Document of The World Bank Report No.: 26197 PROJECT PERFORMANCE ASSESSMENT REPORT GHANA MINING SECTOR REHABILITATION PROJECT (CREDIT 1921-GH) MINING SECTOR DEVELOPMENT AND ENVIRONMENT PROJECT (CREDIT 2743-GH) July 1, 2003 Sector and Thematic Evaluation Group Operations Evaluation Department Currency Equivalents (annual averages) Currency Unit = Cedi (C) 1993 US$1 = C649 1994 US$1 = C957 1995 US$1 = C1200 1996 US$1 = C1637 1997 US$1 = C2050 1998 US$1 = C2314 1999 US$1 = C2669 2000 US$1 = C5455 2001 US$1 = C7171 Abbreviations and Acronyms AfDB African Development Bank AGC Ashanti Goldfields Company CAS Country Assistance Strategy DIC Divestiture Implementation Commission EPA Environmental Protection Agency GGL Goldfields Ghana Limited GOG Government of Ghana GSD Geological Survey Department ICR Implementation Completion Report IRS Internal Revenue Service JV Joint Venture MC Minerals Commission MD Mines Department MDF Minerals Development Fund MOF Ministry of Finance NDF Nordic Development Fund OED Operations Evaluation Department PMMC Precious Metals Marketing Company PPAR Project Performance Assessment Report SAR Staff Appraisal Report SGMC State Gold Mining Corporation SSM Small-scale mining TA Technical assistance Fiscal Year Government: January 1-December 31 Director-General, Operations Evaluation Mr. Gregory K. Ingram Director, Operations Evaluation Departnent (Acting) Mr. Nils Fostvedt Manager, Sector and Thematic Evaluation Mr. Alain Barbu Task Manager Mr. Andres Liebenthal OED Mission: Enhancing development effectiveness through excellence and Independence In evaluation. About this Report The Operations Evaluation Department assesses the programs and activities of the World Bank for two purposes: first, to ensure the integrity of the Bank's self-evaluation process and to verify that the Bank's work is producing the expected results, and second, to help develop improved directions, policies, and procedures through the dissemination of lessons drawn from experience. As part of this work, OED annually assesses about 25 percent of the Bank's lending operations. In selecting operations for assessment, preference is given to those that are innovative, large, or complex; those that are relevant to upcoming studies or country evaluations; those for which Executive Directors or Bank management have requested assessments; and those that are likely to generate important lessons. The projects, topics, and analytical approaches selected for assessment support larger evaluation studies. A Project Performance Assessment Report (PPAR) is based on a review of the Implementation Completion Report (a self-evaluation by the responsible Bank department) and fieldwork conducted by OED. To prepare PPARs, OED staff examine project files and other documents, interview operational staff, and in most cases visit the borrowing country for onsite discussions with project staff and beneficiaries. The PPAR thereby seeks to validate and augment the information provided in the ICR, as well as examine issues of special interest to broader OED studies. Each PPAR is subject to a peer review process and OED management approval. Once cleared internally, the PPAR is reviewed by the responsible Bank department and amended as necessary. The completed PPAR is then sent to the borrower for review; the borrowers' comments are attached to the document that is sent to the Bank's Board of Executive Directors. After an assessment report has been sent to the Board, it is disclosed to the public. About the OED Rating System The time-tested evaluation methods used by OED are suited to the broad range of the World Bank's work. The methods offer both rigor and a necessary level of flexibility to adapt to lending instrument, project design, or sectoral approach. OED evaluators all apply the same basic method to arrive at their project ratings. Following is the definition and rating scale used for each evaluation criterion (more information is available on the OED website: http://worldbank.org/oed/eta-mainpage.html). Relevance of Objectives: The extent to which the project's objectives are consistent with the country's current development priorities and with current Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, Operational Policies). Possible raUngs: High, Substantial, Modest, Negligible. Efficacy: The extent to which the project's objectives were achieved, or expected to be achieved, taking into account their relative importance. Possible ratings: High, Substantial, Modest, Negligible. Efficiency: The extent to which the project achieved, or is expected to achieve, a return higher than the opportunity cost of capital and benefits at least cost compared to alternatives. Possible ratings: High, Substantial, Modest, Negligible. This rating is not generally applied to adjustment operations. Sustainability: The resilience to risk of net benefits flows over time. Possible ratings: Highly Likely, Likely, Unlikely, Highly Unlikely, Not Evaluable. Institutional Development Impact: The extent to which a project improves the ability of a country or region to make more efficient, equitable and sustainable use of its human, financial, and natural resources through: (a) better definition, stability, transparency, enforceability, and predictability of institutional arrangements andlor (b) better alignment of the mission and capacity of an organization with its mandate, which derives from these institutional arrangements. Institutional Development Impact includes both intended and unintended effects of a project. Possible ratings: High, Substantial, Modest, Negligible. Outcome: The extent to which the project's major relevant objectives were achieved, or are expected to be achieved, efficiently. Possible ratings: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Bank Performance: The extent to which services provided by the Bank ensured quality at entry and supported implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of the project). Possible ratings: Highly Satisfactory, Satisfactory, Unsatisfactory, Highly Unsatisfactory. Borrower Performance: The extent to which the borrower assumed ownership and responsibility to ensure quality of preparation and implementation, and complied with covenants and agreements, toward the achievement of development objectives and sustainability. Possible ratings: Highly Satisfactory, Satisfactory, Unsatisfactory, Highly Unsatisfactory. Contents Contents ............................................ iii Principal Ratings ........................................... .v Key Staff Responsible ............................................v Preface ............................................ vii Summary ............................................. ix Background ............................................1 Macroeconomic Context ............................................ 1 Ghana 's Extractive Industries ..........................................2 Bank Involvement in Ghana 's Extractive Industries ..........................................4 Project Assessments ............................................4 Mining Sector Rehabilitation Project ............................................4 Project Objectives ............................................4 Concept, Design, and Quality at Entry ............................................5 Project design ............................................5 Quality at entry ............................................6 Project Implementation ............................................7 Ratings ........................................... 10 Efficacy ........................................... 10 Efficiency ...................................................... 10 Sustainability ........................................... 11 Institutional development impact ........................................... 11 Outcome ........................................... 12 Bank performance ........................................... 12 Borrower performance ...................................... 13 Mining Sector Development And Environment Project ..................................... 14 Project Objectives ..................................... 14 Design and Quality at Entry ..................................... 14 Project design ..................................... 14 Appraisal and quality at entry ..................................... 15 Project Implementation ..................................... 15 Support to small-scale mining ..................................... 16 Environmental mitigation ..................................... 