finance & development · u tun wai david williams christopher willoughby shahid yusuf finance...

Bahrain NowzadEDITORShuja NawazMANAGING EDITORJoslin Landed-MillsASSISTANT EDITOR

EDITORIAL OFFICERRichard StoddardART EDITORDeborah SalzerGRAPHICS ARTIST

ADVISORS TO THE EDITORAndrew CrockettBarend de VriesHoward HandyPaul IsenmanArturo IsraelClaudio LoserGuy PfeffermannAlexander ShakowAlan TaitU Tun WaiDavid WilliamsChristopher WilloughbyShahid Yusuf

Finance & Development is published quarterly in English, Arabic,French, German, Portuguese, and Spanish by the InternationalMonetary Fund and the International Bank for Reconstruction andDevelopment, Washington, DC 20431, U.S.A. (DSPS 123-250). Secondclass postage is paid at Washington, DC and at additional mailingoffices. • English edition "ISSN 0015-1947. • Opinions, expressed inarticles and other material are those of the authors; they do notnecessarily reflect Fund or Bank policy. • New readers who wish toreceive Finance & Development regularly should-apply in writing toFinance & Development, International Monetary Fund, Washington,DC 20431, U.S.A., specifying the language edition and briefly statingthe reasons for their request. • The contents of Finance & Devel-opment are indexed in Business Periodicals Index, Public AffairsInformation Service (PAIS), and Bibliographic Internationale desSciences Sociales. An annual index of articles and book reviews iscarried in the December issue.

^^^^te^ga

New Fund publicationsEmerging Financial Centers: Legal andInstitutional FrameworkEdited by Robert C, Effros, Assistant General Counsel (Legislation)of the Fund's Legal Department, this volume sets out the legal andinstitutional framework of seven developing financial centers—theBahamas, Hong Kong, Ivory Coast, Kenya, Kuwait, Panama, andSingapore. After an introduction by the editor, in which theprincipal characteristics of the centers are summarized and com-pared, each center is dealt with separately in a common format.Essays detail each center's economic background and reviewrecent developments. This is followed by a reproduction of themonetary, central banking, and general banking laws, as well asselected laws of the main specialized financial institutions and thelaws, if any, relating to the regulation of securities.

Pp. xvi + 1150. Price: US$35.

Occasional Paper No. 15

Hungary: An Economic Surveyfay a staff team headed by Patrick de Fontenay

This paper is based in part on a study prepared by the staff whenHungary applied for membership in the Fund. (Hungary enteredthe Fund on May 6, 1982.) It reviews the gradual reform of theHungarian economic system, the performance of its economy inrecent years, and some of the problems facing Hungary at present.

Occasional Paper No. 16

Developments in International Trade Policyby S.J. Anjaria, Z. Iqbal, N. Kirmani, and L.L. Perez

The paper focuses on the main current issues in trade policies ofthe major trading nations. It discusses increasing pressures forprotection arising from both cyclical factors and structural rigidities,the proliferation of restrictive measures in the industrial and agricul-tural sectors, and the implications of recent trade developments fordeveloping countries.

The paper assesses the shift away from a liberal trade policystance, the increasing preference for dealing with problems ofimport competition through measures that may not be as easy tochallenge under existing multilateral rules as were more traditionalforms of restrictions at the border, and the tendency to alleviatetrade frictions through sectoral and bilateral approaches to tradepolicy. It also outlines recent international initiatives that affecttrade policy, including structural adjustment, export subsidies, anddispute settlement procedures. Statistical tables present data tosupport the analysis of current developments in trade policy.

Price of Occasional Papers: US$5 each (US$3 for universitylibraries, faculty, and students).

Advice on payment in currencies other than the U.S. dollar willbe giuen on request.

Copies of the above publications are available from: Publications UnitInternational Monetary FundBox A-101Washington, DC 20431 U.S.A.

Shella meshan

©International Monetary Fund. Not for Redistribution

March 1983/Volume 20/Number 1

A quarterly publication of the International Monetary Fund and the World Bank

An/aria et al. 2 ProtectionismIncreasing restrictions on trade; implications for developing countries

Helen Hughes 6 Capital utilization in manufacturingA study of optimal levels of utilization in selected developing countries;some policy advice

Jacques Artus 10 Toward a more orderly exchange rate systemExchange rate instability and how to reduce it

Karim Nashashibi 14 Devaluation in developing countries: the difficult choicesOften bitter, but better than the alternatives

Stephen Heyneman 18 Improving the quality of education in developing countriesBetter instruction and teaching tools are needed

22 External debt—the continuing problem

David Williams 24 Opportunities and constraints in international lendingImplications of increased bank lending for the international monetarysystem

Russell Kincaid 28 What are credit ceilings?

Takamasa Akiyama and 30 Coffee and COCOa trendsRonald Duncan A cloudy outlook

Shuja Nawaz 34 Economic impact of defense expenditures

Keith Palmer 36 Private sector petroleum exploration in developing countriesLess than anticipated activity in the 1970s; why?

39 World economy in transition: Trade shares and trends, 1950-81

40 Cofinancing—new World Bank approaches

41 Recent staff studies

42 Book noticesGuy Pfeffermann and The Role of Economic Advisers in Developing Countries,

Vinod Thomas by Lauchlin CurrieClaudio Loser International Financial Cooperation: A Framework for Change,

by Frances Stewart and Arjun Sengupta; edited by Salah Al-ShaikhlyOther books received

45 Letters

The Editor welcomes views and comments from readers on the contents of the magazine.7»e contents of Finance & Development may be quoted or reproduced without further permission. Due acknowledgement is requested.

_ FinancesDeveloDrnent


ProtectionismA review of the recent resort totrade restrictions, factorsinfluencing trade policies, andthe implications of protection forthe developing countries.

For much of the period from the end ofWorld War H to 1973, world trade grew rap-idly, aided by significant liberalizationachieved in successive rounds of multi-lateral trade negotiations. The world econ-omy became increasingly integrated as thegrowth of trade outpaced production.Since 1973, although export performancehas differed widely between and withinbroad country groupings, global trade vol-umes have been erratic, growth has slowedconsiderably, and the pace of integrationhas stabilized. By 1981, when output roseby only 1 per cent, world trade hadstagnated.

A major problem currently facing theworld economy is a substantial underutili-zation of resources. The unemploymentrate in seven major industrial countries,which averaged just over 3 per cent in1963-72, exceeded 6Vi per cent in 1981, andfurther increased in 1982. Capacity utiliza-tion is generally low, and employment op-portunities have been eroded severely insome industrial sectors.

In 1981-82, restrictive actions on tradebecame more widespread in both the in-dustrial and the agricultural sectors. Themeasures affecting trade in industrial prod-ucts have been recognized, in principle, asbeing exceptions from the existing liberalframework of the General Agreement onTariffs and Trade (GATT), to be appliedtemporarily. For legal and historical rea-sons, GATT rules have not applied equallyrigorously to agricultural trade. Policy-makers continue to recognize the need toovercome structural rigidities and to en-courage goods to be exchanged on the basisof comparative costs. Not surprisingly,however, there is little agreement on spe-cific measures to encourage structural ad-justment. Agricultural protection has againbecome a debated issue—both in the pro-tecting countries, because of its economic

2 Finance St Development / March 1983

and budgetary costs, and in the competi-tive exporting countries, because of theburden of adjustment it imposes on theiragricultural sectors. But the central prob-lem of international trade policy persists—the urgent need for structural change re-mains greater than the adjustment that isunder way.

The rise in protectionist pressures is wor-risome, because the likelihood of chain re-actions toward more protectionism gener-ated by individual restrictive actions isgreatest in a setting of slow economicgrowth and highly interdependent econo-mies. Two broad developments are of par-ticular concern. First, with the conclusionof the Tokyo Round of Multilateral TradeNegotiations (MTN) in 1979, the momen-tum toward trade liberalization that hadbeen sustained for three decades appearedto have been dissipated. Protectionist pres-sures and protective actions that weretaken or intensified since the conclusion ofthe Tokyo Round affect sectors accountingfor more than one fifth of world trade inmanufactures, including iron and steel,automobiles, textiles, and clothing. In ad-dition, industrial countries apply restric-tions at the border or use other measuresthat affect or distort trade in temperatezone agricultural products that account forone third of international trade in agricul-ture, including sugar. Apart from the in-crease in the number of restrictive actions,domestic and export subsidies have alsoincreased in recent years in many majortrading nations.

A second cause for concern is that thereappears to have been a perceptible shift inthe attitude of policymakers, who seem tobe increasingly preoccupied with bilateraltrade balances. Bilateral approaches also re-flect a concern that foreign competitors arereceiving an undue advantage as a result oftheir domestic policies and institutional ar-rangements. In addition, efforts to protectparticular sectors have become more en-trenched, jeopardizing the prospects fordeveloping and strengthening the rules ofgeneral application that constitute the mul-tilateral trading system. This shift couldbring about greater international accept-ance of existing sectoral restrictions or leadto further regulation of international tradeflows; at worst, it could lead to balancing

trade flows, bilaterally or sectorally, atprogressively lower levels. Political andstrategic considerations also appear to haveinfluenced trade policy in several recentinstances.

Factors in trade policies

Protectionist pressures tend to increaseduring periods of cyclical downturns ineconomic activity and the associated rise inunemployment, as industries and workersseek government action to insulate sectorsof the economy from import competition.The groups threatened by import competi-tion tend to be the traditional industries,where political influence is relatively welldeveloped. On the other hand, those hurtby protectionist policies, including con-sumers and taxpayers, generally do not in-fluence policy as actively. Pressures be-come particularly acute when recessionaryconditions spread worldwide—as they did,for example, in 1974-75 and as they havesince 1980—and the possibilities for main-taining employment through increasedexport activity diminish. Furthermore, pro-tection granted through restraints on im-ports and subsidies for exports during peri-ods of rising unemployment is, in practice,difficult to dismantle speedily when eco-nomic activity improves. Unless the recov-ery is sustained over relatively long peri-ods, there is a tendency for protection topersist when the need for it has passed.

'

: s

TTtis article draws selectively on a fuller study,Developments in International Trade Poli-cy, IMF Occasional Paper No. 16, by S. ].Anjaria, Z. labal, N. Kirmani, and L. L. Perez.The study, completed in July 1982, focused onthe main current issues in the trade policies of thema/or trading nations. In addition, it outlinedthe principal developments in the recent evolu-tion of the international trading system, includ-ing international initiatives in such areas asstructural adjustment, safeguards, agriculturaltrade policies, and dispute settlement proced-ures. The focus of this article is on protectionistmeasures or pressures in major industrial coun-tries, because these have a pronounced effect onthe world economy. The negative effects of devel-oping countries' own protectionist policies,which are also important, are not covered in thisarticle. See inside front cover for ordering infor-mation on Developments in InternationalTrade Policy.


Protectionist pressures are also closelyrelated to underlying structural rigidities;such constraints had become increasinglyapparent after 1974-75 in several industrialcountries. They have been manifested as:(1) successive rounds of inflation over thelast two decades, each beginning at a high-er level; (2) high and growing unemploy-ment, particularly of semiskilled laborers intraditional industries, who are difficult toretrain and relocate; and (3) sluggish pri-vate investment. The Organization for Eco-nomic Cooperation and Development(OECD), in 1978, adopted guidelines onpositive adjustment that emphasized theimportance of pursuing policies consistentwith efficient resource allocation. Commit-ment to this principle was renewed by gov-ernments in the OECD Declaration onTrade Policy, adopted June 4, 1980.

In addition to the OECD initiatives, re-cent discussions at the GATT and the Unit-ed Nations Conference on Trade and De-velopment (UNCTAD) have focused on thegeneral policy issues relating to structuraladjustment. The practical difficulties ofadopting more positive adjustment policieswere compounded by the cyclical down-turn that began in 1980. The problems en-suing from structural rigidities in many ma-jor trading nations have been made worse,too, by the rapid emergence of the newlyindustrializing countries as competitors ina wide range of sectors.

The recent pressures for protection havesometimes been attributed to exchange ratedevelopments. Some exchange rates havecontinued to diverge from levels consistentwith the underlying economic fundamen-tals. The effects of these on trade have attimes encouraged protectionist forces incountries experiencing trade deficits to de-mand bilateral restrictions. Moreover, in-creased exchange rate instability is said tohave impeded rational long-run policy-making by business. Finally, it is some-times contended that trade liberalization isless important under a floating exchangerate system than under fixed exchangerates because the effects of arduously nego-tiated tariff concessions are likely to beswamped by large exchange rate changes.

The importance of exchange rates thatreflect economic fundamentals has longbeen recognized and is central to Fund sur-veillance over members' exchange rate pol-icies. However, protectionism is not a vi-able solution to the problems arising frominappropriate exchange rate levels or ex-change rate instability. Protectionist mea-sures have direct, industry-specific effectson resource allocation. Exchange ratechanges and trade liberalization do not off-set each other. Protectionism is likely to

lead to a misallocation of resources, to slowthe pace of necessary structural adjust-ment, and to invite retaliation in othercountries. Indeed, protectionist measurestend to perpetuate inappropriate exchangerate levels.

Industrial trade policies

The loss of dynamism in world trade hasbeen particularly marked in the case ofmanufactured products, which make up 55per cent of world trade and 70 per cent oftrade among industrial countries. Thegrowth of world exports of manufacturesdecelerated from 11 per cent annually dur-ing 1963-73 to 5 per cent during 1974-80and further to 3 per cent in 1981. In1981-82, a few existing import restrictionswere liberalized, but the main trend wastoward increased protection that affectednot only historically protected industriesbut also new sectors.

An important set of recent protectionistactions has been in the automobile sector,which accounts for 8 per cent of total worldtrade in manufactures. In 1981, bilaterallimitations or other forms of restraints wereintroduced on Japanese exports to severalindustrial countries, including the UnitedStates, that together account for two thirdsof Japan's exports of automobiles. Com-bined with previous restrictions, the mea-sures now affect virtually all Japaneseexports of automobiles to industrial coun-tries. The spread of protection in this im-portant sector is symptomatic of the prob-lems of structural adjustment in the indus-trial countries.

International trade in iron and steel ac-counts for 4 per cent of world trade and hasbeen growing relatively rapidly despite re-strictions imposed periodically by one ormore industrial countries since the 1960s.Since early 1981, pressures for protectionhave remained high. In October last year, anew restraint arrangement covering the Eu-ropean Community's steel exports to theUnited States was concluded. Given thepersistence of trade and adjustment prob-lems in this sector and the increasing in-roads that competitive developing countrysuppliers have recently made in the mar-kets of the industrial countries, it wouldnot be surprising if suggestions for nego-tiating still wider restrictions on steel con-tinued to surface from time to time.

Trade restrictions in the textile andclothing sector, which constitutes about 5per cent of world trade, also raise difficultissues of trade policy. In December 1981,the Multifiber Arrangement, which autho-rizes importing countries to restrict importsfrom individual exporting countries, wasextended by a protocol of extension to July1986 (MFA HI). Although it is still too early

to judge the effect of MFA III on the exportprospects of developing countries, indi-cations are that it may be used by someimporting countries to restrict exports of"dominant"—that is, efficient—develop-ing country suppliers more severely thanthose of other countries. At the expirationof MFA III, multilaterally negotiated re-strictions in textiles and clothing will havebeen applied for nearly 25 years. Duringthis period, they have become pro-gressively more complex and comprehen-sive. In other industrial sectors that havebeen subject to protectionist pressures inrecent years, such as footwear, petrochem-icals, shipbuilding, and electronics, the pic-ture is somewhat mixed.

Sectoral pressures for protection inNorth America and Europe have com-monly arisen with respect to the industrialperformance of Japan. The flexibility of theJapanese economy and its ability to shiftresources rapidly to more promising lineshave been generally recognized. Never-theless, pressures in North America andEurope have increased to address allegedspecific problems in their bilateral tradewith Japan. In late 1981 and early 1982,Japan announced measures to facilitate im-ports and streamline import proceduresand to deal more effectively with com-plaints from abroad through the establish-ment of a trade ombudsman's office. Themeasures included the elimination and re-duction of some tariffs, and the simplifica-tion and relaxation of certain proceduresand standards perceived as nontariffbarriers.

Agricultural trade policies

In the agricultural sector, the majortrading nations pursue the following objec-tives in varying degrees, though they arenot necessarily mutually consistent: self-sufficiency or security of supplies in food;parity or fair income for the domestic farmsector; market stabilization; and reasonableprices to the consumer. In addition, agri-cultural policies also take account of social,regional development, environmental, andhealth considerations. Agricultural produc-tion is inherently subject to the vagaries ofthe weather, which can bring about sharpvariations in supplies. These factors, com-bined with generally low short-term de-mand and supply price elasticities, tend toproduce relatively wide price fluctuations.

International trade policy in the agricul-tural sector is often designed to achieve themain objective of income support for do-mestic farmers through a variety of mea-sures, including import quotas, tariffs,variable import levies, and export sub-sidies, and is often designed to mitigate theextent of market fluctuations. Market de-

Finance & Development / March 1983 3


velopments are also related to policies onfood security and food aid, and they areinfluenced by preferential access commit-ments undertaken by some industrialcountries. In addition, international agree-ments exist for sugar, dairy products,grains, and meat, but only the sugar anddairy products agreements contain pricingand other economic provisions.

The Fund's study focused on the fiveprincipal agricultural commodities pro-duced in temperate zones—dairy products,

porting countries for the commodities con-cerned. For the United States, the ratioshave generally remained, or have comedown to, around or below unity.

Although trade policies and market con-ditions for each of the main temperate zoneproducts differ, a broad generalization thatemerges from the survey of agriculture isthat national policies have given an un-equivocal priority to the achievement of do-mestic social and political objectives, to thedetriment of trade liberalization. In the few

World production and exports, 1963-81

Growth of worldcommodity output

1963

5.4

1973

8.7

1975

-1.1

1977

In per cent

4.6

In billions of current U.S

World exportsOf which

Agricultural productsMinerals1

Manufactures

Growth of volume ofworld exportsSource: LiAl I , Internationa/ Trade— Indicates zero.

154 574 873 1,125 1

1979

2.8

.dollars

635

1980

1.4

1,985

1961

1.3

1,970In per cent of world exports

29.216.953.2

7.1

1981/82.

21.116.760.5

11.1

17.223.757.4

-2.9

16.723.657.6

In per cent

4.3

16.024.557.8

5.6

15.028.555.2

1.7

14.727.456.1

—

'Including fuels and nonferrous metals.

fats and oils, grains, meats, and sugar.While the specific means of protection varyfrom product to product, virtually all majorindustrial countries protect their domesticagricultural sectors to a considerable ex-tent. In the fats and oils sector, protectionis relatively low. In the other sectors, highdomestic price supports often encourageproduction that cannot be absorbed at pre-vailing prices; this obliges the disposal ofstocks in international markets, displacingmore efficient producers.

An indication of the extent to which in-dustrial countries' agricultural sectors areinsulated is given by the generally low ratioof international trade to global production.This insulation of domestic markets pro-duces a considerable divergence in domes-tic wholesale prices among countries. Acomparison of nominal protection coeffi-cients, that is, the ratio of domestic whole-sale to "international" price, for selectedagricultural commodities in the three majorindustrial areas—the European Commu-nity, Japan, and the United States—isillustrative of the degree of agricultural pro-tection, although subject to the many qual-ifications outlined in the study on whichthis article is based. The nominal protectioncoefficients are well above unity, reflectingpositive protection, in both the EuropeanCommunity and Japan, the major im-

products where international trade has re-mained relatively free, such as cereal sub-stitutes, the liberal stance of policies is in-creasingly in jeopardy. Long-term tradeagreements that involve elements of mar-ket organization could become increasinglyimportant in the future. Currently, suchagreements reportedly cover principallygrains (accounting for at least 15 per cent ofinternational grain trade), beef, soybeans,and dairy products.

The economic effects of agricultural pro-tection can take the form of overproductionand underconsumption of agricultural out-put in the protecting country, as well asdistortions in the allocation of resources be-tween agriculture and other sectors. Largeexportable surpluses may develop that maybe disposed of in the international marketthrough export subsidization, imposingcosts on consumers or taxpayers in the pro-tecting country and on efficient producersabroad.

According to some studies, including arecent report by the OECD Secretariat, re-form of agricultural policies might not pro-voke the massive structural dislocationsthat are sometimes feared. Moreover, evena relatively small reduction in surpluses inthe traditional importing countries wouldbring about a significant improvement inconditions of access for the traditionally ef-

ficient suppliers of agricultural productssuch as dairy products and meat. In spite ofthe somewhat higher rate of growth ofworld agricultural trade since 1973, the pro-portion of total production entering trade isstill generally small. More open agriculturaltrade policies may reduce market instabilityand possibly raise the average level of freeinternational market prices for certain com-modities, while lowering high consumerprices in certain countries. Even so, the es-tablishment of more liberal internationaltrading conditions in agriculture is still notgenerally accepted as a priority. In the ab-sence of new policy initiatives to promotereform of agricultural trade policies, an in-tensification of restrictions through a prolif-eration of bilateral arrangements cannot beruled out; this would pose serious con-sequences for efficient producers.

Implications for LDCsIn the aggregate, non-oil developing

countries' exports have since 1973 grownrather well, but within this group therehave been wide disparities in export per-formance. Non-oil developing countries'export performance and growth rates de-pend on a number of factors in addition tothe stance of trade policies in the industrialcountries. Among such factors, analyzed ina recent study by Goldstein and Khan (Ef-fects of Slowdown in Industrial Countries onGrowth in Non-Oil Developing Countries, IMFOccasional Paper No. 12), are the rate ofeconomic growth of the industrial coun-tries; the stance and general orientation ofthe developing countries' own economicpolicies (including their trade policies,which are not discussed here); the scope ofdemand and production linkages betweentheir export and domestic sectors; and thequantity and efficiency of physical and fi-nancial capital available in these countries.The observed variations in the developingcountries' export and growth performancemust therefore be assessed with referenceto the sensitivity of their existing produc-tion and trade patterns to the interaction ofthese factors. In the past three years, inparticular, the impact of the economicslowdown in industrial countries on the ex-port performance of many developingcountries has been marked.

Although there are obvious and impor-tant differences in export and growth per-formance among the developing countries,there is little doubt that protectionism inthe major trading nations has hampereddeveloping countries' efforts to specializein production in accordance with theircomparative advantage. The MTN most-favored-nation tariff reductions on indus-trial products exported by developing

Finance & Development I March 19834


countries were, on average, less deepthan the overall cut. Tariff escalation—in-creasing nominal protection by stage ofprocessing—on products of export interestto developing countries remains a problem,and it has received increased attention re-cently in the GATT and elsewhere. The in-cidence of nontariff barriers is frequentlyhigher in sectors where developing coun-tries have a comparative advantage; the im-position of nontariff barriers on agriculturalcommodities produced in both industrial

the GATT. Attempts to apply this clausehave, however, generated trade frictions,and the precise criteria for what constitutes"development" and "improvement intrade situation" are—and probably will re-main—controversial issues. If successfulefforts by more advanced developing coun-tries to pursue export-oriented develop-ment strategies are frustrated by selectiveprotection, export diversification of otherdeveloping countries and their greater inte-gration in the world economy may well be

necessitate or validate restrictions or per-petuate trade distortions, longer-term in-vestment planning becomes more uncer-tain. This can occur even if individual traderestrictions are not directed specifically atimports from developing countries.

Several developing countries, includingsome newly industrializing ones, have re-cently begun to rationalize and simplifytheir import regimes, as well as to reduceimport restrictions and their reliance on ex-port subsidies. These efforts would be

and developing countries is widespread;and in textiles and clothing, restrictions areapplied to exports from developing coun-tries, while those from other countries thataccount for a significant portion of worldexport are generally unrestricted.

During the MTN and subsequently, con-siderable controversy has arisen on theconcept of "graduation." The enablingclause agreed in the MTN states that, withthe progressive development in their econ-omies and improvement in their tradesituation, developing countries would beexpected to participate more fully in theframework of rights and obligations under

hindered. Inward-looking bilateral and re-gional approaches to trade policy could beencouraged, and these would be contraryto the interests of all countries.

The existing structure of protection andthe tendency of major trading nations toapproach trade issues on a bilateral or sec-toral basis also have detrimental effects ondeveloping countries' longer-term pros-pects for increasing exports. Even in sec-tors where access to major markets is rela-tively liberal, these approaches generateuncertainty about future continued access.If major trading nations avoid pressures forstructural change by applying policies that

greatly discouraged if it appeared thatopen, multilateral approaches to tradeproblems were to be abandoned by the ma-jor trading nations. Experience indicatesthat when restrictions, such as those in theagricultural and textile and clothing sec-tors, become entrenched, the relativelysmaller trading nations are obliged to seekthe best possible accommodation of theircommercial interests in order to avoid theloss of markets. This further weakens thebasis for the pursuit of more fundamentalliberalization or reform of trade policies de-riving from multilateral rules of generalapplication. HD


GATT ministerial meetingThe Genera! Agreement on Tariffs and Trade (GATT) establishes theeconomic, legal, and institutional framework for the conduct of worldtrade. Asa result of successive rounds of trade negotiations conductedunder GATT auspices, tariff barriers to trade in manufactures arerelatively low. Because of the rise in protectionist pressures and re-strictive actions in recent years, the GATT CONTRACTING PARTIESheld a meeting at ministerial level—the first since 1973—in Geneva inNovember 1982 to assess the functioning of the multilateral tradingsystem and to agree on its direction for the coming years. The resultsof the GATT meeting are embodied in a wide-ranging MinisterialDeclaration. In it, the Ministers recognized that "protectionistpressures on governments have multiplied, disregard of GATT disci-plines has increased, and certain shortcomings of the GATT systemhave been accentuated," and they resolved to overcome these threats tothe system.

A key element of the 16-page Ministerial Declaration is the un-dertaking by CONTRACTING PARTIES "to make determined efforts toensure that trade policies and measures are consistent with GATTprinciples and rules, and to resist protectionist pressures in the formu-lation of national trade policy and in proposing legislation." Ministersalso undertook "to refrain from taking or maintaining any measuresinconsistent with GATT and to make determined efforts to avoidmeasures which would limit or distort international trade." In addi-tion, a detailed work program to be pursued under GATT auspices wasagreed.

Both the President of the World Bank, A. W. Clausen, and theManaging Director of the Fund, Jacques de Larosiere, speaking on thefirst day of the meeting, issued strong warnings about the dangers of

protectionism. Mr. Clausen made three points: • ".. .free and ex-panding trade is crucial for Third World development." • "... it isneeded urgently now to help some developing countries overcome theirdebt problems." And, • "... in order to liberalize commerce... weneed to overcome a fundamental weakness in decisionmaking abouttrade: National and international procedures... are one-sidedly sen-sitive to interest groups that stand to lose from international com-petition."

Liberal trade, he stressed "can help us escape from the stop-and-gobetween recession and inflation that has hamstrung the world economyin recent years. Free trade fosters growth—and therefore jobs andhigher incomes—-and, at the same time, it tends to restrain inflation.The benefits that a nation gains from trade are more than enough tocover the costs.. .."

Making the point that the "smooth operation of the internationalfinancial system and the efficient working of the adjustment pro-cess are dependent upon the maintenance of open trade policies,"Mr. de Larosiire emphasized the dangers of protectionism. In theimmediate future, "It is absolutely essential that debtor countries'access to world markets is not restricted if they are to be in a positionto service their external debt." For the longer term, he stressed threepervasive consequences of protection: it obstructs the adjustment pro-cess, impeding the optimal utilization of resources; it endangers thewhole outward-looking strategy that has fostered postwar develop-ment; and it threatens the prospects for medium-term recovery in theworld economy. "The Fund," he said, "has a vital interest in pre-serving the open trading system. It also has an important role to playin supporting that objective... through the exercise of surveillanceand through its financing in support of adjustment programs."


The optimum utilization ofcapital in developing countriesis an important but complexaspect of development strategy.

Helen Hughes

Most countries use only a small proportionof the capital invested in manufacturing (inthe form of buildings and equipment) atfull capacity. Even though interest must bepaid on that capital stock (at least no-tionally), often the only plants that operatecontinuously involve processes such aspetrochemicals and iron and steel that can-not be shut down without very high costs.

