Report No. 14547

GhanaCountry Assistance Review

June 1, 1995

Operations Evaluation Department

. 4

X. = lF~~~~~~~~

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

Abbreviations and Acronyms

AEF Africa Enterprise FundCAM Country Assistance Management (System)CAR Country Assistance ReviewCAS Country Assistance StrategyCEM Country Economic MemorandumCFF Compensatory Finance FacilityCG Consultative GroupCIR Country Implementation ReviewCOCOBOD Cocoa marketing boardCPBP Country Program and Budget PaperCPP Country Program PaperCSP Country Strategy PaperDAC Development Assistance CommitteeDEC Development Economics Vice PresidencyDFC Development finance companyERP Economic Recovery ProgramESW Economic and Sector WorkGDP Gross domestic productGLSS Ghana Living Standards SurveyGNP Gross national productIBRD International Bank for Reconstruction and Development1D Institutional developmentIDA International Development AssociationIFC International Finance CorporationILO International Labour OrganizationIMF International Monetary FundMFEP Ministry of Finance and Economic PlanningMIS Management information systemMTADS Medium-term Agricultural Development StrategyNATCAP National Technical Cooperation and Assessment Programme (UNDP)NGO Non-governmental organizationOD Operational DirectiveODA Overseas Development Administration (United Kingdom)OED Operations Evaluation DepartmentOMS Operational Manual StatementsOPR Operational Policy Review DepartmentPAMSCAD Program of Actions to Mitigate the Social Costs of AdjustmentPAR Performance Audit ReportPBD Planning and Budgeting DepartmentPCR Project Completion ReportPER Public expenditure reviewPFP Policy Framework PaperSAC Structural Adjustment CreditSAIS Structural Adjustment Institutional Support ProjectSAL Structural Adjustment LoanSAR Staff appraisal reportSDR Special Drawing RightsSECAL Sector Adjustment LoanSOE State-owned enterpriseSPA Special Program of Assistance for AfricaTA Technical Assistance

THE WORLD BANKWashington, D.C. 20433

U.S.A.

Office of Director-GeneralOperations Evaluation

June 1, 1995

MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT

SUBJECT: Ghana. Country Assislance Review

Attached is the report Ghana: Country Assistance Review prepared by the Operations EvaluationDepartment. This country assistance review (CAR) is the first of a series of "new style' OED studiesthat will, for each country concerned, evaluate (i) the relevance and efficacy of the Bank's overallcountry assistance strategy; and (ii) the effectiveness of various lending and nonlending instruments ofBank assistance. The objectives are to establish accountability, derive lessons of experience, andprovide recommendations for future actions.

Ghana was selected to be the focus of this first CAR in part because it is frequently cited as a"success case" of sustained adjustment and resumed growth. Its GDP growth rate averaged nearly fivepercent over the past decade - higher than in any African country other than Botswana and Mauritius -and poverty has decreased. At the same time, a number of outstanding questions remain regarding theextent, sustainability, and pace of Ghana's economic recovery. For instance, even a continuation ofthe five percent growth of the last decade, at the current rate of population growth, would require some35 years to double the per capita income to a still low level of US$900.

The study finds that World Bank loans of more than US$2 billion since 1984, including morethan US$1 billion for adjustment operations, have helped Ghana to make considerable progress over thepast decade. Bank contributions in terms of nonlending services (policy dialogue, economic and sectoralstudies, and aid mobilization and coordination) were particularly significant. The study cautions,however, that the progress may not be sustained unless Ghana speeds up the implementation of a largeunfinished agenda of adjustment reforms. In particular, the current strategy has not succeeded instimulating the needed private sector response or in raising agricultural productivity, two elements onwhich the sustainability of future development depends.

The study recommends that the Bank should adopt a strategy that gives central importance toan interrelated set of "sustainability issues" including: private sector development; the 'nexus' ofdeclining agricultural productivity, higher population growth and environmental degradation; economicgovernance; and institutional development.

The study also recommends that the Bank's process of country assistance strategy formulationbe strengthened through: self-evaluation of past strategies; more rigorous risk and options analysis; andthe monitoring of performance against a set of suggested "sustainability indicators."

Attachment

Contents

Ghana at a Glance. 5Preface. 7Executive Summary ............. 9

1. Overview of Bank/Ghana Relations .17

Overview of Bank/Ghana Relations, 1983-1993 .17Key Events and People .18The Congruence of Government and Bank Strategies .20Overview of Bank Lending .24

2. Relevance of the Bank's Assistance Strategy .25

Main Conclusions .25Selective Coverage .25Bank Strategy Papers .26Evolution of the Bank's Assistance Strategy .27

Overview .27TIhe Strategy in the 1981 CPP .27The Strategy in 1982-83 .28The 1986 CPP .31The Strategy in the 1990 CSP ... 32T'he Strategy as Described in the 1993 CSP .33

Relevance of the Bank's Strategy .34Right Diagnosis? Short-term, Yes. Long-term, More Work Needed .34Clarity of Definition and Monitorability of Objectives:

Good in the 1980s; Fair in the 1990s .35Realism of Basic Assumptions: Partial .36Timeliness? Yes .36Appropriateness of Priorities? Short-term, Yes.

Longer-term, Uncertain .37

The task manager and principal author of this study is Robert Armstrong. Background papers were preparedby Ronald Ridker (education), Jacob Meerman (agriculture), William Branson (macroeconomic isues), and NielsDabelstein (poverty). Sulaiman Wasty contributed analysis of portfolio performance and ownership, and Sanliang Yueprepared many of the charts and tables. Many others in and outside the Bank provided helpful comments. The workwas carried out under the gencral direction of Mark Baird and Manuel Pcnalvcr. Norma Namisato provided wordprocessing assistance.

2

Mix of Instruments: Appropriate in the 1980s,Evolving in the 1990s . ..................................... 40

Adequacy of Risk Assessment? Uncertain ....... ................... 40Feasibility of Implementation and Ownership?

1980s, High; 1990s, Uncertain .......... ...................... 41Defining the Bank's Role vis-a-vis Other Donors:

Not Much Done .......................................... 43Sustainability? The Uncertain Path from Recovery to

Sustainable Growth . ....................................... 43Conclusions ............................................. 44

3. Effectiveness of the Bank's Instruments ......... ..................... 45

Introduction ............................................ 45Most and Least Effective Instruments .......... ..................... 46Overview of Bank and Total Donor Assistance ....... .................. 46The Effectiveness of Lending - Performance Trends ...... ............... 48Projects with Unsatisfactory Outcomes ......... ...................... 51Status of the Portfolio ........................................... 54The Role of IFC ............................................ 58Allocation of Staff Resources . ..................................... 59Economic and Sector Work (ESW) ........... ...................... 59Aid Mobilization and Coordination ............ ...................... 64

4. Agriculture and Education: Objectives and Performance ..... ........... 69

Agricultural Objectives and Performance ......... .................... 70Bank Strategy for Agriculture in the 1980s ....... ................... 71Policy and Output ............................................ 72Bank Performance and Recommendations .......................... 73Natural Resource Conservation: An Impending Crisis? . 75

Educational Objectives and Pcrformance .. ........................... 75Bank Support to the Sector . .................................... 76Outcomes ........................................... 76Conclusions .......................................... 79

5. Progress Towards the Strategic Objectives ......... .................. 81

Overview of Outcomes ......................................... 82External Shocks/Exogenous Factors . ................................ 83Evaluation of Performance in Economic Management ...... ............. 88Progress Against Growth and Poverty Alleviation

Objectives ........................................... 97Sustainability Indicators ......................................... 102

Annex to Chapter 5 ......................................... 107

3

6. Conclusions and Recommendations ................................. 115

Overview .................................................. 115Key Strategic Issues ............................................. 116Recommendations . ............................................ 117Design of Bank Strategy .............. ........................... 118For Achieving Objectives . ........................................ 122For Making Bank Instruments and Processes More Effective .... ......... 126

Bibliography 129

Boxes

1.1 Selected Notable Events, 1981-94 ................................... 191.2 Conditions for Successful Adjustment in Ghana ........ ................ 211.3 Ghana Compared with Other Sub-Saharan African and Low-income

Countries ................................................... 232.1 Bank-Fund Coordination .............. ........................... 292.2 IMF Views on the Sequencing of Ghanaian Reforms .................... 392.3 The Downside of High Aid Dependency: Dangers and Distortions .... ..... 423.1 Outstanding Projects ............................................. 493.2 Lessons from the Institutional Support (SAIS) Project .................. 533.3 Attainment of Institutional Development Objectives ..................... 555.1 Projected vs. Actual Cocoa Prices ................................... 855.2 Aid and The Dutch Disease: Macroeconomic Management When Everybody

Loves You ............................................... 885.3 Is Credibility the Binding Constraint to Private Investment? ..... .......... 935.4 Rent-Seeking and Patronage Relations ......... ...................... 965.5 Program of Actions to Mitigate the Social Cost of Adjustment .... ......... 995.6 The Social Dimension of Adjustment Project ......................... 1005.7 Main Findings About Poverty in Ghana ............................. 104

Tables

1.1 Commitment and Disbursements in Ghana (FY79-94) ................... 242.1 Ghana: Phasing and Sequencing of Adjustment Process ..... ............. 303.1 Aid Flows, 1981-92 .............................................. 473.2 IDA Lending by Major Lending Instrument, FY79-94 ................... 483.3 Unsatisfactory Projects . .......................................... 523.4 Ghana - Selected Indicators of Bank Portfolio Performance

and Management ......................................... 573.5 Staffyears by Main Instrument . .................................... 594.1 Selected Agricultural Indicators ..................................... 734.2 Education Sector Indicators ....................................... 775.1 External Shocks and Performance Response Measures ...... ............. 845.2 Selected Incentive Indicators ...................................... 905.3 Selected Fiscal and Monetary Performance Indicators ...... .............. 91

4

5.4 Selected Indicators of Public Sector Reform . .......................... 955.5 Selected Growth and Structural Indicators .985.6 Selected Social Indicators, 1970-92 .1015.7 Recent Trends in Poverty, 1988-92 .1035.8 Selected Sustainability Indicators .106

Figures

5.1 Changes in Terms of Trade .865.2 Changes in Export Prices .86

5

GHANA AT A GLANCE

Ghana is a low-income country, with a Ghana's economy grew in the initial years afterpopulation of about 16 million, located on the west independence but declined from the mid-1960s intocoast of Africa. Its land area is about the same as the early 1980s. By 1983, when the Econonicthat of the United Kingdom, Uganda and the Lao Recovery Program (ERP) was adopted, Ghana'sPDR. Per capita income in 1992 was US$450 economy had collapsed. The ERP was therefore(World Bank Atlas method). Average life expectancy launched at a time when a broad consensus foris about 55 years and the literacy rate is about 60 change prevailed. Initial reform efforts focussed onpercent. stabilization and liberalization and were in general

successful until a weakening of fiscal discipline, inPolitical Developments the wake of the 1992 elections, resulted in a

resurgence in inflation and in a rapid depreciationGhana achieved its independence in 1957 and of the now market-determined exchange rate.

was ruled by a succession of civilian and militaryregimes. In 1992, a new constitution was approved Average annual GDP growth over the decadeby referendum and multi-party presidential and following the introduction of the ERP was nearlyparliamentary elections were held. The current 5 percent, with services being the fastest-growingPresident of the elected civilian regime is Flight sector. Agricultural sector growth over this decadeLieutenant (retired) Jerry Rawlings. The next was 2.3 percent overall, with 3.1 percent growth inelections, both presidential and parliamentary, will cocoa production but only 2.2 percent growth inbe in 1996. food production. This recovery was supported by

substantial aid from the IMF, World Bank, andThe Economic Structure other donors.

Ghana is well endowed with a broad range of Adjustment measures were implemented withnatural resources such as arable land, forests, and varying degrees of intensity and success. Goodsizeable deposits of gold, diamonds, bauxite, and progress was made, for example in exchange rate,manganese, as well as a considerable capacity of pricing and trade policy reform. There remains,hydroelectric power. The economy has traditionally however, a substantial agenda of reform to bedepended to a high degree on primary (agricultural undertaken in public sector management and civilas well as mineral) production and exports. Exports service reform, public enterprise divestiture, and inof gold, cocoa, and timber still account for the creating an enabling environment for private sectorbulk of total merchandise exports, with respective development. The weakest aspect of the economicshares of 29 percent, 31 percent, and 13 percent. recovery has been the low private savings and

investment rates, which need to increaseThe agricultural sector, inclusive of forestry substantially to sustain growth in the medium term.

and fishing, remains the dominant sector, Issues of high population growth, decliningemploying some two-thirds of the labor force and agricultural productivity and environmentalaccounting for nearly half of total GDP. problems also need to be addressed if growth is toAgricultural production, which is primarily small be sustained over a longer run.scale, is concentrated in cocoa and staple foodcrops. Ghana is currently ranked as the world'ssecond largest (after C6te d'lvoire) producer and A Few Key Indicator (1993)exporter of cocoa.

%ofServices comprise the second largest sector in GDP %

the economy, accounting for an increasing share in Private investment 4 PopulationGDP (37 percent in 1993), while the industrial Gvt. investment 10 (annual growth) 3.1sector accounts for the remaining 16 percent. consumption 012 Primary 7r

Broad money 17 Secondary 39Economic Developments Foreign financing 9 Inflation 25

7

Preface

"The Bank is not a reflective place... We haven't had time to look back.Bank Director, 1993

This Country Assistance Review (CAR) on Ghana is the first of a series of 'new style'OED studies. These studies will, for each country concerned, "look back," evaluate, andreflect upon:

(a) the relevance and efficacy of the Bank's overall country assistance strategy(taking into account country conditions, the impact of external/exogenousfactors, and the role of other donors); and

(b) the effectiveness of the various instruments of Bank assistance, namely:investment, adjustment and technical assistance (TA) lending operations,economic and sector work (ESW), policy dialogue, portfolio management, andaid mobilization and coordination.

In addition, the report is also "forward-looking" in its recommendations for future actions byboth the Bank and the Government of Ghana.

This CAR is, in effect, a performance audit report on the Bank's total assistanceprogram to Ghana. As such, it has approximately the same relation to the Bank's overallassistance program as OED project audits have to individual Bank projects. But here thecountry program as a whole, rather than the individual project, is the "unit of account." Thetwin objectives of the study are to establish accountability and to derive lessons of experience.

In contrast to OED's previous country studies which comprised a comprehensivehistorical reviews of Bank/country relations over many years, the new CARs will be highlyselective in their coverage of time span, sectors, instruments and issues. The time focus ofthis CAR is upon the period 1983-94, since 1983 is when Ghana adopted an adjustmentprogram strongly supported by the Bretton Woods institutions.

Sectoral coverage is limited to agriculture and education. Agriculture was selectedbecause of its large and critical role in the economy, and because the sustainability of Ghana'sdevelopment will depend heavily on the extent to which agricultural productivity can beimproved. Education was selected because of the importance of human resourcedevelopment to Ghana's longer-term prospects, and also because the Bank has played animportant role in supporting Ghana's comprehensive education reform program, launched in1986.

8

Among the Instruments of Bank assistance, emphasis is given to adjustment lending,the policy dialogue, technical assistance and aid coordination.

Why Ghana?

Ghana was selected to be the focus of OED's first 'new style" CAR because it isfrequently cited as a success case of sustained adjustment and resumed growth. The mainindicator of this success is the nearly five percent annual GDP growth rate achieved in thelast decade-higher than in any African country other than Botswana and Mauritius. It isimportant to ascertain the extent to which the Bank's assistance contributed to Ghana'srelative success. In Africa especially, examples of success are at a premium. There are stillquestions, however, about the extent and sustainability of the success of the Ghanaian reformprogram. Hence, an objective of this study is to assess this success and to formulateindependent judgments regarding the main contributions (and shortcomings) of the Bank'sassistance program.

Sources and Acknowledgements

This study is founded upon a variety of sources. On the project side, PCRs andproject audits were used to identify common denominators of project success and failure overtime. Supervision reports and Country Strategy Implementation Review reports providedinformation on the performance of ongoing projects, and on systemic project implementationissues. Extensive reviews were made of Bank archives and of the substantial literature onGhanaian development. Interviews were held with Bank and IFC staff, the IMF and otherdonors, and Ghanaians in both the public and private sectors. Background reports wereprepared by consultants on agriculture, education, and macroeconomic management issues.Bank staff from DEC and PBD provided analyses of external shocks and of the costs ofmanpower and other resources employed by the Bank in carrying out its assistance strategy.The kind cooperation and valuable assistance of those who contributed their time and effortsto the preparation of this report are gratefully acknowledged.

A draft of this report was sent to the Ghanaian Government. The comments receivedwere taken into account in the final report.

9

Executive Summary

1. Bank assistance was generally effective in helping Ghana make considerableeconomic progress over the past decade. But this progress cannot be sustained unless (i)the implementation of a large unfinished agenda of adjustment is accelerated in the shortrun; and (ii) the Government and donors alike mount a concerted effort to address thekey long-term "sustainability issues." The watchword is sustainability. And the criteria forassessing the success of future Bank assistance should be focused largely on theperformance of a specific set of sustainability indicators.

2. It is now more than ten years since Ghana launched its far-reaching but"gradualist" Economic Reform Program (ERP). The Bank Group has lent Ghana morethan US$2 billion since 1984-more than US$1 billion for adjustment operations. Muchhas been achieved owing to the joint efforts of the Government, and the Ghanaianpeople, supported by the World Bank Group, and other donors. The IMF and WorldBank played major roles in helping the Government shape and underwrite the ERP as itevolved over the years. And it was clearly owned by, not forced on, the Government.

3. In the 1980s, against a background of highly strained relations and mutual mistrust,the Bank succeeded, by and large, in establishing excellent Bank-country relations and atrue sense of partnership with a core group of Ghanaian officials responsible for preparingand implementing the ERP. A key factor in sustaining the reforms since 1983, despite asetback in 1992, was the continuity and competence of this core group.

4. The main accomplishments over the past decade are well known. Exchange rateand trade reforms were the centerpieces of the ERP, and Ghana now has an interbankforeign exchange market, few quantitative restrictions, and low levels of protection.Extensive price and distribution controls were dismantled. The fiscal situation was greatlyimproved and rationalized, especially on the revenue side. Inflation was brought downfrom more than 75 percent in the early 1980s to 10 percent in 1992. More was investedin rehabilitating Ghana's moribund infrastructure. Poverty was reduced and socialindicators improved. The regulatory framework has been improved. Cocoa and otherexports have recovered. And GDP growth has averaged about five percent over the pastdecade.

5. Moreover, unlike the experience in many adjusting countries, the stabilizationobjectives were achieved simultaneously with increased real public expenditures, especiallyon social services and on public investment. This "win-win outcome" facilitated the pursuitof "adjustment with a human face." Poverty in Ghana is less than in many other Africancountries and has been reduced in recent years. Progress has continued in getting theprices right, in rehabilitating the infrastructure, in rationalizing the revenue structure, andin public investment programming-areas where both the Government's and the Bank'sstrengths can be built on.

10

6. But fiscal problems have resurfaced in parallel with the move to greater democracysince 1992. And a long agenda of unfinished adjustment remains. Performance has beendisappointing in private-sector development, public enterprise restructuring, civil servicereform, expenditure control, agricultural development, educational achievement,environmental control, and institutional development. Progress in privatization has alsobeen slow, although some important new privatization measures were taken in 1994 andearly 1995.

7. The prospects for sustaining a satisfactory rate of per capita growth and reducingpoverty further are uncertain, unless Ghana continues to rely on exceptionally highvolumes of concessional aid, something that may not be feasible nor indeed desirable forreasons discussed later. Agricultural growth remains well below what is necessary andfeasible-and may well be no better than the rate of population growth. Total and privatesavings and private investment remain extremely low. Educational testing results in theprimary schools are very low. There are serious environmental problems. And even acontinuation of the five percent growth of the last decade, at the current populationgrowth rates, would require some 35 years to double the current per capita income and toreach a still low level of US$900 by the first third of the twenty-first century.

Rethinking the Strategy

8. For the Bank's assistance to Ghana in the years ahead to be as relevant,efficacious, and cost-effective as possible, it is recommended that the Bank andGovernment should together rethink some of their priorities and focus increasingly on theissues mentioned in the two previous paragraphs. This implies giving more attention toinstitutional development, economic governance, aid dependency, and sustainability issues.Two of the key sustainability issues are how to increase substantially the very low privateinvestment rate (recently as low as four percent of GDP) and the even-lower nationalsavings rate (less than one percent of GDP).

9. It is further recommended that the Bank should redefine some of its comparativeadvantages and refine its performance indicators. And it should sharpen its processes forformulating and monitoring the implementation of its country assistance strategy asarticulated in the Country Assistance Strategy Papers (CAS).

10. In the 1980s, the Bank's strategy and instruments were effective in helping theGhanaian Government to stabilize, liberalize, and rehabilitate an economy in crisis.Particularly in the mid-1980s, the Bank's program of assistance to Ghana gets high marksfor the relevance of objectives, efficacy of implementation, cost effectiveness, and staffperformance. The Government also gets a high rating for borrower performance.

11. The hardest parts of Ghana's adjustment agenda are yet to come, and theobjectives and instruments have to be adjusted accordingly. The remaining hardadjustment measures need to be taken and deepened. Urgent attention needs to be givento several longer-term issues if the growth and gains achieved are to be sustained-oraccelerated. The binding constraints are on the institutional and managerial side.

11

12. Key issues. Questions that need to be addressed explicitly in the Bank's countrystrategy are: What can the Bank do to improve the sustainability prospects? And whatpace of change is needed, on which key fronts, to minimize the risks that Ghana's progresswill not be sustained? Most of the truly important issues-as distinct from the urgentissues that tend to command attention-can be called the "sustainability issues." Two ofthe main ones are the "leading sector" and aid dependency issues.

13. In the 1980s, recovery relied necessarily on the public sector as the leading sector,and was foreign-aid led. During the 1988-92 period, annual net disbursements of officialdevelopment assistance (ODA) averaged 10 percent of GDP (this figure does not includesome of the IMF arrangements). Aid-supported expenditures on infrastructure and socialsector development can keep growth going for a time. But sustained growth and povertyalleviation requires a far more vigorous private-sector response, based on higher domesticsavings and private investment, than has been achieved so far. Ghana's heavy dependenceon foreign aid also has other disadvantages, and the future strategy should provide forreducing this dependency.

14. Following the 1992 election of a civilian government under a new constitution, thepolitical economy is now quite different from that in the 1980s. The challenge for theBank is to help Ghana manage a transition to a more sustainable development path whileit also transforms its political forms and processes. The sustainability of the political andsocial transitions depends partly on how the economy fares, and vice versa. The Bank'sstrategy and instruments therefore need to be forged to fit the new political economy.

15. A shared vision. The future strategy should be based on a shared vision of whatkind of society the Ghanaian Government and people wish to bring about. As a seniorGhanaian official said, "When we know what we want and take the initiative, things gookay. But when the World Bank takes the initiative, things don't work so well." It shouldbe a central role of the Bank to help Ghanaians elucidate the options, and then to fosterthe needed policy changes, investments, and institutions. If in the 1980s a main thrust ofthe strategy was to "get the prices right," for the 1990s it should be in helping Ghana to"get the institutions and economic governance right."

16. The policy measures that most need to be taken by the Government, and besupported by the Bank and other donors, include developing and implementing a "nexusstrategy" (defined below) that gives priority to agricultural development, family planningand environmental protection, civil service reform, capacity building (but not through thetraditional technical assistance projects that have so far proven ineffective), and thedivestment and restructuring of large and inefficient public enterprise sector.

17. Two critical issues are: How to stimulate the needed private-sector response? Andhow to reverse the declining productivity in agriculture without harming the environment?

18. More gradualism-or breakthroughs? The government strategy has, since 1983,been based on a gradualist approach in which it took the better part of a decade for mostof the agenda of stabilization and liberalization measures to be implemented. Thisgradualism was facilitated by the substantial aid provided by the donors, especially in the

12

form of quick-disbursing aid, much of which financed current expenditure. The strategynow needs to accelerate the pacing and revisit the sequencing of future policy andprogram actions in the light of the failure of the current strategy to foster the neededprivate sector response or to raise agricultural productivity, upon which the sustainabilityof future development are so dependent.

19. Accelerating the adjustment will undoubtedly be difficult in the face ofdevelopments in democratization that may slow decision-making. But it must berecognized that the binding constraints have changed and that the readiness of donors tosupport reforms that are not accelerated is uncertain in an era of scarce donor resources.What is needed today is a strategy that makes a breakthrough in the investment climateand business environment for the private sector. Accelerated programs of privatization,unequivocal support for private sector development, and in-depth administrative reformsappear to be the foundation for persuading the private sector that Ghana's increasinglymarket-friendly policies are unlikely to be reversed.

20. For the Bank The Bank had a well-established comparative advantage in helpingGhana to correct its distorted macroeconomic and sectoral policies and to rehabilitate itsinfrastructure. The Bank now needs to further develop and improve its capabilities-tobecome more effective in key areas where its past assistance has been least effective:

- In fostering private-sector development, an area in which IFC needs to playan increasingly active role, including through the African Enterprise Fund.

i In promoting institutional development and capacity building.

3 In helping the government to become smaller and more efficient.

- In elucidating policy analyses and options to promote enlightened publicdebates on key policy issues among wider constituencies.

Recommendations for the Design of Bank Strategy

21. Focus on sustainability. The Bank's strategy needs a longer time horizon and acentral focus on identifying and addressing the binding constraints to sustainable, long-term development. This implies a need to formulate the strategy with specific referenceto evolving political and social as well as economic institutions, that is, in the context ofthe political economy. The strategy should ultimately be based upon a shared vision ofwhat kind of society the Ghanaian Government and people wish to bring about.

22. Monitor perfornance against sustainability indicators. There is no agreed orestablished set of core sustainability indicators for Ghana. It would be useful if such a setwere identified and used in the Bank's Country Assistance Strategy (CAS) for purposes ofoptions analysis, policy dialogue and performance monitoring. The report suggests asubset of such indicators (Table 5.8). It suggests further that an additional subset ofgovernance-cum-institutional development indicators be developed and applied. Macro-level monitoring needs to be supplemented by more micro-level monitoring of attitudes

13

and behaviors of farmers, informal sector participants, women in the labor force, the poor,the civil service, and students and teachers in the classroom.

23. Key indicators of the sustainability of growth and poverty alleviation include:fertility rate, productivity growth and structural change in agriculture, total and privatedomestic savings rates, the "genuine" savings rate (national savings less depreciation offixed capital and depletion of natural resources), private investment ratio and annual rateof private investment growth, export growth rate, especially of nontraditional exports,primary school enrollment rate (especially of girls) and criterion reference test scores, andthe government revenue to GDP ratio. Either the levels or the trends in these indicatorsare worrisome. Relevant governance indicators (which are admittedly hard to define inobjectively monitorable form) would include: the predictability, openness and competenceof policy-making; the professional ethos of the bureaucracy; accountability of the executivebranch of government; participation of civil society in public affairs; rule of law; andincidence of civil disturbances.

24. Develop and implement a "nexus strategy." The CAS should give higher priority toagricultural development, family planning and environmental protection. The "nexus" hererefers to the combination of rapid population growth, declining agricultural productivityand environmental deterioration. This combination of problems warrants priorityattention in both the lending program and in non-lending services (economic and sectorwork, policy dialogue and aid coordination). A better information base is needed (forexample, on levels and trends in agricultural production and productivity) and newmodeling approaches should be encouraged to elucidate the critical linkages.

25. The poor in Ghana are disproportionately rural and located in the northernregion, where the nexus problem is most pronounced. Dealing with the linkages istherefore critical to alleviating poverty. It is also critical to avoiding tribal strife, such asoccurred in the north region in early 1994-which appears at least partly attributable tothe growing scarcity of fertile land and tensions over land rights.

26. Focus on institutional development. A greater focus on sustainability implies aneed to focus in turn on institutional constraints. Both the Bank and the Governmentneed to formulate and implement a coherent and comprehensive strategy for publicadministration reform. This strategy should address the root causes of underutilization ofGhana's existing human capacities and capabilities, the overstaffing and low productivity ofthe civil service and public enterprises, and the poor performance of most technicalassistance projects (not only the Bank's).

27. Focus on govemance. A greater focus on sustainability also implies a need to focuson governance factors. The Government is still in some areas part of the problem ratherthan part of the solution. For example, Ghana's overstaffed and poorly functioning publicadministration and public enterprises are impediments to private sector developmentthrough financial crowding out (excessive credit to the parastatals), through the "unlevelplaying field" between public and private firms, de facto monopolies, rent-seeking,obstruction and harassment by public officials, inefficient delivery of infrastructure andother public goods and services and through the "signals" received (if not intentionally

14

sent) that statist or populist policies and intimidating practices are not yet permanentlyrelegated to the past.

28. The Bank's strategy, therefore, and its policy dialogue with the Government shouldemphasize the need to address those governance factors that tend to diminish private-sector confidence and hence private-sector savings and investments. And the longer-termstrategy also needs to be geared to fostering social and political sustainability, whichimplies ensuring that the benefits of growth, and of external assistance, are broadly sharedthrough the society, especially by the poorest segments.

29. Make the Bank's strategy more strategic. Future Bank strategy papers (includingsectoral and other strategy documents, as well as country assistance strategies) should bemade "more strategic" by ensuring that they address head-on questions such as: Whichlong-term and short-term constraints are most binding? What are the critical"sustainability issues," and what can donors do about them? Which risks can be managedand which cannot? How are lessons of past experience being heeded? What "criticalmasses" of Bank/donor lending and non-lending services can improve the prospects ofgetting significant "breakthroughs" on the binding constraints? And what are the Bank'scomparative advantages in assisting Ghana? The previous sections of this summary haveaddressed some of these questions and they are further developed in the main report.Future strategy papers should focus, more so than in the past, on assessing risks and onelucidating options (especially regarding risks to be taken or avoided) for the Bank'ssenior managers and the Board.

30. Strengthen CAS evaluation and accountability. A corollary to making the Bank'sstrategy "more strategic" is that the CAS and its discussion should serve, among otherthings, to clarify what the Region should (and should not) be held accountable for inimplementing its country assistance strategy. It may be useful and possible to identifysome strategy implementation performance indicators that serve to strengthenaccountability without compromising the flexibility needed by the Region at the tacticallevel. In any event, it is recommended that future CASs should contain a more substantialsection comprising a self-evaluation of Bank successes and failures in achieving targets andobjectives specified in previous CASs. It should also explain how the lessons derived fromthat evaluation are being applied in the current strategy.

Recommendations for Achieving Objectives

31. Reach out, and listen to, broader constituencies in and outside Ghana. There is anunmet demand among Ghanaians for information and analysis on the country's economicand social situation and prospects. The Bank can help meet this demand by disseminatingits documents more widely, by encouraging the Government to discuss Bank and jointly-authored reports with broader audiences, and by undertaking more "outreach" activities inregard to its provision of non-lending services. The outreach should be a two-way streetand involve more listening by Bank staff to views and concerns of Ghanaians and otherssuch as academic analysts of Ghanaian social and political issues.

15

32. Develop a more strategic approach to in,stitutional development and technicalassistance. Ghana lacks a coherent strategy for civil service and parastatal sector reform,and it is a matter of highest priority for the Bank to assist in formulating it. Anotherproblem has been the poor efficacy of technical assistance to Ghana. Such assistance hastoo often been characterized by the dominance of short-term products and events overlonger-term processes geared to institutional development, capacity-building, and learning-by-doing. The Bank should encourage the Government to convene a local consultativegroup to focus on these technical assistance issues and come up with an agreed actionplan.

33. Get at the real causes of inadequate expenditure management. Despite many publicexpenditure reviews and the provision of technical assistance in financial management,Ghana's poor expenditure management continues to be a major problem. Priority shouldbe given to solving this problem, but not by further relaxing conditionalities regarding theGovernment's required share of local expenditures.

34. Don't oversell Ghana as a success story. Given the large unfinished agenda ofreform and the challenges of making the gains sustainable, overselling Ghana as "successcase" may be a disservice to the country by detracting attention from the needed reforms.Nevertheless, the relative success of the government policies under the ERP so far may beheld as an example to other countries not yet committed to adjustment.

35. Beware the down-side of excessive aid dependency. The Bank should give greaterconsideration to minimizing the potential adverse effects of "too high" levels of externalaid, including upon the behaviors of government officials. Excessive aid may allowgovernments to postpone adoption of needed but politically difficult reforms, thusreducing savings and private investment and delaying the expected supply response. But itmay also lead to excessive numbers of projects relative to the country's absorptivecapacity, dependency behaviors (such as "let the donors do it"), distortions in incentivescreated by multiple scales of donor-financed supplements and allowances, and thedominance of donor-driven objectives and donor-specified performance indicators. Thisattention needs to be made explicit in the design of non-lending as well as lendingservices.

36. Develop improved aid coordination policies and practices. All of the above-mentioned recommendations have implications for donor coordination, as the Bank has arole to play in getting all donors to focus on sustainability, to work in the context of acommon strategy for institutional development, to minimize the downsides of aiddependency, and so on. Aid coordination is currently effective in some areas but not inothers, and the Government needs to be urged to take a much greater role itself. Inparticular, sectoral and sub-sectoral coordination needs to be strengthened in a context ofagreed sectoral strategies and action plans.

17

1. Overview of Bank/Ghana Relations

"We need Bank people who can speak our cultural and political economylanguage ... who can talk to us in our own terms."

Senior Ghanaian official

1.1 At the time of its independence, in 1957, Ghana was considered a middle-incomecountry with a per-capita income high by African standards, an educational system judgedto be the best in Africa, fairly well-maintained infrastructure, and adequate governmentinstitutions. Ghana's growth of the 1950s continued into the early 1960s, aided by highworld cocoa prices. But falling cocoa prices and dirigiste economic policies led tostagnation in the late 1960s. The economy followed a persistent downward spiralthroughout the 1970s and into the early 1980s owing to weak producer incentives, pooreconomic management, and declining external aid levels. The outcome was high inflation,diminished import capacity, and a steady decline in per capita incomes.

1.2 The policy environment of this period was one of large budget deficits, pervasivecontrols, and a pricing framework that, together with an overvalued exchange rate,discouraged private sector savings and investment, provided inadequate incentives forproduction and exports, and encouraged corruption, rent-seeking and smuggling. Otherimportant facets of Ghana's decline were the deterioration in institutions, especially theschools and civil service, and the exodus from the country of many weli-educatedGhanaians. On the political front, the 1970s and early 1980s were a period of politicalinstability, with three successful coups (1972, 1979, 1981), one palace coup (1978), andother attempts to overthrow the Government. This political volatility was both a causeand consequence of the poor economic performance. The combination of politicaluncertainty, poor economic policies, and general economic decline rendered it impossiblefor the Bank to maintain either a satisfactory policy dialogue with successive governmentsor a satisfactory portfolio of projects.

1.3 In the early 1980s, three shocks precipitated a virtual collapse of the economy:a prolonged drought, a marked deterioration in the terms of trade, and the unexpectedreturn to Ghana of over a million Ghanaian workers expelled from Nigeria.

Overview of Bank/Ghana Relations, 1983-1993

1.4 Bank/Ghana relations were strained and difficult between the mid-1960s and early1980s (largely for the reasons described above) but improved markedly in the decade1982-83 to late 1992. These relations became very good from about 1985 onwards, asmutual mistrust was replaced by a respectful partnership in which the Bank became thelead donor. Strains emerged between November 1992 and mid-1993 as Ghana's

18

adjustment program came "off track" and the Bank withheld disbursements on itsadjustment operations. But these strains were moderate and transitory, and currentrelations are again good, and the outlook is for continued good relations at both thetechnical and the policy levels.'

1.5 The nature of the relations is changing as a consequence of Ghana's movementtoward democracy and the Bank's increased attention to participation, institutionaldevelopment, and governance issues. In the 1980s, the Bank dealt with only a small groupof leaders and technocrats accountable to an unelected head of state. In the 1990s,Ghana has an increasingly active Parliament (notwithstanding its one-party composition)and new forms of decentralized organization and accountability. If these trends continue,the Bank will increasingly deal with a wider set of Ghanaian constituencies.

Key Events and People

1.6 The main "events" that served to put Bank/Ghana relations on a new footing werethe change of government in December 1981, the economic collapse of 1982-83, and theappointment of a small core team of highly competent Ghanaian technocrats (also in1982-83) with a common, coherent view of what needed to be done and commanding theinstruments of central executive authority.

1.7 Given the concentration of power in the hands of the head-of-state, Flight Lt.Jerry Rawlings, the single most important "event" was his decision to turn to a market-oriented, outward-looking structural adjustment program-against the background of yearsof failed dirigiste/statist policies, external shocks, the failure of aid from the eastern bloc,and disruptions occasioned by early efforts to install a radical populist regime. Theculminating event was the official adoption of the ERP in April 1983.

1.8 Box 1.1 provides a chronology of both a number of important Ghana events andsome notable events in Bank/Ghana relations. Changes in the Bank's "Ghana team" in1982-85 were too gradual to be classified as an "event," but over these years the Bankmobilized a very strong core country team for Ghana. The coincidence of these twocounterpart teams, with their compatible personal as well as intellectual qualities, goes along way to explain why Bank/Ghana relations improved so much during the mid-1980s.Many individuals-both in the Government and in the Bank (and in other donoragencies)-did in fact exert heroic efforts to make good things happen in Ghana. In manyinstances, the key individuals who made a difference in the Bank/Ghana relationship, andin some of the Ghanaian outcomes, were at the "working level."

1. This distinction between relations at the policy vis a vis the technical level is an important one; in some cases relations maybe good at one level but not at another. It was not unusual, for example, for relations between Bank staff and country officialsto remain good at the technical level, e.g. regarding investment projects, even during periods when the policy dialogue was pooror absent. Ethiopia during the Mengistu regime was an example of this phenomenon. In Ghana. however, the relations at bothlevels have tended to move more or less together.