16 Health of artisanal miners ..................................... 17 Ratings ...................................... 17 iv Efficacy ..................................................... 17 Efficiency ..................................................... 17 Sustainability ..................................................... 17 Institutional development impact ...................................................1 l8 Outcome ..................................................... 19 Bank performance ..................................................... 19 Borrower performance ..................................................... 20 Current Sectoral Issues ..................................................... 20 Sharing of Resource Rents ..................................................... 20 Development of Artisanal/Small-Scale Mining ......................................... 21 Environmental Mitigation in the Small-Scale Mining Sector .................... 22 Post-mine Economic Development ..................................................... 22 Future of Large-Scale Mining ..................................................... 23 Lessons Learned ...................................................... . . 23 Annex A. Basic Data Sheet . ..................................................... 25 v Principal Ratings MINING SECTOR REHABILITATION PROJECT ICR ES PPAR Outcome Highly Satisfactory Satisfactory Moderately Unsatisfactory Sustainability Likely Likely Unlikely Institutional Substantial Substantial High Development Impact Bank Performance Satisfactory Satisfactory UnsaUsfactory Borrower Satisfactory Satisfactory Unsatisfactory Performance MINING SECTOR DEVELOPMENT AND ENVIRONMENT PROJECT ICR ES PPAR Outcome Satisfactory Moderately Satisfactory Moderately Satisfactory Sustainability Likely Non-evaluable Unlikely Institutional Substantial Modest Substantial Development Impact Bank Performance Satisfactory Satisfactory Satisfactory Borrower SaUisfactory Satisfactory Unsatisfactory Performance Key Staff Responsible MINING SECTOR REHABILITATION PROJECT Project Task Manager/Leader Division Chief! Country Director Sector Director Appraisal J. Moose S. Aiyer C. Koch-Weser Mid-term L. Maraboli M. Oakes Smith E. Lim Completion L. Maraboll P. van der Veen S. Michailof MINING SECTOR DEVELOPMENT AND ENVIRONMENT PROJECT Project Task Manager/Leader Division Chief! Country Director Sector Director Appraisal L Maraboli M. Oakes Smith 0. Lafourcade Mid-term L. Maraboli P. van der Veen P. Harrold Completion F. Remy P. van der Veen P. Harrold vii Preface This is the Project Performance Assessment Report (PPAR) on two mining projects in Ghana: the Mining Sector Rehabilitation (Cr. 1921-GH) and the Mining Sector Development and Environment (Cr. 2743-GH) projects. The forrner project was partly funded by an IDA credit of SDR 29.3 million, approved in June 1988, and closed three years late in December 1996. It was cofinanced by the European Investment Bank. The Mining Sector Development and Environment Project was supported by an IDA credit of SDR 7.9 million, approved in June 1995, for which the original closing date of December 31, 2000, was extended by one year. The Nordic Development Fund provided cofinancing for this project. This report is based on the Implementation Completion Reports (ICRs) prepared by the Africa Region (Report no. 16731, dated June 19, 1997, and Report no. 24196, dated June 12, 2002), the appraisal documents (Report no. 7039, May 10, 1988, and Report no. 13881, May 19 1995) loan documents, project files, and discussions with Bank staff. An Operations Evaluation Department (OED) mission visited Ghana in October 2002 to discuss the effectiveness of the Bank's assistance with the government and the project implementing agencies. The collaboration and assistance of all their officials are gratefully acknowledged. This PPAR is the first OED review of Bank operations in Ghana's mining sector. It will serve as an input to an ongoing OED evaluation of Bank Group activities in extractive industries, for which Ghana has been selected as a case study. Following standard OED procedures, a draft PAR was sent to the borrower and cofinanciers for comments before it was finalized, but no comments were received. In accordance with the Bank's disclosure policy, the final report will be available to the public following submission to the World Bank's Board of Executive Directors ix Summary This is the Performance Assessment Report prepared by the Operations Evaluation Department on two mining sector projects in Ghana. The Mining Sector Rehabilitation Project (MSRP) was approved for a credit of SDR 29.3 million in FY88 and closed in FY97. It was cofinanced by the European Investment Bank. The Mining Sector Development and Environment Project (MSDEP) was approved for an IDA credit of SDR 7.9 million in FY95 and closed in December 2001. It was cofinanced by the Nordic Development Fund. The key objectives of the MSRP were to (i) rehabilitate economically viable mines, (ii) help attract private investment in mining, (iii) strengthen the governmental agencies dealing with the sector, and (iv) increase the benefits to the country from small- scale mining. A significant environmental cleanup component was added late in the implementation period, but the project's objectives were never modified to reflect this. The basic premise of the first objective was flawed. Two, and arguably all three, of the state gold mines slated for rehabilitation proved to be unviable for technical and financial reasons. Nevertheless, the state mines were kept going for more than a decade (1983-95), at a cost to GOG in excess of US$ 100 million, largely funded by IDA and other donors. All three mines were shut down by their private operators within a few years of their takeover, due to a combination of poor geology, technical problems and low gold prices. In light of the project's physical and financial shortcomings, this PPAR assesses the outcome of the project as moderately unsatisfactory. The institutional development impact of the project is evaluated as high because of the legalization of small-scale mining and the, strengthening of two key sector agencies. Project sustainability is rated unlikely because the Minerals Commission, which was one of the project's successes, is in financial difficulty, threatening its future performance. Both Bank and borrower performance are assessed as unsatisfactory, given poor project quality at entry, inadequate supervision, and the inability of the government to make hard decisions. The main objectives of the MSDEP were to (a) enhance the capacity of the mining sector institutions to carry out their functions of encouraging and regulating investments in the mining sector in an environmentally sound manner; and (b) to support the use of techniques and mechanisms that would improve the productivity and financial viability of small-scale mining operations and reduce their environmental impact. The project substantially succeeded in its capacity building objective but failed to improve the performance of small-scale mining (SSM). Efforts to improve the observance of health and safety precautions by artisanal miners had meager results. Despite the failure of the SSM component, the overall outcome of the project is assessed as moderately satisfactory in light of the project's continuing relevance and the substantial progress in its capacity-building objective. x Overall the project's institutional development impact is assessed as substantial because of the positive results achieved in strengthening the Minerals Commission and the Mining Department and because of the improved legal and fiscal provisions for mining (currently before Parliament) that were drawn up as a direct result of the project. Heightened awareness of and emphasis given to the environmental impact of mining can also be attributed to the project. However, the sustainability of the capacity building achievements of the project is now in doubt and overall project sustainability as rated as unlikely. Bank performance is assessed as satisfactory, albeit only marginally, given shortcomings in the project's quality at entry. Borrower performance is assessed as unsatisfactory because the Government did not provide proper sector leadership, failed to reform the Geological Survey Department and latterly did not provide adequate funding for the Minerals Commission to operate effectively. The main lessons learned from these projects are: * hnproving