The reasons for this "waste" are obviousin high-income countries. Workers andmanagers do not like to work in the earlymorning, late afternoon, at night, on week-ends, or on holidays, and society can affordthe cost of such preferences, which are ex-pressed as premiums for work outside reg-ular hours. However, in developing coun-tries, the failure to utilize capital aroundthe clock is a luxury that can ill be afforded.A few rapidly developing economies, suchas Hong Kong, have a very high level ofcapital utilization; studies conducted in the1960s and early 1970s in several less indus-trialized countries, however, estimatedthat manufacturers used their buildingsand equipment for only about 15 per cent ofthe year. It was generally believed that suchpercentages were extremely and unneces-sarily low, and that higher levels of utiliza-

tion would increase the productivity of cap-ital. Further, by doubling or trebling thenumber of shifts worked, higher levels ofcapital utilization would also enable newemployment to be increased without addi-tional investment.

To arrive at policy recommendations tobridge the gap between utilization levelsthat are appropriate and those that exist indifferent countries, a World Bank studyfound that levels of capital utilization were,on average, higher than expected but stillreflected considerable underutilization. Inmost cases, the actual level was intendedand was rational, given domestic policiesand costs; distorted domestic policies andcosts were the main reasons for this appar-ently "inefficient" use of capital in eachcountry studied. Eliminating a variety ofpolicy distortions is thus necessary toachieve "optimum" levels of capital utiliza-tion for particular countries at particularstages of development.

The study

Manufacturing enterprises in four coun-tries were selected for the comparativestudy—Colombia, Israel, Malaysia, and thePhilippines. All four were at a relativelyadvanced stage of development, althoughdifferences in natural endowments, in thesize of the domestic market, and particu-larly in policies meant that manufacturingplayed a somewhat different role in each.The management of between 350 to 400firms were interviewed in each country, sothat the overall results were based on asample of about 1,700 firms. Question-naires were designed to identify the pro-portion of time a plant had operated duringthe year of the survey and to provide in-formation on the determinants of capitalutilization.

The measurement of capital utilizationproved very complex. The low estimatesderived from earlier studies had beenbased on electricity utilization. As a result,actual utilization was consistently under-valued; in Israel and the Philippines in1972-73, for instance, average electricityutilization was 15 per cent and 18 per centrespectively, while average capital utiliza-tion (weighted by capital) was 43 per centand 61 per cent. Likewise, the number ofshifts worked turned out to be a poor indi-cator, as many plants extended their hoursof operation by working more than eighthours a shift and by weekend work, as wellas by multiple shifts. Manufacturers alsotended to operate the capital-intensivesections of their plants intensively—fre-quently at three shifts a day, seven days aweek—but operated the more labor-intensive sections on a one-shift basis. Thehigh level operations of the capital-intensive parts of a plant were usually notreported. This meant that capital utiliza-tion could not always be increased simplyby increasing the number of shifts workedin the rest of the plant. Increasing shiftwork usually also requires additionalinvestment.

The measure used in the study wastherefore the amount of time machineryand buildings were in use as a proportionof the time they could be used (the latterbeing 24 hours a day or 8,760 hours a year).

This article is based on a comparative study ofcapital utilization in Colombia, lsradt Malaysiaand the Philippines m Capital Utilization inManufacturing by Romeo M. Bautista, HelenHughes, David Vim, David Morawetz, andFrancisco Thomm (Oxford University Press forthe World Bank, 1981).

Finance & Development I March 19836

Capital utilization in manufacturing


Development, manufacturing, and capital utilizationIn Colombia, Israel, Malaysia, and

Population (in millions)GNP per capita (in U.S. dollars)Real GNP per capita growth

(in per cent)Value added in manufacturing as per-

centage of GNP (in current prices)Real growth in value added in

manufacturing (in per cent)Manufacturing exports as share of

total merchandise exports(in per cent)2

Capital utilization weighted bycapital (in per cent)

Years

19781978

1960-78

1978

1960-78

1977

1972-73

the Philippines, 1960-78

Colombia

25.6900.0

2.9

19.0

6.7

19.0

79.0

Israel

3.74,150.0

4.2

18.01

4.0

80.0

43.0

Malaysia

13.31,150.0

3.9

18.0

12.4

17.0

71.0

Philippines

45.6530.0

2.6

24.0

7.0

25.0

61.0

Source: Bautista et al, Capital Utilization in Manufacturing, p. 5.'Includes mining.Includes Standard International Trade Classification (SITC) categories 5 through 8, minus division 68.

According to this measure, for instance, ifcapital is used for one eight-hour shift aday, five days a week, fifty-two weeks ayear, it would be operating at 24 per cent ofcapacity.

The findings

Even allowing for the peak in economicactivity that was taking place in the selectedcountries in 1972-73 when the survey wascarried out, it was surprising to find thatutilization levels were substantially higherthan previous studies had indicated. Dur-ing the year covered by the study, capitalwas in use for 43 per cent of the time itcould be in Israel, 61 per cent in the Philip-pines, 71 per cent in Malaysia, and 79 percent in Colombia (see the table). Theseorders of magnitude were supported bysubsequent research in the other devel-oping countries. (These average figures, ofcourse, gloss over the considerable range inthe level of the capital utilization amongand within the various branches of manu-facturing, and, indeed, among individualfirms in the same sector of industry, andtherefore have to be used with greatcaution.)

The principal hypothesis confirmed bythe study was that the less than 100 percent levels of capital utilization were in-

tentional, even in the peaks of businesscycles, except for continuous processes.(Only in Israel did actual use fall below in-tended levels of utilization: this was be-cause labor shortages made second shiftsdifficult to mount.)

There seemed to be no simple correlationbetween the level of development and cap-ital utilization. The country with the high-est per capita income, Israel, had the lowestcapital utilization. Colombia, with the sec-ond lowest per capita income, had thehighest capital utilization. This confirmsthe findings of other studies and reflectsthe fact that capital utilization levels de-pend to a large degree on the appropri-ateness of economic policies. Capital utili-zation, moreover, should be viewed not inisolation from, but in relation to, the otherfactors of production, notably labor andmanagement. For example, it was foundthat high capital utilization was not neces-sarily optimal for development, as it oftenmeant unduly low employment. To achieverapid growth, it is total factor utilizationthat must be optimized.

Main determinants of utilization

Identifying the determinants of capitalutilization was tantamount to choosing apath through a minefield of theories, data

measurement difficulties, and econometricoptions. The determinants are so intricate,subtle, and overlapping that all types ofapplied economic analysis leave somethingto be desired. The initial decisions of a firmon products and production techniques,location, the intended level of utilization,and subsequent expansion are all inter-connected. The resulting levels of utiliza-tion largely reflect those planning deci-sions, but unexpected changes in demandand supply conditions can also affect them.

Nevertheless, despite the variations be-tween countries and between industrieswithin countries, some consistent deter-minants of the levels of utilization didemerge. They relate to the following:

• The capital-labor intensity of produc-tion;

• Economies of scale;• Fluctuations in the demand and sup-

ply of inputs or outputs;• Effects inadvertently caused by poli-

cies that have other objectives;• Forms of industrial organization; and• Infrastructural and noneconomic fac-

tors.Some of these will be discussed in detail.

The capital/labor ratio emerged as one ofthe two principal determinants of utiliza-tion in each country and in the cross-country regressions over the long run.Capital-intensive industries, plants, andparts of plants tended to operate relativelylong hours, implying that the larger thecapital cost, the more responsive manu-facturers were to the incentive to econo-mize. Labor-intensive processes, in con-trast, often had low capital utilization ratesto avoid the payment of the higher costs ofnight and weekend shifts. Since continu-ous process industries, which are usuallyalso highly capital intensive, have veryhigh stop-and-start costs, such industriesas steel and chemicals tended to be in oper-ation all the time. Of the 347 manufacturingenterprises surveyed in Colombia, for in-stance, the 8 plants that reported full capac-ity utilization were all producing industrialchemicals and petrochemicals, glass, iron,and steel. The plants with low utilization,on the other hand, tended to be relativelylabor-intensive export industries or pro-

Finance & Development /March 1983 7


ducers for a limited but highly protecteddomestic market.

The question arises on the reasons amanufacturer would have to choose a rela-tively more capital-intensive process in adeveloping country. The impact of relativecapital and labor costs on the choice ofproduct and production techniques provedto be critical in determining the level of cap-ital utilization. If capital was cheap relativeto labor, capital-intensive technologieswere likely to be chosen. Thus, in Israel,subsidized credit, low tariffs on importedmachinery, and the overvaluation of the ex-change rate caused the cost of capital goodsto be unduly low. This led to a shift towardmore capital-intensive techniques of pro-duction than would otherwise have beenthe case. Moreover, sectoral patterns ofmanufacturing investment were influencedby government allocations of financial re-sources to particular enterprises at low in-terest rates. Equally, a high price of capitalrelative to labor was expected to encourageexisting plants to expand by increasing thehours worked rather than by purchasingextra machines. The cost of borrowing tofinance capital expansion had to beweighed against the cost of hiring labor foradditional shifts.

The scale of production typically affectsutilization in two ways. Some productionprocesses are "indivisible"; they have, forengineering reasons, a single most efficientsize. Such indivisibilities may result in theunderutilization of some pieces of equip-ment, such as printing machinery thatmakes labels for a food canning plant; oncethe requirements for a given period havebeen met, the equipment becomes idle. Onthe other hand, a small plant cannot installonly half the smallest available steamboileror furnace to correspond to its need.

The existence of economies of scale thusimplies that smaller firms will generallyhave low levels of capital utilization. A pro-ducer may also decide to overbuild plantcapacity because the cost of built-in excesscapacity lying idle is less than the lack ofeconomies of scale. (This is especially likelyif there is some expectation of expandingproduction in the future.) Hence, large-scale plants will tend to utilize capital morefully than small-scale ones. The study con-firmed these assumptions.

The indivisibilities of entrepreneurshipand management were also important. Anowner-managed plant rarely operates morethan one shift (albeit a "stretched" one thatmay last 12 or even 14 hours) because hir-ing a supervisor for a second or third shiftwould raise costs disproportionately. Fur-ther, the owner, besides being plant super-visor, generally performs the functions offinance officer, sales manager, and so on;

and a night supervisor typically could notcarry out these daytime functions. A largerplant, on the contrary, is able to specializethe services of supervisors more, makingthe operation of second and third shifts rel-atively less expensive.

Daily or seasonal fluctuations in the sup-ply of inputs were also sources of capitalidleness. Bakeries geared to supplyingfresh bread early in the morning tended tobe idle the rest of the day. The strongestseasonal demand factor was noted in Col-ombia, where the Christmas buying sea-son, fortified by wage and salary bonuses,created a very high seasonal peak in De-cember. Off-peak utilization rates were cor-respondingly low. High short-term interestrates for working capital, compared withthose for fixed capital, limited off-seasonstockpiling. In agricultural processing, uti-lization levels depended on the length ofthe season and the perishability of the in-puts, and whether the plant had alternativeuses. In general, operations producingperishable goods had lower utilization thanothers. If output can only be stored for ashort time, producers will adjust utilizationto respond to short-term fluctuations in de-mand. In Israel, however, capacity utiliza-tion in plants producing perishable prod-ucts was 7 percentage points higher than inother plants. The reason was that the prod-ucts—newspapers, bread, meat, milk, andother food goods—were mainly producedat night to be ready for sale the next morn-ing. But the managers were able to find away to use the plant during the day as well,using cheaper labor.

Shift, weekend, and public holiday pre-miums represented rhythms in the work-ing day, week, and year. The higher theovertime and shift work premiums, thelower the level of capital utilization tendedto be. Problems were encountered in mea-suring wage premiums, but it seems thatthese premiums were less important as de-terrents to high capital utilization in devel-oping than in industrial countries.

Helen Hughesan Australian, studied atMelbourne University andat the London School ofEconomics. She was aneconomic consultant forAustralian mining andmanufacturing firms, andalso taught at theUniversities of New SouthWales and Queensland,

and at the Australian National University. She hasbeen at the World Bank since 1969 and is at presentDirector of the Economic Analysis and ProjectionsDepartment.

Although policies such as those favoringcapital intensity may be misguided, theireffects turn out largely as intended. How-ever, in some cases, policies also havewholly unintended impacts on capital utili-zation. Where "capacity in existence" wasused as a criterion for granting import orproduction licenses, for instance, it becameprofitable for entrepreneurs to invest in ex-cess manufacturing capacity so that amonopoly of inputs or production could beobtained. This was the case with flour mill-ing in Colombia, and similar experiencehas been observed in many other coun-tries, including India and Pakistan. In so-cial terms, the result was capital wastagebecause of large-scale planned capitalidleness.

A high level of capital utilization requiresworking capital to finance raw materials,wages, and so on. But in the repressed cap-ital markets that are common in many de-veloping countries, the cost of workingcapital is usually very high, much higherthan that of long-term capital for in-vestment. As a result, when their marketsexpand, manufacturers buy additionalequipment rather than mounting extrashifts, thus creating a bias in favor of alower utilization of capital. In the inter-views in the Philippines and Colombia, thisproblem was emphasized, and it was alsofound to exist in other Latin Americancountries.

In sum, the more distorted manpower,capital, and taxation policies are, the lowerthe level of capital utilization will be. Fur-ther, while trade policy distortions may notnecessarily lead to low capital utilization, itseems that low protection and high exportlevels do lead to a relatively high utilizationof the most plentiful factor of production inthese countries—labor.

Policy implicationsTheory offers no single conclusion that

can be drawn about the relationship be-tween the levels of capital utilization anddevelopment. Rather, the relationshipchanges over time. In the early stages ofindustrial development, when capital isscarce and labor ample, small-scale oper-ations, management difficulties, and lim-ited social infrastructure are likely to keepcapital utilization at low levels in all butthose processes in which technology dic-tates high capital intensity and continuousprocessing. As development progresses, itbrings improvements in management andincreases in the scale of production, partic-ularly in such infrastructural facilities astransport. In semi-industrial countries,capital utilization generally rises to peaklevels when production, trade, and em-ployment in manufacturing are growing

8 Finance & nevelrnmtmt I Mnrrh 1983


most rapidly. If appropriate policies arepursued, both labor and capital utilizationwill tend to be very high. In time, however,the trend toward high factor utilization willbe offset as workers' preferences change: asfull employment is approached, higherpremiums must be paid to compensatepeople for working extra shifts and week-ends. Capital utilization then drops.

Thus, while raising the utilization ofcapital and increasing employment areinterconnected and to some extent comple-mentary objectives, at low levels of devel-opment they may also be competitive. Inmost developing countries, the optimumcombination and utilization of both at suc-cessive levels of development requires con-siderable sophistication in policymaking.

When making policy with respect to utili-zation levels, it is also important to takeinto account social attitudes. Industrialcountries had a sharp, and in many in-stances permanent, increase in shift workduring World War II, mainly as a result ofexhortation to entrepreneurs and workersto back the war effort by round-the-clockshifts. In many developing countries itwould pay to promote higher capital utili-zation as a social goal by similar "jaw-boning." Still, exhortation is of limitedvalue if existing policies lead to price dis-tortions that make high utilization levelsunprofitable for private firms.

Several conclusions can be drawn on thebasis of this study regarding actual policies

being implemented in developing coun-tries. In many countries, the price of capitalto manufacturers is lower than its cost tosociety in terms of the lost employment op-portunities it represents. Reducing tariffson imported capital equipment, whiletariffs on finished goods remain high, en-courages the choice of capital-intensiveproducts and production techniques. Thisresults in relatively high levels of capitalutilization but can also lead to lower levelsof employment than might otherwise pre-vail. Subsidies to capital have a similar ef-fect, whether provided directly or throughreduced interest charges. Moving the priceof capital closer to its true social cost incountries with urgent employment needsencourages a more labor-intensive andsmaller-scale structure of manufacturing. Ifa scarcity of managerial skills and social fac-tors militates against shift work, the initialresult is usually increased employment butwith lower levels of capital utilization.However, in time, as management gainsexperience and the scale of production in-creases with economic development (andwith export growth), this trend is reversed.

Other considerations are-.• If the cost of working capital is high

relative to that of long-term borrowing, re-moving interest rate distortions increasescapital utilization.

• In those developing countries wherewage premiums for night shifts appear tobe too high to reflect workers'—or social—

priorities, reduced shift premiums will leadto higher employment.

• Tax policies may also be used to im-prove utilization levels. Corporate taxstructures tying depreciation rates to utili-zation levels may be a direct incentive aswell as a useful educational tool in in-creasing capital utilization.

• Direct controls generally have unfore-seen side effects and should be avoidedwhere possible. Controls that tie "capac-ity" to import quotas or production licensesshould particularly be avoided.

• Measures encouraging the develop-ment of management skills can be helpful.Clearly, where inadequate infrastructureinhibits shift work, improving public trans-port and safety will raise utilization levels.Finally, taking remedial measures in a co-herent framework aimed at increasing capi-tal utilization will be more effective thanundertaking them ad hoc.

It is evident that a wide range of macro-economic policies, together with direct con-trols, contributes to the price distortionsthat lead to low utilization. As such, thereis no single policy solution to the problem.While the changes in the particular policiesoutlined above are considered the principalones needed to improve capital and laborutilization, optimal utilization of the factorsof production will only result from a total-ity of "good" policies. Unfortunately, thismeans that optimal utilization is likely tocome about only slowly. ED

Changingyour

address?In order for us to

change your address,please return the

mailing label fromthis copy of

the magazine.If you cannot findthe mailing label,

you must send usboth your previous

address and yournew address.

Attach label from latest copy of

My new address is (please print):

NAME

ADDRESS

COUNTRY

Place in envelope, stamp, and mail to:Finance & DevelopmentAttention: PublicationsInternational Monetary FundWashington, DC 20431, U.S.A.

POSTAL CODE


VevJJ^L,

city

Changeof

addressform


Toward a more orderlyexchange rate systemWays of reducing fluctuationsinclude more stable andcredible domestic policies,greater intervention in foreignexchange markets, and, undercertain conditions, the formationof currency blocs.

Jacques R. Artus

There is, at present, a widespread feelingof dissatisfaction with the exchange ratesystem born out of the 1971-73 crisis, inparticular because of the large movementsof floating exchange rates. These move-ments have often exceeded 30 per cent overa period of two to three years, a variabilitymuch greater than the variability of therelative national price levels (see chart).Some contend that such extremely largedifferences between exchange rate move-ments and prices were justified—bychanges in international comparative ad-vantage, by the underlying determinants ofdomestic saving and investment, or simplyby the stickiness of domestic prices. But itis difficult to accept these views; structuralfactors cannot have changed so much, sofast, and so frequently, and exchange rateshave been much more variable than relativemoney stocks, which are not subject to anystickiness problem.

Reasons for this instability give groundsfor concern. The first reason is that an un-anticipated disturbance affecting the sup-ply or the demand for money tends to havea larger initial effect on the exchange ratethan is ultimately necessary to offset theeffect on relative national price levels. Thatis, the exchange rate "overshoots." An un-anticipated reduction in money growth, forexample, is followed by an initial periodduring which prices keep rising as fast asbefore because of built-in rigidities in thegoods and labor markets. The ratio of avail-able money to nominal income declinesand the real interest rate rises. The ex-change rate then has to appreciate to offsetthe favorable international differential inreal interest rates and the time these areexpected to persist.

When the overshooting related to theinterest-rate effect was first emphasized

(Dornbusch, 1976), it was thought to ac-count for small and brief changes in theexchange rate as monetary shocks weresupposed to have only small and brief ef-fects on real interest rates. Now, data onshort-term real interest rates suggest that amajor monetary contraction may causeshort-term real rates to rise by as much as5 percentage points and that the effect maylast for several years. As an illustration, a 5percentage point increase in the short-termreal interest rate expected to last for threeyears should cause an initial exchange-rateovershooting of 15 per cent.

A second reason for the instability is thatan unanticipated disturbance that affectsthe current account balance also tends tohave an initial effect on the exchange ratethat exceeds the ultimate change. For ex-ample, an adverse change in the terms oftrade will lead to a worsening of the currentaccount and a period of adjustment duringwhich private wealth is shifted from do-mestic residents to nonresidents. Thesenonresidents have to be compensated forthe exchange and political risks of holdingthe weakening currency and will bid downthe exchange rate to a level that is below theone expected to prevail in the longer run. Itis difficult to assess the size of the over-shooting arising from a current account im-balance, but there is evidence that even asmall international transfer of privatewealth may at times imply a significant de-preciation of the currency. This is particu-larly the case when the share of assets de-nominated in this currency in the portfoliosof nonresidents is small.

The third reason for longer-run insta-bility is that the future is highly uncertain.The size of overshooting from unantici-pated disturbances on the real interest rate,or on the current account, is related to theexpected duration of their effects, which, inpractice, are extremely difficult to forecast.Forecasts are therefore revised on the basisof current developments, and these re-visions lead to exchange rate movements.Such uncertainty also leads to indetermi-nate expected long-run equilibrium ex-change rates. Private market participantshave fairly limited information on what un-derlying economic conditions are going tobe a few years ahead and change theirviews frequently. This is particularly true atpresent, as it has been during most of the1970s, because practically all industrial

countries are engaged in efforts to reducestagflation, efforts whose ultimate successis highly uncertain. In addition, nationalauthorities are at times directly responsiblefor exchange rate swings because they leteconomic conditions and expectations de-teriorate gradually to a point where the ex-change rate has been pushed so far thatthey feel obliged to shift suddenly to a pol-icy of strict monetary restraint.

Nevertheless, over the past ten years, rel-ative gross national product deflators orrelative money stocks in the three mainindustrial countries—the United States,Japan, and the Federal Republic of Ger-many—have often followed relatively regu-lar trends with fluctuations of limitedamplitude. It is difficult to see why suchtrends would lead to wild gyrations in ex-pectations, unless political factors were asimportant as economic factors in affectingexchange rate expectations, in part becausethey affected the perceived riskiness of thephysical assets located in, or the financialassets issued by, the various countries.

These influences on long-run instabilityhave nothing to do with market inefficien-cies. Indeed, the view is that because ex-change rates reflect all available informa-tion, they often tend to overshoot and tovary considerably from year to year.Whether it should also be recognized that,at times, private market participants arevictims of speculative crazes or other"bandwagon forms" of behavior is morecontroversial. It seems that there is no sci-entific way to prove or to disprove the exis-tence of such forms of behavior. For exam-ple, one may find it difficult to believe thatthe change in the outlook for underlyingeconomic and political conditions in theUnited States and the Federal Republic ofGermany from January 1980 to August 1981was so large that it justified a 54 per centappreciation of the U.S. dollar vis-a-vis thedeutsche mark in real terms. But that is amatter of judgment. Possibly the informa-tion available in January 1980 suggestedthat the United States was on the verge ofrunaway inflation. Possibly the informa-tion available in August 1981 suggestedthat the situation in Eastern Europe was onthe verge of dealing a deadly blow to theGerman financial system. Neither expla-nation seems plausible, but it cannot beproved that they were inconsistent with theinformation available at the time.

10 Finance & Development I March 1983


Bilateral exchange rates, relative prices, money stocks, arid relative current

Germany vis-a-vis United States Japan vis-a-vis United States

account positions, 1973-811

Japan vis-a-vis Germany

Source: IMF, International Financial Statistics. Note: Indices 1973-81 = 100. 'The bilateral real exchange rate is calculated as the bilateral nominal exchange rate (periodaverage) adjusted for changes in GNP deflators expressed in local currencies. The data onrelative prices also refer to the GNP deflators. The data on money stocks refer to M2 (averageof end of month series). The relative export/import ratios refer to goods and services. All dataexcept for bilateral exchange rates are seasonally adjusted.

The costs of instability

The static welfare loss from exchangerates that are out of line with long-runneeds is probably small. From a welfarestandpoint, far more important are the ac-tual costs of moving resources. Given exist-ing rigidities in goods and labor markets,such shifts will lead to transitional unem-ployment of both labor and capital, andperhaps also to a permanent loss of re-sources as some workers are forced to gointo early retirement and some capitalequipment is scrapped. There may also beinflation when the industries that start toexpand bid for the necessary resources.These sectoral shifts, with the accompany-ing loss of resources, can be on a fairly largescale. Fluctuations in nominal and real ex-change rates of 30 per cent or more in twoor three years have been frequent. Suchfluctuations can have major effects on thestructure of production. (See the table forestimates of the effects of a 30 per centappreciation.)

Effects of a 30 per cent appreciation of the effective exchange rate1

(In per cent)

Export volumeSemi-finished manufacturesFinished manufacturesOther goods

Import volumeSemi-finished manufacturesFinished manufacturesOther goods

Terms of tradeVolume of production

Semi-finished manufacturesFinished manufacturesOther tradable goodsNontradable goods

PricesConsumption deflatorGNP deflator

France

-16.5-22.4-14.1

15.624.34.0

11.3

-10.3-7.2-3.2

3.2

-16.2-6.8

Germany

-13.7-17.8-9.1

15.424.74.9

10.4

-9.6-9.5-2.1

5.8

-15.2-6.5

UnitedKingdom

-7.2-12.5

-5.0

8.017.94.28.0

-10.1-7.6-1.1

3.3

-18.8-8.0

Japan

-23.0-31.1-23.5

20.133.33.3

15.0

-6.8-7.7-6.4

3.5

-9.6-4.1

UnitedStates

.I

-26.4-31.5-18.2

27.230.46.3

14.2 ji

-1.8-6.3-4.3

1.5

-10.1-4.6 !

'The estimates presented here are derived from the Fund's Multilateral Exchange Rate Model (see Artus and McGuirk(1981)). They refer to medium-term effects (three to four years) of an appreciation of the exchange rate over and above theexchange rate change corresponding to the differential in underlying rates of inflation. In the simulations used to derive theseestimates, the level of real GNP is constrained to remain unchanged in order to focus attention on the effects of theappreciation on the structure of the economy.

Finance & Development I March 1983 11


Another source of concern is that the un-certainty concerning longer-run changes inreal exchange rates that is generated by thetype of instability considered here may leadto a shift of resources away from the moreexposed traded-goods sectors. The conse-quences may be slower growth of foreigntrade, a weaker competitive climate, anddecreased incentives for growth and pro-ductivity. Such consequences are quitecomplex and difficult to measure. The mainproblem is that this uncertainty cannot beeasily hedged in the forward markets.Other forms of hedging have to be used,and some of them may be quite costly interms of resource allocation. Firms may, forexample, set up separate production unitsin each of their major national markets tominimize the risk of sudden movements inthe ratio between costs and product prices.They may also try to decrease their net riskposition by importing more from countriesto which they are exporting, or by export-ing more to the countries from which theyare importing.

As a result, the decrease in the volume ofinternational trade could be small, but themove away from a full exploitation of com-parative advantage could be quite signifi-cant. Similarly, even if the level of foreigndirect investment is sustained, resourceallocation may be impaired by excessivediversification of production units amongcountries. More relevant in the presentcontext is that, if investors are risk-averse,an increase in exchange rate uncertaintyshould, in itself, generate a decrease in theamount of savings transferred amongcountries. There could also be a fragmen-tation of the international capital markets,so that the comparative advantage of eachmarket over different types of transactionsand maturities could not be fully exploited.The econometric evidence available thus farcasts little, if any, light on the issues con-cerning the effects of exchange rate uncer-tainty, particularly on long-term interna-tional capital flows. Available surveys ofhow businesses view exchange rate devel-opments are also inconclusive on thesepoints.

A further cost of exchange rate instabilityis that it can make it more difficult for theauthorities to achieve stable domestic eco-nomic conditions. One reason is that, asdiscussed above, a disturbance in a majorcountry may push its exchange rate to over-shoot because of its effects on real interestrates and the current balance. The over-shooting will present authorities in tradingpartners with a dilemma: to accept thechanges in their exchange rates, with theiraccompanying effects on domestic costsand prices, or to change their monetary

policies. Both alternatives may be ex-tremely costly.

To be fair, exchange rate instability in aninflationary world has certain advantages.It favors authorities that are inclined to fol-low a policy of monetary restraint, becausethe appreciation of their exchange rates willhelp the domestic fight against inflation,and it similarly penalizes loose monetarypolicies. The problem is that there arecountries such as Japan, and to a lesser ex-tent Germany, that have already largely re-duced their domestic inflation—in partthanks to relatively well-functioning labormarkets. These countries do not appreciatethe choice given to them between exchangerate depreciation and abnormally high in-terest rates.

Policy remedies

Given the costs of instability in real ex-change rates, major changes in real ratesthat are not needed from the standpoint ofadjustment in the goods markets should beviewed with concern. The issue is how theycan be reduced without introducing equalor even greater costs.