19

Box 1.1: Selected Notable Events, 1981-94

Year Ghana Events Events in Bank/Ghana Relations

1981 * Coup led by Flight Lt. Rawlings assumes * Bank/IDA lending suspended (Jun 81-Jan 83)power (Dec)

1982 * Severe drought/bush fures (1982-83) * Pdt. Clausen visit to Ghana* Close relations with Libya (1982-83) * Bank economic mission (Nov)* Peoples' and Workers' Defense Committees

mobilized* Unsuccessful coup attempts (1982-89)

1983 * Return of one million Ghanaians expelled from * Reconstruction Import Credit I (June)Nigeria * Consultative Group resumed (after 13-yr. hiatus

* Economic Recovery Program (ERP) launched (Nov)(Apr)

1984 * Attempted coup * Subsequent annual CG mtgs. (1984-93)* University reopened * Export Rehab. Cr. (also TA Cr.) (Jan)

1985 * National Economic Commission established * Reconstruction Import Credit II (March)

1986 * Strained relations with Togo * Industrial Sector Adjustment Cr. (Mar)* Coup attempt failed * Education Sector Adj. Cr. I (Dec)

1987 * Togo border reopened (May) * SAC I (April)* Student disturbances (Mar-June) * Pdt. Conable visit to Ghana (July)

* World Bank reorganization

1988 * District-level political reforms/elections * Financial Sector Adj. Cr. I (May)

1989 * District-level political reforms/elections * SAC II (April)

1990 * Investors Conference (Jan) * Education Sector Adj. Cr. 11 (May)* Formation of National Commission for

Democracy

1991 * Clashes in Northern Region (Gonja/Nawuri) * SAC III (May)* Financial Sector Adj. Cr. II (Dec)

1992 * Government awards 80% wage increases to * Agriculture Sector Adj. Cr. (Mar)civil service (Aug) * Adjustment credit tranche releases withheld (late

* Presidential elections won by Rawlings (58%) 1992/early 1993)(Nov) * Ghana 2000 and Beyond report issued (Nov)

* Parliamentary elections (NDC wins 189 of 200seats; opposition boycott) (Dec)

1993 * Opposition parties recognize legitimacy of * Bank releases tranche on SAC IIIgovernment (Aug) * Consultative Group (June) results in donor

* Over 10,000 Ghanaians evacuated from C8te pledges of over US$2 billiond'Ivoire (Nov)

1994 * Clashes in Northern Region, * Tranche releases withheld on FINSAC 11 andKonkomba/Nanumba (est. 1000 killed, refugee ASAC owing to delayed privatizationsmovements) (Feb)

* Privatization of Ashanti Goldfields

20

1.9 Compared to the dramatic events of the early 1980s, the remainder of the decadewas relatively uneventful. There has been a high degree of continuity in the politicalleadership and in the core economic team, relative peace with Ghana's neighbors, internalpolitical stability, and no major exogenous shocks (other than declining terms of tradeoffset by increased aid from the Bank and other donors). The main "event" of the rest ofthe decade was Ghana's gradual implementation of the ERP and adjustment program intowhich the ERP evolved.

1.10 Two closely-related events in 1992 are important to the Bank/Ghana relationship.One was the adoption of a new constitution and the elections of that year, which havealready changed, and probably will continue to change, the nature of government and theforms of governance in Ghana. By themselves, these changes have so far neitherstrengthened or weakened Bank/Ghana relations. They will however change the natureand forms of the relationship in significant ways. The implications for the Bank's strategyare discussed in Chapter 2. The other 1992 event was the Government's decision, shortlybefore the elections, to increase civil service wages by some 80 percent-an action thatwas not provided for in the budget and that triggered a large fiscal deficit, renewedinflation and exchange rate depreciation, and strains in government/donor relations.

1. 11 Following nearly a decade during which the stabilization components of theadjustment program remained "on track," with the inflation rate reduced to about 10percent (from over 100 percent), the increased fiscal deficit had a significant effect on theBank/country relationship. New concerns arose about the sustainability of adjustment, andabout whether the pursuit of votes would entail a serious erosion of fiscal discipline. TheGovernment was disappointed and to some extent angered by the Bank's suspension ofdisbursements on its adjustment operations-just at the time, from the Government's pointof view, when it most needed quick-disbursing funds. In 1993 and 1994 the Governmenthas taken new revenue and other measures-though not to the extent advocated by theIMF and Bank-to put the adjustment program back on track. The gains have howeverbeen offset in part by larger-than-planned borrowing by the state enterprise sector. Bycustomary criteria, however, donor/government relations in general and Bank/governmentrelations in particular are, again, quite good.

The Congruence of Government and Bank Strategies

1.12 The fact that the Bank and Government had shared objectives and an agreedstrategy for undertaking a reform program explains the good working relations and thegenerally high quality and effectiveness of the policy dialogue that evolved during the1980s. The two parties perceived each other as partners in Ghana's development. Therewere differences, but these had more to do with tactics, pacing, and sequencing ofreforms, than with the overall paradigm of reform. By around 1990, however, there wasless congruence between the strategy and policies proposed by the Bank and those thatthe Government wanted to pursue. But the differences and disagreements rarely becamecontentious.

21

Box 1.2: Conditions for Successful Adjustment in Ghana

A central question in the political economy of Ghana's reform program is: How did itprove possible for the government to mobilize and maintain the political support that made thereforms possible? Some answers are:

* There was a crisis, and a broad public perception of that crisis in the period 1980-83; therewas a consensus that the old order had failed. And especially because this crisis culminated along period of decline, there was a greater readiness by various interest groups to accept a newdirection of policy-even if this involved actions that had previously been anathema, e.g.devaluation.

* The political opposition was initially fragmented, demoralized, and intimidated. But by thetime of the 1992 elections, the main opposition parties were themselves advocates of continuedand even accelerated adjustment.

* Although the leadership did not share with the public a long-term vision of the kind ofsociety the reforms were intended to bring about, the ERP was a public and comprehensiveprogram that served as an action plan for transformation.

* The reform program was conceived and managed by a core team of technocrats with acommon, coherent view of what needed to be done and commanding the instruments ofconcentrated executive authority.

* There was continuity and long tenure by this core team and by the head of state underwhom they served.

* The government comprised an authoritarian regime headed by a charismatic leader whodid not have to run for re-election or be unduly concerned about delivering fast results.

* A fast albeit modest improvement in conditions followed adoption of the ERP; althoughthis was initially more attributable to recovery from the drought than to the ERP, it made forbetter ERP acceptance.

* Relations with donors improved considerably, and large aid flows permitted the benefits tobe greater and the adjustment process to be more gradual than would have been possible witha lower volume of aid.

* Notwithstanding many governance problems, corruption was relatively contained on thegovernment side. Donors on their side largely refrained from pushing their own favored whiteelephant projects.

* GDP growth (at about two percent per capita per annum) was sustained, withredistribution in favor of the richest and poorest, the latter largely in the rural areas. Thepoorer, rural beneficiaries of the reforms became the main supporters (in the 1992 elections)of the government party that had launched the ERP.

22

1.13 In the same month in which Flight Lt. Rawlings came to power (December 1981),the World Bank published a document that articulated its view of the problems andprospects of African countries. The document, entitled Accelerated Development in Sub-Saharan Africa (sometimes called the 'Berg Report") did not have specific directapplications to Ghana but was consistent with much of the Bank's advice to Ghana andwith many of its adjustment operation conditionalities. One central theme of the BergReport was the need for countries (such as Ghana) to 'get the prices right." This was amain objective of Ghana's ERP and of the conditionalities attached to Bank lending in the1980s. As this report will describe, Ghana has made much progress in getting its pricesright; it now faces a new set of binding constraints to sustained growth and povertyalleviation. The focus now is more on the need to 'get the institutions right," and to givemore attention to longer-term issues that were the main focus of the Bank's long-termperspective study on Africa published in 1989.2

1.14 Ghana has frequently been cited by senior Bank managers and others as an"African success case" insofar as it has sustained its reform program for over a decade,achieved five percent GDP growth and reduced poverty. This report will consider in somedetail the respects in which Ghana has been most and least successful in achievingdevelopment objectives. To put the Ghanaian economy and achievement in someperspective, the following box shows how Ghana compares with other African and low-income countries in terms of a number of key indicators.

2. Sub-Saharan Afico: Fromn Cnns to Sus2i,wbke Growth A Long-tam Peripecdve Snrdy, World Bank, 1989.

23

Box 1.3: Ghana Compared with Other Sub-Saharan African and Low-income Countres

How does Ghana compare with other Sub-Saharan African (SSA) countries? A comparison between key social and economicindicators for Ghana and these other countries shows that Ghana compares favorably with other African comparators, especiallyin terms of recent trends in these indicators, but lags behind low-income countries taken as a group.

Social PerformanceGhana compares favorably with SSA countries in terms of most social indicators (namely, incidence of poverty (headcountratio), total fertility rate, life expectancy, infant mortality and primary school enrollment) even though it has a lower per capitaGNP. Average daily caloric intake in Ghana, though slightly increased, failed to rccover its previous position (2,089 in 1986).The recent trend for these indicators is favorable, in contrast to much of the rest of Africa where per capita incomes havedeclined and poverty has worsened. But as the table below shows, the comparison is not so favorable with respect to lowincome countries (including China and India, which have comparablper capita income levels). Particularly notabl are thedifferences in fertility rates and primary school enrollmcnt rates.

Economic PerformanceWith regard to the economic indicators, Ghana also compares favorably in comparison with SSA countries but is far behindthe average for low-income countries, inclusive of Asian countries. Moreover, because Ghana's per capita income had declinedso much before 1983, and because population growth remains high, even the 5 percent annual growth in GDP since 1983 hasso far served mainly to restore average living standards to levels reached long before. As may be seen from the table, Ghanuis highly dependent on external aid but not more so than other SSA countries on average. But its savings performance is notablypoor in comparison to all low-income countries. This is also the case with respect to its private investment rate (not shown)of only 4 percent of GDP in 1993.

Social ldicatorsGhana

198 1993 Sub- Saha ran Low-incomelfr itca Cowunies

Poverty' (headcount ratio) 36 31 50(44) 19Total Fertility Rate 6.5 5.9 6.2 3.6Life expectancy at birth (years) 52 56 52 62Infant mortality per 1,000 98 79 93 64Average daily calorie intake' 1,948 1,974 2,144 n.e.Primaqry chool enrollment (% of age group) 80 74 67 101

Economic IndkatorsGhana

19W0 1993 SuSb-Saharan Low-incomeA19ica CourlesPer capita GDP(S) 370 430 520 380Average annual growth of GDP per capita (1980-93) n.a. 0.1 -0.S 3.7Gross Domestic Investnment (% of GDP) 6 15 16 30Gross Domestic Sevings (% of GDP) 5 -I 15 27External DependencyP (% of GDP), 1992 4 9 11 n.a.Net present value of external debt as % of GNP 35 47 n.a. 37Average annual growth of exports' -8 5 2 6Average annual growth of import -I 3 -2 2

a' Kilocalories per capita, 1992.b. Net ODA disbursements from all donors as percent of GDP.c. Comparing growth rates between 1970-80 and 19S0-93, respectively.

24

Overview of Bank Lending

1.15 During the four years prior to the launching of the ERP, IDA commitments anddisbursements to Ghana totalled about US$100 million. During the four years, FY83-86,IDA commitments rose to nearly US$500 million and disbursements exceeded US$200million. Thereafter, the lending program was further built up to commitment averagingabout US$250 million annually (FY87-FY94) and disbursement of close to US$200million. Of the total commitments of US$2.4 billion during FY83-94, adjustment lendingamounted to about US$1 billion, or roughly 40 percent. Table 1.1 below shows thedistribution of IDA lending and disbursements by major sector group. Assessments of therelevance and efficacy of this lending, and of the deployment of the Bank's non-lendinginstruments, are provided in the next chapters.

Table 1.1: Commitment and Disbursements in Ghana (FY79-94) (US$ million)

Variable 1979-82 1983-86 1987-90 1991-94 Total1983-94

Sector Group Name

Agriculture Sum of L/C $ 30 25 116 217 358Sum of Disb. $ 34 22 43 158 223

Industrial/IDF + Sum of L/C $ 19 60 177 141 378Financial Sum of Disb. $ 4 29 128 160 317

Infrastructure Sum of L/C $ 54 114 160 307 581Sum of Disb. $ 80 55 148 145 348

Social Sectors Sum of L/C $ 0 50 190 230 470Sum of Disb. $ 2 14 66 132 212

Multi-sector + PSM Sum of L/C $ 0 220 285 148 653Sum of Disb. $ 0 105 344 220 669

Grand Total Sum of L/C $ 103 469 929 1042 2439Sum of Disb. $ 120 226 730 814 1769

Source: Task Manager's Workstation Version 3.0.

25

2. Relevance of the Bank's Assistance Strategy

"The citical test of the success of any adjustment policy is growth. TheGhanaian program went for growth and has provided the wherewithal toenable Ghanaians to dare to hope."

Senior Ghanaian official

2.1 This chapter describes the evolution of the Bank's assistance strategy for Ghana overthe past decade and evaluates its "relevance."

Main Conclusions

2.2 The main conclusions are that: (i) the strategy for the 1980s of "going for growth"through a public sector-led recovery program was highly relevant in its priority objectives andits instruments; (ii) the strategy of the late 1980s and early 1990s has only gradually come torecognize the unsustainability of the growth pattern of the 1980s and the critical importance ofinstitutional constraints ("implementation issues" and "sustainability issues"); and (iii) Bankstrategy documents have been strong in their diagnosis of constraints to near-term growth butweaker with regard to risk assessment, priority setting, long-term issues, and accountability.The changed binding constraints and political economy of the 1990s require the Bank to adjustits priorities and its mix of instruments to reflect the new realities, needs and perceptions.

2.3 The Bank's country assistance strategy over the past ten years has been relativelystrong with regard to its near-term "adjustment agenda" but weaker with regard to integratinginto this agenda the objectives of poverty reduction, institutional development andenvironmental sustainability. The strategy has so far been successful in fostering growth ledby Ghana's public sector. But both the country's and the Bank's strategies were slower toconfront the constraints to private sector development. What is now needed on the Ghanaianside is a clear vision and enunciation of the country's own long-term development strategy.In turn, what is needed on the Bank side is a clear elucidation of what the Bank and itspartners can and cannot hope to accomplish in helping Ghana to achieve an economictransition to a more sustainable, less aid-dependent, private sector-led growth process.

Selective Coverage

2.4 For purposes of this report, the main "objectives areas" will be public sectormanagement and policy reform, institutional development, educational sector development,and agricultural development. Two overarching objectives that have not been much focussedupon in past Bank strategies, but which are of great importance, are good governance andsustainability. Given the need for selectivity, this report will not have much to say aboutother important objectives and sectoral issues, e.g., gender issues, environmental protection,

26

infrastructure, family planning, urban development, and so on. The main question on privatesector development is: Why has the private sector response in Ghana been so sluggish despitethe many favorable macroeconomic measures taken? With regard to the Bank's instruments,particular attention is given to adjustment lending and the related policy dialogue, ESW(mainly as it related to adjustment lending and the dialogue), and to selected TA operationsand aid coordination initiatives.

Bank Strategy Papers

2.5 The Bank's assistance strategy for the period 1983-93 is embedded in many strategy-related documents, including Business Plans, annual Country Assistance Management System(CAM) statements of objectives, Country Program and Budget Papers (CPBP), sectoralstrategy papers, Policy Framework Papers, among many others.3 Between the early-1970sand 1988, the basic Bank document used for country assistance strategy review was theCountry Program Paper (CPP). The CPP was superseded in 1988 by the Country StrategyPaper (CSP), which was intended to serve more distinctly as a strategic rather than aprogramming document. For the present CAR, the four CPPs/CSPs prepared between 1981and 1993 provide these benchmarks.

2.6 After the 1981 CPP, a new CPP should have been prepared by 1984 (at the latest) toreflect the major changes in the Bank's country assistance program being made to respond tothe ongoing changes in Ghanaian policy. In the event, the next CPP was not reviewed bysenior management until January 1986. In the absence of a CPP for the period 1983-85,some internal Bank documents (especially Back-to-Office Reports) from missions in late 1982and early 1983 are taken as a proxy. A Country Strategy Paper was prepared and reviewedin 1990 and again in 1993.4 No further CSPs will be prepared because the CountryAssistance Strategy (CAS) process is subsuming the CSP process and document. The latestCAS for Ghana was reviewed by the Board in April 1994.

3. The Policy Framework Paper (PFP) is of course different from the other documents in that it represents a statement of agovernment's development program, rather than of the Bank's strategy and program. But PFPs are nevertheless strategydocuments for the Bank as well as the govemnment in that the Bank ofFicially endorses such programs and pledges to help thecountries implement them.

4. The CPP guidelines in effect between 1979 and 1988 called for CPPs to be prepared at intervals of not longer than threeyears. The CSP Guidelines issued in 1988 called for CSPs to be prepared at two-year intervals for countries receiving IDAcomnritments in excess of US$100 million. Hence, the Region was not in compliance with OMS 1.12in this regard. Accordingto the strict 'leter' of the guidelines, a CSP should have been prepared in 1988 and in 1992 rather than 1993. It mnay howeverbe argued that there was compliance with the 'spirit' in that both senior management and the Board were continuously apprisedthrough various other documents (e.g., reports on adjustment operations) of the Bank strategy and lending program. This wasmoreover a period in which there was considerable continuity in the Government's strategy and Bank's program, so that newstrategies were not required by signilicant changes in country conditions.

27

Evolution of the Bank's Assistance Strategy

Overview

2.7 The Bank's strategy for Ghana over the period 1983-93 was characterized byconsiderable continuity in its basic objectives and instruments. The central objective was tohelp Ghana adopt and implement an "adjustment agenda" largely in keeping with the tenets ofstabilization, liberalization, and "market-friendly" policies as advocated in particular by the1991 World Bank Development Report entitled The Challenge of Development. In particular,IMF and World Bank policies during the period under review put emphasis on: exchange raterationalization (geared to restoring export competitiveness and inducing rapid growth in non-traditional exports); fiscal discipline; tax and tariff reform (geared to improving incentives toproduction, broadening the tax base and improving efficiency while at the same time retainingequity objectives); price and trade liberalization; deregulation; legal reforms and measures tosecure property rights, promote competition, and stimulate private investment (includingforeign direct investment); privatization and reform of state enterprises; and financialliberalization. This agenda of reforms geared to "getting the balances, prices, and incentivesright" dominated the policy dialogue and the Bank's (and IMF's) lending conditionalities inthe decade 1983-93.

2.8 Initially the IMF took the lead in the policy dialogue owing to the high priority givento stabilizing the economy and adjusting the exchange rate. A distinguishing characteristic ofthe strategy was its "gradualism." The sequencing was orthodox in that it moved from initialemphasis on stabilization and rehabilitation projects to liberalization and adjustmentoperations. Sectoral adjustment operations preceded the structural adjustment operations.Both the dialogue and the lending program moved gradually towards deepening the adjustmentprogram in areas involving more difficult governance and institutional development issues,e.g. in privatization, public enterprise restructuring, civil service reform and private sectordevelopment.

The Strategy in the 1981 CPP

2.9 The 1981 CPP was prepared against the background of longstanding deterioration inBank/country relations and in the country's economic situation. In the three prior years to(FY78-80) only three IDA operations (DFC II, Rural Development 11, and Roads 111) wereapproved for a total of US$54 million. This was tantamount to a "core program" or holdingoperation.' Non-lending services wvere also minimal since the CG was suspended and therewas low demand for ESW. The 1981 CPP was nevertheless an "advocacy document," and itmade a case for a large increase in the lending program, from the average US$18 million ofthe FY78-80 period to a proposed commitment level averaging US$80 million for the periodFY82-86. In addition, two large SALs were identified as reserve projects for FY82 and

5. With the benefit of hindsight, it appears remarkable that the Bank found Ghana to be still creditworthy for IBRD lendingas late as FY76-77-and not for enclave operations. The FY76-77 IBRD operations were for power, industry, a DFC, and roads.But since the 1970s are prior to the period covered by this CAR, no attempt has been made to assess how this came to be orthe lessons to be derived from that period.

28

FY84 in the event the Government moved more vigorously to undertake economic reformsthan was assumed in the base case.

2.10 The CPP noted the poor economic performance and policy shortcomings but madea case for increased lending and ESW based on a perception that the (then) new civiliangovernment "seemed eager to receive outside advice on economic and sectoral policymatters."6 But the CPP hedged its bets: the "most likely" scenario was a muddle-throughscenario entailing modest policy change at best-well short of what was needed to justifyadjustment lending. Yet, the prospects for improving the policy dialogue were argued tobe sufficiently good as to warrant a substantial increase in investment lending. No specifictargets or events were identified as "triggers"to a larger or smaller lending program.

2.11 The Bank's senior management did not accept the Region's proposed strategy andlending program; the internal contradictions were all too apparent. Proposed projectlending was curtailed in the absence of substantial policy changes, including exchange rateadjustments. Staff were authorized to prepare for non-project lending in case of acoherent reform program and instructed to focus on longer-term structural macroeconomicissues and the cocoa sector. No new loans were approved between June 1981 andJanuary 1983.

The Strategy in 1982-83

2.12 No CPP was prepared between 1981 and 1986, but the evolution of the strategycan be traced from other documents and interviews with Bank staff. A key document(December 1982) noted that whereas the Bank had kept the lending program "captive" tomacroeconomic reform (which the Ghanaian authorities were starting to considerseriously), the Ghanaian authorities' receptivity and interest in pursuing an economicdialogue with the Bank was very much dependent upon an early resumption of Banklending. Given this perceived linkage, the case was made for resumed lending even if theGhanaian policy measures fell short of expectations, on the grounds that providing thesupport needed to make the reform package workable would enable the Bank to establishits credibility with the authorities and influence the course of events.

2.13 In the event, this high-risk approach was adopted and the Bank did resume lendingin 1983, even though the Ghanaian program was perceived to be far from a "firstbest" andthe prospects for sustaining the reforms were highly uncertain. Within the Fund and Bankthere were two schools of thought. Some advocated going ahead with lending, despite thegradualism of the 1983 ERP, in the hope that the Government would find the politicalwill to sustain and deepen the reforms with the aid and advice of the Fund and Bank, andsubsequently other donors. Others were skeptical of the Government's commitment toreform and its ability to sustain even a modest program. It is not clear to what extent thepro-lending group in the Bank may have had any influence on the Fund's decision to goahead with its 1983 standby arrangement, but in any event this paved the way for theBank's own resumption of lending (see Box 2.1 on Bank-Fund Coordination).

6. In retrospect this appears to have been wishful thinking.

29

Box 2.1: Bank-Fund Coordination

Even before the PFP's introduction in 1986-87, Bank-Fund collaboration onGhana was comparatively close. From the beginning of Ghana's ERP in 1983, the twostaffs worked in tandem, and Ghana was chosen as one of six pilot countries for"enhanced collaboration" between the Bank and Fund. This led to more frequent andtimely exchanges of information and draft documents than typically characterizedBank/Fund relations. In addition, it led to a joint seminar to discuss a variety of issuesrelating to Ghana's development. This seminar included numerous staff from theBank's (then) Projects Department who normally had little or no contact with theFund. This kind of initiative was exceptional at the time.

The division of labor between the Bank and Fund largely followed traditionallines. Typical differences arose over the years when, for example, Bank staff felt thatFund requirements on the overall budget deficit and credit ceilings were restrainingrecovery, and having adverse effects on Bank projects. Fund staff on their part hadreservations about the Bank's endorsing Ghana's gradualism in adjusting the exchangerate (implying multiple exchange rates) and about Bank proposals for the foreignexchange auction. Differences also arose concerning the matter of cocoa producerpricing where, after the initial win/win situation (in which raising cocoa producer priceshad the effects of increasing both production and revenues as well as reducingsmuggling), there were some trade-offs between fiscal and incentive objectives.

A major Bank/Fund disagreement arose over the size of the Government's wagebill as the Bank supported a civil service reform program that called for a higher ceilingthan the Fund proved willing to accept. Bank staff felt strongly that in that instancethe Bank "capitulated" to the Fund for reasons other than the technical merits of theFund's position. A government view of this episode is that the earlier failure to moveahead with the civil service reforms, including the Bank-supported selective salaryincreases, was partly to blame for the 1992 "explosion" in the wage bill.Notwithstanding the occasional Bank/Fund disagreements, the staff of both institutionsbelieve that Bank/Fund collaboration on Ghana was effective over the years. The staffcooperated closely in the preparation and monitoring of Ghana's economic andfinancial programs, with some joint participation in missions as well as overlapping ofmissions facilitating this cooperation. The PFP served in fostering this collaboration.

2.14 The Bank's 1983 economic report laid out in considerable detail a sequencedprocess of adjustment moving from stabilization to rehabilitation to liberalization andgrowth. Table 2.1, taken from that report, shows the range of policy and institutionalreforms that the Bank was to advise and assist with for the next ten years. This economicreport was, therefore, an important, high-quality strategy document, as it laid out theBank's recommended program for Ghana to get its prices right, to improve capacityutilization, and to reduce government intervention in the economy. It also provided

30

Table 2.1: Ghana: Phasing and Sequencing of Adjustment Process

Stabilization Rehabilitaton LiberalzatonPhase Phase and Growth

Phase

Polcy Reform(a) Pricing Reform Package1. Exchange Rage x o +2. Interest Rate x o

3. Wage Rate x o +

4. Energy Prices x o +

5. Infrastructure Prices x o +

(b) Trade and Industrial Policy Package

6. Industrial Incentives x o

7. Nontraditional Export Promotion x

8. Price Controls x o

9. Trade Restrictions x

(c) Investment Package

10. Public Investment Program x

1 1. Private Sector Initiatives x

(d) Taxation and Subsidy Package

12. Tax Reform x o +

13. Consumer Subsidies x o +

14. Producer Subsidies x o +

(e) Human Resource Development Institutional Reforms x o

15. Planning Process x o

16. Investment, Appraisal, Monitoring and x oEvaluation

17. State Enterprises x o

18. Export Marketing Boards x o +

19. Mining Companies x o +

20. Statistical and Information System x o +

21. External Debt Management x o +

22. Financial Intermediation x o

23. Agriculture Support Services x o

24. Civil Service Reform x o

Legend: x: initiate actiono: continue action, evaluate and adjust

+: sustain action

Source: Ghana: Policies and Program for Adjustment. World Bank Report No. 4702-GH, Oct. 1983.

31

recommendations for sectoral reforms and recovery measures, and projected thesubstantial external assistance necessary to make the reform program succeed, and servedas an instrument of aid mobilization.

The 1986 CPP

2.15 By the time the next CPP was reviewed in January 1986, there had been a seachange in Bank/Ghana relations and in Ghana's economic situation, triggered by thelaunching of the ERP. The Government had by then implemented a wide-ranging set ofreforms including exchange rate devaluation, increased cocoa producer prices, removedprice controls, rationalization (partial) of petroleum and utility prices, reduction inexternal arrears, and the exercise of fiscal and monetary discipline. These reform effortswere supported between mid-1983 and end-1985 by two IMF standby arrangements(totalling over SDR 400 million) plus two CFF drawings (totalling another SDR 178million).

2.16 The Bank had established a good policy dialogue, taken the lead in aidmobilization (including through chairing resumed annual CG meetings) and mounted asubstantial lending program, up from zero in FY82 to US$73 million in FY83, US$125million in FY84 and US$175 million in FY85. Cofinancing of IDA projects in those yearstotalled US$127 million. The lending program was a mixture of low-conditionalityreconstruction import loans, sector rehabilitation loans and specific investments. IFCmade its first investment in Ghana in 1984 in a project to increase the output of theAshanti Goldfields Corporation. Against this background, the 1986 CPP proposed tofocus the policy dialogue an1 an increased number of adjustment operations onbroadening the scope of the Government's reforms into a full-fledged structuraladjustment process. This was to include more action on the exchange rate, furtherir creases in the producer price of cocoa, domestic resource mobilization efforts, increasesin both current and development expenditures, reforms in the financial sector, publicenterprise reforms, and progressive liberalization of the trade regime.

2.17 The CPP proposed to maintain an average of US$85 million per year throughFY90 with sector adjustment loans (in agriculture, industry, education and health), twoSACs with conditionalities focussed on the areas mentioned above and a number ofinvestment projects.7 The CPP focussed on agriculture, with 6 projects (out of 21) and 19percent of total lending. Industry and transport were both allocated about 14 percent,education and hLalth each about 7 percent, and adjustment lending about 32 percent.

2.18 The CPP attached a "relatively high probability that needed policy changes by andlarge will be achieved," but it noted that "should the [reform] program seriously falter,both the level of lending and the number of operations would be scaled down." The CPPdid not however contain any specific performance criteria or triggers, and it provided onlya single lending scenario. Three possible risks were identified and briefly discussed. Thepolitical risk was seen as a risk of instability that might arise from the regime losing the

7. Not including resources from the Africa Facility that were projected at about US$20 million per annum.

32

support of the political base of the trade unions and working classes (on which it came topower) before it could forge new alliances in the rural areas, the main beneficiaries of thenew policies. The implications of these risks being realized were not however identified.The other risks identified were those of drought and falling cocoa prices. In both cases,implications were drawn for the balance of payments and financing gap, but nocontingency plans were identified, nor were other implications assessed.

The Strategy in the 1990 CSP

2.19 The 1990 CSP was the first CSP for Ghana. By that time, Ghana had attained thestatus of a model client. In the Bank's annual country performance ratings (used toestablish "norms" for IDA allocations), Ghana had received very high ratings for severalyears. This was reflected in an "excellence premium" in both IDA allocations andadministrative resources for economic and sector work, policy dialogue and lending (CAMresources).8 Thus, actual IDA allocations to Ghana in the late 1980s were averaging overUS$200 million per year-more than double the level requested in the 1986 CPP. TheFY88-91 CAMs allocated to Ghana an average of 32 staff years per year-more than forany African country save Nigeria.

2.20 The Government's ability to pursue its reform agenda during the years between1986 and 1990 was enhanced by the fact that this period was relatively uneventful in termsof either major political changes or external shocks. In keeping with its plan to movetowards an elected national government, the Government established elected districtassemblies during the late 1980s, but as of 1990 the timing and modalities had not beenclarified. The 1990 CSP envisaged that a broader sharing of power on the political frontwould help foster support for further economic reforms to maintain the pace ofdevelopment. It did not envisage an alternative scenario-such as the one that cameabout in 1992-when an elections-related wage increase for civil servants put themacroeconomic program off track and threatened to dampen the pace of development.

2.21 The 1990 CSP was a more sober document than other reports on Ghana at a timewhen the country was being extolled as a "success story" and when some Bank staff andmanagers' thoughts were turning to how to accelerate Ghana's growth to the level of thefast-growing countries of East Asia.9 The CSP did recognize Ghana's "success" inimproving production incentives, restoring fiscal and monetary discipline, increasing publicinvestment, and in managing well its external financing strategy. But it also flagged thedaunting tasks that lay ahead, noting that inflation was still high, the private investmentresponse weak, the Government's implementation capacity overstretched, absolute povertystill widespread (estimated at a third of the population), and the environment threatened.

2.22 The 1990 CSP defined three priority objectives. The first was to intensify effortsto promote private sector development, recognizing that this would require the

S. lhe performance ratings and IDA allocation exercises remained procedures intemal to Bank management.

9. In the early 1990s. references were being made to Ghana as a prospective 'tiger of Africa.

33

Government to take a more "hands off' approach to managing the economy than it haduntil then been willing to adopt. Specific measures recommended included improving theregulatory framework for private investment, with comprehensive financial sector reforms,accelerating the divestiture and commercialization of state-owned enterprises, andimproving services and infrastructure crucial to agricultural smallholders. The secondpriority was given to improving public sector management and implementation capacity.Implementation weaknesses were flagged as a "binding constraint" and attention was givento the deterioration in the Bank's portfolio performance as indicated by a decline in grossdisbursements, longer intervals to effectiveness, and an increase in total undisbursedbalances for investment projects relative to appraisal estimates. The CSP proposed torelate the volume of future lending to implementation performance, with undisbursedbalances comprising the key indicator.'°

2.23 The CSP also proposed a few measures to encourage the Government to addresshuman resource development, poverty and environmental issues that were seen toultimately determine the economic and social sustainability of the reform program. As inother areas, however, no specific performance criteria or targets were established, and inconsidering what would warrant a reduction in lending, the CSP argued that owing to the.complex and subtle nature of the judgments that will have to be made ..., no singleindicator or group of indicators can be used to trigger a decision." The CSP proposed abase case of lending during the period FY91-95 averaging about US$250 per annum, ofwhich US$80 million annually for adjustment lending. This compares w.th average annuallending of US$21' million during the period FY86-90, when about half of total lendingwas for quick-disbursing operations.

The Strategy as Described in the 1993 CSP

2 24 The 1993 CSP was prepared in the first half of 1993, shortly after the electionsand the 80 percent wage increase to civil servants. The wage increase resulted in anincrease in the fiscal deficit by about 6 percent of GDP, a more than 50 percent increasein the money supply, a sharp depreciation in the exchange rate, and a collapse of privateinvestment. This was characterized as a (presumably one-time) "cost of the politicaltransition," indicating that the economy seemed to be stabilizing satisfactorily in the firsthalf of 1993. The tone was generally sanguine. The 1993 CSP was significantly influencedby the completion in 1992 of Ghana 2000 and Beyond, a major Bank report producedwith large Ghanaian participation, which emphasized the need for action in three areas inorder to accelera:e the growth rate well above the 5 percent rate achieved during theprevious decade. These areas were: private sector development and export promotion,public sector management, and human resources development. In these respects the 1993CSP appropriately put increased emphasis on the long-term issues.

10. It was proposed that if total undisbursed balances for investment lending were higher then aggregate appraisl estimatesby more than a 50-75 percent range, then the volume of lending in the following year would be reduced to US$150 million, withcutbacks focussed on sector- of week performance.

34

2.25 The CSP acknowledged that a major impediment to private sector confidence(private investment dropped to less than 4 percent of GDP in 1992) was uncertainty aboutgovernment attitudes toward the private sector. Steps seen as essential to the emergenceof a vigorous private sector included removal of old inhibiting regulations, increasedtransparency and due process in enforcement, elimination of implicit transfers to publicenterprises, more clarity in the limits to private sector activity, and speedy divestiturecutting through red tape on asset valuation and labor laws.

2.26 With regard to public sector management, the CSP identified the need for a re-invigoration of civil service reforms plus steps to make management more effective, reduceuneconomic subsidies on water, electricity and transport, strengthen expendituremanagement, improve tax collection, etc. Capacity building was to become a major focusof Bank efforts, based on a capacity building assessment then under preparation. Withregard to human resources development the CSP called for a big push in primaryeducation, which was seen as critical to both growth and poverty reduction. The CSP alsonoted that the Bank "would change the way it does business in Ghana" by paying greaterattention to the overall issue of ownership and avoiding crowding out Ghanaian initiatives.

2.27 The CSP noted that although some improvements had been effected in portfolioperformance following a major restructuring of slow disbursing projects, the average paceof IDA projects remained slow, partly owing to the non-availability of counterpart funds.It proposed to reduce the share of adjustment lending to below 20 percent in FY94-96(compared to 33 percent in FY90-93 and 47 percent in FY83-89) subject to improvementsin the investment project disbursement rate that would enable total IDA disbursements toremain at a level of about US$200 million annually.

Relevance of the Bank's Strategy

2.28 The criteria used to assess relevance include: adequacy of the diagnoses; realismof assumptions; clarity of definition of objectives; timeliness; appropriateness of thepriorities given to component objectives; appropriateness of the mix of instruments;adequacy of risk assessment; realism/feasibility of implementation; and sustainability."The following paragraphs consider the Bank's strategy for the full 1982-93 periodaccording to each of those criteria. The chapter concludes with an overall assessment ofthe strategy's strengths and weaknesses.

Right Diagnosis? Short-term, Yes. Long-term, More Work Needed

2.29 Getting the strategy "right" in 1982-83 depended on a good diagnosis of the stateof the Ghanaian economy. The Government was poorly equipped to do its own in-depthESW. The Bank was also handicapped by the fact that its ESW on Ghana had been

I1 In project evaluation, austainability is assessed independently of relevance, and there can be cases where projects are foundto have a utisfactory outcome even when the benefits are unlikely to be sustainable. In the case of an overall long-term Bankauiatance program, however, the targeted outcome is itself a sustainable improvement in the country's welfare, so it is impossibleto inagine a satisfactory outcome that is not sustainable. Hence, in CARs, unlike in PARs, sustainability will be treated as acriterion of relevance.

35

negligible in the years prior to FY83. This vacuum needed to be filled in a hurry. Thisneed however could not be met as fast as was desirable because of the lack of qualifiedstaff with pre-existing country knowledge and experience. With the decline in lending, theeconomists and sectoral specialists had been transferred to work on other countries. Ittook some time, therefore, for the stock of people and the stock of needed countryknowledge and understanding to be built up.

2.30 The costs of the delays were the weaker designs of some initial operations (e.g. theexport rehabilitation project and associated TA project of 1983, both rated unsatisfactory,and an industrial sector adjustment credit with some serious design flaws).'2 It tookseveral years for the Bank to build up its lending for adjustment, for institutionaldevelopment, and for agriculture.

2.31 Notwithstanding these handicaps, the Bank did manage to produce some high-quality economic work, beginning with the 1983 CEM, that provided a generally sounddiagnostic basis for the strategy of the mid-1980s. In the view of a number of seniorGhanaian and Bank officials, however, it has proven difficult to get similarly robustdiagnostic foundations in some of the sectors, and on some cross-cutting issues such as ID,privatization, and public enterprise reform. With few exceptions, moreover, most of theBank's ESW tended to focus on relatively near-term adjustment issues rather than uponthe longer-term issues.