the legal and fiscal framework for private investors can be highly effective in attracting inflows of capital to the mining sector. * Legalization of artisanal mining is desirable but needs to be accompanied by measures to tax and control the environmental damage caused by such mining. * Effective, transparent, and equitable mechanisms for transferring resources back to communities affected by large-scale mining are essential sociopolitical tools to ensure that tangible economic benefits are felt by local people and that mining has legitimacy in the eyes of the public. Gregory K. Ingram Director-General Operations Evaluation Background 1. The mining sector in Ghana, as described in this part of the report, is a major exporter and significant contributor to government revenues. Since the mid-1980s, extractive industries have been undergoing a massive transfornation fueled by large infusions of foreign direct investment. The World Bank has contributed to this transformation through assistance to rehabilitate government-owned mines and by its wide-ranging support to sectoral supervisory bodies. MACROECONOMIC CONTEXT 2. Ghana has a GDP of about US$5.7 billion. GDP per capita for its population of slightly more than 20 million (which is increasing at 2.2 percent annually), is about US$280. Its relatively high external debt of about US$7 billion, 125 percent of GDP, has enabled Ghana to qualify for relief under the Highly Indebted Poor Countries (HIPC) initiative. Real growth has averaged about 4 percent annually over the five-year period 1997-2002, but the economy still suffers from high inflation and a rapidly depreciating currency. These are partly the legacy of fiscal laxity driven by electoral considerations that created regular bouts of macroeconomic instability throughout the 1990s. 3. Agriculture dominates the Ghanaian economy, accounting for over 35 percent of GDP and 60 percent of employment. However, by the standards of Sub-Saharan Africa Ghana has a fairly large and diversified industrial sector that accounts for about 25 percent of GDP. 4. Mining represents about 5 percent of GDP and its contribution to total government tax revenues in 2001 was about 4%, or about US$31 million, including US$10 million of income tax paid by mining industry workers. Royalties of US$18 million, which are levied at 3 percent of the value of gold production at the free market price, accounted for most of the mining sector's contribution to government revenues'. Various tax allowances means that corporate income tax payments by mining companies are modest, despite their combined turnover in excess of US$600million. Mining exports of about US$ 700 million account for about a third of Ghana's total export revenues. Large scale mines and mining service companies employ about 30,000 workers. Considerably more people are involved in artisanal and small-scale mining. 5. In the late 1970s to early 1980s, the Ghanaian economy experienced a severe decline. The contraction of the Ghanaian economy was so severe that real per capita incomes in 1984 were nearly 30 percent below the level of a decade earlier. The Rawlings government launched its Economic Recovery Programn (ERP) in 1983-84 and received strong support from the Bank through some of the earliest structural adjustment operations in Africa. Initially there was suspicion of this new lending instrument both within Ghana and in the international community, particularly in academic circles with which Ghana was well connected. Because of the heavy emphasis on reducing the role of the state, privatizing state enterprises, and creating a climate attractive to foreign direct 1 GOG also receives dividends on its 10% shareholding in most mining enterprises. 2 investment, some critics saw structural adjustment as alienating national assets and returning the nation to a dependent relationship and peripheral position vis-a-vis the global economy. In Ghana, the gold mines were particularly sensitive, as they were culturally symbolic of the pre-colonial state. GHANA'S EXTRACTIVE INDUSTRIES 6. Ghana's extractive industries produce gold, diamonds, manganese, bauxite,2 and salt. Modest offshore oil and natural gas reserves in the Saltpond basin were briefly exploited in the early 1980s, but then shut down as they were uneconomic. Gas production from the offshore Tano field is due to begin soon, to feed a power plant. 7. Gold has been Table 1: Mineral Production in Ghana, 1985-2001 produced in Ghana for Mineral 1985 1990 1995 2000 2001 hundreds of years. Modem Gold (000 ounces) 300 541 1709 2315 2336 mining methods were Diamonds (000 carats) 636 637 632 627 870 introduced in the late l9t Dimns(0cat) 63 67 62 67 80 Manganese (000 tons) 357 247 187 895 1077 century and production Bauxite (000 tons) 124 364 530 504 678 million ounces in the early N.B. The total value of minerals production in 2001 was US$662 million. 1 960s. Thereafter the industry went into a 20-year decline that was finally halted in the mid-1980s. Since then the nature and scale of the mining sector has gone through a massive transformation. Foreign direct investment of about US$5 billion3 has flowed in and probably exceeds the value of FDI in all other sectors combined. The new investment has been heavily concentrated in surface mining, which is now the predominant method of extraction. While diamond, manganese, and bauxite have shown large increases in production, their financial importance to Ghana is dwarfed by gold (Table 1), which accounts for over 90 percent of mineral exports and, at about US$650 million in 2002, is by far Ghana's largest export.4 2. The bauxite is not of metallurgical grade and is exported while an aluminum smelter operates on imported aluniina. 3. Including mining service companies. 4. In 1992 gold overtook cocoa, historically Ghana's principal export. 3 8. Gold mining in Ghana today bears little Table 2: Gold Production in 2001 resemblance to that of the 1980s when virtually Mine Quantity Value all operations were in underground, state- (000 ounces) (US$ m) owned or controlled mines.5 In the mid-1990s Ashant Goldfields 998 271 the Ashanti Goldfields Ltd. (AGC) was Goldfields Ghana 527 143 successfully floated on the London stock Abosso Goldfields 303 82 exchange6 and the three main state-owned gold Resolute Amansie 111 30 mines were divested. Gold production was Others 231 61 valued at nearly US$640 million in 2001 and is Small-scale miners 166 50 entirely controlled by the private sector. Two Total 2336 637 companies, AGC and Goldfields Ghana (GGL)7 account for over 65 percent of production (Table 2). 9. There is also a large, primarily illegal small-scale/artisanal sector that produces gold and diamonds8 and employs much more labor than large-scale mining. It is a long- standing feature of the sector and is likely to outlive large-scale mining. By absorbing large amounts of unskilled labor (as many as 100,000 people, including children, although not necessarily full time)9 this activity is a valuable socio-economic safety valve where there is permanently high unemployment. Unfortunately, it has markedly poor health and environmental impacts. 10. Gold output is now about 2.3 million ounces, a tenfold increase (Tables 1-2) in volume since its record low of about 280,000 ounces in 1983-84. The ERP received massive backing from the donor community and the Bank quickly approved an export rehabilitation project'" and two structural adjustment operations. They provided a crucial injection of foreign exchange into the economy to enable the purchase of essential spares and other inputs needed to restore Ghana's main export industries, cocoa, timber, and gold. 11. As a result of the improved macroeconomic conditions brought about by the ERP and the new minerals law passed in 1986, private foreign investment quickly began to flow in. The first new mine for over 40 years (Konongo) started production in 1988. Three new open-pit gold mines entered production in 1990-91. One of these (Bogosu) was backed by the IFC. 5. Before 1988 all gold was produced by AGC (55 percent state-owned) or SGMC (100 percent public), which together employed about 19,000 workers. 6. AGC is Ghana's largest company (but with a substantial British shareholding) and also has mines in Guinea, Zimbabwe, and Tanzania. In 2001 it was ranked as the tenth largest gold producer worldwide. The Ghanaian government is a minority shareholder. 