Report No. 6082 Financing Adjustment With Growth in Sub-Saharan Africa, 1986-90 February 1986 FOR OFFICIAL USE ONLY Document oftheU4rId Bank This documenthas a restricted distribution and may be used by recipients only inthe performance of their official duties. itscontents maynot otherwíse bedisclosed withoutWoridBank authorization. Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized
120
Embed
Financing Adjustment With Growth in Sub-Saharan Africa ...documents.worldbank.org/curated/en/515791468003872373/pdf/multi0page.pdf · Financing adjustment with growth in Sub-Saharan
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Report No. 6082
Financing Adjustment With Growthin Sub-Saharan Africa, 1986-90
February 1986
FOR OFFICIAL USE ONLY
Document of the U4rId Bank
This document has a restricted distribution and may be used by recipientsonly in the performance of their official duties. its contents may not otherwísebedisclosed without Worid Bank authorization.
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
THE WORLD BANKWashngtUonID. C. 20433
U.5.A.
A.W. CLAUSENPheit ~February 21, 1986
MRMORANDUM TO THE EXECUTIVE DIRECTORS
Subject: Financing Adjustment with Growth in Sub-Saharan Africa, 1986-90
This i8 the World Bank's fourth report focusíng on developmentissues and requirements in sub-Saharan Africa. Since 1981, the developmentstrategíes of many African countries have changed dramatically. Majorstructural reforms are being uindertaken and some promising results arealready evident. The tragedy of drought and famíne heightened the urgencyof these reforms while often weakening the capacity to implement them.Nevertheless, many African policy-makers have demonstrated farsighteddetermination in pursuing necessary reforms. While much remains to bedone, especially in addressing long-term issues of human and naturalresource development, there has been marked progress by many Africancountries in redressing major macro-economic and sectoral distortions.
The response of the donor community to Africa's famine wasgenerous and effectíve, and represe.ated a heartening example of how a widearray of donors can pursue a shared obiective to the great benefit of thepeople of Africa.
However, the major structural reforms undertaken by many Africancountries to address their long-term development problems have not receivedadequate donor support. As noted in the Report, growth and equityenhancing reform programs already under way are foundering because ofinadequate donor fundlng, which ís often inappropriate in form andtiming. Countríes consídering major reform programs can find littleencouragement from the donor support demonstrated for those countries withreforms under way. In the absence of adequate financial support,structural reforms cannot be achieved with growth. Adjustment throughfurther economíc contraction ls not a feasible alternatíve in a continentwhere per capita income levels today are no higher than they were twentyyears ago.
-2-
In the past, the Bank has argued that resource availability isonly one element ín addressing Africa's development; the political will ofAfrican leaders, essential to effectively utilize both domestic and foreignresources, is also critical. But there is now growing evidence that manyAfrican countries are exercising this political will. Theír comnitment andefforts must be matched now by the political will of the donor community toincrease the resources available to implement growth-oriented adjustment 3programs.
The resources needed to restore growth prospects in Africa are notlarge but they exceed by far those currently available or envisaged. Theseresource requirements reflect the deepening debt problem _n Africa, with arapidly increasing share of resource flows required to service debt, aswell as a decline in non-concessional flows due to the deterioratingcreditworthiness of the region. While-some increases in bilateral andmultilateral flows are foreseen, because of an enlarged Lome III and theSpecial Facility for Sub-Saharan Africa, a further US$2.5 billion per yearin concessional flows and debt relief is required ín 1986-90 to restoreAfrican import capacity per capita to its 1980-82 level, even assuming asignificant increase in exnorts. The increased flows from multilateralagencies including an enlarged IDA-8, IMF Trust Fund reflows, and anenhanced replenishment of the African Development Fund, when due, mightprovide up to $1 biLlion per year of these additional funds. For theremainder, a major effort is required in bilateral assistance -- whetherthrough new money or more liberal debt relief.
Donor action is also required in several areas to enhance theeffectiveness of assistance to Africa. Many of these actions are not new;they have been discussed and agreed upon in principle in various fora.However, their implementation has lagged far behind the implementation ofstructural reforms by many African countries, and urgent action is nowrequired on matters relating to the quality, form, timeliness andcoordination of aid. Action in the following areas would do much tostrengthen the system:
- Donors must be willing to work within coherent and realisticadjustment and investment programs designed by African governments.Projects outside thcese programs are of lower priority and should not befinanced unless the agreed prioríty program is also fully financed.
- Donor decisions on aid and debt relief must be made together inlight of an overall financial package and a program of adjustment withgrowth. This could be achieved Sn the context of adapted consultativegroup meetings or roundtables.
- To enhance the effectiveness of aid group meetings, the ¡najorparticipants should discuss the key elements of the financial package andadjustment program required for the country, in advance of the meeting, andarrangements should be made to monitor implementation of aid commitments insupport of each adjustment program.
- Donors should provide meaningful medium-term indications of aidand debt relief levels-, to enable countries to program their investmentstrategies in light of an indicative resource envelope.
For its part, the World Bank stands ready to further strengthenits role in assisting countries to design programs of growth-orientedadjustnent and in their funding. In addition, as requested by theDevelopment Committee, the Bank will continue to exercise a leading role inestablishing mechanisms to enhance the effectiveness of aid flows throughimproved coordination and monitoring.
There are opportunities ahead to restore many African countries toa sustainable growth path. But substantially increased donor support isessentlal to capitalize on these opportunities.
FOR OFFICIAL USK ONLY
Financing adjustment with growthin Sub-Saharan Africa, 1986-90
The World BankWashington, D.C.February 1986
This document has a restricted distribution and may be used by recipientsonly in the performance of their offícial dutíes. Its contents may notothervise be dísclosed without World Bank authorization.
.i
Contento
Country coverage y
Acrongym aud definitiona vi
Sumuary aUd conclusion. 1
Progreso in domestic policy reform 1Resource scarcity 2Measuring the resource gap 2Bridging the resource gap 3Improving aid coordination 4A year of opportunity 4
1 Introduction 5
The end of the drought 5Continuing long-term decline 7The deepening debt problem 8A strategy of adjustment with growth 9
2 Adjusttent programa 11
Correcting overvalued exchange rates 12Correcting urban-rural bias 14Rationalizing the public sector 17
3 Long-ter= constrainta on growth 20
Population 20Human resources 23
Health 23Education 24
Deforestatíon 25Agricultural research 26 _
4 External capital and aid coordination 30
External resource flows, 1986-90 30Foreign exchange requirements 31Sources of external finance 33Bridging the resource gap 36
Institutional reforms in aid coordination 37Recent progress 37Further reforms 38
Appendix A: The debt problea and ita implications for import capacity 41
Profile of the African debt problem 41Projected debt-service burden and ímport capacity 42
Appendix B: Macroeconomic indicators 48
Statistical Annez 51
Kap 109
- ív
Text tables
1.1. Develop¡Aents vi the trade environmentof IDA-eligible countries, 1970-86 6
1.2* Output, populecion, income, andconsumption in IDA-elígible countries, 1960-86 7
1.3. Availabilíty of resourcesin IDA-eligible countries, 1960-84 8
2.1. Earnings as a multiple of per capitaGDP: Sub-Saharan Africa and Asia 16
3.1. Desired family size and actual fertility rates 22
3.2. Agricultural research efforts:Sub-Saharan Africa, Asia, and Latin America, 1980 27
4.2. Supply of external finance for IDA-eligiblecountries, 1980-90 34
4.3. Concessional resource gap forIDA-eligible countries, 1986-90 35
Text figures
2.1. Real effective exchange rate indexesin developing countríes, 1971-84 13
This Report was prepared by a team led by Ramgopal Agarwala and comprisingCharles Humphreys, Kathie Krumm, and John Underwood, with the assistance ofmany World Bank staff members and Stanley Please, principal consultant. Itdraws heavily on the ideas and views expressed by scholars and officials inAfrica and in donor countries. The work was carried out under the generaldirection of Xavier de la Renaudiere. Sang Lee, in collaboration with RameshChander and with the assistance of Zafar Ahmed and Maria Cristina Germany andothers, prepared the Statistical Annex. Clive Crook was principal editor andJane Carroll and Carol Rosen were technical editors. Special thanks go to thesupport staff, headed by Jean Ponchamni and including Lynn CIark-MCCarthy andAnn Van Aken.
v
Country coverage
The report covers all countries south of the Sahara, except SouthAfrica. However, for statistical analyuis and for estimation of resourcerequirementa, the text focuses on twentv-nine countries of sub-Saharan Africathat are eligible to receive credíts from the International DevelopmentAssociation (IDA>, an affiliate of the World Bank.
Eenin Kenya SenegalBurkina Faso Lesotho Sierra LeoneBurundi Liberia SomaliaCentral African Republic Madagascar SudanChad Malawi TanzaniaEthiopia Mali TogoGambia Mauritania UgandaGhana Mozambíque ZaireGuinea Niger ZambiaGuinea-Bissau Rwanda
As a group, these countries are referred to in this report as tlow-íncomecountries" or "IDA-eligible countries." Thus, the definition of low-incomeused here differs slightly from the definition of low-income economies in theStatistical Annex and in other World Bank publications (per capita GNP of $400or less in 1984). In this report, the low-income countries include not onlythose with per capita GNP of $400 or less in 1984, but also Liberia, Lesotho,Mauritania, and Zambia, which had per capita CNP below $550 in 1984.
The share of these twenty-nine countries in some major economicaggregate.i in sub-Saharan Africa is shown in the charts below.
IDA eigibte counhGs in sub-Soharan Alico 1984
popuúon hGC
e7-a,
Eagofts DeS-
- ví -
Acronyms and defiuitions
DAC Development Assistance Committee of the OECDFAO Food and Agriculture OrganizationIDA International Development AseociationILO InLernational Labour OrganisationIMF International Monetary FundODA Official development assistanceOECD Organisaton for Economic Co-operation and DevelopmeatOPEC Organization of the Petroleum Exporting CountriesUNICEI United Nations Children's FundUNCTAD United Nations Conference on Trade and DevelopmentUnesco United Nations Educational, Scientific and Cultural OrganizationUNIDO United Nations Industrial Development OrganizationWHO World Health Organization
Unless otherwise indicated, tables and figures in the text andappendices are based on World Bank data. Growth rates are least-squaresestimates in constant prices unless otherwise stated.
Billíon is 1,000 million.
Dollars are US dollars.
.. Not available.
.) Less than half the unit shown.
Summary and conclusiona
Sub-Saharan Africa ia emergín& from one of the worst famines inrecent history. Cood rainas have come to much of the region. Per copitaincomes should rise this year for the first time since 1980. Even so, thereis little cause to celebrate. Low-income Africa ¡s poorer ín 1986 than it wasa generation aga in 1960. And though the recent good news interrupts thetrend of chronic economic decline, it will not be enoug- to reverse it.Population growth is largely unchecked, productivity all but stagnant. Ifpresent trends continue, the human disaster of 1983-84 in sub-Saharan Africavill return to haunt the world community.
This year gives a breathing opace. Africa shoutd use it to makeatructural adjustmenta aimed at reviving growth, especially in agriculture,and to halt the deterioration in basic health and education services. Donorcountries, for their part, must not see 1986 as a time to relax. Theprincipal theme of this report ¡s that Africa's attempts to help itself willfail with"u additional resources in the form of new aid and debt relief.
Progreuo la domestic policl refor.
Many Afri-an goverments are now making significant progresain structural adjustmen.. Sut they atill have much to do tocorrect the accumulated policy distortiona of the past.
In many African countries, government policy has long discriminatedagainst agriculture--the sector which ¡a not only crucial for any attempt toraise output and exporte, but whích also gives most of the poor theirlivelihood. Dismantling such policies will hela the region to increase outputand exporte, and help the poor at the same time. Thís proceso has begun.Goveramenta have started to reduce the overvaluation cf their currencíes--oneway in which agriculture had been penalized. They have increased adrículturalprices and lowered real urban wages. They have reduced public spending, vitbite bias toward expanding employment in urban areas.
These are welcome reformst but no more than the first steps.Exchange retes remain severely overvalued in many African countries.Producers' shares of export yrices are still too low. Urban wages are higherthan in other low-income developing countries.
Africa's governnments have done even less to overcome the region'slonger-term obstacles to development. Family planning, education and health,resource conservation and agricultural research are vital in this respect. Insome cases, however, services have declined, canceling out progress made inearlier years. And governments still have much to do to improve theallocation of resources--by giving a greater role to prices, markets and theprivate sector, by increasing the supply of domestic savings, and by runningpublic enterpríses more efficiently.
Resource scarcity
Declining importa and investment threaten to undermine atructuraladjustment in low-income Africa.
Adjustment with growth requíres extra resources, but the recent trendin resource flows has been just the opposite. The investment rate in low-íncome Africa has been falling sínce 1980 and ¡a now the lowest amongdeveloping regíons. It is not sufficíent to maintain and rehabilitateexisting productive capacity, nor to provide for new capacity. This declinein investment mirrors declines in Afríca's supply of both domestic and foreign.savings; the shortfall in foreign savings in turn reflecta a worsening debtcrisis and a diminishing flow of commercial capítal. A dozen countries nowface acute debt difficulties, with little prospect of improvement.,
Africa needs more imports, not just of investment goods, but also ofessential consumer goods and raw materials. Importa per capita have beendeclíníng since 1970s and the rate of decline has accelerated in the 1980s.In spite of hígher coffee prices and lower oil prices, the outlook for theterms of trade is poor. Without new aid and debt relie4, imports per capitawill contínue to fall for the rest of this decade and beyond.
Measuring the resource gap
To continue its progreso toward economic adjustment, lov-income Africa vill need at least $11 billion a year inconcessional flows during 1986-90. Alloving for known andexpected aid comhnitents, a gap of $2.5 billion remains.
In calculating external resource requirements, the objective is tohalt the trend of decline in per capíta consumption by 1990 and achieve somegrowth thereafter. This minimum objective would require a GDP growth rate ofat least 3-4 percent a year by 1990. Qn the basis of the Bank's countryexperience, it is unlikely that such growth rates can be achieved, unless thedecline in import capacity is reversed. At a mninimum, imports per capitashould return to theír level of 1980-82. This implies imports of $28.5billion a year in current prices during 1986-90. Provision must also be madefor debt-service payments that are projected to rise to an average of $6.8billion a year over the períod. Altogether, then, the region needs $35.3billion a year for the next five years.
The possible sources of that finance are as foLlows. Exports mightprovíde at best $20 bíllíon a year--and this assumes that governments continueto adopt export-oriented policíes. New bank borrowing on commercial termsmight provide $1 billion a year; in view of the region's poor debt-servícingcapacity, it would be imprudent to seek more. In aLl, another $1 billion ayear ín nonconcessional finance might come from the IMF, froma direct privateforeign investment, and ín short-term capital flows. Debt. reschedulings,under existing procedures, should provide around $2.3 bíllion a year. Knownand expected aid commítments imply concessional flows of $8.5 biíLion a year--an increase of about 25 percent from tte level of 1980-82. Altogether, theregion might tap $32.8 bíLlion, leaving a resource gap of $2.5 billion a year.
-3-
1986-90
projected
annual average
(bilitons of dollars)
Forcign exchange requirements for IDA-eligible countries
Imports of go"d end services (excluding Interest> 28.5
Debt service payments (plus payments to IMF) 6.8
Total 35.
Sources of f¡nance
Exports of goods and services 20.0
Nonconcessional flows 2.0
Debt rescheduling (based on existing practices) 2.3
Concesslonal flows (based an existing and expected commitments> 8.5
Total 32.8
Resource gapTo be fWiled by additional concessional flows
and debt rescheduling 2.5
Bridgíng the resource gap
Multilateral agencies might provide $1 bíllion of the
resource gap of $2.5 billion a year. This leaves $1.5
bhllion a year to be met from new bilateral aid and
additional debt relief.
Multilateral agencies could, at best, meet about $1 billion of the
gap íf the IDA-8 replenishment is st least $12 billíon and low-income Africa
receives a large part of the SDR 2.7 billion expected in the IMF Trust Fund
reflows. Bilateral aid and additional debt relief must bridge the remaining
gap of $1.5 billion a year. That vould mean an increase of 30 percent over
the aid and debt relíef given in 1984 and a 20 percent increase over the
levels currently projected.
The additional resources will be needed to pay for importe during
1986-90 and to cover debt service payments, so they must be in quick-
disbursing form. The correct pattern of assistance--in particular, the split
between quick-disbursing aid and debt relief--will vary from case to case.
But one general rule should be observed: no donor country should be a net
recipient of resource flows fros any African country undertaking credible
economic reforma.
Donors might provide new bilateral assistance in several ways, for
example:
o By converting exísting official loans into grants, through retroactive
termsa adjustment, as proposed in the UNCTAD declaration of 1977
-4-
O By enabling their export credit agencies to refinance debt servicepayments on more concessional termo
o By widening the circLe of creditors that participates in debt rescheduling
o By increasing their ai* 5udgets, and especially by increasing quick-disbursing funde avíilLole in support of atructural adjustment.
Improving aid coordination
Dovora must act more in concert-vith each other and vithrecipiente. This report suggests cix vaya to improvecoordinastioQ.
First, donora must be willing to work within adjustment programadesigned by African governments. Second, they should bring decisions on aidand debt relief together, using the existing framework of consultative groupmeetings. Third, the major participants should discuss the elements of therequired financial package in advance of full-scale aid coordinatíonmeetings. Fourth, to provide effective support for medium-term adjustment,donors should be more willing to give medium-term indications of aid. Fifth,secretariats should be established to monitor progress toward economic reformand toward implementing governments' and donors' agreements. Sixth, themultilateral agencies must assume a larger role in orchestrating donorassistance--both in designing adjustment programs and in financing them. TheWorld Bank and the IMF, in particular, must work together with Africangovernments, first to develop adjustment and investment programe aimed atrestoring growth, and second to assess the requirements for, and sources of,external finance.
A year of opportunity
Recovery from the drought has brought lower food prices, and importedpetroleum will be cheaper because of the decline in oil prices. This makes1986 a good year for Africa to accelerate ite procese of correcting overvaluedexchange rateg: lower food and petroleum prices will soften the inflationaryimpact of devaluation on urban dwelLers; at the same time, devaluation wouldhelp raise the farm prices of agricultural exporta and partly offset theeffects of lower food prices on farmere. Lover food prices have alzo createdan opportunity for Africa to dismantle, without undue hardship, more of itacontrole on food pricing and marketing.
It ¡J a year of opportunity for donors too. Decisions on the IDA-8replenishment and on the IMF Trust Fund reflows will affect the resourcesavailable for Africa over the next five years. The Development Comnittee'smeeting in April and the epecial session of the UN General Assembly in May canestablish the principles on which the concerted effort of the Africancountries and the international community can be based.
1. Introduction
The famine of 1983-84 ín sub-Saharan Africa ¡i videly recognized 8sbeing part of a longer-term production crisis. The World Bank's previousreport on the region argued that this deeper problem has ite roots in high andaccelerating pqpulation growth and in low and declining efficiency in the useof resources. _ The report predieted that, without additional donor support,inadequate resources vould soon become another major constraint, and Africa'sdecline would continue.
This year has brought some velcome signs of relief. Most of Africahas had good rainfall. An increasing number of countries are starting torecognize the policy mistakes of the p8st and are taking steps to correctthem.- Por low-income Africa, termo of trade have improved because of highercoffee prices and lower oil prices.