Clarity of Definition and Monitorability of Objectives: Good in the 1980s; Fair in the 1990s

2.32 A good strategy shotuld have a limited number of well-defined and well-prioritizedobjectives, as objectively monitorable as possible. In all of the Bank's strategy papers,however, the objectives tended to comprise either such a large number of broad goals(G)DP growth, stabilization, improved public sector management, environmentalprotection, etc.) or so many specific "key" and "main" and "high priority" objectives, that itwas hard to discern the truly highest priorities.

2.33 The guidelines for CPP and CSP preparation were at least partly responsible. TheCSP guidelines, for example, called for the document to "listthe Government's keydevelopment objectives ... land] discuss central issues in the economy and its main sectors,"with such listings and discussions not necessarily calling for an evaluation of the strategicfocus of the Government's program-or lack thereof. And while the CSP section on theBank's assistance strategy was required to "recommend priorities for policy change' and to"focus only on the main macroeconomic and sectoral issues,"the requirement to prioritizewas not much enforced.

2.34 In the 1980s, the nature of the main goals (stabilization, liberalization,rehabilitation) made it relatively easy to define the relevant key performance indicators,e.g. movements in the inflation rate, real exchange rate, recovery of cocoa exports, etc. Inthe 1990s, the emphasis on private sector development and capacity building, for example,

12. The quality and impact of specific projects will be reviewed in Chapter 3.

36

makes it harder to sort out which of ttie underlying constraints are more binding thanothers, to identify the relative priorities of actions needed, and to define the appropriatekey performance indicators. This problem needs to be approached from two directions:One is to put more precision and prioritization into the specification of objectives; theother is to develop performance indicators that constitute the best available proxies formeasuring qualitative data in a quantitative way."3

Realism of Basic Assumptions: Partial

2.35 Some of the Bank's underlying assumptions about key variables were off the mark.For example, the terms of trade deteriorated significantly more in the late 1980s and early1990s than was envisaged, inflation proved harder and slower to bring down than waspredicted, aid inflows (disbursements) lagged behind forecasts, and there was over-optimism in the projected trends in agricultural output and in the growth of domesticsavings and private investment. And the Bank now readily acknowledges that, in the earlyand mid-1980s, there was an institution-wide underestimation of the time required forcountries such as Ghana to implement the full range of reforms and to reap the benefitsin the forms of increased investment and productivity and reduced poverty. Underlyingthis optimism was an under appreciation of the importance of non-policy constraints. Thissaid, however, none of the above-mentioned projections was so far off the mark as toundermine the strategy so long as the Government remained committed to pursue thereforms-including making adjustments as necessary when some of the initial assumptionsproved wrong. As the Bank made the right assumption about the Government'sownership of and commitment to its reform program, this sufficed to make realistic theassumption that the consequences of other wrong assumptions could be managedsatisfactorily. "

Timeliness? Yes

2.36 It was a main contribution of the Bank (and the IMF) that they supported in atimely fashion the Government's emergent reform program, a risky thing to do in 1983.This was not only because the new Ghanaian team lacked any track record in managingreform, but also because some of its members had a record of anti-market actions andanti-Bank/Fund rhetoric. Against this background, staff in charge of the Ghana program

13. There is often some resiaLance to the adoption of any set of performance indicators on the grounds that they imply a*targetry, or may be used as trigers in a mechanistic way, or may lead to excessive micromanagement. The 1990 CSPreflected this resistence in seeking to avoid quantitative triggers that could result in a lowering of the lending program. Thpoaition advocated later in thia report ia that a et of 'sustainability indicators" should comprise a core set of indicators by whichthe performnce of the country's strategy and Bank's assistance should be assessed. There is considerable truth to the sayingthat 'what gets measured gets done'-especially to the extent that the list of priority indicators is short. But this emphasis on"getting the right indicators" does not at all imply that such indicators should be instruments of micromanagement. Quite thecontrry!

14. If the Rank has been too widely off the mark in some assumptions, e.g. about the extemal financing requirements oravailabilities, or the severity of terms of trade shocks, then it is true that even a govemment fully committed to reform couldno have managed the shocks or shortfalls, and the strategy might have failed for lack of realism in the assumptions. But thisdid not happen.

37

made a strong case for accepting these risks-at the risk of their own credibility-and theBank made a substantial upfront commitment of resources.

Appropriateness of Priorities? Short-term, Yes. Longer-term, Uncertain

2.37 A corollary of "getting the priorities right" is "getting the sequencing right." And asnoted above, this is highly dependent upon getting a right diagnosis of which constraintsare most binding, and which are most amenable to remedy and relief through changedpolicies, financial inflows, and projects and programs. In general the Bank'smacroeconomic analysis and policy advice in Ghana were sound. The Bank advocated atighter budget policy to reduce inflation, a more flexible and realistic exchange rate policyto restore competitiveness, and a move to positive real interest rates both on efficiencygrounds and to restore and maintain external balance. The Bank supported themovement toward a free foreign exchange market, the removal of quotas and internalprice controls, and tariff levelling and reduction. It supported financial sector reform andinstitution building in the two financial sector credits. The Bank was also aware from thebeginning of the problem of low private saving and investment and with partial successsupported through the private investment credit measures to raise them.

2.38 The Bank strategy as manifested in its economic studies, policy advice, andadjustment operation conditionalities was largely in line with the tenets of consensus, andother benchmarks of "best practices" in adjustment advice and lending of the times. Therewere, however, some flaws in the design." The first was the early sequencing of theadjustment program. The Industrial Sector Adjustment Credit (approved in March 1986),for example, promoted investment in sectors that were soon to become obsolete due totrade liberalization. It reversed the sequence of stabilize, liberalize, invest, so that thecredit supported some non-viable firms. A second problem was the lack of provision ofassistance for potentially viable firms in the wake of the trade reform, resulting in a largenumber of bankrupt firms still in receivership. A third deficiency was in the financialsector, where an initial emphasis on cleaning the balance sheets of the banks meanttreating symptoms rather than root causes, and the lack of pressure to make the financialsystem more competitive left the credit and foreign exchange markets with oligopolystructures. This contributed to the high real interest rates and a lack of availability offoreign exchange. "

2.39 The Bank's strategy may also be faulted for being slow in identifying andaddressing the ir.stitutional impediments to the reforms. It was not until 1987, forexample, that the Bank approved a project to strengthen the capacities of the coreeconomic ministries/agencies. A more appropriate sequencing would have aimed atimproving implementation capacities earlier. And as will be discussed in Chapter 3, the

15. 'fhe strengths and shortcomings on the implementation side will be discussed in Chapters 3 and 4.

16. The main cause of recent inflation, high real interest rates and forex scarcity was the government's lack of fiscal disciplinein 1992-93, with the consequence that too much of the burden of stabilization fell on monetary policy. This was however afailure in the government's implementation of its own stabilization program, not a shortcoming in the design of the Bank'sstrategy per Se.

38

Bank's overall strategy is still handicapped by the lack of a comprehensive and coherentID strategy that deals adequately with demand side as well as supply side constraints,'7and with economic governance factors.

2.40 As identified in the Bank's latest report on governance,'" some important facetsof economic governance with which the Bank may legitimately be concerned (because oftheir direct bearing on development prospects) are the following: predictability, openness,and competence of policymaking; professional ethos of the bureaucracy; accountability ofthe executive branch of government; participation of civil society in public affairs; rule oflaw; and incidence of armec. conflict among factions, tribes, and clans (within and acrosscountry borders). Neither the formal strategy papers nor other internal documentsreviewed in the archives gave much attention to these issues uI.til very recently.'9

2.41 Finally, and perhaps most importantly, the Bank should have sought, earlier andmore forcefully, actions by the Government to foster private sector development,especially to improve investor confidence; to accelerate the slow processes of privatization,public enterprise restructuring, and civil service reform; to identify and solve fundamentalproblems in expenditure management (a main cause of the "counterpart fund problem"),to address some of the above-mentioned governance issues seen to impede both privatesector development and ID; and to give more attention to longer-term issues such asfamily planning, agricultural productivity, and environmental deterioration-the "nexusissues."

2.42 IMF staff who were involved in designing the arrangements of the 1980s alsobelieve that the IMF (and implicitly the Bank) should have pressed for earlier actions infinancial sector reform and state enterprise divestiture in particular (Box 2.2). This IMFdiagnosis is consistent with the views of many Bank staff. The Fund's report emphasizesthe crowding out effects, financial and otherwise, that made the slow pace of privatizationand public enterprise reform a constraint to private sector development. TheGovernment's lack of action on these fronts was perceived by the private sector as a'danger signal," given its fears that the "unlevel playing field" in favor of state enterprisesmight persist, and that the Government might even reverse the trend towards promoting amore 'market friendly" enabling environment for private sector development.

2.43 Some government officials and Bank staff argue that it would have been futile oreven counter-productive for the Bank and Fund to "press"for more and faster stateenterprise reform and privatization when the political leadership was strongly resistent to

17. The demand side constraints refer to those factors which impede effective utilization of existing capacities, individual andinstitutional, e.g. the government's bypassing (at a political level) of the civil service when it comes to decision-making onimplementation isues.

IS. Govntan=c: 7hc World Bank's Expericnce, World Bank, 1994.

19. This is not to uy that the Ghana team lagged behind the rest of the Bank in this regard. Yet even the 1994 CAS hesitatedto mention the erious tribal conflict in the north of Ghana in February 1994-an event that seemingly called for interpretationas to ita significance and implications.

39

these measures. To be sure, strong Bank conditionalities in these areas, even if agreed toby the government, would not have been "owned" by the government and would not havebeen implemented effectively. The view of this report is not that the Bank and Fundshould have "pressed" for reforms in these areas through conditionalities, but rather shouldhave concentrated upon doing more and better ESW and establishing a better andbroader dialogue on the options, including non-conventional forms of privatization.

2.44 Important reasons why liquidations and divestitures were probably more needed inGhana than in many other countries were that (a) the SOE sector was (and still is) large;and (ii) the level of private sector confidence was so low that a "non-gradualist" approachto dealing with the enterprises seems all-the-more necessary to provide a signal to theprivate sector that the government is truly serious about its market-friendliness, and willnot soon reverse its policies.

2.45 The point is often made that reforms such as SOE rationalization, privatizationand financial sector reform require much more administrative and institutional capacity tocarry out than, for example, price and trade reforms. This is quite true. But the bindingconstraint to movements on these fronts in the 1980s and early 1990s was not so muchGhana's lack of implementation capacity as the lack of political will and the failure toeven design a program that could in time be implemented, even if only gradually in theface of the implementation constraints. Finally in 1994-95, however, the government didbegin to move ahead with both the design and implementation of a number ofprivatizations.

Box 2.2: IMF Views on the Sequencing of Ghanaian Reforms

Overall, while the sequencing of reforms was broadly appropriate, with the benefitof hindsight it is apparent that the economy would have benefitted from a more forceful,and perhaps less gradual, implementation of the state enterprise and financial sectorreforms as well as the liberalization of the institutional and administrative framework.

This would have speeded up the development of private sector activity, includinghigher rates of saving and investment, as a consequence of the strengthening of thefinancial position of state enterprises and a deepening of the financial intermediationprocess resulting from a more efficient banking system.

Admittedly, progress in these specific areas has been hampered partly by thelimited availability of domestic technical skills and by the complexities inherent to theprocess of divestiture of state enterprises and of bank restructuring. The increasingdemands generated by the policy and institutional reforms already implemented, orunder preparation, have unavoidably strained the public sector's management andimplementation capacity.

Source: Quoted from Kapur and others (1991), Ghana- Adjmusun and Growth, 1983-91, IMF Occasional Paper No. 86.

40

Mix of Instruments: Appropriate in the 1980s, Evolving in the 1990s

2.46 The Bank's ESW and aid mobilization/coordination were well suited to address thepriority needs of the 1980s. The studies fed into the policy dialogue, and the Bank wasrelatively well equipped to provide advice in areas where Ghana's policies were weakest,e.g. the large distortions in prices and incentive structures. The Bank was also wellequipped, initially through of the Consultative Group and staff visits to donor capitals, andsubsequently through the Special Program of Assistance for Africa (SPA) and GlobalCoalition for Africa, to be a major force in aid mobilization and coordination. And with acomparative advantage in infrastructural investment, the Bank was also well equipped toundertake the rehabilitation projects needed in the 1980s. The modus operandi wassimple, with a relatively small and cohesive Bank team dealing with a small and cohesiveteam of technocrats on the Government side.

2.47 Thus, the Bank's instrument mix of the 1980s made a good match to the priorityobjectives. And these in turn made a good fit to the country's political economy, forreasons outlined in Box 1.2 and because the initial measures (stabilization, pricedecontrols, etc.) did not require much institutional capacity to implement. And this was amajor strength of the strategy.

2.48 During the 1990s, however, the needs have changed and so has the politicaleconomy. With democratization has come the need for the Bank to relate to moreconstituencies. With the emergence of institutional and governance constraints has comethe need for the Bank to pursue a less familiar agenda, to develop new instruments, andto mobilize new skills. And the main tasks now on Ghana's unfinished agenda ofadjustment, such as privatization, civil service reform, and public enterprise reform, dorequire substantial institutional capacities to achieve. So the Bank is now very much intransition in trying to accomplish these new tasks and to "get the instrument mix right."'

Adequacy of Risk Assessment? Uncertain

2.49 The Bank's CSP guidelines called for CSPs to "assess the sensitivity of the basecase (of macroeconomic projections) to variations in parameters subject to greatuncertainty", and also to deal with exposure and creditworthiness issues, including "countryrisk." Given that Ghana was an IDA country throughout the review period, nocreditworthiness analysis was required per se. But an in-depth risk assessment is anintegral part of any strategy. The Ghana CPPs and CSPs were relatively weak on thisfront. The two main areas of risk that most documents flagged were the possibilities thatgovernment commitment would falter and that cocoa prices would fall by more than wasassumed in the base case. In neither case, however, did the strategy papers outlinecontingency plans for dealing with such eventualities (except by cutting the lendingprogram in the case of faltering progress on reform).

20. The apt observation that there was closer "lit" in the 1980s than in the 1990s between Ghana's political economy, the bindingconstraints, and the Bank's instruments and comparative advantages is attributable to Mr. Ravi Kanbur, current Chief Economistof the Africa Region.

41

2.50 In following up on the Wapenhans Report, the Bank has sought to make morerobust its assessments of risks at the project level, and its identification of measures thatmay be taken to manage or minimize those risks. It has not however taken similar actionto ensure that the CAS includes a robust risk assessment.2 ' Since some aspects ofcountry risk and creditworthiness assessment typically call for more confidentiality thanother issues, this may not make the CAS (which has a less restricted distribution than theCPP and CSP) an appropriate vehicle for such assessment. At the same time, there is nomore fundamental matter for both senior management and the Board to discuss than howmuch of which kinds of risks the Bank should be prepared to take, and which riskmanagement strategies are most appropriate.

Feasibility of Implementation and Ownership? 1980s, High; 1990s, Uncertain

2.51 In judging the relevance/appropriateness of a Bank strategy, it is necessary toconsider how much congruence there is-or is not-between the economic managementstrategy of the government and the economic management strategy the Bank isadvocating. It is appropriate for the Bank to advocate a "firstbest strategy," where firstbest refers not to some textbook ideal but to the best set of policies that seems feasible ofimplementation, taking into account real-world constraints.

2.52 It is also appropriate that the Bank be pragmatic and make compromises with itsfirst-best solutions if second-best solutions can be found that are more "owned" by theGovernment and so have a better chance of sustained implementation. Both the IMF andBank appreciate that it is better to have a relatively "soft" program that is owned by theGovernment than a "harder" one not-so-owned. In Ghana's case, both the IMF and Banksettled for a more gradualist program of reforms than they preferred. In retrospect, thisseems to have been a correct decision. It does not however imply that Ghana's gradualismwas optimal.' Of course no one knows the counterfactual, i.e. whether a faster pacewould not have led to reversals. And one does observe that Ghana's reforms,notwithstanding the setbacks on the fiscal side since 1992, have been sustained for longerthan in any other African country.

21 The CAS guidelines call for some assessment of the past and prospective effects of changes in a country's externalenvironment, and also of the sustainability of the reform process in the case of countries in the process of economic adjusmnt.

22. Other things being equal, the more congruent the Bank'a preferred strategy for a country and the government's actualstrategy for that country, the more 'relevant' is the Bank's strategy in the enwe of ita probability of implementation. On theother hand, a Bank assistance strategy that leads the Bank to compromie so much that it supported a non-viable governmentstrategy would be congruent but not relevant. To be relevant, a Bank dategy has to support a govrnment strategy that is atleast minimally viable, however close or far that nay be from a 'first best' or best practice strtegy. If that test is not mat (amit was not in 1981), then the Bank should refrain from providing asistunce other than in the context of a 'core program.'

23. In attempting to think systematically about the 'counterfactuals', OED did underake sonm formal mnacro analysis (cf.Branson, W., (1994) 'Macro and Trade Issues', Working Paper for Ghana CAR, mimeo). In addition, it consulted such relvantpublished analyses as Roe, A. and H. Schneider, (1992) Adjussnent and Equity in Ghana, Paris: OECD, that asesed linksbetween social developmenu, investor confidence, and the sustainability of Ghana' reforms, plus the substntial liati an trGhana's political economy. This facilitated the analysis of both 'counterfactual scenarios' of the past and some 'usauinabilityscenarios' of the future. Substantial additional formal modelling work in these areas is needed to make such aalyses morrobust and more relevant to future policy-making and future Bank ategy design.

42

2.53 The downside of the large aid flows which resulted from this posture, however, hasbeen a number of "dependency effects," summarized in Box 2.2, which now constitute adanger to the sustainability of the reforms. These effects, including new forms of rent-seeking that serve to undermine governmental capacities, have tended to be too muchignored by donors.

Box 2.3: The Downside of High Aid Dependency: Dangers and Distortions

A main finding of this study is that too little attention has been given by the donorcommunity in general, and the Bank in particular, to the downside of Ghana's high aiddependency. A corollary finding is that more attention deserves to be given tosustainability indicators that presuppose declining, rather than increasing or continuing,high aid dependence.

The main dangers and distortions arising from Ghana's high aid dependence are:

* Dutch disease (see Box 5.2 in Chapter 5).

* Donor-driven agendas and priorities; accountability to donor constituencies vis avis local constituencies.

* Donor-driven budgets and processes.

* Rent-seeking and patronage relations; other "external diseconomies" and"addictive behaviors."

* The use of leverage and reverse-leverage; conditionality conflicts.

* The use of aid to facilitate avoidance rather than implementation of neededadjustments, especially those involving downsizing the public sector.

* Over-reliance on donor initiatives and actions; "letting the donor do it."

Examples of each of these distortions are provided in the report.

2.54 An important feature of the Ghanaian reform program from its beginning in 1983is that the government had a high degree of ownership of its program. There isconsiderable evidence of this, including public statements of the authorities that made itclear that they were responsible and accountable for the "bitter medicine" that had to beswallowed, and that the World Bank and IMF were their supporters, not their scapegoats.Similarly, the Ghanaian officials were known in the Bank as people who knew their ownminds and were tough negotiators, able to define the limits of which issues they were orwere not prepared to take on, and determine how far they could go. And when thegovernment delivered on most of what it had agreed to do, this performance builtconfidence in the Bank and donor community that led to the gradual buildup of assistance

43

pipelines. The downside of the "soft" and gradualist approach was that certain problemareas were left unaddressed for too long (e.g., civil service reform and privatization).24

Defining the Bank's Role vis-o-vis Other Donors: Not Much Done

2.55 Guidelines for both the CPP and CSP called for Bank strategy to be defined in theframework of a country's overall external assistance requirements and the roles of otherdonors and creditors.' In practice, the Bank's strategy papers have given little attentionto the matter of donor comparative advantages, whether at a point in time or as theymight be developed over time, even with regard to partners as close as the IFC.26

Considerable attention was given to ensuring that adequate aid could be mobilized, butmuch less attention was given to aid coordination matters such as the "downside" of thedependency effects that may be entailed when so much aid is being provided by so manydonors.

Sustainability? The Uncertain Path from Recovery to Sustainable Growth

2.56 The Ghanaian strategy was to "go for growth" above other objectives, and theBank supported this strategy. This led to the designation of the rate of GDP growth asthe dominant indicator of the success of the strategy, for both the Government and theBank. And although growth in agriculture and manufacturing was much lower thanprojected, not much attention was paid to these shortcomings because the mining andservices sectors-especially trade, government services and transport-grew enough toenable aggregate output to rise by the targeted-and seemingly satisfactory-5 percent perannum. No one wanted to "argue with 5 percent" when the rest of Africa and most of theworld was doing so poorly.

2.57 A number of NGOs raised concerns and complaints that the Government's 'jumpstart strategy" of the economy was hurting the environment by pushing logging and otherresource exploiting activities.27 Meanwhile, UNICEF, ILO, and others raised concernsthat the Government, Bank, and Fund were giving too little attention to poverty and thesocial costs of adjustment. Independently of these latter pressures, however, theGovernment launched, with Bank support, ambitious programs of educational reform and

24. This is not to say, however, that a 'tougher' posture by the Bank would have ensured the needed acceleration and deepenrngof the reforms. Ideally, the Bank would have been more effective in helping the Government to see for itself the benefits ofaccelerated reform versus the costs of delay. But this is not tantamount to saying the more leverage should have exercisdthough tougher conditionalities. Bank staff correctly recognized that government ownership was critical.

25. The CAS guidelines go further and call for 'an asseusment of the role the Bank might play relative to the plans andcapabilities of other multilateral and bilateral donors'-a formulating which invites the Bank to identify and purue itscomparative advantages.

26. Relations with the RAF were of course a special case (cf. Box 2. 1). lbero were close reations and good commnicatioebetween Bank and IFC (including FIAS) stff who commented upon draft CPPI and CPsn, but there is litle evidence of mucbattempt to achieve fully integrated BankI1FC strategies. See also Chapter 3.

27. See, for example, the document published in 1993 by Friends ofthe Earth entitled Plunder7it Ghana's Rain Fo,tu fotiUqlProfir. The OED mission cannot attest one way or the other to the validity of allegations made in such documents.

44

of restoring basic social services, including in the rural areas. For this they deserve goodmarks.

2.58 Probably the main shortcoming in the design of the Bank's assistance strategy hasbeen the lack of attention given to what may be called a broader set of "sustainabilityissues." In the CPPs and CPSs, a few passing references were made to environmentalproblems, but none of the strategy documents discussed the matter of sustainability withreference to the political or social dimensions, financial dependency dimensions, orstructural dimensions. The Ghana brief for the last two SPA meetings has contained asection on "sustainability indicators," but this frame of reference did not appear in the lastCSP or in the latest CASs. How the current strategy might focus upon the sustainabilityissues and incorporate a set of sustainability indicators will be discussed in some detail inChapter 5.

Conclusions

2.59 The above review of the "relevance criteria" does not readily give rise to a rating.It shows, however, that the Bank's strategy had some notable strengths, includingespecially initial sound diagnostics based on good-quality economic work; close alignmentbetween the Bank's country assistance strategy and the Government's strongly "owned"own strategy; timeliness; acceptably realistic assumptions about external events; especiallyfor the 1980s, an appropriate mix of instruments; and a generally appropriate sequencingand an acceptable "second-best" pacing.

2.60 Among the relative weaknesses identified were a lack of clarity in objectives andtheir priorities; too much focus on aggregate (GDP) growth as the indicator by whichsuccess of the strategy would be gauged; insufficient "strategizing" (e.g., identification ofwhich constraints are most binding at each stage, and concentrating resources on keyconstraints); a lack of robust risk assessment; delays in giving attention to implementationand institutional constraints; and most importantly, an overriding focus on near-termstabilization and adjustment issues that crowded out attentions and resources available todevote to the "sustainability issues," of which the "nexus issues" and the "dependencyissues' are important subsets. This last point will be further explained and elaboratedupon in the next chapters.

45

3. Effectiveness of the Bank's Instruments

"When we know what we want and take the initiative, things [projects,programs, etc.] go okay. But when the World Bank takes the initiative, thingsdon't work so well."

Senior Ghanaian official

Introduction

3.1 This chapter focuses on the effectiveness of the various Bank instruments, namelylending (for investment projects, adjustment operations and technical assistance), ESW,and aid mobilization/coordination.' As in other parts of this report, the coverage isselective rather than comprehensive. Emphasis is given to identifying instruments thatappear to have been most and least effective. Hence, particular attention is given to thefour projects approved since 1983 whose outcomes have been rated as unsatisfactory, andto the two projects whose outcomes were deemed by OED to have been highlysatisfactory. Attention is also given to OED's findings regarding project impacts oninstitutional development (ID) and ratings of the sustainability of project benefits.

3.2 The main sources for drawing conclusions about the efficacy (and where possiblethe cost-effectiveness) of Bank projects in Ghana are OED's Performance Audit Reportsand Project Completion Reports prepared by the Africa Regional Office and theGhanaian Government. The current OED database contains evaluations of 41 Bank/IDAprojects undertaken in Ghana since 1968, of which 16 are PARs and 25 are PCRs. Forthe period 1983-93, the database includes 19 evaluations of which 14 are PARs and 5PCRs. The aggregate commitment amount for these 19 projects was US$744 million,equivalent to 35 percent of the US$2.1 billion in commitments for all IDA projectsapproved in 1983-93.

3.3 Only one OED impact assessment has been made of a Bank project in Ghana.This was the assessment of the resettlement component of the Kpong hydroelectric

28. Sometimes "policy dialogue and "portfolio management" are treated as separate instruments. But since policy dialogue iso much interwined with ESW discussions and adjustment operation preparation and negotiation, and portfolio managementis so intertwined with project supervision, these will be treated in this chapter as subsumed in ESW and lending respectivly.There are of course particular portfolio management instruments such as the Country Strategy and Implementation Review(CSIRs) and Country Implementation Reviews (CIRs) at which Bank staff and managers meet periodically with seniorgovernment officials to take stock of project implementation and sometimes other issues. Documents relating to such review(especially the CSIRs of 1989, 1991 and 1993, known as the "Akuse Reviews" were of course reviewed by the mission and provedespecially valuable in identifying systemic problems in project implementation. Such problems are identified and dicusse Inthis chapter. But it was not possible to undertake, on the basis of the mainly hearsay and anecdotal information collected, aprocess evaluation portfolio management in Ghana, eg., an evaluation of the efricacy and cost-effectivenes of the "Akueprocese perse.

46

project. That project was implemented between 1977 and 1982, before the period coveredby this review.

Most and Least Effective Instruments

3.4 The Bank's most effective instruments have been its macro- and policy-orientedeconomic work, the macro-level policy dialogue, the structural adjustment operations,lending for infrastructure rehabilitation, and its aid mobilization efforts (includingleadership of the Consultative Group for Ghana and the Special Program of Assistancethat mobilized substantial cofinancing). The Country Economic Memoranda of 1983 anu1985 were particularly useful documents that helped the Bank, Government, and donorsto share a common set of priorities, and Bank ESW was also effective in helping theGovernment with its tax reforms and public expenditure programming.

3.5 Less satisfactory or mixed results have been associated with lending for agriculture,social sectors, finance and industry; with some sector work; and with efforts to foster aidcoordination (as distinguished from aid mobilization). The least effective instruments havebeen those (including technical assistance) aimed at fostering institutional development ingeneral and institutional reforms in the civil service and public enterprise sector inparticular. One general problem that made for lowered effectiveness of lending was thatthe Bank and other donors sometimes pushed more operations and financing on Ghanathan the limited absorptive capacity could manage effectively. For the Bank, this reflectedin part the fact that Africa had a more-or-less fixed share of total IDA resources, and thatfew other African countries were comparably aidworthy.

Overview of Bank and Total Donor Assistance

3.6 Aid flows to Ghana should be regarded largely as endogenous because they havebeen very much tied to Ghana's economic management performance. Through most ofthe 1970s and into the early 1980s, when the Bank and other major donors judgedGhana's economic management performance to be poor, aid flows were relatively low(about three percent of GDP in the years 1981-83). After Ghana adopted its ERP in1983, the aid pipeline was built upon only gradually. But since 1987 Ghana has been afavored aid recipient, and ODA commitments have averaged about US$800 million peryear between 1987 and 1993. The 1993 recorded commitment was about US$700million.9 By end 1992 the undisbursed aid pipeline exceeded US$1 billion.'

29. Al the June 1993 Consultative Group meeting, donors were reported to have "pledged" over S2 billion in assistance. SuchCG pledges, however, frequently, involve some double-counting and do not provide a good basis on which to predictdisbursements.

30. Since the ODA flows include highly concessionary aid flows only, the above figures do not include, for cxample, some ofthe IMF arrangements and other official flows that were on harder-than-ODA terms.

47

3.7 Table 3.1 shows that Ghana has become a very aid-dependent country, with netODA disbursements equivalent to about 10 percent of GDP over the five years 1988-92.Ghana has enjoyed both an "adjustment premium" (entitling it to IDA allocations above-and-beyond the IDA norm) and SPA eligibility (entitling it to large cofinancings in theIDA adjustment operations).

Table 3.1: Aid Flows, 1981-92 (ODA terms, US$ millions)

Yearly Average for Periods

1981-83 1984-86 1987-89 1990-92

Commitments, Total 186 356 737 888of which: IDA 33 137 258 280

Gross Disbursements 152 288 554 873of which: IDA 14.4 94 189 190

Net Disbursements 133 264 522 687of which: Bank/IDA 19 89 177 176of which: IDA 14 93 187 188

* as % of GDP 3 5 10 10

* per capita ($) 12 21 37 45

Net Transfers 105 238 478 635of which: IDA 12.7 90.3 181.3 178

a. Calculated from yearly averages.Source: OECD, World Bank Debt Tables.

3.8 IDA assistance to Ghana over the period 1984-92 was equivalent to 34 percent oftotal ODA commitments, 28 percent of gross ODA disbursements, and 32 percent of netODA disbursements. In recent years, IDA commitments have levelled off at aboutUS$250 annually-equivalent to about one-third of the annual average ODAcommitments. Table 3.2 shows IDA commitments and disbursements to Ghana, by majorlending instrument, for the fiscal years 1979-94.

3.9 A matter of considerable concern to the Bank has been the stagnation in the levelof IDA disbursements on investment projects at a level of about US$80 million overseveral years. This reflects a variety of problems in Ghana's implementation capacity,including its management of counterpart funds. These issues will be discussed below.

48

Table 3.2: IDA Lending by Major Lending Instrument, FY79-94 (US$ million)

Variable 1979-82 1983-86 1987-90 1991-94 Total1983-94

Major L1nding Instr.

Financial Intermediary Sum of LJC S 19 0 30 0 30Loans Sum of Disb. S 4 18 14 20 53

Sector Adjustment Loans Sum of IC S 0 257 191 16 633Sum of Disb. S 0 106 257 231 595

Structural Adjustment Sum of UC S 0 0 264 133 397Loans Sum of Disb. S 0 0 190 206 396

Specific Investment Loans Sum of UIC S 30 103 237 390 730Sum of Disb. S 80 43 135 178 357

Sector Investment & Sum of LJC S 54 75 186 318 579Maintenance Loans Sum of Disb. S 36 45 103 160 307

Technical Assistance Sum of LVC S 0 35 21 15 71Loans Sum of Disb. S 0 13 30 18 61

Grand Total Sum of UC S 103 469 929 1042 2439Sum of Disb. S 120 226 730 814 1769

Source: Task Manager's Workstation Version 3.0

3.10 The Bank built up its commitment and disbursement rates until levelling off in theearly 1990s. Sectoral adjustment lending was undertaken ahead of structural adjustmentlending, and combined sectoral and structural adjustment lending reached about US$1billion-equivalent to 42 percent of the total IDA lending of US$2.4 billion.

The Effectiveness or Lending - Performance Trends

3.11 PCR-PAR performance ratings were available for 41 completed operations inGhana as of November 1994 (representing total commitments of US$1.1 billion). Abouthalf of these completed operations were approved prior to 1983, but in terms of volumeof lending the projects approved since 1983 comprise about two-thirds of the total.Among these rated projects, 32 (or 78 percent) were judged as satisfactory by theevaluators. A significant proportion of the satisfactory projects and programs have alsobeen marked by likely sustainability albeit moderate achievements of institutionalobjectives in several instances.

3.12 The satisfactory proportion of Bank's portfolio in Ghana compares favorably withthe Bank-wide satisfactory average of 74 percent and with the Africa Region's satisfactoryaverage of 65 percent. Equally important, the 80 percent of program and policy loans(principally, adjustment operations) judged to have been satisfactory is considerably higherthan the Bank-wide average of 73 percent and the Africa Region's average of 59 percent.Two highly satisfactory operations are summarized in Box 3.1. One reason why theseSALs succeeded better than the SECALs was that they involved decisions only at the

49

macro level, whereas the SECALs required coordination at the sectoral levels where thedeficiencies in information, institutions, and management were more severe.

Box 3.1: Outstanding Projects

Ghana's first and second and second structural adjustment credits (SACs I and U) werejudged as outstanding programs by the OED evaluators.

The major factor in the successful outcome of these programs was that the government washighly committed to adjustment and implemented a wide range of measures promptly andwholeheartedly. Weaknesses of implementation capacity, owing to deficiencies in the structure andstaffing of institutions, and in management information systems in the public sector, were the majorconstraints on the adjustment process. Facing this, the government set up a Structural AdjustmentProgram Secretariat (SAPSEC) under the office of the Chairman of the Provisional NationalDefense Council (PNDC) Secretaries, to enable continuous high-level political monitoring of theprogram, and to ensure that measures calling for decision by the Council were brought before itpromptly. In addition, ten special units and task forces, coordinated by SAPSEC, were set up toimplement various parts of the program.

The pace of implementation was markedly slower where adjustment required changes ininstitutions and recruitment of skilled staff rather than mere changes in rules. In addition, thepaucity of up-to-date financial and other information for proper management decisions slowed thepace at which changes could be made, e.g., the divestiture of state enterprises. The pace was alsoslower where adequate studies had not been done and where a basis for agreement at the technicallevel was lacking.

Main strengths of these SAC operations lay in the following:

* timing-these operations were natural next steps in Bank support to Ghana's ERP, given themutual confidence built up in the post-1983 period;

* consultations-there were extensive consultations within the Bank, between the Bank andgovernment, between the Bank and Fund, and between Bank and other donors;

* complemnentarity-the SACs and rest of the Bank/IDA lending program were mutuallyreinforcing in support of the adjustment program;

* time frame-an optimistic time frame conveyed a sense of urgency at the outset, but was flexibleenough to permit more modest targets under SAC II;

* ESW-a considerable amount of high quality studies underpinned the success of the program;* ownership-the consultative approach to the design of the structural adjustment program fostered

an unambiguous sense of ownership of the program by the Govemment;* implementation-arrangements, especially for SAC I, kept the political directorate and senior

officials informed, and provided for adequate coordination and monitoring; and* supervision-Bank staff hands-on supervision facilitated work on the public investment program,

budget, tax reform, public expenditure review, and, to a much lesser extent, civil service reform.Staff devoted twice as much time to supervision of these operations as to the average AfricaRegion SAL operation.

50

Box 3.1: Outstanding Projects (cont'd)

The OED audits identified also some weaknesses in these generally successful operations. Theseincluded:

* ESW: in cocoa marketing and SOE reforms, the economic and sector work was inadequate tounderpin the Bank's efforts to convince the Government regarding the need for and direction ofchange;

* inexperience: Bank staff lacked experience in public enterprise and in cocoa marketing reforms;* narrowness of perspective: Bank insistence on specific targets for divesting state enterprises did

not appear to take into account sufficiently the policy and institutional constraints facing privateinvestment at the time; other options were inadequately considered. Similarly, the financialimplications of widespread retrenchment in a public sector dominated economy were notsufficiently considered;

* institutions: Bank study of institutions and analysis underlying institutional reform was limited.Thus, there was no definition of future roles of institutions that would lose regulatory functionsas liberalization proceeded; and

* continuity: frequent turnover of Bank staff, regarded by the client as disruptive of the adjustmentprocess, was an undesirable feature of the Bank's conduct of these operations.

Lessons

* Brainstorming sessions involving high levels of government at the design stage of the adjustmentprogram have considerable merit as a means of enhancing the client's sense of ownership andstrengthening commitment to implement it. The Ghana experience shows that the inclusion ofthe private sector, unions, and other interests in such sessions can foster greater understandingand acceptability of necessary policy measures.

* Where, as in Ghana, the public sector has been the dominant operator in the formal economy forover thirty years and the culture or tradition of private entrepreneurship has been substantiallyweakened, private entrepreneurs should be expected to respond only cautiously to partial changesin the policy environment. They need to be convinced that policy change will be permanent andcomprehensive, and that the 'official' attitude to private profit and wealth will be consistent withthe enhanced role envisaged for the private sector. Frequent consultations between thegovernment and the private sector can be useful in making the latter comfortable with the policychanges and in identifying the continuing impediments to their response.

* The adjustment process in an economy such as Ghana's requires that in addition to providingincentives to stimulate private investment, institutional arrangements for assisting the privateinvestor in mobilizing resources need to be put in place, and the legal basis and administrativeprocedures need to be clarified for potential investors as well as for bureaucrats. All theserequirements have to be addressed early in the adjustment process.

* Ghanaian officials concur with the view that the progress of the adjustment program in manyareas owed much to the 'hands on' supervision of Bank staff. This allowed Bank expertise tobuttress Ghanaian efforts in critical areas in a manner not feasible with contract assistance, andpermitted flexibility in resolving unforeseen and unforeseeable implementation problems. Normsof staff time provided for supervision of adjustment operations should be reviewed in light of thisexperience.