7. A subsidiary of Goldfields of South Africa, the world's fourth largest gold producer in 2001. The government has a 10 percent stake in GGL. 8. Over half of Ghana's total recorded diamond production comes from small-scale operators. 9. This is a 'guesstimate'. No accurate or reliable number is available for the extent of employment in this activity. 10. Cr. 1435-GH, approved January 1984. It was accomnpanied by a parallel TA project (Cr. 1436-GH). 4 BANK INVOLVEMENT IN GHANA'S EXTRACTIVE INDUSTRIES 12. In the early 1980s, as part of IDA's support to the ERP, nearly US$50 million was made available to the mining sector from the previously mentioned IDA adjustment credits. The funds were used for the rehabilitation of the three gold mines run by the State Gold Mining Corporation (SGMC), which employed about 8,000 workers and was on the point of collapse. IFC also assisted AGC in raising funds (through a syndicate loan offering) for its own rehabilitation program in the mid-1980s. Hence, although the Mining Sector Rehabilitation project (approved in mid-1988) was the Bank's first lending operation tailored exclusively to the needs of the mining sector, it was the logical continuation of assistance that had begun several years earlier. The Mining Sector Development and Environment Project followed in 1995, seven years later. These two operations overlapped for about two years, due to a major delay in implementing the former project. The latter project closed at end-2001 and there has been no follow-on operation. 13. By 1986, when preparation of the first free-standing lending operation for mining began, the decline in gold production had been arrested, but SGMC was still making losses due to its high costs (in part due to overmanning), decayed plant, and low output. The mining industry was still penalized by an overvalued exchange rate, which also provided strong incentives to smuggling of gold and diamonds. In 1985 an expatriate contract management team was put into SGMC with Bank-CIDA funding" to help fill the many gaps in technical and managerial positions. Many skilled Ghanaians had left the public sector (many also left the country) due to low salaries and the prolonged macroeconomic crisis during the 1970s and early 1980s. Project Assessments MINING SECTOR REHABILITATION PROJECT Project Objectives 14. The stated objectives of the Mining Sector Rehabilitation project were to (i) rehabilitate economically viable mines, (ii) help attract private investment in mining, (iii) strengthen the governmental agencies dealing with the sector, and (iv) increase the benefits to the country from small-scale mining. The outright sale of the state-owned mines was not an explicit project objective, although joint-venturing of the three SGMC mines (Prestea, Tarkwa, both underground, and Dunkwa, alluvial dredging) with private partners was seen as the best way to improve management and possibly obtain additional funding for mine rehabilitation"2. 11. From the Export Rehabilitation TA Project (Cr. 1436-GH). 12. The Region pointed out that the evolution from rehabilitation through joint-venturing to outright sales was a logical one and fully consistent with the objectives of the project and the evolving thinking about the role of Government from owner/operator of mining assets/facilities to regulator/administrator at that time. 5 15. These objectives were highly relevant to the issues in the sector at that time. Viewed from the standpoint of current sectoral priorities, with the exception of mine rehabilitation, objectives (ii) to (iv) are assessed as substantially relevant. However, one surprising omission was any environmental objective, given that the mines had been operating in poor condition for many years. A significant environmental cleanup component was added to the project at a late stage of implementation, but the project's objectives were never formally modified to reflect this'3. 16. The basic premise of the first objective, which was the core of the entire project,"4 was flawed. Two, and arguably all three, of the state gold mines slated for rehabilitation proved to be unviable for technical and financial reasons. A more thorough technical and financial evaluation prior to deciding the scope of the project would have revealed that they were marginal investments, but this information only came to light after the credit had already been approved. 17. From GOG's perspective, the project also had an unstated objective: the preservation of mining sector jobs, regardless of the financial or economic viability of the mines, particularly as Prestea and Dunkwa were company towns with no other raison d 'etre. In addition, the mineworkers union was politically powerful. The Bank appears to have tacitly accepted this objective, judging by its willingness to finance unviable mine rehabilitation and to condone SGMC's financial losses over many years. Concept, Design, and Quality at Entry Project design 18. The lack of an explicit environmental goal is a serious shortcoming of the original project design'5. Including a component for the mitigation of past (as well as ongoing) environmental damage would have enhanced the prospects of attracting private operators'6. In any event, GOG would have to bear the cost of environmental cleanup of SGMC's past operations and this would have been suitable for IDA financing. Instead, the Staff Appraisal report (SAR) erroneously treats lightly the environmental impact of mining, stating that "the current activities of the mining sector in Ghana have very limited environmental effects in the form of subsidence, surface disturbance or disposal of waste from mining or metallurgical processing."'7 This statement was patently untrue, particularly of AGC's mines and the environmental damage caused by them in the 13. The Region pointed out that the evolving concerns of the project team and government responsiveness resulted in the inclusion of substantial additional environmental activities during implementation. 14. At appraisal 85 percent of the project cost was directed to this component and ultimately two-thirds of the credit was used for it. 15. The Region argues that the project reflected and complied with the environmental requirements and practices of the Bank at that time. 16. Contrary to information given to the PPAR mission in Ghana, the Region commented that the majority of the companies which expressed an interest in the privatization indicated they did not want environmental initiatives prior to their actual involvement. 17. Report # 7039-GH, para 3.15. 6 Obuasi area. The credit conditionality, however, did require that effluent discharges and air quality monitoring be carried out and a study be undertaken of the environmental impact of surface mining'8. The first mention of the seriousness of the environmental damage from mining was slow to emerge in Bank files. It was the result of a separate reconnaissance mission"9 in late-1 991, three and a half years after appraisal, to review environmental issues in Ghana. Formal recognition of the need to retrofit the project with environmental mitigation components took even longer, emerging in mid-1993, just a few months prior to the original project closing date. 19. In a somewhat novel feature, this traditional SL was split into two tranches, only the first of which was available for use at credit effectiveness. While the first tranche provided funds on safety grounds for the urgent rehabilitation of underground mine shaft winders, and for the institutional strengthening of government entities dealing with the sector, the balance of the credit was held back in the second tranche. Access to this depended on satisfactory implementation arrangements and adequate external financing for a more exhaustive rehabilitation program under joint ventures. Tranching the credit was a useful safeguard and helped keep up the pressure on GOG to come to closure on mine divestiture. But its potential usefulness was not exploited fully since the opportunity to entirely drop mine rehabilitation and use the second tranche for other activities (such as a major environmental cleanup or severance pay for redundant miners) was not taken. Quality at entry 20. The overall quality at entry of this project is assessed as unsatisfactory because the technical and financial evaluations at appraisal of the SGMC mines were inadequate20. The SAR takes it as given that the rehabilitation of these mines was justified, even though the results of evaluations (due diligence funded by the Project Preparation Facility) of the Prestea and Tarkwa mines by mining consultants were not available at that stage.2' The results were available before credit effectiveness and revealed that the real FRR on both was likely to be too low to attract outside investors. But this did not lead to a rethinking on the part of either IDA or GOG. The assessment of Dunkwa's potential, though available earlier, was incorrect regarding the ore reserves as well as the operational capabilities of the dredgers. Experience later showed that none of these mines had long-term viability and they did not justify the major injection of new funds from both public and private sources. 