-velopment Success,” in African Development:Lessons from Asia, 1991, pp. 23–48.

———. “Another Look at the East Asian Miracle,”World Bank Econ. Rev., Sept. 1995, 9(3), pp.509–34.

REDDING, S. GORDON. The spirit of Chinese capi-talism. Berlin and New York, Walter deGruyter, 1993.

———. “Societal Transformation and the Contri-bution of Authority Relations and CooperationNorms in Overseas Chinese Business,” in WEI-MING TU, ed., 1996, pp. 310–27.

RIDDELL, ROGER C. Manufacturing Africa. Lon-don: James Currey, 1990.

ROSENSTEIN-RODAN, PAUL N. “Problems of In-dustrialization of Eastern and SoutheasternEurope,” Econ. J., June–Sept. 1943, 53(2), pp.202–11.

———. “International Aid for UnderdevelopedCountries,” Rev. Econ. Statist., May 1961,43(2), pp. 107–38.

SHEAHAN, JOHN. “Import Substitution and Eco-nomic Policy: A Second Review.” ResearchMemorandum No. 50. Center for DevelopmentEconomics, Williams College, Williamstown,MA., 1972.

SINGER, HANS W. “The Distribution of Gains be-tween Investing and Borrowing Coun-tries,”Amer. Econ. Rev., May 1950, 40(2), pp.473–85.

———. “Dualism Revisited: A New Approach tothe Problems of the Dual Society in DevelopingCountries,” J. Devel. Studies, Oct. 1970, 7(1),pp 60–75.

SINGER, HANS W. AND ROY, SUMIT. Economicprogress and prospects in the third world.Aldershot: Edward Elgar, 1993.

STERN, JOSEPH J. ET AL. Industrialization and thestate: The Korean heavy chemical industrydrive. Cambridge; Harvard Institute for Inter-national Development and Korean Develop-ment Institute. Distributed by Harvard U.Press, 1995.

STIGLITZ, JOSEPH E. “Learning to Learn, Local-ized Learning and Technological Progress.”Princeton U., Department of Economics, Re-print No. 7, 1987.

STREETEN, PAUL, ed. Trade strategies for devel-opment. London: Macmillan Press, 1973.

TOKUNAGA, SHOJIRO, ed. Japanese foreign invest-ment and Asian economic independence. Tokyo:U. of Tokyo Press, 1992.

TOYE, JOHN. Dilemmas of development. Oxford:Basil Blackwell, 1987.

TSAO, YUAN. “Growth without Productivity: Singa-pore Manufacturing in the 1970s,” J. Devel.Econ., 1985, 19(1). pp 25–39.

TSURU, SHIGETO. Japan’s capitalism: creative de-feat and beyond. Cambridge: Cambridge U.Press, 1993.

TU, WEI-MING, ed. Confucian traditions in EastAsian modernity. Cambridge: Harvard U. Press,1996.

TYBOUT, JAMES. “Linking Trade and Productivity:New Research Directions,” World Bank Econ.Rev., May 1992, 6(2), pp. 189–211.

UNITED NATIONS, COMMISSION FOR ASIA ANDTHE FAR EAST. Programming techniques foreconomic development. Bangkok, United Na-tions, 1960.

VINER, JACOB. International trade and economicdevelopment. Glencoe, Illinois: Free Press, 1952.

WADE, ROBERT. Governing the market. Prince-ton: Princeton U. Press, 1990.

WATERSTON, ALBERT. Development planning.Baltimore: Johns Hopkins U. Press, 1965.

WESTPHAL, LARRY E. “The Republic of Korea’sExperience with Export-led Industrial Develop-ment,” World Devel., Mar. 1978, 6(3), pp. 347–82.

———. “Industrial Policy in an Export PropelledEconomy: Lessons from South Korea’s Experi-ence,” J. Econ. Perspectives, Summer 1990,4(3), pp. 41–60.

WINSTON, GORDON C. “The Theory of CapitalUtilization and Idleness,” J, Econ. Lit., Dec.1974, 12(4), pp. 1301–20.

WOLFE, MARSHALL. “Approaches to Develop-ment: Who Is Approaching What?” CEPALRev., 1976, pp. 131–71.

WORLD BANK. The East Asian miracle. NewYork: Oxford U. Press for the World Bank,1993.

YOSHIHARA, KUNIO. The rise of ersatz capitalismin South East Asia. Singapore: Oxford U. Press,1988.

YANG, YA-HWEI. “Government Policy and Strate-gic Industries: The Case of Taiwan,” in TAKA-TOSHI ITO AND ANNE KRUEGER, eds., 1993,pp. 387–408.

YOUNG, ALWYN. “The Tyranny of Numbers: Con-fronting the Statistical Realities of the EastAsian Growth Experience, Quart. J. Econ., Aug.1995, 118(3) pp. 641–81.


bruton: a reconsideration of import substitution a ...faculty.nps.edu/relooney/bruton.pdf · of the...

Documents