51

3.13 Portfolio performance in Ghana has fluctuated considerably in recent decades.For projects approved in the late 1960s until 1974, the share of evaluated projects whichreceived a rating of satisfactory remained around 88 percent. For projects approved in1974-79, portfolio performance ratings deteriorated sharply, with the percentage ofprojects rated satisfactory declining to the mid-60s range. And for projects approved inthe 1980s, performance trends show the satisfactory ratio improving again to the mid-80slevel. The differences in satisfactory ratings over time reflect a combination of:(a) changes in the external environment, particularly in the terms of trade; (b) the extentto which Ghana managed to buffer domestic economic activity from exogenous shocksthrough appropriate economic management; and (c) portfolio-specific characteristics.

3.14 There are also substantial variations across sectors. The sectors at the upper endof the range have been energy, human resources, power, transport, and water andsanitation. The sectors with below-average trends include technical assistance,telecommunications, finance, and agriculture. The performance of program and policyloans has generally been consistent with the overall country average. Power, transport,and water and sanitation sectors have maintained consistently highly satisfactory ratingswhereas, the agriculture sector performed worse during the latter half of the 1970s butshows some improvement in performance ratings in the recent period.

3.15 Shifts in the sectoral composition of the portfolio had a considerable impact onoverall portfolio performance. This was measured by comparing Ghana's portfolioperformance trends with "normalized" trends. Had the Bank continued to lend to thesame sectors after 1974 at the same proportionate levels as in the years prior to 1974,Ghana's overall performance trends would have witnessed a commensurate decline duringapproval years 1974-79 largely because of the poor performance of finance sector loans.

Projects with Unsatisfactory Outcomes

3.16 Of the 41 projects evaluated by OED through June 1994, nine were rated ashaving unsatisfactory outcomes (Table 3.3). Five of the nine projects, including the threeagricultural projects, were approved in the 1970s, and were characterized byimplementation delays and financial and other problems attributable mainly to the overalleconomic and political environment described in Chapter 1.

3.17 Of more interest and pertinence to the present situation are the two rehabilitationloans that received unsatisfactory ratings. Evaluation experience indicates that these twoprojects were premature and that their preparation was hurried. They were based oninsufficient information and were implemented when preparation of the sector workprograms was not well advanced. There was also a lack of consultation with implementingagencies at the design stage that hindered implementation at a later stage. Theimplementation of the technical assistance loan was problematic; in timber, the

52

Table 3.3: Unsatisfactory Projects

Year of Approval

Sugar Rehabilitation 1972

Livestock Development 1974

Telecommunications 1975

Upper Region Agricultural Development 1976

Second National Investment Bank (NIB) 1979

Energy 1983

Export Rehabilitation 1984

Export Rehabilitation Technical Assistance 1984

Structural Adjustment Institutional Support (SAIS) 1987

deployment of advisors was out of phase with the rehabilitation work; in cocoa, Bank staffdid not consult sufficiently with Ghanians, the ownership level was low, and there was noagreement on the technical assistance that was needed; and in mining, the Bank'sinsistence on foreign managers taking over functions of local managers proved to bedisastrous. So, while exports did improve, there were substantial problems withimplementation and the two loans were rated as unsatisfactory.

3.18 Among the recently-evaluated projects, the Energy/Petroleum InfrastructureOperation (Credit 1373) was hampered by enormous complexity of project objectivesleading to implementation delays and, eventually, unsatisfactory outcome. Finally, theSAIS project (Credit 1778) failed to meet its objectives mainly because the project did nothave either an adequate degree of local ownership or sufficient discipline in the process ofsupervision by the Bank. Other factors having an adverse effect on the outcome were thelegacy of mutual mistrust between the political leadership and the civil service, and themanagement styles of those overseeing the project.

3.19 Among the key lessons learned from this last project were that the "enablingenvironment" for institutional development in the public sector is more important than thechoice of instruments applied in technical assistance operations, and that the governance-related constraints to effective Bank assistance for institutional development arefundamental, binding constraints. The SAIS project confirmed the desirability of pursuinga 'process" rather than a "blueprint" approach to the design of such projects. But it alsopointed to the need to put discipline into the process and to ensure that appropriate staffskills are mobilized (see Box 3.2). A fundamental problem is that civil service salary andother conditions are so poor that there are few, competent counterparts to work withconsultants (local or expatriate), as trained staff leave the service in large numbers. In-service training is woefully inadequate, but the benefits of training cannot be sustained in

53

the absence of reforms that improve motivations as well as salaries of productive civilservants and public enterprise employees.

Box 3.2: Lessons from the Institutional Support (SAIS) Project

This project, whose outcome was deemed to be satisfactory by the PCR butunsatisfactory by the PAR, confirms several lessons of experience recently codified inthe Bank's Technical Assistance (TA) Handbook and provides a few others that havenot yet entered the lexicon of TA best practice. These lessons are as follows:

* The central challenge is to establish strong discipline in terms of process, in orderto ensure a constant reassessment of issues, and progress toward long-termobjectives.

* Externalities matter. Much more explicit attention needs to be given to the"environmental" aspects of TA projects (where the environment is defined as the"institutional environment").

* There are no substitutes for ownership on the Borrower's side and for in-depthcountry knowledge on the Bank side. Bank staff who appraise and design TAprojects need to know how local institutions really work and how political, culturaland personal factors must be "factored in" to project design and supervision.

* The importance of having local "champions" should not be underestimated.Participation and ownership by "steering committees" and advisory counterpartgroups are no substitutes for individual leaders/managers who know what TA theywant and know how to use it.

* Taking a long-term, strategic approach to TA implies not letting the Bank/donoragendas, work programs, and incentives "drive" project identification, design andimplementation. It also implies the needs to be patient, to help the Borrower learnby doing and by being accountable, to accept delays and some flops as costs ofkeeping accountability where it belongs, rewarding those who take the "right" risks.

* The lack of TA coordination is a problem in Ghana that needs to be recognized andaddressed by new means. The variant donor policies and practices regarding salarysupplements, for example, require rationalization as a matter of urgency. A local aidcoordination group for TA should be organized and should launch a study on how toimprove, collectively, the developmental effectiveness of TA in Ghana.

* Improve the performance indicators and focus them more on quality, process, andbehaviors. Disbursement rates may be highly misleading indicators of projectperformance and absorptive capacity. More attention needs to be paid to monitoringthe quality of TA outputs of all types.

Source: PAR, Report No. 13262, June 1994.

54

3.20 The effectiveness of the Bank's lending program as a whole has been hindered byinstitutional and managerial weaknesses on the part of the implementing ministries andagencies. In many cases, it is not easy to ascertain whether the more binding constraint toeffective implementation is a lack of commitment or a lack of institutional and managerialcapacity. But the latter are clearly serious constraints. In some cases, moreover,ownership has been lacking. The Agricultural Diversification Project was reportedly aproject pushed by the Bank on a government reluctant to accept it.

3.21 In some cases, technical assistance (TA) has helped to improve institutionalcapacities, but the record on TA effectiveness is itself very mixed and on the whole poor.How to improve the effectiveness of TA is a subject that requires more concertedattention by both the donors and government. The United Nations DevelopmentProgramme's National Technical Cooperation and Assessment Programme (NATCAP)was supposed to help in this regard but this program is ineffective and lacks donorconfidence. And TA does not much help to address systemic problems such as the poorethos of the bureaucracy and distorted incentive structures, which problems themselvestend to render TA ineffective.

3.22 Box 3.3 describes a cross section of projects in which varied progress was achievedon the ID front.

Status of the Portfolio

3.23 As of FY93, 40 projects and programs under implementation had receivedsupervision ratings (for their overall status as well as progcess toward their developmentobjectives) as reported by the Bank's Operational Policy Review (OPR) Department. Inaddition, audits of six PCRs were in progress. About 13 percent of the ongoing projectswere marked by considerable implementation difficulties. The overall status of the Bank'scurrent portfolio was then below the Bank-wide average. Problem projects then included:Mining Sector Rehabilitation, Small and Medium-Scale Enterprises, Rural Finance, CocoaRehabilitation, Forest Resources Management, Telecommunications II, PetroleumRefining and Distribution, Power V, Public Enterprise Technical Assistance, andTransport Rehabilitation.

3.24 Despite slipping over FY89-93, Ghana's supervision measures remained roughly ona par with the all-Regions averages, and compared favorably with the Africa Regionaverages. The ARPP ratings initially improved from 1.72 in FY89 to 1.59 in FY90, butthen fell to 1.92 by the end of FY93. A major cause for concern was the proportion of"problem' projects in the portfolio, which have yet to show any significant improvement.

3.25 Some of the implementation delays in the portfolio were attributable to thedeterioration of disbursement ratios. The undisbursed balance rose steadily over fiveyears, FY89-93. Some is this increase was accounted for by the depreciation of the dollaragainst the SDR, but the bulk was due to a substantial deterioration of the disbursementratio, from 33.1 percent in FY88 to 14.3 percent in FY92. This compared to a Bank-wideratio of 17 percent and the Africa Region's average of 15 percent in FY92.

55

Box 3.3: Attainment of Institutional Development Objectives

Institutional weaknesses in Ghana are frequently cited as a persistent problem in mnanagingBank's country porLfolio. To overcome these shortcomings, several operations have contained asignificant institutional development component. The record of achievement has been mixed, asindicated by the following.

Substantial Progress

* Power System Rehabilitation Project (Credit 1628, approved in 1985). The project wasconceived in part as a means to strengthening the Electricity Corporation's institutional capacityto reliably deliver electricity services at reduced costs, and one of its principal component wastechnical assistance for operational support, training, work planning, and the preparation andimplementation plan of ECG. The ECG benefitted widely as a result of the project.

* Road Rehabilitation and Maintenance Project (Credit 1601, approved in 1985). The projectwas particularly successful in carrying out many institutional arrangements on which earlierhighway projects had produced mixed results. The once overstaffed Ghana Highway Authority(GHA) reduced its manpower by about 2,000 employees.

* SACs I and II (Credits 1777 and 2005, approved in 1987 and 1989). Amongst the principalachievements of policy reform was a move toward efficiency in use of public sector resourcesthrough: (a) imorovements in public expenditure management; (b) civil service reform resultingin the retrenchment of an average of 12,000 staff per year over the period 1987-91; and (c) theselling of 15 SOEs to the private sector and the closing of 23 others.

* Railway Rehabilitation Project (Credit 1170, approved in 1981). The project had two mainobjectives: (a) to improve managerial efficiency by instituting modem methods of operations,maintenance, and financial management; (b) to provide the revitalized management withimproved equipment to enable Ghana Railway Corporation to function as a viable transportenterprise, in particular to move export traffic to Ghanaian ports.

* Volta Agricultural Development Project (Credit 1009, approved in 1980). Amongst thevarious components of the project were measures to improve extension services, credit andtraining for farners, and construction of farm service centers for the distribution of inputs.

Marginal Improvements

* RIC n (Cred.t 1573, approved in 1985). The otherwise satisfactory program yielded variedresults across and within sectors. While the reorganization of the Cocoa Board was started, thedivestiture of some activities in plantations, transport, and processing was slow. In timber, theForestry Products Inspection Bureau and the Timber Export Development Board wereestablished, but the latter remained understaffed. Similarly, no worthwhile institutionalimprovement was achieved in gold mining activities.

56

Box 3.3: Attaimnent of Institutional Development Objectives (cont'd)

* Health and Education Rehabilitation Project (Credit 1653, approved in 1986). Despite theemergency nature of the project, several long-term benefits were derived from the project: (a)work in the critical area of drug supply and extension of immunization coverage; and (b)developing middle-level leadership in the health sector and in thc Ministry of Health.

* Petroleum Refinery Rehabilitation and Technical Assistance Project (Credit 1446, approvedin 1984). The project was designed also to provide technical assistance for a managementimprovement program and for training of refinery staff. But it took considerably more time thanwas originally expected. The study on the MIP was seriously delayed, and implementation ofthe recommendations had to be carried out under a succeeding project (Credit 1819).

* Water Supply Technical Assistance and Rehabilitation Project (Credit 1342, approved in1983). The primary objective of the project-to strengthen the managerial capabilities of theGhana Water and Sewerage Corporation (GWSC) both at headquarters and in the districts and toimprove its operating efficiency and financial performance-was not as successfully achieved ashad been envisaged during project preparation.

Negligible Results

* RIC I (Credit 1393, approved in 1983). Notwithstanding the achievement of the objective ofproviding quick-disbursing foreign exchange to finance urgently needed rehabilitation imports,the program made very little progress toward building the procurement and implementationcapacity of the Ghanaian agencies.

* ERP and ERTAP (Credits 1435 and 1436, approved in 1984). The implementation of theseexport rehabilitation projects, except perhaps in the ports sector, left much to be desired. Intimber, the timing of the deployment of advisors was out of phase with physical rehabilitation.In cocoa, expertise could not be efficiently provided and effectively utilized. Theimplementation of the project in the mining sector was very poor.

* Second National Investment Bank (NITB) Project (Credit 0901, approved in 1979). Thetechnical assistance components of the project consisted of: (a) assisting the Government informulating and implementing an export and marketing program; and (b) improving the intemaloperations of Ghana's primnary development finance institution. The project had a minimal effecton improving NIB's institutional capabilities.

* Structural Adjustment Institutional Support Project (Credit 1778, approved in 1987). (SeeBox 3.2.)

57

Table 3.4: Ghana - Selected Indicators of Bank Portfolio Performance and Management

FY84 FY88 FY91 FY92 FY93

Portfolio Performance

No. of Projects under implementation 15 25 38 38 40

Average implementation period 3.3 2.9 3.3 3.5 3.5(years)'

Average Ratings

Development objectives6 1.67 1.40 1.43 1.58 1.57

Overall statusc 1.93 1.68 1.71 1.95 1.92

Percent of Projects rated 3 or 4

Development objectivesb 13.3 .0 5.3 10.5 7.5

Overall status' 26.7 4.0 7.9 18.4 12.5

Canceled during FY in US$m .01 .00 1.85 .18 .89

Disbursement ratio (%)d .5 .5 17.4 14.5 17.5

Disbursement lag (%)' .0 .0 4.0 12.9 25.9

Supervision resources (Total in 260 358 453 601 631staffweeks)

Average supervision (sws/project) 17.4 14.3 11.9 15.8 15.8

Supervision resources by location(in %)

% Headquarters 100 99 100 88 76

% Resident Mission 0 1 0 12 24

Supervision resources by ratingcategory

Projects rated I or 2 14 14 12 13 15

Projects rated 3 or 4 27 13 24 30 24

a. Average age of projects in the Bank's country portfolio.b. Extent to which thd project will meet its development objectives (see OD 13.05, Annex D2).c. Assessment of overall performsnce of the project based on the mtings given to individual aspects of projects

implementation (e.g.,m nsgement, availability of funds, compliance with legal covenants) and to development objectives(see OD 13.05, Annex D2). The overall status is not given a better rating than that given to project developmentobjectives.

d. Ratio of disbursements during the year to the undisbursed balance of the Bank'a portfolio at the beginning of the year:investment projects only.

e. For all projects compriring the Bank's country portfolio, the percentage difference between actual cumulativedisbursements and the cumulative disbursement estimates so given in the 'Original SAR/PR Forecast' or, if the loanamounts have been modified, in the 'Revised Forecast.' The country portfolio disbursement lag is effectively theweighted average of disbursement lags for projects comprising the Bank's country portfolio, where the weights used arethe respective project share in the total cumulative disbursement estimates.

58

3.26 The lack of timely availability of matching funds for the Government's contributionto projects has often become a bottleneck to Bank disbursements. The Government hasargued that the Bank's practice of requiring up-front payments into project accounts inthe relevant ministry or agency ties up resources inefficiently, and that the proliferation ofaccounts also leads to loss of control by the Ministry of Finance and Economic Planning.

3.27 Concerns have also been expressed regarding delays in contract processing andprocurement both at the level of the Ministry and the implementing agencies. The Bankhas undertaken a training seminar for project coordinators to address the issue ofimplementation delays. Especially for procurement of goods, the Ghana SupplyCommission's monopoly position adversely affects its efficiency and performanceaccountability. However, the alternative of competitive contracfing through private sectorprocurement agencies remains closed.

The Role of IFC

3.28 Notwithstanding some continuing important constraints to private investment, andespecially direct foreign investment (see Chapter 5), IFC has progressively increased itsinvolvement in Ghana in recent years, with generally satisfactory results to date.

3.29 As of June 1994, IFC's Ghanaian portfolio amounted to 20 projects for totalcommitments of US$155 million, of which US$148 million (95 percent) was related to loaninvestments and US$7 million to equity. The mining sector accounted for 76 percent ofthe total. Ghana now represents one of IFC's largest exposures in Sub-Saharan Africa.The substantial growth of IFC's activities in Ghana led to the opening of an office inAccra in May 1992. Until 1990, IFC involvement in Ghana was heavily oriented towardsthe gold sector, in keeping with the Government's strategy to rehabilitate and expand asector which represents almost one quarter of export resources. Ashanti GoldfieldsCorporation (AGC), with a total exposure of US$79 million, represents one of the highestsingle-company exposures in IFC's portfolio.

3.30 From IFC's point of view, its most satisfactory involvement has been in the goldsector, where it has brought some US$386 million to rehabilitate the sector on its ownaccount, as well as on account of participants through mobilizations. In FY92, itintroduced AGC to the derivative-products market through the approval of a US$140million gold-link loan to finance its future expansion, and a gold hedging facility to enablethe company to purchase forward-sale and option contracts from international bulliondealers. IFC also assisted the Government in its divestiture of AGC, through a successfulflotation in the international market.

3.31 While in volume terms IFC's financing has been concentrated in the mining sector,it has diversified into a range of activities in steel manufacturing, aluminum, plasticsproducts, tourism, packaging, soap, wood products, agribusiness and the financial sector.IFC has also become actively involved in the development of the small business sectorthrough its Africa Enterprise Fund (AEF) which provides loan and equity financing tosmall businesses. Ghana is by far the most active AEF country, with 17 approved projects

59

totalling about US$9.2 million. With the opening of a satellite office in Accra last year,IFC expects a continued increase in its involvement.

3.32 Several AEF projects have matured with assistance from the Africa ProjectDevelopment Facility (APDF). Ghana is currently the most active APDF country. TheFacility has processed 20 projects since its inception in 1986, for which it developedconcepts, marketing strategies and in many cases helped mobilize local funding. Twenty-two new projects are currently in the pipeline.

Allocation of Staff Resources

3.33 The following table describes trends in the Bank's deployment of staff resources, asallocated among the main instruments. As may be seen, the trend in the total amounts ofstaff time allocated for Ghana was very much in line with the trends in lending volumedescribed earlier-a rapid buildup in the mid-1980s but a levelling off (albeit still someincrease) in the early 1990s.

Table 3.5: Staffyears by Main Instrument (SY, annual average)

1 983-86 1987-90 1991-94

Lending 8.8 13.8 12.3

Supervision 4.8 8.0 12.3

ESW 6.0 7.8 7.5

Aid Coordination 0.8 0.8 0.5

Technical Assistance 1.0 1.8 1.5

Other 1.0 1.0 1.3

Totals 22.4 33.2 35.4

Economic and Sector Work (ESW)

3.34 Some of the main conclusions on the Bank's ESW on Ghana, 1983-93, are: (i) ithad to "filla vacuum" especially in the mid-1980s, and to some extent today as well; (ii) itwas of mixed although generally high quality; (iii) it had a substantial impact on theformulation of the Bank's country assistance strategy in general and on the lendingprogram (and the quality thereof) in particular; and (iv) it had a substantial impact on theGovernment as an instrument of policy dialogue and on the donors as an instrument ofaid coordination.

3.35 Some weaknesses were that: (i) it was slow to turn from short-termmacroeconomic issues to matters of poverty, environment, the "nexus,"institutional

60

development, economic governance, and other so-called long-term issues; (ii) it wassometimes too "economistic," i.e. without sufficient regard to the politica' economy orcultural context; (iii) it tended to be "inward looking" (e.g. without much reference toother work on Ghana outside the Bank) and was disseminated to only a narrow audience;(iv) it sometimes presented as prescriptions the Bank's preferred options, rather thanelucidating options for consideration by policymakers; (v) it was sometime presented inless than the most user-friendly fashion; and (vi) its availability led in some cases toneglect by the Government to do its own economic analysis. The criticismsnotwithstanding, and subject to the caveats below, the Bank's ESW on Ghana was on thewhole highly relevant, especially in the mid-1980s but also thereafter, efficacious to asatisfactory degree, and cost-effective.

3.36 The judgments above cannot be defended on the basis of a robust cost-benefitmethodology. The costs (Table 3.5) were an average of six SY p.a. in the period 1983-86and at 7.7 SY on average for the period 1987-94.31 But it is difficult to identify andmeasure quantitatively the benefits of activities such as ESW, and to gauge what impact orinfluence any particular study had, directly or indirectly, on any particular audience, or anyparticular decision.

3.37 The above assertions that the Bank's ESW on Ghana had a "substantial impact" onthe government, donors, and Bank managers are largely based on the answers of morethan 150 respondents drawn from these audiences to interviews in which questions wereposed regarding the perceived impact/influence and the strengths and shortcomings of theBank's ESW. The judgments rely not on the views of the authors as to the impact theythought (or hoped) they had, but rather of the audiences who might be expected to havesome bias towards underreporting the actual impact. Given the close linkages betweenthe Bank's ESW on Ghana and its country assistance strategy, it is probably inevitable thatthe "relevance" of the ESW is found to parallel the relevance of the assistance strategy, aswas assessed in some detail in Chapter 2.

3.38 In the early 1980s when lending was minimal, there was a hiatus in Bank economicwork on Ghana. Consequently, it took some time to fill the vacuum of countryknowledge, especially since there was a lack of flexibility in the Region in reassigning staffacross divisional and departmental lines. As one Bank lead economist observed: "Thisshows how the Bank pays a heavy price for budgeting according to lending size, as it tookus about three years to properly staff ourselves after the Government of Ghana decided tolaunch the ERP." As neither the Government nor anyone else had been producingeconomic or sectoral studies for some years, only the Bank and Fund were able andwilling to fill the void.

31. 'Men may be convened to curmnt dollar prices. As with other Bank MIS data, in the case of the considerable adjutmntlending that took place in Ghana, the convention adopted wss to count as ESW any preparatory work up to the Issuer Paperdtge and as lending the appraisal and subsequent stages of the lending cycle. In practice, especially in the late 1980a when ESWesam under some attacks as being of questionable efficacy, there became a tendency to record as lending some preparatoryactivities previously recorded as ESW.

61

3.39 Fortunately, in spite of the lag by the Bank in filling the void, the quality of whatlimited economic studies the Bank was able to produce in the mid-1980s was very high,not only in terms of economics but also in strategic content. As early as October 1983,the Bank laid out a proposed phasing and sequencing of Ghana's adjustment process thatstands well the test of "best practice" over a decade later. The same can be said of the1985 Country Economic Memorandum that provided a strong analytic basis for the well-constructed 1986 Country Program Paper, as well as for that year's Consultative GroupMeeting.

3.40 In succeeding years, the CEMs had a more varied quality, and in some years theywere largely descriptive and not so robust analytically. By and large, however, the Bank'smacroeconomic analysis (and in consequence its policy advice) was sound in advocatingtighter budget policy to reduce inflation, a more flexible and realistic exchange rate policyto restore competitiveness, movement towards free foreign exchange markets and positivereal interest rates, the removal of quotas and internal price controls, tariff levelling andreduction (though the speed of the last was probably excessive), institutional reforms inthe Ministry of Finance and Economic Planning (MFEP) (which advice was not taken),and so forth. Chapter 2 has also provided examples of where, with the benefit ofhindsight, it appears that certain advice given to the government or action taken by theBank on the basis of ESW was somewhat faulty.32

3.41 One hypothesis supported by the Ghana experience is that the quality of theBank's policy advice and in turn the likelihood that this policy advice will be heeded is afunction of the quality of the underlying ESW. Thus, where high quality ESW wasdone-as for example in the case of cocoa pricing, tax reform, and some aspects of publicexpenditure analysis-the Government did in fact pursue the recommended reforms. Thiswas also the case with the first two SACs that OED deemed to be outstanding projects,based on high-quality ESW.

3.42 In other cases where there was either a lack of Bank ESW (say, on institutionaldevelopment and environmental economics), or when the ESW was judged (both by Bankand government staff) to have been of lesser quality and persuasiveness (privatization,public enterprise reform), there was less inclination on the part of the authorities toaccept the proffered advice.33

3.43 One apparent exception to the good analysis, good advice, good results links, is ineducation where the Government accepted Bank advice, deemed to have been based uponmore or less "best practice" sectoral work, and where the sectoral work also underpinned

32. See for example paras. 2.3941 and 2.55.

33. This presentation has not discussed what criteria are appropriate to measure the quality and efficacy of ESW. One curmrtview in the Bank is that the Bank's expenditure on ESW is an inveatment in knowledge and undersandings (intellectual capital)rather than an opersting coat of turning out products called E'SW reponu. The "real' products are seen to be the improvedinternalized knowledge of the staff involved, and the associated institutional memory. Some benefits accrued to stff,government officials and Board memben who read the reports, but these benefits are seen as basically externalities. It is theimproved competence and knowledge of staff that are the key payoff to the ESW studies. Another view defines the quality ofESW as virtually identical tm its impact.

62

the Bank's two education sector SECALs. Given the poor outcomes in education testingto date, one may have to revisit a judgment of the quality of the underlying ESW, forreasons described in Chapter 4.

3.44 Similarly in agriculture, there are strongly variant views within the Bank as to thetechnical quality of the sectoral work underlying the medium term agricultural strategy,which provided a basis for numerous agricultural operations funded by the Bank and otherdonors. That the medium-term strategy, which was prepared jointly by the Bank andgovernment, served to raise the government's ownership of those projects is notquestioned. The fact that agricultural projects as prepared after the Medium-termAgricultural Development Strategy (MTADS) appear to have better supervision ratingsthan pre-MTADS projects tends to validate the importance of ownership. But this reportcannot shed any light on possible tradeoffs between technical quality and greaterownership, if in fact that happened.

3.45 As in other low-income countries, an important area in which Bank ESW of the1980s "missed the boat" was in expecting that getting the balances and prices right wouldproduce rapid supply responses through increased private savings and investment. InChapter 5, other examples are provided of where the Bank's projections, based on itsESW, were reasonably accurate or otherwise (and for the right or wrong reasons). AnAnnex to that chapter provides a set of charts comparing actual and projected values ofselected macroeconomic variables.

3.46 The Bank's work on its Public Expenditure Reviews has in general been of goodquality and helped to rationalize the programming and budgeting of public expenditure.But one problem that the PERs have not yet served to solve is that of poor expenditure(cash) management. This problem is adversely impacting the rate of projectimplementation across the board. A recent PER has focussed on these problems and theGovernment is reportedly preparing a strategy and action plan to address them. The Bankshould give high priority to following up on this matter to ensure that necessary correctiveactions are taken without delay.

3.47 One of the most notable, controversial, and expensive pieces of ESW on Ghana tobe produced by the Bank (together with a team of Ghanaians) was the report entitledGhana: 2000 and Beyond. This study was launched following the CG meeting of 1991 inresponse to a perception of "adjustment fatigue" and a slowing of the reform momentum.Its objectives were to raise the sights of Ghanaians and donors alike as to the possibility ofaccelerating growth to well above the five percent annual growth rate achieved since 1983.A related aim was to lengthen horizons and to shift the focus of attentions from near-termto longer-term issues including human resource development and capacity building. Thereport also constituted a poverty assessment. Well over 300 SY of resources were devotedto this task alone, making it an extremely expensive piece of ESW by Bank standards.

3.48 The report got a wide readership, and prompted diverse reactions amongst donors,Bank staff, and Ghanaians inside and outside of the Government. Some were of the viewthat it was a "marvelous, readable report, for all the criticisms about its being over-

63

ambitious," and that it did indeed succeed in raising sights and understandings aboutGhana's longer-term constraints and prospects.

3.49 Others took a more critical view, feeling that "it gave exactly the wrong message"and was not credible in targeting a dramatic acceleration in Ghana's growth rate when, atthe very time of its release in draft (the latter part of 1992), the Government's reformprogram had come off track. According to this more critical school, the report was akin toadvocating a "great leap forward" just when the real issue was how to prevent backslidingand even sustain the five percent growth previously achieved. The diverging views cutacross all audiences, including some Bank staff who felt that the Bank team took tooacademic an approach, and had too little experience of Ghanaian realities.

3.50 Both views are warranted. The main problem with the report was its timing. Itsmessages would have been far more appropriate and timely in the absence of the 1992backsliding. The report can be criticized for its lack of a full poverty strategy and for theshort shrift it gives to issues such as environmental degradation and land tenure. But thisreport seems likely to have, and to deserve, a substantiai shelf life. It served a needed anduseful function by raising sights, lengthening horizons, putting long-term issues on thetable, and getting some Ghanaians to become more outward looking, including throughtaking heed of lessons of experience from Asia that could usefully be applied in Ghana.

3.51 One lesson of experience of Ghana: 2000 and Beyond is that s-ch high profile,high-expense ESW/ undertakings are risky and likely to be controversial. But the risk ofraising such controversies and debates, within and among the various audiences, may wellbe worth taking if they serve to shift attentions of policy- and opinion-makers away from"merely urgent" towards "truly important" issues. Ghana 2000 and Beyond was surelywarranted, and the Bank should not hesitate to raise its own sights and undertake suchambitious (albeit expensive) undertakings when circumstances again warrant.

3.52 Other issues concerning the management and processing of ESW include the feltbut unmet need on the part of many Ghanaians, within as well as outside theGovernment, to have access to more Bank documents such as Ghana 2000 and Beyondand other Bank reports that provide otherwise unavailable information and analyses. Thisconsideration, plus the fact that a democratized Ghana will call for an increasingly well-informed electorate if economic policy is to be enlightened, suggests that increased priorityis warranted for the dissemination of Bank ESW documents.

3.53 Finally, in view of the need for economic policy-makers in Ghana to giveincreasing attention to the "sustainability issues," priority should be given to developingnew analytic frameworks that integrate, for example, environmental and economic andinstitutional dimensions of development. Recent work by Ramon Lopez on modelinginterrelationships in Ghana among agricultural production, environmental, and trade policyvariables is just the kind of innovative ESW the Bank should be encouraging.

64

Aid Mobilization and Coordination

3.54 The Bank played a leading and highly effective role in helping the Governmentmobilize large amounts of external assistance since 1983. It has also played a leading rolein fostering communication and coordination amongst the donors, with some success. Theevidence suggests that such aid mobilization/coordination efforts as have been undertakenby the Bank have been highly relevant, fairly efficacious (in terms of meeting the limitedcoordination objectives set), and on the whole cost-effective (even if the Bank's MIS datathat indicate that less than one Bank staffyear of resources per year has been devoted toaid coordination is a clear underestimation).34 In no small part because of these efforts,aid coordination is deemed to have been more successful in Ghana than in most otherdeveloping countries, and is not now generally perceived to be a serious problem inGhana.35

3.55 This report, however, concludes that Ghana does have an aid coordinationproblem. This is not a problem of the Bank's making, nor one for which it has theprincipal responsibility for solving. The basic problem is that high volumes of aid and largenumbers of donors and donor-financed projects give rise to a need for many kinds ofcoordination. These range from getting consistency in basic national and sectoralobjectives to getting donors to pursue their own comparative advantages (e.g. in terms onconcentrating on types of projects they do best) to harmonizing disbursement procedures.Progress has been made on all these and other fronts, but not to the extent needed. Amain source of the problem has been the limited role taken by the Government in aidcoordination.'

3.56 The overall role of aid mobilization and coordination in the Bank's countryassistance strategy was discussed in Chapter 2. As was indicated, the Bank correctlyrecognized, beginning in 1983, that more resources would be required to support Ghana'sreform program than the Bank and IMF could provide alone. But because most donorsthen had strained (if not broken) relations with Ghana, and were lacking in confidencethat the initial reforms would be sustained, Bank staff engaged in an active program to

34. Considering the time spent by Headquarters staff on consultations with the IMF on preparing semi-annual SPA meetingsand annual CG meetings, visiting donor capitals to discuss sectoral and project issues, and the frequent interaction of ResidentMission stff with their counterparts in other donor agencies, it appears that aid coordinstion has utilized considerably morethan the very modest amounu shown in Table 3.5. The judgment that the Bank's aid mobilization/coordination efforts wereprobably cost-effective is based on a counterfactual scenario in which the Bank did not underuke those aidnobilizAtion/coordination efforu. In this framework, even a small acceleration of government policy actions and donorcommitments that might be attributable to the Bank's aid mobilization/coordination activities could justify the marginalexpenditure of even several manyears of staff time.

35. This finding is based upon extensive structured interviews with donor representatives (mainly in Accra and in a few donorcapitals). To one of the least structured opening questions ('1s there an aid coordination problem in Ghana, and if so what isits nature and how serious is it?'), most respondents answered no, even when they believed that excessive aid and excessivenumbers of projects (in relation to the country's limited absorptive capacity) were indeed fairly serious problems.

36. The principle that the main responsibility for aid coordination must lie with the recipient Government has been an agreedDAC principle for nearly a decade. See Aid for Improved Development Policies and Programs and Implications for AidCoordinan on, DAC, Paris, 1986.

65

persuade other donors to extend aid to Ghana, including through numerous visits to donorcapitals.

3.57 As was indicated in Chapter 2, the Bank's aid mobilization objectives with regardto the composition as well as the level of aid sought from other donors were highlyrelevant, and there was a high degree of complementarity between the Bank's ESW andaid coordination instruments. The Bank and IMF studies underpinned the Bank's aidmobilization/coordination required, and gave credibility to the analysis of aid requirements.

3.58 The Consultative Group meetings, resumed in 1983 and chaired by the Bank,became the main formal instrument of aid mobilization, though not of coordination. Infact, however, the mobilization efforts took place in advance of the CG meetings and themain focus of the meetings, at least until recently, was on communication (exchange ofinformation and views) rather than upon coordination (moving towardsintegrated/harmonizing donor actions or policies).

3.59 The CGs have served the main function of being forums that (a) served a catalyticfunction by accelerating some government policy actions (including preparation of budgetsand annual Public Investment Programs); (b) allowed the government to "make its case"regarding its aidworthiness and to answer questions about its policies and how it intendedto deal with its main problems; (c) facilitated exchanges of information amongst donorsabout their ongoing and planned activities in Ghana; (e) provided an opportunity forstocktaking on both sides; and (f) strengthened the position of the reform elements in theGovernment by demonstrating the international support for such reforms, and so probablyhad some role in helping to keep the reforms on track.

3.60 A few donor representatives expressed more critical views of CGs as "set pieceswvth donors and government alike overstating and bragging about their accomplishments"and with so-called "pledges" (or indicative commitments) being "fudged if not farcical."But even these critics did not dispute that the above-mentioned benefits were to someextent realized and all seemed to agree that recent CGs are characterized less by suchformal recitals of accomplishments and more by more constructive and substantivediscussions of issues, generally at the macro level.

3.61 In addition to the CG, the Bank organized a social sector donors' meeting inVienna in 1986 and a PAMSCAD donors' meeting in Geneva in 1988. The bi-annualSPA meetings th.at have taken place for the past several years have served mainly asinstruments for aid mobilization for adjusting African countries, including Ghana,characterized by heavy external indebtedness. In recent years, the SPA has becomeinvolved in coordinating activities including the design of a standard audit for theadjustment operations of all donors, harmonization of certain disbursement procedures,untying of some forms of aid, etc. The SPA has also organized a number of workinggroups to address issues such as public expenditure reviews, poverty, civil service reform,and governance. While highly relevant, the benefits of such activities are hard to quantify,

66

and the Bank does not have line-item budget or MIS entries for "SPA inputs" or "GCAinputs "" against which to measure the costs.

3.62 Whereas the annual formal CG meetings were important as catalysts forgovernment actions and for the mobilization of aid, the true coordination of donor policiesand programs has come about more through periodic (generally ad hoc) sectoral CGs(often following major Bank sector reports) and through local aid groups meeting inAccra. The extent and nature of the Bank's participation in these activities seems to be afunction not so much of Bank policy or country strategy as of the approach followed byindividual task managers and Resident Mission staff. When staff are inclined to be pro-active in aid coordination, then actions may be taken. But there is little compulsion,within the Bank or other donor agencies, to require certain types of coordination becausethe country assistance strategy requires it. At the macro level, the World Bank ResidentRepresentative has hosted monthly meetings geared mainly to brief other donors on thestatus of the Government's macro policies and indicators.3 " These meetings are howevermore geared to improving communications than coordination.

3.63 At the sectoral level, monthly health meetings among donors (begun in 1986) havebeen followed by monthly donor meetings on education since 1990. There is currently amonthly lunch of donors involved in the health sector and other local donor groups forprimary and tertiary education, water supply, and other forms of infrastructure. Leadagencies of such groups tend to extol their benefits, but other donors are moreskeptical.9 And owing to their ad hoc nature, their sustainability is questionable. Mostdonors appear to want to make these sectoral groups more effective, but specificproposals for how to do so seem to be lacking.

3.64 The two main constraints to more effective coordination at the sectoral levelappear to be (i) the failure of the Government to take the lead in aid coordination; and(ii) the lack of clearly defined sectoral strategies and project priorities consistent withmedium-term macro and expenditure frameworks. A major exception was the medium-term agricultural strategy, seen by some donors as a prototype warranting emulation inother sectors. The essence of the problem is best characterized by a senior official whoobserved that: "We have many needs, and if a donors comes to help meet these needs, we

37. The GCA refers to the Global Coalition for Africa, a political-level forum for bringing together top-level representativesof African governments, donors and other leaders to focus attention upon and mobilize support for actions to address some ofAfrica's najor problems, e.g.the 'nexus,'low savings, and the defense burden. There are surely some benefits to Ghana fromthe OCA activitiea, but it is not possible to quantify the costs, much less the benerits, attributable to Ghana per se.

38. The RIbF, in contraat, consider that it is for the government to decide how much it wishes to communicate to donors aboutthe macro situation. The Bank tries to steer the fine line between being helpful and communicative to other donors withoutcommunicating more than the government may wish to communicate.