18. The final report, "Study on the effects of mining on Ghana's environment" (by NSR Environmental Consultants) was issued in October 1991. It made an important contribution to raising awareness within the Bank of the urgency of addressing the sector's poor environmental performance and thus helped trigger the preparation of a follow-on project with an explicit environmental focus. 19. The mission back to office report, dated 12/23/91 states, "Gaseous emissions [from AGC], fallout from the ore roasting stack, dust dispersion, tailing spillage, tailing dam decant liquor and various liquid effluents have over a very long period of time produced widespread contamination in the Obuasi area and in the downriver drainage." 20. The Region pointed out that the mines were considered viable once rehabilitated, based on professional assessments by the CGM group, (the management contractor) and studies conducted by other consultants. 21. The consultants (RTZ) were asked inter alia, to assess the ore reserves, estimate rehabilitation costs, and derive present value estimates for the rehabilitated mines. 7 21. Clearly there was insufficiently reliable information at the time of the Bank's appraisal on the reserve potential of the three mines. Reserve information cited in the SAR was overly optimistic: 15 years of expected production at Tarkwa22 and 40 years at Dunkwa, although these estimates partly depended on a gold price of at least $400 per ounce, which also proved to be far too high13. The SAR states that this price was 'substantially below recent long range gold price forecasts'. 22. The assessment of risks presented in the SAR correctly identified the two main risks facing the project as lower-than-expected production and falling gold prices. But the financial sensitivity analysis at appraisal (based on data that was soon revealed to be overly optimistic) showed that only a small 14 percent decline in the gold price would jeopardize the case for rehabilitation by pushing the FRR to under 10 percent. Given the volatility of gold prices, this seems to have been an insufficient safety margin. Although forecasts of commodity prices are notoriously unreliable, the SAR (para. 6.03) asserts that the base case price and production parameters used in the analysis "appear conservative" because the price was an average of the previous twelve years24. This was probably a reasonable assumption at that time. Project Implementation 23. Due to the lack of a full account in the Implementation Completion Report (ICR) of the decade-long process of divesting the SGMC mines to private operators, this section of the PPAR provides a summary of the divestiture process and a critique of both its relevance and outcome. 24. A debate within GOG and with the Bank on whether to seek a single joint venture (JV) partner for all three SGMC mines (despite great technical and operational differences between them) or to offer them separately remained unresolved for several years. It was complicated by the discovery that long-term prospects for Tarkwa were poorer than initially assessed. Offers were sought for the SGMC mines in mid-1989. No bids were initially received to take over Tarkwa. Prestea, though it had always been viewed as the less attractive of the two underground mines, received offers from potential JV partners and was the first to reach preliminary agreement in early 1991, but this failed to come to closure and ultimately Prestea was the last to be divested after three attempts. The Dunkwa alluvial dredging operation was seen as the most profitable and had attracted unsolicited expressions of interest from private firms. But this assessment proved to be unfounded as the dredgers had major technical problems that could never be satisfactorily resolved and the assessment of the remaining ore reserve was not accurate. 22. The consultants (RTZ) also found that the gold reserves at Tarkwa were of a lower grade of ore than previously assumed. 23. However, it should be recognized that the project was appraised in 1987, when gold prices were somewhat higher - around $440/ounce. They began to fall almost immediately thereafter, declining to a 20- year low of $279/oz. in 1999. 24. Expressed in 1987 dollars and excluding the 1980 peak year price of $612/oz. 8 25. Concluding the JV process proved to be much slower than anticipated (in part due to GOG indecision, unfamiliarity with privatization, and the multiplicity of players involved on the Ghanaian side - the Minerals Commission (MC), SGMC, Ministry of Lands, Divestiture Implementation Commission, MOF). Inevitably, this delayed the mine rehabilitation works as well. Over two years after Board approval, no JV was in place and bidders for Dunkwa had withdrawn their offers. In the interim, SGMC continued to lose about $5 million annually, but by 1992 its losses had risen to $12.5 million on revenues of only $17 million. 26. In mid-1991, at a cofinanciers meeting, GOG agreed to adopt the Bank's suggestion to halt production at Tarkwa and Dunkwa and put them on "care and maintenance" as a way of halting SGMC's losses, but this decision was later reversed. Nor did GOG undertake any significant downsizing of SGMCs headquarters or carry out other cost-reduction measures as promised. A second round of bids for the SGMC mines was solicited in late 1991, but a timely GOG decision proved hard to obtain, given the South African nationality25 of the bidder for Prestea and Tarkwa. Finally, Goldfields of S. Africa took over management26 of the Tarkwa mine in July 1993, five years after the credit had been approved and just six months ahead of the original closing date. Other offers were later received for Prestea and Dunkwa, eventually resulting in the successful transfer of management to private operators in mid-1995 (Dunkwa) and mid-1996 (Prestea). 27. However, these "successes" proved to be short-lived: Prestea's new operators (Barnex) pulled out in 1998, citing the low gold price and high production costs. The management and employees then took over Prestea until 2000. They too had to shut down the mine, having used up all their pooled end-of-service benefits27 and run up substantial debts with a local bank and the power company. Dunkwa also ceased operating in early 2000, due to the low gold price, poor reserves, and technical problems with the dredgers28. Finally, production from the Tarkwa underground mine also ceased in 1999, when it was replaced by a new larger, lower-cost surface mine, but the majority of workers were retained. 28. The impact of the mine rehabilitation on Ghana's foreign exchange earnings was neutral at best. There is no evidence that SGMC's gold exports increased as a result of the project. The SAR expected the project to help raise SGMC's output from 40,000 ounces to 140,000 ounces by the early 1990s. This did not happen. When Tarkwa divested (in 1993), SGMC's production was still only 43,000 ounces, which yielded significantly less foreign exchange, as the gold price had fallen to $360 per ounce from $440 per ounce at the time of appraisal. Furthermore, SGMC's substantial arrears for 25. The apartheid regime was still in power in S. Africa at that time. 26. It did not purchase the assets. 27. Paid to them by Barnex. 28. The region commented that the. owner of the Dunkwa mining operation has diversified into other activities, including agro-processing, thereby preserving employment in the area. At the time of the PAR mission in late 2002, these non-mining activities were also virtually at a standstill. 