However, the long-term tren¿ of decline persists. As yet, too fewcountríes are embarking upon the atructural adjustment process; those thathave begun atill have far to go. Only a handful of countries are seriouslytrying to control extremely high rates of population growth. The availabilityof resources in Africa ¡a shrinking, and without additional donor support,will continue to do so.
The challenge for Africa and the donor countries ¡i to reverse thelong-term trend of decline. Adjustment with growth must become a realistictheme for the lov-income countries of the region--as for the high-debt middle-income developing countries. Over the next five years, this calls both fornational programa of policy reform and for substantially higher donorasmistance: each will fail without the other. To achieve adjustment withgrowth during this period, decisions must be made in 1986 on the level snd theform of assistance that the developnment community will provide. The focus ofthis report is on the needs of low-income African countries eligible to borrowfrom IDA, '2ut many of íts conclusions hold for the other sub-Saharan countriesas vell. -
The end of the drought
Sub-Saharan Africa ¡i emerging from one of the worst famines inrecent history. The drought that began in southern Africa in 1983 went on toaffect more than twenty countries and 35 million people by 1985. Theínternational community responded on an unprecedented scale, vithcontributions of food, medicine, and other critical supplies. Relief workersfrom donor and recipient governments and from prívate agencies moved massive
1. Tovard Sustained Development in Sub-Sahsran Africa: A Joint Programof Actíon (Washington, D.C., 1984>.
2. Por a list of the IDA-eligible countríes, see the note on countrycoverage at the beginning of this report. In this report, the terms "IDA-eligible" and 'lov-income" countries are synonymous.
-6-
samounts of material under extremely difficult condítíons, helping to savethousands, if not millions, of líves. Yet the famiíne took its toll. Manyfarmers and nomads have joined the ranks of the urban poor. Families havebroken up, systems of community support have weakened, and the strain on over-stretched goverrment servi'es has increased still further. In these difficultcircumstances, people are sL.dggling, with remarkable spirit, to resume theirnormal lives.
Fortunately, most of Africa received good rainfall in 1985. In spiteof the famine's devastation, agriculture is recovering. According to the FAO,agricultural output in 1196 will equal or exceed pre-drought levels in mostIDA-eligible countries. - The main exceptiona are Ethiopia, vhere faminepersists in some areas, and Mozambique; both will need special assístance thisyear, according to the UN Office for Emergency Operations in Africa. Most ofthe region, however, shares in a sense of relíef.
Another favorable development is that low-income Africa' s terms of
trade are expected to improve this year by more than 10 percent--enough torestore them to their average for 1975-80 (Table 1.1). Coffee prices haveíncreased sharply, mainly because of drought in Brazil. They should be around50 percent higher in 1986 than in 1985; this vill add roughly $750 million tothe export earnings of the low-income countries. More recently, the price ofoil has fallen dramatically; if it stabilizes between $15 and $20 a barrel,they will save at least $750 million a year in import costs. These twofactors more than make up for falls in the prices of some other exports--suchas cotton, tea, and edible oil--and the rise in the dollar price ofmanufactured imports.
Table 1.1. Developments in the trade environment of fDA-eligible
countries, 1970-86
1970-75 1975-80 1980-85 1986
(prel iminary) (projected)
Terms of trade (1975-80=100) 124 100 95 101
Merchandise export volume
(annual percentage changeJ -2.2 1.4 -1.0 6.4
Together with the first steps toward policy reform, thesedevelopments should bring a pause in Africa's decline. In 1986, per capitaGDP should rise for the first time since 1980--though only just. Thanks tothe improvement in the terms of trade, per capita income will rise by a largermargin--perhaps by 2.5 percent (Table 1.2).
3. FAO, "Food Cropa and Food Shortages," Special Report (Rome, December 16,1985).
-7-
Table 1.2. Output, popufation, incomo, end consumption in IDA-etlgible countries" 1960-86
e. GDY Is GDP adjusted for changes i¡ teras of trade.
b. 1986 es a percentage of 1910.c. 1984 es a percentege of 1960.
Continuing long-term decline
Welcome as these developments are, they wíll not make up the groundthat low-inconie Africa has loSt in recent years. Even after thís year'sprojected increase, per capita income will have fallen by about 12 percentsince the atart of the decade. In countries such as Chad, Niger, Tanzania,and Togo, the drop since 1980 will be roughly 30 percent--similar to that inthe United States during the Creat Depression of the 1930s. The decline inAfrica's per capita output during the 1980s, together with the decline in the19709, will wipe out all ita rise in per capita output since 1960. As aresult, low-income Africa is poorer today than in 1960. Improvements overthose years ín health, education, and infrastructure are increasingly atrísk. For the first time since World War II, a whole region has sufferearetrogression over a generation.
The underlyíng trend of economic decline in low-income Africa has notchanged. Movementa in the terms of trade are hard to predict, but the long-term projection is for renewed deterioration; by 1990 the index is likely tobe a third lower than in 1970, and 3 percent lower than in 1986. Rapid growthin population continues largely unchecked. Afríca's policy reforms shouldeventually lead to greater efficiency and higher output, but this process hasbarely started.
Now a new problem must be reckoned with. In the 1970s, as the WorldBank's 1984 report on Africa poínted out, the region was not short ofinvestment as compared with the 1960s. The increase in foreign capitalinflows more than balanced lower terms of trade and the fall in the domesticsavings rate. Africa maintained its investment at around 18 percent of GDP.But in the 19809 the investment rate has slipped. It fell to an estimated 14percent in 1984 (Table 1.3). It seems that Africa is acquiring anothernegative superlative; its investment rate ja now the lowest of any developingregion. Investment ¡a leas than the region needa for sustained development--too little to provide for new productive capacity or even to maintain andrehabilitate existing capacity.
Table 1.3. Avallability of resaurces in ¡DA-elígible countries, 1960-84
The fall in the investment rate during the 1980s reflects a declinein both domestic and foreign savings. Domestic savings averaged about 15percent of GDP until the mid-1970s. Since then, the savings rate has droppedsteadily. In 1984, it may have fallen to 6 percent--an extraordinarily lowfigure. Two factors go far to explain this trend. One js that per capítaincomes have been falling. More important, perhaps, public sector def£cita(that is, negative public savings) have gone up because of uncheckedgovernment budget deficits and losses incurred by state-owned enterprises.
The net inflow of fo-.eign savings has fallen from 11 percent of GDPin 1980-82 to 8 percent of CDP in 1983-84. This has happened primarilybecause of the increase in debt service payments and lower inflows ofcoomnercial capital due to reduced debt servicing ability. (See Chapter 4 formore details.)
The deepening debt problem
Most of the countries in low-income Africa face mounting difficultiesin servicing theír debts. Between 1980 and 1984, the region's debt servicepayments (including payments to the INF) increased from 18 percent of exportearnings to 26 percent. In countries such as Malawi, Niger, and Zambia, theratio rose to more than 30 percent. But, these actual debt service paymentsunderstate the problem. If it had not been for reschedulings and, in somecases, a build-up of arrears, debt service payments in 1984 would have beenmuch higher--38 percent of export earnings for low-income Africa as a whole.Fourteen countries rescheduled their debts in 1984-85; in some cases, the newarrangement was just the latest in a series. And several countries wentfurther into arreara on their payments, including to the IMF.
Repeated reschedulings and arrears hinder development. Frequentreschedulinga use up the scarce management time of Africa's policymakers.They can also create a climate of uncertainty that makes sustained developmentmore difficult. Arreara, for their part, can halt new loan commitments anddisbursements from existing loans. In particular, arrears to the IMF candisrupt purchases from the Fund under existing arrangements and make moredifficult the process of formulation of an adjustment program, which in turncan block debt relíef from private and bilateral creditora. In either case,foreign suppliers begin to take account of the uncertainty that they will benaid--and the debtors' imDort Drices rise as a result.
-9-
Without action by the development community, these problems willmount during the next few years (see Chapter 4 and Appendix A). A dozen low-income countries face particularly severe debt problems and for them presentrescheduling arrangements will be inadequate. - Even with rescheduling onconventional termo during the next few years, their debt burden will still behigh at the end of the century.
A strategy of adjustment with growth
If Africa's decline ís to be reversed, action is needed on threefronts. First, the region needs more resources for investment--both foreignand domestic. Second, it must use new and existing resources moreefficiently. Third, it must curb its growth in population.
The basic objective of further adjustment is to stimulate growth.Thar requires more resources. Some of the additional resources can come fromhigher domestic savings. Prudent fiscal and monetary polícies are essentialfor that purpose. But domestic savings cannot carry the whole burden.
A consensus is emerging in the international community thatconcessional capital inflows to the low-income countries of Africa shouldrise, at least for the rest of this decade. To make medium-term programa ofadjustment with growth feasible, donors must give medium-term indications oftheir plans for aid and debt relief. Programs can then be designed on alonger-term basis to give adequate attention to the tasks of developing humanresources, conserving natural resources, and maintaining or extendinginfrastructure. A clearer picture of the resources to be available would alsopromote stability in trade and credit regimes. Domestic and foreign privatesectors would then feel more confident about the outlook for imports mdcredit and could play a more active role in stimulating growth.
In judging the prospecto for adjustment with growth ín low-incomeAfrica, it is important to realize that the region's performance has not beenuniformly poor. The region has its succesa storíes, some of which are betterthan average for the developir.g world. This is true both for sustaineddevelopment and for medium-term adjustment. At least seven countries in sub-Saharan Africa (Botswana, Cameroon, Ivory Coast, Kenya, Malawi, Nauritius, andRwanda) achieved significant growth ín per capita income over the last twenty-five years; four of them (Botswanas Cameroon, Ivory Coast, and Mauritius) havegraduated from low-income status. Some (Ivory Coast, Kenya, Malawi, andMauritius) launched unsustainable public expenditure programs during theperiod of commodity price boom in the mid-1970s, but then adopted atructuraladjustment programs in the early 1980s and by 1985 had made substantialprogress. For each area of policy Africa can report some success. Forexample, exchange rate policy in Malawi; food marketing in Cameroon; exportcrop pricíng in Kenya; export promotion in Mauritius; family planning inMauritius and Zimbabwe; and management of windfall income and technicalassistance in Botswana.
4. Benin, Gambia, Liberia, Madagascar, Mali, Mauritania, Miger, Somalia,Sudan. Tanzania. Toto. and Zambia.
- 10 -
A detailed strategy for 1986-90, which translates general aims intospecific programs, can be formulated only at a country level--but someelements that all such programs should include are discussed in the chaptersthat follow. Two sets of issues are important. The first, discussed inChapter 2, concerns the effíciency of reesorce use. It involves three mainareas of policy that were highlighted in earlier reports:
o Exchange rateso Urban-rural biaso - Public sector management.
The second set of issues, discussed in Chapter 3, covers longer-termconstraints on development; these were also identified in earlier reports:
o Population growtho Healtho Educationo Deforestationo Agricultural research.
Chapter 4 then goes on to assess the external capital and donorinstitutíonal reforms required to support these programs of adjustment withgrowth.
A crucIal objective of the proposed programs is to stimulate outputand exports by correcting discrímination agaínst the rural sector. This is tobe done through reduction ín price distortions and in urban bias in theallocation of government and aid expenditures. Most of the poor in low-incomeAfrica live in rural areas; so these improvements should promote equity asvell as growth. Rarely will the poor be the principal losers from suchprogramse
- 11 -
2. Adjustment programs
For some time Africa's own regional institutions have been leading achange in attítudes toward development policy. The most recent jointreport by the Ec omic Comnission for Africa and the African Development Bankas one example. - It argued that mistaken policies have depressedagricultural output and industrial productivity, promoted inefficiency instate-owned enterprises, and damaged incentives for the domestic privatesector. The Organisation of African Unity's summit meeting in July 1985echoed the theme by adopting "Africa's Priority Programme for EconomicRecovery, 1986-90." This too stressed the importance of agriculture, and theneed for new industrial policies that give the private sector a largereconomic role, and for more active population programs. This shift in Africanthinkiug toward a more pragmatic approach to development. is important andencouraging.
But new thinking is only a first step. Successful adjustment meansadopting new policies, not merely agreeing that they are desirable. How muchhas the region actually achieved? Thia is hard to answer precisely, becauseimprovemento consist of many different changes in many dífferent countries.Nonetheless, progress ¡a clearly under way.
Especially in the past two years, more countries have started toact, and the changes they are making go deeper than before. Several countriesillustrate the trend. Togo, Ghana, and Zaire have reinforced thecomprehensive economic and financial reforme which they introduced in 1983.In 1984-85, Senegal, Mauritania, Zambia, and Guinea adopted similar policies--aiming to improve broad macroeconomic performance, as well as to assistindividual economic sectors. Other countries, such as Niger, Mali andMadagascar, continue to adopt important policy reforms, though more slowly.The reform movement seems likely to endure, and perhaps to gather pace, asmore countríes see the resulta in those that have ted the way.
These reforma cover a wide range of measures aimed at giving príces,marketa, and the private sector a greater role in promoting development inAfrica. In particular, they reflect a desire to reduce administrativeintervention in setting prices; to end monopolies on trade and marketing; andto reduce the government's role in allocating credit and to increase the scopefor prívate sector activity. Changes are needed on a broad front, but thischapter focusea on three areas which were highlighted in the World Bank'sprevious Africa report: exchange rates, urban-rural bias, and public sectormanagement.
1. Economic Report on Africa, 1985.
-12-
Correcting overvalued exçhange rates
Correcting overvaluation of currencies helpa to reduce balance ofpayments deficita. Equally important, it also shifts the internal termsa oftrade in favor of those who produce for export (often mainly rural households)and away from those who consume imports (often predominantly urbanhouseholda). In the past two years, several countries have made progress inthis area. After more than a decade of creeping appreciation, the averagereal effective exchange rate for the IDA-eligible countríes in Africa began todepreciate ín 1984, and this trend seema to have continued into 1985. Agrowing number of governments have adopted more flexible exchange rateregimes, which rely on supply and demand in the foreign exchange market todetermine the exchange rate.
Following a huge devaluation, Zaire introduced a floating ratesystem in 1983, and by 1984 the exchange rate was freely determined ín aninterbank market. By the end of 1984, the real effective exchange rate hadfallen by three-quarters znd the premium on the parallel market was no morethan 10 percent, roughly in line with the cost of bank commissions. In late1985, Zambia introduced an auction system for most trade-related exchangetransactions. Although the government and selected public enterprisescontinue to receive an administrative allocation of foreign exchange, thevalue of foreign exchange is determined by the auction. After theintroduction of the new system, the nominal exchange rate fell by two-thirdsin three weeks.
Guinea and Gambia are establíshing similar systems intended to leadto a unified. market-determined exchange rate; as an interim measure inGuinea, the new rate has been devalued to less than 10 percent of the oldofficial rate. Madagascar took a step toward more flexible exchange rates inMarch 1984; it began to adjust the Malagasy franc every quarter in line withthe change ín the consumer price index during the previous quarter. Betweenthen and the third quarter of 1985, the real effective exchange ratedepreciated by about 15 percent and was brought in line wíth its 1978 level.Other countríes (Ghana and Mauritania) have made substantial devaluationswithin their existing exchange rate systems, by combining vigorous nominaldevaluations with effective measures to contain inflation.
These are promising developments. However, for the low-íncomecountries as a group, the average real effective exchange tate appreciatedsignificantly in the 1970s and early 1980s--a marked contrast to theperformance of Asian and Latir American countries (see Figure 2.1>. As oflate 1985, in spite of the recent adjustments, the real effective exchangerate was about 15 percent above ita 1970-72 levels. Moreover, even the 1970-72 level ¡a too high for some countries, especially where terms of tradeshifts have been large and negative. Por example, in Zaire and Zambia,market-determined real effective exchange rates had fallen at the end of 1985by over 40 percent from their level in 1970-72. The progreso observed in1984-85 is important, but only part of a long proceso that still has far togo.
- 13 -
Real efecthe exchange rte Indoxes In devMeoping countrles, 1971-84
(1971 = 100)
170 u
160 - ~ -A1i su-tSafaan ~co atO - iD A cournes ín Af
_ * _ * _ I.Ist6Y emli1erre<,o --- ,ss..ee.e.ASI/
150 b...
Ait sub-Sahoran Africa /
140
130 --
120 -» --
- «/ IDA countriesb Afnca110
100
-
i ,
o~~ S
80 < s *'-",."..A5se
70 '
Western HeiBmsphee %. * _ '41 _ *o
1972 1974 1976 1978 1980 1V82 1984
Note Regional ove s ae weghted by GDP in 1983Sourca iMF Prch m WoW Bank - 3352 2
One of the reasons for delaying or minímizing exchange rate
adjustments is the fear of the social and political consequences. Resistance
to adjustment, of course, comes from groups--often in the urban sector--who
benefit from the implicit subsidies of currency overvaluation. The
dístributional effects of devaluation diffet aecording to the underlyingatructure of the economy, but they are likely to benefit lower-income
groups in many African countries. Moreover, devaluation allows the removal of
controls and trade restrictiona which create artificial scarcities and bring"rents" to those with acceso to the acarce rights. Rarely will the poor be
the principal losera from devaluation.
- 14 -
However, because of the impact on the urban population, it isimport-ant to time devaluations with care. The economic developuients of1985-86 make 1986 a good year to act decisively on exchange rate adjustment.Lower food prices, following recovery from the drought, together with lowerpetroleum import prices, can cushion the impa,t of devaluation on the towns;at the same time, devaluation would help raise the farm prices of agriculturalexports and partly offset the effects of lower food prices on farmers.
Correcting urban-rural bias
There is a growing consensus in Africa that policies whichdiscriminate against ag-riculture and favor the urban sector must be changed.There are at least three links between farming and development in low-incomeAfrica. First, the largest part of the population and the majority of thepoor depend on agriculture for a living. It will therefore be difficult toattack poverty on a broad front so long as the sector ¡a discriminatedagainat. Second, the region's output ís still heavily agricultural. Higheroutput will require incentives for improved productivity in farming. Third,exporta, too, will be mostly agricultural for the medium term. But they willbe slow to rise as long as governments channel the profits mostly to benefittheir urban populatíons.
Devaluation, as noted earlier, can be a powerful instrument toredress the urban-rural bias. It can simultaneously mean higher farm pricesfor agricultural exports and higher prices for the imported goods that areconsumed mostly in the towns. But correcting the urban bias requirescomplementary changes in domestic policy. These include hígher farm prices,lower agricultural taxation, more flexíble marketing arrangements, and wagerestraint in the urban sector, as well as more public expenditure in the ruralareas. Over the past two to three years, many governments in low-incomeAfrica have introduced reforms in these areas.
Many governments have improved agricultural incentives by increasingoffícial príces ín conjunction with a currency devaluation. Some have done itby deregulating markets, to give farmers access to higher prices on parallelmarkets. Others are introducing more flexible systems of offícial pricing asa step toward lifting controls on agricultural trade.
There ís a clear trend toward higher food-crop prices, In 1984,following substantíal devaluations, the government of Zambia raised theofficial producer price of maize by over 35 percent (a 12 percent increase inreal terms). In Tanzania, the government more than doubled the controlledproducer price of maize between 1983 and 1985 (a one-third increase in realterms). In 1985, Mauritanía set ration shop cereal prices in líne with theprice of imported grain. In Zaire following devaluation and deregulation offood marketing, producer prices are estimated to have doubled for maize andtripled for cassava between 1983 and 1984 (which amounted to at least adoubling of these prices in real terms since 1980). As part of a reformpackage including massive devaluation, in early 1986 Guinea decontrolledproducer prices for food crops and quadrupled the official wholesale price ofimported ríce.