39. Thia skepticiam seems to be particularly pronounced in the case of NATCAP, the UNDP-supported program for reviewing,prioritizing and programming technical cooperation from all donors. This seems to be a case in which an effort at donorcoordination mnay be counterproductive, since several important donors seem not to take seriously either its inputs (requestsfor data, anawers to questionnaires) or outputs (data, reports). This illustrates that more aid coordination is by no mcans alwaysa good thing

67

take what the donor offers. Whether this adds up to a cohesive program or not is anothermatter. "

3.65 A major issue, strategic issue, which is also a matter of aid coordination, is thedetermination of Cte appropriate balances among quick-disbursing aid, project assistance,and various types of sectoral aid, e.g. time slices of sectoral expenditure programs.Neither the Bank's own strategy documents (CSPs, CASs) nor its documents prepared foraid coordination meetings (SPA, CG) have been as explicit or as transparent as would bedesirable in setting forth the assumptions and criteria underlying projections of needs forquick-disbursing vis a vis other forms of aid over the coming years.

3.66 In many countries experiencing overstretched absorptive capacities such asGhana's, it has proven cost-etfective to undertake major project pruning exercises. Suchexercises can serve to reduce the number of projects under implementation (pending animprovement in absorptive capacity) in order to ensure that implementation of the higherpriority projects is not impeded by too much robbing of Peter to pay Paul, or by too few"critical masses" of implementation capacity. Given the consensus about howoverstretched is Ghana's absorptive capacity, it seems surprising that such an exercise hasnot been mounted.

3.67 That it has not implies that the Bank is satisfied that Public ExpenditureReviews-which may now be considered the most important instrument of aidcoordination-are fulfilling their function of striking the right balance between resourceavailabilities and resource utilizations.' It is entirely appropriate that the PERs be suchinstruments, as they do serve to focus the attention of governments and donors alike onthe overarching issue, which is not just to maximize the return on IDA resources but tomake the totality of resources, Ghanaian and aid financed, as productive as possible. Andwhile other donors note that there have been improvements in the quality andtransparency of the PER, and more donor participation/consultation in the PER process,they clearly look for further movements in these directions.

3.68 In conclusion, it would seem that Ghana's aid coordination problems should beaddressed mainly by Government actions to clarify its national and sectoral strategies, toprioritize its projects and programs accordingly, and to take the lead in managing aid anddirecting donors to their areas of comparative advantage.

3.69 On the Pank's part, however, there is scope for the country assistance strategy todefine more clearly how the Bank sees its own comparative advantages evolving vis a visother major donors over time, and what possibilities it sees to minimize donor competitionand to improve cooperation in cost-effective ways. A corollary would be to identifypossible roles the Bank might play to assist such partners as the African Development

40. It is an arbitrary matter of classification that the PER is defined as an instrument of ESW rather than of aid coordination,or lending, or policy dialogue. But given how important the PER is (or should be, even if it isn't) to other donors as aninstrument for both aid mobilization and coordination, and at both the macro and sectoral levels, it mnight well be classified asan aid coordination instrument to emphasize this linkage.

68

Bank and major NGOs to deliver their assistance to Ghana more effectively, whetheralone or in partnership (co-financing, co-studying, co-evaluating, etc.) with the Bank.

69

4. Agriculture and Education: Objectives and Performance41

"We considered agriculture as cocoa for too long; this was a blind spot... The nexusissues are now critical"

Bank Division Chief

"The educational testing results are again dreadful, although there may be amismatch between what is being tested and what is being leamed. In any event, thetest results are provoking a lot of soul searching about what is to be done [aboutimproving educational performance]"

Bilateral Donor Official

4.1 This chapter reviews briefly the objectives of the Bank's country assistance strategyin agriculture and education, and performance against those objectives. Agriculture wasselected because of its large and critical role in the economy. The sustainability ofGhana's development will depend heavily on the extent to wnich agricultural productivitycan be improved. And because most of the poor in Ghana are rural people who remaindependent upon agriculture for their livelihood, Ghana's progress in poverty alleviation isclosely linked to progress in agriculture.

4.2 A key conclusion of this report is that the "nexus" issues (i.e. the inter-relatedissues of declining agricultural productivity, high fertility and population growth rates, andeilvironmental deterioration) are critical. And whether the nexus issues are adequatelydealt with or not will have a large bearing on whether Ghana's growth can be sustained.

4.3 Education was selected because of the importance of human resource developmentto Ghana's longer-term prospects, and also because the Bank has played a substantial rolein supporting a comprehensive education reform, launched in 1986, through two educationsector adjustment credits and five investment credits.

4.4 In both sectors, important progress has been made, yet the outcomes as measuredby trends in output and productivity in agriculture and by test scores in the case ofeducation have been unsatisfactory to date.

41. This chapter is based in large part upon two background papers. See Meerman, J. (1994) "Agriculture", Background Paperfor Ghana CAR and Ridker, R. (1994) "Education Reform Program: Bank Inputs and Progress to Date", Background Paperfor Ghana CAR. With regard to the "nexus issue" in Africa in general, see Cleaver, K. and G. Schreiber (1994), Reversing dheSpiral: 7he Population,Agric 4iure and Environment Ns in Sub-SaharanAfrica, World Bank Directions in Development Series.

70

Agricultural Objectives and Performance

4.5 The Bank's strategy in the 1980s gave high priority to agriculture in principle, butthe Bank was slow in practice to build up its pipeline of sectoral studies and lendingoperations. The Bank did play a useful role (through policy dialogue and projectconditionalities) in pushing for pricing reforms, but the pace of reform in reducing therole of inefficient agricultural parastatals continues to be slow, and the rate of growth ofagriculture is unsatisfactory. A potential crisis looms unless the "nexus issues" areaddressed more vigorously by both the Government and the Bank.

4.6 Before 1983, Ghana's approach to development was reflected in the agriculturalsector by input subsidies, heavy output-taxes, administered prices, marketing by publicmonopolies, and processing by public enterprises for rubber, palm oil, rice, and othercrops. In general, farm prices were too low and processing too costly. Macroeconomicpolicies were also anti-agriculture insofar as the currency was overvalued and industrialprotection excessive. In the several years prior to 1983, cocoa growers received less thana fifth of the value of their output (at a realistic exchange rate).

4.7 The results included a catastrophic deterioration of infrastructure and institutions,the gradual replacement of official commerce by parallel marketing, and a retreat byfarmers into subsistence production. Official cocoa exports fell by more than half, whileother official agricultural exports nearly disappeared. During the 1970s agriculturalproduction grew less rapidly than population. The only exception was rice, which becamea flourishing minor crop that enjoyed good prices and privileged access to highlysubsidized inputs, including fertilizer and tractor-services.

4.8 At the onset of the reforms, Ghana's agriculture, including forestry, accounted forabout half of GDP, 60 percent of employment and nearly 80 percent of total merchandiseexports (cocoa 73 percent, forestry products 4 percent). Cocoa covered half the totalcultivated area, with the main cereal, maize, being grown in all regions. Most productionwas (and is) by small holders using traditional techniques (slash and burn); mechanizationand irrigation are still unimportant.

4.9 The decline in Ghana's cocoa production from 400,00 tons per year (tpa) in theearly 1970s to only 158,000 tpa by 1983/84 was largely due to the decline in real producerprices; other factors included swollen shoot virus disease, drought, bush fires, aging oftrees, ineffective extension services, irregular input supply, and costly marketing services.Industrial crops were also held back by low producer prices and inefficient parastatals.Public expenditure in agriculture was relatively low and concentrated on provision offertilizers and on high-cost irrigation investments. Technological packages were availablefor several important crops but extension services were weak and ineffective.

4.10 Between 1969 and 1980 the Bank approved eight agricultural sector projectsgeared to helping Ghana to achieve greater self-sufficiency in food and raw materials foragro-industry, and to rehabilitate cocoa. The projects (for cocoa, integrated ruraldevelopment, oil palm, sugar, livestock and fisheries) were relatively unsuccessful exceptfor oil palm, mainly because of the overall country economic difficulties. Bank staff

71

involvement in these projects served to maintain some stock of sectoral knowledge in theBank. But owing to the hiatus in sectoral work, the Bank's sectoral strategy formulation,project development work was handicapped pending completion of an agricultural sectorreview in 1985.

Bank Strategy for Agriculture in the 1980s

4.11 The Bank's strategy in the early 1980s recognized that agriculture was the keysector for short-term recovery and medium- to long-term sustained growth. Withoutgrowth in agriculture, foreign exchange shortages would persist, periodic food shortageswould reappear, the unemployment problem could become unmanageable, and industrialrehabilitation would be hampered. The broad objectives of the strategy were to assist theGovernment's efforts:

* to provide adequate incentives through appropriate price and tradepolicies, with prices for a number of important crops needing to beincreased in real terms.

- to provide inputs, tools and spare parts, and essential rehabilitation ofmachinery and equipment.

* to undertake investment and/or institutional restructuring for export/importsubstitution crops-particularly cotton, tobacco, and groundnuts.

- to shift crop production to the private sector; to implement a food securitysystem for maize.

* to reduce the number of agriculture-related public corporations andeliminate subsidies to these entities.

This required promotion of private sector involvement, reduction in direct governmentintervention (including fertilizer distribution), improved cost recovery on irrigationschemes, streamlining of extension, and a coherent research policy. In retrospect, theseappear to have been appropriate objectives and priorities.

4.12 It took some time, however, for the Bank to develop the necessary ESW andlending instruments. Some agricultural conditionalities were attached to the initialreconstruction credits, e.g., increased producer prices for cocoa, cotton and tobacco in1984-85. But between 1980 and mid-1987 only one Bank project (Oil Palm II) wasapproved. This was a period of "catching up' in sectoral work, and during this period theBank carried out a number of studies on cocoa, irrigation, forestry, and selected crops. Inretrospect, the Bank seems to have given to agriculture in practice less priority than givenit in the strategy statements of the 1980s.

4.13 On the lending side, the SAC I (1987) contained important conditionalities oncocoa pricing while the Agricultural Services Rehabilitation Project (also 1987) tackled thereorganization and strengthening of the Ministry of Agriculture (especially its policy and

72

planning functions), pricing and trade policy, the detailed investment program,rationalization of production parastatals, privatization of input supply (initiated underRIC I), and rehabilitation of cotton. Cocoa subsector problems were addressed to someextent in a FY88 Cocoa Rehabilitation Project, but it was not until the 1992 AgriculturalSector Adjustment Credit that the domestic purchasing monopoly of the Cocoa MarketBoard (COCOBOD) was finally broken. Through the 1980s the focus was instead ontrying to improve COCOBOD's efficiency through downsizing and restructuring.

Policy and Output

4.14 Agricultural GDP estimates (by the Bank) suggest growth of less than two percentper year since 1984, far below the rate of growth of population, and well below the ratestargeted in the Bank strategy papers. If agricultural growth has indeed been this poor,then a key component of the strategy has had a highly unsatisfactory outcome. The dataon agricultural growth are however contradictory, with the FAO estimating thatagriculture production (as distinguished from agricultural value added) grew nearly twiceas fast. It seems hard to reconcile these data.

4.15 Food prices would be expected to increase under a scenario of low supply growth(two percent) when gross national disposable income was rising by over five percentannually and when there is a very high income elasticity of demand for food stables (closeto unity). But real food prices in fact declined since 1984 (see Table 4.1),42 an outcomewhich supports the conclusion that agricultural growth was more than two percentannually. Moreover, the divergence could be due to some combination of three factors:greatly reduced costs in the marketing chain from producer to consumer, increasedimports, or underestimated production. Since there were only minor changes in marketingorganizations and instruments over the period, reduced marketing costs are not theexplanation. Nor do increased imports appear to have been an important factor.'3 Thelast factor, underestimated output, would be consistent with FAO production series basedon official government data. Since agricultural GDP and production were highlycorrelated over the periods in question," it is possible that the Bank has underestimatedthe growth of agricultural GDP. Some Bank staff privately estimate the growth rate to beabout three percent annually-that is, near to the population growth rate. But thatgrowth rate must also be regarded as unsatisfactory. In any event, given the apparent datadiscrepancy, the importance of agricultural production in the Ghanaian economy-abouthalf of GDP-and the influence of perceptions about agricultural growth on strategy and

42. This price decline is consistent with the increase over time in the average purchasing power of the minimum daily wage inkilograms of maize: from 1.02 kilos on average for the three Junes of 1980-82 to 1.86 kilos for the Junes of 1985-90. SeeHarold Alderman, Incomes and Food Securiny in Ghana, Working Paper no. 26, (Cornell Food and Nutrition Policy Program,May 1992), Table 1, page 5. The cross-elasticity of demand of maize with other staple foods is high.

43. For example, grain imports in 1991 at 340 thousand tons (compared to 250 thousand tons in 1980) had a value equivalentto less than 3 percent of aggregate value added in agriculture. World Df'lopmeu Repon 1993, Table 4.

44. FAO production indices are price-weighted and all intermediate inputs of agricultural origin are deducted. Use of inputs(chemicals, fuel, tools, etc.) is low in Ghana so that production may be very close to value added in agriculture. Pertinent hereis the lack of any great increase in input use since 1983.

73

programming, an effort at reconciliation is needed. Table 4.1 shows some key indicatorsof agricultural outcomes over the decade.

Table 4.1: Selected Agricultural Indicators

1982 1986 1990 1992

Cocoa Data

Production ('000 t.) 177 218 261 262

Producer Price ('000 cedis per m.t.) 12 86 224 258

Receipts to Farmer (%) 45 41 56 43

Duties Paid (%) n.a. 58 27 25

Value of Nontraditional Agricultural 6 17 29 22Exports ($ m)

Annual Growth RatesAgricultural GDP (World Bank)

1983-1991: 2.9%1984-1992: 1.7

Agricultural Production (FAO)1983-1991: 7.61984-1992: 3.1

Sourwces FAO Statistical Yearbooks, World Bank Macroeconomic Model, COCOBOD, Ghana Export Promotion Board.

4.16 Since 1983, much of agriculture has been liberalized, but by no means all. Foodcrops and inputs, now largely untaxed and unsubsidized, are traded on open markets.Many of the marketing and production parastatals have disappeared or been made private.Yet at end-1993, parastatals still played a dominant role in tree-crop marketing andprocessing (rubber, cotton, palm-oil) and the Ghana Cocoa Board (COCOBOD) stillmonopolized export marketing. It still remained responsible for cocoa research, extension,seed-production, and quality control, and the producer price of cocoa was still set bygovernment.

Bank Performance and Recommendations

4.17 Several IDA investment projects in agriculture approved in the late 1980s becameproblem projects, mainly owing to weak government ownership, excessive Bank zeal intheir promotion, a failure to work with beneficiaries during project preparation, and weakcommitment to achieving project objectives by those charged with their implementation.With the important exceptions noted below, during the decade 1983-93 the Bank correctlyemphasized the most pressing issues in agriculture-those involving basic structural reformof prices, marketing and enterprise ownership.

74

4.18 Agriculture received insufficient attention in the evolving Bank strategy for Ghanafor several reasons. One was the difficulty the Country Operations and Agriculturaldivisions had in working together, with the COD having lead responsibility for cocoasector issues. Another reason was that when Nigeria launched its own adjustmentprogram in 1986, this created a large demand for staff in AF4AG to work on Nigeria,resulting in a "crowding out" of sectoral work on Ghana for some years.

4.19 The slow pace in eliminating the parastatal monopoly in cocoa marketing anddeveloping a more efficient, competitive domestic and export market was a majorshortcoming. In 1994, eleven years after the initiation of structural adjustment, Ghanawas beginning to open local cocoa marketing to private operators and cocoa producerswere still getting less than half the value of their production.45

4.20 In 1991, a Medium-Term Agricultural Development Strategy (MTADS), wasprepared with unusually intense collaboration between Bank and Government officials incarrying out upstream analytical work. On the whole, the MTADS paid big dividends insubsequent government ownership, in improved project design, and in improvedimplementation of Bank-financed and other projects, even though it has weaknesses,notably the lack of integration of cocoa sector issucs into the analysis. But aftersubstantial analytical work had been carried out, the Bank seemed to lose interest in food-security issues, notwithstanding that poverty is mainly a rural problem and a sizeableproportion of the population is seriously malnourished. The MTADS work could still beused to help develop interventions to enhance food-security and thereby help to alleviatepoverty. More attention is warranted in particular for roots and tubers vis-a-vis rice, onwhich large sums have been expended for low-return irrigation.

4.21 In recent years the Bank has taken useful initiatives to deal directly with privatesector agents, as in the 1993 Agricultural Sector Investment Project and in the areas ofagricultural exports and rural credit. But there remains a need for more in-depthanalytical work to identify specific constraints to increase private investment in agricultureand to propose specific remedies. Particular emphasis needs to be given in this work tofostering investment in nontraditional agricultural exports.

4.22 Following the Bank's suggestion, Ghana created an interministerial AgriculturalPolicy Coordinating Committee (APCC). Chaired by the Deputy Minister of Finance, ithas dealt adequately with difficult policy and programming issues that cut across ministries,including design and implementation of the Agricultural Sector Adjustment Credit.Competent authority to deal with agricultural policy issues has been a problem in somecountries. The APCC approach may be worth considering elsewhere.

4.23 In recent years, the Bank has not worked much with most other donors inagriculture, notwithstanding the important contribution of other donors to Ghana'sdevelopment. Presumably the payoff from closer interaction could be considerable. One

45. This is considerably below whal is received by producers in Cote d'lvoire and other competitor countries in the world cocoamarket.

75

promising area: increased harmonization of donor/Government practices with respect tothe pricing and distribution of commodity assistance, for example of food and agriculturalchemicals.

Natural Resource Conservation: An Impending Crisis?

4.24 Ghana's land frontier has been receding rapidly in the last three decades, andrising population density has led to a shortening of the fallow period and a reduction insoil fertility. Increasingly, agricultural production is obtained at the expense of the soil,the forests, and the wetlands. The long-term sustainability of farming is, therefore,becoming an important issue for policymakers, and it will become more so if there is asignificant increase in the agricultural growth rate. Soil fertility loss, soil erosion, anddeforestation have been estimated to cost as much as four percent of GDP per year.'The report Ghana 2000 and Beyond (pages 31-32) notes that "though land does notappear to be a critical constraint in the immediate future, there is still cause for concern."This concern would seem to warrant much higher priority than it has so far received inboth the Government's and Bank's strategies.

Educational Objectives and Performance

4.25 Ghana once had one of the most developed educational sectors in Africa and alarge stock of well-qualified manpower. But in 1983, it faced a crisis of educationaldecline in quantity, quality and infrastructure. Primary enrollment and adult literacy hadfallen back to the average for low-income Africa, while the proportion of Governmentexpenditures going for education declined to only one percent in 1983 and 1984. A massexodus of trained teachers resulted in their replacement by unqualified teachers (44percent of primary and 33 percent of secondary teachers were still untrained in 1985),instructional materials and supplies were severely lacking; and the planning andmanagement of education was weakened by emigration of specialists and lack of transportto collect statistics, inspect schools, etc. Up until 1986, the Bank had never financededucation in Ghana-a rare exception in Africa-and its knowledge of the sector waslimited.

4.26 Against this background of deterioration, Ghana launched a comprehensiveeducation reform program in 1986. Its goals were to replace the previous English-styleacademic education with a more vocationally and practically oriented curriculum, reducethe number of years required to complete the primary and secondary cycles, increaseaccess to basic education for the rural masses, improve quality, and accomplish all this in afinancially sustainable manner. This was indeed an important way in which Ghana soughtto achieve "adjustment with a human face." The program has been heavily supported bythe World Bank and other donors.

46. The problem should not seem to be much less serious if it turns out that agricultural growth is closer to the populationgrowth rate than Bank data indicate. As was seen above, the Bank seriously overestimated the agricultural/growth rate in itsearlier projections, and in the face of declining productivity owing to environmental deterioration, the nexus problem may getworse whether sectoral growth is now 2 percent or 3 percent.

76

4.27 The objectives and the design of this program were exemplary, and the reformshave been advanced as a model for other developing countries. Progress in implementingthe program has been substantial on the input side. But there are major shortcomings interms of outputs and outcomes. By far the most worrisome of the problems is the failureto achieve better learning outcomes. Recent test results suggest that minimal learning ishappening in the classrooms.

Bank Support to the Sector

4.28 The first Bank-supported project in this sector, the Health and EducationRehabilitation Project (HERP, approved in January 1986), provided inputs (textbooks andsupplies) on an emergency basis to stop the deterioration until a more comprehensiveprogram could be worked out. The next two projects, the First and Second EducationSector Adjustment Credits (EdSACs 1 and 2, approved in December 1986 and May 1990,respectively) provided support for the reform and introduced cost-sharing and efficiencymeasures to allow more resources to be devoted to quality improvements and basiceducation. The four subsequent projects provided support for specific sectors: expansionof the new Senior Secondary Schools (SSS) into underserved areas, non-formal education;tertiary education; and primary education. Altogether, these seven projects havecommitted US$232 million to assist Ghana's education sector. The first two projects havebeen satisfactorily completed; the others are being implemented without major problems.Learning outcomes, however, are difficult to predict.

4.29 Other donors have made significant contributions to the program. USAID hasprovided a US$35 million grant for a Primary Education Program (PREP) while UNDP(along with the Bank) provided funds for a technical assistance project to help establish acentral Planning, Budgeting, Monitoring and Evaluation (PBM&E) Division in theMinistry of Education (MOE).

4.30 The size of this Bank portfolio and the generally satisfactory projectimplementation to date are the result, among other things, of the Government's ownershipand commitment to the reform program. Another contributing factor was a capable taskmanager located in the field since its inception. This has enhanced the dialogue, made itpossible to address problems and potential misunderstandings quickly, and made it easierfor the Borrower to understand and comply with Bank requirements for procurement,disbursement and audit.

Outcomes

4.31 Although the reform program is still being implemented, it is possible to assessprogress so far. In terms of inputs, progress is substantial. Textbooks and teachers'manuals incorporating the new curriculum have been produced and distributed, allteachers have been provided with at least some in-service training for the new curriculum,the structural changes have been fully implemented, and cost-saving and cost-sharingmeasures have been put in place at all but the tertiary level-an important qualification.The gross primary enrollment ratio has increased, and although the enrollment ratio for

77

girls remains somewhat lower than for boys, the growth rate of girls' enrollment hasslightly exceeded that for boys, as may be seen from Table 4.2.

Table 4.2: Education Sector Indicators

Indicator 1970-75 1987/88 1991/92

Gross enrollment ratio %

Primary, total 80 (1980) 11 74 (1993)

Pupil-teacher ratio, %

Primary 30 23 28

Middle/Junior Secondary 23 19 18

Trained Teachers, %

Primary n.a. 56 72

Secondary n.a. 65 72

Average Annual Growth, % 1980181-1986187 1987/88-1991/92

Grade/enrollments 2.3 2.1

Primary enrollments 2.6 2.5

Girls enrolled in primary school 2.7 3.0

Middle/JSS enrollments 2.2 -0.8

Girls enrolled in Middle/JSS 2.6 -0.9

Source: For secondary enrollments in all years and for all 1970-75 indicators and SSA comparisons, Social Indicators ofDevelopment World Bank, 1994. For all other, tables from R. Ridker (1994).

4.32 In terms of outputs and outcomes, however, there is cause for concern for severalreasons:

Enrollments. While enrollment at the secondary level has increased satisfactorily,primary enrollment is expanding less rapidly than population. One reason is the increasedfees imposed by districts or local communities after the government decentralization programinitiated in 1988.

Equity of access. Equity of access has been improved by locating new schools inpreviously underserved areas and by eliminating the examination that previously took place atthe end of the primary cycle. This allows students to continue into junior secondary beforefacing a major screening test. There are, however, two countervailing trends. First, therehas been rapid growth in enrollments in private primary schools in the two major cities. Forthose who can afford the fees, these schools substantially increase the probability of their

78

children getting into good senior secondary and tertiary schools, thereby strengthening thehigh correlation between educational attainment and income. Second, the decentralization ofschool management and the cost-sharing measures are resulting in increased fees, both atprimary and secondary levels, that hurt the poor more than the rich.47 No integrated studyhas been undertaken to determine the net effect of these various trends. There have beenstudies on why parents don't send their children to school, on schooling expenditures (underGLSS), and in connection with the participatory poverty work.

Learning outcomes. Most disappointing and worrisome is that learning outcomesappear to be quite poor. The strongest evidence for this comes from two rounds of criterionreference tests given to sixth grade primary (P-6) students in conjunction with USAID's PREPproject, which indicated in two successive years of tests that only some two percent ofstudents were able to answer more than 60 percent of relatively simple mathematics andEnglish questions correctly.

This outcome is consistent with other surveys and with anecdotal information fromprimary and secondary schools. It is also consistent with information (including circuitmonitoring reports) on within-school activities indicating that learning materials, even whenavailable, are often not used, that actual student-teacher contact hours are very low (becauseof absence of teachers and/or students from class), that in-service teacher training (to orientteachers to the new curriculum) was inadequate, that parts of the new curriculum are overlyambitious (e.g., classes getting through less than half of the syllabus by the end of the allottedtime), and that within-school supervision and discipline of teachers is weak. There is also aproblem of poor linkage between the junior and senior secondary schools. One factor thatmay have affected the early scores was a reported "flight to quality", as the educated andhigher-income classes rushed their children through the old system; consequently, some of thebetter students missed the first years of the new system.

These poor test outcomes have provoked a new stocktaking by the government anddonors. Efforts are underway, in the context of a Committee for the Review of theEducational Reforms (established in 1994) and the donors' roundtable, to formulate a newvision and action plan geared to getting more satisfactory results in future.

Non-formal education. The non-formal program has not enrolled as many students astargeted, although a larger number than expected have come from poorer, rural areas. Moreserious, there seems to be a lack of discipline in the program's operations and management,as training materials are not available on time, new classes are not opening on schedule,meetings take place less than the prescribed four times per week, financial records are indisarray, there are delays in the production of reading materials in local languages, and lagsin research and evaluation activities. In spite of these difficulties, however, an impactassessment of the program does report some encouraging results in terms of learningachievement. One reason may be that adults are taught in their own languages.

47. This has not yet happened at teniary levcl because only feeding subsidies have so far been eliminated and a heavilysubsidized loan scheme has been introduced simultaneously.

79

Monitoring and evaluation. While some capacity building activities have borne fruit,an important one reviewed in this report-efforts to establish a viable PBM&E Division-hashad an unfortunate history. A high level of useful output was produced during projectimplementation, but most of this capacity has proved to be unsustainable.

Financial sustainabilit'. Given projected enrollment figures, it will be difficult tosustain required recurrent expenditures even if per student costs (corrected for inflation)remain constant. But if quality is to be increased, there will be upward pressures on costsrelated to vocationalization, the replacement of untrained with trained teachers, and upgradingmaterials and supplies. This puts a heavy burden on measures to improve efficiency and cost-sharing. Thus, the structural change is resulting in less savings than originally anticipated;revolving funds for replacing donor-provided textbooks are in place but not keeping up withfinancial requirements; room and board subsidies have been eliminated at secondary level butnot at tertiary level if the subsidy element in loans is taken into account; and pressures toincrease staff, while so far resisted, are likely to increase over time. The financialsustainability outlook needs to be regularly monitored in a modelling framework thatincorporates research findings on all important pressures on costs, including the willingness ofparents to pay. Contingency plans also need to be developed. These are jobs for thePBM&E Division, but that division does not yet have the capacity to undertake them.

Conclusions

4.33 The most serious of these problems is the failure to achieve better learning outcomes.It is difficult to discern the root cause of the problem and suggest what should be done aboutit. Projects supporting the reform program provided all the inputs thought to be necessary forthe proper operation of a school system. That they proved inadequate may mean that theBank does not know enough about how to reform an education system, such as Ghana's,which was so close to being moribund. A lesson is, as other countries and cities (including inthe developed world) have found, that educational reform is more complex and difficult thanis often thought at the onset of a reform program. There seems now to be a consensus thatthe Ghanaian reforms were too rushed, without adequate time being allowed for debate,training of teachers, development of curricula, etc., to phase in the reforms.

4.34 Two conclusions can be suggested in the form of hypotheses that need further testing.The first is that the curriculum reforms, imposed on the school system from above, werebased upon an inadequate knowledge of the capacity of the average school to absorb andimplement the changes. To the extent that was the case, the reforms should not have beenimplemented so rapidly: more time should have been taken to learn about the capacity of theschool system, to involve teachers in average schools in the process of developing the newcurriculum, to test the new curriculum and textbooks before applying them wholesale, and tomore properly train teachers in their use.

4.35 The second conclusion/hypothesis is that the system of rewards and punishments forperformance within which teachers operate is very weak, and has almost nothing to do withlearning outcomes. In Ghana today it is still difficult to fire a teacher for absence ordrunkenness, let alone for being a poor teacher. Until this changes-until there is morediscipline and accountability in the system-learning outcomes will continue to be poor if this

80

hypothesis is correct. This is the premise behind the Bank-funded Primary SchoolDevelopment Project, which attempts to use community pressure and improvements in thecalibre and motivation of headteachers to obtain better performance. The measures employedin that project, however, may be less powerful than required, given the severity of theproblem.

4.36 At least two things are needed for parents and community leaders to exert effectivepressure on teachers to perform: an objective way to judge performance, and budgetary orother mechanisms to affect the situation. The first requires examination scores for individualclasses or schools that can be compared over time and across schools. The second requiresthat the Government delegate budgetary or employment authority to local communities and, toreduce inequalities in the capacity to pay, provide funds to poorer communities (or parents)for use by their schools. Neither of these conditions is present in most of Ghana's ruralcommunities. Community involvement is however an important objective of the Bank'sprimary school development project in which 1,500 communities have so far participated inselection of their headmaster.

81

5. Progress Towards the Strategic Objectives

"As a sign of how times have changed, just think of terms and words like'queuing, ' 'essential commodities, ' 'hoarding, ' 'middlemen, ' 'black marketeer,'economic saboteur'! Our children today don't even know what these mean."

Ghanaian businessman

"We didn't have a clear vision after SAC I of what to do in institutionaldevelopment, private sector development, public sector management, or stateenterprise reform."

Senior Bank staff member

5.1 This chapter evaluates the extent to which the stated objectives of the Bank'scountry assistance strategy were achieved. One benchmark for the evaluation is the set oftargets identified in the CPPs and CSPs. The impact of the external environment is alsoconsidered. A principal question addressed is whether the outcomes seem likely to besustainable. The evaluation is made largely through assessing the performance of anumber of economic, financial, and social indicators. There is already a substantialliterature on Ghana's adjustment record (see Bibliography); this chapter will therefore givemore attention to some selected issues less well covered in the literature, e.g. institutionaldevelopment and sustainability.

5.2 A substantial part of the "evidence" underlying many judgments arrived at in thischapter lies in the performance of a variety of economic indicators shown in several tablesin the following pages.' In brief, the "story" told by the "incentive indicators" (Table 5.2)is that great progress was made in "getting the prices and incentives right." The story toldby the fiscal and monetary performance indicators (Table 5.3) is of more mixed success,e.g. with good desirable trends, for example, in tax revenues and public investmentexpenditures but with inflation and real interest rates at undesirably high levels andgovernment savings and the M2/GDP rations being desirably low. (M2 is defined as thesum of money and quasi-money.) The story told by the indicators on public sector reform(Table 5.4) is one of very slow progress in that area. Tables 5.5-5.7 describe a story ofsatisfactory growth in aggregate output, mining, services, and most social indicators butdismal growth in manufacturing, agriculture (with qualifications cited in Chapter 4), andespecially in savings (both public and private) and private investment.

5.3 A set of charts (presented in an Annex to this chapter) compare how actualoutcomes (in terms of the behavior of some key variables) compare with the values

48. See "Performance Indicators for Adjustment Program - Country Case Studies," DECVP, World Bank, September 1994(unpublished).

82

targeted/projected in the several CPPs and CSPs described in Chapter 2. Thesecomparisons serve to facilitate the "efficacy analysis" of this chapter, where efficacy refersto the extent to which objectives and targets are achieved.

5.4 The final set of indicators presented in this chapter (Table 5.8) comprises some"sustainability indicators" that the report advocates should be key/core benchmarks forassessing future progress.

Overview of Outcomes

5.5 Ghana achieved notable success over the past decade in macro-economic policyreform as measured by various macro-level indicators (GDP growth, export recovery,removal of price distortions, etc.), but far less in terms of institutional and structuralchange and as measured by various micro-level indicators (school test results, soil fertility,productivity of civil servants, etc.). The successes have been in rehabilitatinginfrastructure, getting the prices right, freeing up foreign exchange and other markets,reforming taxes and reducing subsidies, initiating cost recovery measures, and putting inplace some important elements of an enabling environment for private sectordevelopment.49

5.6 The record on poverty alleviation is mixed. Owing to a lack of baseline data forthe early 1980s, it is difficult to establish changes in poverty up to 1987/88, when suchbaseline data became available. But the available evidence suggests that povertydecreased somewhat in the mid-1980s and that the proportion of the population inpoverty has dropped marginally from about 43 percent in 1987/88 to about 42 percent in1991/92.

5.7 Ghana also has a mixed record on stabilization, although budget deficits werereduced considerably between 1983 and 1992. Intlation remained high and too much ofthe burden of stabilization was put upon monetary policy, with crowding out effects on theprivate sector. The least satisfactory performance has been in agriculture' andmanufacturing output growth, in institutional reform (including the civil service and publicenterprise divestiture and reform), in investment project implementation, and mostimportantly, in the "sustainability indicators." The disappointing response of privatesavings and investment is reflected in these sustainability indicators. Clearly the heavilyaid-dependent, public sector-led growth of the 1980s and early 1990s will not besustainable indefinitely. As will be discussed below, the interrelated issues of dependence,economic governance, and institutional development themselves form a nexus of"sustainability issues' that comprise core issues for the Bank's future assistance strategy.

49. The evidence underlying these and similar assertions in this section will be provided below, largely in the form of tables inthe text and charts in the Annex.

50. Even if the agricultural growth rate is higher than Bank data indicate, it is unlikely that the growth has reached the ratestargeted; more seriously, the trends in agricultural productivity and in environmental deterioration are the most worrisome.

83

5.8 Thus, the performance of the Ghanaian economy is a qualified success story. Theattention and praise given by spokesmen for the Bank and other donors to the successesat the macro-level may have served to mask some shortcomings at the sectoral and microlevels. While GDP growth was at or near the targeted five percent annual rate, thecomposition of this growth was markedly different from what was projected/targeted, withservices (largely trade) growing much faster and agriculture and manufacturing growingmuch slower than targeted. In some cases, moreover, there seems to have been a "haloeffect.""5 And some apparent successes at the design stage have not been translated intosuccesses at the implementation stage. For example, the Ghana education reformsdescribed in Chapter 4 were still, as of early 1994, being represented as highly successfuleven though "results on the ground" suggested that the outcome was far from a greatsuccess.

5.9 In sum, different stories can be told about Ghana depending upon the choice ofindicators, and this choice may focus attention on which actions and outcomes will be mostclosely monitored and measured in future. For this reason, a main finding of this report isthat it is exceedingly important to "get the right indicators" as a precondition to "gettingthe indicators right."

External Shocks/Exogenous Factors

5.10 Ghana's performance was affected by events and shocks beyond the control of theGovernment, but which called for a response by the Government and donors. Table 5.1shows the nature of the shocks experienced by Ghana over the fifteen years 1978-92.(Note that a positive sign for the shocks refers to adverse shocks rather than to beneficialwindfalls.)

51. This was clearly the case, for example, with the Structural Adjustment Institutional Support Project that tended to beregarded (including by the PCR) as more successful than the facts warranted, probably at least in part because of that project'sassociation with the successful first two SAL operations.

84

Table 5.1: External Shocks and Performance Response Measures (as percent of GDP)'

Period Average Period Average Period Average1978-82 1983-87 1988-92

External Shocks

Terms of trade 3.4 -0.3 2.8

Export volume 0.7 -0.1 0.5

Total 4.1 -0.4 2.3

Performance Response Measures

Additional net external financing 1.6 -0.8 2.3

Export promotion -0.4 0.0 0.9

Import intensity 2.8 1.0 -0.5

Economic compression 0.0 -0.6 -0.3

Total 4.1 0.4 2.3

a. For an explanation of the concepts and methodology employed in deriving this table, see Desmond F. McCarthy andAshok Dhareshwar, Economic Shocks and the Global Environment (Policy Research WPS 870, March 1992).

5.11 The terms of trade deterioration was quite substantial in the years just precedingthe ERP and in recent years. During the mid-1980s Ghana's terms of trade improvedslightly until 1987 (104, with 1984 = 100), but the index has deteriorated steadily sincethen, (to 74 in 1992 and 66 in 1993). Falling cocoa prices were a main determinant of thisdecline, as they declined from 90-110 cents per pound in the period 1984-87 to between50-56 cents over the years 1989-93.

85

Box 5.1: Projected vs. Actual Cocoa Prices

It was noted in Chapter 2 that the Bank's projections of Ghana's terms of tradewere not very accurate, mainly because the Bank did not correctly foresee how largewould be the fall in cocoa prices in the late 1980s. The differences between the Bank'sprojections and the actual outcomes in Ghana's export prices and terms of trade areshown in Figures 5.1 and 5.2.

Several questions can be posed about these projections. One is whether theBank's forecasters were particularly wrong in this instance, or only about as wrong asother forecasters (given that no one has proven consistently accurate at forecasting).Another is whether the error in the forecasts had serious adverse consequences-forexample whether large and lumpy investment projects were launched on the basis ofcommodity price (and consequent revenue) forecasts that subsequently could not befully funded. A related question would be whether there was sufficient contingencyplanning by either the government or Bank for dealing with unexpectedly adverseoutcomes in the terms of trade.