9 unpaid electricity consumption, as well as its borrowings from local banks, had to be cleared by GOG. 29. The state mines were kept going for more than a decade (1983-95) for political reasons. The cost was met from GOG's budget in turn massively backed by IDA and other donors. The SGMC bailouts increased GOG's foreign indebtedness by more than $50 million, but there was no payback in return in the form of higher taxes or royalties. It was fiscally imprudent of GOG to have borrowed (and in foreign currency) to perpetuate a non-viable business that was incapable of generating a positive cash flow. No hard budget constraint was put on SGMC. Revenue received from the divestiture of the SGMC mines was extremely modest29 and was used to cover the cost of severance pay of redundant workers not retained by the new managers and the pension liabilities of SGMC's retired employees. 30. When the ICR was written in early 1997, the authors understandably felt that the rehabilitation and divestiture of the SGMC had been successful since there were encouraging signs of increased output and reduced losses. But these turned out to be unsustainable, for geological (Dunkwa and Prestea), technical (Dunkwa dredgers), and financial reasons (low gold price). In the case of Prestea and Dunkwa the closure followed within five years of divestiture, while in Tarkwa underground mining was phased out as soon as a major new surface mine could be brought into production. 31. Dunkwa and Prestea made only modest royalty and negligible tax contributions after divestiture. Both shut down permanently within about five years of the change in management. With hindsight, it was not worth the trouble to divest them or keep them going, except for the protection of employment. It would have been cheaper to have paid the workers to stay at home, saving GOG the cost of attempted rehabilitation, especially of the Dunkwa dredgers, which suffered many technical design and operational difficulties. 32. The mine divestitures were only a partial privatization because SGMC still exists as a legal entity (although it has no operations or staff). It remains the owner of some assets of the Prestea and Tarkwa mines.30 The assets were never sold but given to the new operators on a "concession" basis. SGMC has yet to be liquidated and is likely to exist for several more years while the old equipment is gradually disposed of. 33. On a more positive note, it can be argued that divestiture of the Tarkwa underground mine serendipitously acted as a "carrot" to get Goldfields to invest in a major new surface mine. It is unclear whether they would have done so without having acquired the first-hand experience of operating in Ghana as well as access to data and information that only a local presence can bring. The opening of the new open pit operations at Tarkwa in 1998 is a major success story, but is only indirectly attributable 29. Only about $2 million was paid to GOG for each of the two underground mines and $0.4 mrillion for Dunkwa. 30. However, the Dunkwa mine was disposed of as an outright sale. 10 to the Bank's project.3" Goldfields recognized that there was an upside potential in the area and took over the Tarkwa concession. As a result of their exploration, (which was in no way connected with the Bank project), they found new reserves, albeit within the original concession area and developed a major new surface mine. 34. Having significantly improved the legal and incentive framework for new investment in mining, the best way of exploiting Ghana's rich gold potential was to pursue new prospects rather than sinking additional money into old pits. Yet at that time GOG and the donor community were unable or unwilling to recognize this, despite the evidence of major inflows of foreign capital for exploration and for greenfield mining projects. Ratings Efficacy 35. Overall efficacy of the project is assessed as substantial because three out of four objectives were largely met. However, the principal stated objective, the rehabilitation of viable mines, cannot be said to have been achieved. The implicit project objective to provide GOG with the financial resources to keep the SGMC mines in operation in order to avoid job losses, was largely achieved, even if there are major reservations about the validity of this objective. Efficiency 36. The SAR estimated the EIRR on the combined rehabilitation of the SGMC mines to lie between 17 and 29 percent, but did not provide details showing how these estimates were derived. The ICR did not attempt to recalculate the EIRR on the grounds that it "has become irrelevant as all the former operations of SGMC are now in the private sector." This is a fallacious argument since a change in ownership or management does not prevent the economic analysis of an investment. 37. The short period and low level of financial benefits accruing from the rehabilitation investments strongly suggest that the net benefits from the SGMC component of the project would have been negative32. Nor were these investments least- cost. The ICR indicates that about US$105 million was spent on mine rehabilitation. Even a very generous redundancy package for all of SGMC's 8,000 staff would have cost significantly less. 38. The project's efficiency is therefore assessed as negligible. 31. The Region disagrees with this conclusion, arguing that although 'the new project used different technology, mining methods and know-how, (it) is an integral part of the original privatization package'. 32. Ignoring the social benefits from preserving jobs. Sustainability 39. The main physical component of the project, upon which most of the IDA credit was spent, has not proved to be sustainable. Rehabilitation of the SGMC mines, which were not technically or financially robust at a time of falling gold prices, proved to be unviable. On the other hand, the inflow of private investment to the sector proved to be sustained for the past decade, although it is now increasingly uncertain. Strengthening of the Geological Survey has not been successful and the Mines Department continues to suffer from skill shortages. Even the Minerals Commission, one of the project's successes, is in financial difficulty, which threatens its future performance. Overall project sustainability is therefore rated unlikely. Institutional development impact 40. A new mining law was passed33 and a Minerals Commission was established in 1986 (while the project was being prepared). These were two key steps in enhancing the attractiveness of Ghana's extractive industries for potential investors. The law was prepared without any Bank or external assistance. However, further improved terms were contained in the modified regulations issued in mid-1987, upon Bank and IFC advice. The mining laws were later amended by Act 475 in 1995. 41. Upon advice from the Bank, gold production by small-scale artisanal miners was legalized in 1989 by passage of the Small Scale Mining Law.4 The establishment of legal purchasing arrangements initially by a public (and later by private) buying agents offering world prices for gold and diamonds to artisanal miners was the result of an active policy dialog with the Bank. It was an important liberalization of the sector and helped to boost recorded gold exports and Ghana's foreign currency earnings. However, the Bank's recommendation to privatize or close the state buyer, the Precious Minerals Marketing Corporation (PMMC), has not been acted upon. Of late, the PMMC's share of gold and diamond purchases has fallen sharply, despite offering miners 99 percent of the world price. It is handicapped by not being able to pay cash for its purchases and does not offer credit to miners, which private buyers are able to do. 42. In the 1980s, GOG supervision of sector activities by the Mines Department was severely handicapped by low salaries, shortages of professional staff, and lack of equipment (such as vehicles) to carry out mine inspections. The Geological Survey had virtually ceased operating for similar reasons. The project provided office equipment and vehicles to enable these bodies to carry out their functions, but made no headway on the issue of salaries. Support to the newly created MC in the form of training and TA was also an effective contribution of the project, which in its later years came increasingly to be managed by the MC. 33. Minerals and Mining Law, PNDCL No. 153, 1986. 