*- 15 -
A number of countries are making similar efforts to improveincentives for export crops. Ghana, after large devaluationa in 1984-85,tripled the price of cocoa between 1983 and 1985 (a 50 percent íncrease inreal terms). Zambia raised the price of cotton, and Zaire raised the pricesof both coffee and cotton; in both countries, these price increases followedsubstantial currency devaluatíons.
Another clear illustration of this trend is that the countriesimplementing reforms duríng the 1980s have seen crop prices rise much fasterthan urban incomes. The disparity was sometimes dramatic--as in the case ofmaize in Zaire and rice in Madagascar. In most cases, farm prices foz exportcrops also improved relative to urban incomes, though less so than food-cropprices.
Examples of some dramatic shifts in urban-rural íncome ratíosinclude Tanzania during 1980-84, when real farm incomes are estimated to haverisen by 5 percent, while urban vage earners faced a decline of 50 percent.In Ghana, during the same petiod, farm incomes stagnated but urban incomesfell by 40 percent in real terms. Per capita incomes have fallen in low-income Africa during the 1980s, while real producer prices have risen or atleast remained constant ini most of the countries that have adopted reformprograma. This leaves little doubt that the reforming governments have helpedto raise the terma of trade between the countryside and the city.
In agriculture, it is hard to separate the effects of better weatherfrom the effects of better prices and better policy. But the response ofagriculture in the countries that have adopted improved incentives has beentoo dramatic to dismiss. In Ghana, cocoa output in 1985-86 is expected to be25 percent above the level of two years earlier--though it will still be lowerthan the pre-drought level of 1981-82. After a price increase of over 300percent in 1983, maize production tripled in 1984, rising above the pre-drought level by almost two-thirds. In Zambia, cotton and maize prodxctionrose by over 20 percent during 1983-85. The amount of maize marketed lsestimated to have risen by over 55 percent during 1984-85; it came close tomeeting Zambia's domestic consumption requirementa for the first time since1976. In Togo, cotton production is estimated to have doubled between 1984and 1985.
Incentives for farmers have improved, but there is room for furtherprogreso. The producer ahare of the international price of exports remainsrelatively love For coffee, cocoa, and cotton--low-income Africa's principalagricultural exports-farmers in the main exporting countries generallyreceived only about half of the export unit value in 1984. Coffee farmers inKenya were an important exception, usually receiving as much as 90 percent ofthe export price. By contrast, in Tanzania, the producer share for coffee was40-50 percent, and in Madagascar it was only about 20 percent in 1984-85.During these years, for cocoa, the share ín Togo was about 40 percent, and forcotton, the share in Mali and Togo was as low as 30 percent.
Two recent reforma in this area may be a guide for othercountries. in 1984-85, Cameroon's national marketing board atartedímplementing a system of rebates to coffee and cocoa farmers; it is estimated
- 16 -
that ít will increase theír share of the export price by 7-8 perce.nt. In late1985, Guinea said the state marketing agency would pay coffee and palmproducera 80 percent of export prices, and raised official buying pricessubstantially.
Conditions in 1986 seem right not just for initiatives on exchangerate policy, but also for other policies to correct the urban-rural bias.Higt food output has helped ease consumer prices in many countries. Thiscreites an opportunity for governments to dismantle price and marketingcontrols an¿ to eliminate food marketing subsidies with fewer fears of theeffects on inflation. Lower prices for food and petroleum should also dampeninflation. Hence, 1986 will be a good year to try to restrain urban andpublic sector wage increases. Furthermore, in those areas where excessproduction threatens to depress the market, deregulation, incl.uding freerintr¿regional trade, could help to maintain producer pricés and at the sametime help consurmera in areas with shortages.
Price adjustments and market deregulation can go only part of theway to correct the urban bias. Personnel policies in the public sector mustalso change: they influence urban earnings because the government and publicenterprises are the largest employers in most countries. Public sector hiringend wage policies have ínflated wages and in many cases left them out of línewith productivity and labor costa in other developing countries. Althoughearnings data are poor and difficult to compare between countries, indicatorsfor selected African and Asian countries reveal that government and urbanwages in low-income Africa are relatively high (see Table 2.1). This salarystructure was adopted at the time of independence and then maintained in manyAfrican countries through most of the 1970s; it must now be adjusted toreflect current budgetary realities. Salaries for some positions in thepublic sector--especially senior offícials and highly skilled technicians--may need to be raised in some countries, but the average public sector wagewill have to come down in most of Africa.
Table 2.1. Earnings as a multipie of per capita GDP: Sub-SaharanAfrica and Asia
Sub-Saharan AfricaRange Asia
Earnings Average Low High (average>
Central government 6.9 3.9 15.1 3.3Manufacturing 4.3 1.7 6.5 2.8Construction 3.1 2.0 4.1 2.0
Note: Data cover thirteen IDA-eligible countries in sub-SaharanAfrica and three low-income countries in Asia for various years from1978 to 1983.
Source: Petar S. Heller end Alan A. Tait, "Government Employmentand Pay: Some International Comparisons," IMF Occasional Papar 24,October 1983; 110 Yearbook of Labour Statistlcs, 1984;and World Bank estimates.
- 17
Rationalízing the public sector
The role of the atate in the economies of sub-Saharan Africa islarger than in other regions. Public sector employment is half of all modernsector employment, compared with only one-third in Asia. Allowing for countrysize, public enterprises are more numerous than in most other developingcountries, and they engage in a wider array of activities. Public investmentaccounts for the bulk of investment in the formal sector.
As the Bank's previous reports on Africa have argued, the rapidgrowth of the pubLic sector, along with its inefficient management andoverambitious public investment programs, are ímportant reasons for theeconomic difficulties facing sub-Saharan Africa today. In addition, the scopeof public sector activities and the operating subsidies to which publicenterprises had access atifled private entrepreneurial activity inagriculture, industry and commerce. Many governments in sub-Saharan Africanow face financial dífficulties and are therefore trying to reduce the size ofthe public sector and to improve its management. However, these reforms arestill at an early stage.
Financial stabilization programa are begínning to reduce governmentexpenditures. Public consumption as a percentage of GDP in low-income Africadeclined from 17.1 percent in 1980-81 to 16.2 percent in 1983-84. Thisdecline happened alongside the fall in GDP. As a result, public consumptiondeclined by 1.5 percent a year in real terms during 1980-84, compared with anincrease of almost 5 percent a year in 1975-80. These cuts have beenachieved, in part, by restraints on public employment. In the past few years,Rwanda, Mali, Somalia, and Togo have abandoned their policy of automaticemployment of secondary school and university graduates. In 1984, the shareof personnel ín total central government expenditures declined for the firsttime in several years in Mali because of slower growth in governmentemployment as a result of the introduction of competitive examinations forentry to the civil service. The SomaLi government froze government empLoymentin 1985. Togo reduced total government employment by 1.6 percent in 1984.
Budget restraint, reflected in lower public employment, is desirablein much of Africa, but ín some cases it has led to excessive cuts in financingfor equipment, maintenance, operating costs, and materials. The result hasbeen a steady deterioration in the quaLity of public services and furtherdeclines in the productiíity of public employees. The impact is plain whenroad maintenance crews lack fuel and bitumen to accomplish their work, whenteachers lack books, chalk, and even classrooms for their students, and whenhealth workers have no medicines to distribute. This deterioration in publicservices is especially disruptive for programs designed to deal with the basicconstrainta on development. Unless a better balance is established betweenpersonnel and other expenditures, these programs are in jeopardy. To maintainnonwage expenditures, most countries must reduce government employment and, insome instances, civil service wages.
Efforts to rationalize the public enterprise sector have focused onimproving management as well as reducing it9 size. Studies and audits aregenerally followed by closure of nonviable enterprises, by divestiture of
- 18 -
nonstrategic enterprises, or by carefully taílored rehabilitation ofenterprises (such as utility companies) -selected to remain ín the publicsector. Most of these efforts involve institutional reforms to strengthengovernment supervision.
Low-income Africa's governments have closed down or divested about 5percent of their public enterprises during the 19809. For many of those thatremain, better management has cut operating losses--through hígher prices,better cost accounting, and (usually) reduced employment. In Ghana, theaggregate net loases (before taxes) of approximately a hundred publicenterprises felL in 1983 to 20 percent of their 1982 level. In the Gambia,net losses after taxes of seventeen public enterprises fell in 1984 to about30 percent of the 1982-83 level. Be en 1981 and 1984, the operatingdeficita of thirteen key public ente.prises in Mali fell by over 50 percent.In 1985, Ghana's Cocoa Marketing Board--the country's largest public sectoremployer--dísmiased 19,000 employees, a third of its payroll. In Mali, about10 percent of public enterprise employees were fired in early 1984. However,reductions made in one area can sometimes be offset by increased employmentelsewhere. In Benin, for example, the central government rehired many of theworkers who lost their jobs when some provincial public enterprises closeddown.
Of the countríes that are trying to reform their public enterprises,Senegal and Zaire are especially interesting. In Senegal, the government andaix public enterprises have signed formal agreemento (contract programs) whichestablish objectives and performance indicators and set out the reciprocalobligations of the government and the enterprises. Compared with other publicenterprises, these firma have had higher sales growth and lower personnelcosta.
In 1984, Zaire introduced a series of reforms, and for some largepublic enterprises, these measures have started to bear fruit. Substantialrestructuring of the mining sector, together with devaluation and tax reforms,have vastly improved the financial situation of CECAMINES. After two years ofsubstantial losses, and despite the continuing loa price of copper, profitsafter taxes amounted to $45 million in 1984 and are expected to have been muchhigher in 1985. In Zairels transport sector, ONATRA--the company responsiblefor river and rail transport between Kinshasa and the port of Matadi--tooksteps to reduce wage costs and raise tariffs. Nov it ía paying divídends tothe treasury for the fírst time in several years. And Zaire's nationalraílway company reduced its losses by over two-thirds since 1983.
The region s governmento have made efforta to reduce the size oftheir public investment programa and to improve the allocation of public fundsto sound, high-priority projectc. Some of the discipline results fromunavoidable financíal austerity. Much of it, though, reflects a greaterappreciation of the importance of including rehabilitation in publicinvestment programa, of keeping investments in line wíth the capacity tofinance recurrent costa, and of using existing productive capacity beforebuilding more. There ja also a growing awareness that amall-scaleinvestments, especially in agriculturea may be more productive than large-acale investments and that large projects of great political appeal butdoubtful economic viability are best postponed or dropped altogether.
-19 -
However, doubtful projects continue to be financed--ostentatiousuniversítiest overdesigned highways, luxury sports stadiums, and uncompetitiveagroindustrial ventures. Often these are financed by donors responding topolitical requests. Sometimes donors fail to appreciate the scarcity ofinvestable funds and the need to devote aid resources to projects that areeconomically sound. And it is odd that finance ¡a sometimes easier to findfor a politically sttractive project of doubtful viability than for support ofexchange and trade adjustment programs or for maintenance and rehabilitationprojects. Most African countries atilL need to strengthen the authority ofthe government agencies that are in charge of planning and project evaluation.
This year, strengthening the planning process is all the moreurgent. The coffee price boom will augment resources availahle to manyAfrican governments. It ís essential that mistakes of the miu-1970s followingcommodity price booms are not repeated, and that extra resources be used tostimulate growth in the near term and to attack the basic constraints todevelopment in the long term. In most of low-income Africa in 1986-90,rehabilitation and maintenance should remain the focus for expenditures oninfrastructure, agriculture, and industry--together with some carefully chosennew investments to remove bottlenecks. However, as the next chapter explaina,in areas concerned with overcoming long-term constraints on growth, it isnecessary to go beyond rehabilitation and maintenance and finance new projectscovering both capital and recurrent expenditure. Projects of doubtfulviability--"white elephants," as the 1984 report called them--must not beallowed to reappear.
- 20 -
3. Long-term constraints on growth
The thrust of structural adjustment in Africa has been toward agreater role for prices, markets, and the private sector in promotingdevelopment. This emphasis needs to continue. However, in areas such asfamíly planning, human resource development, natural resource conservation,and agricultural research, governments must play a larger and more effectiverole. Sub-Saharan Africa has made little progresa in these areas in recentyears. In some cases, services have declined, undoing earlier progresa.
On issues such as family planning, resource conservation, andagricultural research, governments must commit themselves to change andpromote a social consensus in its favor. Consensus must spring from a clearerunderstanding of the link between these long-term factors and prospects for abetter quality of life.
Policy can then build on this by giving programs in these areasgreater priority in the allocation of scarce budgetary and human resources, bystrengthening public institutions where necessary, and by giving privateagencies (including nongovernmental organizations) a much larger role.Increased health and education coverage for a growing population requires,when resources are short, restructured programs which emphasize low-costpreventive health care, primary education, and greater efficiency. Withoutnew policies, the quality and relevance of these services will decline, andthe objectives of universal literacy and greater technical and managerialíndependence will be out of reach.
Population
Lower rates of population growth may well be one of the mostimportant requirements for sustained development in Africa. A recent paperfrom the Worldwatch Institute stated the problem graphically:
If African governments take a serious look at futurepopulation/resource balances...they may discover thatthey are forced to choose between a sharp reduction inbirth rates or falling living standards and, in somecases, rísing death rates. If [they] choose to donothing, the "demographic transition" that has marked theadvance of all developed countries may be reversed forthe first time in modern history. -
1. L. R. Brown and E. C. Wolf, Reversing Africa's Decline (Washington,D.C.: Worldwatch Institute, June 1985), pp. 31-32.
-- 21 -
Until recently, very few African governments recognized theimportance of family planning. Africa's vast empty spaces suggest that manymore people can be accommodated. The problem,-however, ja not that Africalacks the land to support a larger population, but that substantial investmentand other inputs will be needed to make that land sufficiently productive. Ittakes time to mobilize these resources and use them effectívely. The existingrate of population growth is higher than the rate at which the supply ofnutrition, water, health, and education can be increased.
In 1984, at both the Second African Population Conference in Tanzaniaand at the UN International Population Conference in Mexico, Africangovernments appeared ready to adopt a new approach to their policy onpopulation. A recent World Bank study found that about three-quarters ofAfrican countries now endorse family planning, at least fo health reasona,and some have set explicit targets for population growth. _ Increasingly,national leaders are ready to discuss the population problem and the need foraction to deal with it--as, for example, the presidents of Nigeria and Kenyahave done in recent apeeches.
In some cases, such as Malawi, the turnaround in official attitudeshas happened remarkably quickly. In Zimbabwe, provision of contraceptiveservices has expanded rapidly in the past five years, and contraceptive usehas more than doubled. Kenya and Botswana have programs well under way, andseveral other countries--for example, Tanzania, Liberia, Nigeria, Rwanda,Burundi and Malawi--have recently started programs too. As a result, fertilityseens to be falling in some countries. Birth rate- are starting to drop inZimbabwe, which provides good access to family planning services in ruralareas. Kenya, one of the first countries with a major population program,received considerable aid for family planning in the 1970s; ita most recentpopulation surveys show that fertility rates have declined, though onlymarginally, from 8.0 in 1977 to 7.7 ín 1984.
However, the progress achieved so far in sub-Saharan Africa as awhole is small in relation to the task of achieving viable populationgrowth. First, several African countries still take a pronatalist position orshow virtually no support for population control. Second, experience aroundthe world shows that in the difficult area of family planning, governmentsupport is only a begínning. It is not encugh to widen contraceptive choiceand expand channels of access to family planning assistance, however importantthat may be. Poor people in Africa, as elsewhere, find it in their interestto have large families. Desired family size in Africa is the highest in theworld (Table 3.1); in contrast wíth Asia and Latin America, it is higher thanthe fertility rate.
2. World Bank, "Population Growth and Policies in Sub-Saharan Africa"(forthcoming).
-22-
Because the demand for contraceptíon is so small, family planningprogramo that do nothing more than supply contraceptives will have limitedímpact on population growth. Clearly, action is needed to raise demand forfamily planning and to build a national consensus in itS favor. Over themedium term, demand for family planning will increase with progress ineducation (especially of women), with better standardq of health (especiallyredvced infant mortality), and with improvements in the status of women.Equally important are policies that reduce the gap between private and socialrates of return from family planning by increasing the private financial costsof large families.
Table 3.1. Desired fam¡iy slze and actual fert¡llty rates:Sub-Saharan Africa, Asia, and Latin America
Des Ired TotalRegion fam;ly size fertility rate
Sub-Saharan Africa(10 countries> 7.5 6.7
s¡a(10 countries> 4.0 4.7
Latin America(13 countries . 4.3 4.7
Note: Data are from surveys conducted in the late 1970sor early 1980s.
Source: Worid Bank, "Population Growth and Policies in Sub-Saharan Africa" (forthcomíng).
One measure of the progress of family planning efforts is thecontraceptive prevalence rate (CPR); that ¡s, a percentage of married women ofchíldbearing age using contraception. The average CPR in sub-Saharan Africai9 3 to 4 percent, compared with 50 percent or more ín much of Asia (Thailand59 percent, China 71 percent, Sri Lanka 55 percent, Indonesia 58 percent, andPhilippines 48 percent). However, in Mauritius, the CPR is 51 percent and inZimbabwe 22 percent. Other examples of better than average CPRs in Africa areBenin 18 percent, Ghana 10 percent, and Cameroon 11 percent. Experienceelsewhere suggests that to bring population growth down to about 2 percent ayear, the CPR has to ríse to at teast 25 percent. To reach that target beforethe end of the century, African governments and the donor community alike willhave to increase their efforts.
National governments at present finance a very small share ofexpenditure on family planning, and it is essential for them to give higherpriority to population control in budgetary allocations. Donors also need toincrease their support for famíly planning activities, which in Africacurrently receive only 0.5 percent of ODA, compared with 1.5 percent for alldeveloping countries.
- 23 -
Human resources
Pespite the recognized need to develop human resources in Afríca,there have recently been signs of shrínking access to, and a deterioration inthe quality of, services for health and education--especially in ruralareas. Basic policy reforms, as well as additional resources, will be neededto reverse this trend.
Health
Health conditions in sub-Saharan Africa have improved over the pastfew decades, but they remain among the worst in the world. In most Africancountries, acceas to health care is extremely limited. According to 1985estimates by U14ICEF and WHO, childhood imzunization levels, for example, arethe lowest in tLe world: polio 32 percent, measles 35 percent, tuberculosis41 percent, and diphtheria, pertussis, and tetanus 33 percent. An immuni-zation level of about 80 percent is generally considered necessary to controlthe transmissíon of these diseases, so there is still a long way to go.
The 1980s have seen decline, not improvement, in health services.Despite the lack of reliable and timely statístics, most observers report thatpublic health delivery systems have deteriorated in the past few years,particularly in rural areas. In many countries, thís has meant a loweroverall standard of health. Immunization levels have actually dropped inrecent years in countries such as Zambia, Tanzania, and Ghana. One reason forthis dismal picture is a worseníng shortage of financíal resources. Healthexpenditures are often vulnerable to cuts during períods of fiscal austerity;in some West African countries the share of public health in the nationalbudget fell in the early 1980s.
Inefficient use of the resources that are available has aggravatedthe problem. Government policy has contributed to this in three ways. First,at times of budget austerity, growth of employment in the health sector hasbeen maintained, so a rising salary bill has crowded out funding for essentialsupplies, transport, and maintenance of facilities. Second, most governmentscontinue the practice of concentrating health care facilities in urban areas(and especially in high-cost hospitals), neglecting peripheral facilities thatare much more cost-effective for treating the vast mejority of diseases.This, together with the policy of providing such facilities free or at anominal charge, puto pressure on high-cost urban facilities and leavessecondary facilities underused. Third, some governments have reduced thesupply of private health services by making medical facilities a publicmonopoly.