Apropos of the first question, the Bank has been accused of indulging its "vestedinterest in commodity price optimism" in order to "validate the export-led growthstrategy for primary product exporters."" But while there were flaws in the technicalanalysis, the same technical reasons (too much reliance on extrapolation) led the Bank'sforecasts below actual prices for the period 1974-82-a mistake for which the Bank wasnot accused of "commodity price pessimism."b Since then, the Bank has improved itstechnical analysis by taking into account the vintages of cocoa trees planted around theworld and by relying more on futures prices for near-term forecasts, and oneconometric models only for medium- to longer-term forecasts. It is also consultingmore with private forecasters, and since 1988 its cocoa forecasts have been quiteconsistent with private sector forecasts.

a. Toye, J. (1991) "Ghana" in Mosley, P. and others, Aid and Power, NY: Routledge, p. 197.b. This conclusion is based upon a study by Prof. Angus Deaton comparing actual vs. forecasted cocoa prices. See A.Deaton (1992), "Commodity Prices, Stabilization, and Growth in Africa." Princeton University Research Program inDevelopment Studies, mimeo.

86

Projected vs. Actual Values(Projections from Bank Strategy Papers)

Flgure 5.1: Changes in Terms of Trade

30 -

Flgure~ ~~~ 5.2 Chne inpcA Exor rie

20

I~ 0

-10

-20

-3019501 9811 98219831984198519861987198819891990199v11g92

Figure 5.2: Changes in Export Prices

20-

5

-20

2019801 9811 98219'8319'841 985198619i7l 9881 98919901 991 199 2

87

5.12 Ghana was able to offset these external shocks mostly through its access toincreased concessional external financing. Since a substantial share of this externalfinancing was in the form of quick disbursing non-project aid, the Government retainedconsiderably more financial degrees of freedom than had it been forced to undertakeimport intensification or economic compression. But if the terms of trade deterioration ispermanent, one cannot ignore also the undesirable effect of raising the country's aiddependence and the consequences of the "Dutch disease" (see Box 5.2).

5.13 Natural disasters and exogenous circumstances are other external shocks. InGhana, with over 40 percent of GDP originating in agriculture, weather conditionssignificantly affect GDP; performance can vary severely owing to periodic droughts such asin 1981-83 and again (to a lesser extent) in 1990. Nigeria's expelling of over one millionGhanaians in 1983 resulted in about a 10 percent increase in Ghana's population thatyear. Conversely, the overvaluation of the CFA franc relative to the Ghanaian cedi andhard currencies up to 1993 was a positive exogenous development, whereas the CFAdevaluation in 1994 constituted a shock.

5.14 There was relative peace between Ghana and its neighbors, political stability withinthe country, and no major natural disasters to cope with between 1984 and 1993. Periodictensions with Togo and CMte d'Ivoire did not cause major disruptions to the economy.There was however a man-made disaster in February 1994 when tribal strife in thenorthern region resulted in deaths of more than a thousand persons, in large movementsof refugees fleeing this strife, and in the interruption of all developmental activities in theregion. The causes and consequences of this strife need to be better understood if theBank's strategy is to be as "relevant" as may be necessary to help avert recurrences infuture. Such events clearly highlight the need for donor strategies to be grounded in agood understanding of a country's political economy.

5.15 Because the aid inflows largely offset the exogenous shocks over the period underreview, the latter cannot be blamed for those aspects of Ghana's performance that wereless-than-satisfactory. Yet the impact on the economy, and on the behavior of the variousagents in the economy, are very different depending upon whether Ghana's foreignexchange resources come from export earnings or from aid. This requires that the Bankpay close attention to how changes in macro variables affect micro-level incentives andbehaviors, as in the case of cocoa earnings versus aid transfers. It also means that positivewindfalls (such as the discovery of oil, or a surge in cocoa prices, or very plentiful suppliesof external assistance) typically have some distorting, downside effects that need to bemonitored and minimized insofar as possible. These effects may, in some cases, outweighthe benefits when a "behavioral approach" and a long-term view are taken. In Ghana'scase, the benefits of plentiful aid have more than compensated for the windfall losses fromthe terms of trade. But one type of distorting effect that large aid flows can create is thatknown as the "Dutch disease" effect.

88

Box 5.2: Aid And The Dutch Disease: Macroeconomic Management When EverybodyLoves You

The article with this title (by Stephen Younger) appeared in the November 1992issue of World Development. In it, Younger shows how the large aid inflows into Ghanain recent years have sometimes worked at cross purposes with both stabilization andstructural adjustment objectives.

Younger points out that of the 5-6 percent [in some years more] of GDP that hasbeen flowing into the government's coffers each year in aid, less than one percent hasbeen channelled to the financial sector via repurchase of debt or bank deposits. Therest has gone to government spending. Younger notes that although some governmentexpenditures are ill-advised, the problem is not too-high government expenditure per se.The problem is that "when the government accepts foreign aid to finance localexpenditures, it forces the money base to increase as dollars are converted intocedis"-with the resultant inflation causing an appreciation of the real exchange rate.Then, in trying to hold down inflation by restricting money growth, Younger argues, thegovernment has pursued a tight domestic credit policy, at the expense of crowding outprivate investment.

The remedy? In seeking to limit aggregate demand, Younger argues that thegovernment should look for ways to raise revenue through more incentive-neutral taxes(such as income tax or a value-added tax), thus constraining private sector expendituresin general rather than just limiting investment through tight credit.

In practice this means that the government should try to balance the budgetinclusive of project aid (i.e. by running a surplus of say four percent of GDP beforeprojects) and channelling the surplus (through debt reduction) to the private sector."This is what Ghana must do," Younger concludes, "if it wants to shift the engine ofgrowth from public sector recovery projects to new productive investments in the privatesector."

Evaluation of Performance in Economic Management

5.16 This section reviews Ghana's performance in economic management against theobjectives of the Bank's strategies as outlined in Chapter 2. The relevant performanceindicators are those relating to policy and institutional measures. A subsequent sectionreviews performance against the "bottom line" objectives of accelerating growth andalleviating poverty. The final section reviews progress against some key "sustainabilityindicators."

5.17 Exchange Rates: Excellent. Ghana proceeded gradually, through a number ofstages, to achieve its present unified, largely market-determined exchange rate system.The foreign exchange markets were unified in 1990 with the introduction of a weeklywholesale auction of foreign exchange. In 1991 all export proceeds were allowed to be

89

sold to banks and bureaus except for cocoa and gold, and in 1992 the Bank of Ghanareplaced its weekly auction with an interbank market. More than half the foreignexchange in the interbank market is supplied by the Central Bank. The black marketpremium (see Table 5.2) has become negligible. These developments have helped tofoster export development. But Ghana still lacks organization for export promotion.There is no export credit facility, export bank or import duty drawback system, and an"export mentality" has yet to develop. The IMF took the lead on exchange rate mattersbut the Bank played a substantial role and made a significant contribution to the exchangereforms through its economic work and policy dialogue.

5.18 Fiscal Discipline: Fair. Achievements under the ERP included increased resourcemobilization and budget deficit reduction (see Table 5.3 and Figure 5.6 in the Annex tothis chapter). Fiscal policy formulation is however hampered by insufficient data toconstruct accurate consolidated public sector accounts. In any event, there was recurrentslippage in meeting fiscal targets, and the policy mix became too weighted towards fiscalease and monetary tightness even before the 80 percent wage increase of October 1992.New revenue measures were taken in 1993-94, but the deficits are still larger than isconsistent with low inflation and the provision of adequate credit to the private sector.The IMF took the lead in negotiating deficit reduction targets under its arrangements, butthe Bank also contributed to the achievement of budgetary reform through its policydialogue, its economic studies relating to taxes, subsidies and cost recovery, and through itseffective technical assistance to the National Revenue Secretariat.

5.19 The Bank's strategy documents underestimated revenue growth before 1988 andoverestimated it since then, while it consistently overestimated expenditures (see Figure5.6). The revenue overestimate was related to the underestimation of the decline incoffee prices, while the expenditure overestimate stemmed from over-optimism regardingb')th aid flows and project implementation capacities. The deficits in most years werelarger than projected in the CPPs, but in the more-frequently updated PFPs, the Bankproved overly optimistic about the rate of deficit reduction.52

5.20 The 1992 election-related fiscal shock had a greatly adverse impact insofar as itweakened the response to the positive private sector development measures. And becausethe fiscal situation remained problematic into 1994 and 1995, with large credits to publicenterprises being the main problem, this continued to have a depressing effect on privateinvestments and savings.

5.21 Tax Reform: Good. There have been major improvements in resource mobilizationand in the rationalization of the tax structure. The tax base has been broadened, marginaltax rates have been reduced, and the indirect tax regime has been rationalized. A ValueAdded Tax (on which the Bank did considerable preparatory work) is to be implementedin 1995. Other Bank contributions on the revenue side were mentioned above.

52. See R. Faruqee, How Macroeconomic Projections in PFPs for the Africa Region Compare with Outcomes, Africa Region,Office of the Chief Econorrist, August 1993.

90

Table 5.2: Selected Incentive Indicators

1984 1987 1990 1993

Exchange Rate PolicyNominal exchange rate 36 162 330 652Real exchange rate 45 100 102 134Black market premium 147 33 10 3

Trade PolicyAverage of import tariffs 27 24 13 13Import duties/total imports 11 10 13 n.a.

Liberalization of PricesTradable CPI 100 225 459 665Non-tradable CPI 100 206 494 799Tradable CPI/Non-tradable CPI 100 109 93 83

Incentives for CocoaReal produce price of Cocoa 758 2132 1565 n.a.Producer price/FOB price 36 38 51 n.a.

Real interest rate91 day T-Bill -16.6 -14.5 -7.1 7.6

Financial Sector MeasuresReal deposit rates -17.0 -13.3 -14.0 -2.6Real lending rates -12.3 -9.9 -8.8 3.6

Spread 4.7 3.4 5.2 6.2

5.22 Public Expenditure Priorities: Fair. Expenditure allocation has improved markedlysince the early 1980s through implementation of a rolling, three-year public investmentprogram and regular public expenditure reviews. Inefficient subsidies have been reduced,white elephant projects minimized, debt arrears have been cleared, and expenditures havebeen directed towards priority areas such as primary education and infrastructurerehabilitation. Ghana's expenditure levels and composition are however in large part"donor facilitated" if not "donor driven." The huge public sector in terms of employment(nearly 600,000 employees in a country of 16 million persons, an extremely large numberby a comparative standard) and the consequently large wage bill underlie the

91

Table 5.3: Selected Fiscal and Monetary Performance Indicators

1984 1987 1990 1993

Fiscal Indicators (as % of GDP)

Total revenues' 8.0 14.1 11.8 15.9

Tax revenues 6.6 12.7 10.8 12.7

Total expenditurek' 11.1 19.1 17.0 26.3

Recurrent expenditureE' 8.6 11.2 10.2 15.5

Capital expenditure 2.5 7.9 6.8 10.8

Current savings- -0.6 2.9 1.6 -0.4

Overall deficit -3.1 -5.0 -5.2 -10.4

Financing

Monetary (net) 0.7 -1.0 -1.0 0.0

Other domestic 0.4 0.6 -0.4 1.6

Foreign 2.0 5.4 6.6 8.8

Monetary Policy (annual percentage changes)

Broad money 61 53 18 n.a.

Credit to rest of the economy 103 30 110 n.a.

Quasi-money/M2 16 21 23 34

M2/GDP 11 17 17 17

Real M2 growth 25.8 14.0 -13.2 n.a.

Real GDP growth 8.6 4.8 3.3 n.a.

Inflation (avg) 39.6 39.8 37.2 25.2

Real interest rate, 91 day T-bill -16.6 -14.5 -7.1 7.6

a. Excluding grants.b. Including capital expenditure financed by externally financed grants and loans.c. Includes special efficiency.d. Total revenue minus recurrent expenditure.

92

Government's difficulties in controlling the deficit. Major weaknesses in budget execution,accounting and monitoring systems remain to be corrected. The Bank rrade a majorcontribution in this area through its economic work on public investment and publicexpenditure reviews.

5.23 Trade Liberalization: Mixed. The Bank considers Ghana's trade liberalization tohave been a success since imports are not subject to quantitative restrictions, the importlicensing system has been abolished (as of 1989), export taxes have been abolished exceptfor cocoa, and tariffs have been rationalized and brought to low average levels (three tierswith rates of 0, 10, and 25 percent). The removal of import quotas and tariff reductionwas however done quickly and to some extent unexpectedly, resulting in widespreadbankruptcies in the private sector. Both Bank staff and Ghanaians are divided on thequestion of whether the trade liberalization went too far too fast (partly as a result ofBank conditionalities), resulting in the demise of potentially viable firms that did not havethe time or wherewithal to adjust and to survive.

5.24 Deregulation: Fair-to-Good. There has been substantial deregulation in recentyears, and a new Investment Act now makes the approval of all investments automatic.Regulations were liberalized on minimum capital requirements for foreign investors,expatriate immigrant quotas, technology transfer rules and the number of industrialactivities exclusively reserved for Ghanaians. But Bank staff cite considerable anecdotalreports of remaining "informal" regulation, intimidation, and rent-seeking behavior by thepublic administration that inhibits private investors and underlies the "credibility problem"(see Box 5.3). Both the Bank and IFC have provided studies and advice on how toimprove the business environment.

5.25 Foreign Direct Investment. Fair. Barriers impeding the entry of foreign firms havein large part been abolished, but foreign investment inflows have so far been minimal-inlarge part for the same reasons that inhibit domestic private investment. Both the Bankand IFC played substantial roles in advising the Government on how to create a morefavorable investment climate for foreign as well as domestic investors.

5.26 Pnivatization: Slow (at least until 1994). Ghana's divestiture program progressedslowly, with only 54 of 300 non-financial public enterprises fully divested by 1993 in thesense that they were liquidated or leased or sold to the private sector with all proceedscollected; in another 16 enterprises there has been a partial collection. Bank staffattribute the slow pace of privatization partly to the initial preparation time but mainly toinsufficient political support. There seemed to be a lack of ownership of the program(contrary to the ownership of the macroeconomic reforms) partly because of anideological heritage (shades of Nkrumahism and populist sentiments of the early PNDC),and partly because of vested interests. Performance contracts were signed with some 13state enterprises but pricing and procurement decisions are still taken by the Governmentand efficiency improvements in the enterprises have been very limited. Capacityutilization is very low and many enterprises need to be liquidated. The retrenchment oflabor is however frustrated by prohibitively expensive end-of-service benefits. There is still

93

Box 5.3: Is Credibility the Binding Constraint to Private Investment?

A recent article by Ernest Aryeetey published in World Development addresses the criticalquestion: 'Why is the private sector not investing [in Ghana] as is desirable and expected?"

Aryeetey notes that there are a number of different views on this subject. 'There are those,"he says, 'including the World Bank who believe that the reforms have not gone far enough inaddressing substantive issues (mainly institutional) affecting the operations of potential investors, inspite of the fact that all manner of incentives arising from macroeconomic and other sectoralreforms have been provided. Other views expressed by many Ghanaians suggest that the issue isone of credibility with regard to the sustainability of reforms in the medium-to-longer terms."

There is a growing literature, authored both by economists and political scientists, on theextent to which "credibility' is a major determinant of the behavior of private agents, particularlydecisions of private entrepreneurs as to whether and where to invest. One recent study, in whichGhana was one of 28 countries surveyed, found that the degree to which entrepreneurs expectgovernments to stick to major policies explains half the variance in growth rates.b

AryeeKey, drawing upon principal/agent and game theoretic models and upon several surveysmade of Ghanaian businessmen, concludes that a "major reason" for the poor private investmentresponse is the a "deep mistrust by the private sector for a govemment whose earlierpolicies... indicated an antagonistic attitude to private investment." Aryeetey does not find anexpectation that the government's market-friendly policies will soon be reversed, but he notes that"the sentiment is often expressed among Ghanaians that various macroeconomic reforms are pursuedonly to satisfy donor conditions for assistance'-implying a perception of weak ownership by thegovernment of its reform program.

Aryeetey finds that the very high degree of risk aversion found among Ghanaian investors isrational only under relatively high levels of uncertainty, where the uncertainty pertains to thegovernment's real ownership of its reform program, and is reflected in the limited credibility ofinvestors in the sustainability of the reforms. Aryeetey argues that the binding constraints to privateinvestment are not therefore so much in the regulatory and incentive frameworks emphasized somuch by the Bank as in the "lack of consensus on a development philosophy and what role theprivate sector is expected to play." This uncertainty would only be removed, heconcludes, if adequate legislation is passed to assure investors that property rights would berespected in future.

Aryeetey does not use the word "governance" in his article, but his analysis is fully consistentwith this report'- conclusion that only by addressing the core 'sustainability issues' will Ghanaovercome the "credibility constraint" to increased private investment.

a. See E. Aryeetey, "Private Investment under Uncertainty in Ghana," World Development, vol. 22, no. 8 (1994).b. A. Brunetti and B. Weder, "Credible Policies and Economic Growth: A Cross-national Analysis,"HarvardUniversity, mimeo (1995).

94

clear action plan for enterprise divestiture or rationalization. The Bank's approach toprivatization (as well as to deregulate public enterprise reform and civil service reform)may have been too "formulaic,"and not sufficiently heedful of important political andsocial dimensions of privatization.

5.27 In 1994 there appears to have been some acceleration of activity on the fronts ofpromoting privatization and private sector development. Ashanti Goldfields was privatizedand a bill was sent to Parliament to launch privatization of Ghana's telecommunications.A liberal investment law was adopted along with Ghana's subscription to Article VIII ofthe IMF whereby Ghana renounced recourse to quantitative restrictions on its externaltrade. 1994 also witnessed increased activity on the Ghana stock market and a moreactive dialogue between the public and private sectors. The results in terms of privateinvestment levels, however, remain to be seen.

5.28 Financial Liberalization: Fair. Interest rates have been liberalized and sectorallending guidelines lifted. Bank portfolios have been restricted, prudential regulationsrationalized and banking supervision strengthened. But interest rate spreads are high(Table 5.2) and service quality is low, pointing to a lack of competition. The oligopoly infinancial markets contributes to high real interest rates and lack of availability of foreignexchange. Divestiture of banks under the Bank's Second Financial Sector AdjustmentCredit (FINSAC 2) has proceeded slowly.

5.29 Civil Service Reform.: Very Slow. Ghana has a huge civil service that is widelyacknowledged to be dispirited, de-motivated, oversized, and under-productive. The ethosof the bureaucracy is a major problem, and with donors providing salary supplements in avariety of overt and covert ways, rent seeking is rife. A civil service reform program(CSRP) was launched in 1987, with conditionalities prescribed in SAC I. But the CSRPlacked ownership, at least partly because there was a lack of participation by the variousgroups that had a stake in the outcome, and the program had few positive results. Netretrenchments were negligible notwithstanding substantial gross retrenchments, owing tounplanned new hiring (Table 5.4). Efforts to control the wage bill at 4.5 percent of GDP,which were finally achieved in 1990, were undone by the 1992 wage hike that raised thewage bill to 8 percent of GDP. Other reforms such as the automated payroll system andjob evaluation and planned links of pay to performance have so far been ineffectual.Bank staff report, however, that the new Parliament is now making demands on the civilservice to become more effective, and a new CSRP is underway that will supposedlyinvolve a more participatory process to build ownership and the CSRP within a frameworkof an overall strategy regarding the role of the public sector.

5.30 Institutional Development (ID) and Capacity Building. Within the last two yearsthe Africa Region has given greatly increased priority to capacity building as an objectiveof its country assistance strategy. Among the country-specific reasons for this emphasiswas the recognition that Ghana's project implementation capacity was inadequate tomanage even existing Bank and donor projects, much less the more demanding sectoraloperations that the Bank wanted to be able to launch. "Absorptive capacity" rather thanfinance was seen as the binding constraint to project implementation, and the remedy was

95

to build capacity. Box 3.3 in Chapter 3 illustrated the very mixed record of Bank projectsin achieving the ID objectives.

Table 5.4: Selected Indicators of Public Sector Reform

1987 1992

Numbers of Public Sector Employees('000)

Core civil services 131 102

Education services 159 167

Subvented organizations 81 70

Security organizations 29 29

Total Government services 400 368

Net retrenchment, 1987 to 1992 - (32)

Public enterprises 250 227

Net retrenchment, 1987 to 1992 - (23)

Total pu5Jic sector 650 595

Net retrenchment, 1987 to 1992 - (55)

Civil Service Wages/GDP 2.0 5.7

1984 1987 1990 1991 1992 1993

Performance of State-owned Enterprises

Profits/GDP (%) 0.3 1.9 0.7 na na naEstimated number of public firms na 329 291 266 na naDivestitures (no. of firms)

Targeted for divestment (no.) na 39 42 na naActual divestments na 23 25 na naProceeds collezted 7 3 7 1Proceeds partially collected 7 4 4 1No proceeds collected 1 10 2 3Lease/management contract 3 0 1 0Liquidation 22 2 2 6

Source: Adjusmneni in Africa: Lessons from Country Case Studies, World Bank Regional and Sectoral Studies, 1994.

96

Box 5.4: Rent-Seeking and Patronage Relations

A subject much talked about by Ghanaians and by Bank staff working on Ghana,but not written about, is how the large donor presence in Ghana has given rise to newforms of rent-seeking through new forms of donor/Ghanaian patronage relationships,which have replaced old forms of rent seeking (often associated with access to scarceimport licenses or foreign exchange).

Incentives in Ghana's public sector are significantly distorted by multiple scales ofdonor-financed salary supplements, and by the non-transparent manner in which donor-financed benefits (vehicles, external training, study tours abroad, appointment toadvisory groups, commissioning of studies, etc.) are allocated through patronagerelationships. Another distorting factor has been the dominance of objectives that areessentially short-term, product- and event- oriented, and often donor-driven, overlonger-term objectives that are more "process oriented" and geared to ID and capacitybuilding. Technical assistance projects that give considerable discretion to project andtask managers regarding the allocations of vehicles, study tours and the like are amongthe principal targets for this rent-seeking.

An example of the above was to be found in a Bank technical assistance projectapproved in 1987.' This project helped to foster a system in which much of the regularwork of the Ministry for Finance and Economic Planning, whether preparing the publicinvestment program or for a Consultative Group meeting, came to be undertaken by adhoc task forces and short-term consultants rather than by the institutional units chargedwith these tasks. This de facto privatization often served to postpone if not underminethe development of the capacity of the institution to do its assigned tasks. But sincethe task forces and steering committees and advisory groups all got supplements, it isno wonder that this approach became popular. Many donors now acknowledge thatwork does not get done without the payment of supplements of some sort.

Behaviors induced by these distortions are sometimes disabling, as in the case ofpublic servants who have become resentful and de-motivated because they cannot sharein donor-financed benefits received by a favored few. The Skills Mobilization Schemeof the late 1980s clearly had this kind of adverse external effect. Yet when the valueof a vehicle is worth several years of a Ghanaian civil servant's salary, and a day'sconsulting fee or honorarium is worth a month's salary, much time is spent away fromassigned jobs in pursuit of fees and favors.

a. Structumi Adjustment Institutional Suppori Project (Credit 1778-GH).

97

5.31 The Government has created an inter-agency Capacity Building SteeringCommittee, a joint UNDP-Government team has prepared a study and action plan forcapacity building, and work is underway to define a strategy "for both better utilization ofexisting capacity and development of the specialized skills needed to achieve acceleratedeconomic growth." But the OED findings on past efforts to assist capacity buildingthrough technical assistance (TA) operations emphasize the need to ensure that thesystemic, governance-grounded constraints to more effective capacity utilization are beingaddressed at the same time. Otherwise the interventions treat symptoms rather than rootcauses of the inefficiency of the public service. Unfortunately, the UNDP study did notaddress the demand side, utilization issues.

5.32 Future TA and capacity building activities in Ghana should be framed in thecontext of a comprehensive strategy for institutional development. Many of Ghana'sinstitutional problems stem from demand side rather than supply side constraints, and the"ethos of the bureaucracy" and the overextension of government are central, systemicissues. Until and unless these systemic problems are addressed more directly andvigorously by civil service reforms and other relevant measures than they have been todate, the sustainability of the Bank's TA/ID projects and project components will beuncertain at best.

Progress Against Growth and Poverty Alleviation Objectives

5.33 GDP growth: Right rate (excepting 1992 and 1994), unintended composition. GDPgrowth averaged nearly five percent over the decade-close to the targeted rate. But thecomposition of the growth was quite different from that intended and projected (Table 5.5and Figures 5.9 to 5.12 in the Annex to this chapter). The growth of both agriculture andmanufacturing was substantially lower and that of the services sector-especially thedistributive trade subsector-was substantially higher than projected. The rapid growth intrade was stimulated in part by the large amount of non-project lending that was spent inlarge part on consumption and intermediate goods. The slower-than-projected rate inagriculture was partly attributable, as was discussed in Chapter 4, to the relative neglect ofagriculture by the Government and donors in the mid-1980s.

5.34 Poverty alleviation: Some gains. The main thrust of Ghana's ERP was to "go forgrowth" as the primary objective vis-d-vis poverty alleviation, income distribution,environmental conservation or other objectives. Given the country's circumstances,however, it was clear that accelerated growth was a precondition for poverty alleviation.The Bank was more concerned in the mid-1980s with helping Ghana to get its prices rightand to get stabilized and to get growth going than with reducing poverty-especially sincenot very much was known about the extent and incidence of poverty. And both Ghanaiantechnocrats and Bank staff were more than fully occupied initially with crisis managementand with mobilizing sufficient resources from home and abroad to enable the ERP to beimplemented.

98

Table 5.5: Selected Growth and Structuial Indicators

1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

GDP and Sectoral Growth

GDP 8.6 5.1 5.2 4.8 5.6 5.1 3.3 5.3 3.9 5.0

Industry 9.1 17.6 7.6 11.5 7.3 2.6 6.9 3.7 5.7 4.3Mining 13.5 6.5 -3.0 7.9 17.8 10.0 6.4 6.7 10.4 9.1Manufacturing 12.9 24.3 11.0 10.0 5.1 0.6 5.9 1.1 2.7 2.2

Agriculture 9.7 0.6 3.3 0.0 3.6 4.2 -2.0 4.3 0.6 2.8Cocoa -8.4 13.2 18.2 3.3 -6.3 3.2 3.0 -2.0 2.1 3.5

Services 6.6 7.5 6.5 9.4 7.8 6.7 7.9 6.3 6.6 7.0

Seetoral Structure (% GDP)

Industry 11.3 12.7 13.0 13.8 14.0 13.7 14.1 14.0 14.2 14.1Agriculture 52.5 50.3 49.5 47.1 46.1 45.8 43.5 43.2 41.8 41.0Services 36.2 37.0 37.6 39.1 39.9 40.5 42.4 42.9 44.0 44.9

Savings

Savings (Gross Nat'l) 4.0 5.4 5.1 5.6 6.2 7.2 7.4 9.3 7.2 0.2(% GDP)

Public -0.6 0.1 1.7 2.9 2.7 2.5 1.6 3.2 -2.2 -0.9Private 4.6 5.3 3.4 2.7 3.5 4.8 5.8 6.1 6.4 1.2

Investment (% GDP) 6.9 9.6 9.4 10.4 11.3 13.2 14.4 15.9 12.8 13.9

Public 2.5 4.2 7.3 7.9 8.0 7.8 6.8 7.7 8.5 9.9Private 4.6 5.3 2.1 2.5 3.3 5.5 7.6 8.1 4.3 4.0

Memo Item

Direct Foreign Investment(USSm) 2.0 8.0 4.3 4.7 5.0 15.0 14.8 20.0 22.1 25.0

5.35 By about 1985, however, the Bank was becoming increasingly concerned with the need toaddress the social costs of adjustment. In the case of Ghana, UNICEF played a catalytic role ingetting both the Government and the Bank to pay more attention to "human concerns," albeit notwithout antagonizing the Government. The tensions were resolved and in 1987 two poverty-oriented initiatives were launched. One was a Ghana-specific, multi-donor program known asPAMSCAD (see Box 5.5), the other another multi-donor multi-country program know as theSocial Dimensions of Adjustment Project (SDA) (see Box 5.6).

99

Box 5.5: Program of Actions to Mitigate the Social Cost of Adjustment

PAMSCAD was conceived as a component of the Economic Recovery Program to cushion the impact of structuraladjustment on vulnerable groups and to support those poor who would not benefit from the ERP in the short run. TheGovernment and donors alike considered the program a "flagship" in the efforts to counter critique of structural adjustment.

At the CG Meeting in 1987 multi- and bilateral donors enthusiastically supported the PAMSCAD concept and aninteragency preparation mission, led by the World Bank, was fielded to assist the government's Social Sector Task Forcein preparing the program. Preparationwas strained by two factors. First, it was difficult to identify precisely those householdsthat were particularly vulnerable. Second, a sectoral approach led to competition between line ministries for projects which,in many cases, were off-the-shelf projects lacking financing or extensions of existing projects.

The resultant US$85.7 million program was a patchwork of 23 individual projects to be implemented by several agenciesof seven ministries coordinated by two special units. Employment generation projects were to provide productive jobs andincrease incomes for the urban and rural unemployed and underemployed while at the same time produce useful social andeconomic assets for the country. Retrenched workers were to receive financial compensation, training, counselling andresettlement assistance. Basic needs projects were to provide essential drugs, supplementary feeding of malnourishedchildren, safe and reliable water supply to rural inhabitants, and deworming of primary school children. Education sectoractivities comprisea bulk food purchase to reduce cost of feeding secondary school children and paper commodity aid forprimary school books. Community initiative projects were to enable communities themselves to identify and implementactivities to improve their social and economic infrastructure.

At the pledging conference in Geneva in February 1988, 13 donors pledged a total of IJS$85.7 million. SubsequentlyUS$83.9 million was conmitted. Disbursements were, however, delayed considerably. Of the 13 donors supportingPAMSCAD, four began to commit resources in 1988, seven in 1989 and two in 1990 (as of 31 July). In terms of the volumeof disbursements, only 19 percent of total funds had been released by the end of the first year of implementation. Acontributory factor was donors' require nent for reappraisal (after the Bank's appraisal) and additional information on projectactivities prior to release of funds. About half of the 23 projects were cofinanced by three or more donors.

Implementation was plagued by the complexity of the program and institutional arrangements with the 23 projects,dispersed across nine sectors and all regions. Despite the intentions to design the program for decentralized implementationwith modest institutional requirements, there was a strong tendency towards centralization of key program managementresponsibilities within national ministries or agencies and to establish and equip implementation committees and units withstaff, computers and vehicles. The monitoring system, which was to be financed under the SDA program, was neverestablished.

A multi-donor evaluation in 1990 found that PAMSCAD had not shown significant benefits in terms of mitigating thesocial cost of adjustment. Some of the deficiencies noted were a short time horizon; dispersion of interventions in all regionsdespite the geographical concentrations of the poor; contrary to the analysis of poverty and vulnerability and the statedintentions there was a bia; in resource allocation towards communities and male beneficiaries in urban and/or southern areas.While these groups were affected by structural adjustment in the short term, i.e. some 45,000 retrenched governmentemployees who in general found the severance pay (which was not donor funded) adequate and few took advantage of varioustraining and credit schemes offered, they were not particularly vulnerable compared to small farmers in the northern regions.

The important lesson from PAMSCAD is that, being primarily palliative in design and function, it dealt with short-termeffects rather than causes. This orientation led to a segmentation of the poverty problem and partial rather than holisticapproach to interventions. It marginalized issues of poverty and vulnerability from mainstream economic policy and planning,by appearing to address them within a distinct program, at a time when the more fundamental issues of poverty and thepolicies and strategies needed to address them remained unaltered.

100

Box 5.6: The Social Dimension of Adjustmnent Project

The Ghana SDA project was initiated by the "Assessment of Social Dimension of StructuralAdjustment for Sub-Saharan Africa Project" which was launched in 1987 with UNDP, AfDB and theBank as partners. The regional project and the Ghanaian project have identical objectives: to supportthe Government in formulating and implementing a medium- and long-term poverty alleviation strategyin the context of its Economic Recovery Program. In this perspective, the project would foster thedevelopment of a close interaction between the entities responsible for the design and follow up ofpoverty alleviation programs and projects, the execution of policy studies on the social dimension ofadjustment, and the productinn of statistical data on the structure and evolution of living conditions ofhouseholds. The Ghanaian SDA project was one of the first country projects to be developed.

Partly because the SDA concept was new, project preparation and appraisal took an unduly longtime-27 months. The SDA project was seen by the MFEP as 'supply driver1" by the Bank (SDA Unit)because of the amount of input to and frequency of missions dealing with GSS, and the Bank's veryactive role in securing bilateral cofinanciers. After one year of preparation, the MFEP intervened inthe negotiations, which until then had primarily been between the Bank (SDA Unit) and GSS. Theproject proposal was unacceptable to MFEP because it was too much focussed on statistical datacollection and too little on linking the work of GSS with Ghanaian policy makers' needs. It then tookanother 10 months before a revised project was ready for negotiations.

The Ghana Living Standard Survey is of a scale and complexity that few statistical departmentsin the developed world would have undertaken. Neither the processing or the analytical capability wasavailable in Ghana. The first LSS was conducted in 1987/88, i.e. prior to project effectiveness, andwas designed to accommodate the Regional SDA data requirements. The results were emerging fromprocessing in 1990 and most analytical work took place in the Bank HQ. The publication of A PovertyProfile for Ghana, 1987- 88 provided substantial and important information for poverty analysis inthe Bank -and outside, and much poverty analysis and programming is based on this report. Ghanaianresearchers, however, point out that the database is accessible in Washington only. Processing of twosubsequent surveys in 1988/89 and 1990/91 have not yet been completed. Both have been plagued bya number of technical problems, difficulties of compatibility between the rounds, shortage of qualifiedstaff and a management who had no experience with this type of survey, and therefore could notprovide the necessary professional supervision.

In terms of capacity building in the Ghana Statistical Services, the project has not been verysuccessful. Permanent TA is supplemented by numerous trouble-shooting missions from the Bank (SDAUnit) and consultants. Throughout the project, government agencies and donors complained that theGSS did not make data available soon after processing, and that the GSS was slow to react to requestsfor specific data. The Socio-economic Analysis Unit in GSS has not been established and the StudiesFund has not been utilized for its purpose. Few outside GSS and MFEP knew about it's existence, andthe Panel for vetting proposals was never established. The frustration was clearly expressed by a BankMission in March 1991: "The GSS should not be in the business of undertaking its own policyanalysis function, until such time as it can first demonstrate it's ability to meet it's own deadlines forproviding macroeconomic and socio-economic data in a user-friendly format to MFEP and lineMinistries." The fund is now used to finance statistical trouble-shooting consultancies.

The Evaluation and Monitoring Fund was never implemented. Initially delays were caused by lackof donor funding. Switzerland withdrew its commitment in 1991 and the component was dropped.

101

5.36 PAMSCAD in time became known more as a political, palliative success but it haddesign and implementation problems that prevented very significant results from beingachieved "on the ground." One counterproductive effect of PAMSCAD may have beenthat it delayed actions by both the Government and the Bank to focus more seriously ondesigning a true poverty alleviation strategy. The SDA project was also plagued withimplementation and capacity building problems that delayed getting comparable multi-yearanalyses of the impact of adjustment on Ghana's poor, and so also delayed preparation ofthe poverty strategy.

5.37 Because of the lack of baseline data prior to the Ghana Living Standards Survey(GLSS) of 1987/88, it is difficult to know what effect the adjustment program had on thepoor in the mid-1980s. Bank staff have written that "the adjustment program in Ghanahas had a positive impact on the poor," citing the major expansion in the provision ofsocial services, particularly in health and education, plus the agricultural pricing policy andliberalization which increased farm income generally.

5.38 While data on incomes are lacking, it is evident that several quality-of-lifeindicators, inicluding child malnutrition, infant mortality, literacy and access to water allimproved since the adjustment program started. Other indicators such as primaryenrollment ratios were less changed while a few indicators, such as population per doctorand nurse, pointed to deteriorating services. On balance, hovever, most social indicatorswere moving in the right direction. Table 5.6 shows the movements of several of theseindicators over the past decade.

Table 5.6 Selected Social Indicators, 1970-92

Indicator Unit of Measure 1970-75 1980-85 1987-92

Public expenditures on basic social services % of GDP 6.3 3.2 7.4

Expenditure on social security % of total govt. expenditures 6.9 4.0 6.4

Primary School enrollment % of age group 71 76 77

Literacy % of population (age 15+) 30 53 60

Female literacy % of female (age 15+) n.a. 42 51

Infant mortality per 1000 live births 107 98 81

Life expectancy years 50 52 56

Total fertility rate births per woman 6.6 6.5 6.1

Population growth rate annual % 2.2 3.7 3.0

Energy consumption per capita kg of oil equiv. 125 78 96

Population per physician persons 12,900 14,894 22,970

Population per nurse persons 693 640 1,669

Source: Social Indicators of Development, World Bank, April 1994.

102

5.39 It was certainly a positive albeit unusual aspect of Ghana's adjustment thatgovernment expenditures rose in real terms even during the early stages of stabilization.This was facilitated by the combination of rising revenues (consequent upon the exchangerate, pricing and fiscal reforms), rising aid inflows, and the government's concern to giveadjustment a human face by improving social services. This was a "win-win"situation. AsTable 5.6 shows, public expenditures on basic social services increased significantly as ashare of GDP while expenditures on social security also rose as a proportion of totalexpenditures.

5.40 New data on poverty in Ghana, from the Extended Poverty Study (EPS) show thatbetween 1988 and 1992 the incidence of poverty in Ghana as a whole fell significantly.Some of the more important data are reported in Table 5.7 belaw. Some of the moreimportant conclusions drawn from both the EPS and a Participatory Poverty Assessmentare summarized in Box 5.7.