34. PNDCL No. 218 12 43. The overall institutional development impact of the project is assessed as high because of the legalization of small-scale mining and the strengthening of the two key sector agencies. Outcome 44. The main stated project objective of mine rehabilitation proved to be irrelevant. The crucial sectoral priority of environmental mitigation was never formally adopted as a project objective. However, there was progress toward achieving the secondary project objectives-attracting private investment, strengthening sectoral agencies, and increasing benefits from small-scale mining. But in light of the project's physical and financial shortcomings, this PPAR assesses the overall outcome of the project as moderately unsatisfactory. Bank performance 45. As mentioned earlier (paras 20-22), the project's shortcomings at entry reflect the inadequacies of the appraisal process. Most importantly, a thorough due diligence exercise had not been completed when the project was approved by the Board. Due to the haste in bringing the project to the Board, the appraisal team was unwilling to wait for the outcome of ongoing PPF-funded studies that revealed that the mines were not financially viable. It was irresponsible of the Bank to lend for what proved to be unviable investments. 46. The Bank's supervision of the project, particularly during the first three years of implementation, was unsatisfactory. Bank files show that after the change in task manager in 1990, there was no financial analyst assigned to assist with project supervision until project closure in 1997, despite the importance of financial issues. Neither did an environmental specialist participate in supervision during the first four years of the project. Despite consistently poor supervision ratings ['3'] the Bank management went along with GOG's slow pace of decision making on divestiture and mine closure and did not take a strong position against the SGMC financial losses, which over the 5 years prior to divestiture exceeded the value of the assistance from IDA. 47. Corrective action was not taken at the (much delayed) mid-term review in late 1992, which presented an opportunity to cancel the mine rehabilitation components of the project. Other post-mine options could have been examined if the Bank had been less willing to be carried along by GOG in the SGMC bailout. It appears that even at this late stage (just a year ahead of the original closing date) the Bank believed that closure was less desirable than offloading the mines for whatever modest sum they would fetch. The IDA funds could have been used instead to carry out a comprehensive environmental cleanup at the mines and for payment of severance benefits to the miners. 48. Instead, the project underwent only minor modification and was prolonged by three extensions of the closing date. No formal project restructuring was done, even though new components such as environmental mitigation and aerial and geophysical surveys were added at a late stage. 13 49. On a more positive note, significant savings in the use of the credit (due to the inordinate delays in proceeding with mine rehabilitation works) enabled environmental mitigation of $1 million for each of the three mines to be added to the scope of the project. The inclusion of these components shows the extent to which Bank mining staff had assimilated the importance of addressing the sector's environmental impact in the five year period since the SAR had incorrectly minimized it (para 18). But the amounts were determined on the basis of the availability of funds rather than on an evaluation of actual needs at the three sites. For Prestea, the IDA funds were far less than was required and another donor agency (NDF) had to be approached to cover the shortfall. Environmental impact assessments and audits were undertaken when the environmental mitigation components were retrofitted35. 50. The ICR (prepared in the pre-1999 format) is unsatisfactory because it does not adequately describe the project's implementation experience and difficulties and its restructuring to address environmental issues. There is no assessment of projecct design or quality at entry. Coverage of financial and economic issues is weak (no financial perfornance information is provided on SGMC, nor data on royalty and tax payments or even proceeds received by GOG from divestiture). No account is given of the enviromnental mitigation works carried out.36 51. Overall Bank performance is therefore assessed to be unsatisfactory. Borrower performance 52. GOG's mining sector policies were improved considerably during the project period. The private investment climate for mining and the policy on artisanal mining both yielded substantial benefits. But viewed narrowly from the perspective of this project, which was dominated by the SGMC mine rehabilitation, these were relatively minor aspects. 53. Decision making by GOG on divestiture of the SGMC mines was very slow. In addition, GOG did not impose a hard budget constraint on SGMC. This allowed the agency to keep operating and its losses to keep rising. The state-owned mining industry and its workers benefited from very substantial subsidies for more than a decade, when these scarce funds could have been used elsewhere for activities that yielded (at least) a positive return on the investment. Finally, it was financially imprudent of GOG to borrow in foreign currency to keep an unviable state enterprise afloat. 54. Overall performance by the borrower is therefore rated as unsatisfactory. 35. It should be noted that Bank environmental staff at that time regarded these two projects as "flagships" for introducing environmental concerns into the sector dialogue on mining in Africa. 36. The PPAR rnission learned that the IDA-funded anti-pollution equipment to allow off-barge mercury processing at Dunkwa was procured but never installed before the shutdown of operations. 14 MINING SECTOR DEVELOPMENT AND ENVIRONMENT PROJECT Project Objectives 55. In overall terms, this project sought to support the sustainable development of Ghana's mining sector on an environmentally sound basis. The stated project objectives were (a) to enhance the capacity of the mining sector institutions to carry out their functions of encouraging and regulating investments in the mining sector in an environmentally sound manner; and (b) to support the use of techniques and mechanisms that will improve the productivity and financial viability of small-scale mining operations and reduce their environmental impact. 56. This was essentially a capacity-building project that redressed the relative neglect of socio-economic and environmental issues in the Bank's previous interventions in the sector. As the SAR explicitly recognizes, "[flor many years, the adverse environmental impact of mining was ignored." Addressing the problems of small-scale mining (SSM) was also a welcome rebalancing of effort and attention away from large-scale mining that had dominated the sectoral agenda since the mid-i 980s. 57. The project was in line with the declared 1995 CAS objectives of capacity building, improved governance, and environmentally sustainable growth. The 2000 CAS places much more emphasis on rural poverty alleviation, which this project addresses through its SSM/artisanal mining component. Hence, the relevance of project objectives both at the start and at project completion was high. Design and Quality at Entry Project design 58. The design of the small-scale mining component failed to address key legal, institutional, and financial issues relating to the new equipment that the project sought to introduce to raise SSM productivity, albeit on a pilot basis. Successful adoption of the machinery would have required either some form of credit scheme for the sale of the equipment, or the setting up of a service provider to own, operate, and rent the equipment out for a fee. These issues were not addressed at appraisal or incorporated later, even though they had been identified in the Executive Project Summary,3" which envisaged a sigriificant small business development component to accompany the technological improvements. However, by the time of appraisal this had been dropped as premature and much greater emphasis was given to the technical characteristics and testing of equipment to be introduced38. 37. Dated 1/13/94. 38. The Region pointed out that the reason that so much attention was given to the technical suitability and acceptability by small scale miners of the equipment introduced was the result of failed previous efforts where both Government and small scale operators had purchased and brought equipment to the country, which did not perform under the particular conditions of the Ghana deposits/circumstances. 