Raising the standard of health in Africa wíll require several policyreforms to increase the efficiency of resource use. The first step is to aimto deliver a minimum package of health care--especially immunization againstmajor communicable diseases, oral rehydration therapy, early diagnosis and
- 24 -
treatment of acute respiratory infections and malaria in chíldren under five,and other elemento of mother and child health cave. To do this, governmentsneed to reorient their public health systems by expanding access to community-based, primary health care, particularly in rural areas. These communityprogramo must then be backed up by institutions to monitor public health.
Covernments nust also change their hiring and cost recovery policiesto ensure that they have adequate funding to cover their nonlabor costa.Several governments are already emphasizing primary health care programs thatwill help to correct the urban bias of past policies. Many have introducedco0t recovery measures. Senegal, for instance, has a program in which userfees are collected and managed by community heaith committees. Othercountries--Ghana, Malawi, and Zambia, for example--are seriously consideringvarious schemes for user charges and community financing. Some have takensteps to reduce unit costs--for example, Mali now imports generic drugs.
Reforms to improve the level and use of resources are important; theymust be supported by donor financial assistance. Donors should coordinatetheir efforta not only in financing projects, but also in promotíng policyreforme that emphasize primary health care, cost recovery, and funding forrecurrent costs. Donora should also consid2r funneling a larger ehare oftheir resources through nongovernmental organizations. These are already theprincipal providera of health care in many rural areas, where they can serveas a well-developed vehicle for targeted primary health care.
Education
In many African countries, the education sector expanded afterindependence. Now it is suffering a reversal in both the quality and thequantity of the services it provides. According to the most recent Unescodata, primary enrollment ratios declined during 1980-83 in twelve IDAcountries in sub-Saharan Africa. In some cases (Somalia, Togo, andMozambique), the decline in these three years was 10 percentage points ormore. Evidence of declining quality ¡8 more scattered and anecdotal, but thefact that spending--on items other than teachers' pay--has fallen in the pastthree years is an indication. A lower qualíty of education in turndiscourages families from sending their children to school--especially whenthe decline happens alongside higher school fees and other charges.
To restore the momentum of primary education, African governmentswill have to allocate additional resources to the sector. This iS necessarynot only to meet the enrollment requirements of a rapidly growing population,but aleo teo ensure that students have basic books and supplies. Some savingsmay be possible on primary teacher salaries--without jeopardizing either thequality or quantity of primary educati,n--through increased qualitydifferentials in the salary structure, increases in teachers' working hours(to bring them more in line wíth civil service norms), and, in some cases,higher student-teacher ratios.
- 25 -
In postsecondary education, the immediate prierity ja to reduce thepublic costs-per student. In sub-Saharan Africa, the annual pubLíc cost ofhigher education amounts to almost $3,000 per studert. Thit ís eight timeshigher than the cost of higher education ín Asia and aimost double its cost inLatin America. The costs are high for several reasons. Student-teacherratios are only seven to one, compared with twelve or fifteen to one in Europeand the United States. Nonteaching costs and student subsidies are aliohigh: a recent survey of twenty-four African countries showed that twenty-twoprovide free tuition, twenty-one cover board aid lodgíng for most studenta,and sixteen provide additional cash allowances. Covernments can reduce thesepublic costs in several ways. For example, they should close or consolidotesmall or low-priority programs (for example, the degrees in Spanish and Polishgranted by Ghana's universities). They should spend leas on arta andhumanities and more on scientific and engineering training. They shouldinstitute tuition charges and provide fewer subsidized services and studentstipends. Some countries, such as Cameroon, Ghana, Malawi, and Nigeria, areworking on programs to cut costs in their universities.
Donors should coordinate their efforts to ensure that scholarships,technical assistance, and construction programs are consistent with thenational objectives discussed above. External assístance to education ls atpresent heavily skewed to postsecondary education. Bilateral assistance isconcentrated in technical assistance and scholarships, multilateral aid incivil works and construction. Only 14 percent of ODA for education :.s spenton operating costS and supplies, and only 11 percent of total foreigi.assistance for education is channeled to primary education. Foreign donorsundoubtedly have a comparative advantage in supplying higher education andcapital assistance, but they need to restructure their assistance to avoid adeterioration in the primary schools: they should increase the share ofasaistance for school supplies at the primary level.
Deforestation
Deforestation and, more broadly, degradation of the land--soilerosion, declíning soil fertílity, and desertification--are serious andmounting problems over large parts of Africa. The area of forest and savannahwoodland has halved since the turn of the century, and there have been majorlosses in farm tree stocks. The decline in tree stocks ís accelerating underseveral influences: consumption of fuelwood ís rising with population growth;land clearance has removed trees from farm boundaries and groves that wereneeded to maintain soil moisture and nutrient contents or to protect soalsfrom erosion; seedlings and mature trees have been lost to ill-managedlivestock; commercial logging in the higher rainfall zones goes on withoutadequate reinvestment in, or maíntenance of, forest reserves; and the meanannual yield of tree stocks has rallen in proportion to the decline in thestocks themselves.
A recent World Bank 3 aper reviewed the major issues in this area andproposed an action program. 3 It concluded that virtually every governmenthas recognized the threat of deforestation and taken some steps to protectforeste and encourage tree planting. Yet forestry programs are often treatedas a low priority and, in some cases, even countries with vell-designedpolicies fail to implement them effectively. There is a clear need for publiceducation and government commitment.
In the more favorable ecologícal conditions of the humid zone,farmers and local communities are planting trees, especially on privatelyowned farmlands. The issue there is how to design policies on incentives,pricing, harvesting, forestry extension, and expansion of forestry research.In the Sahelian/Sudanian zone, where trees grow,slowly, the natural rangelandsare overstocked, and land is owned communally, the hostile environmentpresente more formidable problems. These cannot be solved by reforestationalone. Forestry programs must be integrated with policy on agriculture,livestock, land settlement, forestry, energy, and irrigation. When anintegrated approach has been tried, it has often been successful as, forexample, in Ethiopia, Kenya, and Uganda. Nongovernmental organizations canplay a vital part by supporting the community actions that are essential tothe succeas of this approach. Judged against the scale of the problem,however, the programa adopted so far to check deforestation have been grosslyinadequate.
Deforestation in Africa presents a formidable challenge. Answeringit wíll require sustained political commitment and appropriate conservationpolicies on the part of the African governments, with substantial technicaland financial support from donors extending over many years.
Agricultural research
Over the past few decades, large sums have been allocated toagricultural research in sub-Saharan Africa. The number of researchocientists more than doubled in the decade 1970-80; scientific staff years inrelation to the value of agricultural output in Africa now compare favorablywith other developing regions. Funding, especially from outside donors, hasbeen substantial by international standards. It amounted to about 1 percentof agricultural GDP in 1980, which compares with 0.5 percent in Asia and about0.75 percent in Latin America (Table 3.2). All of the research centers in theConsultative Group for International Agricultural Research support some aspectof African food and agricultural production.
There have been some encouraging results. Hybrid maize varieties forsubtropical climates, adopted in the 1960s or earlier, have enabled some East
3. "Deforestation, Fuelwood Consumption and Forest Conservation inAfrica: An Action Program for FY86-88," January 1986.
- 27 -
African countries to become self-sufficient in maize or even to becomeexportera (Kenya, Malawi in good years, and Zimbabwe). The Ivory Coast hasdeveloped a hybrid coffee (arabusta) to raise the quality of production wherearabicas cannot be grown. "Jab planters" have been developed ín Nigeria tominimize soil tillage and soil erosion. Controlled droplet sprayers haverevolutionized insect c-'ntrol in, for example, Sahelian cotton production.Progress has been made toward controlling leaf mosaic in cassava and livestockdiseases such as east coast fever.
In apite of this, most observera agree that the technology shelf insub-Saharan Africa is nearly bare. Most farming remains at rudimentarysubsistence levels: farmers make little use of fertilizer, and hand-hoecultivalion is still the most common. Major advances such as those thatrevolutionized wheat and rice cultivation in Asia have not happened since the1960s, when some countries wíth favorable climates adopted hybríd maize.There has been no breakthrough for millet and sorghum, which account for 80percent of cultivated land in the Sahel and other dry areas, or for rootcrops, which are major staples in more humid zones. For some crops, thebacklog of basic research is enormous, despite the apparently high levels ofresources devoted to it. Even where improved technology exists, it oftenlacks proper testing or is poorly dispersed to farmers.
Table 3.2. Agricultural research efforts: Sub-Saharan Africa, Asia, and Latin America, 1980
Expenditure on agricultural research Agricultural scientistAs percentage Per scientist staff years per S10of agriculural staff Vear milIlon agricultural
GDP lthousands of dollars) GDP
Sub-Saharan AfricaWest Africa 1.19 83 1.42
East Africa 0.81 46 1.76
Southern Africa - 1.23 50 2.47
Asia
South Asia 0.43 34 1.29
Southeast Asia 0.52 25 2.07
Latin AmerícaTemperate South America 0.70 53 1.32
Tropical South America 0.98 56 1.77
Caribbean and Cen-tral America 0.63 52 1.20
a. Includes South Africa.Source: R. A. Evenson, "The FARCs: Evidence of Impact on National Research and Extension on
Producttvity," for the Consultative Group for International Agricultural Research (draft), 1985.
- 28 -
Existing research capacity ís not only costly to maintain but oftenunderused. Costs per researcher are high by international standarda (onaverage, over $50,000 per scientist a year, compared with about half thatamount in Asia). Researchers are isolated and frequently lack the incentives,equipment, funding, and supporting personnel to conduct researcheffectively. Highly traíned national researchers become demoralized or aresiphoned off into administrative roles. The local institutiona responsiblefor research are generally weak; they lack strong political backing andrecurrent budget support.
A long-term program to strengthen agricultural research in sub-Saharan Africa must aim to build up scíentific knowledge, to strengthen thecountry-level institutions, and to make better use of existing local researchcapacity. A short-term príority, however, ís to adapt existing technology tolocal farm conditions. The results can then be spread by input suppliers inthe prívace sector (as in many developing countries in other regions) orthrough the extension system. In the longer run, attention must turn to landconservation--and not just to agronomic practices (cultívation, seeds,fertilizer, weed control), but also to land tenure and farming systems.Especially in dry areas, it will not always be enough, even in the short run,to adapt existing technology. Africa needs major research on new cropvarieties to diversify agricultural production, on new techniques to conservesoil moisture, on improved irrigation practices, on control of livestockdiseases, and on agroforestry.
Donors pay for much of the agricultural research in sub-SaharanAfrica. In the early 1980s they spent about $300 million a year--perhaps halfof the total. As a result, they shape research polícies, institutions, andpriorities, often without adequate national involvement. The large number ofdonors and the small síze of many of the countries in sub-Saharan Africa haveled to confusion and to duplication of effort. Donors have compounded theproblem by frequently changing their priorities and by demanding quantifiedresults within periods that are unrealistic for agricultural research. Donormissions have produced an increasing amount of often contradictory docu-mentation, proposals, and solutions to Africa's research problems. Largenumbers of separate donor-financed projects make competing demands on staffresources which are limited, particularly in the smaller countries. Moreover,these "enclave" projects typically last only a short time: they do little toimprove the institutional framework.
In order to assist the African countries in this area, the World Bankis helping te develop a special program for African agricultural resear:h.The inítial objectives of the program are to improve the exchange ofinformation among donors on research activities, to collect and assessresearch results on promising technologies, to help develop national researchstrategies supported by regional programs and networks, and to provideresearch funda to selected national scientists. Thís is a modest butimportant initiative to improve the effectiveness of aid for agriculturalresearch.
- 29 -
In conclusion, it is worth emphasizing that sub-Saharan Africa hasmade little progreso in overcoming the long-term obstacles to development. Inseveral countries, facilities for basic services have deteriorated. Poorgovernment policies have sometimes contributed to the problem--for example, byfsiling to provide adequate funde for recurrent costa and by allowing wages tocrowd out apending on equipment, maintenance, materiala, and other supplies.Unit costs have often been unaustainably high. Some governments have ruledout more efficient nongovernmental organizations or private alternatives.Policy reforma are essentíal: they can help to mobilize more local resourcesand to reallocate existing resources to their most efficient use. Rut bythemselves policy reformas will not be enough to reverse the deterioration.Additional resources--from donors as well as African governments--will beneeded if the initiatives outlíned in this chapter are to succeed. Theamounts are smaLl in relation to total resource requirements, snd in astrategy of adjustment with growth, allocations to these sectora should beprotected, even when there is an overall resource constraint.
- 30 -
4. External capital and aid coordínation
Medium-term programs of growth-oriented adjustment are e3sental ifAfrica's decline i8 to be reversed. But the effectiveness of programa thatare already in place and the further spread of such reforma are threatened byinadequate external resource flows. Adjustment programa need additionalfinancing to meet urgent needs for raw materials and spa.ne parta to increasethe use of existing capacity. External financing is also needed, along withincreases in domestic savings, to raise investment rates, which are now lowerín Africa than in any other region of the developing world.
The scarcity of foreign exchange imperils the success of adjustmen programa in country after country. Because of it, administrative controls arereta.ned, spending on maíntenance and rehabilitation is zqueezed, and existingcapacity ja not fulty used. More important, without extra resources, theprice changeo caused by greater reliance on market mechanisms can have asericus social and political impact; this, jn turn, strengthena those whoargue for continued public intervention in allocating resources.
The external capital required to support these programs in low-incomeAfrica is not large in relation to the needs of highly indebted middle-incomecountries. However, most of it must be granted on a concessional basis, sinceseveral of Africa'9 IDA-eligible countries have prolonged debt problems, andall have limited credítworthinesa. Official flows require budgetaryappropriationa. These can be made only through a political process in thedonor countries. As a result, obtaining new fundo from official sourcesraises íssues which are different from those that face countries wíth accessto commercial sources. For instance, commercial lenders provide freely usableforeign exchange, while official lendera often concentrate an projectfinancing to ensure political identification with the funda provided. Andwhen official lenders provide import fínancing, it may be tied, thus requiringthe recipient country to maintain import controls, in order to enforce donorprocurement regulations. Moreover, unlike commercial lendera, donors oftenhave. their own view of the priorities for which recipienta can use thefunda. Another complication ja that in low-income Africa a large part of debtservice is owed to preferred creditora; this límíta the acope of debtrescheduling in reducing the region's debt servicing problem.
To provide the necessary external capital in the appropriate form,political decisions are required to increase resource flows to Africa througha combination of additional aid and debt relief, and -o relax the rigiditiesconcerning the use of aid. This chapter discusses these issues withparticular attention to the resource requirements of the next five yeara,1986-90.
External resource flows, 1986-90
As earlier Bank reports on Africa have emphasized, external resourceflows cannot guarantee development. Without appropriate policies, no amountof external assistance can reverse Africa'a decline. At the same time, policy
- 31 -
reforms designed to promote growth will be unsustainable without additionalresources. Quantifying the resource requírements of particular countríesseveral years ahead ¡a difficult at the best of times. The problem now ¡sgreater than usual, because massive shifts in relative prices and in patternsof resource use in the past few years have overturned the historicalrelationships that might be a basis for calculation--and these shiftacontinue. The prospects for commodity prices, the future trenda in exchangerates, and the outlook for the world economy are all uncertain. tt isimpossible to say precisely how many countries will develop credibleadjustment programa and thus be eligible for additional resources.
Despite these uncertainties, this report estimates the minimumrequirements for IDA-eligible countries &s a group. The basic objectiveunderlying these estimates ¡a to halt the trend of decline in per capitaconsumption by 1990 and achieve some growth thereafter. Because of the needfor higher savinga to support essential investment, and because all Africancountries have rapidly growing populations, tLis objective can be achievedonly if the GDP growth rate reaches at least 3-4 percent a year by 1990. Thereport estimates external resource requirements during 1986-90 by sssessingthe minimum importa needed to achieve investment levela consístent with suchgrowth rates in light of the maximum exports and domestic savings feasíbleduring the period. These are minimum estimates for IDA-eligible countries asa group. Though some of the countries may fail to develop credible programsto be supported with additional assistance, others might need more than theminimum to support faster growth.
This report goes on to estimate the mínimum flow of concessionalassistance needed to pay for imports and debt servicing--after allowing forthe maximum feasible effort to increase export earnings, continued favorabledebt relief. and the highest level of nonconcessional flows that would beprudent. Low-income Africa will need concessional flows of $11 billion a yearduring 1986-90 to continue the proceso of growth-oriented adjustment. Of thetotal, about $8.5 billion a year can be expected on the basis of known andprojected commítmente. This leaves a gap of $2.5 billion a year to be filledby additional multilateral and bilateral asaistance, including debt relief.
Poreian exchan&e_rE!ql!iementa
The Bank's experience suggests that in much of low-income Africainadequate import capacity *s thwsrtíng adjustment programa. Real importa percapita have been declining for low-income countries as a group since the 1970s(see Table 4.1), and the rate of decline accelerated in the 1980s. To financegrowth-oriented adjustment programs, and to lay the foundation for sustainedand more rapid growth in total output and per capita consumption, thís trendmust be reversed. At a minimum, real importa per capita should return to thelevel of 1980-82. This implies importe of about $28.5 billion a year atcurrent prices during 1986-90, or about 24 percent more than in 1980-82(Table 4.1).
- 32 -*
Table 4.1. Foreign oxchange payments by IDA-eliglble countries, 1980-90
(bltions of dollara)
1986-901980-82 projected
annual average 1983 1984 annuel average
Imports of goods and servíces (excluding Interest) 22.9 19.0 19.5 28.5 a
Scheduled debt servlce payments on public debt 3.6 4.3 4.1 5.8 b
IMF repurchases and charges 0.4 0.6 0.7 1.0
Total 26.9 23.9 24.3 35.3
Memorandum items:
Import volume per caplta (1970=100) 74 62 62 74 a
Import prico índex 100 92 90 101
Amortization before rescheduling 2.0 2.7 2.4 3.1 b
<of which: multilateral) (0.2> (0.2> (0.3) (0.6)
lnterest before rescheduling 1.6 1.6 1.7 2.7 b
(of which: multilateral) (0.2) (0.2> (0.3) (0.4)
Total debt servlce before rescheduling 3.6 4.3 4.1 5.8 b
(of which: multilateral> (0.3) (0.4) (0.6) (1.0)
A. Proposed objectíve.
b. Scheduled debt servlce payments on existing and expected new comíitments Including
reschoduled 1oans.
Scheduled debt service payments, plus payments to the IMF, areprojected to rise to about $6.8 billion a year during 1986-90, as againstscheduled payments of $4.0 bíllion a year (and actual payments of $3 billion ayear7 in 1980-82. The situation will be particularly serious forabout a dozen low-income African countries (Benin, Gambia, Liberia,Madagascar, Mali, Mauritania, Niger, Somalia, Sudan, Tanzania, T*go, endZambia). For these countries as a group, acheduled debt service payments in1986-90 will be over $2 billion a year more than they were ín 1980-82.
- 33 -
Sources of external finance
The principal source of financing for external payments is exportearnings. Adjustment policies along the Ulnes discussed above should heLp toraise exports. However, because of the region's economic structure, exportawill be relatively slow to respond. They are mostly primary commodities,which have limited prospects. For several kinds of exports--especiallymínerals such as copper, bauxite, and iron ore--prospects are poor both forprices and for volumes. And nontraditional exports start from a very lowbase. Therefore, it will be very hard to achieve a rapid expansion of exportsin the near term. A country-by-country analysis indicates that exports incurrent prices during 1986-90 could, at best, be about 25 percent higher thanin 1980-82. And this assumes export-oriented policy reforms in Africa and noincrease in protectionísm in the industrial countries.