It is a truism, but one that cannot be re-peated too often, that the important pre-requisite for more exchange rate stability isto have national authorities follow morestable, credible, and balanced domesticpolicies to deal with their inflation prob-lems. First, stable expectations are neces-sary for the floating exchange rate systemto function properly, and these depend onmarket participants having some reliableinformation on where exchange rates arelikely to be a few years in the future. Sec-ond, only stable and credible domestic pol-icies can lead to the gradual winding downof inflationary expectations that must takeplace if inflation is to be reduced.

Domestic policies must also be balanced.This does not only mean that a restrictivemonetary policy should be accompanied bya restrictive fiscal policy. The overshootingthat occurs with a program of monetary

Jacques R. Artusa French citizen, is Chief

of the External

Adjustment Division in

the Research Departmentof the Fund, which hejoined as an economist in

1969. Mr. Artus has a

BA from the University ofParis and a PhD from theUniversity of California at

Berkeley. He has published widely on economic andfinancial matters.

restraint in the main reflects the failure ofprices and nominal factor incomes to adjustquickly to changes in monetary policies, aswell as to changes in supply and demand inthe goods and labor markets. There areonly two ways to facilitate the decline ininflation and reduce overshooting: (1) torely on market forces, which assumes thatmost of the rigidities that limit the role ofthese forces are eliminated; or (2) to relyon incomes policy, which assumes that asuitable social and political framework isestablished.

National authorities should also keep aneye on the exchange rate in conductingtheir monetary policy. Technological andregulatory developments in financial mar-kets are currently affecting the meaning ofthe various monetary aggregates in a num-ber of countries. In addition, in countriessuch as the United Kingdom where resi-dents are free to substitute assets denomi-nated in foreign currencies for domesticmoney holdings, shifts in expectationsabout exchange rates or relative interestrates can affect the normal relationship be-tween the monetary aggregates and do-mestic nominal demand. Even in othercountries, this normal relationship can beaffected by changes in foreign demand forthe local currency. Thus, seeking a con-stant rate of growth of certain monetaryaggregates may, at times, destabilize thereal economy and the exchange rate. Onsuch occasions, movements in the ex-change rate can be viewed as a useful in-dicator of shifts in money demand thatshould be offset by shifts in domesticmoney supply. (The interest rate can alsobe used as an indicator to detect suchshifts, but there is a risk that a change in thenominal interest rate may reflect a changein inflationary expectations rather than ashift in money demand. This can lead tocostly policy mistakes; in particular, an ex-pansion of domestic money supply mayfeed the rise in inflationary expectations ata time when they should be subdued by acontraction.)

However, the use of the exchange rate asan additional indicator of monetary policyrequires careful judgment. Exchange ratemovements that are needed to offset in-flation rate differentials and other changesin underlying economic conditions do notcall for changes in domestic money supply.Similarly, the authorities may rightly hesi-tate to change domestic money supplywhen the instability of the exchange ratereflects the instability of monetary policiesabroad. Finally, even if a convincing casecan be made that an exchange rate appre-ciation reflects an increase in money de-mand, the authorities must assess whetherallowing a sustained expansion in domestic

12 Finance & Development / March 1983


money supply will not be misinterpretedby private market participants as an indi-cation that the monetary authorities aregiving up the fight against inflation.

There are cases where exchange rate in-stability results from a difference of inter-ests between countries. For example, therecan be a difference of interests, at least inthe short run, between countries wheredeeply embedded wage rigidities make itnecessary to maintain high real interestrates for a sustained period in order toreduce inflation, and countries where,thanks to more flexible nominal wages,there is no such need. Some reconciliationmust take place, and some exchange rateinstability may be unavoidable in the shortrun to achieve a reduction of inflation incountries with inflexible nominal wages.But, in the longer run, there is no doubtthat the domestic and international objec-tives would best be reconciled by a gradualelimination of wage rigidities.

Intervention and currency blocsAlthough more stability could be intro-

duced into the system through more effi-cient domestic policies, it would be unreal-istic to expect that such a change wouldcompletely solve the problem of exchangerate instability. In any case, progress onthis front will be difficult and slow. Bothintervening in foreign markets and formingcurrency blocs could help control exchangerates, but only under certain conditionsthat are not easy to satisfy in practice.

Intervention can be financed fromowned reserves, use of swap facilities, orborrowing from private international mar-kets. Directives or incentives given to stateagencies or private banks to lend or borrowabroad are also forms of intervention. (It isassumed here that the effects of interven-tion on high powered money are being off-set to prevent domestic impact.)

First, it must be recognized that coun-tries differ substantially as to how much theauthorities can rely on intervention to con-trol their exchange rate. There is a wholespectrum, ranging from countries that haveprimitive domestic trade and financial mar-kets completely isolated from world mar-kets that can keep the exchange rate fixedfor many years by direct controls (althougha black market may develop), to countrieswith large and highly developed domesticmarkets fully integrated with world mar-kets, where intervention is unlikely to havea sustained effect.

Recent experience also suggests that theeffectiveness of intervention depends onwhether it has specific objectives (either aspecific rate or a lower or upper limit) overthe short run, or aims only to influence theevolution of rates that are basically floating.

In the first case, members of the EuropeanMonetary System (EMS) have demon-strated that countries can maintain ex-change rates within narrow bands for sub-stantial periods of time, even though theyhave widely different domestic policies andinflation rates. In the second case, the ex-periences of countries such as Germany(with respect to the deutsche mark/dollarrate), Japan, and the United Kingdom sug-gest that intervention as part of a managedfloat tends to have a limited effect. Thereare two main reasons for this difference ineffectiveness. First, a major factor influenc-ing the exchange rate is the expectation ofa significant and persistent change in theshares of domestic and foreign securities inprivate portfolios at home and abroad. Ifthere is no clear intervention policy, inter-vention on any given day is unlikely to giverise to such an expectation. Second, if theauthorities have a specific exchange ratecommitment, and this commitment is cred-ible because it is realistic, private specu-lators will greatly enhance the effectivenessof intervention by moving funds in thesame direction.

This evidence would suggest that an ef-fective way to avoid the instability of afloating rate system would be to use someform of flexible pegging. This is what prac-tically all developing countries, as well asthe countries that participate in the ex-change rate arrangements of the EMS, havedone. But the major risk is that it becomesa system that postpones needed exchangerate changes. A further problem is that, inthe longer run, currency blocs such as theEMS can work well only if member coun-tries are able to adjust their rates of growthof labor costs to levels that are consistentwith the rate in the least inflation-pronecountry. This requires countries to be will-ing to adopt financial policies that fosterthe achievement of this target within a fewyears. It also requires a certain flexibility ofnominal wage rates if restrictive financialpolicies are not to lead to a vicious circle ofprofit squeeze, low investment, and de-clines in productivity calling for reductionsin real wage rates.

There are, therefore, a number of ap-proaches that can be used to reduce ex-change rate instability. By far the mostpromising is to move toward more stable,credible, and balanced domestic policies.Given the present inflationary conditions,of particular importance would be policymeasures that would reduce rigidities in la-bor markets in order to enhance the effectof monetary restraint on the growth ofnominal wages. This would shorten the pe-riod of high real interest rates and reducethe overshooting of the real exchange rate.However, progress in this direction is likely

to be slow. Another approach is to relymore heavily on official intervention. Nodoubt, there are clear limits to the effec-tiveness of intervention, as well as cleardangers, but the usefulness of interventionunder certain circumstances should not bedismissed. In this context, regional mon-etary blocs could contribute to a more or-derly global system, but only if membercountries are able to adjust their rates ofgrowth of labor costs to the rate in the leastinflation-prone country. Finally, a subjectnot considered here, coordination of re-serve policies among important centralbanks, will have to continue in order toprevent potentially destabilizing changesin the composition of their portfolios of for-eign currencies. ED

This article is based on a paper presented at aconference on the Evolving International Mone-tary Arrangements, at Wingspread in Racine,Wisconsin U.S.A. in July 1982.

Related reading

Victor Argy, Exchange Rate Management inTheory and Practice, Princeton Studies inInternational Finance, No. 50 (October1982).

Jacques R. Artus, "Effects of U.S. MonetaryRestraint on the DM-$ Exchange Rateand the Gentian Economy," Working Pa-per No. 926, National Bureau of Eco-nomic Research (July 1982).

Jacques R. Artus and Anne K. McGuirk, "ARevised Version of the Multilateral Ex-change Rate Model," IMF Staff Papers,(June 1981).

Michael P. Dooley, "An Analysis of Ex-change Market Intervention of Industrialand Developing Countries," IMF StaffPapers (forthcoming).

Rudiger Dornbusch, "Expectations and Ex-change Rate Dynamics," Journal of Politi-cal Economy (December 1976).

Group of Thirty, foreign Exchange MarketsUnder Floating Rates (Group of Thirty,New York, 1980).

Ronald I. McKinnon, "Floating Foreign Ex-change Rates 1973-74: The Emperor'sNew Clothes," in Institutional Arrange-ments and the Inflation Problem, edited byKarl Brunner and Allan H. Meltzer,Carnegie-Rochester Conference on Pub-lic Policy, Vol. 3 (Amsterdam, 1976).

Michael Mussa, The Role of Official Interven-tion, Occasional Paper No. 6 (Group ofThirty, New York, 1981).



DevaluationIn developing countries:the difficult choices

Devaluation is not painless; it isoften better than thealternatives and it may be theonly way to reduce distortionsand restore profitability toexports.

Karim Nashashibi

A country's current account position maydeteriorate because of expansionary finan-cial policies, a deterioration in the terms oftrade, price distortions, higher debt ser-vicing, or—and this has been increasinglythe case—a combination of these factors.On an ex post basis, external receipts andpayments always balance: a current ac-count deficit can only be sustained if thereare capital inflows to match it, unless thecountry can draw down reserves. Other-wise, adjustment must take place. It mayhappen over a number of years by appro-priate policies to restructure the economyand minimize social and economic dis-locations, or it may occur through a forcedreduction in imports by direct controls anda buildup of arrears on external payments.

To facilitate adjustment, many countriesseek balance of payments support fromoutside sources (including the Fund) anddebt relief from creditors in the frameworkof a planned adjustment process. Exchangerate depreciation is frequently an essentialpart of this. Yet it is fraught with contro-versy, particularly its role in Fund-supported programs. It has been associatedwith reduced economic growth and in-creased unemployment, and, in countrieswith an active public sector, it has beenlinked to undermining public sector invest-ment and development strategies. The risein the cost of living and the redistribution

of income that a devaluation entails havealso frequently been criticized for shiftingthe burden of adjustment to the lower-income groups who are least able to affordit. A final criticism has been that deval-uation is often ineffective, in that it can in-crease inflation without sufficiently im-proving the external payments position,which eventually induces another deval-uation and plunges the country into aninflation-cum-devaluation spiral.

Underlying factors

As already indicated, countries face bal-ance of payments problems for many rea-sons. One may be a deterioration in theterms of trade. In principle, this should notbe a convincing argument for a devalua-tion. (For instance, the deteriorating termsof trade experienced by many developingcountries during the past few years couldnot have been reversed by devaluation.) Ifexport prices decline—say, in response to arecession in the industrial countries- de-veloping countries could simply await thecyclical turnaround and the improvementin prices. If, however, the decline in exportprices is secular (for instance, it may reflecta long-term decline in market share due tothe substitution of some export such asrubber, jute, or other textile fibers by man-made products), it would be appropriate tofind some other way of alleviating the so-cial burdens of obsolescence while pro-moting the diversification of the exportbase by encouraging the production ofother export commodities.

In practice many developing countriesmay not have the external reserves or theborrowing capacity to await a cyclical up-turn nor the financial or human resourcesto bring about sufficient export diversifica-tion within a reasonable period of time. Itmay be easier—and quicker—to improvethe profitability of exports by a devaluationof the domestic currency. When export

prices decline while the exchange rate re-mains unchanged, major export activitiesbecome unprofitable in terms of the domes-tic currency. A devaluation restores profit-ability without reducing foreign exchangereserves or public sector savings. Other ex-ports whose world prices have not fallenwill reap a windfall gain from devaluation,which will stimulate their expansion.

A more convincing case can be made fordevaluation when currency overvaluationis primarily caused by a rapid rise in thedomestic cost of production resulting fromexcess domestic demand rather than by adecline in export prices. A typical exampleis the small developing country with a fixedexchange rate that does not influenceworld prices either of its exports or its im-ports and whose capital markets are rudi-mentary. Inflationary pressures usuallyarise from internal political expectations forexpenditures on social programs, develop-ment projects, or defense, which are higherthan can be financed from taxation and vol-untary private savings. To finance this in-crease in expenditures, the governmentwill need to borrow from the banking sys-tem. Excessive credit may also have beenextended to the private sector, which findsit difficult to raise equity in the financialmarket.

This expansionary monetary policy willincrease incomes and aggregate demand,raising the prices of the "home" or non-traded goods—that is, of subsistence agri-cultural products, various services, or con-struction—and the prices of factors ofproduction which use them intensively.Given an inflationary environment, andthe resulting fall in real interest rates, sav-ings will be diverted into real estate, furtherbidding up wages and land values. Theprices of tradables (import substitutes andexports) that are set in world markets willbecome cheaper relative to home goods,prompting domestic consumers to switch



to them. Domestic consumption of trad-ables will therefore increase with incomesbecause of higher aggregate demand andrelative price effects.

Meanwhile, the production of tradableswill be discouraged as the rise in the pricesof domestic inputs and factors of produc-tion spreads from the home goods sector,increasing costs and squeezing profits,since, as already indicated, prices of trad-ables are fixed by international markets. In-centives to produce them will diminish,marginal producers will drop out, and therate of production will level off or decline,exacerbating the external deficit and caus-ing a drop in reserves. Concurrently, as aresult of high domestic inflation, real inter-est rates may become negative as institu-tional factors often prevent the rise in nom-inal rates, discouraging savings in domesticcurrency, and encouraging conversion intoforeign currency, which often offers thedual advantage of an anticipated appreci-ation vis-a-vis the domestic currency andhigher real interest rates.

Alternative approaches

Under the circumstances outlined above,the authorities have two alternatives tomeet their external imbalance: they may re-sort to restrictions on imports and ex-change transactions or they may reduce thedeficit through financial and price policies,by cutting expenditures and restoring theprofitability of producing tradables.

A politically favored route, mainly be-cause its real costs only emerge after a sub-stantial lag, is to reduce the trade deficitdirectly, by imposing restrictions on lessessential imports and on foreign exchangemovements. While these measures may re-duce the external deficit temporarily, theydo not restore profitability to exports—infact they will eventually contribute to theirfurther decline. Nor do they reduce domes-tic expenditures to a level that the countrycan afford. With the underlying inflationunaffected by controls—indeed, frequentlyexacerbated by them—production will shiftto the more profitable home goods sectorand to import substitutes that now benefitfrom higher domestic prices due to the im-position of import licensing and quotas.

However, since efficient domestic substi-tution for imports is often limited in manydeveloping countries because of a lack oftechnological know-how or an absence ofeconomies of scale, these initial controlswill not achieve significant savings in for-eign exchange. Shortages, too, will developin imported inputs, which will eventuallyaffect output directly, reducing capacityutilization in the traded goods sectors, andresulting in unemployment. If domesticprice controls are imposed on the prices of

some import substitutes or commoditiescontaining imported inputs, the allocationof resources will be further distorted. Theblack market for foreign exchange will ex-pand as official and private rates divergeand capital is sent abroad, exacerbatingthe external payments deficit. The econ-omy will then become more inward-looking as farmers move from cash crops tosubsistence agriculture and foreign tradedeclines.

Restrictions on imports and capital canbe particularly harmful to efficiency; theytend to shift resources to areas with highdomestic profitability (such as import sub-stitutes) and away from more efficient ex-ports whose profitability is domesticallyrepressed by inflation and an overvalueddomestic currency. By undervaluing im-ports they also tend to bias investmenttoward capital-intensive techniques. Even-tually, special exchange rates may be pro-vided for expatriate remittances, tourism,and nontraditional exports—betraying therecognition that price incentives areneeded to stimulate foreign exchange earn-ings.

The alternative approach would bethrough financial policies designed to re-duce the demand for goods and services toa level consistent with the country's in-come, while promoting a shift in domesticdemand from tradables to nontradablesand providing producers with incentivesfor expansion. These objectives could beachieved by demand management alone.Government spending and credit to the pri-vate sector can be cut back to an extentwhere there is an absolute reduction inwages and prices. This would lower thecost of producing tradables and improvetheir profitability relative to home goods asprices of home goods fall with the declinein demand while prices of tradables remainat the same level. However, if relativeprices need to be adjusted significantly, adrastic reduction in aggregate demand maybe necessary. Since nominal wages are of-ten sticky downward, this reduction couldresult in substantial unemployment and so-cial dislocation. Therefore, this course islikely to be slow, painful, and politicallyunacceptable to most governments.

But rather than induce an increase in theprofitability of producing tradables bypushing down their costs through painfuland time-consuming demand managementpolicies, a country can raise the profitabilityof tradables by raising their domestic pricethrough devaluation.

How devaluation works

The obvious advantage of devaluation isthat it is easier and quicker than demandmanagement alone. While the costs of

traded inputs will rise by the full percent-age of the devaluation, costs of domesticinputs should rise by less, and profits—theresidual—should rise proportionally more.The change in relative prices will also in-duce exporters to search for cheaper do-mestic substitutes for imported inputs,which will further reduce the demand forimports and strengthen the profitability ofdomestic production.

For a devaluation to shift enough re-sources from home goods to tradables toimprove the balance of payments, thechange in relative prices must be perceivedas lasting. Domestic costs should not riseenough to erode the newly restored profit-ability of tradables—yet the increased do-mestic demand for home goods coupledwith the rise in the production of tradableswill tend to put upward pressure on do-mestic costs. But these pressures can bereduced since devaluation also contractsaggregate demand, at least in the shortterm. The rise in the prices of traded goodsentailed by devaluation reduces the realvalue of cash balances that people are ac-customed to hold—so, in effect, a deval-uation amounts to a reduction in the realmoney supply. To restore the real value oftheir balances, people spend less, and de-mand falls for all goods, including trad-ables. This contraction is essential to rectifythe initial imbalance resulting from an ex-cess of domestic expenditure over output.If aggregate demand is kept in check, therewill be no reason for the price level toincrease beyond the initial rise from thehigher prices of tradables.

But behind every devaluation looms thedanger of inflationary pressures for higherincomes, which will raise costs and in-crease the likelihood of monetary expan-sion to accommodate the higher demand.The result of monetary expansion will be ajump in prices, fueling inflationary expec-tations and the demand for foreign cur-rency—both as a hedge against inflationand in anticipation of a further devaluation.Apart from the negative impact this willhave on the external payments position, itwill set in motion another devaluation anda new jump in prices. Such an inflation-devaluation spiral can be very damaging,not only because it fails to correct the dis-equilibrium but also because it underminesthe credibility of exchange rate depreci-ation as a policy instrument. Whether thishappens or not depends to a large extenton the government's determination to pre-serve the new structure of relative pricesthrough adequate demand managementpolicies. Thus, while the sole reliance ondemand management for correcting an ex-ternal imbalance may be long drawn outand costly in terms of unemployment and

Finance fr Development I March 1983 15


output, reliance on devaluation alonewould be ineffective. The answer lies ina mix of policies whereby expenditure-dampening measures preserve the changein relative prices induced by a devaluation.

Impact on output

The effects of a devaluation on output arecomplex. In the short run, before exportshave expanded, there will be a drop in do-mestic demand for tradables that helps thebalance of payments. But the drop in de-mand—from lower real incomes—for homegoods will be accompanied by an increasein demand for them, since they are nowcheaper than tradables. If the devaluationis large and if there is little substitution be-tween tradables and home goods as there isin most developing countries, there will bea net reduction in demand, since theamount of money available for home goodswill fall. If consumers are able to substitutedomestic goods readily for tradables, a netrise in demand for them will tend to in-crease their prices and bid up wages, whichwill spread to tradables and reduce theirrelative price advantage. Additional mea-sures may then be needed to dampen do-mestic expenditures.

Moreover, for the production of trad-ables to increase under conditions of fullemployment, fiscal and monetary policieswill also have to be stringent enough todepress demand and reduce the output ofhome goods. In the short run and due tothe lagged export response, this will resultin a decline in total output, some unem-ployment, and a fall in real wages. As theoutput of tradables expands, unemployedlabor will be gradually absorbed but at alower real wage and a lower income sharethan previously.

But few developing countries have,strictly speaking, full employment. If be-fore devaluation there existed quantitativerestrictions on imports and price controls,there may even have been substantial ex-cess capacity and unemployment. Even ifthis were not the case, most developingcountries have ample labor resources in"disguised unemployment"—in low pro-ductivity or seasonal jobs, and withinhousehold units. If nominal wages in-crease, exports could expand by drawingon these underemployed resources. Deval-uation will then shift resources to sectorswith higher labor productivity and pro-mote growth, without contracting the out-put of home goods, which will then merelyhave a lower share of national income. Sim-ilarly, to restore the profits of tradableproduction to their normal level, theirshare in the export revenue will have torise. While this implies that the share ofreal wages will need to decline, nominal

wages could still increase. Such a chain ofevents will make devaluation much morepalatable, since growth in output and em-ployment will make up for the redis-tribution of income shares.

The success of a devaluation in develop-ing countries will essentially be measuredby its ability to stimulate the growth of ex-ports. If imports had been scarce before-hand, devaluation may not result in the ex-pected shrinkage of imports in response tohigher prices, particularly when domesticsubstitution possibilities are limited. Moreexports, too, may depend on a higher in-flow of imported raw materials as well ascapital equipment and spare parts to reha-bilitate previously neglected plant andequipment. Consequently, in the shortrun, the current account balance may verywell deteriorate following a devaluation asthe import volume will increase imme-diately, while export expansion will occuronly after a substantial lag.

Will exports rise?Depending on the exports produced, the

response to a devaluation will occur withvarying time lags. The initial increase inproductivity in response to better incen-tives can often occur within a year of thedevaluation. If the exports are agriculturalproducts, even if a devaluation takes placewell into the crop cycle, improved crophusbandry and greater efforts in harvestingcan result in significantly higher yieldsquite quickly. It may be possible to bringmarginal land easily into production andthere may be excess capacity in the export-servicing facilities; in these situations a fur-ther expansion in output can take placewithin two years. New investment will alsoresult in export growth but may take con-siderably longer; the installation of newmanufacturing capacity may have a longgestation period, and producers may needtime to regain confidence after the policychanges. These lags, coupled with infra-structural bottlenecks and rigidities oftenencountered in the transfer of resourcesacross sectors, may imply an adjustmentperiod stretching over several years.

Experience with devaluation in many de-veloping countries, particularly exportersof manufactures, has shown that the vol-ume of exports is generally quite respon-sive to price changes. However, with thecontraction in world trade that has takenplace over the past two years, and the risein protectionism in industrial countries,policymakers must carefully evaluate thepotential for additional exports. Further-more, some agricultural and primary com-modities for which growth in world de-mand is slow or stagnating can onlyincrease if the market share of another ex-

porter declines. But once a market share islost because of inflationary policies, a de-valuation can help a country recoup it.Moreover, should it want to diversify itsexport base away from slow-growing ex-ports, a devaluation will also provide theincentives and the profit opportunities innontraditional export areas with bettergrowth possibilities.

Real wage; income distribution

In the short run, and if only as a result ofthe rise in the prices of tradables, the realwage will fall in all activities after a deval-uation. And the larger the devaluation, thelarger the decline in the wage earner's stan-dard of living. This is its major political lia-bility. But given the internal imbalance be-tween output and domestic expendituresthat preceded the devaluation, the existinglevel of real wages could not have been sus-tained anyway. Nor is there any alternativeto a decline in the real wage if internal bal-ance is to be restored; although, as men-Honed earlier, it could be achieved moregradually but with higher unemploymentthrough deflation. Yet because devaluationentails sudden price changes and a fall inpurchasing power, it does capture morepublic visibility and understandably gener-ates more political opposition.

How the real wage will be affected as theadjustment process unfolds depends onthe proportion of tradables consumed bywage earners and their readiness and abil-ity to substitute home goods. The extent towhich labor succeeds in raising its nominalwages depends on its degree of organiza-tion and relative bargaining power. Al-though these factors are strong in some de-veloping countries, they tend to be weak inmost low-income developing countries andit may take considerable time before thereal wage is restored. But toward the end ofthe adjustment process there can be an in-crease in the real wage without adverselyaffecting the balance of payments, pro-vided output has increased by more.

Beyond the general principle that a de-valuation would tend to increase the rela-tive share of profits in national income andreduce the relative share of wages, the im-pact of a devaluation on income distribu-tion can vary widely according to the insti-tutional characteristics of each country andthe supportive measures taken by the gov-ernment. It is worth emphasizing that thepredevaluation income distribution couldnot, in any case, have been sustained. Ingeneral, if the export sector is primarily ag-ricultural, devaluation will shift incomesfrom urban to rural areas, as the demandfor urban services will tend to decline andurban dwellers will be more affected by theprice increases of tradables. In addition,



government expenditures tend to concen-trate on urban areas, not only in sustainingthe civil service but also in providing theurban population with educational andhealth facilities and sometimes food sub-sidies. Since the real value of these ex-penditures will fall, the beneficiaries of allthese services will experience a reductionin their real income.

Even if incomes move from urban torural areas, it does not necessarily implythat they will move from the poor to therich. It will imply this if export crops aregrown on large agricultural estates whilemost of the rural population is engaged inproducing nontraded food crops; but ifmost of the rural population is actively en-gaged in producing export crops, they willall benefit from a devaluation. Moreover,the ultimate effect on incomes will be deter-mined by the policies established to sup-port the devaluation.

Supporting policies

As already indicated, how successfullydevaluation is translated into an expansionof exports and import substitutes dependsto a large extent on the supporting mea-sures taken by the government. These fallinto three categories: price policies, de-mand management measures, and policiesthat transform an initial increase in supplyinto an actual increase in exports.

From the outset, a devaluation is a signalthat greater reliance will be placed on theprice mechanism. Hence, price policiesmust ensure that the change in the ex-change rate is fully reflected in the pricestructure of the economy. If the economyhas had price controls, quantitative restric-tions on imports, and some reliance on ad-ministrative decisions in production, de-valuation will have to be accompanied bypolicy measures consistent with its objec-tives. Most prices will need to be adjusted,restrictions on imported inputs for trad-ables will need to be relaxed, and cost andincentive distortions discouraging the pro-duction of tradables will have to be elimi-nated. In state enterprises, price changesbrought about by the devaluation will notin themselves result in the desired supplyresponse and reallocation of resources. Forinstance, the cropping patterns in statefarms, which may be administratively de-termined, will have to be directed toward aconfiguration that best reflects the profit-ability of alternative crops according to thenew relative export prices and domesticcosts.

It was stressed earlier that the relativeprice and profit advantage of tradables canonly be maintained if demand for homegoods is kept in check. This would signalthe government's firm intention to pursue

an anti-inflationary policy—which is im-portant to reduce inflationary expectationsand the destabilizing anticipation of a sec-ond devaluation. While supportive budget-ary policy should reduce consumption, thequestion is who should bear the cuts and inwhat proportion: the private or the publicsectors, the low income or the high incomegroups, the rural or the urban sectors?

In many developing countries the basicneeds of the poor are partly met by budget-ary expenditures. It is often claimed thatdevaluation taxes the poor in three ways:by reducing the real wage, increasing un-employment, and reducing governmentservices and entitlements. A reduction inthe real incomes of the poor may not betolerable and government expenditures onhealth, education, and food subsidies mayhave to be maintained in real terms. In ad-dition, devaluation will cause other govern-ment expenditures to rise, as the cost oftheir import content and that of debt ser-vice go up, requiring larger increases inrevenues to reduce the budgetary deficit.While devaluation will also increase gov-ernment revenues—through higher taxrevenues on trade and incomes and higherprofits from public enterprises—the impacton expenditures is more immediate andwill often tend to be larger. If higher profitsin tradables do not lead to higher privateinvestment because of lack of confidence oruncertainty, the government will need toencourage it; this, in itself, may limit thegovernment's options in reducing privatesector consumption. The alternative to re-ducing consumption is to cut back invest-ment that would reduce future growth.Either way these would be unpopular deci-sions for any government.