5.41 The decline in poverty between 1988 and 1992 was most marked in the rural areas,and especially in the poorest region, rural Savannah. But poverty has clearly become amajor problem in Accra, where the incidence of poverty tripled from 7 percent to 21percent of the population and income distribution become more unequal. The causes arestill poorly understood and warrant more research, but the consequences from a politicaleconomy standpoint are fairly obvious. It is also not fully understood why rural povertydecreased so much in a period of declined agricultural prices and productivity. But a largepart of the answer seems to be the growth in the share of income from non-farm self-employment in the poor households. And over half of that income came from serviceactivities, mainly wholesale and retail trading." The sustainability of these trends isamong the more important "sustainability issues" in Ghana.

5.42 The most common indicator for relative income inequality is the Gini coefficient.For Ghana, the indicator remained essentially unchanged between 1985 and 1992 at 40(zero is perfect equality, 100 perfect inequality), which compares favorably with levels ofinequality in Latin America and a number of African countries in the range of 40-60 butunfavorably with Asia where the range is 25-40.

Sustainability Indicators

5.43 The sustainability of Ghana's growth means different things to different people.At a project level, project designers or reviewers focus mainly on financialsustainability-as in the case of schools which will not be sustainable unless growinggovernment revenues permit growing allocations for teacher salaries and other O&Mexpenditures. For many, the central sustainability issue is the need to accelerate privatesector development, as it is recognized that the public sector led growth of the pastdecade cannot go on indefinitely. For others, sustainability is seen as an environmental

53. These findings ere reported in L. Demery and L. Squire (1995).

103

Table 5.7: Recent Trends in Poverty, 1988-92(headcount and poverty gap ratios by region)

1988 1992

Ratio Sharea Raho Share'

(%) (%)

Headcount:bGreater Accra 7 2 21 6Other urban 32 23 27 22Rural coastal 37 12 27 12Rural forest 38 31 33 32Rural Savannah 50 31 39 29

Ghana 36 100 31 100

Sub-Saharan Africa 48 48

Poverty-gap ratio:'

Greater Accra 2 1 5 6

Ghana 12 8

Sub-Saharan Africa 18 19

Squared poverty-gap (%)Ghana 6.36 5.11

a. Shares may not add to 100% owing to rounding.b. Headcount ratio: Percentage of the population whosc income or consumption is less than a pre-specired poverty line.c. Poverty-gap ratio: The gap or distance between the mean income (expenditure) of the poor and the poverty line.

The poverty line was USS31.23 per person per month at 1985 prices.

Sources: Data for Sub-Saharan Africa from World Bank (1990 and 1993a); data for Ghana from Ghana: E.aended PoverySurvey, 1995.

104

Box 5.7: Main Findings About Poverty in Ghana

* Between 1988 and 1992, the incidence of poverty in the country as a whole fell from about 36percent to 31 percent. This is clearly a major achievement.

* Almost all of the improvement comes from economic growth and the resultant increase inaverage incomes and expenditures.

* Income distribution has remained relatively stable over the period.

* Other indicators such as the depth of poverty and the food expenditure ratio also show markedimprovement.

* The gains appear to have benefitted most regions, and especially rural areas. In rural Savannah,for example, the incidence of poverty fell from 50 percent in 1988 to 39 percent in 1992. Itappears that the reduction in poverty that has occurred in rural areas may be due to increasedreliance on off-farm income.

* Urban areas other than Accra have benefitted from the economic recovery. Poverty in Accrahas, however, increased markedly (from just 7 percent in 1988 to 21 percent in 1992).

* The gender incidence of poverty also seems to have improved.

* The geographical incidence of poverty is important. Poverty is still, predominantly, a ruralphenomenon (rural areas account for more than 70 percent of national poverty). The northernrural Savannah region is significantly poorer than the national average. The Northern, UpperEast and Upper West regions are the poorest regions, followed by the Volta region. Despiteincreasing poverty, the Greater Accra region is the least poor in the country.

* Informal sector wage employees have gained most from the economic recovery, but export cropand food crop farmers, in spite of their relative poverty, have also seen an improvement inwelfare in recent years.

* The majority of the poor are food crop and export crop farmers with average per capita incomeabout one-third of national average.

* The Participatory Poverty Assessment suggests that living standards among Ghana's poorestgroups remain seriously low.

* The EPS analysis of the incidence of government expenditures suggests that the poor have sofar gained very little from social spending. The estimates show that targeting has not improved.

Source: 1995 CAS, p. 5.

105

issue-reversing trends in deforestation, soil erosion, etc. And for still others,sustainability is an issue of self sufficiency in foreign exchange, where the central questionis how fast export expansion and import substitution will proceed to enable a country torestore creditworthiness and no longer be dependent upon foreign aid.

5.44 Under the auspices of the SPA, the Africa Region developed its own set ofsustainability indicators.'4 Taking that set as a departure point, Table 5.8 presents a setof sustainability indicators for Ghana. The rationale for the choice of these variables andthe interrelationships among them are fairly self-evident, but a few Ghana-specific factorsmay be emphasized. One is that the emphasis on the growth of non-traditional exports isall-the-greater considering that some of Ghana's traditional exports are non-renewable(gold, other minerals), some are based upon already over-exploited resources (logging),and cocoa is a product with very low income and price elasticities, making it undesirablefrom a producer's standpoint to expand world production. Thus, if the long-term outlookfor traditional exports is not favorable, and the size of the domestic markets limits thescope for import substitution, there is all-the-more reason to regard the growth of non-traditional exports as a critical policy variable.

5.45 The need to regard these variables together is underlined by the fact that someof the relations among the variables are complementary, some competing. Acceleratingthe growth of private investment, for example, will be essential to get progress in all theother variables, although in the short run there may be tradeoffs between increasinggovernment revenues and providing adequate incentives to private investors. By the samnetoken, foreign aid may be supportive of private investment in the short run (as throughthe financing of infrastructure that improves the enabling environment) but couldundermine it by financing government expenditures that would crowd out, rather thancrowd in, private investment. And in both the short run and longer run, efforts to dealadequately with the "nexus issues" are essential to raise "green" savings, which in turn areneeded to finance a more adequate level of private investment.

5.46 The behavior of all the variables shown in Table 5.8 has been discussed elsewherein this report. But by pulling these key indicators together, attention is focussed on howserious is the problem of putting Ghana's development on a more sustainable path. Theseindicators point to a potential "sustainability crisis"which should command the urgentattention of government and donors alike. The indicators may also usefully serve as keylong-term target variables to be monitored at the highest levels.

54. Annex 2.3 of Special Program of Assistance: Launching the Third Phase, Africa Region, The World Bank, 1994.

106

Table 5.8: Selected Sustainability Indicators

1984-87 1988-92 1993

Savings, Investment, and Aid DependencyPrivate Investment (% of GDP) 4 7 4National Savings (% of GDP) 5 7 0.2"Genuine Savings Rate"' (% of GNP) -12 -4 -1 (1992)External Aid Dependencyb (% of GDP) 5.8 10.7 9 (1992)

Nexus Indicators

Total Fertility Rate 7 6 6Agricultural Growth Rate 3 2 3

Cocoa 7 0 4Agri., livestock fishery 3 3 3Forestry 1 3 1

Environmental IndicatorsAnnual loss from deforestation, erosion, . 4.0soil fertility (% of GDP)Deforestation rate (annual % of total 1.3forests)

Export GrowthTraditional I 1 5 20Nontraditional 32 8 -0.1

Government Revenue (% of GDP) 12 13 17Poverty/Income Distribution

Proportion of population below poverty line 36 (1988) 31 (1992)(%)Distribution Inequality Index (Gini 41 41coefficient)

a. The 'Genuine Savings Rate' is obtained by subtracting depreciation of fixed capital and depletion of naturalresources from Lhe national savings rate. See K. Hamilton and J. O'Connor (1994) 'Genuine Saving and theFinancing of Investment,' World Bank, Environment Department, mimeo.

b. Net ODA disbursements as % of GDP.

107

ANNEX TO CHAPTER 5

Charts of Projected vs. Actual Values for Selected Macroeconomic Variables

Note: The following charts compare how actual outcomes (in terms of the behavior ofsome key variables) compare with the values targeted/projected in the several CPPs andCSPs described in Chapter 2. These comparisons serve to facilitate "efficacy analysis'which focuses on the extent to which the Bank's (and Government's) policy and outputobjectives were achieved.

109

PROJECTED vs. ACTUAL VALUES(Projections from Bank Strategy Papers)

Figure 5.3: Official Assistance

700 -

Pro*W OWdal /600 - "sAwe

500 /

400-

9 300- /\t z jl V YurMNU

200-

100 ~~~~~~~~~~~~Dobunmm

198119821983198419851986198719881989199019911992

Figure 5.4: World Bank Lending Program

350

300

250-

9200 - Act/ <I150 100 -

50

019811982 1983198B419'851986 19871988 19891990 199119,9-2

110

PROJECTED vs. ACTUAL VALUES(Projecthns from Bank Strategy Papers)

Figure 5.5: Savings and Investment

20

18 - prmt GD

1 4-

12-

K0 -0

4 -Ac tual avnp

2 -

01985 1986 1987 1988 1989 1990 1991 1992

Figure 5.6: Government Accounts

25

20 Bpdm

15-

10-

l;5 Re~ R .nJrn

0*

-5O

-10- tdP*

-15- 1985 1986 1987 1988 19B9 1990 1991 1992

III

PROJECTED vs. ACTUAL VALUES(Projections from Bank Strategy Papers)

Figure 5.7: Resource and Trade Balances

200-\ Aotu

7Thdf Ewanoe

-200 -

--400

-600-

-800

-1000 1980198119821983198419851986198719881989199019911992

Figure 5.8: Current Account Balance

100-

-100

3-200-

1-300

-400 -

-500 -

-600 I

19801 9'81 198219831 9 841 9851 98619871 9881 98 91 9'901 9911 992

112

PROJECTED vs. ACTUAL VALUES(Projections from Bank Strategy Papers)

Figure 5.9: GDP Growth

10 _ -l0

6 ~~~~~~~~~~~~~~~~~~~~~6

4- \4

2 -2

0 -0

-2 -2

S -4 -4

-6 --- 6

198019811982 1983 1984 19851986619871988 1989199019911992

Figure 5.10: Growth of Agriculture

6\4

4-

2

10

-4kt1u9

-6-

81 918019I8 I119 821 9831 98 419'8 51 9861 9 871 9881 9 8919 9'01 9'9119'92

113

PROJECTED vs. ACTUAL VALUES(Projections from Bank Strategy Papers)

Figure 5.11: Growth of Industry

20 -20

15 Ata 15

10 1 0

5 5

_ 0

-5 -_5

-10 \ -I10

-15 -15

-20 -1980198119821983198419851986198719881989199019911992 -- 20

Figure 5.12: Growth of Services

20l

15-

10-

5-

0-

-5o -120- 198019 \1

1920 1o1 981 1 98219'83 1 984 19'85 198619'8719'88 1989 190 1 99 11 992

115

6. Conclusions and Recommendations

"We need longer time horizons. Even the CSP/CAS has a short-termhorizon."

Senior Bank official

"We need to define better [in our country assistance strategies] what we needand want to hold ourselves accountable for."

Bank director

Overview

6.1 In order for the Bank's assistance to Ghana in the years ahead to be as relevant,efficacious and cost-effective as possible, this report recommends that the Bank shouldrethink some of its priorities; give more attention to institutional development, economicgovernance and sustainability issues; redefine some of its comparative advantages;reformulate its performance indicators; and refine its process of country assistance strategyformulation.

6.2 The Bank's strategy and instruments were effective, in the 1980s, in helping theGhanaian Government to stabilize, liberalize and rehabilitate a moribund economy incrisis. Particularly in the mid-1980s, the Bank's program of assistance to Ghana warrantshigh ratings for the relevance of objectives, efficacy of implementation, cost effectiveness,and staff performance. The Ghana Government warrants a high rating for Borrowerperformance during those years.

6.3 In recent years, however, the performance on both sides has been more mixed.Progress has continued in getting the prices right, in rehabilitating the infrastructure, inrationalizing the revenue structure, and in public investment programming. These areareas where both the Government's and the Bank's strengths and comparative advantagescan be built upon.

6.4 Performance has however been disappointing in private sector development,privatization (at least until 1994 when more privatization activities were launched) andpublic enterprise restructuring, civil service reform, expenditure control, agriculturaldevelopment, educational achievement, environmental control, and institutionaldevelopment.

6.5 Ghana's agenda of unfinished adjustment is a long one and the prospects forsustaining a satisfactory rate of per capita growth and poverty alleviation are uncertain.Agricultural growth remains well below what is feasible (and may well be no better than

116

the rate of population growth), private savings and investment rates remain extremely low,educational testing results in the primary schools are very low, and there are seriousenvironmental problems. This being the case, proclaiming Ghana to be a "success story"may not be accurate nor in the country's best interest.

Key Strategic Issues

6.6 The objectives and instruments of the Bank's country assistance strategy for Ghanatoday are not so well adapted to current conditions as they were in the 1980s. Thehardest parts of Ghana's adjustment agenda are yet to come, and the objectives andinstruments have to be adjusted accordingly. More urgent attention needs to be given toseveral longer-term issues if the growth and gains achieved are to be sustained, oraccelerated. The binding constraints are on the institutional and managerial side.

6.7 Key issues that need to be more explicitly addressed in the Bank's country strategyare: What can the Bank do to improve the sustainability prospects? And what pace ofchange, on which key fronts, is needed to minimize the risks that Ghana's progress will notbe sustained? Most of the truly important issues (as distinguished from the urgent issuesthat tend to command attention) can be called the "sustainability issues."

6.8 Two main categories of sustainability issues are the "leading sector" and aiddependency. During the 1980s, recovery had to be public sector and foreign aid led. Aid-supported expenditures on infrastructure and social sector development can keep thegrowth going for a time. But sustained growth and poverty alleviation will depend upon afar more vigorous private sector response than has so far been achieved. Ghana's heavydependence on foreign aid also has its disadvantages, and the future strategy shouldprovide for a reduction in this dependency.

6.9 The binding constraints and the prevailing political economy are now quitedifferent from those in the 1980s, the latter being characterized by a transition towardsdemocracy following elections in 1992 of a civilian government under a new constitution.The current challenge for the Bank is how best to help Ghana manage a transition to amore sustainable path while a transition to democratic political forms and processes isunderway. The sustainability of the political and social transitions depends partly on howthe economy fares, and vice versa. The Bank's strategy must consequently be a 'politicaleconomy" strategy.

6.10 A new strategy should be based on a shared vision of what kind of society theGhanaian Government and people wish to bring about. It should be a central role of theBank to help Ghana elucidate the options, and then to forge the needed new instrumentsand institutions. If in the 1980s a main thrust of the strategy was to "get the prices right,"a main thrust for the 1990s should be in helping Ghana to "get the institutions andeconomic governance conditions right."

6.11 The new strategy also needs to accelerate the pacing and revisit the sequencing offuture policy and program actions in the light of the failure of the current strategy tofoster the needed private sector response or to raise agricultural productivity. One of the

117

report's conclusions is that unless Ghana's adjustment strategy becomes less "gradualist"than in the past decade, growth may slow down to the extent that political and socialstability would be endangered.

6.12 Two critical issues are: How to stimulate the needed private sector response? and,how to reverse the declining productivity in agriculture without harming the environment?The trends in several environmental indicators (e.g., deforestation, declining soil fertilityresulting from shortened fallows) need to be reversed.

6.13 The Government strategy (as of 1994) was still based on continued gradualism inadjustment, including civil service reform, privatization, public sector restructuring,agrarian reform, and family planning. This gradualism was facilitated, during the lastdecade, by the substantial aid provided by the donors, especially in the form of quickdisbursement policy, much of which financed current expenditure. And as noted earlier,the gradualist strategy did lead to growth although not necessarily of a sustainable nature.

6.14 But the binding constraints have changed. What is needed now is a strategy thatachieves a breakthrough in the investment climate and business environment, andimproves the confidence of private sector investors and actors. Lack of confidence nowappears to be the binding constraint to private sector development. Accelerated programsof privatization, administrative reform, and "right-sizing" of government appear to be sinequa nons of increased confidence by the private sector that Ghana's increasingly market-friendly policies are unlikely to be reversed. Accelerating adjustment will undoubtedly bedifficult in the face of the developments on the democratization front that may tend toslow decision making, but it appears as a necessary condition for sustainable growth.

6.15 The Bank has a well-established comparative advantage in helping Ghana tocorrect its distorted macroeconomic and sectoral policies and to rehabilitate itsinfrastructure. The Bank now needs to further develop and improve its capabilities, andchange its comparative advantages, to become more effective in key areas where its pastassistance has been least effective, i.e. in fostering private sector development (an area inwhich IFC needs to play an increasingly active role, including through the AfricanEnterprise Fund), in promoting institutional development and capacity building, in helpingthe government to become smaller and more efficient, and in elucidating policy analysesand options to promote widened and more enlightened public debates on key policy issuesamong constituencies formerly excluded from such debate and decision-making.

Recommendations

6.16 The main recommendations that flow from these findings and conclusions are asfollows:

118

For the Design of Bank Strategy

* Focus the strategy on sustainability, institutional development, andeconomic governance.

* Monitor performance against specified sustainability and micro-levelindicators.

* Develop a "nexus strategy" and sectoral action plans.* Make the strategy "more strategic" and longer-term.

For Achieving Objectives

* Build on Bank comparative advantages (ESW, policy dialogue, aidcoordination).

* Focus the dialogue(s) on hard choices and options.* Reach out, and listen, to broader constituencies within and outside Ghana.* Develop a more strategic approach to institutional development and

technical assistance.* hGet at the root causes of inadequate expenditure management.* Don't oversell Ghana as a success case.* Beware the downsides of aid dependency.* Develop new aid coordination policies and practices.

For Making Bank Instruments and Processes More Effective

* iGet in-house ownership of, and accountability for, the Bank's own strategy.* Put more "evaluation" into CASs.* Change the skill mix of staff to better meet the skill requirements of the

strategy.* Reassess the mix of Headquarters vs. Resident Mission assignments.

6.17 The remainder of the chapter expands upon these recommendations and theirrationale. None of the recommendations is directly contrary to the Bank's currentstrategy, and for most it may be even said that "the Bank is already doing that." But theoutcomes to date described in Chapters 3-5 suggest that additional movement in therecommended directions is still warranted.

Design of Bank Strategy

* Focus the strategy on sustainability, institutional development, andeconomic governance.

6.18 Ghana has been doing well by some criteria but not in terms of several"sustainability indicators." More attention to sustainability will itself imply giving priorityto relieving those institutional and governance constraints that are now the bindingconstraints to private sector development, which is necessary to maintain growth and toachieve the ultimate objective of poverty alleviation.

119

6.19 The Bank's current strategy focuses on portfolio implementation and private sectordevelopment. The emphasis on implementation is appropriate insofar as implementationcapacity has replaced policy distortions and resource scarcity as the binding constraint topublic investment and to the cost-effective delivery of crucial public services. So far,however, Ghana's (and the donors') approach to capacity building has emphasized thesupply side (increasing the volume of skills through more donor-financed trainingprograms) and has neglected the demand side, i.e. the institutional context and governanceconditions which determine how effectively capacities are utilized.

6.20 Ghana's underutilized human capacities and the poor "ethos of the bureaucracy"are arguably more important constraints to the effective functioning of government thanthe scarcity of skills per se. A strategic emphasis on institutional development willtherefore serve to focus attention on the binding constraints to better public sectormanagement, as well as to private sector development.

6.21 Ghana's overstaffed and poorly functioning public administration and parastatalenterprises are impediments to private sector development in several ways: throughfinancial crowding out, through the "unlevel playing field" between public and privatefirms, through de facto monopolies, through rent-seeking, obstruction and harassment bypublic officials, through the inefficient delivery of infrastructural and other public goodsand services, and through the "signal" sent that statist or populist policies and intimidatingpractices are not yet permanently relegated to the past.

6.22 This CAR therefore recommends that the Bank strategy pay more attention tothese and other "behavioral" and political economy dimensions of Ghana's development,including the factors that give rise to the values and attitudes of the principal actors onthe economic scene and how they interact with each other. The focus on institutionald:;velopment, governance and sustainability will help to ensure that these behavioraldynamics are taken into account. One example in shifting the public/private line would beto explore more "populist" approaches to privatization (e.g. vouchers, lotteries) that takeinto account Ghana's legacy of Nkrumahism and prevailing fears of change in ownership.Another example would be to analyze more explicitly the prospective winners and losersfrom the next steps in the adjustment agenda, and adjust the strategy accordingly.

6.23 Both the pacing and sequencing of future policy measures need to be analyzed in apolitical economy context. It is often said by Bank staff, for example, that Ghana'spolitical reforms, involving a greater role for the Parliament, may slow decisions andactions in the short-to-medium term. But insofar as these broaden ownership and improvethe prospects of sustainability, they will serve to accelerate development in a longer run.The Bank needs to more explicitly build the analysis of such factors into its countryassistance strategies.

* Monitor performance against sustainability and micro-level indicators

6.24 Key indicators of the sustainability of growth and poverty alleviation include:fertility rate; productivity growth and structural change in agriculture; total and privatedomestic savings rates; "genuine" savings rate; private investment ratio and annual rate of

120

private investment growth; export growth rate, especially of nontraditional exports; primaryschool enrollment rate and criterion reference test scores; and the gover.ament revenue toGDP ratio. Trends in these indicators are not encouraging.

6.25 The Bank recognizes that there is a close relationship between governance andsustainability. Hence, the sustainability indicators should include a subset of governanceindicators to complement the more traditional economic ones. Relevant governanceindicators would include: the predictability, openness and competence of policymaking; theprofessional ethos of the bureaucracy; accountability of the executive branch ofgovernment; participation of civil society in public affairs; rule of law; and incidence ofcivil disturbances. Given the low productivity in the public service and the link betweenthe "ethos" and the productivity of the civil service, the Government might wish toinstitute periodic attitude surveys of civil servants to monitor progress in developing theneeded "ethos."

6.26 The Bank's country assistance strategy needs to pay more attention to theimplications of movements in these indicators and to what the Bank and other donors cando to help improve the indicators.

6.27 This study has emphasized the importance of monitoring performance at the microas well as macro level. In education, for example, the performance of the educationalreform program appears satisfactory when measured by expenditures on education, orgrowth in enrollment. But at the micro, classroom level, recent criterion reference testresults suggest that little learning is happening in the classroom. The need for timelymicro-level monitoring should be recognized in all sectors.

6.28 The "sustainability indicators" can usefully serve both to focus the dialogue on keystrategic issues, options and trade-offs and to provide benchmarks for judging the extentto which the Bank's assistance strategy is succeeding or not in achieving its most importantobjectives.

6.29 However, because most of the sustainability indicators shown in Table 5.8 areessentially "outcome" indicators (as distinguished from "policy" or "intermediate"indicators), they are not recommended as "triggers" for the determination of Bank (orother donor) lending levels, at least not in the short run. Other indicators that monitorspecific governmental policy or programmatic actions are better suited to that purpose.

6.30 Over a medium-to-longer term, sustainability indicators may be relevant as"development effectiveness" indicators, since one might expect some correlation betweenthe level of donor assistance and the country's performance as reflected in thesustainability indicators. But the level of Bank lending should take into account manyother considerations such as the country's absorptive capacities and the levels and types offinancing available from other official and private sources.

121

Develop a "nexus strategy" and sectoral action plans

6.31 Enhanced attention to institutional development, governance and sustainabilityissues should be built into sectoral strategies. The process used in preparation of themedium-term agriculture strategy could serve as a model collaboration so as to ensuregovernment and stakeholder ownership.

6.32 The strategy should address the linkages between rapid population growth (3.0percent per annum, with a total fertility rate of 6.1), slow agricultural growth, decliningagricultural productivity, and environmental deterioration. The poor in Ghana aredisproportionately rural and located in the Northern region, where the nexus problem ismost pronounced. Dealing with the linkages is therefore critical to the poverty alleviationstrategy. It is also critical to the avoidance of tribal strife such as occurred in theNorthern region in February 1994-which appears at least partly attributable to thegrowing scarcity of fertile land and consequent tensions over land rights. The linkages or"nexus" issues have been relatively neglected over the years by both the government andthe Bank, and should be addressed in future strategies.

6.33 Ghana elected not to participate in the UNDP-financed program to assist Africancountries in developing their own national long-term perspective studies (the NLTPSprogram). This was a surprising decision. The Bank should urge the Government to getinto the NLTPS process-or something equivalent-involving debates amongstconstituencies regarding Ghana's long-term development options.

* Make the Bank's strategy "more strategic" and longer term

6.34 The Bank's current CAS papers have their origins in CPPs and CSPs that tendedtc be "advocacy papers." In those earlier documents, Regional staff and managers typicallymade a favorable case-within some bounds of credibility-for each country'saidworthiness, creditworthiness, readiness to pursue appropriate policies, and ability toabsorb and use effectively both Bank loans and non-lending resources. Projections wererequired by Bank operational directives to represent "most likely" outcomes, but inpractice the projections tended to reflect several optimistic biases; the Ghana CPPs/CSPsshared these biases.

6.35 The CAS documents of recent years are more in the nature of true strategydocuments than the early CPPs and CSPs, but they could and should be made "morestrategic" if they are to address head-on questions such as: Which long-term and short-term constraints are most binding? What are the critical "sustainability issues," and whatcan donors do about them? Which risks can be managed and which cannot? How arelessons of past experience being heeded? What "critical masses" of Bank/donor lendingand non-lending services can improve the prospects of getting significant "breakthroughs"on the binding constraints? And what are the Bank's comparative advantages in assistingGhana? Future strategy papers should focus, more so than in the past, on assessing risksand on elucidating options (especially regarding risks to be taken or avoided) for theBank's senior managers and the Board.

122

6.36 Future CASs should focus, more so than in the past, on elucidating options for theBank's senior managers and for the Board regarding t'ie various types and levels of riskthat may or may not warrant taking. This is not the same as presenting high and lowlending scenarios contingent upon the country's adopting more or fewer reforms, orimproving its absorptive capacity by more or less.

6.37 The strategy should also assess whether there are possible "critical masses" of Banklending and non-lending services that can and should be mobilized and concentrated upongetting "breakthroughs" on the country's binding constraints. Above all, sharpening thestrategic focus has to do wi,h priority setting, realism about what the Bank can accomplishin both the short-term and longer-term, defining appropriate leadership and supportingroles, and pursuing the Bank's comparative advantages vis-a-vis the Government and otherdonors.

For Achieving Objectives

0 Build on Bank comparative advantages

6.38 Many donor representatives interviewed for this study felt that the Bank's greateststrengths and its comparative advantages vis-d-vis other donors lie in the areas of ESWand ESW-based policy dialogue, "macro level" aid mobilization and coordination, andadjustment lending. There was also a majority view that the Bank has been least effectivein projects and policy dialogue involving institutional development (including free-standingtechnical assistance projects and technical assistance components) and in rural projects.These responses were broadly consistent with this report's reviews of project performanceand the efficacy of other instruments.

6.39 But in specific projects and activities "the Bank" is mainly the set of individualsthat represent the Bank at any particular time and place, and efficacy depends greatlyupon individual skills. Thus, for example, some Bank Resident mission heads were seen tohave been highly effective at policy dialogue while others were less so. Bank staff havingsimilar or identical titles come with highly varied experiences and skills. A lesson ofexperience is that great care needs to be taken to fit the right persons to the right jobs, ona case by case basis, especially in the case of Resident mission staff and above all in thecase of the head of the mission.

6.40 Insofar as the binding constraints have changed from those where the Bank hasbeen most effective to those where it has been least effective, the Bank needs to focus onways to overcome its relative weaknesses as well as to build upon its established strengths.(See also below on changing the skill mix.)

* Focus the dialogue (and related ESW) on hard choices and options

6.41 This report concludes that Ghana's unfinished agenda of reform needs to beimplemented more expeditiously in the near future. There is, however, some tensionbetween the goals of rapid implementation and Ghanaian ownership, probably bestexemplified in the area of public service reforms. This tension reflects in turn the

123

Government's multiple objectives-social and political as well as economic-objectives thatare sometimes complementary, sometime conflicting.

6.42 It follows that the Government/Bank dialogue should be focussed largely on thehard choices and options, and that Bank ESW (or more appropriately, jointgovernment/Bank ESW) should be geared to illuminating the trade-offs and elucidatingthe options. The government cannot be compelled to deepen and accelerateimplementation of its unfinished agenda of adjustment, or to greater priorities to thesustainability issues. The lesson is clear that conditionalities imposed on unwilling andunpersuaded governments are often futile, if not counterproductive. The "tension"between the needs for ownership and accelerated action needs to be addressed throughESW and dialogue geared not so much to presenting the Bank's prescriptions for action asto assessing implications of alternative course of action-including the likely adverseimplications of too-gradual action.

6.43 Thus, the Bank's response should be an active one that serves to make thediagnostics and prognostics so compelling that there emerges a shared perception ofproblems and needed solutions that will warrant Bank financial and technical support.

* Reach out, and listen, to broader constituencies within and outside Ghana

6.44 At present, there is excess demand in Ghana for information on the economy, andespecially for assessments of policy and program options/alternatives. Official documents,including those of the Bank, currently have a very restricted circulation and few people,even within the Government, have seen a PFP. This demand will grow with the transitionto democracy, with a new parliament, local assemblies and other constituencies withresponsibilities and accountabilities whose fulfillment depends on better information abouttU e economy.

6.45 The Bank should play an important role in helping to meet this demand bydisseminating reports and encouraging the Government to discuss Bank and jointly-authored reports with broader audiences. Within tight ESW resource constraints this mayrequire fewer studies and more "outreach" functions. But many Ghanaians in governmentand in the private sector indicated that what they most want is for Bank reports to discusspolicy options, and the pros and cons of those options, to inform public debate, not havingBank staff disseminate and defend pre-packaged messages.

6.46 Thus, the outreach should be a two-way street, and involve more listening by Bankstaff to views and concerns of Ghanaians, and others. Bank reports could gain by citingmore non-Bank sources, including serious critics of the Bank. Annotated bibliographicalreferences in Bank reports would be a useful way of alerting readers to this substantialnon-Bank literature.

6.47 There has been a great deal of collaboration in recent years between Bank staffand Ghanaians (academics and consultants as well as government officials) in producingmajor studies such as Ghana 2000 and Beyond and the medium-term agricultural strategy.This collaboration has generally been fruitful. Other donors have collaborated in

124

preparing Public Expenditure Reviews and other studies. This collaboration often resultsin longer production times, and sometimes in compromises that Bank staff areuncomfortable with. While no rigorous study of the costs and benefits of thesecollaborative efforts could be made, anecdotal evidence strongly suggests that suchcollaboration, although sometimes costly in the short run, is likely to be cost-effective inthe long run.

Develop a more strategic approach to Institutional development andtechnical assistance

6.48 Ghana lacks a coherent strategy for civil service and parastatal sector reform and itis a matter of highest priority for the Bank to assist in formulatCng it. The strategy shouldrecognize that the "root causes" of the problems (overstaffing, low productivity, rent-seeking behaviors, lack of motivation) may lie in the environment and governanceframeworks in which the institutions function.

6.49 A related need is for better coordination of donor policies and practices ontechnical assistance, much of which is ineffective. Attention needs to be given inparticular to minimizing the "negative externalities" of many technical assistance projects.These externalities include distortions in incentive structures due to the multiple scales ofdonor-financed salary supplements and by the non-transparent way in which technicalassistance-financed benefits (vehicles, external training, appointments to advisory groups ortask forces, etc.) are allocated. Another distorting factor has been the dominance ofshort-term, product- and event-oriented objectives, over longer-term objectives that aremore "process oriented" and geared to institutional development, capacity building andlearning-by-doing. The Bank should encourage the Government to convene a localconsultative group to focus on these issues and to come up with an agreed action plan.

* Get at the root causes of inadequate expenditure management

6.50 A major cause of slow project disbursement/implementation has been the non-availability of counterpart funds, a problem that stems more from inadequate expendituremanagement (loose correspondence between budgeted and actual expenditures due toweaknesses in Ghana's budget execution, accounting and monitoring systems) than fromrevenue shortfalls. This has been raised at successive country implementation reviews thathave concluded with assurances that the underlying problems would be addressed. At thesame time, the Bank has relaxed its requirements that the government finance increasingshares of local expenditures, over time, on several projects. This was an expedientresponse to the problem of slow project disbursement, but went against sustainability andaccountability objectives. It is recommended that no further compromises of this type bemade with the sustainability objectives.

6.51 A recent Public Expenditure Review has focussed on the expenditure controlproblems, and the government is reportedly preparing a strategy and action plan toaddress them. The Bank and other donors should give high priority to ensuring that thenecessary actions are taken without delay.

125

* Don't oversell Ghana as a success case

6.52 The Bank has often proclaimed Ghana to be a case of successful adjustment, andput it forth as an example to other African countries. This has served a useful purpose,but it may have contributed to some slackening in Ghana's own efforts from around 1989.While the Bank certainly did not intend to send a message to the authorities that theycould rest on their laurels, a substantial number of Ghanaians (as well as non-Ghanaians)interviewed suggested that the effusive praise of Ghana by the Bank had some sucheffects. This report could not attest to the validity of these perceptions. Nevertheless,given the large unfinished adjustment agenda and the challenge of making the gainssustainable, the possible costs of overselling Ghana as a success case should be borne inmind when the Bank uses Ghana as a model of success in its external relations activities.

6.53 This report does not purport to proffer specific guidelines for putting thisinjunction into practice. The main general guideline is that Bank staff and managers needto consider not only (i) the real benefits of giving credit where credit is due (as it hasbeen to the Ghanaians for their considerable achievements) and of pointing out to otherAfrican countries the benefits that adjustment has brought to Ghana; but also that (ii)there is a potential downside to giving rise to perceptions that the Bank isoverrepresenting Ghana as a success case, as that may have adverse effects on Ghana'sresolve and/or the Bank's credibility.

* Beware the downsides of aid dependency

6.54 This report has flagg.ed several aspects of Ghana's dependency on foreign aid. Atthe macroeconomic level, aid levels are high, as measured by aid as percentages of GDP,savings, or import capacity. Negative effects may include "Dutch disease," donor-drivenagendas and budgets, and distortion of behaviors. But there are no clear-cut criteria forjudging how much aid is enough, too much or too little. From the Government'sstandpoint, given that aid is available on highly concessionary terms, there is incentive tomaximize the flow, even if the rate of return on the additional resources is low.

6.55 The Revised Minimum Standard Model (RMSM) framework has been a standardtool used by Bank staff for estimating resource requirements and the "financing gap." Tothe many familiar shortcomings of using an accounting consistency model as a normativeinstrument, it may be added that estimating the "foreign exchange gap" is now less relevantto Ghana because of its floating exchange rate. Attention should be put instead on theconsolidated public sector deficit. But with big gaps in information, departures fromefficiency in expenditure, and poor correspondence between budgets and expenditures, a"resource requirement" approach to determine the "right level" of foreign aid is not veryhelpful.

6.56 The more useful approaches are micro and sectoral, and involve the identificationof ex post and ex ante rates of return on actual and prospective projects, taking account oflimited implementation capacities. Ghana has been fortunate in having enough externalfinancing resources to be able to push its absorptive capacities to the limit. But the record

126

of declining project implementation azid slow disbursement on donor projects points to theconclusion that this limit has been passed in some areis.

6.57 The project-level impacts and the "behavioral aspects" of the aid flows discussed inthis report suggest that the Bank (and other donors) should (i) define the trajectory alongwhich Ghana's high aid dependence will be reduced; and (ii) treat the dependency effectsof aid as costs, including for example costs of delays in undertaking civil service andparastatal sector reforms that might have been forced by a lesser availability of aid.Donors also need to pay more attention to the behavioral dimensions of the dependency.

* Develop new aid coordination policies and practices

6.58 This report concludes that a lack of aid coordination is more of a problem thanthe Government, the Bank, or other donors perceive it to be. Annual Bank-led CGs wereparticularly useful in the mid-1980s in providing a forum for Ghanaian and Bank advocatesof larger aid to make their case. The CGs and subsequent SPA meetings (the latter notattended by Ghanaian representatives) have remained as useful aid mobilization forums.But they do not serve much as instruments of aid coordination.

6.59 True aid coordination (i.e. coordination that focuses on improving the quality ofaid) takes place through a few local groups organized by sector; these are rated (by thedonor representatives) as having mixed success. Costs are low but benefits are hard toassess. Several donors want these groups to be made more effective, but specificproposals for how to do so seem to be lacking. The Bank could usefully play a moreproactive role in developing such proposals, without necessarily infringing upon theinclinations of other donors to playing leading roles in sectoral coordination.

6.60 Progress has been made, including through the SPA, to harmonize procedures forimport finance and to untie some aid. But there remain problems of differing donorstandards for procurement, disbursement, accounting and reporting, paying of salarysupplements, etc. Overproliferation of projects and project management units and ofdonor missions overextend the government and make it too donor-oriented--amanifestation of the dependency syndrome remarked above. These concerns and issuesadd up to an 'aid coordination problem." The sheer size of aid and the large presence ofthe donors makes aid coordination quality control important. The overarching issue is tomake the totality of resources, Ghanaian and aid-financed, as productive as possible.

For Making Bank Instruments and Processes More Effective

* Get in-house ownership of, and accountability for, the Bank's ownstrategy

6.61 One of the more surprising findings (from statements made by Bank staff andmanagers) of this study is that there was weak in-house ownership of the Bank countryassistance strategies articulated in the 1990 and 1993 CSPs. Ownership of the 1985 CPPseems to have been greater. The Ghana country team was particularly cohesive in themid-1980s but has been somewhat less so since then. The large turnover in Bank

127

managers and senior staff since the 1987 Bank reorganization has been disruptive to bothstrategic continuity and in-house ownership.

6.62 The lack of ownership of the last two CSPs stems partly from the fact that theywere prepared in the Country Operations Divisions (CODs), and reflected views andpreferences not fully shared by sectoral staff who were invited to comment on draft CSPsbut felt that their comments were not much heeded. Also, one division chief remarked ofthe 1990 CSP that "the CSP strategy was endorsed by senior management one day but thenext day it was hack to business as usual." At the time, this meant giving higher priority togetting new projects launchect than to getting ongoing projccts implemented satisfactorily.