In order to meet Africa's growing foreign exchange requirements,resource flows from external sources will need to increase during 1986-90, butthe recent trend has been just the opposite. In 1984, gross capital inflowsto low-income Africa from all sources were about 15 percent lower in currentprices than in 1980-82. Thíi was due to a decline in nonconcessionaL flows,in particular private capital flows (Table 4.2).
Nonconcessional capital flows were more than 25 percent of totalgross capital flows in 1980-82; however, with increasing debt servicedifficulties, they have been declining steadily. The decline was particularlymarked for private commercial flows, which dropped from about $1.6 billion ayear during 1980-82 to about $0.4 billion in 1984. Nonconcessional flows fromofficial sources (bilateral and multilateral) have also declined, though notso aharply. Given the weak debt-servicing capacity of most low-income Africancountries, it would not be desirable for them to obtain more than about $1billion a year in nonconcessional funds during 1986-90.
Concessional flows to low-income Africa were largely stagnant during1980-84; worldwide concessional flows declined during thís period. As aresult, the ahare of low-income Africa in global net ODA disbursements went upfrom about 19 percent in 1980-82 to 21 percent in 1984--a small butcommendable shift of resources to Africa. During 1985, food aid deliveriesincreased substantially (about double in volume compared with 1984), butnonfood aid may have declined because of famine-related disruptions. On thebasis of known and expected levels of commitments --including the World Bank'sSpecial Facility for Sub-Saharan Africa, the increases under the Lome IIIConvention, and the fourth replenishment of the African Development Fund(1985-87)--concessional flows during 1986-90 would be about $8.5 billion ayear, some 25 percent higher than 1980-82 levels.
In recent yeara, debt reschedulinga under the Paris and London Clubshave shown considerable flexibility. For countries with serious debt-servicing difficulties and credible adjustment programs, rescheduling of85-95 percent of principal has been common, and it has sometimes covered 100percent of the principal and sometimes amounts due from previousreschedulings. Rescheduling by bilaterals of 50 percent or more of interestpayments due has also become common.
- 34 -
Table 4.2. Supply oQ external finance for IQA-eligible countrles, 1980-90
a. Based on disbursements out of existing or currentiy expected commitments.
Assumes $10.5 bilíion for IDA-8 and 1 percent a year increase in realterms for bilateral grmnts and loans between 1980-82 and 1986-90.
b. Includes direct foreign private Investment, net short term capital. IMFpurchases, and, for the historical data, changes in reserves and errors and omissions.
A large number of African countries wíll have to undertake furtherrescheduling of debt service to private and bilateral official creditors in1986-90. If reschedulings cover 100 percent of principal and$ in the case ofofficial creditors, 50 percent of interest due (excluding payments onpreviously rescheduled debt), they can provide relief of about $2.3 bíllion ayear for IDA countries es a group, of which about $1.5 billion a year would befor the countries facing prolonged debt problems. Reschedulings along theselinea are consistent with recent practice, though they will need to be appliedto more countries than heretofore. Creditors must also continue to tailorrescheduling actions to individual country circumatances. Even after such
- 35 -
relief, de4t service will remain at over 25 percent of export earnings during1986-90 for the IDA-eligible countries. For the dozen countries with the mostserious debt problems, the ratio will remain above 35 percent during 1986-90and beyond (see Appendíx A for greater detail).
Other nonconcessional flows--such as IMF purchases, direct foreignprivate investment, and short-term capital flows--alsk. declined during 1980-84. Purchases fróm the IMF dropped from about $1.2 bíllíon a year in 1980-82to about $0.6 billion in 198i. The long-term nature of the balance ofpayments problema of many of these countries, and the revolving character ofIMF resources, make it reasonable to assume that new purchases from the IMF(excluding Trust Fund reflows) will remain low during 1986-90. As adjustmentprograma are introduced, they should attract direct foreign privateínvestment; such inflows can be expected to increase over time, but theirvolume is likely to remain modest in 1986-90. Similarly, inflows of short-term, trade-linked capital are likely to remain low. At best, these threesources can provide a further $1 billion a year during 1986-90.
To sum up: after taking into account the maximum feasible effort onexport earnings, the maximum prudent level of all nonconcessional flows, andcurrently committed or expected flows of concessional assistance and debtrelief, only $32.8 híllion a year of external finance can be expected during1986-90. External payments of $35.3 billion a year and inflows--withoutfurther action--of $32.8 billíon a year leave a gap of $2.5 bíllion a yearduring 1986-90 (Table 4.3). To bridge thís gap, concessional flows shouldrise from $8.5 billion a year to $11 billíon a year over the period.
Table 4.3 Concessional resource gap for IDA-el igible countries, 1986-90(bit <lons of do ars)
1986-90projected
annual average
Foreign exchange payments required 35.3
External finance in the absence of further special act;ons 32.8
Gap to be filled by additional concessionalassistance and debt rolief 2.5
Possible edditional fínance from multilateral agencies 1.0
Gap to be filled by additional bilateral concessionalassistance and debt relief 1.5
With concessional flows of $11 billíon a year, nonconcessional flowsof $2 billion a year, and actual debt service payments of $4.5 billion a year,external flows net of debt service payments wilL be $8e5 billion a year. This
- 36 -
ís the amQunt of external assistance that will be avaílable to supplementdomestic savings. Such external assistance would amount to 10-12 percent ofGDP for these countries as a group. Assuming that with improved policies thedomestic savinga rate can recover to 8-10 percent of GDP by 1990, this wouldpermít inveatment at about 20 percent of GDP, a rate which is essential for aCDP growth rate of at least 3-4 percent a year by 1990.
Bridgiflgthe resource «ap
Because of the time lag between commitments and disbursements, asubstantially larger increase in commitment authority will be required tobridge the resource gap of $2.5 billion a year. Since the additional financeis needed to meet import and debt service requirements in 1986-90, most of itwill have to be in a quick-disbursing form. However, donors should alsodirect additional assistance to projects that are addressed to the long-termconstraints on development discussed in Chapter 3.
These needs should be taken into account in decisions on the IDA-8and the next African Development Fund replenishments, and in decisions on theuse of the SDR 2.7 billíon expected from IMF Trust Fund reflows during1986-91. It has already been agreed that the Trust Fund reflows should beused to provide additional balance of payments assistance on concessionalterms to the low-income countries in sub-Saharan Africa and elsewhere that areeligible for IDA resources, are in need of such assistance, and faceprotracted balance of payments problems. These funds can go far to meet theneeds of the African countries with prolonged debt and balance of paymentsdifficulties. If there is, in addition, a positive outcome on funding for themultilateral lending institutions (including an IDA-8 replenishment of atleast $12 bíllion), these sources could probably meet about $1 billion of thedisbursement gap.
The remaining gap of about $1.5 billion a year would have to befilled by additional bilateral aid and debt relief. This would amount toabout a 30 percent increase in bilateral assistance and debt relief over 1984,or 20 percent above the levels currently projected (Table 4.2>. Suchassistance can be provided by a combination of additional fast-disbursing aidand debt relief. Each package will, of course, depend on the individualcircumstances of donor and recipient. However, as a generel rule, no donorcountry should be a net recipíent of resource flows from an African countrywhich is undertaking credible reforai programs. This is particularly importantfor countríes whose aid agencies want to maíntain a preferred creditorstatus.
Additional bilateral assistance could be provided in differentvays. For example, donors could retroactively convert existing official loai¿sinto grants--as proposed in the UNCTAD declaration of 1977; enable theirexport credit agencies to refinance scheduled debt service payments onconcessional terms; widen the group of creditors that participate ínrescheduling of principal and interest; and increase their aid budgets,particuLarly in the form of balance of payments support for adjustment
- 37 -
programa. Sweden provídes an example of the latter approach. The Swedishgovernment recently added to its aid budget a special supplement for use byrecipiente to repay commercial credits.
Institutional reforma in aid coordination
These increases in aid and debt relief will only be enough to supportminimum import levels. But much still needs to be done to increase theeffectiveness of aid in support of comprehensive adjustment. This sectiondiscusaes the key areas for attention and reform.
Recent progress
The creation of the Special Facilíty for Sub-Saharan Africa in 1985was an endorsement of the principle that donors should focus their aid,especially nonproject assistance, on countries that are willing to adoptoverall or sectoral adjustment programs. It has also shown how bilateral andmultilateral agencies can work together to support policy reforms.
At recent consultatíve group and other aid coordination meetings,increasing numbers of donors have recognized the need for additioralnonproject aid. Ghana and Senegal are cases in point. Bilateral aid agencieshad long been reluctant to join the multilateral ínstitutions in providing anadequate volume of nonproject assistance to Ghana. However, at theconsultative group meeting in November 1985, several bilateral donorsindicated that they were willing to increase their nonproject assistance andto cofinance with the multilateral agencies. In Senegal the two mainsuppliers of nonproject assistance, France and the United States, have beenworking closely with the Bank and the IMF to províde support for major policyreforms.
Another recent development is that many bilateral and multilateraldonors are concentrating on projects that maintain or rehabilitate existingcapacíty. This is particularly true of expenditure on infrastructure, large-scale irrigation, and public enterprises; lately few such projecta have beenwholly new investments. Increasingly, donora are cofinancing rehabilitationschemes. Examples include a major agricultural project in Zambií and a powerproject in Senegal.
Donors are trying to improve their project portfolios by cancelingnonperforming projects and reallocating funds to areas of higher priority.For example, in 1983-85 the World Bank increased disbursement rates andredirected funda to areas of high priority in seventy-five projects in twenty-five countries of sub-Saharan Africa. At a recent consultative group meetingfor Zambia, the aid agencies of several countries, in particular of theFederal Republic of German-, announced their intention to redirect part oftheir committed assistance toward higher priority projects and quickdiabursing operations. In Níger, the government and several donore haveagreed to scale back construction of a major trunk road and to focus insteadon higher priority activities such as road maintenance.
- 38 -
In the past two years, donors have expanded their formal arrangementsfor coordinating aíd, and their use of consultative groups and roundtables hascontinued to increase. In 1983-84, the pace of consultative group activityincreased substantially, and meetings were held in eight countries. In 1985consultative groups were established for the first time ín Mauritania andSenegal. Roundtables began in five countries in 1984 and in another three in1985. As a result, by the end of 1985, some kind of formal aid coordinatingmechanism was established or agreed in principle for all but one of the low-ancome African countries (the exception ís Ethiopia). The Bank and the United
Nations Development Programme aim to provide similar documentation for bothconsultative groups and roundtables.
Aid coordínation has also been strengthened at the local level. Forexample, special meetings on aid for selected sectors have been held--usually-under the sponsorship of a lead donor--in níne countries (Benin, Burundi,Kenya, Madagascar, Malí, Rwanda, Senegal, Tanzania, and Zaire). Similarmeetings are scheduled in eleven others. In eight countries (Ghana, Guinea-Bissau, Lesotho, Senegal, Somalia, Sudan, Zaire, and Zambia>, localrepresentatives of major donors also hold regular meetings--often under thechairmanship of a lead donor and in conjunction with a government aid steeringgroup--to discuss broad issues of aid and economic policy.
These are promising signs. But the progress toward more effectiveuse of aid needs to continue. There are at least six areas in which furthersteps should be taken now to strengthen the system.
Further reforma
First, governments in Africa must be seen to have the primeresponsibility for designing their adjustment and investment programs and forcoordinating aid and other financial flows. Donors can assist in the task--but they should not undermine this responsibility by trying to negotiate theirown favoríte package of policy reforma or by promoting their own petprojects. To this end, the African countries must strengthen their coreministries of finance and planning, and the unita that coordinate theirforeign assistance. Better coordination within the government--between thecentral bank, the ministries of finance and planning, and the sectoralministries--is also necessary. Such improvements may call for technicalasaistance from outside. Some countries have made progresa in this area (forexample, Zambia> but most 9till have a long way to go. Changes in donorattitudes and programa are essential to ensure that theír assLtance fits intothe national development priorities.
Second, decisions on aid and debt relief are not yet adequatelyintegrated. The current pattern of discussions means that aid levela areindicated before agreement is reached on debt relief. As a result, larger aidcomíitments have often allowed creditors to harden their terms for debtrestructuring. In such cases, the increase in resources necessary for theeconomy to grow out of ita problems never materialized. In Latin America's
- 39 -
debt negotiations, fresh money ard the restructuring of existing debt arenormally considered together. In Africa, donors and creditors are notnecessarily one and the same, but the principle of simultaneous determinationís just as important. Debt rescheduling for low-income African countries willbe needed regularly for many years to come, so debt relief and aid must beviewed as elements of a single financing plan in support of growth-orientedadjustment. The responsible donor country institutions can then work out thespecific details of aid commitments and debt relief, as at present. Over thenext five years, increases in aid--and the expected rise in export earníngs--should be used to support growth, not to reduce debt relief.
One vay to integrate the two financíal decisions is to bring togetherthe different representatives handling aid and debt wíthin creditor and donorgovernments. This was done in India in 1968 and 1972 and in Pakistan in 1973and 1974; debt relief was committed as part of the annual aid-pledgingexercise. In 1981, Pakistan's aid consortium meeting focused on balance ofpayments problems, and a part of the meeting was devoted to a discussion ofdebt relief--under the chairmanship of the French representative, who waschairman of the Paris Club. More recently, in Africa, the Mauritania meetingof March 1985 considered financing needs for 1984-85 and adopted an integratedpackage of aid and debt relief. In this meeting, Arab donors evaluated debtrescheduling and fresh money requirements together; and some OECD donorscommitted new aid and at the same time indicated their willingness toreschedule debt at the subsequent Paris Club.
Third, consultative group meetings have sometimes been so large thateffective policy focused discussion became almost impossible. Such groupshave worked best when the major donors have had discussions with thegovernment, the IMF, and the World Bank in advance of the formal meetings.This enables the major partíes--not least, the government--to recognize whatls required of them in devising a consistent internal and external financepackage. Such prior consultation míght be essential if the aid and debtdiscussions are brought together. Private banks have formed small steeringcommíttees to address the debt problems of the middle-income countries. Pormost of the major African debtors, they have also established small steeringcommittees led by one or two banks. In the case of countries with prolongeddebt and balance of payments difficulties, detailed discuasions on debt andaid will be required; prior consultation between the principal agencies willtherefore be particularly important.
Fourth, Africa's attempted policy reforms are likely to fail unlessdonors can give indications of the resources that will be available in themedium term-national programs of medium-term adjustment cannot be formulatedwíthout this informatíon. At present, creditors will reschedule only the debtservice that is due over the following twelve to eighteen months. As a rule,donors do not commit themselves to financial support for more than a yearahead. Often they lack legislative authority to make multiyear commitments.But even a nonbinding indication of aid to come could provide help in medium-term planning. Donors have already given multiyear pledges as part of the IDAreplenishments. In a recent consultative group for Ghana, at least onebilateral donor (Canada) was willing to make multiyear indicative pledges. Totackle the special problems of Africa, other donors must do the same.
- 40 -
Fifth, follow-up to the agreements reached at debt and aid meetingaleaves much to be desired. Too often, it takes a year after a Paris Clubagreement to finalíze the arrangements for bilateral debt relief. Donoraoften fail to abide by consultative group agreements on the volume andcomposition of assistance. Moreover, there is no procedure to identify suchahortfalla quickly.
As proposed ín the 1984 Africa report, consultative groups shouldestablish appropriate monitoring capacity. For instance, a one or two personsecretaríat could be assigned to the office of the Wor d Bank's residentrepresentative to work closely with the host government in monitoring theimplementation of consultative group agreemento and bring any developmentsthat might be inconiístent with them to the attention of the group. Such astaff function might suitably be financed jointly by donots.
Sixth, multilateral agencies must play a central role. Increases inaid should be largely in the form of nonproject assístance--for reasonsalready explained. This makes coordination through multilateral agencies allthe more important. The Bretton Woods institutions are already using more oftheir resources to support adjustment programs. The European Community,acting under the Lome III agreement, has started to consider sectoral policiesalongside its assistance programs. The African Development Bank is beginningto direct more resources to quick-disbursing assistance in support of policyreforma--as in ite recent support for Ghana's economic recovery program.These are welcome signs of an increasing emphasis on quick-disbursing, policy-oriented tending. But there is a risk of fragmentation and inconsistency inpolicy reform programs negotiated by different agencies. To avoid this, thecountry's own medium-term adjustment programs and its action program for thenext twelve to eighteen months should be the focus of adjustment lending bythe World Bank and other agencies.
Cooperation between the World Bank and the IMF should continue to bedeepened. The Bank and the Fund should work together with governments, first,to develop an adjustment and investment program aimed at restoring growth, andsecond, to assess financial needs and sources. Programs of this kind wouldthen become a basis for financial support--which would be agreed inconsultative group meetings and delivered in the form of concerted, policy-based lending.
In conclusion, 1986 offers several oppportunities to meet the Africanchallenge. Decisiona on IMF Trust Fund reflows and the IDA-8 replenishmentwill help to determine the resources available for low-income Africa during1986-90. The Development Committee's meeting in April and the apecial sessionof the UN General Assembly on Africa in May can establish the principles onwhich the concerted effort of the African countries and the internationalcommunity can be based.
- 41 -
Appendix A
The debt problem and ita implications for import capacity
Duríng the 1970s, IDA countries in Africa borrowed heavily despite arise in their export earnings and increasing ODA. Between 1970 and 1980,their external debt rosç by over 21 percent a year (Table A.1)--more rapidlythan in Latin America. - Several countries increased their debt ten times ormore in these ten years. In some cases, the debt roughly tripled in two tothree years. The growth rate of debt slowed down considerably during the1980s (to 7 percent a year during 1980-84). But purchases from the IMFincreased rapidly and arreara on short- and long-term debt accumulated(Table A.2).
Profile of the African debt problem
At the end of 1984, the low-income African countries' debt (includinglong-term and short-term arrears and IMF purchases) stood at about $45billion. For this group of countries, total debt represented 74 percent oftheir GNP and 349 percent of their exporta. Both ratios were considerablyhigher than those for major borrovers among developing countries (which were51 percent and 239 percent, respectively). For several countries, theexternal debt was more than 100 percent of GNP and 300 percent of exports.<Table A.2).
A large part of the outstanding debt of the IDA-eligible countries atthe end of 1984 was owed to official creditors, often on concessional terms.As a result, despite the heavy burden of debt, the scheduled debt service onlong-term debt (plus payments to the IMF) was 35 percent of exporta in 1984.However, the actual debt service ratio was only 22 percent, because ofreschedulings and the accumulation of arrears. This distinction can be veryimportant: some countries had scheduled debt service ratios that ranged from80 to 145 percent, while their actual debt service payments ranged from 20 to40 percent.
The composition of the debt service reveals two interesting points(Table A.3).
O The predominance of official creditors, especially multilateralagencies. Official lenders plus the IMF account for more than 75 percent ofdebt service due in 1986-81 for the IDA-eligible countries. Of the 75percent, 20 percent is due to the IMF, and 17 percent to other multilateralagencies. For some countries, multilateral lending agencies (plus the IMF)account for 50 percent or more of the debt service. Since neither payments to
1. The tables in this appendix present data on debt Lor thirty-seven ofthe principal borrovers of sub-Saharan Africa, although the discuasionconcentrates on twenty-five IDA-eligible countries.
- 42 -
the IMF nor debt service due to other multilaterals can be rescheduled, thísreduces the scope for debt relief.