Given these narrow options, the govern-ment in most developing countries has totread a thin line between a credible adjust-ment effort necessary for devaluation to beeffective and the stress it imposes on the

Karim Nashashibia Jordanian citizen, is anAdvisor in the MiddleEastern Department of theFund. He studied at .universities in France andthe United States andholds a PhD in economics

from ^e Un*versity °fCalifornia at Berkeley. Hehas published articles in

the field of international trade and exchange ratepolicy and is the co-author (with Bent Hansen) ofForeign Trade Regimes and EconomicDevelopment: Egypt (National Bureau ofEconomic Research, 1975).

social fabric. This stress can be eased, forinstance, by attempting to raise taxes paidby upper-income groups, by reducing gov-ernment expenditures where they leastaffect the basic needs of the poor, or bytightening credit for housing and consumerdurables. These efforts will, however, beresisted by the higher-income groups whousually wield most of the political power.

Nevertheless, the government has somechoice. In setting the level of the devalua-tion, it may devalue further than necessaryto restore both export profitability and al-low for some rise of wages. But such acourse will be more inflationary and maycause destabilizing expectations. Alterna-tively, the authorities may opt for a smallerdevaluation than necessary, coupled withstricter demand management measures.Both alternatives can be combined withsome restrictions on less essential imports.In any case, a specific rate of currency de-preciation should be measured in its effectsin conjunction with the entire mix of poli-cies chosen by the authorities.

The government will also need to pursuea concerted policy in a number of areas toensure that foreign exchange earnings andsavings do increase. Development expen-ditures may need to be redirected to easeexisting bottlenecks, such as often arise inelectricity generation and transport andcommunication. The capital stock andexport-servicing facilities may also have tobe rehabilitated to facilitate channelingmore exports to foreign markets. An activepolicy of developing commercial channelsabroad will also be required to generatehigher demand.

Such a program geared toward the trans-formation of the productive structure of theeconomy and alleviation of sectoral imbal-ances will require a sustained effort andpossibly some increase in development ex-penditures. Moreover, because of the lag inthe higher volume of exports that may re-sult in a short-term deterioration in thetrade account of the balance of payments, itmay be—as officials in developing coun-tries often maintain—that a devaluationcan only be effective if it is initially backedby substantial foreign exchange resources.Apart from helping to support the newrate of exchange, such resources can alsoease the initial reduction in real income ofthe adjustment process and shorten itsduration.

For many developing countries experi-encing a balance of payments problem, fur-ther commercial borrowing from abroadmay not be sustainable, and concessionalfinancial assistance (including, in somecases, debt relief) could provide the re-sources necessary to tide them over untilexport growth resumes. HD

Finance & Development /March 1983 17


Improving the quality of education

The experience of World Bankprojects shows that betterquality teaching and teachingtools are crucial.

Stephen P. Heyneman

The contribution of learning to economicgrowth has long been recognized, althoughthe precise measurement of its effects isstill subject to debate. Parental demand foreducational opportunity, coupled with theeconomic and political incentives for au-thorities to supply it, account for the ex-pansion of formal schooling to unprece-dented levels. Universal primary schoolenrollment has now been achieved in 35 ofthe richer developing countries since WorldWar II, including Argentina, Gabon, Ma-laysia, and Trinidad and Tobago (Table 1).In the 36 poorest countries (whose percapita gross national product (GNP) wasUS$265 a year or less in 1975) average en-rollment in all levels of primary educationincreased from 48 per cent of the school-age population in 1960 to 70 per cent in1977. There are, moreover, about 50 percent more children enrolled in grade onethan in grade four, so this 70 per cent en-rollment figure significantly understatesthe proportion of children who beginschool. By the end of this century, everychild will probably be able to start school.

The quality of education received bythese children is the subject of this article.Typically, primary school teachers in devel-oping countries have few teaching tools,and even these are of poor quality. Theirown salaries absorb most available funds.Not surprisingly, despite common objec-tives in teaching mathematics, reading,and science, their pupils leave school withfar fewer skills than do their counterpartsin the developed world, who are exposedto schools of substantially higher quality.And yet the Bank's experience shows thateven poor education in these basic skillsmakes for better farmers, better adminis-

trators, and better mothers with healthierchildren. Analysis that attempts to isolatethose factors that have the most effect on apupil's achievement in developing coun-tries shows conclusively that better qualityteaching and teaching tools—particularlymore and better textbooks—have a sub-stantial effect. As a result, there has been ashift in emphasis within educational in-vestments.

Poor resources

Perhaps the most serious oversight of ed-ucational planners during the 1960s was tobypass the issue of how well schools wereteaching skills. This is no longer possible.Schools are asked to transfer an increasingnumber of cognitive skills and amount offactual knowledge. The mathematics andscience taught to primary school studentstoday were, ten years ago, often expectedonly of secondary school students. Simi-larly, secondary school students are taughtnow what before was taught in universityor, in some cases, not at all because thefield is new.

Yet in spite of the significant advancesmade in providing places in primaryschools, these schools in the developing

world generally have poor resources andtherefore pass on less effectively than theycould the increasingly complex skillsrequired of school leavers today. The edu-cation of teachers and the availability of fur-niture, equipment, and materials are nor-mally well below the standards consideredminimal for schools in industrial societies.

In 1977, for instance, there were tenpupils for each available primary schooltextbook in the Philippines. In Bolivia, in1978, the monetary value invested annuallyin furniture and materials in the averagefourth-grade classroom was approximately80 U.S. cents a pupil, one sixtieth of theinvestment for each pupil in Maryland(U.S.A.) during the same year. In a surveyconducted in Malawi in 1979, 1 pupil in 8was found to have a chair, and only 1 in 88a desk. Primary schools were withoutsafety standards. Walls frequently col-lapsed after a rain; roofs had large holes;wind and storms disrupted classroom ac-tivity as a matter of course. The normalclassroom was dark and stuffy; students saton the ground, balancing an exercise bookor slate on their knees to write. Teachershad no offices, chairs no backs, and stoolshalf a seat. The results of inadequate sup-

Table 1Primary school enrollment ratios, 1960-78

Country income group by GNP per capitaNumber ofcountries

Gross enrollment1

(tn per cent)

1960 1965 1970 1975 1978

Low income,less than $2652

Lower middle income,$265-$5202

Intermediate income,$520-$1 ,075"

Upper middle income,$1 ,075-$2,5002

Low and middle incomeHigh income, above $2,500

36

21

21

149217

48

59

71

8557

114

58

65

84

9366

118

61

69

95

9671

120

64

73

102

10275

120

70

80

107

10378

116Source: World Bank, Education Sector Policy Paper, 1980.'The gross enrollment ratio is the total enrollment of all students in primary school divided by the population that

corresponds to the age group of primary schooling. Over- or under-aged students can frequently inflate the figures, andaccount for the percentages above 100. See United Nations Educational, Scientific, and Cultural Organization, StatisticalYearbook 1978-79.

2ln 1975 U.S. dollars.



in developing countries

Table 2Allocation for nonsalary schoolresources as a per cent of totalrecurrent expenditures, 1978-79

Primary Secondary Tertiary

Africa 3.8 12.7 13.1Asia 8.8 13.8 22.7Industrial countries 14.4 14.8 25.5

Source: United Nations Educational, Scientific, andCultural Organization, Statistical Yearbook 1978-79.

port for educational systems have a nega-tive impact on an economy by reducing thevalue of the already low investment ofscarce resources.

Such inadequate teaching environmentsare partly a result of the fact that the lion'sshare of the funds available for education—on average 95 per cent of the money setaside for primary schools—has to go to-ward teachers' salaries. But paying a teach-er to copy from a worn-out textbook onto ablackboard and to supervise its memo-rization by 50 students is an ineffective uti-lization of expensive talent. The question iswhat level of classroom resources teachersshould have available to justify their salary.In industrial countries today, 14 per cent ofrecurrent costs of primary schools are allo-cated to classroom resources—books,maps, visual aids, furniture, and the like—and 86 per cent is spent on salaries. Theaverage in Asia is 9 per cent and 91 per centfor salaries; and in Africa 4 per cent with 96per cent for salaries (Table 2). One policydilemma that developing countries are nowconfronting with regularity is the minimumstandard of resources a primary schoolshould have before new teachers are takenon. Experience has shown that, at a min-imum, expenditures on teaching toolsshould account for approximately 10 percent of public recurrent expenditures.

The large difference in educational qual-ity between low- and high-income coun-tries is also widening. As more and morepupils enter school there is less available toteach them with. In 1960, the average Or-ganization for Economic Cooperation andDevelopment (OECD) country invested 14

times more for each elementary school pu-pil than did any of the 36 countries with percapita incomes of below $265 per year. Butby 1977 the difference had grown to 50:1.

Education and productivity

Although conclusive evidence is scarce,it is clear that the poor quality and limitededucational materials available in devel-oping countries adversely affect the level ofcognitive skills a student can acquire atschool. Recent studies show the averagestudent from a developing country scoringat a level that falls in the bottom 5 to 10 percent of students from a high-income coun-try, and the poorer the country in economicterms, the fewer cognitive skills are ac-

quired by the end of the primary schoolcycle. Though any comparison—whetherbetween pupils or schools or districts orcountries—should be approached cau-tiously, there is general agreement thatachievement is lower in schools in poorcountries.

The implications of these results forgrowth in the developing world are sober-ing. The general education of a populationin science, mathematics, and reading has asignificant bearing on the degree of pro-ductivity to be expected from it. The linkbetween irrigation-based farming and edu-cation is illustrative. There are approxi-mately four different levels of technologyinvolved in irrigation-based agriculture

Farmer-entrepreneurstechnology level

Level A:Traditional farming

(Techniques passedfrom parent to child)

Level B:Intermediate technology

Agricultural Inputs

Local varieties of seedsand implements.

Small quantities offertilizer.

Minimum learningrequirements

Addition and subtraction—notnecessarily acquired throughformal education.

Addition, subtraction, division,and rudimentary literacy.

Source: Risto Harma, "The Farmer Entrepreneur and Hts Prerequisite Prior Education in Agricultural Development,"mimeograph, World Bank, 1979.

1 Direction from agricultural extension agents is helpful at any stage, but the essence of a farmer-entrepreneur is his abilityto calculate his own production function.

Finance b Development I March 1983 19

Four basic stages of agriculture productivity and their learning requirments'

Level C.Fully improvedtechnology

Level D.Full irrigation-basedfarming

All above inputs; tubewellaccess during theoff-season; and waterrates acre

Mathematics, independentwritten communication, highreading comprehension,ability 10 research unfamiliarwords and concepts;elementary chemistry,blotogy. physics; and regularaccess to information fromprint and electronic sources.

High-yielding varieties,proven seeds; seedrales acre, fertilizerrates/acre; and pestcontrol rates'acre.

Multiplication, long division,and other more complexmathematical procedures,reading and writing facililies,and rudimentary Knowledgeof chemistry and biology.


(see chart). The most elementary is whereknowledge and skills are passed fromfather to son (see chart, level A); littleschooling is required here. The secondlevel of technology (B) includes a singlemodern input, such as fertilizer, whose uti-lization is substantially improved if thefarmer has rudimentary literacy and aknowledge of addition, subtraction, and di-vision. If a farmer has none of these skills,he will have to follow by rote the one-to-one advice of an extension agent—an ex-pensive and comparatively inefficientmethod of learning. The third level (C) in-cludes several modern inputs simulta-neously, such as high-yielding varieties ofseed and careful allocations of pest controland fertilizer. Here, having to follow theadvice of an extension agent by rote is evenmore expensive, but for the farmer to takehis own initiative requires an under-standing of long division, multiplication,and other mathematical procedures; anability to read and write; and rudimentaryknowledge of some chemical and biologicalprinciples. The fourth level of technology(D) is the most modern and includes all theabove inputs plus tubewell access duringthe off-season. For a farmer to operate effi-ciently at this level he needs to be able tocommunicate in writing, to research un-familiar words and concepts himself, tounderstand basic concepts drawn fromchemistry, biology, physics, and to havedependable access to new informationfrom print as well as electronic sources.This is the ideal: that every farmer be ableto calculate his own "production-function"anew every year and with each change incrop.

Irrigation farming has been used here asa proxy for national development schemesthat aim to increase farm productivity byusing new technologies in the (C) and (D)categories. For these schemes to be success-ful, certain minimum cognitive skills are re-quired of the farming population. A lack ofunderstanding or a misunderstanding ofthe principles on which new technologiesare based can result in catastrophic agricul-tural yields, and harmful side effects aswell. Such skills are required no less ur-gently by the manufacturing and serviceindustries.

Basic cognitive skills are also required iffurther training in other, often nonformal,areas is to be successful. The Bank's experi-ence has shown that if people are to be ableto make the most of technical or on-the-jobtraining, they not only need to spend a cer-tain amount of time in formal education,they also need to emerge from formal ed-ucation with a sufficient grounding in cog-nitive skills. Being exposed to 6-12 years ofthe low quality education provided in

many countries may not be enough. Sev-eral faculties of engineering have remainedup to one third empty—not because of alack of demand for local engineers but be-cause there has not been a sufficient supplyof qualified entrants with the requisitegrounding in mathematics and science.This experience has forced the Bank, forexample, to rethink the process of man-power planning very carefully and to incor-porate a factor of quality in addition to thenormal supply figures for numbers of yearsexposed to formal schooling.

Textbooks crucial

Since the 1960s, social scientists havebeen trying to isolate the characteristicsmost closely associated with achievementin basic cognitive skills. The approach isrelatively recent and imperfect; but, thoughtechnical caveats abound, it has come toone important conclusion: in wealthy coun-tries much of the learning in school is ac-counted for by factors not connected withthe school environment, implying that ad-ditional physical facilities, teaching equip-ment, and textbooks only help the acqui-sition of new knowledge a little. But thequality of the physical and particularly theclassroom tools in low-income countries isshown to explain three and even four timesthe differences in achievement that it can inhigh-income countries. In fact, the poorerthe country in economic terms, the largerthe impact on achievement school qualityseems to have.

The evidence is scarce on which aspectsof the educational environment can be ex-pected to have the most impact on studentachievement in the developing world. Butone conclusion is consistent: higherachievement is associated with the avail-ability of textbooks and other printed mate-rials. Of the 20 assessments that have beenmade of the relationship between the avail-ability of printed material and studentachievement scores, 17 have reported posi-tive effects.

Stephen P. Heynemana U.S. citizen, was edu-cated at the University ofChicago. He joined the Ed-ucation Department of theWorld Bank in 1976,where he is now a seniorsociologist, and has par-ticipated in the assessmentof education projects inAsia, Latin America, andAfrica.

Providing books where there is nothingto read in the schools sounds like a simpleundertaking but may entail substantial lo-gistical challenges—the books have to beacquired or produced, stored, distributed,and so on. But there have been successes inthe efforts made to improve this aspect ofschooling. The Bank's Third EducationProject in the Philippines assisted the Gov-ernment to design new, high-quality text-books in mathematics, science, and Fili-pino. Approximately 97 million of themwere distributed sufficiently widely to alterthe average ratio of pupils to books from10:1 before the loan to 2:1 afterward. Aver-age student scores were raised signifi-cantly; in fact, the achievement gains in thePhilippines were approximately twice themagnitude to be expected in North Amer-ica were class size to be reduced from 40 to10. Moreover, the gains from the projectwere frequently the greatest where theywere least expected—among studentswhose family background and school lo-cation have been associated with severepoverty and who came to the school withlow levels of knowledge.

Where there are enough textbooks, andthe school system has met certain stan-dards of efficiency and good management—as in Algeria, Greece, Ireland, or Korea—the Bank tends to lend for science labora-tory equipment, supplementary readers,library facilities, electronic media, or spe-cialized teaching facilities. These too can beexpected to increase achievement by sig-nificant amounts.

Thus, the demand for educational in-vestment is slowly experiencing a shift inemphasis away from providing places foruniversal primary education and towardimproving the quality of the learning. Asfar as we can predict, the bulk of the Bank'scapital investment in education will con-tinue to assist the expansion of schools, butthis proportion is expected to decrease. Infiscal year 1983 it should decline to lessthan 85 per cent of resources disbursed foreducation—from 93 per cent between 1975and 1978. The decline is accounted for byincreases in lending for curriculum devel-opment, for radio, television, educationaladministration, and particularly for in-creases in the development, production,and distribution of learning materials. Byfiscal year 1983 the Bank's education sectorlending is expected to be $50 million a yearin classroom "tools"—up from $1.6 milliona year a decade earlier. For example, 5 percent of education projects contained fundsfor textbooks in fiscal year 1975; 10 per centin 1976; and 25 per cent in 1977. By fiscalyear 1978 this had risen to 40 per cent. Asa result of the Bank's investments, a verylarge proportion of students have or will

20 Finance & Development /March 1983


have textbooks in Benin, Brazil, ElSalvador, Indonesia, Lesotho, Malawi, thePhilippines, and Swaziland.

Future policy developments

Two results seem to be emerging in thewake of the demonstrated importance ofthe quality of educational materials. First,developing countries want to design, man-ufacture, and distribute their own edu-cational materials, including textbooks.The World Bank has, where it is economicto do so, made loans for printing presses,storage and distribution systems, papermills, and the training of editors, de-signers, and production experts. This is thecase in both Indonesia and the Philippines.

But this area needs to be analyzed care-fully. In some cases, the cost of producingbooks locally from scratch may be prohib-itive and the technical experience, equip-ment, and raw materials (particularly pa-per) may be imported more cheaply fromWestern Europe or North America. Thepublication process demands substantialexperience in editing and production, inprinting, testing, and marketing. Six to tenyears is normally required to develop a newgeneration of textbooks for primary schoolgrades one through six. This may, giventhe availability of the necessary skills, beeconomic for books on local history, civics,and literature; but it may be cheaper toadapt already published materials in math-ematics and the sciences. However, it is of-ten more economic for countries to publishtheir own textbooks than to print them.Printing in large quantities requires special-ized machinery and a constant supply ofraw materials and maintenance skills; pub-lishing requires editorial and design skillsbut not necessarily the local hardware formanufacture.

A second effect of the interest in the qual-ity of education is the increased awarenessof the need for equality of educational op-portunity within and between populations.There is a standard for educational oppor-tunity today that varies only slightly fromone high-income country to the next. Indeveloping countries, however, equal edu-cational opportunity has, so far, meantonly a place at school. But when a pupil inBolivia, India, or Malawi, who must learnsimilarly complex skills of mathematics orscience, has access to only one sixtieth thelevel of learning resources as the child inEurope or North America, there is noequality of educational opportunity be-tween Bolivia, India, and Malawi on theone hand, and Europe or North Americaon the other. For the first time in history itwill soon be possible for every individual ata given age to have an opportunity to beginschooling. But this is not a sign that equal-

ity of opportunity has been reached be-tween nations. Substantial new resourceswould have to be made available to pupilsin developing countries if they are to haveanything approaching equality of opportu-nity with pupils in the developed worldafter entry in school has been obtained.

This poses a substantial economic di-lemma. Should the quality of education indeveloping countries be the same as that inthe industrial countries? Some areas ofeducation—such as electrical engineering,architecture, or computer sciences—arefairly similar in all countries; consensusexists on the requirements for training andthe physical inputs necessary to deliver it.This consensus is similar to the agreementthat exists on the physical requirements forbuilding a bridge, a road, or a dam withcertain specifications. Moreover, minimumphysical requirements, and therefore min-imum cost standards, are predictable.

Primary and secondary schools also haveminimum requirements for achievement indifferent skills, but there is little consensuson the minimum resources required to pro-vide them. The major industrial countriesinvest 50 times more per pupil to meet sim-ilar curriculum objectives than do certaindeveloping countries. This disparity leavesdeveloping countries with an unenviablechoice: establish a ceiling on enrollments;possibly also decrease science and readingcurriculum objectives to a level well belowthose of industrial countries; or increase ex-penditures on education to provide the

quality of resources needed if their curric-ula are to be efficient. If the latter option ischosen, some consensus must be reachedon the level of inputs required and whereresources can be found to obtain them.These will be the central issues in most de-veloping countries in the years to come.

Related reading

Stephen P. Heyneman, The Evaluation of Hu-man Capital in Malawi, World Bank StaffWorking Paper No. 420 (1980).

Stephen P. Heyneman, Joseph P. Farrell,and Manuel A. Sepulveda-Stuardo, Text-books and Achievement: What We Know,World Bank Staff Working Paper No. 298(1978).

Stephen P. Heyneman and William Loxley,"The Impact of Primary School Qualityon Academic Achievement AcrossTwenty-Nine High and Low IncomeCountries," American Journal of Sociology(forthcoming).

Torsten Husen, Lawrence). Saha, and Rich-ard Noonan, Teacher Training and StudentAchievement in Less Developed Countries,World Bank Staff Working Paper No. 310(1978).

Peter H. Neumann, Publishing for Schools:Textbooks and the Less Developed Countries,World Bank Staff Working Paper No. 398(1980).

International readership surveyIn October 1982 Finance & Development maited a questionnaire to2,500 readers randomly selected from the mailing lists of our variouslanguage editions. The survey was conducted on our behalf by OpinionResearch Corporation, a professional polling organization based in NewJersey, U.S.A.

An earlier readership survey, conducted in 1976, gave us a goodindication of the composition of our readership. The new surveyupdates and expands this information, but its primary purpose is to elicitthe views of our readers on the journal's contents, presentation, andmany other matters. We hope the results will provide us with guidanceregarding the needs and preferences of our readership.

The response to the survey has been excellent and we should like tothank those of you who took the time to participate. The results arebeing tabulated and a report will be published in our June issue.



External debt—the

"The external indebtedness of non-oil de-veloping countries has been of growingconcern in recent years." So began a studyon debt published by the Fund in May1981. The same could equally well be saidtoday, with even more reason: not onlyhave the debt problems of many non-oildeveloping countries become more acute,but also there has been mounting concernabout the external debt of other countries,including some developed countries. Thelong-term external debt of non-oil devel-oping countries increased by 52 per centbetween 1979 and 1982. Meanwhile, theirdebt service ratio on long-term foreign bor-rowing—interest and amortization pay-ments as percentages of exports of goodsand services—rose from 18.1 to 22.3 percent, and much more so for some individ-ual countries. The accumulation of short-term debt (excluded from the above fig-ures) has added significantly to the debtburden problem.

The serious implications of the risingburden of debt in today's recessionary con-ditions were highlighted by A. W. Clausen,President of the World Bank, in a recentspeech at the United Nations: "Countrieswhich could manage their debts com-fortably as long as trade was expanding,"he said, "now have more difficulty payingtheir debts, and commercial banks seem tobe responding by curtailing credit." Thissituation, combined with stagnating tradeand declining assistance, was a matter ofconcern. "The fabric of international coop-eration," he warned, "just might not beable to bear up under the strains of pro-longed economic stagnation...."

In the past three years the external debtsituation of some developing countries hascome under increasing pressure. The num-ber of countries seeking a rescheduling oftheir debt has increased sharply: those un-dertaking a rescheduling of their officialdebt under the auspices of the "Paris Club"increased from 3 in 1980 to 8 in 1981, fallingto 6 in 1982. In addition, during this period

many countries, including some major bor-rowers such as Argentina and Mexico, en-countered serious problems regarding theservicing of their commercial banking debtand had to undertake a restructuring of thisdebt. Five such agreements were con-cluded in each of the years 1981 and 1982,and many more are under negotiation. Afurther development was the external debtdifficulties experienced by a number ofEastern European countries, notablyPoland—a development that shattered thewidely held belief that the U.S.S.R. wouldultimately stand behind an Eastern Euro-pean country in financial trouble.

These developments have stimulated agreat deal of literature and much debate,particularly regarding the impact of debtproblems on the international financial sys-tem. The discussion has focused on a num-ber of points: the nature of debt problems

Non-oil developing countries:external debt service payments, 1973-81

40

35

30

25

20

15

10

5

0

(In per cent of exports of goods and services)

Net oil exportprs'

non-oil developingcountries .A

Other countries

i / Low income/ countries*'Maior exportersof manufactures'1

1973 1974 1975 1976 1977 1978 1979 1980 1981

Source: IMF, Annual Report, 1982.'Bahrain, Bolivia, People's Republic of the Congo, Ecuador,

Egypt, Gabon, Malaysia, Mexico, Peru, Syrian Arab Republic,Trinidad and Tobago, and Tunisia.

'Argentina, Brazil, Greece, Hong Kong, Israel, Republic ofKorea, Portugal, Singapore, South Africa, and Yugoslavia.

ith per capita gross domestic product not exceeding theequivalent of US$350 in 1978.

Consisting of middle-income countries that, in general,export mainly primary products.

(whether they are due to transient factorsor whether they reflect more basic eco-nomic factors); the pervasive effects of theworld recession on developing countries'exports and thus their capacity to servicedebt; the impact of high interest rates, par-ticularly since an increasing amount of debthas been lent at floating commercial rates;and the lending practices of commercialbanks. Another very important questionhas been whether current rescheduling ar-rangements are satisfactory and to what ex-tent commercial banks should assist coun-tries experiencing debt difficulties. TheManaging Director of the Fund, Jacques deLarosiere noted that "Although some re-duction in 1983 in the rate of increase of thebanks' overall exposure is not unlikely, toosharp a reduction would clearly be harm-ful, even supposing it were feasible.... Anabrupt curtailment of bank lending wouldnot only run counter to the interests of thedebtor countries, but it would also be con-trary to the interests of the banks them-selves and would weaken the financial sys-tem as a whole."

On the particular question of debt man-agement, the role of the Fund in assistingcountries with debt problems has also beenan issue. In the words of Mr. de Larosiere," . . . bank lending that gives a country anunwarranted sense of financial ease is notsustainable over the medium term.. . . Inthis regard, the Fund can play a helpful rolein the context of its regular consultationsand reviews of policies with member coun-tries in counseling them as to how best tostay on a sustainable policy path." Mean-while, as a result of the Ditchley II con-ference, 31 industrial country banks agreedto establish an international banking insti-tute to improve the availability and qualityof financial and economic information onmajor borrowing countries.

The World Bank and the Fund—apartfrom their role in assisting member coun-tries with their debt problems—have donetheir part in contributing to the under-

22 finance & Development I March 1983


continuing problem

Table 1Public and private debt of developing countries, 1973-82

(In billions of U.S. dollars)

1973 1975 1976 1977 1978 1979 1980

Total debt 109.4 161.9 195.5 240.1 298.8 352.4 404.8Total debt service 16.0 23.0 26.1 33.1 47.9 62.3 70.4

Low-income Africa 0.4 0.5 0.5 0.6 0.7 0.8 1.1Low-income Asia 1.0 1.3 1.3 1.4 1.6 1.8 1.9Middle-income oil

importers 9.7 14.0 15.4 18.8 27,9 34.6 40.8Oil exporters 4.9 7.1 8.9 12.4 17,7 24.9 26.7

1981

462.183.0

1.31.9

49.830.8

1982'

529.095.0

Source: World Bank, World Debt Tables.• Indicates data not available.'Estimates

Table 2Long-term debt ratios for non-oil developing countries, 1973

Ratio of externallong-term debt toexports of goodsand services' 88.7 97.7 102.6 104.9 111.2 101.9 92.9

Ratio of externallong-term debtto GDP1 16.6 18.3 20.5 22.1 23,7 22.7 21.8

Debt service ratio2 14.0 14.0 13.9 14.0 17.3 18.1 16.3

-82

1981

103.2

24.321.0

1982

109.1

25.222.3

Source: IMF, Occasional Paper No. 9, World Economic Outlook, 1982, Tables 31 and 33.'Ratio tor year-end debt to exports or GDP (or year indicated.?Payments -interest, amortization, or both — as percentages of exports of goods and services.

Table 3Non-oil developing countries: distribution of debt by class of creditor,

end of year, 1973-82'

(In per cent)

1973 1975 1976 1977 1978 1979 1980

Total outstandingdebt of non-oildevelopingcountries 100.0 100.0 100.0 100.0 100.0 100.0 100.0

To official creditors 49.9 46.2 45.3 44.3 42.5 41.1 41.4Governments 36.9 33.0 31.7 30.4 28.8 27.4 27.2International

institutions 13.0 13.2 13.6 13.9 13.7 13.7 14.2

To private creditors 50.1 53.7 54.7 55.7 57.5 58.9 58.6Financial

institutions 35.7 42.7 44.4 45.8 46.2 49.5 49.6Other private

creditors 14.4 11.0 10.3 10.0 11.3 9.4 9.0

1981

100.0

40.125.9

14.2

59.9

51.4

8.5

1982

100.0

39.425.1

14.3

60.6

52.7

7.9

Source: IMF Occasional Paper No. 9. World Economic Outlook, 1982, Table 32.1 Excludes data for the People's Republic of China prior to 1977.

standing of debt issues. The Bank pub-lishes regular World Debt Tables, and debtquestions are covered annually in its An-nual Report and the World Development Re-port. It has also published a Staff WorkingPaper by Nicholas Hope entitled Develop-ments in and Prospects for the External Debt ofthe Developing Countries: 1970-80 and Be-yond. The Fund has published the studyquoted in the first paragraph, External In-debtedness of Developing Countries, and pub-lishes a yearly report, International CapitalMarkets that, though not specifically ondebt, gives a fairly comprehensive coverageof developments and prospects in interna-tional bank lending. In addition, the Funddiscusses the problem of indebtedness inits Annual Report and World Economic Out-look, which appear once a year.