6.63 The gap between officially sanctioned strategy and implementation reflects notonly that some staff responsible for its implementation may not "own" the strategy, butalso that there were no processes to ensure accountability at the country strategy level.The only formal mandated monitoring of actions and outcomes was the provision thateach CSP should review recenit performance in light of the previous CSP's objectives.Such review sections, however, tended to be short and perfunctory (not just in the GhanaCSPs), and so the monitoring and accountability mechanisms remained weak.

6.64 The Africa Region recognized the need to build better in-house ownership of the1994 CAS and used a more participatory process in preparing it. The Region notes that anew role of Country Teams in the formulation of CASs is now improving the proecss andproviding more sectoral and institutional focus to the strategy. It may nevertheless beuseful to conduct periodically some kind of in-house survey to assess how much of a"C'AS ownership problem" persists in the Region, and to determine what more could bedone to solve that problem. The responsibility of Bank managers and staff to adhere tothe approved strategy, and to be accountable for outcomes, is now enhanced by theBoard's role in defining the mechanisms for monitoring and accountability of CASdocuments. The Regions should, however, propose to the Board, as part of each CAS,what results from the Bank's strategy should be used as performance indicators. Theseresults would normally be in the form of movements in key indicators, e.g. of portfolioperformance, actions to improve aid coordination and ESW dissemination, etc. Theadoption of such performance indicators should not however imply a reduction in theflexibility needed by Bank managers to manage effectively.

6.65 Increased accountability for country strategies implies a need for CASs to be frankand forthright about the risks inherent in the strategy, and how it proposes that these risksbe managed-if in fact the Bank can do anything to manage the risks. This path wouldallow the Board discussion of CASs to focus on which risks the Board thinks the Bankshoald take, how risks can be minimized or managed, and, given the riskss, what should theRegions be accountable for delivering.

* Put more evaluation into CASs

6.66 A corollary of the above recommendation on accountability for Bank strategyimplementation is that CASs should contain a more substantial section on self evaluationof the Bank's successes and failures in achieving the targets/objectives of the previous

128

strategy paper. The section should make clear which assumptions underlying the paststrategy paper proved right or wrong (and the reasons) and how lessons derived fromevaluating the past strategy have been applied to the proposed current/future strategy. Inaddition, it would be desirable for the Bank and the Government to conduct ex-postevaluations of the CAS in a similar way to that of project evaluations (throughImplementation Completion Reports).

* Change the skill mix of staff to better meet the skill requirements of thestrategy

6.67 The strategy shift of the 1990s and the changed political economy in Ghana implya need to change the skill mix of those responsible for designing and implementing thestrategy. The appropriate mix of skills and experience to enter into dialogue with a fewtechnocrats at the center of government about exchange auction systems in the mid-1980sis not the same required to enter into dialogue with tribal elders in rural areas about local-level governance and institutional issues. Understanding the systemic problems that hinderthe effectiveness of institutional development/technical assistance projects, and theinteractions of interest groups involved in private sector development, is a challengingtask. And it takes special skills to engage in participatory poverty assessments and todialogue with new parliamentarians about options for solving "nexus" problems. Yet theseare among the skills that will be needed to deliver what the new challenges require.Recruitment, retrenchment and retooling criteria need to be adjusted accordingly.

6.68 If the strategy were to be focussed especially on the institutional development,govemance and sustainability issues as recommended, this would put a premium on staffand consultants with considerable field experience, private sector or NGO experience,private sectoral managerial experience, and institutional development skills. The increasedemphasis on dissemination, outreach and listening would also call for staff and consultantswith attributes relevant to performing those functions most effectively.

* Reassess the mix of Headquarter vs. Resident Mission assignments

6.69 The above strategy would constitute a significant change in the Bank's modusoperandi, with a premium put on field skills and understandings based upon fieldexperience. The new priorities would also seem to call for a shift in favor of staff in theResident Mission vis-a-vis Headquarters.

129

Bibliography

Abbey, J.L.S. (1989). On Promoting Successful Adjustment: Some Lessonsfrom Ghana. Per Jacobsson Foundation.

Alderman, H. (1991). "Downturn and Economy Recovery in Ghana: Impacts on the Poor."Ithaca, NY: Cornell Food and Nutrition Policy Program, Monograph 10.

Anyemedu, Kwasi. (1991). "Export Diversification Under the Economic Recovery Program".In Ghana: The Political Economy of Recovery. Ed. Donald Rothchild. Boulder, CO:Lynne Rienner.

Aron, Janine and Ibrahim Elbadawi. (1993). Some Lessons for the Design andMicro-Management of Foreign Exchange Auctions from Sub-Saharan Africa.Washington, D.C.

Aryeetey, Ernest. (1991). "The Relationship Between Formal and Informal FinancialSectors." African Economic Research Consortium, Research Paper, Nairobi.

-----. (1994). "Private Investment under Uncertainty in Ghana." World Development, vol.22, no. 8.

Bahai, Jagdish et al. (1993). Strengthening Local Initiative and Building Local Capacity inGhana. Washington, DC: World Bank.

Bates, Robert H. (1981). Markets and States in Tropical Africa. Berkeley: University ofCalifornia.

------ (1991). "Agriculturai Policy and the Study of Politics in Post-Independence Africa."In Africa 30 Years On. Ed. Douglas Rimmer. London: Royal African Society andJames Currey.

Bheenick, R. et al. (1989). "Successful Development in Africa: Case Studies of Projects,Programs, and Policies." EDI Development Policy Case Series, Analytical CaseStudies no. 1. Washington, D.C.: World Bank.

Boateng, E. Oti et al. (1990). "A Poverty Profile for Ghana, 1987-88," Social Dimensions ofAdjustment in Sub-Saharan Africa Working Paper no. 5. Washington, D.C.: WorldBank.

Branson, W. (1994). "Macro and Trade Issues for Ghana CAR," Background Paper. TheWorld Bank, June.

130

Brunetti, A. and B. Weder. (1995). "Credible Politics and Economic Girowth: A Cross-national Analysis." Mimeo, FHarvard University.

Bruton, H. (1993). "Asian Lessons for African Development," Williams Collcge, Center forDevelopment Economics, Research Memorandum Series, RM-136, October.

Callaghy, Thomas M. (1989). "Lost Between State and Market: The Politics of EconomicAdjustment in Ghana, Zambia and Nigeria." Paper presented to World Banksymposium (mimeo).

-and John Ravenhill (eds.). (1991). FHemmed In: Responses to Africa's EconomicDilemma. New York: Columbia UP.

Centre for the Study of African Economies, University of Oxford and Department ofEconomics, University of Ghana. (1992). "Economic Reform and the ManufacturingSector in Ghana." Country Background Paper for the Africa Regional Program onEnterprise Development. Washington, D.C.: World Bank.

Chand, Sheetal K. and Reinold van Til. (1988). Ghana: Towards Successful Stabilization andRecovery." Finance and Development, March.

Chazan, Naomi. (1982). "Ethnicity and Politics in Ghana." Political Science Quarterly. Fall

-. (1983). An Anatomy of Ghanaian Politics: Managing Political Recession, 1969-1982.Boulder, CO: Westview.

----. (1988). "Ghana: Problems of Governance and the Emergence of Civil Society." InDemocracy in Developing Countries: Africa. Ed. Larry Diamond, Juan Linz, andSeymour Lipset. Boulder, CO: Lynne Rienner.

-----. (1991). "The Political Transformation of Ghana Under the PNDC." In Ghana: ThePolitical Economy of Recovery. Ed. Donald Rothchild. Boulder, CO: Lynne Rienner.

-----, and Donald Rothchild (eds.). (1988). The Precarious Balance: State and Society inAfrica. Boulder, CO: Westview.

Chhibber, Ajay and Nemat Shafik. (1990). "Exchange Reform, Parallel Markets, andInflation in Africa: The Case of Ghana," PRE Working Paper no. 427. Washington:World Bank.

Cleaver, Kevin. (1993). "A Strategy to Develop Agriculture in Sub-Saharan AFrica and aFocus for the World Bank." Technical Paper no. 203, Africa Technical DepartmentSeries. Washington, D.C.: World Bank.

-----. and G. Schreiber. (1994). Reversing the Spiral. The Population, Agricultural, andEnvironment Nexus in Sub-Saharan Africa. Direction in Development Series.Washington, D.C.: World Bank.

131

Cobbe, James. (1991). "The Political Economy of Education Reform in Ghana." In Ghana:The Political Economy of Recoverv. Ed. Donald Rothchild. Boulder, CO: LynneRienner.

Cornia, Giovanni, Richard Jolly and Francis Stewart (eds.). (1987). "Ghana." In Adiustmentwith a Human Face. Oxford: Clarendon Press. vol. 2.

de Merode, Louis. (1991). "Civil Service Pay & Employment Reform in Africa: SelectedImplementation Experiences." Washington: WB Africa Technical Department,Institutional Development and Management Division Study Paper no. 2.

Demery, L. and L. Squire. (1995). "Poverty in Africa: An Emerging Picture." Mimeo, TheWorld Bank.

Denoon, David. (1986). Devaluation Under Pressure: India. Indonesia, and Ghana.Cambridge: MIT.

Elliot Berg Associates. (1990). "PFP Study: Case Study - Ghana." A Review of the PolicyFramework. Paper Process. Alexandria, VA.

Friends of the Earth, The. (1992). Plunder in Ghana's Rain Forest for Illegal Profit.London: Friends of the Earth Ltd.

Frimpong-Ansah, Jonathan. (1991). The Vampire State in Africa: The Political Economyof Decline in Africa. London: James Currey.

Government of Ghana. (1991a). "Ghana: Technical Cooperation Policy," NATCAP,MFEEP, April.

Government of Ghana Documents: See separate list.

Green, Reginald. (1987). Ghana. Helsinki: World Institute for Development EconomicsResearch.

-----. (1988). "Ghana: Progress, Problematics and Limitations of the Success Story," IDSBulletin. January.

Gyimah-Boadi, E. (1991). "State Enterprises Divestiture: Recent Ghanaian Experiences."In Ghana: The Political Economy of Recovery. Ed. Donald Rothchild. Boulder, CO:Lynne Rienner.

Hansen, Immanuel and Kwame Ninsin (eds.). (1989). The State, Development and Politicsin Ghana. London: CODESRIA Book Series.

Haynes, Jeffrey. (1989). "Ghana: Indebtedness, Recovery and the IMF, 1977-1987." In TheAfrican Debt Crisis. Eds. Trevor W. Parfitt and Stephen P. Riley. London:Routledge.

132

Helleiner, G.K (1986). "Outward Orientation, Import Instability and African EconomicGrowth: An Empirical Investigation" In Theory and Reality in Development. Eds.Sanjaya Lall and Frances Stewart. London: Macmillan.

Herbst, Jeffrey. (1993). The Politics of Reform in Ghana, 1982-1991. Berkeley: Universityof California.

Hopkins, AG. (1986). "The World Bank in Africa: Historical Reflections on the AfricanPresent." World Development 14(12), December.

Huq, M.M. (1989). The Economy of Ghana: The First 25 Years Since Independence.London: McMillan.

Husain, I. (1994). "Why Do Some Economics Adjust More Successfully Than Others?Lessons from Seven African Countries." Policy Research Working Paper no. 1364.Africa Regional Office, Office of the Chief Economist. Washington, D.C.: WorldBank.

------ and R. Faruquee (eds.). (1994). Adjustment in Africa: Lessons from Country CaseStudies. World Bank Regional and Sectoral Studies. Washington, D.C.: World Bank.

Hyden, Goran. (1986). "African Social Structure and Economic Development." In Strategiesfor African Development. R.J. Berg and J.S. Whitaker. Berkeley, CA: University ofCalifornia.

Jebuni, C., S. Asuming-Brempong and KY. Fosu. (1990). "The Impact of the EconomicRecovery Programme on Agriculture in Ghana." Report Submitted to the UnitedStates Agency for International Development, Accra.

----- -A Oduro and KA. Tutu. (1991). "Trade and Payments Liberalization and EconomicPerformance in Ghana." Paper Presented at the African Economic ResearchConsortium Workshop, Nairobi.

-----, N.K Sowa and KA. Tutu. (1991). "Exchange Rate Policy and MacroeconomicPerformance in Ghana," AERC Research Paper, Nairobi.

Jeffries, Richard. (1978). Class Power and Ideology in Ghana: The Railwaymen of Sekondi.Cambridge: Cambridge UP.

-----. (1991). 'Leadership Commitment and Political Opposition to Structural Adjustmentin Ghana." In Ghana: The Political Economy of Recovery. Ed. Donald Rothchild.Boulder, CO: Lynne Rienner.

-----, and Clare Thomas. (1993). "The Ghanaian Election of 1992." African Affairs. vol 92:331-366.

133

Kanbur, Ravi. (1993). "The Economic Reform Program and the Poor in Ghana,"Unpublished note (3pp). Accra: World Bank Resident Mission.

-----. (1994). "Welfare Economics, Political Economy, and Policy Reform in Ghana." WBPolicy Research Working Paper no. 1381. Washington, D.C.: World Bank.

Kapur, Ishan et al. (1991). "Ghana: Adjustment and Growth, 1983-91." IMF OccasionalPaper no. 86. Washington, D.C.: IMF.

Killick, Tony. (1978). Development Economics in Action: A Study of Economic Policies inGhana. New York: St. Martin's.

-----. (1991). "The Developmental Effectiveness of Aid to Africa." In African ExternalFinance in the 1990s. Eds. I. Husain and J. Underwood. Washington, D.C.: WorldBank.

-----. (1992). "Explaining Africa's Post-Independence Development Experiences." ODIWorking Paper 60. London: Overseas Development i'nstitute.

-----. (1993a). "issues in the Design of IMF Programmes." ODI Working Paper 71.London: Overseas Development Institute.

----. (1993b). "Economic Inflexibility in Africa: Evidence and Causes." ODI Working Paper73. London: Overseas Development Institute.

Kraus, Jon. (1991). "The Political Economy of Stabilization and Structural Adjustment inGhana." The Political Economy of Recovery. Ed. Donald Rothchild. Boulder, CO:Lynne Rienner Publishers.

Lele, Uma (ed.). (1992). Aid to African Agriculture. Baltimore, MD: Johns Hopkins UP.

Lewis, John (ed.). (1988). Strengthening the Poor: What Have We Learned? Washington,D.C.: Overseas Development Council (see chapter by Jolley on "Poverty andAdjustmerit in the 1990s" re PAMSCAD).

Lewis, W.A. (1953). "Report on Industrialization and the Gold Coast." Accra: GovernmentPrinter.

Lopez, R. (1993). "Economywide Policies, Agricultural Productivity, and EnvironmentalFactors: The Case of Ghana." Pollution and Environmental Economics Division.Environment Department. Washington, D.C.: World Bank.

Loxley, John. (1988). Ghana: Economic Crisis and the Long Road to Recovery. Ottawa:North-South Institute.

134

Martin, Matthew. (1991). "Neither Phoenix nor Icarus: Negotiating IMF Programmes inGhana and Zambia, 1983-1989." In Hemmed In: Responses to Africa's EconomicDilemma, (1993). Eds. Thomas Callaghy and John Ravenhill. New York: ColumbiaUP.

May, Ernesto. (1985). "Exchange Controls and Parallel Market Economies in Sub-SaharanAfrica: Focus on Ghana," WB Staff Working Paper no. 71. Washington, D.C.: WorldBank.

Meerman, J. (1994). "Agriculture: Ghana." Background Paper, OED.

Meier, Gerald and William Steel (eds.). (1989). Industrial Adjustment in Sub-SaharanAfrica. New York: Oxford UP (see chapter on "Removing Price Controls in Ghana").

Mikell, Gwendolyn. (1989). Cocoa and Chaos in Ghana. New York: ParagonHouse.

Mosley, Paul, Jane Harrigan and John Toye. (1991). Aid and Power: The World Bank andPolicy-Based Lending. (2 vols). New York: Routledge.

------ and John Weeks. (1993). "Has Recovery Begun? 'Africa's Adjustment in the 1980s'Revisited." World Development, vol. 21, no. 10.

Nelson, Joan M. (ed.). (1990). Economic Crisis and Policy Choice. Princeton: Princeton UP.

Norton, Andrew and Ellen Bortei-Dorku. (August 1993). "Ghana Participatory PovertyAssessment." Draft Report.

Nugent, Paul. (1991). "Educating Rawlings: The Evolution of Government." In Ghana: ThePolitical Economy of Recovery. Ed. Donald Rothchild. Boulder, CO: LynneRienner.

Paul, Samuel. (1989). "Private Sector Assessment: A Pilot Exercise in Ghana." WB PPRWorking Paper no. 199. Washington, D.C.: World Bank.

Pinto, Brian. (1989). "Black Market Premia, Exchange Rate Unification and Inflation inSub-Saharan Africa." The World Bank Economic Review 3(3): 321-338.

Ray, Donald. (1986). Ghana: Politics, Economics and Society. Boulder: Lynne Rienner.

Ridker, R. (1994). "Education Reform Program, Bank Inputs and Progress to date."Background Paper, OED.

Rimmer, Douglas (1992). Staying Poor: Ghana's Political Economy. 1950-1990. New York:Pergamon.

135

Roe, Alan and Hartmut Schneider. (1992). "Adjustment and Equity in Ghana."Paris: OECD Development Centre Studies.

Rothchild, Donald. (1991). "Ghana and Structural Adjustment: An Overview." In Ghana:The Political Economy of Recovery. Ed. Donald Rothchild. Boulder, CO: LynneRienner Publishers.

-----, and E. Gyimah-Boadi. (1986). "Ghana's Economic Decline and DevelopmentStrategies." In Africa in Economic Crisis. Ed. Jofln Ravenhill. New York: ColumbiaUP.

Sandbrook, Richard, with Judith Barker. (1985). The Politics of Africa's EconomicStagnation. Cambridge: Cambridge UP.

Sarris, Alexander. (1992). "Household Welfare During Crisis and Adjustment in Ghana,"Mimeographed. Ithaca, NY: Cornell University Food and Nutrition Policy Program.

Schadler, Susan et al. (1993). "Economic Adjustment in Low-income Countries: ExperienceUnder the Enhanced Structural Adjustment Facility." IMF Occasional Paper no. 106.Washington, D.C.: IMF.

Sowa, N.K, A. Baah-Nuakoh, KA. Tutu and B. Osei. (1992). "Small Enterprises andAdjustment: The Impact of Ghana's Economic Recovery Programme." ODI ResearchReport. London: Overseas Development Institute.

Steel, William and Leila Webster. (1990). "Small Enterprises in Ghana: Responses toAdjustment." WB Industry and Energy Working Paper no. 33. Washington, D.C.:World Bank.

-----. (1992). "How Small Enterprises in Ghana Have Responded to Adjustment." TheWorld Bank Economic Review, vol. 6, no. 3.

Stewatt, Frances, Sanjaya Lall and Samuel Wangwe. (1992). "Altemative Strategies: AnOvervievA." In Alternative Strategies in Sub-Saharan Africa. Eds. Frances Stewart,Sanjaya Lall and Samuel Wangwe. New York: St. Martin's.

Stryker, J. Dirck et al. (1990). Trade, Exchange Rate and Agricultural Pricing Policies inGhana. Washington, D.C.: World Bank.

Svedberg, Peter. (1991). "The Export Performance of Sub-Saharan Africa." Trade andDevelopment in Sub-Saharan Africa. Eds. Jonathan Frimpong-Ansah, Ravi Kanburand Peter Svedberg. Manchester: Manchester UP.

Toye, J. (1991). "Ghana." Aid and Power, vol. 2. Eds. P. Mosley et al. New York:Routledge.

136

UNDP. (1989). "National Technical Cooperation Assessment and Programmes: Principlesand Procedures." RBA, February.

World Bank. (1994a). Adjustment in Africa: Reforms, Results, and the Road Ahead.World Bank Policy Research Report. Oxford: Oxford UP.

World Bank. (1994b). "Status Report on Poverty in Sub-Saharan Africa 1994: The ManyFaces of Poverty." Human Resources and Poverty Division. Africa TechnicalDepartment. Washington, D.C.

Younger, S.D. (1992). "Aid and the Dutch Disease: Macroeconomic Management WhenEverybody Loves You." World Development, vol. 20, no. 12 (November).

137

World Bank Economic and Sector Reports & Publications, 1983-94

Report # Type Date Tiile

13423 SEC 12/29/94 Financial sector review: Bringing savers and investors13567 SEC 12/22/94 Urban development strategy reviewWPS1395 WPS 12/01/94 A typology of foreign auction markets in Sub-Saharan

AfricaWPS1396 WPS 12/01/94 Foreign exchange auction markets in Sub-Saharan

Africa: Dynamic model for auction exchange ratesT6320 TAN 11/02/94 Private Sector Development ProjectWPS1381 WPS 11/01/94 Welfare economics, political economy, and policy

reform in GhanaWPS1364 WPS 10/01/94 Why do some economies adjust more successfully than

others? Lessons from seven African countries13661 INT 09/01/94 Staff loss and retention at selected African

universities: A synthesis report10816 SEC 12/17/93 Energy sector reviewWPS1212 WPS 10/01/93 The links between economic policy and research:

Three examples from Ghana and some generalthoughts

11321 SEC 06/01/93 Private sector assessment11369 SEC 02/01/93 Strengthening local initiative and building local

capacityESM148 INT 11/01/92 Industrial energy efficiency activity, phase 111486 ECO 11/01/92 2000 and beyond: Setting the stage for accelerated

growth and poverty reductionWPS1009 WPS 10/01/92 Labor and women's nutrition: A study of energy

expenditure, fertility, and nutritional status in Ghana11174 JRN 09/01/92 Small enterprises in Ghana: Responses to adjustment11322 INT 08/01/92 Formal and informal finance for small enterprises in

Ghana10993 INT 06/01/92 Human resource development and economic growth:

Ghana in the next two decades10888 JRN 05/01/92 The economics of farm fragmentation: Evidence from

Ghana and Rwanda11100 INT 05/01/92 Household demand for improved sanitation services:

A case study of Kumasi, GhanaWPS884 WPS 04/01/92 Intercommodity price transmittal: Analysis of food

markets in GhanaWPS792 WPS 10/01/91 Reducing labor redundancy in state owned enterprises10048 SEC 10/01/91 Public enterprise sector reviewWPS773 WPS 09/01/91 Are Ghana's roads paying their way? Assessing road

use cost and user charges in Ghana

138

Report # Type Date Title

8914 SEC 06/01/91 Medium term agricultural development strategy(MTADS): An agenda for sustained growth anddevelopment (1991-2000)

WPS672 WPS 05/01/91 The macroeconomics of public sector deficits: Thecase of Ghana

9475 ECO 04/01/91 Progress on adjustment8911 SEC 10/01/90 Towards a dynamic investment response9220 INT 09/01/90 Small enterprises in Ghana: Responses to adjustment8099 SEC 06/01/90 Housing sector review: sustaining the structi ral

adjustment program: A shelter strategy8616 ECO 06/01/90 Public expenditure review: 1990-92WPS429 WPS 06/01/90 Ghana's cocoa pricing policyWPS427 WPS 05/01/90 Exchange reform, parallel markets, and inflation in

Africa: The case of Ghana8221 SEC 12/01/89 Fiscal decentralization7673 ECO 10/01/89 Public expenditure review: 1989-91WPS287 WPS 10/01/89 The consistency of government deficits with

macroeconomic adjustment: An application to Kenyaand Ghana

INU51 INT 10/01/89 Cost and benefits of rent control in Kumasi, GhanaWPS365 WPS 10/01/898944 JRN 09/01/89 Black market premia, exchange rate unification, and

inflation in Sub-Saharan AfricaWPS198 WPS 05/01/89 Equity in unequal deductions: Implications of income

tax rules in Ghana and NigeriaWPS199 WPS 05/01/89 Assessment of the private sector: A case study and its

methodological implications7593 SEC 03/01/89 Basic education for self-employment and rural

development: A sub-sector study7597 SEC 03/01/89 Population, health and nutrition sector review7515 ECO 01/01/89 Structural adjustment for growthESM74 INT 10/01/88 Sawmill residues utilization study7384 SEC 07/01/88 Reviving the urban sector7307 SEC 06/01/88 Prospects for selected export productsESM84 INT 06/01/88 Energy rationalization in the industrial sector6817 SEC 06/01/87 Forestry sector review6635 ECO 03/01/87 Policies and issues of structural adjustment6173 SEC 12/01/86 Irrigation subsector reviewDRD215 INT 12/01/86 The political economy of industrialization in primary

product exporting economies: Some cautionary tales6234 SEC 11/01/86 Issues and options in the energy sector

139

Report # Type Date Title

5854 ECO 10/01/85 Towards structural adjustmentCPD8556 INT 10/01/85 Fiscal dcficits, inflationary expectations and parallel

exchange markets in Ghana: Monetarism in thetropics?

5716 SEC 09/01/85 Industrial policy, performance and recovery5366 SEC 08/01/85 Agricultural sector review5824 ECO 08/01/85 Priorities for public expenditures, 1986-885737 SEC 06/01/85 Transport sector strategy note5289 ECO 11/01/84 Managing the transitionCPD8434 INT 08/01/84 Exchange controls and the parallel market economies

in Sub-Saharan AfricaUDD45 INT 11/01/83 Asante culture and low - income housing policies: An

examination of antithesis4702 ECO 10/01/83 Policies and program for adjustment

Codes: SEC= Sector report; ECU= Economic report; WPS= Working paper; JRN= Journal paper, INT= Internal paper

140

OED Documents Concerning Projects in Ghana

Report # Type Date Title

13818 PCR 12/29/94 Second Oil Palm Development Project13575 PPAR 09/30/94 First Education Sector Adjustment Credit13314 PPAR 06/30/94 Road Rehabilitation and Maintenance Project13262 PPAR 06/30/94 Structural Adjustment Institutional Support

Project13240 PPAR 06/29/94 Health and Education Rehabilitation Project13211 PPAR 06/29/94 The Seventh and Eighth Power Projects13287 PPAR 06/29/94 Accra District Rehabilitation Project13286 PPAR 06/29/94 Water Supply Projects13211 PPAR 06/29/94 The Seventh and Eighth Power Projects13149 PCR 06/14/94 Port Rehabilitation Project12622 PCR 12/22/93 Education Sector Adjustment Project12555 PCR 11/29/93 Priority Works Project12502 PCR 11/15/93 Structural Adjustment Institutional Support

Project12414 PCR 10/01/93 Northern Grid Extension Project12225 PCR 07/01/93 Accra District Rehabilitat;on Project12141 IER 06/01/93 Kpong Hydroelectric Project12137 PPAR 06/01/93 Industrial Sector Adjustment Credit Project11844 PCR 04/01/93 Industrial Sector Adjustment Credit11528 PCR 01/01/93 Energy Project11498 PCR 12/01/92 Power System Rehabilitation Project11080 PCR 08/01/92 Health and Education Rehabilitation Project10747 PCR 06/01/92 Road Rehabilitation and Maintenance Project10686 PPAR 05/01/92 First and Second Structural Adjustment

Program Project10240 PCR 12/01/91 Petroleum Refinery Rehabilitation and

Technical Assistance Project9732 PCR 06/01/91 Water Supply Technical Assistance and

Rehabilitation Project9580 PPAR 05/01/91 First and Second Reconstruction Import

Credit and Export Rehabilitation Projects9311 PCR 01/25/91 Railway Rehabilitation Project8765 PCR 06/01/90 Volta Agricultural Development Project8717 PCR 06/01/90 Second National Investment Bank (NIB)

Project7328 PPAR 06/01/88 CIMAO Regional Clinker Projects - Togo -

Ivory Coast6946 PCR 09/01/87 Third Highway (Emergency Maintenance)

Project6796 PCR 05/01/87 National Investment Bank Project

141

Report # Type Date Title

6755 PCR 05/01/87 Upper Region Agricultural DevelopmentProject

6400 PPAR 09/01/86 Telecommunications Project5731 PPAR 06/01/85 Power Projects5189 PPAR 06/01/84 First and Second Highway Projects5105 PPAR 05/01/84 Oil Palm Project5092 PCR 05/01/84 Ashanti Region Cocoa Project4600 PCR 06/01/83 Accra - Tema Water Supply (Stage Two)

Project3972 PPAR 06/01/82 Livestock Development Project3526 PPAR 06/01/81 Cocoa Project3525 PPAR 06/01/81 Sugar Rehabilitation Project2794 PPAR 12/01/79 Fisheries Proje(;t1683 PPAR 07/01/77 Accra Tema Water Supply and Sewerage

Project1568 PPAR 04/01/77 First and Second Power Distribution Projects1363 PPAR 11/01/76 Volta River Hydroelectric Expansion Project

142

World Bank Project Documents on Ghana, 1983-94

Report # Type Date Title

P6518 MOP 01/19/95 Mining Sector Development and EnvironmentProject

13881 SAR 01/19/95P6218 MOP 01/18/95 Thermal Power Project13877 SAR 01/18/95 Fisheries Sub - Sector Capacity Building ProjectP6516 MOP 01/18/95P6471 MOP 11/15/94 Vocational Skills and Informal Sector Support

Project13691 SAR 11/15/94P6320 PR 11/02/94 Private Sector Development ProjectP6180 MOP 03/23/94 Community Water and Sanitation Project12406 SAR 03/21/9412252 SAR 01/18/94 Thermal Power Project12332 SAR 01/01/94 Local Government Development ProjectP6164 MOP 01/01/94P6128 MOP 11/10/93 Agricultural Sector Investment Project12222 SAR 11/10/9311760 SAR 05/01/93 Primary School Development ProjectP6045 MOP 05/01/93P6019 MOP 04/01/93 Urban Transport Project1674 SAR 04/01/93P6004 MOP 04/01/93 Private Enterprise and Export Development Project11607 SAR 04/01/9311256 SAR 02/01/93 National Electrification ProjectP5899 MOP 02/01/9311058 SAR 11/01/92 National Livestock Services ProjectP5705 MOP 11/01/9210691 SAR 09/01/92 Environmental Resource Management ProjectP5315 MOP 09/01/92P5840 MOP 09/01/92 Tertiary Education Project10869 SAR 09/01/92P5627 MOP 02/01/92 National Agricultural Extension Project9847 SAR 02/01/92P5523 PR 02/01/92 Agricultural Sector Adjustment ProgramP5706 MOP 02/01/92 Literacy arid Functional Skills Project10164 SAR 02/01/92P5643 MOP 11/01/91 National Feeder Roads Rehabilitation and

Maintenance Project9823 SAR 11/01/91P5659 PR 11/01/91 Second Financial Sector Adjustment Program9313 SAR 05/01/91 National Agricultural Research Project

143

Report # Type Date Title

P5471 MOP 05/01/91P5556 MOP 05/01/91 Community Secondary Schools Construction

Project9556 SAR 05/01/91P5499 PR 04/01/91 Program To Promote Private Investment and

Sustained Development ProjectP5446 PR 03/01/91 Economic Management Support Project8945 SAR 11/05/90 Second Transport Rehabilitation ProjectP5373 MOP 11/01/90P5385 MOP 11/01/90 Second Health and Population Project9029 SAR 11/01/90P5168 MOP 09/01/90 Agricultural Diversification Project8133 SAR 09/01/908331 SAR 05/09/90 Second Urban ProjectP5314 MOP 05/01/90P5244 PR 04/01/90 Second Education Sector Adjustment Credit8207 SAR 02/06/90 Sixth Power Project (Volta River Authority)P5193 MOP 02/01/90P4869 MOP 07/01/89 Fifth Power Project7405 SAR 07/01/897598 SAR 05/01/89 Water Sector Rehabilitation ProjectP4972 MOP 05/01/897554 SAR 04/01/89 Rural Finance ProjectP4965 MOP 04/01/89P4893 MOP 03/01/89 Private Small and Medium Enterprise

Development Project7440 SAR 03/01/89P4915 PR 02/01/89 Second Phase of the Structural Adjustment

ProgramP4836 MOP 11/01/88 Forest Resource Management Project7295 SAR 11/01/88P4768 MOP 06/01/88 Second Telecommunications Project7163 SAR 06/01/88P4814 MOP 05/01/88 Mining Sector Rehabilitation Project7039 SAR 05/01/88P4781 PR 05/01/88 Financial Sector Adjustment Project7000 SAR 12/01/87 Priority Works ProjectP4697 MOP 12/01/876912 SAR 11/01/87 Transport Rehabilitation ProjectP4621 MOP 11/01/87P4631 MOP 10/01/87 Cocoa Rehabilitation Project

144

Report # Type Date Title

6818 SAR 10/01/87P4606 MOP 08/01/87 Public Enterprise Project6881 SAR 08/01/876628 SAR 05/01/87 Petroleum Refining and Distribution ProjectP4564 MOP 05/01/87P4538 MOP 04/01/87 Agricultural Services Rehabilitation Project6645 SAR 04/01/87P4403 PR 03/01/87 Structural Adjustment Program Project6693 SAR 03/01/87 Structural Adjustment Institutional Support

ProjectP4493 MOP 03/01/87P4370 MOP 01/01/87 Northern Grid Extension Project6301 SAR 01/01/87P4311 PR 11/01/86 Education Sector Adjustment ProjectP4415 MOP 11/01/86 Structural Adjustment Technical Assistance

ProjectP4248 MOP 03/01/86 Ports Rehabilitation Project5907 SAR 03/01/86P4217 MOP 02/01/86 Industrial Sector Adjustment Credit ProjectP4157 PR 12/01/85 Health and Education Rehabilitation ProjectP4146 MOP 09/01/85 Power System Rehabilitation Project4932 SAR 08/01/855479 SAR 05/01/85 Road Rehabilitation and Maintenance ProjectP4076 MOP 05/01/85P3987 PR 03/01/85 Second Reconstruction Imports Credit Project5327 SAR 02/01/85 Accra District Rehabilitation ProjectP3896 MOP 02/01/853724 SAR 05/01/84 Second Oil Palm Development ProjectP3825 MOP 05/01/84P3725 MOP 02/01/84 Petroleum Refinery Rehabilitation and Technical

Assistance ProjectP3695 PR 12/01/83 Export Rehabilitation ProjectP3696 MOP 12/01/83 Export Rehabilitation Technical Assistance

ProjectP3554 PR 06/01/83 Reconstruction Import Credit ProjectP3191 PR 05/01/83 Energy ProjectP3479 MOP 03/01/83 Water Supply Rehabilitation Project and

Technical AssistanceP3261 PR 01/01/83 CIMAO Restructuring Project - Togo / Ivory

CoastP3180 PR 12/01/81 Public Administration ProjectP3088 MOP 06/01/81 Railway Rehabilitation Project

145

Report # Type Date Title

2326 SAR 05/01/81P2716 PR 05/01/80 Third Highway (Emergency Maintenance) ProjectP2762 MOP 04/01/80 Volta Agricultural Development Project2527 SAR 03/01/80P2518 MOP 04/01/79 Second National Investment Bank (NIB) Project1743 SAR 04/01n92117 SAR 06/01/78 Feeder Roads ProjectP1971 MOP 03/01n7 Kpong Hydroelectric Project1299 SAR 03/01/771196 SAR 03/01n7 Third Power ProjectP 1972 MOP 03/01/771061 SAR 06/01/76 Upper Region Agricultural Development ProjectP1868 MOP 06/01/761071 SAR 06/01n6 CIMAO Regional Clinker Project - Togo - Ivory

CoastP1812 MOP 06/01/76P1687 MOP 12/01/75 Ashanti Region Cocoa ProjectP1703 MOP 12/01/75 Second Highway Project903 SAR 12/01/75P1718 MOP 11/01/75 National Investment Bank Project880 SAR 11/01/75827 SAR 11/01/75 Ashanti Region Cocoa ProjectP1609 MOP 05/01/75 Telecommunications Project686 SAR 05/01/75173 SAR 02/01/75 Oil Palm ProjectP1431 MOP 02/01/75P1469 MOP 06/01/74 Accra - Tema Water Supply (Stage Two) ProjectP1390 MOP 06/01/74 Livestock Development Project425 SAR 05/01/74 Accra - Tema Water Supply (Stage Two) Project277 SAR 05/01/74 Livestock Development ProjectP1254 MOP 10/01/73 Highway Project76 SAR 06/01/73152 SAR 05/01/73 Rice Development Project27 SAR 12/01/72 Telecommunications ProjectP1159 MOP 12/01/72 Sugar Rehabilitation ProjectPAIOO SAR 11/01/72PU 105 SAR 09/01n2 Accra - Tema Water Supply (Stage Two) ProjectPE37 SAR 10/01/71 Education ProjectP958 MOP 06/01/71 Second Power Distribution ProjectPU52 SAR 10/01/70P845 MOP 06/01/70 Cocoa ProjectPA43 SAR 06/01/70

146

Report # Type Date Title

PA7 SAR 09/01/69 Fisheries ProjectP740 MOP 09/01/69PU12 SAR 07/01/69 Accra Tema Water Supply and Sewerage ProjectP735 MOP 07/01/69P723 MOP 06/01/69 Highway Engineering Credit ProjectPTR11 SAR 06/01/69P703 MOP 05/01/69 Volta River Hydroelectric Expansion ProjectPU8 SAR 05/01/69P605 MOP 05/01/68 Power Distribution ProjectT0629 SAR 05/01/68P263 MOP 08/01/61 Volta River Hydroelectric ProjectT0281 SAR 08/01/61T0249 SAR 06/01/60

| 0 - - -

5 _ _ _ _

._

_ -

i C'- ;;SP. E= =w5 -. R

_ _i;,,_ ,j j _ i F

3

.. 3

i E . _ * l:

_ _

-

I_

.

._

X Xe E I | | | i

lEQ=-- | . | ,l I, j ,,

l

|

|, .. .__ i__

| i

|

_ IMAGIN&_

P-port No: 14547, , Type: OES

_i| i

_ _ _