O The great díversity of bilateral creditors. For some countries theprincipal bilateral creditors are countries represented at the Paris Club, butfor othere (such as Guinea, Mali, Mauritania, Somalia, Sudan, and Guinea-Bissau) the principal bilateral creditors are OPEC members or centrallyplanned economies. Since these creditora have different procedures for debtrelief, the case-by-case approach is particularly important for Africandebtora.
Projected debt service burden and import capacity
For many African countries, the reschedulings and arreara arebuilding up obligations for the future. The implications of Africa's debtburden become clear when the projected debt service payments are seen inrelati,n to the export prospects. Country-by-country projectiona wereundertaken to aseses the scheduled debt service burden in the rest of the19809 and beyond. The scheduled amounts are defined as the payments due onexisting debt and new credits, including rescheduled debts (Table A.4).
The projections of exports of goods and nonfactor services were basedon export projections done for the World Development Report 1986, which startfrom a common set of external assumptions, including developed-country growthrates, primary commodity prices and world demand, prices of manufacturedimporta, and changes in the rateo of exchange of the major currencies. Theprojections aiso assume that the sub-Saharan African countries will adopt muchbetter domestíc policies than those of the past ten years. Exports wereprojected under the assumption that there would be no increase inprotectionism in developed countries.
Gross disbursements of external capital were asaumed to be maintainedat 1980-82 levels in real terms through 1990. The distribution of newdísbursements among concessional and nonconcessional creditora was assumed tobe the same as the distribution of disbursements in 1984. As a result, theterms of the projected new commítments are, in general, somewhat softer thanthe terma of the 1980-82 commitments. New IMF purchases were assumed to behalf the level of repurchases in each year. The projections assume thatarreara are rescheduled on the same terma as other debt and, for countrieswithout debt rescheduling, paid off in equal annual installments over 1986-90.
Reschedulings were assumed to take place in most of the cases whereprojected debt service ratios significantly exceed those of the recent past--the earlier ratios were taken as a rough indicator of a country'e ability toservice debt over the next five years. No reschedulings were assumed forcountries where projected capital inflows would in any case result in asubstantíal increase in import capacity. On this reasoning, seventeer. of thetwenty-five IDA-eligible countries were projected to reschedule debt between1986 and 1990; most of them have already rescheduled their debt at least onceduring 1980-85.
- 43 -
The terma assumed for the reschedulings were:
o A ten-year repayment period, including five years of grace
o Interest ratea at or near market rates for rescheduled private debtand for the claims of export credit agencies
o Interest rates at their original level for rescheduled ODA loana
o A contract cutoff date of December 31, 1984.
Up to 100 percent of principal and 50 percent of interest payments--includingarrears and debt rescheduled before 1985--were assumed to be rescheduled. Inthe case of private financiai credits without external guarantees, however,only principal payments were assumed to be rescheduled. Debc service paymentson debt contracted or rescheduled during the 1985-90 period were not assumedto be rescheduled.
While these terms are relatively generous, they are roughly in linewith those agreed at recent Paris and London Club reschedulings for sub-Saharan Africa-n countries. A single set of terma vas used for convenience.Creditora have sometimes granted more generous terms, and this may benecessary jn the future to provide sufficient debt relief for certaincountries. In other cases, less generous debt relief may be adequate.
For twelve countries in sub-Saharan Africa the scheduled debt serviceratios ín 1986-90 are projected to be significantly higher than the actuallevels in recent years (Table A.4). In these cases, debt rescheduling jslikely to continue for some time--as indicated by the high debt service ratiosprojected for 1991. Moreover, their import capacity in 1986-90 is likely tobe lower than jn 1980-82. even after reschedulings and even if grossdisbursemento of external capital during 1986-90 are maintained at 1980-82levels. Without additional capital flows or additional debt relief, theirimport capacity wil1 decline sharply.
For the other IDA countries, the debt problem is more manageable.Their average scheduled debt service ratio in 1986-90 js hígher (28 percentathan in 1980-82 (19 percent>. However, rescheduling operations are likely toreduce the actual average to 21 percent in 1986-90; and the acheduled debtservice in 1991 ¡a projected to remain at this level, relative to exporte.With gross capital inflows at their 1980-82 level, their importa of goods andnoninterest services would rise by about 13 percent ín real terms between1980-82 and 1986-90. When population growth ¡9 taken into account, thisimplies an 11 percent decline in per capita import capacity. As noted inChapter 4, resource flows to low-rncome Africa will have to íncreasesubstantíally during 1986-90, i£ the minimum objective of restoring per capitaimports to their 1980-82 level is to be achieved.
Table A.¡. Lonr-ter.. pubílc and pua911e17 guaranteed debt, disburiad mnd1 outatandíng(alilItona of dollara)
Niote: Data are for tha cod Of II.Year. Angola ano qpzaabfque are excludad frol. ail the table.i In Appadtx A h,rsusem romprehe-nsíve d'.'e are nol ag*1sbI-.,a. Period growt:h ratea £annot be ealculatad beausne thar, vas no dabt ¡a ¡970.
?sb1. 8.2. 89te~oe1 debt by coe." st tbb sed .f 1084
(aríllima, of dol*¡6e)
N610o. cd ~s141.1 .cc0000ed Ioo2wttml, ______ _____t___ f*tíe,tc,1 orrOAe On t¡t e.
e. moletd. a cmcli MOct *t pOicto. <1*800In frm o9 eeefíe.. t .etOtotlo.e.b. Lroecr. oc 1088-.0 Ptte04pc pay*caco arelled¡o ¡ocg4 " ~.r.o debt. c,.d oc~ .*~t-teom ~y80e. h. le-Ided 10.6#*rt-t.0 debO: coe r.Colt 7,1, ttlar .,t coedob.ocol
Table A 1. Debt eerricef palmants (latrgodíng 7tIF reporcha nwd cUergeo) 4ue on pubitc deb- outatrapttrg 4t che en4 of 1984
1984 debt ""lee Scbedulod 4*bt_sebeduled mettsi42 _ e_.7tt 1986-87 Percont4aRt sb4re oí total debt eervtee due In 19bó-87
A4 Ao An percentegep-rcentxgs AA perecntace of projaferd _ _ ,_ Oflli _ P Y t¿_of OxporUe p rcent- of exporte export. of Flvinenetii
of good *nd *Se of of goode pa4 Militóa, goode end DILAter l IltugletersLsipilr Instíttu*arytceo CYP *rvlees of d.11tro *."le* DAC VPESC 0ther. TotAl 1131t IDAb thr O tth ~d4r. enon llO'
tiot-: Debt oerrvíe excludes interese on bhort-term debt enn4 rreero *nd p 7seto on prí.ate nonguerantet-d debt. te MAY* .. t ádd c. totul ner,sus of roundl.R.;-. nterept *nd exchanse rete fluctustioen eo we11 *s reuchedultngs debt forgtvences, end tbr eceuuletton of arreara aff-ct the dtfferon,e bet~een ochoduled dobt servce .. id 41,inal
pA~_td.
t bl- 4.4. Total debt *ece .<neludáao *13 repor t es sad ít,gea) dm projeeted tmport cpaottp of IDA-eitglbie *ob-9-be afrio. 'oootrte.<auml o0erte tu *lliton of ¿ollar.>
AMtul dabt ser* ce. 1980-82 Schoduled debt ervioe. 1986-90 Prletd dbt srvie. 1986-91> S hed.led 4eb, o.rvioe. 9I91As par- LA par- As per- *o par- ovi
reataG of ent.5e of entage of Onta3eg o1 $.porto of gooda andexporto of exporto of exporta of exporte of aoni.tera.t *ervicesgnod nd goode *nd goods and 4odol ond In 1980 dollart
Interear prioepaI l.tal aerwoea loeremO Procípial Total or---o Intereot Prnclpael Tbtotal rvloaa Tnterfet PVI-rlipl Total oervt ed 9rO-82 1986-10
bte: Angola is exclude fron all aggregates becauee of missing data. Grawth rates for sumary measuresare caLeulated from agrgates In US dollars, exrept the CDP deflator, which is omstructed fron loacurreccy deflators, weigted by CDP in 1980 US dollara and GrY, which is derived directly fron samauryseaaires. For definitions and aources, me the technical notes to the Statistical Annex.
a. Cbnstant prices.b. toss damiRtic inxmme (GDY) ls derived from per capita CDP adjusted by gains or loses fra the terns
of trade.c. These grawth rates are derived from UN Trade Data in cornzt US dollar deflated by appropriate price
indexes (exports by welghted unit price Indexes constructed from the Wbrld Bank's primary cmnmodity priced<at antd tfe iiwnu£acturing unit value ir~ex, and ~imqorta by lmplicit Import deflators). Unit value s_riesare caleluU from the dollar series in current and constant prices. x1cudes Cad, QGnea, GbjnearBissau,
- 49 -
T~bie B.2. Key uauroenolc ratios, 1960-84(aml avere es percte of CDP)
ID-eligible acuntries .. .. -6.1 b -6.9 -7.1 -7.5 -7.9 -6.3 -5.6Otier sub-Saharan Afrca .. 4.4 b -6.0 -1.8 -7.6 -7.2 -9,0 -4.5Total ahSahrn Africa .. .. -S.1 b -6.3 -3.7 -7.6 -7.5 -8.0 -5.0
Nte: ~ga la iewhc f ran the aggregates. The averages are calculated from aggregaes expressed inUS doAlIra, exopt for t1e ontral gowerorumt deficit, which i í based aon a try veges wet by GDP in1980 US dollar.
a. Exlim net off lca]. tranafers. BEblules Qai, Guinea, Ginea-Bissa, 1.bzaque, ad 9wazilard.b. us are for 1977-79.
Nbte: Angola is excLxted frum the aggregates.a. Figuras differ sllghtly f ro those in Table 4.2. (bncesslonal flows here exclude grants not allocated on a country basts, and
tmooessa flo. bere incluie private noraateed short-term loans that have been comversed lnto uediu- anl longterm loass.b. Public &d publicly guaranteed nediun- ad log-term loansad rnuaranteed private nediun- and 1ongterm loans.c. bcludes IMF purchl.d. Deccludes Chdu, Ginea, lisBisa, bzambmique, and Swazil.e. Levels art eri of the year.
"4uu
.pw
1 ~~~uu19,m
»~ l>,c,Q /á ks./
_. -
Iatroductiou 54
Tables
1 Basic indicators 51
Production2 Growth of production 583 Structure of production 594 Growth of consumption and investment 605 Structure of demand 616 Commercial energy 62
Trade7 Level and growth of merchantdse trade 638 Structure of merchandise exports 649 Structure of merchandise importa 6510 Origin and destination of merchandise exports 6611 Termsa of trade 6712 Commodity exports: Volume and price 68
Aid, debt, and capital flows13 Balance of payments, debt service, and international reserves 6914 External public debt and debt serv;ce 7015 Outstanding and disbursed external debt of sub-Saharan Africa 7116 Gross disbursements of external loans to sub-Saharan Africa 7217 Average terms of borrowíng for sub-Saharan Africa 7318 Indicators of aid and total resource flow, 1984 7419 Disbursements of official development assistance 7520 Food aid imports 76
Agriculture21 Growth of agriculture 7722 Production of major crops 7823 Agricultural imports 7924 Major agricultural exports: Growth and shares 80
Social indicatore25 Population growth and projections 8226 Demographíc and fertility-related indicators 8327 Labor force 8428 Urbanization 8529 Health-related indicatora 8630 Education 87
Fiscal data31 Central government expenditure 8832 Central government current revenue 89
Technical notes 90
Principal sources 106
- 54 -
Introduction
The Statistical Annex provides information on the main variablesaffecting social and economic development ín sub-Saharan Africa. The tablesshow data on production, trade, aid, debt, capital flows, and agriculture, aswell as basic indicators, socíal indicators, and fiscal data.
Considerable effort has been made to atandardize the data;nevertheless, atatistical methods, coverage, practices, and definitions differwidely. In addition, the weakness of the statistical systems in most of theAfrican countries affects the availability and reliability of the data.Readers are urged to exercise caution and to take these limitations intoaccount in ínterpreting the indicators, particularly when making comparisonsacross economies.
This annex covers thirty-nine countries in sub-Saharan Afríca with apopulation of more than 0.5 million. They are divided into two groups: low-income (per capita GNP of $400 or less in 1984) and middle-income; for thelatter, a distinction is made between oil ímporters and oil exporters. In thetables, economies in each group are listed ín ascending order of GNP percapita; Angola, Chad, and Mozambique, for which GNP per capíta has not beenestimated, are listed at the end of their groups. The reference numbers belowreflect the numerical order of each country in the tables. Asterisks indicatethe IDA-eligíble countries referred to in the main text.
Summary measures--totals, median values, or weighted averages--werecalculated for the country groups only if data were adequate and meaningfulstatistica could be obtained. The weights used in computing the summarymeasures are described in the technical notes that follow the tables. Theletters y, m, t indicate weighted averages, medians, and totals,respectively.
Worldwide averages or median values for low-income, lower middle-income, end upper middle-income and industrialized countries are included forcomparison at the bottom of some tables. Countries included in the fourgroups are as follows:
1. Low-income countries (per capita GQP of $400 or leas in 1984)
Afghanistan Haiti PakistanBangladesh India RwandaBenin Kampuchea, Democratic SenegalBhutan Kenya Sierra LeoneBurkina Faso Lao People's Democratic SomaliaBurma Republic Sri LankaBurundi Madagascar SudanCentral African Republic Malawi TanzaniaChad hali TogoChina Mozambique UgandaEthiopia Nepal Viet NamGhana Niger ZaireGuinea
2. Lover uiddte-income countries (pet capita GNP of between $401 and $1,635 ín 1984)
Angola Guatemala NicaraguaBolivia Honduras NigeriaBotswana Indonesia Papua New GuineaCameroon Ivory Coast ParaguayColombia Jamaica PeruCongo, People's Korea, Democratic Republic of PhilippinesRepublic of the Lebanon Thailand
Costa Rica Lesotho TunisiaCuba Liberia TurkeyDominican Republic Mauritania Yemen Arab RepublicEcuador Mauritius Yemen, People'sEgypt, Arab Republic of Mongolia Democratíc Republic cEl Salvador Morocco Zambia
Zimbabwe
- 56 -
3. Upper middle-íacome countries (per capita CGP of $l,636 or more in 1984)
Algeria Iraq SingaporeArgentina Israel South AfricaBrazil Jordan Syrian Arab RepublicChile Korea, Republic of Trinidad and TobagoGreece Malaysia UruguayHong Kong Mexico VenezuelaIran, Islamic Panama YugoslaviaRepublic of Portugal
4. Industrial countries (industrial market economies)
Australia Germany, Federal NorwayAustria Republic of SpainBelgium Ireland SwedenCanada Italy SwitzerlandDenmark Japan United KingdomFinland Netherlands United StatesFrance New Zealand
All growth rates shown are in constant prices and, unlegs otherwisenoted, have been computed by using the least-squares method. The least-squares growth rate, r, is estimated by fitting a least-squares linear trendline to the logarithmic annual values of the variable in the relevantperiod. More specifically, the regression equation takes the logarithmicform: Log Xt = a + bt + et, where this is equivalent to the logaríthmictransformation of the compound growth rate equation: Xt = XO (1 * r)t. Inthese equatíons, X, is the variable, t is time, and a = log XO andb " log (1 + r) are the parameters to be estimated; e is the error term. If
is the least-squares estímate of b, then the annual growth rate, r, isobtained as [antilog (b >1 - 1.
The technical notes that follow the tables outline concepts,definitions, and specific data problems. Readers are urged to refer to thesenotes for any use of the data. For comprehensive definitiona and descriptionsof concepto used, see the sources listed at the end of the notes.
Table 1. Basec tndicatore
GNP per caita LSte Average tndex ofArea AQnuel expectancy food productton
Population (thousanda growtb rete Annual tate of at btrth per ceptta(ttl1inas) of aquare Dollare 1965-84 enfgation <Percent> 1983 1981-83aid-1984 kil~letera) 1984 (pereent) 1965-73 1973-83 (yeare) (1974-76-100)
Sub-Saharan Africa 410.9t 22,207t 420v 1.8, 4.2Y 14.8v 49v 94v
A11 ll rincome economeas 2.335.4t e 31,603t 260v c 2.7* d 1.4v 5.4Y 59v 111vA11 lover middle-income
economies 665.1t c 18.446t 750v e 2.9v d 5.6v 17.9v 57v 105,Al1 upper middia-lncoma
economea SOO.lt e Z2,079t 2,050w c 3.8< d 5.3v 34.0O 65v 106vIndustital market econoSías 728.9t e 30,39$t 11.0bov c 2.5w d 3. 8.0v 76v 107v
a. Because date for the entire parSod are not available, figures are for perioda other than that apecífied.b. Pigures are for 1973-82, not 1973-83.e. ftgureas re for 1983. not 1984.d. Figures are for 1965-83, QOt 1965-84.
Table 2. Grouth of pro4uction(mual growth rete; percent)
a. Mauacturing la a part of the Industrial sector, but ite ehare of CDP la shown eeparately because it typicahly te the noatdynamc pat of the Inutrial sector.
a. Due te dífferent converSIOn prceurs these CDP dollar figues are not comparble to OlP ¿ollar fígures usa for GNP>par capita in Table 1. See Tectnical Note.
b. Mauacturing ía a part of thes induatríal sector, but ita ahara of CDP fes hown separately beeaue it typicahly is thaumst dynsmic part of the industrial sector.
e. Pigureo are for 1966, not 1965.d. Figures are for 1982, not 1983.
-60-
Table 4. Growtb of coneunption and inveatmeat<annusl1 growtih rate; percent)
a. Figures are for 1973-82, not 1973-83.b. Publíe consumption figures are not available separately; they are therefore included in prívate consgmption.c. Figures are for 1973-80, not 1973-83.
Table 5. Structure of demnd
Distributioyn of gross dmes~tc product (ereent)1'.xporta of goods
wot. a te8 the ata of A, 8, C, and D. C la the aun of e and PF Coaponenta of C do not *dd to total betauce of overlepning.a. Oaatrally pl7aniad aconontea.
73
Tsbte 17. Average terue of borrówing for sub-Sahbren 4frten
Note: Average teria are for new eoaltaente of pubítc and publiel guaranteed debt.
-74-
Table 18. Indicatora of ald ~o total resaurce floam, 1964
Total recorded Net of fldetl development aselstance (004>: Disburoeeent.0,t flov of
regourceés As parciantaga of Bilateral Frum OPECPopulation per capira Per capitt As percentagle groos docestie (percent of (perceot of total<millions) (do!llars> (dallare) of 01W Investment - total> bilateral 0D4)
Pala kernelaOil expcrters 362 370 -5.2 3.0Other sub-Saharan Africa 344 332 3.0 -1.3Total sub-Saharan Africa 706 702 -1,6 0.9
Pela oilOil exporter. 693 805 0.9 1.8Other sub-Saharan Africa 424 513 4.3 0.3Total sub-Saharan Africa 1,118 1,318 2.0 1.2
Other cropsPulsesOil exporters 839 1,003 0.1 1.9Other sub-Saharan Africa 2,919 3,568 3.6 2.1Total sub-Saharan Africa 3,758 4,571 2.9 2.1
Roote and tuberaOil exporter. 28,046 34,279 2.9 1.4Other sab-Saharan Africa 35,736 44,583 1.4 2.4Total sub-Saharan Africa 63,782 78,862 2.0 2.0
Seed cottonOil exporter. 258 223 2.7 -4.4Other sub-Saharan Africa 1,736 1,745 5.6 0.1Total sub-Saharan Africa 1,994 1,968 5.2 -0.4
SugarOil exporter. 190 186 2.8 3.6Other sub-Saharan Africa 2,082 2,888 4.8 4.1Total sub-Sabaran Africa 2,272 3,074 4.6 4.1
Note: Major crops that are totally or nearly totally exported (such as coffee, tea, cocoe, andrubber) are hoawn in Table 24, which covers exporte of agricultural cossndities.