Finance & Development has also publishedarticles on the topic—in recent years "Theexternal debt of developing countries," byHelen Hughes (December 1977); "Officialborrowing abroad: some reflections," byWalter Robichek (March 1980); "Managingexternal debt in developing countries," byBahram Nowzad (September 1980); "Debtin developing countries: some issues forthe 1980s," by the same author (March1982); and "Eastern Europe and the inter-national banks," by Richard Williams andPeter Keller (December 1982).

Both the Fund and the Bank have beenenjoined by the international financial com-munity and by many governments to playa larger role in debt issues—in the col-lection and dissemination of information,in surveillance of debt matters including as-sistance in the management of externaldebt, and in closer cooperation with coun-tries seeking a renegotiation of their debt.External debt questions are thus likely tooccupy a central position in the concerns ofthe two institutions. Finance & Developmentwill publish material on major develop-ments as they occur and will report on newanalyses and studies undertaken in theFund and the Bank. HD


1973 1975 1976197719781979198019811982

1982


Opportunities and constraintsInJnternatlonal lending*v^l*V^ , *

International bank lending hasbeen increasingly financingdeficits and reserveaccumulation—raising questionsregarding the functioning andcontrol of the monetary system.

David Williams

The growth of international banking, or thenarrower but better known phenomenon,the Eurocurrency markets, has been extra-ordinary over the last 25 years. From virtu-ally nothing in 1957, Eurocurrency depos-its, measured on a gross basis, amountedin 1982 to approximately US$2 trillion—acompound annual rate of growth of over 20per cent.

While the year-by-year fluctuations inthe rate of growth of international lendinghave been marked, at no time—even dur-ing the worrisome days of 1974 when anumber of commercial banks became over-extended in their international operations—has there been a sustained contraction.The expansion of international lending intocurrencies other than the U.S. dollar—in particular, the Euro-sterling, Euro-deutsche mark, and Euro-Swiss franc

markets—has been a very importantsteadying element in the growth of inter-national banking when conditions in somemarkets—particularly the dollar market—were pointing to a slowdown in growth.

There has also been a phenomenalgrowth in the number of financial institu-tions that participate in international lend-ing. At the end of 1981 some 600 banks,from about 85 countries, were operating incountries outside the legal domicile of theirheadquarters. These banks controlledabout 450 subsidiaries, owned a network ofabout 5,000 foreign branches, and had wellover 1,000 direct affiliations. It is strikingthat much of the expansion is in countrieswith well-established banking systems andmoney market centers, such as the morehighly developed countries in Europe andNorth America and South and East Asia.

The physical expansion of internationalbanking is to some extent a matter of thebanks following the growth of overseastrade and international investment by do-mestic corporations. In addition, theemergence of the main centers of inter-national banking reflect the influence of do-mestic regulatory and tax considerationsand, perhaps more important, the absenceof domestic controls and regulations—including the application of domestic inter-est rate ceilings, reserve requirements, andexchange controls—on the foreign oper-

ations of banks, as well as the presence ofhighly efficient international communica-tions systems. In other words, the rapidexpansion of international banking can berelated to a considerable extent to the priv-ileged position of banks in their non-domestic compared with their domesticoperations.

The banks have responded to the stimu-lus of nonregulated banking, which hasprovided not only a high rate of profit-ability compared with domestic bankingbut also wide scope to experiment withnew financial and credit instruments, newinterest rate mechanisms, and new ways ofmeeting customer needs and tapping fi-nancial resources. The expansion of inter-national banking has been a major phe-nomenon of the last quarter of a century;with its growth, however, have come prob-lems for the working of the monetarysystem—both domestic and interna-tional—that are still to be tackled.

The rapid growth of international lend-ing has not, of course, taken place in avacuum. Over the entire postwar period,the scale of the world economy has grown

This article is a shorter version of a paperdelivered at the Colloquium of the Soaite Uni-versitaire Euraptenne de Recherches Financieresin Vienna, in April 1982.



at impressive nominal annual rates. Sincethe mid-1950s, world trade in goods, ser-vices, and private transfers has grown at acompound rate of almost 10 per cent ex-pressed in U.S. dollar terms. Over much ofthe postwar period up to 1973, the periodof the long boom, very impressive rates ofgrowth in volume terms were also re-corded. International financial transac-tions, for which no reliable data exist, havegrown at substantially higher rates. Inshort, despite the slower real growth ratesof the 1970s, and the marked decelerationof the early 1980s, the world economy has,by historical standards, been buoyant overmost of the postwar period, and recessionshave generally been short-lived and com-paratively shallow.

New role for banks

The growth of the international bankingsystem has reflected the rapid growth inthe world economy, and indeed can takecredit for a major role in facilitating thegrowth in international trade and pay-ments that has contributed so much to theunprecedented international prosperity ofthe postwar years. International bankinghas also been an important part of the de-velopment process in the Third World aswell as elsewhere over much of the period.However, the system performed a similarrole in the last third of the nineteenthcentury and the early years of this century.No doubt an efficient international fi-nancial infrastructure is a necessary condi-tion for furthering international economicdevelopment.

But over the last decade or so, the inter-national banks have also been performinganother major function—namely, the pro-vision of balance of payments finance on acontinuing basis and on a large scale to of-ficial borrowers. Formerly, resources tomeet balance of payments deficits were ex-changed mainly, though not exclusively, atan intergovernmental level and entailedbalance of payments adjustment. (This in-volved essentially domestic deflation or, inthe event countries had an effective choice,given their exchange rate arrangements, afloating exchange rate—"going off the goldstandard"—with an accompanying defla-tion of domestic demand.) The possibilityof financing a balance of payments deficitfrom the banking system over a prolongedperiod was not an option that was prac-tically available to countries prior to the1960s.

Over the last decade, the commercialbanks have become the main providers ofbalance of payments financing and the sup-pliers of international reserves to officialentities. In a very real sense, the inter-national financial system is now a network

of international banks whose scale of oper-ations is outside the direct control of thecentral institutional authorities that are di-rectly responsible for keeping the systemon an even keel. The question is whetherself-regulation by the banks represents asufficient control and stabilizing influencein the international monetary system.

Implications of banks' role

During the decade of the 1970s, officialholdings of foreign exchange reserves qua-drupled, which is an unparalleled rate ofaccumulation. During this period almostone half of net current account deficits andthe reserve accumulation of the smaller in-dustrial countries and the non-oil develop-ing countries was financed by net borrow-ing from banks and from the internationalbond market. The international bankingsystem, centered on the main industrialcountries, essentially financed the pay-ments deficits resulting from the sharp in-creases in the price of oil in 1973/74, whichresulted in the major oil exporting coun-tries becoming a major source of funds inthe international banking system. A sub-sequent reduced flow of funds from themajor oil exporting countries has tended tobe compensated by a renewed outflow ofbanking funds from the industrial coun-tries, mainly the reserve currency centers—the counterparts of the growth and cur-rency diversification in the Eurocurrencymarket and the source of the multicurrencyreserve standard, which is a distinguishingfeature of the present phase in theevolution of the international monetarysystem.

The flow of banking resources to thenon-oil developing countries was untilmid-1982 remarkably steady in terms oftheir current account deficits. The com-plementarity of the sources of funds under-scores the demand-determined nature ofthe international operations of commercialbanks. It also underscores the relative in-stability of this system of reserve creation; aslowdown in international lending will af-fect the ability of countries to finance theirbalance of payments deficits and to accu-mulate their desired levels of internationalresources—that is, a generalized shortageof liquidity could arise, which could in-duce disorderly external adjustment bycountries.

The international operations of the com-mercial banks must therefore be looked atnot only from the point of view of theirimpact on the growth of world trade andincome—their traditional commercial fi-nancing functions—but also from the pointof view of their effects on the developmentof the international monetary system as awhole and, in particular, on the stability of

that system, including the process of re-serve creation. To some extent, the impactof these two main functions is indistin-guishable. Countries that borrow for tradeand development can, and do, use thoseresources as holdings of internationalliquidity—or reserves. Indeed, it is difficultto construct a comprehensive series of usesfor the resources provided by the inter-national banks over the last 25 years, al-though a great deal of data exist on thematurity distribution of international bankclaims and on the country distribution ofdebt. A significant question that needs tobe answered, however, is whether thelarge-scale borrowing has generated, or isgenerating, sufficient foreign exchangeearnings over the medium term to pay theinterest and repay the principal as well asto stimulate reasonable domestic economicgrowth in the debtor countries. This ques-tion arises in particular for that part of thedebt incurred as balance of payments fi-nancing, because this "transfer problem"is, in view of the extraordinarily large vol-ume of outstanding international indebt-edness, an integral part of what is now usu-ally referred to as the balance of paymentsadjustment process.

Limits to international banking

In view of the pervasive and funda-mental role of international banking in theworld economy as purveyor and creator ofinternational credit, issues naturally ariseregarding the limits to its growth and howthe system can be controlled in the inter-ests of monetary stability. Domestic bank-ing systems are subject to the ultimate con-trol of central banks, which, despite theirresidual obligation to finance net officialdeficits of the government, can impose asufficient degree of restraint on the domes-tic activities of the banking system toachieve the broad aims of the authorities, atleast over the short term. While no similarinstitutional arrangements exist at the in-ternational level—and it is for considera-tion whether that state of affairs is in thegeneral good—the growth of internationalbanking is constrained over the short runby developments in the international econ-omy itself. The banks' responses to thesedevelopments will help determine thecourse of the international economy in the1980s.

Developments in the world economy areworrisome. The rate of growth of the worldeconomy is exceptionally sluggish, particu-larly as measured by postwar standards.Rates of growth of output of 1 to 2 per centand lower have been common over the lastthree years, and the immediate prospect isfor little improvement. The low rates ofgrowth in output and external trade, and



accompanying high rates of unemployedresources in the labor and product markets,not only make it difficult to effect structuraladjustment but also induce countries toborrow abroad to help finance domestic de-mand, which thereby delays the adjust-ment process and results in acute externalfinancial difficulties.

In addition, the widespread reliance onmonetary restraint to curb the endemic in-flationary pressures of the last decade hasresulted in extraordinarily volatile interestrates and, recently, in very high real levelsof interest rates. Volatile interest rates havebeen associated with—indeed have helpedto induce—very high short-run variabilityof exchange rates between some of theleading industrial countries—the group ofcountries in the European Monetary Sys-tem being a particularly noteworthy excep-tion. With these developments, a shift toshorter maturities on bank loans has be-come noticeable, and many new forms ofborrowing have been introduced in themarket as a means of reducing the higheffective medium-term cost of borrowingand assuring the flow of resources. Float-ing rate note issues, for instance, are nowalmost the standard form of borrowing. Inaddition, however, potential buyers ofmedium-term bonds have been temptedwith bonds convertible into equities, withissues incorporating a currency swap ele-ment, with zero coupon and deep discountissues, and—shades of the 1920s—withcommodity-backed financings (of whichone such issue was linked to the price ofgold and another to the price of oil). Thehigh variability of exchange rates and,more particularly high interest rate levels,has significantly increased the risks, andthereby the costs, of international banking.

Another adverse factor in the currenteconomic environment is that although themassive payments imbalances of the pastfew years have diminished to a consid-erable extent, the aggregate deficit of thenon-oil developing countries (approxi-mately $86 billion in 1980 and about $99billion in 1981) has fallen only slightly.Massive financing will again be needed, ofwhich, on recent past experience, aboutone half can be expected to come from theinternational banks. Without such assis-tance these countries would need to con-tract their imports on a major scale andseverely curb their already inadequate ratesof growth.

However, further large-scale deficits,partly financed by borrowing, will add toan already burdensome load of interna-tional debt. Total foreign indebtedness ofthe non-oil developing countries amountedto approximately $550 billion in 1981,which represents a compound annual rate

of growth of almost 21 per cent since 1971.The ratio of debt service payments to ex-ports of the non-oil developing countries in1981 was approximately 23 per cent, in con-trast to a ratio of approximately 17 per centin 1978. High interest rates compound theburden of large absolute amounts of debtand worsen the current account problemsof servicing it. A 1 percentage point changein interest rates in the Eurodollar market isequivalent to $2 billion in debt service to

the non-oil developing countries based ontheir current debt. Furthermore, the condi-tions are spawning increasing demands fordebt reschedulings on a widespread scaleand thereby spread all debt over relativelylong periods of time. Finally, and perhapsnot unexpectedly, the general quality ofdebt, both official and corporate, in boththe industrial countries and the developingcountries, is deteriorating under the pres-sure of low growth, high inflation, andhigh real interest rates.

In short, the international banking sys-tem is facing an increasingly difficult eco-nomic environment over the short-term.The credit-worthiness of some countries,and corporations, has fallen dramatically.The banking system itself is having toadapt to a new environment of consid-erable instability of interest rates and ex-change rates, declining adequacy of capitalto loan ratios, and increasing demands fordebt rescheduling with consequential re-ductions in creditworthiness. In the light ofthe deterioration in the world economy andthe decline in the former high rates ofprofitability on international lending, andtaking into account the rise in bad ornonperforming assets, there is a risk thatthe flow of bank credit will dry up, particu-larly from the smaller institutions.

There are then, from a number of view-points, grounds for the banks to becomemore cautious in extending new loans.However, there is no reason for the gener-alized indiscriminate slowdown in inter-national lending by the banks that has

emerged during late 1982. Indeed, in thepresent darkening economic circum-stances, it is clearly in the general interestto maintain an adequate flow of fundsthrough the international banking systemand in particular to the non-oil developingcountries.

The banks and adjustment

At the same time, there is need for fur-ther balance of payments adjustment inmany countries, both developed and devel-oping. Clearly, the financing and adjust-ment mechanisms need to be reconciled.However, the present institutional arrange-ments do not assure proper reconciliationbetween adjustment and financing. Thenext phase in the evolution of the inter-national monetary system may well be con-cerned with this dichotomy between thereserve-creating institutions and thereserve-controlling institutions. In themeantime, the banks themselves can takemore steps to help reduce the risks of inter-national lending and maintain some of thefinancial discipline associated with balanceof payments adjustment.

The orderly evolution of internationalbanking would seem to call for commercialbanks to adopt formal measures of self-control. A greater collaborative effort bythe leading international commercial bankswould be an important stabilizing elementin international banking. This effort couldtake a number of different forms. A systemcould be established to provide for a con-tinuous exchange of information betweenbanks in areas critical to their operations—such as the extent of the reporting bank'sinvolvement in each country, the maturitystructure, and the amounts of credit out-standing in individual countries both in ag-gregate and by key sectors. The new ini-tiatives by a number of commercial banksto establish an institute of internationalbanking (a product of the recent Ditchley IIconference) marks an important advance ininternational commercial bank collabora-tion. A comprehensive approach could bedeveloped for the management of externalfinancial risks, including the establishmentof a country risk assessment or ratingsbureau. The major commercial banks couldalso arrange publicly outstanding lines ofcredit with each other. A financial poolingarrangement through the creation of creditlines could provide the banks with an auto-matic short-term source of liquidity andcould also encourage banks to providestand-by facilities to their overseasbranches and to smaller internationalbanks. A further development could be thecreation of an international deposit insur-ance, or credit guarantee fund, whichwould increase the confidence of depos-

26 finance & Development / March 1983


itors and reduce the risks of internationallending. These are some measures that thebanks themselves could initiate and putinto operation.

The role of official institutions

The current overall scale of operations bythe international banks attracts official con-cern, particularly in view of the lack of cen-tral control over the international bankingsystem. The present decentralized systemof prudential control, which rests essen-tially with the commercial banks them-selves and secondarily with individual cen-tral banks, needs further improvement inorder to safeguard the stability of the bank-ing system and the international monetarysystem.

The major responsibility for virtually allthe prudential aspects of internationalbanking continues to fall on the individualcommercial banks. However, since Decem-ber 1974, a great deal of work has beendone under the aegis of the Bank for Inter-national Settlements (BIS) in the form ofthe Committee on Banking Regulation andSupervisory Practices and on the surveil-lance of banking practices, the clarificationof the relationships between local centralbanks, head offices and branches and affil-iates of the international banks, and thesystematic monitoring of internationalbanking. Improvements to the monitoringprocess have included strengtheningprudential control over the internationalbanks by supervising banks' internationalbusiness on a consolidated basis and byassessing country risks better.

The formal arrangements for the super-vision and surveillance of internationalbanking are comprehensive, and the rulesare reasonably fully articulated. Coopera-tion—if not already operational—is in-creasing, though it is an essentially untriedmethod of self-control. The exchange of in-formation between the major centralbanks, the inspection by central banks of—and their control over—the internationaloperations of commercial banks is atpresent at a level that should, at least intheory, if not in practice, provide the cen-tral banks with an early warning system, ifnot the ability to avoid a crisis from devel-oping with the inevitable need for more di-rect official intervention.

However, the growing scale of the for-eign operations of commercial banks,which include both borrowing as well aslending, could result in circumstances inwhich the local central bank could not ade-quately cushion all adverse strains. Underthese circumstances, coordinated centralbank action may be necessary to help halt acrisis or to prevent one from developing.An important example of coordinated cen-

tral bank action at a time of incipient crisiswas the announcement made in September1974 by the governors of the central banksof the Group of Ten and Switzerland thatthey would be prepared to supply liquidityto the markets if the need arose. The sig-nificance of this action was that the majorcentral banks created a precedent to com-mit liquidity to the international bankingsystem if needed and thus acted as a majorstabilizing element.

Uncertainty regarding the capacity of na-tional central banks to stabilize the systemin turn raises the question of the role ofofficial international financial institutions,in particular the Fund. The Fund is a mainsource of conditional liquidity for a mem-ber in need of balance of payments fi-nancing. A Fund member has access at anytime to its various facilities, provided it ac-cepts the policies attached to their use, andthe member is encouraged to use those re-sources earlier rather than later in its exter-nal financial difficulties. The conditional re-sources of the Fund can be made availableonly on the basis of an agreed stabilizationprogram with the member. However, thepolicies relating to the implementation of astabilization program over a period of someyears are not intended, nor are they de-signed, to meet a sudden or emergency sit-uation in a member's balance of paymentsand reserve position. Nevertheless, there isa need to relate in some systematic fashionthe processes of reserve creation, which aredetermined largely by the commercialbanks, with policies to ensure balance ofpayments adjustment over reasonableperiods of time, without members re-sorting to policies destructive of the growthof trade and payments.

The Fund and the commercial bankshave an essentially complementary re-lationship. However, the Fund's legal andpractical responsibilities are solely to itsmembership, and the membership regardsits relations with the Fund as sensitive andconfidential. For example, these relation-ships would not be enhanced in the eventthe Fund decided to publish its reports andevaluations of members' economic policies

David Williamsa British citizen, isDeputy Treasurer of theFund, and a graduate ofthe University of London.He has contributed to anumber of economicjournals.

and prospects; there is also a limit to theamount of information that the Fund, at itsinitiative, could publish or otherwise makeavailable that bears on the economic andfinancial performance of a member. It alsofollows that the Fund cannot in normalcircumstances advise banks regularly ontheir lending activities with individualcountries.

There are, of course, ways in which theFund can and does help its members intheir relations with the commercial banks.For example, there has been concern thatexcessive lending by commercial banks canlead to debt service problems for borrowingcountries, which in turn could threaten thestability of the international banking sys-tem. In response to this concern, the Fundcan play a useful role in helping to resolvecrises that might arise in the servicing ofbank debt, especially as they bear on thebalance of payments position of members.The Fund has also, at the request of itsmembers, advised on members' relationswith the commercial banks, including as-sisting in obtaining adequate informationon their external financial situation. Fur-thermore, banks have been lending tosome countries on a net basis only becausethe country had concluded a stabilizationagreement with the Fund. In this latter re-spect, the commercial banks lend condi-tional resources to support the balance ofpayments adjustment effort of the bor-rower. The Fund itself also closely monitorsdevelopments in international banking,which includes ongoing discussions withthe commercial banks, central banks, andinternational agencies such as the BIS andthe Organization for Economic Cooper-ation and Development.

Under present arrangements, the Fund'srole vis-a-vis the private banks is limited.Its impact on the financial markets and theinstitutions that compose those markets isfelt mainly, though not entirely, indirectly,through its influence on—and ability topersuade—its members. This points up agap in the present international monetarysystem: namely, an absence of central con-trol over lending by commercial banks forbalance of payments purposes, which re-sults in the creation of international re-serves, and the overwhelming need to linkthat financing with policies for balance ofpayments adjustment, which the bankscannot themselves provide. This is a majorchallenge in the improved working of themonetary system—to make reserve cre-ation serve the purposes of balance of pay-ments adjustment. It is a challenge not onlyfor the commercial banking system but alsofor those official international institutionscharged with safeguarding the stability ofthe international monetary system. HD



What are credit ceilings?

Fund-supported stabilization programstypically contain, as one of their provi-sions, limitations on the amount of creditthat may be extended during the programperiod. These limitations are called creditceilings. There is a great deal of confusionabout what these ceilings are, why they areincluded in programs, and what happenswhen they are breached.

The view that monetary and credit devel-opments affect nominal income and thebalance of payments has been corroboratedby both theory and experience. When do-mestic credit exceeds the amount of moneydemanded, nominal income is raised,which leads to greater expenditure on do-mestic goods (causing inflation) and on for-eign goods (resulting in a deterioration inthe external position). Credit policy, in-cluding ceilings to monitor its develop-ment, is therefore an important element inadjustment programs, which seek to man-age aggregate demand, to switch ex-penditures toward tradable goods andinvestment, and to encourage domesticproduction. Specific measures are tailoredto a member's situation, economic struc-ture, and social priorities and are adjustedfor the impact policies have on economictargets. In particular, the amount of do-mestic credit permitted under a ceiling isindividually determined.

Ceilings on domestic credit rather thanon the money supply are used becausemonetary authorities in small open econo-mies cannot control the money supply ex-cept when the exchange rate is floating—that is, when there is no change in theirinternational reserves. As long as there is abalance of payments deficit, the externalsector would have a contractionary impacton the money supply, as domestic currencyis sold to the monetary authorities. More-over, attempts to offset this effect by ex-panding domestic credit would, over time,

be frustrated by larger external deficits. Theexchange rate plays a central role in thisprocess because it influences the size of thebalance of payments deficit in foreign cur-rency as well as in domestic currencyterms. Given a balance of payments deficitin U.S. dollars, for instance, a more depre-ciated domestic currency would have agreater contractionary impact.

Due to insufficient data, econometricmodels are generally unavailable to facili-tate the design of a policy package, so pol-icymakers use national income and flow-of-fund identities, supplemented by statisticalestimates. Consistent financial flows aredeveloped within a monetary survey of theconsolidated accounts of the monetary au-thorities and the deposit money banks. Thetable represents a stylized monetarysurvey—omitting, to simplify the discus-sion, items such as the banking system'scapitalization, retained earnings, valuationgains and losses from exchange rate move-ments, and so on.

The double entry accounting principlesunderlying the survey require that assetsequal liabilities—net foreign assets plus do-mestic assets (or domestic credit) equalmoney and quasi-money or liquidity,which are liabilities of the monetary au-thorities or deposit money banks (that is,NFA + DC =L). This framework containsno behavioral assumptions and is thereforevalid regardless of the economic system orpolicy mix; an advantage that, combinedwith the reliability and timely availability ofmonetary data, also makes the monetaryaccounts a useful instrument to monitorprograms.

For this accounting identity to become auseful economic tool, a behavioral relation-ship—a stable demand for money—is in-troduced; the result is an equation that canbe employed to balance aggregate demandwith available resources. One of the first

Illustrative monetary survey(Outstanding stock at end of period; in millions

of domestic currency units)

Net foreign assets'Domestic credit

Net claims on govern-ment/public sector

Claims on private sectorLiquidity (money and

quasi-money)

Beginning ofprogramActual

50200

50150

250

End ofprogramTarget

30270

70200

300

1 For our purposes, the change in net foreign assets isequivalent to the balance of payments deficit.

tasks in designing credit policy is to esti-mate demand equations both for realmoney and real quasi-money (which in-cludes, for instance, savings deposits).Typically, these equations include, interalia, variables for real income and the op-portunity cost of holding money—the in-terest rate on bonds and/or the expectedinflation rate less the interest rate paid onquasi-money. With the parameter esti-mates from this equation plus targets forinflation and output, as well as for pro-jections for interest rates and inflationaryexpectations, the increase in nominal de-mand for liquidity is forecast. There aremany difficulties involved in making theseprojections, especially when the economyis experiencing significant institutional oreconomic changes. Nevertheless, somejudgment must be reached—not only forthis variable but others as well—if eco-nomic policies are to be formulated.

If domestic credit exceeds the amount ofmoney demanded, a balance of paymentsdeficit—a loss of net foreign assets—mayensue. In the short run, however, an excesssupply of money does more than produce a

28 Finance & Development /March 1983


deficit. Excess liquidity could also increasedomestic inflation and, given excess capac-ity, raise real growth, while a tightening ofcredit policy could, in the short run, lowerdomestic inflation, real growth, and em-ployment. The specific effects on output,employment, prices, and the balance ofpayments depend, of course, on the econo-my's structure and situation, as well asother policies, such as the exchange rate. Inthe table, the weighted average growth rateof nominal money and quasi-money is as-sumed to be 20 per cent. Were domesticcredit to rise above the target level (say to300), then some combination of lower netforeign assets (20)—implying a larger bal-ance of payments deficit—and higher liq-uidity (320) would result.

The planned change in net foreign assetsduring the program period depends notonly on the domestic response to policychanges but also on external conditions,and on receipts and payments that are ex-ogenous during the program period—debtservice, noncompressible imports, andcapital inflows associated with ongoing in-vestment. Export forecasts are based onprojections of demand in trading partnercountries, expected price developments ona commodity-by-commodity basis, and theestimates of the exportable surplus, whichincludes the effects of policies to stimulatesupply. Similarly, import projections arebased on forecasts of global inflation, do-mestic output growth, relative price devel-opments, excess liquidity in the financialsystem, and project implementation. Someitems, such as transportation and merchan-dise insurance, may be related to tradeflows, while others, such as tourism anddirect investment, may be estimated sepa-rately. Debt service projections use data onamortization and interest payments plus aforecast of international interest rates forfloating-interest-rate debt. Many develop-ing countries have exchange control regu-lations, employ debt management policies,or have limited access to international capi-tal markets that simplify the projection ofcapital flows. In countries with a well-developed financial system that is inte-grated into the global market, interest ratedifferentials and expected exchange ratemovements also influence these flows,making the balance of payments even moresensitive to domestic policy.

With agreement on the overall balance ofpayments target, the net foreign assets en-try is also determined. (This target may notbe equivalent to the change in net foreignassets defined in the monetary survey be-cause the latter includes banking systemflows of a medium-term to long-term matu-rity, which are customarily treated as partof the capital account in the balance of pay-

ments.) In the table, a balance of paymentsdeficit of 20 million domestic currency unitshas been envisaged. When exchange ratesare fixed, it is straightforward to go fromthe overall balance of payments deficit tothe net foreign asset target in domestic cur-rency. For example, at an exchange rate oftwo domestic currency units per U.S. dol-lar, the balance of payments deficit wouldbe US$10 million.

The level of domestic credit consistentwith the targeted change in net foreign as-sets and the projected demand for liquidityis obtained by subtracting net foreign as-sets from liquidity (DC = L — NFA), andwith appropriate seasonal adjustments,quarterly ceilings are also obtained. Natu-rally, the appropriateness of deriving a do-mestic credit target as a residual dependson how accurately the targets for net for-eign assets and liquidity reflect the econo-my's structure and prospective develop-ments. Hence, an understanding of theeconomy's behavioral relationships andtheir linkages to financial flows is necessaryto ensure that the policy package is prop-erly designed. In addition, because of thebalance sheet properties of the monetaryaccounts, movements in exchange ratesalter domestic credit by changing the do-mestic currency value of net foreign assets;therefore, when reporting to the Fundunder an arrangement, adjustments aremade in order to insulate domestic creditfrom unanticipated movements in ex-change rates.

The permissible increase in domesticcredit is also allocated between the privateand public sectors. The public sector's fiscaldeficit under the program can be financedfrom either foreign or domestic sources.Foreign financing obviously has implica-tions for debt management and the overallbalance of payments target, as well as fordomestic borrowing. The mix of domesticfinancing—borrowing by selling debt in-struments to domestic nonbank holdersand credit from the banking system—hasdifferent implications for domestic interestrates and the credit available to the privatesector. The domestic credit requirements ofthe public sector are derived by subtractingforeign and domestic nonbank financingfrom its overall deficit. Usually a subceilingon net credit to the public sector is set byadding this flow figure to the initial stock;the remaining domestic credit is allocatedto the private sector. If this amount ofcredit is insufficient to finance the pro-jected activity of the private sector, mea-sures to reduce the public sector's recourseto the banking system would be needed.The subceiling on credit to the public sectoris especially important since larger-than-programmed credit use, resulting from a

greater-than-targeted fiscal deficit, is oftenthe reason domestic credit ceilings arebreached.

Monetary authorities have differing de-grees of control over the domestic credit ofthe banking system. In some countries,they can control it directly through ceilingsassigned to deposit money banks. Wherethe relationship between liquidity andbanking system reserves (known as themoney multiplier) is stable and where theability of deposit money, banks to borrowand lend externally is limited, they can con-trol bank credit indirectly by limiting re-serve growth or increasing reserve require-ments. However, even though stable, themoney multiplier can still fluctuate un-expectedly and commercial banks may stillbe able to lend and borrow abroad to someextent and, thereby, weaken the monetaryauthorities' control over total domesticcredit. In such cases, the authorities mayprefer to have a ceiling set on the domesticcredit of the central bank, which they canmore readily control. Such a ceiling reducesthe risk that it would be accidentally ex-ceeded; however, because the linkage fromcentral bank credit to banking system creditis less stable, the attainment of programtargets may be more problematic.

As has been suggested, credit ceilings inFund-supported programs are a quantita-tive limit to ensure that the domestic creditextended, plus the monetary implicationsof the balance of payments target, willmatch the expansion in liquidity demand.These ceilings are thus an instrument usedto safeguard the economy against financialdisequilibria that could undo the adjust-ment achieved through budgetary, ex-change rate, incomes, and pricing policies.What happens when a breach of a creditceiling occurs? The immediate conse-quence is an interruption in the member'sright to draw on the Fund resources madeavailable to support the adjustment pro-gram. This does not mean, however, thatthe program has collapsed and the stand-by or extended arrangement is canceled.Rather, the first order of business is to de-termine why the ceiling was exceeded. TheFund is well aware that there are numerousuncertainties in designing a financial pro-gram and that a reversible shock may havetemporarily caused the credit ceiling to beexceeded. In such cases, a waiver of thatparticular ceiling is normally granted.However, if the deviation indicates a morefundamental problem with the program'sdesign or implementation, then either theprogram will have to be modified or an en-tirely new financial program will be nego-tiated before access to Fund resources canbe resumed.

G. Russell Kincaid



Coffee and cocoa trendsAn unfavorable outlook for developing countries. T. Akiyama and R. C. Duncan

Between 1977 and 1979, the value of coffeeexports from developing countries aver-aged US$11 billion, while cocoa exportsrealized $3 billion a year; together they rep-resented 7.5 per cent of all developingcountries' nonpetroleum exports. For 24developing countries, many of them low-income, coffee and cocoa exports bring inover 20 per cent of total export earnings.Although they fluctuate widely from oneyear to another, the export revenues theyearn (from cocoa beans as well as semipro-cessed cocoa products—butter, paste, andpowder) have increased substantially in thelast 20 years. These large increases in reve-nues were caused by rapid increases in de-mand, aided by rapid economic expansionand fast population growth worldwide un-til the mid-1970s.

However, since then, the demand forcoffee and cocoa has changed substantially.National income and population growthrates in the industrial countries—the maincoffee- and cocoa-consuming regions—have declined considerably, and do notseem likely to return to their previoushigher levels. As a result, prices of bothcoffee and cocoa have recently fallen. Toprevent them falling further, either de-mand must rise or coffee and cocoa produc-ers should reduce output growth.

Coffee: demand

One of the main causes of stagnatingworld coffee demand in recent years hasbeen the long-term decline in demand inthe United States, the most important con-suming country—followed closely by coun-tries in the European Community. TheUnited States accounts for 30 per cent ofworld imports. Data on per capita con-sumption in the United States show a netdecrease of 1.9 per cent a year over the last15 years. Besides the slowing down of in-come growth, an important factor causingthis decline in consumption appears to beprice competition from other beverages,such as cola drinks and milk (see table).

Consumption of fruit and other juices, andtea—cold teas in particular—has also affect-ed coffee consumption, and concern overthe side effects to health from coffee hasbeen a factor. A recent International CoffeeOrganization (ICO) study shows that in1980 the percentage of the U.S. populationover ten years of age that drank coffee wasdown 18.1 per cent compared with 1963;the corresponding figures for soft drinks,juices, and tea showed increases of 18.5 percent, 3.8 per cent, and 7.0 per cent re-spectively. This declining trend for coffee isparticularly strong among young persons,only one in four of whom are now regulardrinkers compared to one in three a decadeago. Apart from a decline in the percentageof people drinking coffee in the UnitedStates, there is also a decline in the numberof cups consumed by those who do drinkcoffee.

Western Europe, the second largest con-suming area, has been increasing its de-mand for coffee steadily for the past 20years. However, recent ICO studies of cof-fee consumption in the Federal Republic ofGermany and the United Kingdom indicatethat the growth in per capita coffee con-sumption in these countries is also slowingdown, while fruit juice and soft drink con-sumption is increasing rapidly. This devel-opment indicates that coffee demand inWestern Europe may decline as it has in theUnited States. If so, the prospects forgrowth in world demand for coffee becomevery limited. Although Japan, the centrallyplanned economies of Eastern Europe, the

U.S.S.R., and the developing countrieshave also increased coffee consumption,theirs is a small share—5 per cent or less—of total demand. Thus, even if their con-sumption continues to grow at a high rate,the impact on world demand will not besubstantial for many years.

Future growth in the consumption of cof-fee will depend primarily upon incomegrowth, population growth, changes in theprice of coffee, price changes in competingbeverages, and whether or not there is anyfurther erosion in demand as a result of theconcern over the side effects of caffeine in-take. The proportion of decaffeinated cof-fee in U.S. consumption has remained be-tween 12 and 14 per cent since 1974, whichmay indicate that any swing away from cof-fee due to the perceived health risk has alsostabilized.

The projections of world coffee demandthat follow have been made on the basis ofthe expected income and population trendsused for the World Development Report 1982.It is evident that the growth in world con-sumption, at least over the next decade,will depend heavily on consumptiongrowth in the United States and the Euro-pean Community. Population growth inthese countries is projected to be slow forthe next decade, so demand at the turn ofthe century will depend heavily on futuregrowth of income in these countries. Ifcompetition from other beverages con-tinues to depress coffee consumption in theUnited States and to affect, increasingly,coffee consumption in Western Europe,

Indexes of the retail

Coffee (roasted)Cola drinksMilk

1962

100100100

price of coffee and other selected beverages

(In 1978 constant dollars)

1975

162

132

1976

180136139

1977

314108140

1978 1979

261 171113 121— 88

Source: International Coffee Organization.—Indicates data not available.



total world growth in demand for coffeecould be less than 1 per cent per annum forthe 1980s. A less pessimistic view of coffeeconsumption in Western Europe allows forabout 1.3 per cent annual growth in de-mand from income and population growthworldwide during the 1980s.

SupplyCoffee is produced in a large number of

countries in Latin America, Asia, andAfrica. Latin America is the dominant

steadily increasing. The large-scale plant-ings in the wake of the high prices of 1976and 1977 are only now starting to bear fruit.There has also been an increase in the areaplanted to coffee in Brazil, with a shiftnorthward in Brazil's coffee-growing re-gion. Since coffee has proved a good pio-neering crop, large plantings have beencarried out in the Amazon Basin and north-west area. Though well over half of all treesare still in the frost susceptible areas of theSouth, the likelihood of damage from frost

the decade. This prognosis for Indonesia isin sharp contrast to projections of two yearsago when its output was expected to growto about 7.5 million bags by 1990.

PricesDuring the 1960s coffee prices fell and,

measured in real terms, did not rise untilafter the 1975 frost in Brazil, when its pro-duction was more than halved and coffeeprices quadrupled. Since 1979, prices inreal terms have fallen again to levels at or

coffee-producing area, but its share ofworld production has been declining (fromabout 78 per cent in 1961 to about 65 percent in 1980) mainly due to falling Brazilianproduction since the early 1960s (Chart 2).During the same period, African producersincreased their share from 17 per cent to 23per cent and Asian producers went from 4per cent to 10 per cent.

During the 1960s, Brazil followed a policyof coffee tree eradication in a unilateral ef-fort to prop up coffee prices after the col-lapse of the postwar boom. At the begin-ning of the program it was producing up to50 per cent of world output and held stocksequivalent to more than one year's totalworld consumption. As a result of the pro-gram, the number of coffee trees in Brazilfell consistently from 1962-63 to 1970-71.However, soon afterward, the tree eradi-cation program was abandoned, in light ofthe fact that its actions only served to re-duce substantially its share of the worldmarket.

During the 1970s, Brazil pursued a policyof providing incentives for coffee planting.The results of this policy show up in thevigorous growth in the number of coffeetrees planted—from 2.2 billion in 1970/71 to3.4 billion in 1981 (about the same numberas in 1964/65). With the planting of high-yielding varieties, average yields have been

has been lessened by this northwardmovement.

It is the Brazilian Government's declaredintention to maintain in the medium term aproduction capacity of 28-30 million bags.Information is not complete, but estimatesmade by the authors on the basis of treenumbers and likely yields indicate that Bra-zil's average annual production in the 1980scould exceed 30 million bags (preliminaryestimates of the 1983/84 crop range from 29to 35 million bags). This would mean a sub-stantial increase over Brazil's recent aver-age annual output, which has been about23 million bags, and translates into an 8 percent increase in world output. If such anincrease occurs it could only be offset byfairly large decreases by other majorproducers.

Two other important producing coun-tries, Colombia and Indonesia, burdenedby large stocks that have resulted fromproduction in excess of ICO export quotasand other sales, are reported to have re-cently decided to halt expansion of output.Colombia's annual production is expectedto decline from the current level of 14 mil-lion bags to about 12 million bags by theend of the current decade. Indonesia's an-nual production, too, is projected to de-cline gradually from the current 5 millionbag level to 4.5 million bags by the end of

below those realized in the 1960s (Chart 3).The results of analysis by the authors showthat real prices of coffee are expected todecline further from their present low lev-els in anticipation of substantially in-creasing supply. In the absence of severeproduction upsets, they are not expected torecover until the early 1990s. Even then,the increase in prices in real terms is notexpected to be substantial. Consumptionwill respond temporarily to the lowerprices in the next few years, but return toits slow secular trend thereafter.

Cocoa: demand

World consumption of cocoa beansreached a high point of 1.6 million tons in1973. It subsequently fell to about 1.3 mil-lion tons in 1979 in response to the highprices caused by the sharp decline in pro-duction in 1977 as a result of bad weather inWest Africa. There has been some recoveryin consumption in the period since 1979due to the sharp decline in prices since the1977-79 boom. However, since 1979 thegrowth in consumption has been consid-erably less than the increase in production,leading to steadily increasing stocks. Forthe year 1981-82 cocoa stocks are expectedto be about 672,000 tons, compared with278,000 tons in 1976-77.

finance & Development iMarch 1983 31

Chart 1Coffee consumption by major regions,

1961-80

Chart 2Coffee production: world and Brazil

1949-81

Chart 3

Coffee prices: yearly averages1950-81


Cocoa beans are initially processed intococoa liquor, butter, cake, and powder—each a separate stage in processing. Theseproducts are subsequently used as raw ma-terials in a variety of products, but the mainuse, by far, is for chocolates and other con-fectionery. The major consuming regionsshown in Chart 4 are Western Europe(about 40 per cent), North America (27 percent), and the centrally planned economiesof Eastern Europe and the U.S.S.R. (16 percent). Consumption has changed very littlein Western Europe since 1965—if anything,the trend is slightly downward (Chart 4). Inthe United States there has been a markeddecline since 1972, associated in part withthe high prices in the period 1977 to 1979.The slow income growth in the industrialcountries in the latter part of the 1970sseems to have been largely responsible forthe slow or declining trend in world con-sumption in that period.

In fact there has been little growth in thecocoa market since the mid-1960s; theshares lost by the Western European andNorth American markets have been takenup by the centrally planned economies.The tetter's share grew rapidly from 12 percent in 1965 to 23 per cent of the worldmarket in 1975. However, with the highprices and slower economic growth of re-cent years, the centrally planned econo-mies' imports have tumbled—down to anestimated 16 per cent of world consump-tion in 1980. Developing countries as agroup are minor consumers of cocoa prod-ucts. Moreover, their total consumption

has shown no indication of a substantialincrease.

The prospects for growth in cocoa de-mand in the major consuming countries donot look good. The population and incomegrowth projections in the World Develop-ment Report 1982 for the important marketsof North America and Western Europe in-dicate that consumption growth in these

areas will remain as depressed as it hasbeen for the past decade or so. Incomegrowth prospects in the European centrallyplanned economies and the U.S.S.R. arenot optimistic either. The demand for im-ports by the European centrally plannedeconomies and the U.S.S.R. is very re-

sponsive to price and, given current lowprices, may well return to import levels ofthe mid-1970s in spite of falling populationand income growth. However, this is theonly likely area for growth in the cocoamarket. Some increase in cocoa demand inNorth America and Western Europe canalso be expected, if prices fall. Consump-tion growth in the United States has proba-bly been stimulated in the past two yearsbecause of the appreciating U.S. dollar.Our analysis within the framework of aworld model of the cocoa market indicatesthat the growth in world demand for im-ported cocoa arising from income and pop-ulation growth will be about 2 per cent ayear throughout the decade.

Supply

World cocoa production is concentratedin a few developing countries. Brazil, theIvory Coast, Ghana, Nigeria, and Cam-eroon account for about 80 per cent of totalproduction. However, Ghana and Nigeriahave been losing their share of the market—their production actually declining in re-cent years. In Nigeria's case this has beendue to the rise in both the value of the nairaand in wage rates from its burgeoning oilexports since 1973; in Ghana it has beendue to an overvalued exchange rate, ad-verse domestic pricing policies—both ofwhich have drastically reduced incentivesfor Ghanaian farmers to produce cocoa—aswell as to plant disease and depreciatinginfrastructure. Brazil and the Ivory Coastare now the major cocoa producers, hold-


Chart 4

Cocoa consumption by major regions,1961-80

International agreements

International agreements art currently in force for both coffee andcocoa. The International Cufftc Agreement uws established m 1976with the aim of stabilizing toffee prices, using export quotas. Membersof the Agreement include countries accounting for 99 per cent of iivrldexports and 90 per cent of world imports. In September eaih year,before the beginning of the coffee year, members of the Agreementdecide on the global export uuota (an wtll as each exporting country'sauota) and the price range within which to operate. The global quotais based on projections of consumption, production, and stocks for eachcoffee year. Quotas arc adjusted within the year in an attempt tumaintain prif.es within the agreed price range. The export quota pro-visions were put intu effect most recently in October 198(1, follou'ingthe fall in prices beginning June 1980. The level of the global quota uvsprogressively reduced, as prices kept falling until fuly 1981—a tight-ness which led to a buildup of stuck* and heavy competition in non-quota markets. Prices rose in August 1981 and have been somewhatstranger since, but this may have been due mainly to the were frost

that hit the Brazil crop in July 1981. The new Agreement, which is tobe implemented in 1983 with little change from the current Agree-ment, will be in force until September 1989.

ihe International Cocoa Agreement, which came into fane in Au-gust 1981, uses buffer stocks in an effort to stabilize prices. The BufferStock Manager UBS provided with initial funds carried over from theprevious agreement. Additional funds are accrued from levies on eitherexports ur imports of the Agreement members. Many exporting andimporting countries are not members of the existing Agreement, in-cluding the largest producer, the Ivory Coast, and the largest con-sumer, the United States. Faced with increasing stocks and decliningprices, the buffer Stock Manager has been purchasing cocoa wntcAugust 1981 but has not succeeded m raising cocoa prices to the"floor" leivl of Ihe agreed price range. The 1981-82 cocoa crop, forinstance, is expected to exceed 7.7 million tons, about 10 per centhigher than in the prci'ious year and the fifth cemsecutiiv year ofsurplus: stocks are expected to be about 700,000 tons us a result—anincrease of 140 per cent mvr -stocks in 1976-77. Additional funds of$75 million have recently been approved for buffer stock purchases andmade mrailahle through commercial bank loans.


ing shares of 22 per cent and 24 per centrespectively, in 1980-81.

Cocoa production is increasing as it didbetween 1957 and 1964. During this period,world production increased at 7.8 per centa year, stocks rose fourfold and priceshalved. Between 1964 and 1976, productiondeclined and real prices rose again to levelscomparable with the early 1950s (for pricemovements see Chart 5). Since 1976, worldproduction has increased rapidly—at anannual rate of 5 per cent—with Brazil andthe Ivory Coast achieving individualgrowth rates of 10 per cent and 13 per cent,respectively. Further increases in worldproduction seem likely, coming from theplantings induced by the 1977-79 priceboom. Cocoa trees take 3 years to reachproduction, and, for the new high-yieldingvarieties, only 7 years to mature production(compared to about 14 years for oldervarieties).

There are indications that many majorproducers are increasing cocoa productionsubstantially. It has recently been reportedthat Brazil plans to increase its cocoa pro-duction from 340,000 tons to 550,000 tonsby 1992. Other countries also seem likely toincrease production substantially. Withoutany further plantings, production in theIvory Coast could increase by as much as100,000 tons by 1990 due to the maturationof the large plantings made in recent years.South East Asia could also be the source ofa further large increase. Production costs inthis region are estimated to be considerablylower than in other regions and thus cocoais still profitable to growers even at currentlow world prices. Malaysia, the Philip-pines, Indonesia, and Papua New Guineacould therefore well increase cocoa produc-tion for this reason. The area planted tococoa in Malaysia is estimated to have in-creased by more than 200 per cent since1973. In the light of recent and plannedplantings we estimate that Malaysia's pro-duction could increase by more than 100per cent in the next five years to about110,000 tons, and could increase further toabout 130,000 tons by the end of thisdecade.

Relief from this rapidly increasing supplyis likely to come from further reductions inGhana and perhaps Nigeria—though thereis room for doubt in both cases. Ghana maybe able to improve the collection and in-ternal transport of cocoa and therefore in-crease the quantity coming onto the mar-ket; with improved policies, low laborcosts, and abundant land suitable for co-coa, Ghana could readily reverse its down-ward production trend. In an effort to re-vitalize cocoa export earnings to replace thedecline in oil exports, prices now being of-fered to cocoa producers in Nigeria are ac-

tually higher than international prices.However, Nigerian farmers can earn morefrom producing scarce food crops thanfrom producing cocoa for export. If Ghanaand Nigeria hold output constant, worldsupply is likely to increase 40 per cent by1995; should they revert to previous pro-duction levels, supply would increase by 60per cent by 1995.

Prices

Simulations of the world cocoa modelshow the present low real prices continuinguntil the late 1980s, when some recoverywill take place. This recovery is contingenton the expectation that the low prices fore-cast over the next few years will discouragegrowth in the major producing countries inthe longer term. However, if productionthroughout the present decade expands inBrazil, South East Asia, and some other

T. Akiyamafront Japan, joined theWorld Bank in 1981 and iscurrently working in theCommodities and ExportProjections Division. Mr.Akiyama had previouslyworked for UNCTAD.

R. C. Duncanfrom Australia, joined theWorld Bank in 1980 and isat present Division Chief,Commodities and ExportProjections Division. Heworked for the IndustriesAssistance Commission inCanberra prior to joiningthe Bank.

countries at rates that are being proposedat present, cocoa prices will fall even lowerthan currently forecast.

Outlook

Any projections regarding coffee andcocoa are necessarily subject to numerousuncertainties—the price elasticity of de-mand may be higher in the long run thananticipated, and that of supply could alsobe much higher. However, on the basis ofthe authors' evaluation, coffee and cocoaface the prospect of increasing supply andslow growth in demand. Growth in de-mand for these commodities is expected tobe low worldwide due to the expected loweconomic and population growth in themajor consuming regions, the industrialcountries, and also to the fact that per cap-ita consumption levels are reaching satur-ation in many of these countries. In-creasing supplies have already resultedfrom the large-scale plantings that werecarried out in the second half of the 1970sas a consequence of very high prices. More-over, from what is known about plantingintentions, the widespread trend towardincreasing production is disconcerting inthe light of current and likely prices.

International agreements for both coffeeand cocoa have sought to put a floor underprices. However, such agreements cannotbe effective if oversupply persists. Thesecommodities have low price elasticities, soprices have to decrease substantially to in-crease consumption at all. Using widelyagreed medium-term price elasticities it isestimated that for world consumption to goup by 1 per cent, prices have to be loweredby about 3 per cent and about 4 per cent forcocoa and coffee, respectively. Thus, byproducing more of these commodities, theexport revenues of the producing countriesas a whole may go down.

Because of their foreign exchange earn-ing capacity and high labor intensity ingrowing and harvesting, coffee and cocoaproduction are attractive for countrieswhose other sources of foreign exchangeearnings and employment are seen to bescarce. However, the only way to improvethe gloomy outlook is for the uneconomicalproducers to reduce production growth bydiversifying out of coffee and cocoa. Pro-ducing countries could make profitableinvestments within the cocoa and coffeesectors, without increasing supply, partic-ularly in efforts to improve bean quality(through improved fermentation and dry-ing in the case of cocoa) as well as in furtherprocessing of cocoa beans. It has been esti-mated that some producing countries couldincrease the unit value of cocoa bean ex-ports by up to 15 per cent through im-proved fermentation and drying. ED


Chart 5

Cocoa prices: yearly averages,1950-81


Economic impact of defenseWorld military expenditures have beenrising rapidly, increasing by over 22 percent in real terms in 1970-79, even as therate of economic growth slowed in both de-veloped and developing countries. Devel-oping countries' expenditures on defensein the same period rose in real terms fromUS$72.6 billion to $118.7 billion—an in-crease of 63 per cent—while their share ofworld military expenditures rose from 17.1to 22.5 per cent. As a proportion ofgross national product, however, militaryexpenditures in developing countriesdropped slightly in 1970-79 from 5.8 to 5.5per cent.

Data on military and defense expendi-tures are not uniformly available on allcountries. Nevertheless, an establishedbody of information (see Related reading)unmistakably points to a rising trend ofsuch expenditures in developing countries,which can be examined against the back-ground of their economic development.

Most of the developing world as weknow it today emerged as independent na-tions in the aftermath of World War II and,in the case of Africa, after 1960. In the twodecades following that date there was aproliferation of wars and civil disturbancesworldwide, often with costly economic af-

tereffects, and most of these occurred inthe developing world (Sivard, 1980).

Defense expenditures have long beenconsidered necessary, indeed beyondquestion in many countries. Efforts to con-trol or even eliminate them have an equallylong, though less successful, history. Apartfrom their strategic or political implica-tions, defense expenditures have an obvi-ous economic impact that can be particu-larly acute in the developing world, whereresources are scarce and therefore the op-portunity cost of such expenditures is high.This was especially true during the decadeof the 1970s—one of significant economicturmoil, with the economies of the devel-oping world adversely affected by devel-opments beyond their control, includingthe sudden and substantial rise in the priceof energy and the slowing down of growthin the industrial world.

The debate over the economic impli-cations of military expenditures continues—most recently in the United Nations Sec-ond Special Session on Disarmament in thesummer of 1982. However, no clear agree-ment has emerged about the nature andextent of their economic impact.

Many supporters of military expendi-tures justify them not only on the grounds

World military expenditures, 1972, 1976, and 1981(In millions of U.S. dollars, at 1979 prices and exchange rates)

1972

World total 416,304Developed market economies 238,596Centrally planned economies [143,805]Organization of Petroleum

Exporting Countries 1 1 , 1 27Non-oil developing countries

With 1978 GNP per capitabelow US$300 6,333

With 1978 GNP per capitabetween US$300-699 4,695

With 1978 GNP per capitaabove US$699 10,825

Total non-oil developing countries 21,853South Africa 1 ,326

1976

456,045232,980[157,869]

32,934

(6,390)

8,002

16,62031,0122,815

1981

518,727262,137[173,652]

[46,220]

[7,687]

[7,090]

(20,284)35,061

Source: Stockholm International Peace Research Institute, The SIPRI Yearbook of World Armaments andDisarmament, 1982, Taylor and Francis Ltd., London, U.K., p. 140.

Note: Details may not add to totals because of rounding.— Indicates information not available or not applicable.( ) Indicates SIPRI estimates.[ ] Indicates imputed values, with a high degree of uncertainty.

of national security and stability but also ineconomic terms. Defense industries areconsidered seedbeds of technical knowl-edge and skills that are passed on, eitherdirectly or indirectly, to the civilian sector.The military and ancillary services are saidto provide employment and training forlarge numbers of people. Further, the useof military personnel in national health andeducation schemes is given as an exampleof the participation of the military in insti-tution building in the developing world. Anumber of countries in Africa and Asiahave attempted to put such schemes intoeffect.

Some social scientists attribute a majorrole to the military in the modernization ofdeveloping societies. They believe that themilitary is often the leading sector of theeconomy in contact with modern tech-nological advances and that it impartstraining to its personnel in handling so-phisticated equipment. This experience issaid to be transmitted to both the rural sec-tor, where most soldiers originate from andreturn to, and the urban sector, wheremost military organizations operate.

An often quoted major study of the rela-tionship between expenditures on defenseand economic development (Benoit, 1973)provides evidence that even though de-fense expenditures outpaced economicgrowth in 1950-64 in 44 developing coun-tries, their expenditures on defense did nothave a net adverse effect on growth. In fact,the study concludes that in these countries"the more they spent on defense in relationto the size of their economies, the fasterthey grew—and vice versa." Another bene-fit often cited by supporters of defenseexpenditures is the creation of basic infra-structure, such as transport and commu-nication networks.

The basic criticism that is leveled againsthigh defense expenditures is that they re-duce the total resources available for eco-nomic development in poor economies.The growing need of developing countriesfor both domestic and foreign resourcescould be met, it is argued, by freeing someof the current defense allocations, es-pecially where economic difficulties de-mand major structural adjustments. Criticscharge that defense sector expenditurescomplicate the task of adjustment since



expendituresthey escape analysis and scrutiny while us-ing up economic resources.

The high import content of the defensesector is another major cause for criticism.In developing countries as a whole, armsimports represent about 5 per cent of totalimports. Where such imports are financedby external loans, they add to the risingburden of external debt. And, to the extentthey are paid for by export earnings, theyabsorb resources that could have been putto productive alternative uses.

Opposing the view that the defense sec-tor provides employment, and thereforeincome, critics often suggest that defenseexpenditures transfer income to a relativelysmall, privileged group in the cities. Themajority of people of the developing worldlive in rural areas and therefore do not ben-efit from the activities of the defense sector.The modernizing effects of the military, it isasserted, mainly stem from training for theuse of highly sophisticated military hard-ware and weapons, most of which have noapplication in the civilian economy. Theeducation of military personnel to handlesuch equipment and to work within mil-itary organizations does not necessarilymake them valuable additions to the poolof technical talent available within a devel-oping country. Critics point out that it isdifficult to transfer the knowledge of, say,

the operation and maintenance of sophisti-cated fighter aircraft to any sector of thecivilian economy in a developing country.

The suggestion of a positive correlationbetween military expenditures and eco-nomic development in some developingcountries does not, critics assert, establish acausal relationship between the two. TheBenoit study, for instance, covered a periodof large-scale foreign economic aid thatmay have compensated for the diversion ofresources toward defense. Further, theseheightened aid flows may have been a ma-jor factor in the growth of the countriesstudied. A more recent study, of 69 devel-oping countries over the 1950s and 1960s,indicates that growth of military expendi-tures reduced the rates of growth, invest-ment, and agricultural production. A 1 percent increase in the military's share of grossdomestic product (GDP) was associatedwith a 23 per cent and 18 per cent drop inthe respective shares of investment and ag-riculture in GDP (Taylor, 1982).