-79-
Table 23. Agricultural importe
AnimalAverage anunal volue chan8e tu voluie Average annmal value
(Cho- ¡ of *etrte tono) (percent0 (-111los8 of tollars)ComaodTty 1965-73 1973-84 1965-73 1973-84 1965-73 1973-84
Animal and vegetable ollaOil exportera 41 193 8.4 40.5 5 131Othar sub-Saharan Africa 185 253 5.3 11.4 38 165Total sub-Saharan Africa 226 446 5.4 17.7 43 296
Total agricultural taporte cOil exportere 233 1,862Other sub-Saharan Africa . . 902 2"778Total sub-Saharan Africa 1,135 4,640
a. Cheese, butter, and a*lk.b. See Technical Notes.c. Includes producto not listed above.
- 80 -
Table 24. Major agricultural exporte: Grouth and ahares
Average annual Annual change Sub-Saharan exporte asvolume, 1973-84 in volume 2ercentage of total coimodlty exporte(thousands of (percent) Of develonpng countries Of the world
Average annual Annual change Sub-Saharan exporte asvolue, 1973-84 la voluae vercntge o total c0a3dítv exporto(thousaod4 of <.percent) Of developing countries Of the vorld
Percentego r'f total curreoc renenueYax revefne______
Tazas oa tu~m. sen:ia llosstin baSco Taz4e lft r.ntret Totalprof it. and secelrtV o. goodz ieternttooaí trode nontaz current aaoCapital Rat cootrtb.,tlona zod aervínea &,,d t~anoartínv Other tauco c8ve..e .•p ecnt of 081')
The estimates of population for mid-1984 are based primarily on datafrom the UN Population Division. In many cases the data take ínto account theresults of recent population censuses. The data on area are from the computertape for the Food and Agriculture Organization (FAO) Production Yearbook, 1983.
Gross national product (CNP) measures the total domestie and foreignoutput claimed by residente. It comprises grosa domestic product (see theno..e for Tables 2 and 3) and factor incomes tsuch as investment income, laborincome, and interest income) accruing to residento from abroad, lesa theincome earned in the domestic economy acceuing to persona abroad. It ¡s cal-culated without making deductions for depreciatíon.
The GCP per capita figures are calculated according to the nevlyrevised World Bank Atlas method. Perfect cross-country comparability of percapita CNP estimates cannot be achieved. Two obstecles stand in the way ofadequate comparabílity. One concerns CNP numbers themselves. There aredifferences in the national accounting systems of countries and in thecoverage snd reliability of underlying statistical information between variouscountries. The other relates to the conversion of CNP data expressed indifferent national currencies, to a common currency--conventionally the USdollar--to compare them across countries. The Bank's procedure for convertingCNP to US dollarn ¡a essentially based on the use of a three-year average ofthe official exchange rate. For a few countries, hovever, an alternativeconversion factor in used because the prevailing official exchange rate isjudged to diverge by an exceptionally large margin from the rate effectivelyapplied to actual foreign exchange transactiona.
Recognizing that these shortcomings affect the comparability of theCGP per capita estimates, the World Bank has introduced several improvements¡n the estimation procedures. Through its regular review of national accountsof ite member countries, the World Bank systematically evaluates the GNPestimates, focusing on the coverage and concepts employed, and whereappropriate makes adjustments to improve comparabilíty. As noted above, theBank also asseases the appropriateness of the exchange rates as conversionfactors.
The estimates of 1984 CGP and per capita CNP are calculated on thebasis of the 1982-84 base period. With this method, lhe first step ¡a tocalculate the conversion factor. Thís ¡s done by taking the simple arithmeticaverage of the actual exchange rate for 1984 and deflated exchange rates for1982 and 1983. To obtain the latter, the actual exchange rate for 1982 jamultiplied by the reletjve rate of inflation for the country and for theUnited States between 1982 snd 1984; the actual exchange rate for 1983 jimultiplied by the relative rate of inflation for the country and the UnitedStates between 1983 and 1984. The average of the actual and the deflatedexchange ratees i intended to smooth the impact of fluctuations in prices andexchange rates.
- 91 -
The second step ¡a to convert the 1984 GNP at current market pricesin national currencies to US dollars by the conversíon factor as derivedabove. The resulting GNP in 1984 US dollara is divided by the midyearpopulation to derive the 1984 per capita GNP in current US dollara.
The following formulas describe the procedure for computing theconversion factor for year t:
(e* ~~~~~tt ti
t-2,t) 3[t-2 (P t-2 P pS-t-1 t_1 p$ tt-2 ~~~~t-l
and for calculating per capita CNP in US dollars for year t:
*
(Yt$) ¡ Nt t-2,t
Where:
Yt = current CNP (local currency)
pt = GNP deflator for year t
t = annual average exchange rate (local currency/USdollar) for year t
Nt = mid-year population for year t
p$ = US GNP deflator for year tt
The annual rate of inflation ¡a the leasr-squares growth rate of theimplícit gross domestic product (GDP) deflator for each of the periodsahown. The GDP deflator is first calculated by dividing, for each year of theperiod, the value of CDP in current market prices by the value of GDP inconstant market prices, both in national currency. The least-aquares methodis then used to calculate the growth rate of the GDP deflator for theperiod. This measure of inflation has limitations, in particular for the oil-producing countries during the period of aharp increases in oil prices. It jaused as an indicator of ínflation because it ¡8 the most broadly baseddeflator, ahowing annual price movements for all gooda and services producedin an economy.
Life expectancy at birth indicates the number of years a newborninfant would live íf patterna of mortality prevailing for ahl people at thetime of ite birth were to atay the same throughout ite life. Dato are fromthe UN Populatíon Division, supplemented by World Bank estimates.
The índex of food production per capita shows the average annualquantity of food produced per capita in 1981-83 in relation to that in
- 92 -
1974-76. The estimates are derived from those of the FAO, which arecalculated by dividing indexes of the quantity of food production by indexesof total population. For this index, food is defined aS comprising cereals,starchy rotet, sugarcane, sugar beets, pulses, edible oils, nuts, fruits,vegetables, livestock, and livestock products. Quantities of food productionare measured net of animal feed, seeds for use in agriculture, and food lostin processing and distribution. Given the weaknesses in agriculturalproduction statisticá, caution should be exercísed in interpreting them.
The summary measures for GNP per capita and life expectancy ín thistable are weighted by population. The sumnary measures for the annual rate ofinflation are ueighted by the ahare of country GDP valued in current USdollars for the entire period in the particular income group.
Tables 2 and 3. Growth and structure of production
Most of the definitions used are those of the UN System of NationalAccounts.
Groas domestic product (GDP) measures the total final output of goodsand services produced by an ecor.omy--that js, by residents and nonresidents,regardless of the allocation to domestic and foreign claims. It ¡s calculaúedwithout making deductions for depreciatíon. For many countries, GDP byindustrial origín is j easured at factor cost; for other countries withoutcomplete national accounts at factor cost, it ja measured at market prices.GDP at factor cost is equal to GDP at market prices, lesa indirect taxes netof subsidies. The figures for GDP are dollar values converted from domesticcurrency by using the síngle-year official exchange rates, although for a fewcountries an alternative conversion factor is used (see note for Table 1).This procedure does not use the three-year average exchange rate used tocalculate per capita GNP in Table 1.
The agricultural sector comprises agriculture, forestry, hunting, andfishing. In developing countries with high levels of subsistence farming,much of the agricultural production is either not exchanged or not exchangedfor money. Because of difficultíes in assigning subsistence farming itaproper value, the ahare of agriculture in GDP may be underestimated. Theindustrial sector comprises mining, manuEacturing, construction, andelectricity, water, and gas. All other branches of economic activity arecategorized as services.
National accounts series in domestic currencies are used to computethe indicatora in these tables. The growth rates in Table 2 are calculatedfrom constant price series; the sectoral shares of GDP in Table 3, fromcurrent price series.
In calculating the summary measures for each indicator in Table 2,constant US dollar values are first calculated for the perioda covered, andthe values are aggregated for each of the years covered by the períod. Theleast-squares procedure is used to compute the summary growth rates. Theaverage sectoral percentage ahares in Table 3 are computed from groupaggregates of sectoral CDP in current US dollara.
- 93 -
Tables 4 and 5. Crowth of consumption and investment and structure of demand
GDP ¡B defined in the note for Table 2.
Public consumption (or general government consumption) includes allcurrent expendíture for purchases of goods and services by all levels of gov-ernment. Capital expenditure on natienal defense and security ís regarded asconsumption expenditure.
Private consumption is the market value of all goods and servicespurchased or received as income in kind by households and nonprofit institu-tions. It includes imputed rent for owner-occupied dwellings.
Gross domestic investment consists of the outlays for additions tothe fíxed assets of the economy, plus changes in the net value of inventories.
Gross domestic saving shows the amount ñf gross domestic investmentfinanced from domestic output. Comprising public and private saving, it íagross domestic investment plus the net exports of goods and nonfactorservices.
Exports of goods and nonfactor services r-epresent the value of allgoods and nonfactor serv.cee sold to the rest of the world; they include mer-chandise, freight, insurance, travel, and other nonfactor services. The valueof factor services, such as investment income, interest, and labor income, isexcluded.
The resource balance is the difference between exports and imports ofgoods and nonfactor services.
National accounts series in domestíc currency units are used to com-pute the indicators in these tables. The growth rates in Table 4 are cal-culated from constant price series; the shares of GDP in Table 5, from currentprice series. The summary measures are calculated in the same way asexplained in the note for Tables 2 and 3.
Table 6. Commercial energy
The data on energy generally are from UN sources. They refer tocommercial forma of primary energy: petroleum and natural gas liquids,natural gas, solid fuels (coal, lignite, and so on), and primary electricity(nuclear, geothermal, and hydroelectric power)--all converted into oilequivalents. Figures on liquid fuel consumption include petroleum derivativesthat have been consumed in nonenergy uses. For converting primary electricityinto oil equivalente, a thermal efficiency of 34 percent has been assumed.The use of firewood and other traditional fuel, though subs.antial in somedevelopíng countries, is not taken into account because relíable andcomprehensive data are not available.
The summary measures of energy production and consuuption arecomputed by aggregating the respective volumes for each year in the periodsand then applying the least-squares growth rate procedure. The summarymeasure of energy consumption per capita is computed from group aggregates forenergy consumption and population.
- 94 -
Energy ímporta refer to the dollar value of energy imports--Sectíon 3in the Revised Standard International Trade Classification (SITC)--and areexpressed as a percentage of earnings from merchandise exporte. The summnarymeasures are computed from group aggregates for energy imports and merchandiseexporte in current dollars.
Because data on energy imports do not permit a distinction betweenpetroleum imports for fuel and for use in the petrochemicals industry, thesepercentages may overestimate the dependence on imported energy.
Table 7. Level and growth of merchandise trade
The statistics on merchandise trade are from UN publications and theUN trade data system, supplemented by statistics from the UN Conference onTrade and Development (UNCTAD), the International Monetary Fund (INF), and ina few cases World Bank country documentation.
Merchandise exporta and imports cover, with some exceptions, allinternational changes in ownership of merchandise passing across the customsborders. Exporta are valued FOB (free on board), importa CIF (cost,insurance, and freight), unlesa otherwise apecified in the foregoingsources. These values are in current dollars and do not include trade inservices.
The growth rates of merchandise exports and imports are in real termsand calculated from volume indexes of exporta and imports. A volume index isthe ratio of the index for the total value of exporcs or imports to thecorresponding index for the unit values. For most developing economies theseindexes are from the UNCTAD Handbook of International Trade and DevelopmentStatistics and supplementary data. For industrial economies the indexes arefrom the UN Yearbook of International Trade Statistics and Monthly Bulletin ofStatistics. The summary measures are calculated by aggregating the 1980constant US dollar price series for each year and then applying the least-squares growth rate procedure for the periodo shown. These values d notinclude trade in services.
Tables 8 and 9. Structure of merchandise trade
The shares in these tables are derived from trade values in currentdollara reported in UN trade tapes and the UN Yearbook of International TradeStatistics, supplemented by other regular statistical publications of theUnited Nations and the IMP.
Merchandise exports and importa are defined in the note forTable 7. The categorization of exports and imports follows the SITC.
In Table 8, fuels, minerals, and metals are the commodities in SITCSection 3, Divisiona 27 and 28 (minerals, crude fertilizera, and metalliferousores) and Division 68 (nonferrous metals). Other primary commodities compriseSITC Sectiona 0, 1, 2, and 4 (food and live animals, beverages and tobacco,inedible crude materials, and cils and fato) less Divisions 27 and 28.-Textiles and clothiag represent SITC Divisiona 65 and 84 (textiles, yarns,fabrics, and elothing). Machinery and transport equipment are the commodities
9 95 _
in SITC Section 7. Other manufactures, calculated as the residual from thetotal value of manufactured exporta, represent SITC Sectiona 5 to 9 leasSection 7 and Divisions 65, 68, and 84.
In Table 9, food commodities are those in SITC Sections 0, 1, and 4and ín Division 22 (food and live animals, beverages and tobacco, oils andfato, and oilseeda, nuts, and kernels). Fuela are the commodities in SITCSection 3 (mineral fuels, lubricants, and related materials). 9ther primarycommodities comprise SITC Section 2 (crude materiala excluding fuel)lessasDivision 22 (oilseeds, nuts, and kernels) plus Division 68. Machinery andtransport equipment are the commodities in SITC Section 7. Othermanufactures, calculated as the residual from the total value of manufacturedimporta, represent SITC Sections 5 to 9 leas Section 7 and Division 68.
The summary measures jn Table 8 are computed from group aggregatesfor individual export commodity categories and total merchandise exports incurrent dollars; those in Table 9, from group aggregates for individual importcommodity categories and total merchandise importo jn current dollars.
Table 10. Origin and destination of merchandise exports
I4erchandise exports are defined in the note for Table 7. Tradeshares in this table are based on UN and IMF statistics on the value of tradein current dollars. Unallocated exporto are distributed among the countrygroupa ín proportion to their respective shares of allocable trade.Industrial market economies almo include Gibraltar, Iceland, and Luxembourg.The summary measures are computed from group aggregates for individual economygroups and total merchandise exports in current dolLars.
Table 11. Terms of trade
The terms of trade, or the net barter terms of trade, measure therelative level of export prices compared with import prices. Calculated asthe ratio of a country's export unit value index to its ímport unit valueindex, this indicator shows changes over time in the level of export prices asa percentage of import prices. The teri¡s of trade indexes are shown for 1970,1981, 1982, 1983, and 1984, wíth 1980 = 100. The unit value indexes are tromUNCTAD.
Table 12. Commodity exporto: Volume and price
Volume growth rates are computed from quantum data for individualcomaodities. The quantum data for agricultural commodities are from FAO tradetapes. Metals and minerals quantum data and all commodities price data arefrom the World Bank Commodity Data Bank. The príce series representinternational prices in constant US dollara, derived by deflating the currentprices by the unit value index of manufactured exporta (SITC 5-8) from fiveindustrial market economies to developing countries.
Table 13. Balance of payments, debt service, and international reserves
The current account balance is the difference between (1) exports ofgoods and services plus inflows of unrequited offícial and private transfera
- 96 -
and (2) importa of goods and servíces plus unrequited transfera to the rest ofthe world. The current account estimates are primarily from IMF data files.
Interest payments are those on the disbursed and outstanding publicand publiely guaranteed debt in foreign currencies, goods, or services; theyinclude commitment charges on undisbursed debt if information on those chargeswas available. F.>r the defínition of debt, see the note for Table 14.
Debt service la the sum of-actual interest payments and repayments ofprincipal made in foreign currencies, goods, or servíces on external publicand publicly guaranteed debt. The ratio of debt service to exports of goodsand services ís commonly used to assess the ability to service debt. Theratio of debt service to GNP in the summary measures ¡a computed from groupag8regates for debt service and GNP in current dollara; the ratio of debtservice co exporto of goods and servíces in the summary measures is computedfrom group aggregates for debt servíce and total exports of gooda and servicesin current dollars.
Croso international reserves comprise holdings of gold, specialdrawing rights (SDRs), the reserve positíon of IMF members in the Fund, andholdínga of foreign exchange under the control of monetary authoritíes. Thedata on holdings of international reserves are from IMF data files. The goldcomponent of these reserves ¡a valued throughout at year-end London prices;that ís, $37.37 an ounce in 1970 and $308.30 an ounce in 1984. The reservelevels for 1970 and 1984 refer to the end of the year and are in currentdollars at prevailing exchange rates. Because of differences in thedefinition of internat7.onal reserves, in the valuation of gold, and in reservemanagement practices, reserve levels published in national sources are notstrictly comparable. Reserve holdings at the end of 1984 are also expressedin terms of the number *. months of imports of goods and services they couldpay for, with imports at the average level for 1983 or 1984. The sunmmarymeasures are computed from group aggregates for gross international reservesarnd total imports of goods and services in current dollars.
Table 14, External publíc debt and debt service
The data on debt in this and successive tables are from the WorldBank's Debtor Reporting System for deveoping economies.
External debt is defined as debt that has an original or extendedmaturíty of more th8n one year, that ís owed to nonresidents, and that ísrepayable in foreign currency, goods, or services. A dístinction ¡a madeamong (1> publíc debt, which ja an external oblígation of a public debtor,including the national government, a political subdivísíon (or ian agency ofeither), and autonomnoua public bodies; (2) publicly guaranteed debt, whích isan external obligation of a prívate debtor that js guaranteed for repayment bya public eniíty; and (3) prívate nonguaranteed external debt, which íj anexternal obligation of a private debtor that ja not guaranteed for repaymentby a public entíty.
The tables showing public and publicly guaranteed debt do not íncludedata for (1) transactiona with the IMF, with the exceptíon of Trust Fundloans, (2> debt repayable in local currency, (3) direct investment, and (4)short-term debt (that ¡a, debt with original maturity of one year or lees).
97 _
External public debt outstanding and disbursed represente the amountat year-end of all public and publicly guaranteed loans that have been drawnby the borrower net of repayments of principal and vriteoffa.
Debt from official sources comprises (1> multilateral loana (loansand credits from the World Bank, regional development banks, and otherinternational organisationS and intergovernmental agencies> and (2) bilateralloans from governments and their agencies (including central banks) and fromautonomous public bodies. Loans from funds administered by an internationalorganization on behalf of a aingle donor government are cla2sified esbilataral loans.
Debt froa prívate sources comprises loans from (1) supplíers (creditsfrom manufacturers, exporters, or other suppliers of goods), (2) financialmarkets (loans from private banks and other private financíal institutiona,and publicly issued but prívately placed bonda), and (3) other (externalliabilities on account of nationalized properties and unclassified debts toprivate creditor>).
Debt service is defined in the note for Table 13.
Tables 15, 16, and 17. External debt and loan.
The loans referred to in all three tables are medium- and long-termloans whose maturities exceed one year. Lenders indicated in Table 15 and 16are defined in the note for Table 14.