The provision of basic transportation andcommunication infrastructure, often citedas a benefit of defense expenditures, couldjust as easily be justified on purely eco-nomic grounds, according to critics. A ruralroad, they assert, could, in fact, be justifiedon the basis of improved transportation ofrural produce to urban markets and the bet-

Average defense and other central government expendituresin non-oil developing countries, 1976-80

(Per cent of total expenditure)

Other(including administration)-^

Education £—

Health

Housing

Economic services-

Social securityand welfare

Defense

Othertransportation and

communications

Roads

-ElectricityAgriculture

Source: IMF, Government Finance Statistics Yearbook, 1982.

ter flow of goods and services in the otherdirection. There would still be an addi-tional spinoff to the military if it requiredsuch a road for strategic purposes.

The entire debate over defense ex-penditures and their economic costs comesdown to the twin questions: how much de-fense does a country need and how muchcan it afford? The answer does not lie in thearea of economics alone but is related tomany national and international consid-erations of social, political, and economicimport. In the final analysis, it is the pre-rogative of individual governments toassess both the costs and benefits of risingexpenditures on the defense sector and todecide, according to their priorities, theproper allocation of resources between de-fense and other needs.

Snuja Nawaz

Related reading

Emile Benoit, Defense and Economic Growth inDeveloping Countries (Lexington Books,Lexington, MA, U.S.A., 1973).

David Dabelko and James M. McCormick,"Opportunity Costs of Defense: SomeCross-National Evidence," Journal ofPeace Research (Vol. XIV, No. 2, 1977).

Mary Kaldor, "The Military in Devel-opment," World Development (Vol. 4,No. 6, 1976).

Olof Palme, et al., "Military Spending: TheEconomic and Social Consequences,"Challenge (September-October 1982).

Sabi H. Shabtai, "Army and Economy inTropical Africa," Economic Developmentand Cultural Change (July 1975).

Ruth Leger Sivard, World Military and SocialExpenditures 1980 (World Priorities, Lees-burg, VA, U.S.A., 1980).

Stockholm International Peace Research In-stitute, The SIPRI Yearbook of World Arma-ments and Disarmament, 1982 (Taylor andFrancis Ltd., London, U.K., 1982).

Lance Taylor, Military Economics in the ThirdWorld, paper prepared for the Inde-pendent Commission on Disarmamentand Security Issues, 1982.

World Military Expenditures and Arms Trans-fers, 1970-1979 (U.S. Arms Control andDisarmament Agency, Washington, DC,U.S.A., March 1982).

David K. Whynes, The Economics of ThirdWorld Military Expenditure (University ofTexas Press, Austin, TX, U.S.A., 1979).



Private sector petroleum explorationin developing countries

Why did private sectorpetroleum exploration notincrease as much asanticipated in the 1970s?

Keith PalmerWhile real oil prices rose dramatically dur-ing the 1970s, there does not appear tohave been a corresponding increase in pe-troleum exploration in the oil importing de-veloping countries. Total exploration ac-tivity in these countries, measured by thenumber of exploratory wells drilled, was nohigher in 1979 than in 1971, while in theUnited States over the same period it in-creased by about 50 per cent. The samepicture emerges if seismic crew-monthsrather than wells are used as an index ofactivity. This apparent lack of response indeveloping countries cannot be explainedin terms of an absence of geologic poten-tial. A Bank-sponsored survey of oil andgas potential in oil importing developingcountries concluded that of 70 countriesevaluated, 38 had very high, high, or fairpetroleum potential, and of that group only8 countries were considered adequatelyexplored.

Before accepting that private sector ex-ploration has not been responsive to oilprice increases, a number of caveats shouldbe considered. First, as the chart indicates,over half the exploratory drilling in oil im-porting developing countries was under-taken by the government-owned nationaloil companies of a small number of largecountries (notably Argentina, Brazil, andIndia), so trends in exploration in the pub-lic sector are as important as those in theprivate sector. Although reliable data arenot available, fragmentary evidence sug-gests that, in fact, both public and privatesector exploration have been relativelystagnant over the decade, although for dif-ferent reasons.

Second, available data refer to total ex-ploratory wells rather than the more rele-vant "new field wildcats" (exploratorydrilling in areas with no proven petroleum,for which data are not available). Since byfar the largest volume of exploratory activ-ity takes place within producing basins,

these data mask the trends in explorationin frontier areas which are more relevantfor the majority of the oil importing devel-oping countries. Finally, private sectorexploration responds to higher prices onlywith a considerable lag, because of the timerequired to acquire exploration rights, eval-uate existing data, and mobilize equip-ment. Since the second oil price "shock" of1979-80, a modest increase in exploratoryactivity has occurred. Although over halfthis increase is attributable to higher na-tional oil company exploration (in Argen-tina and Brazil), it is at least possible that alagged increase in private sector explora-tion is underway.

Despite these caveats, it appears that pri-vate sector exploration in the oil importingdeveloping countries has not increased at

the rate previously anticipated nor to theextent observed in developed countries,particularly the United States. It is alsoclear that foreign oil company participationin oil importing developing countries hasbeen dominated by a few very large multi-nationals. This is relevant to the analysis ofconstraints on private sector exploration inthese countries. Two types of constraintsappear to exist: those originating in devel-oped countries and those originating in theoil importing developing countries them-selves.

Bias in developed countries

The existence of special tax benefits forincome arising from domestic petroleumproduction for both U.S. companies andindividual shareholders is a major subsidy

principal

Oil1

ExplorationDevelopment

Gas2

Table 1Oil importing developing countries:

investment requirements in oil and gas sector, 1980-90(In billions of 1980 U.S. dollars)

Estimate1980

0.52.11.03.6

Projectedannualaverage1981-85

1.02.51.75.2

ProjectedProjected average annual

annual percentageaverage growth rate1986-90 1980-90

1.5 11.63.2 4.32.7 14.27.4 30.1

Source: World Bank, Energy in the Developing Countries, 1980.1 Based on the investments required to develop oil production from 2.0 million barrels of oil a day in 1980 to

3.6 million barrels of oil a day in 1990.2Based on the investments required to raise gas production from 1 .5 million barrels of oil equivalent in 1 980

to 2.5 million barrels of oil equivalent in 1990.

Table 284 net oil importing developing countries:

estimated remaining ultimate recoverable oil reserves1

(In millions of barrels)

Non-oil producing TotalUltimate Ultimate

recoverable recoverablereserves Proven reserves

Africa 13,730 0 13,820Asia 8,375 0 14,025Latin America 5,510 0 22,185Middle East 1,515 0 3,120

29,130 0 53,150Proven reserves

(In per cent of ultimaterecoverable reserves) 0

Proven

101,7904,950

4507,200

14

Source: Bureau d'Etudes Industrielles et de Cooperation de I'lnstitut Francais du Petrole, PetroleumResources of Oil Importing Developing Countries, 1 975.

'Net oil exporters in Latin America were not included in the study.



for domestic U.S. petroleum investmentbut it discourages exploration overseas, in-cluding in the developing countries. TheUnited States is the major potential sourceof oil exploration capital in the developingcountries. In recent years, progressivetightening of tax rules for the overseas pe-troleum income of United States cor-porations has increased this tax bias in fa-vor of domestic petroleum investment. Taxbenefits include the widespread use of"drilling funds," which permit individualsto deduct from personal income a highshare of exploration costs in the year in-curred (and until very recently at marginaltax rates up to 70 per cent). The drillingfund is probably the single most importantexplanation for the unique presence ofmany small exploration firms in the UnitedStates.

At the corporate level, discriminatory taxbenefits include the retention for mostcompanies of a substantial depletion allow-ance, which effectively shields 22 per centof gross domestic petroleum income fromtaxation, and favorable income consolida-tion rules, which permit the deduction oflosses incurred in domestic petroleumoperations from domestic nonpetroleumincome. A recent study showed that theeffective corporate tax rate on petroleumincome from United States operations isless than 15 per cent. A neutral tax systemwould impose the same effective tax rate onincome from all sources, whether derivedfrom domestic or foreign operations.

Constraints in LDCs

Four main types of constraints within oilimporting developing countries can be dis-tinguished: small fields; government poli-cies; gas-prone areas; and political risk.

The global distribution of petroleum fieldsizes is highly skewed. While there aregood prospects for economic discoveries inthe developing world, there is a high prob-ability that many of the fields encounteredwill be small by the standards of the majormultinationals. It is widely acknowledgedby oil industry representatives that the ma-jor oil companies are primarily interested inareas with potential for large fields—in ex-cess of 500 million barrels of oil—and haveno serious interest in exploring for oil fieldswith reserves of less than 50 million barrels(although they may, nevertheless, developsmall fields discovered while searching forlarger ones). Nor are they interested in ex-ploring for gas fields with reserves of lessthan 3-5 trillion cubic feet—the minimumreserves required for a liquified gas exportproject to be economic. Yet, given the his-torical distribution of field sizes, it can beexpected that over 70 per cent of fields dis-covered in developing countries will be in

the smaller range. From the country's pointof view, at present costs and prices, thesesmall fields can make an important con-tribution to net foreign exchange earnings.For a "typical" range of exploration anddevelopment costs, fields considerablysmaller than 50 million barrels can generatean attractive economic rate of return (afterallowing for the risk of sinking a "dryhole").

If this is the case, why do private compa-nies not search for these fields? The con-

Who does the explorationin developing countries?

(Average data for 70 developing countries in 1979-80)(In per cent)

National oil companies (developing country) 60

National oil companies (developed country) 10

Small majors2 6

arge independents3 2

mall independents" 2

Small i

Large

I

10 20 30 40 50 60

Source: Petro Consultants.1Shell, British Petroleum, Mobil, Texaco, Chevron,Exxon, and Gulf.Fortune 500 assets over US$1 billion.

3Fortune 500 assets US$100 million—$1 billion.'Fortune 500 assets less than US$100 million.

centration of private sector exploration indeveloping countries in a few large firms(see the chart) has a bearing on this ques-tion. The few large firms engaged in in-ternational exploration have limited man-agerial resources; they seek to limit theirexposure in developing countries and fron-tier areas, and ration exploration expen-ditures in line with current and expectedinternal cash generation. Thus they under-take only a limited set of the potentiallyprofitable investments. While from a cor-porate viewpoint it is rational to invest lim-ited resources in the search for large fields,profitable opportunities of vital importanceto the developing countries get deferred.

A related point is that small fields, es-pecially those in large countries, are lesslikely to produce an exportable surplus.The major oil companies and the developedcountry state oil companies are anxious notonly to make a profit but also to securephysical access to crude to maintain secu-rity of supply for their downstream oper-ations and home countries, respectively.For the majority of smaller North Americanindependent oil companies, used to discov-

ering and developing fields in the range of1 to 10 million barrels, the "small fields" indeveloping countries are potentially profit-able. However, their participation remainsvery minor, despite some increase in recentyears. Smaller companies appear to be de-terred by the large scale of internationalexploration expenditures relative to theircash flow, the tax benefits available for do-mestic exploration, and concerns about po-litical risk in developing countries.

Government policy in developing coun-tries can be a constraint to private sectorexploration—either because acreage is notavailable to the private sector or becausethe terms and conditions on which it isavailable are unacceptable when comparedwith alternative investment opportunities.In most oil importing developing coun-tries, public sector petroleum investmentbroadly coincides with national investmentbecause, given the huge capital require-ments of the industry, the only feasible al-ternative to foreign investment is oftenpublic sector investment. The presence ofstate oil companies transcends ideologicalboundaries and is often more a manifesta-tion of nationalism than socialism. (This istrue for most European developed coun-tries too.) While it is understandable thatgovernments of developing countries wishto avoid total dependence on foreign com-panies, in the 1970s state oil companies of-ten held exploration rights over areas wellin excess of what they could hope to ex-plore in the near future. Often they werenot subject to the same requirements ap-plied to private investors to explore an areaor relinquish it, and undoubtedly part ofthe supposed failure of private companiesto respond to new opportunities was, infact, a reflection of the lack of access to themore promising areas.

The fiscal terms in petroleum contractsmay also have deterred private sector ex-ploration in some cases. In response to thelarge oil price increases, countries raisedsubstantially the government share ofproject revenues from royalties, taxes,production shares, and so on to obtain forthe state a larger share of the resource rentand to limit the "windfall gains" made byinvestors. However, the structure of the fis-cal terms in many cases not only limitedwindfall gains but also made it unprofitablefor investors to explore for small fields.

Many basins in oil importing developingcountries have characteristics that suggestthat gas rather than oil may be found. Re-serves have to be very large if liquified nat-ural gas schemes are to be economic, andeven then the capital requirements of suchschemes are so vast that only a minority ofvery large oil companies could contemplatethem. Smaller gas reserves may well be



economic if a domestic market can befound, but private oil companies haveshown little interest in exploring for gasfields where the output would be soldwithin the developing country.

This reluctance to become involved ingas development for the domestic marketin developing countries results from sev-eral factors: the large financial commitmentin distribution and marketing infrastruc-ture, which has low returns on investmentand a long payback period; the high politi-cal risk involved in having an investmenttied to a single (usually government) buyer;the high market risk associated with saleslinked exclusively to the growth of nationaldemand; the absence of clear ex ante gaspricing rules; and the problem of access toforeign exchange to ensure remittance ofearnings. Some of these problems—thoserelating to pricing and access to foreignexchange—can be alleviated through re-vised petroleum contracts; others—such asinfrastructure costs—can be reduced by thefinancial involvement of multilateral lend-ing agencies. But an important number ofsmall economic gas prospects are unlikelyto be sufficiently interesting to be exploredor exploited by the major oil companies,even with contract revisions.

It is widely contended that corporate per-ceptions of high political risk are a seriousimpediment to exploration in the oil im-porting developing countries. The evi-dence suggests, however, that the major oilcompanies, while welcoming measures tostabilize relations with governments, donot see political risk as a fundamental im-pediment to involvement in developingcountries. In part, this is because they candiversify such risk among countries just asthey diversify their geologic risk by ex-ploring in different areas simultaneously.The greatest concerns about political riskare, in fact, expressed by the smaller, pri-marily North American, independent com-panies who have less scope to diversifyacross countries and less experience indealing with governments in the develop-ing world.

It is sometimes suggested that devel-oping countries should respond to com-pany perceptions of political risk by offer-ing more attractive fiscal terms. If this isinterpreted to mean a low governmentshare of profits whatever the size of thefuture discovery, however, that is likely toincrease rather than decrease political risk.Incentive terms following an unexpectedlyprofitable discovery are inevitably widelyperceived as unfair and accelerate demandsfor renegotiation of terms. A preferable re-sponse to political risk is to reduce the vari-ance of the expected return and to acceler-ate payments to the investor in the early

years of production, thus reducing the pe-riod during which the company is at risk.The implication is that the governmentshould adopt a progressive fiscal systemthat accelerates company cash flow andlimits the government profit share in theearly years, but then increases that shareprogressively thereafter as actual field prof-itability increases.

Removing the constraints

In the developed countries, special taxbenefits that deter investment in the devel-oping world could be revised to improveboth tax efficiency and equity. In the devel-oping countries, improved governmentpolicies could accelerate exploration by en-suring that all oil companies—includingthe national oil companies—are subject tospecific requirements to explore areas un-der license within a specified period or re-linquish them, thereby increasing access toand competition for inactive acreage.

Increased exploration for small fields canalso be encouraged through improved taxdesign. Typically, fiscal terms have tendedto emphasize capturing revenues at the ex-pense of exploration for small fields. Better-designed fiscal structures are needed, thatminimize the fiscal burden on marginalfields but are strongly progressive with ac-tual field profitability. This type of systemis not only efficient but also contributes tocontractual stability by automatically ad-justing the government revenue in linewith the actual outcome of exploration anddevelopment. The Bank has been workingwith developing country governments toassist in the design of such efficient fiscalsystems.

The World Bank's oil and gas lendingprogram (through which US$1.5 billion hasbeen lent for 39 projects) has been designedaround the reduction of constraints on ex-ploration and development in the oil im-porting developing countries to stimulategreater private sector exploration on termsthat will encourage investment but ensurea fair share of benefits for the host country.

Keith Palmera U.K. citizen, is acontract specialist in thePetroleum ProjectsDivision of the Bank'sEnergy Department.Mr. Palmer received hisPhD from the Universityof Birmingham. Prior tojoining the Bank, heworked as an advisor to

governments in Africa and Asia on petroleum policyand contract matters, and later in the Fiscal AffairsDepartment of the Fund.

Loans for exploration and appraisal drill-ing, such as those in Tanzania andPakistan, have helped the national oil com-pany evaluate, respectively, a small gas-prone structure and a small oil field. Overhalf the development lending program isfor gas development, particularly the fi-nancing of transmission and distributioninfrastructure. The availability of this typeof financing can stimulate increased ex-ploration and development by private in-vestors (for example, in Thailand andBolivia) by limiting their financial and po-litical exposure. Various forms of Bank"presence" have been formulated in thehope that they will provide an additionalelement of stability to government-company relations and thereby encouragenew entrants (particularly the smaller, in-dependent oil companies) into explorationas well as a higher level of cofinancing ofdevelopment projects with private capitalsources.

At the moment, national oil companiesplay a major role in exploration in many oilimporting developing countries. Even witha shift toward greater private sector par-ticipation, the role of the national oil com-panies will continue to be important, partlybecause countries will not be prepared tohave so strategic an industry exclusively inthe hands of foreign companies and alsobecause they will have a key role in thoseeconomic investments of little interest tothe multinationals. They will, for example,have a key role in the exploration, produc-tion, and transmission of natural gas forthe domestic market; the exploration ofsmall fields; and investments in enhancedrecovery from currently producing fieldsoperated by the national oil company.Thus, there will continue to be a need forincreased investment funds and technicalassistance from multilateral agencies, suchas the Bank, to support these national oilcompanies.

In conclusion, there do appear to be sig-nificant constraints to private sector in-volvement in exploration in the oil import-ing developing countries. In part, theseconstraints are inherent in the nature of thegeological potential, technology, and polit-ical environment, and in part result fromgovernment petroleum policy. The Bank oiland gas program provides a source of tech-nical assistance to assist governments indevising improved petroleum policies to in-crease the flow of exploration investmentwhile ensuring that the country's interestsare fully protected. The lending program isintended to be a catalyst that, by reducingthe constraints, can stimulate an increase inpetroleum investment in the oil importingdeveloping countries far in excess of theamount of Bank lending. ED



WORLD ECONOMY IN TRANSITIONIn many ways, the global economy is in the throes of amajor adjustment to changing circumstances. Beginningwith this issue. Finance & Development will from time to timepresent data on long-term changes in particular sectors or

aspects of the world economy. We hope this series will givereaders a longer-term perspective on these developments.As usual, we welcome suggestions.

Trade shares and trends, 1950-81From 1950 to 1980 the average annual growth rate of the value

of world exports was 12.3 per cent in nominal terms and 6.7 percent in real terms. The latter figure was well above the increase inworld production and several times greater than that of worldpopulation growth.

Area distribution ol world trad aIPn Mr cent)

Industrial countries' share of world trade, 1950 and 1981(In par cent I

100-

•C "0g- M-

UJ

!».

100.I960 1960 1970

'CilcuIMM from U S dollar viluii1960 1981

Indultrlll, countr!*! nOtl exporting [ 1 Non-oil

country I | dav*toplng countriM

Non-oil developing countries' trade indexesI197S - 1001

200

175

ISO

125

100

75

50

2^I960 1965 1970 197& 19B1

Sourco IMF. International Financial Statistics Supplementon Trtda Statistics. 198!

I'inarKf fir DtwlopmenllMarch 1983 39


Cofinancing — new World Bank approachesThe World Bank's Board of Executive Direc-tors has approved the introduction of a setof new instruments designed to increasethe participation of commercial banks inthe financing of Bank projects in devel-oping countries and to provide better ma-turities to developing country borrowersthan are available from direct borrowings inprivate markets.

Announcing the decision on January 13,1983, Ernest Stern, Senior Vice President,Operations, explained that under the newarrangements, in selected cases, the Bank,in addition to making a direct loan for aproject, will also participate in its financingwith commercial lenders. This programwill be tested over the next 12-24 months.During this time, the Bank is authorized toprovide up to a total of US$500 million forparticipation in commercial cofinantingloans.

In the last fiscal year, ending June 30,1982, cofinancing of Bank projects and pro-grams totaled $7.6 billion, including $3.3billion with commercial banks.

Under the existing system, if cofinancingfor a Bank-financed project is needed anddesired by the borrower, the Bank will as-sist in arranging it. However, the Bank andthe commercial banks enter into separateloan agreements with the borrowing coun-try. The new instruments, which are in-tended to help increase the flow of privatecapital to developing nations, will supple-ment the Bank's present methods of co-financing with the private sector.

Three new approaches for cofinancinghave been approved:

• Direct financial participation by the Bankin the later maturities of a commercial loan. Tosupplement the direct Bank loan, the Bankwould participate in a commercial bankloan, in the range of 15-20 per cent of theprivate loan. The resulting maturitieswould be significantly longer than availablein the market. Annual repayments of prin-cipal would be made by the borrower, firstto commercial lenders until their share ofthe loan is fully repaid; then, in later years,principal payments would be applied to theBank's share. As the loan is amortized, theBank's share would be available for sale tothe commercial lenders.

• Contingent participation by the Bank in thelater maturities of a commercial loan that wouldbe initially financed entirely by commerciallenders. Annual debt service of the commer-cial bank loan to the borrower would befixed even though the interest rate wouldbe variable. If the interest rate rises abovethe initial rate, the amortization will not becompleted on the original schedule. TheBank's contingent commitment would beto finance the balance of principal (if any) atthe end of the term of the initial loan if ratesturn out to be higher than anticipated.

The benefit of this approach, in additionto extended maturities, is that it providesborrowers with a fixed payment schedulefor their debt service, enabling better plan-ning. The commercial lenders would havethe opportunity to purchase the out-standing balance for their own account.

• Guarantees of the later maturities of a pri-vate loan by the Bank, instead of direct funding.Under this option, the Bank, to supple-ment its direct loan, would not participatein the commercial bank loan directly, butwould guarantee repayment of the laterpart of the loan. This approach, too, isdesigned to extend maturities. As the loanis amortized, and the guaranteed portioncomes within the normal range of mar-ket maturities, the guaranteed share can bereduced.

It is expected that the pilot phase of theprogram will include from 15 to 20 loansand will be completed within two years.Estimating that the Bank's share of eachcommercial transaction would be approxi-mately 20 per cent, the pilot program couldgenerate total cofinancing of about $2.5 bil-lion. These funds will be used to financehigh quality investment projects alreadydeemed suitable for Bank lending in thosecountries that the Bank judges to be credit-worthy. RD

Finance & Development will continue topresent information on the World Bank's Co-financing program. Readers may wish to refer toa booklet, Cofinancing, that reviews the Bank'scofinancing with private financial institutionsand explains its cofinancing operations. Thebooklet is available, without charge, from TheWorld Bank, Publications Distribution Unit,1818 H St. N.W., Washington, DC 20433,U.S.A.

An earlier article in Finance & Develop-ment, "Cofinancing of Bank and IDA projects,"by Roger Homstein, was published in June1977.

Sources of World Bank coflnandng, fiscal ys«rs 1973-83(Amount In mMons at U.S. dates)


Pl2£?8 Source oCcoflnandng n ko™,, TJU

cofinancing Official Export i> JK PrivAio «Aj4itiKk41on rer M~t

FtoceJyear Number Amount Number Amount Number Amount Number Amount IBRD IDA coats

1973 37 496.2 30 313.0 10 183.2 0 0.0 478.9 308.0 2,813.11974 48 1.463.0 44 788.8 11 589.5 2 84.7 1,092.6 184.3 5,446.01975 53 1,940.3 48 923.3 9 962.0 1 55.0 1,033.6 346.1 8,817.71976 73 2.255.1 61 1,079.7 16 902.9 5 272.5 1,583.3 403.1 9,620.31977 81 2,292.1 72 1.544.7 9 197.5 9 549.9 1,866.2 698.0 9,920.11978 81 2,463.1 71 1,761.6 15 539.3 8 162.2 1,678.5 762.3 11,304.41979 109 3,153.0 88 1,979.9 16 659.2 13 513.9 2,993.3 1,134.1 13,987.71980 92 6,480.7 69 2,475.6 23 2,282.3 19 1,722.9 3,132.9 1,605.2 21,348.41981 82 4,157.6 68 1,614.3 8 737.1 18 1,806.2 3,230.9 1,496.4 21,569.81982 103 7,570.6 79 2,266.7 26 2,047.7 17 3.256.21 4.417.8 1,226.9 23,280.919832 25 2,957.2 23 581.3 5 1,718.9 4 657.0 1,126.2 294.8 9.605.6

784 35.228.8 653 15.328.9 14A 10.819.6 88 9.080.5 22.834.2 8.469.2 137.713.9Source: Work) Bank data.Note: Information tumWied on privite coflnendnu Is tentative, since privet* tank loan airanoements are often completed w long as one to two years after Board presentation when

tunas ere required. The number of InaMdueJ opendone wKh ooBnandna by source Is greater jhan the total amount, since there are projects eofinanoed from more than one source.;Hpu(wlnc»JOeUS|oWm lnprH^oonnaf>c^ """""'Figures Include projects approved by the Board through December 1862, thatlstorthsflrsthaHolthe fiscal year.


Recentstaff studiesPeter H. Neumann and Maureen A. Cunningham

Mexico's Free Textbooks:Nationalism and the Urgency toEducateWorld Bank Staff Working Paper No. 541, US$5.

Increased enrollments in developing coun-tries often mask the reality of teachers with-out the tools with which to teach and childrenwithout the books from which to learn. Text-books are a small item in any school sys-tem's budget, but they are crucial to the ef-fectiveness of the educational effort and in

the developing countries a relatively smallinvestment in increased distribution can pro-vide dramatic returns.

Mexico nationalized the publication of pri-mary school textbooks in 1959 and has for20 years distributed them free to its primaryschool students. Its success in overcomingpolitical, economic, and technical obstaclesin the face of limited resources reflects theimportance of political will and commitmentand suggests that the Mexican experiencemight offer a valuable example for other de-veloping countries.

The authors analyze the factors at work inthe Mexican program and conclude that anysuccessful national effort to provide free

textbooks must have the following: (1) strongand continued government commitment; (2)carefully planned institution building (withproper concern for the indigenous publishingindustry); (3) development of a realistic cur-riculum; (4) staff who were, or were trainedto become, experts in educational publish-ing; (5) regard for existing models of suc-cessful textbooks and series; (6) teacherparticipation in development of textbooks;(7) research and testing of materials; (8)training teachers to use new textbooks; (9)continuous improvement of materials; and(10) an element of choice among textbooks,so that materials can be better matched tolocal needs.

Douglas H. Keare and Scott Paris

Evaluation of Shelter Programsfor the Urban Poor: PrincipalFindingsWorld Bank Staff Working Paper No. 547, US$5.

The World Bank began lending for projectsin urban development in 1972. Now, a de-cade later, the first detailed evaluation offour such projects (in El Salvador, the Philip-pines, Senegal, and Zambia) indicates thatthey have been remarkably successful.

New sites and services developments andschemes to upgrade squatter areas that pro-vide secure tenure and basic serviceswould, the Bank believed, enable and en-courage low-income households to improvetheir housing through self-help financingand/or construction. But could private sav-ings be mobilized; could truly low-cost shel-ter be designed for low-income households;and, if this occurred on a small scale, couldit influence national shelter policies?

While confirming the projects' effective-ness in meeting these objectives, the papernotes some problem areas and yields someunexpected findings. As hoped, housingstocks increased (by as much as 50 per centin Zambia), and fears of a generation of newslums were allayed when the quality of thisnew housing exceeded expectations. Theprojects were effective in reaching low-income groups, and services showed signifi-cant improvement, though early problemsarose in coordinating the large number ofgovernment and private organizations in-volved. (These problems could be avoided, itwas found, by simplifying the first projectsundertaken in a country.)

These urban projects did reach the poor,but one surprising finding is that the range ofhousehold incomes within a project area—

particularly in squatter areas—was muchbroader than expected. This could work tothe advantage of the poorer members of thecommunity—for example, by providingincome-earning opportunities. On the otherhand, the heterogeneity poses problems intargeting benefits directly to low-incomegroups.

Housing construction and upgrading gen-erated substantial employment and incomeand significantly greater rental income than

expected. Self-help proved a relatively effi-cient construction method, though the use ofself-help labor was apparently not as eco-nomical as, and therefore less commonthan, anticipated. Costs were a particularlyimpressive element in the projects'success—running less than one fifth ofthose for conventional housing in Zambiaand less than one half in El Salvador. Costrecovery, however, presented problems inthree of the four projects.

Robert C. Effros (editor)

Emerging Financial Centers:Legal and InstitutionalFrameworkInternational Monetary Fund, Washington, DC, 1982,xvi + 1150 pp., US$35.

The nineteenth and twentieth centuries wit-nessed the rise of a number of great fi-nancial centers, which developed, as a rule,in powerful and growing economies. Thesecond half of this century, however, hasbeen marked by a new phenomenon: theemergence of certain developing countriesas financial centers that are not necessarilylocated in great national economies. Theymay, in fact, be geographically distant fromthe principal sources and destinations oftheir funds. This book, edited by the Assis-tant General Counsel (Legislation) in theFund's Legal Department, examines sevensuch centers: the Bahamas, Hong Kong, theIvory Coast, Kenya, Kuwait, Panama, andSingapore.

The common feature that sets these cen-ters apart from their neighbors is the sophis-tication and development of their financialinstitutions. In each, banks, insurance com-panies, specialized financial institutions, andsecurities brokers—both local and foreign—

offer a wide range of services to residents,and often to nonresidents as well. Whileeach is a success in its own right as a devel-oping financial center, not every one is atthe same stage of development. Some havean effect well beyond their boundaries, whilethe importance of others is still largelynational.

Why have these economies consistentlydistinguished themselves from their neigh-bors in the breadth and drive of their growthas financial centers? The history and devel-opment of the emerging financial centers re-veals numerous accidental circumstancesand diverse causation, yet despite obviousdifferences, the centers have had to confrontcommon problems. Their evident success indealing with similar challenges suggests adegree of commonality in their legislationand experience. The similarity is likely to begreatest among those financial centers thathave evolved furthest, with the most uni-formity among those centers that have ac-quired international significance. These cen-ters, which compete with each other and arejudged against more established interna-tional financial centers, must satisfy morerigorous criteria than those that meet onlythe needs of the local banking system.



Lauchlin Currie

The Role of Economic Advisersin Developing CountriesGreenwood Press, Westport, CT, U.S.A., 1981,xiv + 270 pp., $27.50.

Dr. Currie's book fills an important gap indevelopment literature. After more than 40years' experience advising governments, hetries to pull together the threads and to seeka greater consensus on appropriate devel-opment objectives and policies. Dr. Currieplayed a central role in the formulation ofNew Deal policies in the United States in themid-1930s; he led the first World Bank studymission—to Colombia, in 1949; and there-after he has been advising the ColombianGovernment. He is, therefore, in a uniqueposition to reflect on the role of economicadvisers.

The book arises out of the author's experi-ence with macroeconomic advice—notablyadvice on monetary, exchange rate, tax, andprice policies—and with sector and projectadvice, before addressing more general de-velopment issues such as the role of marketforces versus specific policies. The processof development, according to Dr. Currie,strengthens a country's capacity to adapt—aprocess implying "profound changes in hab-its, attitudes, customs and training." Thesecultural changes are far more important thanthe quantifiable improvements in physicaloutput and economic conditions that somuch of the literature emphasizes. Through-out, the book combines insights specific tohis experience with Colombian develop-ment, many of which are relevant to othercountries as well.

Various principles for fruitful external as-sistance and advice are set out: coherence

and focus in objectives and terms of refer-ence; sensitivity to national goals and condi-tions; modesty and feasibility of proposalsrather than recommendations for sweepingreforms; and provision for policy imple-mentation. The actual results of policy ad-vice, according to Dr. Currie, have beenmixed in Colombia. In fact, the contributionof development and aid agencies—in pro-moting the physical integration of the coun-try, supporting macro and sectoral policies,and in fostering capital markets—has per-haps been greater in Colombia than theauthor recognizes. He may also have under-stated the potential role of the internationalcommunity in the developing world.

The author's personal convictions on de-velopment are reflected in the book's con-clusions. It is noted, for example, that incomplex macroeconomic areas, there is noeffective substitute for trained national econ-omists. It also emphasizes the importance ofnot recommending policies that could im-pede growth—as would measures to slow orreverse rural-urban migration and mobility.Another lesson the book passes on is that a"direct assault" on mass rural poverty re-quires a degree of administrative expertisebeyond the capacity of most developingcountries. The solution lies in increasing thedemand for labor outside agriculture to raisethe incomes of the gradually declining num-bers engaged in agriculture. This lessonruns counter to the practice of many de-veloping country governments and donorinstitutions.

How far can conclusions based on experi-ence in Colombia (or in any single country)be applied to other developing countries?With a coalition government of the presiden-tial form that has experienced relative politi-

cal stability, a strong cadre of technicians,and a sound resource base, Colombia en-joys many advantages over other countrieseven at its per capita income level. Short-term macroeconomic measures are oftenmore effective in steering growth there thanelsewhere. Dr. Currie's focus on marketsand demand expansion as a means to spurgrowth is particularly well placed in the Co-lombian situation, and it is also a messageworth emphasizing in many countries wherethese factors are neglected. However, manyothers have severe supply constraints, notsimply relating to physical capital, but per-haps more important, to human capital, skilllevels, and management capabilities. Thisstudy of Colombia and of the five countrieschosen for comparisons—Brazil, HongKong, Republic of Korea, Singapore, andTaiwan—is not likely to help these othersadequately. On the other hand, a full com-parative discussion of developmental expe-riences is not the purpose of the book.

Dr. Currie sums up his reflections: "Theless developed a country, the greater theneed to satisfy basic human needs andlessen inequality. But the less developed thecountry, the greater the danger that givingtop priority to these objectives will tend toextend the period of underdevelopmentTherefore, advice and aid are likely to bemore effective the more they facilitate andpromote ... transformations and reinforcerather than obstruct natural or marketforces." He concludes that "The greatestcontribution a government ... can make todevelopment is the initial adoption of policiesthat lead to a sustained period (ten years ormore) of high economic growth."

Guy Pfeffermann and Wood Thomas

Frances Stewart and Arjun Sengupta; edited by SalahAl-Shaikhly

International FinancialCooperation: A Framework forChangeWestview Press, Boulder, CO, U.S.A., 1982,xxviii + 204 pp., $24 (cloth), $10.50 (paper).

The central theme of this interesting book isthat the current international monetary sys-tem provides an unsatisfactory frameworkfor international trade and payments andtherefore requires far-reaching reforms. Thebook, issued by Oxford University's Centerfor Research on the New International Eco-

nomic Order, reviews the main sources ofdissatisfaction and provides a range of alter-native proposals to solve the problems of thesystem. The authors claim "the interest ofdevelopment as the major consideration,"but take into account the interests of theother parties in the world economy, namely,the industrial countries and the membersof the Organization of Petroleum Export-ing Countries (OPEC). Unfortunately, thebook's analysis ends in 1981, and some im-portant elements of the system it seeks toreform have changed since then—OPECsurpluses in particular.

The book covers seven major areas ofconcern and presents proposals for a re-formed international monetary order: (1) Re-cycling. The authors argue that developingcountries will continue to require majortransfers of resources. The recycling bycommercial banks—the major source of fi-nancing in the 1970s—should continue, butadditional resources can be channeled bystrengthening existing public institutions andby creating new organizations, such as aThird World Agency, that would move fundsfrom OPEC to the Third World directly.(2) Improvement in international financial


Booknotices


mechanisms. The uncertainty about loanterms could be reduced by loans denomi-nated in special drawing rights (SDRs), byusing other forms of indexing, or (for fi-nancing imports to the developing world) bya global loan guarantee scheme. (3) Low-income countries. Aid could be increasedand redirected to the poorer countries in theform of direct transfers, interest rate sub-sidies, and loan guarantee facilities. (4) Con-ditionality. The conditionality associated withthe Fund's financial support is reviewed crit-ically. Other sources of finance and alterna-tive policy packages are explored. (5) Inter-national currency. The book dismissesproposals for commodity-backed interna-tional currencies as impractical and supportsthe use of the SDR to provide better controlon international liquidity and to transfer re-sources to the developing world through alink between SDR allocations and conces-sional finance. (6) World economic manage-ment. In light of the alleged vacuum in worldeconomic management, it is proposed that aWorld Development Council be founded, ini-tially as an advisory body. (7) South-Southcooperation. A number of proposals arepresented to improve cooperation amongdeveloping countries, to strengthen their"self-reliance," and to promote "South-South" trade and ease payments problems.

Other books receivedBoy Hofhetnz, Jr. and Kent E. CaMer

The Eastasia Edge:Why an entire region Is overtakingthe West In technology, exports,and managementBaste Books, New York, U.SA, 1982, ix+296 pp., $14.95.

In this analysis of the spectacular economicachievements of some East Asian economies, theauthors lay great stress on the political structuresand policies that provide the organization and mo-tivation for rapid growth. They do not see muchchance of successtal duplication of the East Asiansuccess hi other countries and regions In the ab-sence of similar political conditions.

Sheila C. Dow and Peter E. Earl

Money Matters: A KeyneslanApproach to Monetary EconomicsBames&Nobte.Totows, NJ, U.SA, 1982,vttl + 270pp., S28.50.

Economics is full of schools, tendencies, andapproaches—some inevitably growing out of thedevelopment of the discipline and others linked toideology. "Monetarism" and "the monetary ap-proach to the balance of payments" have, for in-stance, occupied center stage for some yearsnow. This book surveys the entire monetary fieldwith occasional references to real situations. Itscoverage is broad rather than deep and, In theprocess, it adds to the proliferation of groupingsby detecting gradations within schools. Its brief,inaccurate sketch of how adjustment programsare designed and how they are to operate (p. 219)

In general the "South" should attempt to in-stitute its own Bretton Woods.

The authors indicate that most of the pro-posals require a degree of cooperation thatis not present in existing organizations thatare viewed as efficient—the World Bank, theFund, and the regional development banks.But the book fails to point out that reformwould require a shift of attitudes within theorganizations as well as a modification of theinstitutional environment. Therefore, it is un-realistic in assuming that success of the re-forms described is likely.

Past developments are discussed withparticular poignancy and balance, and in awell-documented fashion, based on materialissued by international organizations. Therelevance of the changes the book proposesis unfortunately somewhat dated. The largeOPEC surpluses, on which it bases manyproposals, have narrowed for many OPECcountries since 1981 and were reversed intolarge deficits for others. In consequence, theindustrial countries are returning to the tradi-tional surplus position they experiencedprior to the oil price increase of 1973, there-by shifting the source of funds for developingcountry borrowers.

Many of the various proposals describedin the book have been amply discussed butone central issue may be noted: the adjust-

does not inspire confidence. But its main virtuelies in presenting all the varieties of monetary the-ory and it reaches fairly moderate conclusions thatseem to suggest that the schism between Key-nesians and monetarists is not as sharp as it hasbeen made out to be. A nontechnical book, Itwould be of benefit to undergraduate students.

R.C.Q. Matthews, C.H. Feinstein, and J.C. OdDng-Smee

British Economic Growth 1856-1973Stanford University Pan, Stanford, CA, USA, 1982, xxBH-712 pp., $65.

A study of trends between 1856 and 1913, theinterwar, and the post-World War II years, whichexamines not only the immediate causes ofgrowth—the inputs of capital and labor—out alsothe more fundamental causes, such as structuralchange, demand, and international transactions.

John M. LeSche (editor)

International Economic Policies andTheir Theoretical Foundations:A Source BookAcademic Press, Inc., 111 Fifth Avenue, New York, 1982,lx+674 pp., $27.50 (doth).

A nine-part volume comprising essays by dis-tinguished economists on the terms of trade,world food, common markets, cartels and com-modity agreements, multinationals and interna-tional investment, commercial policies, and inter-national payments in the context of the evolutionof the international economic environment. Areadable compendium.

ment and conditionality recommended bythe Fund and to some extent by the WorldBank. The authors question conditionalityand present alternative lines of action. It isindicated that Fund programs place empha-sis on the control of inflation and on a turn-around in the balance of payments withoutan explicit emphasis on the fulfillment ofbasic needs or on income distribution. Whatis not recognized is that initially all correctiveprograms have to deal with major adjust-ments of aggregated demand to keep themin line with available resources. These ad-justments are not determined by differentobjectives, economic philosophy, or the po-litical system, but by the economic realityimposed by prevailing international eco-nomic conditions and the circumstances ofthe particular country. The adjustment has tooccur in any event, and some alternativesproposed in the book may be more painful,with more reduction in efficiency and greatercosts in terms of income distribution, thanwould have occurred in the context of a Fundprogram. Finally other proposals are errone-ously based on financing as an alternativeand not as a complement to adjustment.

Claudia M. Loser

Henry C. Wallich

Monetary Policy and Practice:A View from the Federal ReserveBoard

Lsxfngtofl Books* 0*C* t tooth wxf Cofflpdrty, Lexington, MA,U.S.A., 1962, w+395 pp., $19.95 (doth).

As the subtitle of this book indicates, Wallichwrites from the unique perspective of a seven-year member of the U.S. Federal Reserve Board.The views of an "insider" are always interesting.Add to that Wallich's broad-ranging approach andwell-reasoned and highly readable prose, and youhave a most useful compendium of papers andspeeches. Although there are sections on inter-national monetary problems and internationalbanking, the book is essentially a discussion ofissues in the context of the U.S. economy (there isa section on "investments and the stock market,"hardly an international concern). For the inter-ested reader, there is a convenient summary atthe beginning of the book to guide him. inflationhas always appeared to Wallich "in part at least asa moral problem, transcending economics. In-flation is a form of cheating. It also has seemed tome that economists have had a great deal to dowith paving the way for the inflation we are nowexperiencing." Monetary policy Is important butnot as a panacea nor because there is any mys-terious causal relationship between money supplyand output and prices. Wallich does not like theexcess movements and unpredictability of ex-change rates, but sees no alternatives at thepresent time. Altogether, a valuable collection.



Russel J. do Lucia, Henry D. Jacoby and others

Energy Planning for DevelopingCountries: A Study of BangladeshThe Johns Hopkins University Press, Baltimore, MD, U.S.A.,1982, xx+298 pp., $24 (doth).

A useful reference work for developing countriesthat are starting to engage in energy planning.This work provides a readable account of themethodology behind energy demand estimates,energy price analysis, agricultural analysis for en-ergy planning, and the links between energy andmacroeconomic models. However, the efforts putinto each of these activities should be carefullytailored to the specific needs of energy managersso that the output is timely and capable of beingachieved. In many cases what the authors call"crashing through"—a rapid diagnosis of energyissues and options—is likely to be more valuableto countries in the short to medium term.

Essays and papers on economicissues from the InternationalMonetary Fund, the Group of Thirty,and the Institute for InternationalEconomicsIn economic literature there has always beenroom for short, topical, and well-written essays onspecific issues. The advantage of this genre is thatessays are more focused than a book, more freshthan collections of seminar papers, and more "rel-evant" and accessible than articles in academicjournals. Readers will be familiar with the long-established Essays in International Finance, pub-lished by Princeton University and currently underthe direction of Peter Kenen. Three relatively newseries have been added to the distinguishedPrinceton line. The first is the Occasional Paperspublished by the Fund. These constitute fairly longpapers on a variety of issues originally preparedby Fund staff members for internal purposes. Theunique feature of this well-researched series isthat they present a great deal of information that isgenerally not available elsewhere, since much ofit is based on primary sources to which Fund staffhave unique access in the course of their work.The September 1982 and March 1983 issues ofFinance & Development carry a list of the mostrecent papers in this series, with details of how toobtain them.

The second series consists of the OccasionalPapers published by the Group of Thirty (obtain-able from The Group of Thirty, Two World TradeCenter, Suite 9630, New York, NY 10048 U.S.A.).These essays are quite short and somewhat sche-matic in nature; recent publications include Co-ordination of National Economic Policies byJacques Polak and Exchange Rate Policy Recon-sidered by Otmar Emminger. The third series of-fers po//cy analyses in international economicspublished by the Institute for International Eco-nomics (11 Dupont Circle NW, Washington, DC20036 U.S.A.). These comprise longer and moresubstantive papers with a distinctive policy bent.Publications to date include The Lending Policiesof the International Monetary Fund by John Will-iamson, "Reciprocity": A New Approach to WorldTrade Policy? by William Cline, and Trade Po/;cyin the 1980s by Fred Bergsten and William Cline.

These three series constitute a major con-tribution to the literature on international eco-

nomics and, together with the Princeton series,provide the generalist as well as the specialist withsolid analysis and considered judgments on theeconomic issues of our time.

Robert Solomon

The International Monetary System1945-1981Harper and Row, New York, 1982, xvi+432 pp., $26 (cloth).

Jacob S. Dreyer, Gottfried Haberler, and Thomas D. Willett(editors)

The International Monetary System:A Time of TurbulenceAmerican Enterprise Institute for Public Policy Research,Washington, DC, 1982, iii+523 pp., $25.95 (doth), $14.95(paper).

If the international monetary system continues tobe beset by problems and difficulties, it is surelynot because of a lack of writing on the subject.Much of this avalanche of literature, as is the casewith these two books, consists of reissues of pre-vious writings and the proceedings of numerousconferences on this subject. The Solomon book isan updated and expanded version of an earlierwork that covered 1945-76. It is essentially a his-torical book by someone who for many years wasvery close to the center of international monetarydevelopments. It is quite an enjoyable book for thepersonal touch the author brings to his subjectand for the flow and clarity of his writing. By con-trast, the book edited by Dreyer ef al is more un-even and somewhat more technical. But it con-tains a number of papers that can complementand add analytical focus to issues brought up bySolomon. The Dreyer book records the pro-ceedings of a 1980 conference that brought to-gether a large number of distinguished academicsand policymakers, too numerous to name. Theissues these two books discuss (such as ex-change rate volatility) have not gone away; but ina sense both seem somewhat dated in light of thehuge debt problems that have afflicted the inter-national economy in the past year. Unemploymenthas become a major international concern, butgiven the recent course of events, authors andconference organizers need not fear this fate.

Jagdish Bhagwati (editor)

Import Competition and ResponseThe University of Chicago Press, Chicago, 1982, ix+410 pp.,$32.50 (doth), $15 (paper).

A conference volume of mathematical papers thataims to fill a gap in trade theory, especially regard-ing issues such as import penetration and com-petition. The papers fall into three major cate-gories: economic theory (papers on adjustment toimports, the efficiency cost of dynamic adjust-ment, and the case for protection and adjustmentassistance, respectively); political economy ofprotection (including papers on the potential forlobbying for changes in policy on labor migrationand capital flows, a trade model with economies ofscale and product diversification, a model of tarifflobbying in the context of a simple two-sectoreconomy, and an analysis of the possibility of gov-ernments buying out protectionist lobbies); andempirical analysis of adjustment problems andpolicies (including a very interesting discussion ofthe impact of adjustment in a Lancashire town).For academia and trade specialists.

T.E. Barker. A.S. Dowries, and J.A. Sackey (editors),Perspectives on Economic Development, UniversityPress of America, Washington, DC, 1982, xx + 304 pp.

David Bigman, Coping with Hunger: Toward a Systemof Food Security and Price Stabilization, Ballinger,Cambridge, MA, U.S.A., 1982, xxxii + 351 pp., $35(cloth).

Landrum R. Boiling with Craig Smith, Private ForeignAid: U.S. Philanthropy for Relief and Development,Westview Press, Boulder, CO, U.S.A., 1982, xv +330 pp., $25 (cloth).

Rod Cross, Economic Theory and Policy In the UK,Martin Robertson, Oxford, U.K., 1982, ix + 223 pp.,£10.

Carl K. Eicher and Doyle C. Baker, Research on Agri-cultural Development In Sub-Saharan Africa: ACritical Survey, MSU International Development Pa-per No. 1, Michigan State University, East Lansing, Ml,U.S.A., 1982, xi + 335 pp., free to individuals and insti-tutions in the Third World; $8 in higher incomecountries.

T. Scarlett Epstein, Urban Food Marketing And ThirdWorld Rural Development, Croom Helm Ltd., Totowa,NJ, U.S.A., 1982, 260 pp., $28.50 (cloth).

Pere Escorsa (editor), El desarrollo industrial en losanos 80, Marcombo Boixareu Editores, Barcelona,Spain, 1982, x +464 pp.

Just Faaland, editor, Population and the World Econo-my In the 21st Century, Basil Blackwell, Oxford, U.K.,1982, viii + 264 pp., £17 (cloth).

Richard E. Feinberg, Subsidizing Success: TheExport-Import Bank In the U.S. Economy, Cam-bridge University Press, New York, 1982, xi +189 pp., $34.50.

Bent Hansen and Samir Radwan, Employment oppor-tunities and equity In a changing economy: EgyptIn the 1980s, International Labor Office, Geneva,1982, xviii + 293 pp., SwF 35.

John Hudson, Inflation: A Theoretical Survey and Syn-thesis, Allen & Unwin, Inc., Winchester, MA, U.S.A.,1982, xii +171 pp., $24.50.

Prof. Ryushi Iwata, Japanese-Style Management: ItsFoundations and Prospects, Asian ProductivityOrganization, Tokyo, 1982, iv + 123 pp.

Melvyn B. Krauss, Development Without Aid: Growth,Poverty and Government, McQraw Hill, New York,1982, xi + 208 pp., $17.95 (cloth).

Richard Medley (editor), The Politics of Inflation: AComparative Analysis, Pergamon Press, New York,1982, xii-t-249 pp., $29.50.

Pierre Pascallon, Le Systems Monetalre Interna-tional—Theorie et Realite, Les Editions deL'Epargne, Paris, 1982, 528 pp., F 120.

Hubert Schmitz, Manufacturing in the Backyard: CaseStudies on Accumulation and Employment InSmall-scale Brazilllan Industry, Allenheld Osmun &Co., Totowa, NJ, U.S.A., 1982, 232 pp., $26.50.

Jonathan E. Sanford, U.S. Foreign Policy and the Mul-tilateral Development Banks, Westview Press,Boulder, CO, U.S.A., 1982, xiv + 279 pp., $20.



Selling Bank publications

It was good on your part to have invitedcomments and views from the readers. TheWorld Development Report and the WorldBank Atlas have now been transferred fromthe list of free publications to the list ofpublications for sale. This must have beendue to escalating costs and the rising de-mand for these publications. But it shouldhave been avoided and some other waysfound to cut costs, as these publicationsprovide basic statistical data and are veryuseful for purposes of reference.

R. S. DangayachJaipur, India

James Feather, Director of the World Bank'sPublications Department, commentsOne of our main objectives is to make Bank pub-lications available at a low cost in developingcountries. Thus, while World DevelopmentReport 1982 is published in the United States atUS$8, it is available in India from Oxford Uni-versity Press, New Delhi, at 25 rupees (approx-imately $2.50). This subsidized price puts thebook within the reach of individuals as well asorganizations.

Additionally, all nonrestricted Bank publica-tions are accessible at some 300 depository li-braries in 80 countries throughout the world. InIndia, there are 46 such institutions that receiveour publications free of charge. Their existenceguarantees that no one who needs World Bankmaterial is denied access to it by reason of cost.

By any other name

For several years in the classroom I havenoted that we need a less unwieldy namefor the form of international reserves thathas evolved from the Fund's bookkeepingentries tagged "special drawing rights" in1970. William J. Byrne's excellent review ofthe evolution of the special drawing right(SDR) in the September 1982 issue of Fi-nance & Development prompts me to writeand suggest the acronym spedra (spadra) asan appropriate name for what will perhapsbecome a truly international currency. Thiswould replace the three English words orthree separately pronounced letters nowused. Of course, a check of the world's nu-merous languages would be necessary toassure that adoption of this word for our

"paper gold" would not cause linguisticproblems. I have experimented in classwith the use of "spedra" in place of "SDR"and it works very well.

Professor Charles E. Ratliff, Jr.Davidson CollegeDavidson, NC U.S.A.

William J. Byrne, Senior Operations Officerin the SDR Division of the Treasurer's De-partment, repliesFinancial innovation inspires linguistic inno-vation. Sir Joseph Gold was once moved to writean entire article on the subject of renaming theinfant SDR (IMF Staff Papers, July 1976,pp. 295-311). Now that the infant has become ateenager, a new name would lead to a severeidentity crisis (not to mention acronymy).

Effective family planning

Authors Cochrane and Meerman are to becommended for their article, "Populationgrowth and food supply in sub-Saharan Af-rica" (September 1982); their analysis issound and, regrettably, their conclusionsvalid. In one area must I take issue and thisis with their statement, apropos familyplanning programs, "Nevertheless, there islittle sense of urgency to the programs; norhave techniques or approaches beenworked out that would be most suited toAfrican circumstances." The first part ofthe statement is correct, but the referenceto techniques and approaches is not.

In the decade that I managed the fam-ily planning program in Rhodesia, nowZimbabwe, we demonstrated the efficacyof a program conducted by a well-fundedvoluntary agency. Our data demonstratedthe danger of relying on the bureaucracy toexecute a national program and underlinedthe fallacy in thinking that a populationprogram could be grafted onto the healthservices. A well-managed voluntary pro-gram must take the lead; the health servicesprovide backup. We took our services tothe "doorstep of the people," using lay ed-ucators and distributors drawn from thecommunity within which they were towork. Our well-trained lay personnel car-ried an effective range of oral and injectablehormonal contraceptives and were built in-to the community as a "source," providinginformation, education, and services.Newly initiated clients were referred to our

family planning clinic-based services, con-ducted by nursing personnel, for theirmedical checks; in the absence of our ownclinic services, clients were referred tostaff of the health services whom we hadtrained.

In 1981 the U.S. Agency for InternationalDevelopment (USAID) agreed to fund theprogram with US$7.5 million over threeyears. This permitted a major expansion ofthe field force; provided equipment and ex-pert manpower to monitor and analyze theachievements of the program; and spon-sored a training facility for the benefit ofother sub-Saharan African countries.

Based on a 1969 census, Zimbabwe's an-nual population growth rate was estimatedat 3.6 per cent. In April 1981 the PopulationReference Bureau concluded that our pro-gram reduced this rate to 3.2 per cent.Zimbabwe's recent census indicates an an-nual population growth rate of 3 per cent,which is close enough to vindicate the find-ings of the Population Reference Bureau.We did reduce the rate of populationgrowth; what we did could have been doneby others.

Peter DoddsFormer Director,Zimbabwean FamilyPlanning Association

The Editor respondsMr. Dodds records a success story in Africanfamily planning programs. There is little doubtthat the program has increased contraceptive usesubstantially. The exact magnitude of the effecton birth rates is uncertain, however, due to lackof recent data on fertility levels. We are told thatforthcoming data from Zimbabwe's own censusshould clarify the picture.

The important thing, however, is not the exacteffect of the program but why it succeeded in thefirst place and whether that experience can berepeated in other African countries. Among themany critical variables that will determine itsreplicability are, primarily, literacy and mor-tality rates, which affect the desired number ofbirths in society and which seem to be moreconducive to low desired fertility in Zimbabwethan generally in Africa. Other factors includethe efficiency of public and private organizationsin supporting family planning, and the commit-ment of political leaders and administrative staffto the success of such programs.


Letters


Monthly Operational Summary carries up-to-datedetails on over 700 projects every year. Reports oneach stage of project development—identification,preparation, appraisal, negotiation, and approval—and enables you to contact the appropriate agency fordetails on procedures for bid submissions.

Project Technical Data Series enables consultants,civil works contractors, and equipment suppliers toreact quickly to new opportunities. Complements theMonthly Operational Summary with the first publisheddetails of goods and services to be procured. Listsname and address of implementing agency.

Mail this coupon with your check payable to: The Johns Hopkins University Press, Journals Division,Baltimore, MD 21218 U.S.A.In Europe, mail to: The World Bank, 66, avenue d'lena, 75116 Paris, France.

C3 Yes. I wish to take advantage of business opportunities resulting from loans made by The World Bank.Please enter my subscription to:

n Monthly Operational Summary. Q Project Technical Data Series.I understand this US$75 subscription entitles I understand this US$150 subscription entitlesme to 12 issues of this valuable guide. me to weekly mailings of Technical Data Sheets.

Here is my check for US$. .. Start sending my new subscription immediately to:

NAME. TITLE .

ADDRESS .

CITY. STATE/COUNTY . POSTAL CODE .

COUNTRY .

Learn how to benefit

from worldwidebusiness opportunities

with these two guides


finance & development · u tun wai david williams christopher willoughby shahid yusuf finance...

Documents