Interest rates, maturities, and grace períods are averages weightedby the amounts of loans. Interest is the major charge levied on a loan and isusually computed on the amount of principal drawn and outstanding. Thematurity of a loan is the interval between the agreement date, when a loanagreement is signed or bonds are issued, and the date of final repayment ofprincipal. The grace period is the interval between the agreement date andthe date of the first repayment of principal.
Loans with a grant element of 25 percent and above are defined asconcessional. The grant element of a loan is the grant equivalent expresasedas a percentage of the amount committed. It is used as a measure of theoverali cost of borrowing. The grant equivalent of a loan is ita commitmentvalue, less the discounted present value of itu contractual debt service;conventionally, future debt service payments are discounted at 10 percent ayear.
Tables 18 and 19. Foreign assistance and other resources
Officíal development assistance (ODA> consists of loans and grantsmade at concescional financíal terma by official agencies of the membera ofthe Development Assistance Committee (DAC) of the Organisation for EconomicCo-operation and-Development (OECD) and members of the Organízation ofPetroleum Exporting Countries (OPEC) with the objective of promoting economicdevelopment and welfare. Net disbursements equal grosa disbursementa leaspayments to donora for amortization.
- 98 -
Total recorded net flow of resources includes ODA, grants fromprivate agencies (private aid), and transactiona at commercial terms: exportcredite, bilateral portfolio investment (including bank lending) by residentoor institutiona in DAC countries, direct investment (including reinvestedearnings>, and purchases of securities of international organizations activeín development. Net bilateral flows exclude unallocated and unspecifiedbilateral flows and all disbursements to multilateral institutions. Grants,including the value of technical cooperation and assistance, are gifts inmoney or in kind for which no repayment ís required.
The sumnary measurep of per capita net resource flows and per capitaODA in Table 18 are computed from group aggregates for population, netresource flows, and ODA. ODA os a percentage of GNP íj computed from grouptotals for ODA and GNP in current US dollars. OA es a percentage of grossdomestic investment is weighted by country GNP. Net bilateral ODA as apercentage of total ODA and ODA from OPEC as a percentage of net bilateral ODAare computed from the respective group aggregates.
The summary measures in Table 19 for total grants and technicalasaistance as percentages of net ODA are average country percentages weightedby each country's share in the aggregate ODA.
Table 20. Food aid imports
Food aid includes cereals only and is expressed in metric tons ofgrain equivalent. The dat- are for marketing years, from July 1 of thepreceding calendar year to June 30 of the current ctlendar year. The suwmarymeasures for per capita food aid are computed from group aggregates for foodíaid and population.
Tables 21 and 22. Growth of agriculture and crop production
Food includes commodities that are considered edible and containnutrients. Nonfood compt-ises all inedible and nonnutritive agriculturalcommodities. Accordingly, coffee and cea are classified as nonfood because,although edíble, they have virtually no nutritive value. (The definition offood used here is not the same as in Tables 8 and 9, where all beverages,regardless of nutritive value, are considered as food items.)
In Table 21, summary measures for annual growth rate of volume offood and agricultural production are country growth rates weighted by eachcountry's share in the aggregate value added in agriculture for the end yearof each period; those of growth rate of total production per capita aredifferences between growth rates of aggregate production and aggregatepopulation for each country group.
Tables 23 and 24. Agrícultural importa and exports
Subcategories of daíry importa are sporadically reported from oneperiod to the next. Thus, the corresponding rates of growth pertaín only tothose subcategories of dairy importa for which data are reported, and theymust be interpreted wíth caution. All data are for the calendar year.
_ 99 _
Table 25. Population growth and projections
The growth rateo of population are period averages calculated frommidyear populations. The summary measures are computed from group aggregates.
The projections of population for 1990, 2000, and the year in whichpopulation wíll eventually become stationary are made fo" each economy separ-ately. Starting with information on total population by age and sea,fertility rates, mortalíty rates, and international migration rates in thebase year 1980, these parameters are projected at five-year intervalo on thebasis of generalized assumptions until the population becomes statSonary. Thebase year estimates are from updated computer printouts of the United Nationsand data from the World Bank, the Population Council, the US Buresu of theCensus, and recent national censuses.
The net reproduction rate (NRR) indicates the number of daughtersthat a newborn girl will bear during her lifetime, assuming fixed age-specificfertility rateo and a fixed set of mortalíty rates. The KRR thus measures theextent to which a cohort of newborn gírls will reproduce themselves undergiven schedules of fertility and mortality. An NRR of 1 indicates thatfertility ji at replacement level: at this rate chíldbearing women, on theaverage, bear only enough daughters to replace themselves in the populatíon.
A stationary population is one in which age- and sex-specificmortality rates have not changed over a long period, while age-specificfertílty rates have simultaneously remained at replacement level (NRR=1). Insuch a population, the birth rate is constant and equal to the death rate, theage structure also ¡S constant, and the-growth rate ¡s zero.
Population momentum in the tendency for population growth to continuebeyond the time that replacement-level fertility has been achieved; that ¡s,even after NRR has reached unity. The momentum of a popuiation in the year tis measured as a ratio of the ultimate stationary population to the populationin the year t, given the assumption that fertility remains at replacementlevel from the year t onword.
A population tends to grow even after fertility has declined toreplacement level because pas- hígh growth rates will have produced an agedistribution with a relatively hígh proportiol of femaLes who are or will beof reproductive age. Consequently, the birth rate will remain higher than thedeath rate and the growth rate will remain positive for several decades. Apopulation takes fifty to seventy-five years, depending on the initialconditions, before its age distribution fully adjusts to the changed fertilítyrates.
To make the projections, assumptions about future mortality rates aremade in terma of female life expectancy at birth (that is, the number of yearsa newborn girl would live if subject to the mortality risks prevailing for thecross oection of population at the time of her birth). Economies are firstdívided according to whether their primary school enrollment ratio for fetalesis aboye or below 70 percent. In each group a set of annual incremento infemale life expectancy ¡s assumed, depending on the female life expectancy in1980-85. Por a given life expectancy at birth, the annual incremento during
- 100 -
the pro ection period are larger in economies having a higher primary schoolenrollment ratio and a life expectancy of up to 62.5 years. At higher lifeexpectancies, the increnients are the same.
To project the fertility rates, the year in which fertility villresch replacement level is estimated. These estimates are speculative and arebaqed on information on trends in crude birth rates (defined in the nete forTable 26), total fertilíty rates (also defined in the note for Table 26>,femsle lífe expectancy at birch, and the performance of f amilv planningprograms. For most countries in sub-Saharan Africa total fertility rates areassumed to remain constant until 1990-95 and then to decline until replacementlevel is reached; for a few they are assumed to increase until 1990-95 andthen to decline.
International migration rates are based on past and present trends inmigration flow. The estimates of future net migration are speculative. Formost economies the net migratiop rates are assumed to be zero by 2000, but fora few they are assumed to be zero by 2025.
The estimates of the hypotheticaL size of the stationary populationand the assumed year of reaching replacement level fertility arespeculative. They should not be regarded as predictions. They are includedto provide a sumnary indication of the long-run implications of recentfertility and mortality trends on the basis of highly stylizqd assumptíons. Afuller description of the methods and assumptions used to calculate theestimates is available from the Population, Health, and Nutrition Departmentof the World Bank.
Table 26. Demographic and fertility-reLated indicators
'he crude birth and death rates indicate the number of live birthsand deaths per thousand population in a year. They are from the same sourcesmentiene4 iin the note for Table 25. Percentage changes are computed fromunrounded data.
The total fertilíty rate represents the number of children that wouldbe born per woman, íf she were to live to the end of her childbearing yearsand bear children at each age in accord with prevailing age-specific fertilityrates. The rates given are from the same sources mentioned ín the note forTable 25.
The percentage of married women of childbearing age uuingcontraception refers to any form of contraception used by the women or theirhusbands. The forms generally comprise male and female sterilization,intrauterine devices (IUD), condoms, injectable and oral contraceptives,spermicides, diaphragms, rhythm, withdrawal, and abstinence. Women ofchidbearing age are generally womien aged fifteen to forty-nine, although forsome countries contraceptive usage is measured for other age groups. Data aremainly derived from the World Fertility Survey, the Contraceptive PrevalenceSurvey, the World Bank, and the UN report, Recent Levels and Trends ofContraceptive Use as Assessed in 1983.
- 101 -
The summary measures in this table are country data weighted by eachcountry's share in the aggregate population.
Table 27. Labor force
The population of working age refers to the population aged fifteento sixty-four years. The estinates are based on the population estimates ofthe World Bank for 1983 and previous years. The summary measures are countrydata weighted by each country's share in the aggregate population.
The labor force comprises economically active persons aged ten yearsand over, including the armed forces and the unemployed, but excludinghousewives, students, and otLar nonparticipanto in the labor force.Agriculture, industry, and services are defíned in the same manner as inTable 2. The sunmmary measures are country percentages weighted by eachcountry's ahare in the aggregate labor force.
The labor force growth rates are derived from the Bank's populationprojections and from ILO data on age-specific activity rates in the sourcecited above. The summary measures for 1965-73 and 1973-83 are country growthrates weighted by each country's ahare in the aggregate labor force in 1973;those for 1980-2000, by estimates of each country's share in the aggregatelabor force in 1980.
The application of ILO activity rates to the Bank's latest populationestimates may be inappropriate for some economies in which there have beenimportant changes ín unemployment and underemployment, in international andinternal migration, or in both. The labor force projections for 1980-2000should thus be treated with caution.
Table 28. Urbanization
The growth rates of urban population are calculated from the WorldBank's population estimates; the estimates of urban population shares arecalculated from UN sources. Data on urban agglomeration are also from theUnited Nationa. Because the estimates in this table are based on differentnational definitiona of what is "urban," cross-country comparisona should beínterpreted with caution.
The summary measures for urban population as a percentage of totalpopulation are calculated from country percentages weighted by each country'sshare in the aggregate population; the other summary measures ín this tableare veighted in the same fashion using urban population.
Table 29. Health-related indicators
The estimates of population per physician and nursing person arederíved from World Health Organization (WHO) data, some of which have beenrevised to reflect new information. They also take into account revisedestimates of population. Nursing persons include graduate, practical,asuistant, and auzíliary nurses; the inclusion of auxiliary nurses enables abetter estimation of the availability of nursing care. Because definitions ofnursing personnel vary--and because the data ahown are for a variety of years,
- 102 -
-generally not more than two years distant from those specífied--the data forthese two indicatora are not strictly comparable across countries.
The daily calorie supply per capita ¡a calculated by dívíding thecalorie equivalent of the food supplies in an economy by the population. Foodsupplies comprise domestic production, importo less exports, and changes inatocks; they exclude animal feed, seeds for use in agriculture, and food lostin processing and distribution. The daily calorie reguirement per capitarefera to the calories needed to sustain a person at normal levels of activityand health, taking into account age and sex distributions, average bodyweights, and environmental temperatures. Both sets of estimates are from theFAO.
The suumary measures in thís tabl-e are country figures weighted byeach country's ahare in the aggregate population.
Table 30. Education
The data in this table refer to a variety of years, generally notmore than two years distant from those specified, and are mostly from Unesco.
The data on number enrolled in primary school refer to estimates oftotal, male, and female enrollment of students of all ages in primary school;they are expressed as percentages of the total, male, or female populations ofprimary achool age to give gross primary enrollment ratios. Although primaryschool age js generally considered to be six to eleven years, the differencesjn country practices in the ages and duration of schooling are reflected inthe ratios given. For countries with uníversal primary education, the grossenrollment ratios may exceed 100 percent because some pupila are below orabove the official primary achool age.
The data on number enrolled in secondary school are calculated in thesame manner, with secondary school age generally considered to be twelve toseventeen years.
The data on number enrolled in higher education are from Unesco.
The sumimary measures ín this table are country enrollment ratesweighted by each country's share in the aggregate population.
Table 31. Central government expenditure
The data on central government fjnance ín Tables 31 and 32 are fromthe IMF Covernment Finance Statistics Yearbook, IMF data files, and World Bankcountry documentation. The accounta of each country are reported using thesystem of common definitions and classificationa found in the IMF Draft Manualon Government Finance Statistics. Owing to differences in coverage ofavailable data, the individual componenta of central government expenditureand current revenue shown ín these tables may not be atrictly comparableacross all economies. The ahares of total expenditure and revenue by categoryare calculated from national currencies.
- 103-
The inadequate statistical coverage of state, provincial, and localgovernments has dictated the use of central government dota only. ThiB mayseriously understate or distort the statistical portrayal of the allocation ofresources for various purposes, especially in large countries where lowerlevels of government have considerable autonomy and are responsible for manysocial servíces.
It must be emphasized that the data presented, especially those foreducation and health, are not comparable for a number of reasons. In manyeconomies private ixealth and education services are substantial; in otherspublic services represent the major component of total expenditure but may be-financed by lower levels of government. Great caution should therefore beused in comparísons across countries.
Central government expenditure comprises the expenditure by allgovernment offices, departments, establishments, and other bodies that areagencies or instruments of the central authority of a country. It includesboth current and capital (development) expenditure.
Defense comprises all expenditure, whether by defense or otherdepartmento, for the maintenance of military forces, including the purchase ofmilitary supplies and equipment, construction, recruiting, and traíning. Algofalling under this category is expenditure for strengthening the publicservices to meet wartime emergencies, for training civil defense personnel,and for foreign military aid and contributions to military organizations andalliances.
Education comprises public expenditure for the provision, management,inspection, and support of preprimary, primary, and secondary schools; ofuniversities and colleges; and of vocational, technical, and other traininginstitutions by central governments. Also included is expenditure on thegeneral administration and regulation of the education system; on researchinto its objectives, organization, administration, and methods; and on suchsubsidiary services as transport, school meals, and medical and dentalservíces in schools.
Health covers public expenditure on hospitala, medical and dentalcenters, and clinics with a major medical component; on national health andmedical ínsurance achemes; and on family planning and preventive care. Alsoincluded ¡a expenditure on the general administration and regulation ofrelevant government departments, hospitala and clinics, health and sanitation,and national health and medical insurance achemes.
Housing, amenities, and social security and welfare cover publicexpenditure (1) on housing (such as income-related schemes), on support ofhousing and slum clearance activities, on community development, and onsanítary services; and (2) for compensation to the sick and temporarilydisabled for loss of income, for payments to the elderly, the permanentlydisabled, and the unemployed, and for family, maternity, and childallowances. The second category alzo includes the cost of welfare servitessuch es care of the aged, the disabled, and children, as well as the cost ofgeneral administration, regulation, and research associated with socialsecurity and welfare services.
- 104 -
Economic services comprise publíc expenditure associated with theregulation,-support, and more efficient operation of business and witheconomíc development, redress of regional imbalances, and creation ofemployment opportunities. Research, trade promotion, geological surveys, andinspection and regulation of particular industry groups are among theactivities included. The five major categories of economic services are fueland erergy, agriculture, industry, transportation and communication, and othereconomic affairs and 3ervices.
Other covers expenditure for the general administration of governmentnot included elsewhere; for a few economies it also includes amounts thatcould not be allocated to other componenta.
Overall surplus/deficit is defined (on,a cash basis> as current andcapital revenue and grants received less total expenditure less lending minusrepayments.
The summary measures for the componenta of central governmentexpenditure are computed from group totals for expenditure components andcentral government expenditure in current dollars; those for total expenditureas a percentage of CNP and for overall surplus/deficit as a percentage of GNPare computed from group totals for the above total expenditures and overallsurplus/deficit in current dollars and GNP in current dollars respectively.
Table 32. Central government current revenue
Information on data, including sources and comparability, is given inthe note for Table 31.
Tax revenue is defined as all government revenue from compulsory,unrequited, nonrepayable receipts for public purposes, including interestcollected on tax arreara and penalties collected on nonpayment or late paymentof taxes. Tax revenue is shown net of refunds and other correctivetransactions. Taxes on income, profit, and capital gain are tsxes levied onthe actual or presumptive net income of individuals, on the profits ofenterprises, and on capital gains, whether realized on land sales, securities,or other assets. Social security contributions include employers' andemployees' social security contríbutions as well as those of self-employed andunemployed persons. Domestic taxes on goods and services include generalsales, turnover, or value added taxes, excises on goods, taxes on specificservices, taxes on the use of goods or property, and profits of fiscalmonopol-ies. Taxes on international trade and transactiona include importduties, export duties, profits of export or import marketing boards, transfersto government, exchange profits, and exchange taxes. Other taxes includeemployers' payroll or manpower taxes, taxes on property, and other taxes notallocable to other categories.
Current nontax revenue comprises all current government revenue thatis not a compulsory nonrepayable payment for public purposes. Proceeds ofgrants and borrowing, funds arising from the repayment of previous lending bygovernments, incurrence of liabílíties, and proceeds from the sale of capitalassets are not included.
- 105 n
The sumuary measures for the components=of current revenue arecomputed from group totals fot revenue components and total current revenue incurrent dollars; those for current revenue es a percentage of GNP are computedfrom group totala for total current revenue and GNP in current dollars.
- 106 -
Principal sources
The information in the Statistical Annex has been compiled from theWorld Bank data files and the followíng sources:
National A System of Natíonal Accounts. New York: UX Departmentaccounts of International Economic and Social Affairs, 1968.andeconomic Statistical Yearbook. New York: UN Department ofindicatora International Economic and Social Affairs, various
issues.
Draft Manual on Government Fínance Statistícs. Washíngton, D.C.:114?, 1974.
FAO and IMF data files.
National sources.
Energy World Energy Supplies. UN Statistical Papers, Series J. NewYork: UN Department of International Economic and SocialAffaírs, various years.
Trade Direction of Trade. Washington, D.C.: IMF, variousissues.
International Financial Statistics. Washington, D.C.:IMF, various issues.
Handbook of International Trade and Developmen.Statistics. New York: UNCTAD, various issues.
Monthly Bulletin of Statistics. New York: UN Departmentof International Economic and Social Affairs, various
issues.
Yearbook of International Trade Statistics. New York:UN Department of International Economic
and Social Affairs, various issues.
United Nationa trade tapes.
- 107 -
Balance Balance of Payments Manual. 4th ed. Washington, D.C.:of payments, IMF, 1977.capital flows,and debt IMF balance of payments data files.
Development Co-operation. Paris: OECD, various.annualissues.
Geographical Distribution of Financial Flows to DevelopingCountries. Paris: OECD, various annual issues
Population Demographic Yearbook. New York: UN Department ofInternational Economic and Social Affairs, various issues.
Estimates and Projections of Urban, Rural, and CityPopulations, 1950-2025. The 1982 Assessment.New York: UN Department of International Economic andSocial Affairs, 1985
Population and Vital Statistics Report (Quarterly).New York: UN Department of International Economic andSocial Affaírs, 1985
World Population Trends and Policies: 1983 Monitoríng Report.New York: UN Department of International Economicand Social Affairs, 1983.
World Populacion: 1983. Washington, D.C.: US Bureau ofthe Census, International Statistical ProgramsCenter, 1983.
World Populacion Prospects by Country, 1950-2025. New York:UN Department of International Economic and Social Affairs,1985
Labor Labour Force Estimates and Projections, 1950-2000.force 2nd ed. Geneva: ILO, 1977.
ILO tapes.
- 108 -
Social Demographic Yearbook. New York: UN Department ofindicators International Economic and Social Affairs,
various issues.
Statistical Yearbook. New York: UN Departnent ofInternational Economic and Social Affairs,various issues.
Statistícal Yearbook. Paris: Unesco, various issues.
World Health Statistics Annual. Geneva: WHO,various issues.
World Health Statistics Report. Special Issue on Waterand Sanitation, voL. 29, no. 10. Geneva: WHO, 1976.