15 Appraisal and quality at entry 59. Given the wide scope of the project, it would have been appropriate to have undertaken a sectoral environmental review as part of the appraisal process. Instead, as the project was placed in the Bank's category B, only a somewhat limited and superficial environmental analysis was done.39 Had it been known before appraisal that the project would later include a tailings dam, it would have been classed in category A. The sectoral environmental review was finally undertaken in the last year of the project and its findings were not available until after credit closure. 60. The SAR correctly identified two risks to the project, but dismissed both of them as unlikely. In practice, the project suffered when both risks materialized. First, the development and adoption of new technology appropriate for artisanal mining proved problematic (para 64). Second, not all the implementing agencies proved capable or willing to implement their components successfully. 61. The ICR correctly points out that the project's quality at entry would have been significantly enhanced if a PPF had been used to carry out the study of institutional arrangements in the mining sector. The institutional study was finally completed two and a half years after Board presentation, which, given the need for debate and consensus building, led to difficulties in completing implementation of its many recommendations40 before credit closure. Finally, a social analysis of artisanal mining during project preparation would also have helped improve the design of this component4". Project Implementation 62. Although numerous studies were undertaken in the early phases of project implementation, the MC was slow to initiate action on the recommendations flowing from these studies. This was particularly the case for the institutional reforms, which took three years to be adopted as GOG policy and which have still not been fully implemented. 63. The project design put the MC in charge of overall project coordination, while actual implementation was appropriately delegated to the various project beneficiaries. However, for reasons related to inter-institutional rivalries and difficulties in coordination and decision-making, the overall progress of the project was hindered by the multiplicity of bodies involved in its implementation. The MC did not have authority over the other entities involved and was seen by some as high-handed or lacking in objectivity. Its role was resented by some, particularly as regards institutional reforms that emerged from studies carried out under the project. 39. Both Annex G of the SAR and the PID consist of only 3 pages of identical text. 40. Some of which were hotly contested. 41. Such social studies were carried out as part of the Bank's small-scale mining projects in Niger and Burkina. 16 Support to small-scale mining 64. A substantial portion of the IDA credit (US$2.5 million) was initially earmarked for the procurement, testing, and dissemination of mechanical equipment designed to raise productivity of the SSM operators. Virtually all the equipment proved to be unsuitable for local conditions, even though study tours involving the artisanal miners, as well as MC staff, had been undertaken as part of the equipment selection process. But no feasibility study was done and the equipment procured under the project was too large for the scale of most artisanal mining, too heavy and too costly to move to different sites, and too expensive for SSM operators to purchase. These shortcomings appear to have come to light only in the closing stages of the project and were not flagged in the supervision reports or adequately covered in the ICR. 65. Equally important to the failure of the mechanization component was the lack of legal, managerial, or financial arrangements for operating the equipment procured by the MC. Since the latter was only the procurement agent and financier, it needed another entity to manage, operate, and charge a fee for the equipment if it was to be run on a rental basis, given that most of it was too large to interest potential buyers. Although the SAR envisaged experimentation with technology delivery and credit mechanisms, surprisingly, no study was carried out under the project to address these aspects, which are crucial to any future efforts to mechanize SSM operations. Hence, at project closure, no suitable equipment for SSM had been disseminated or an appropriate business model for mechanization been designed. Environmental mitigation 66. The construction of a tailings dam for the Tarkwa underground mine was initially part of the mitigation activities funded by the previous Bank project. However, it came to be included as part of the second project during its implementation because the works could not be completed before the closure of the preceding project. This concrete/earth structure prevents the finely ground waste rock (produced in the form of slurry from the old processing plant42), which was dumped into a valley, from being carried down the valley by a minor river. The area behind the dam has been stabilized and is being revegetated. Water quality is monitored downstream for evidence of chemical seepage. 67. The PPAR mission was able to confirm the positive findings of the ex-post environmental audit of the land reclamation undertaken in the Nueng North forest reserve near Tarkwa. This area had been degraded by abandoned workings of artisanal miners. The project successfully restored the forest to an acceptable environmental standard. Although undertaken by a Ghanaian contractor from outside the region, the labor input was essentially from local villagers. The MC needs to analyze the experience of the three reclamation projects funded under the credit to determine the cost of a bigger program of environmental reclamation that could be funded through a levy (para 94) on SSM operators. 42. Now pernanently closed down. 17 Health of artisanal miners 68. The efforts under the project to improve the observance of health and safety precautions by artisanal miners had meager results. Most workers who crush the ore- bearing rock do not use facemasks and inhale large amounts of fine dust particles. Mercury, the most widely used chemical to separate gold from ore, is highly toxic. Retorts imported under the project and sold on a subsidized basis to miners did not prove to be a success for a variety of reasons. The potential users were dissatisfied with the size, design, and fragility of the glass retorts, the difficulties in obtaining spares and the price (about US$60), despite the two-thirds subsidy43 from the MC. They also had insufficient financial incentives to recycle mercury, which was often provided to them by private gold buying agents. 69. Finally, many artisanal miners have a poor awareness or are skeptical of the health risks of handling and inhaling mercury. The failed efforts of the MC at disseminating mercury retorts on a large scale leaves the large number of mercury users exposed to unacceptably high risks to their health as well as that of their unborn children. Clearly more attention is needed to identify technically and financially acceptable mercury retorts. Apart from an improved, more robust model that could be manufactured for the Ghanaian market, the case for an initial free distribution of retorts on health grounds should also be studied. Ratings Efficacy 70. The project substantially succeeded in its capacity-building objective but failed to improve the performance of small-scale mining. Overall efficacy is therefore rated as modest. Efficiency 71. In the absence of any financial or economic analysis for this technical assistance project, it is not possible to assess its cost effectiveness or evaluate its efficiency. Sustainability 72. The overall progress in institutional development achieved under the project and its sustainability are threatened by the ongoing severe squeeze on the operating budgets of all the sector supervisory bodies. The annual allocations from MOF to those bodies are grossly inadequate. In recent years they have been able to top up what they receive from MOF by using their share of the Minerals Development Fund (paras 85-89). However, the disbursement arrangements have been altered since 1999 because of GOG's budgetary crisis. These monies are now credited to the Consolidated Fund rather than being disbursed directly by the IRS. GOG has been withholding releases to the MDF 43. The cost of the retorts was about $180. 18 from the Consolidated Fund, with an adverse impact on the operational capabilities of the MC, MD, and GSD. A loss of skilled staff is also to be feared if there is no early resolution to the financial crisis faced by the sector entities. The sustainability of the capacity building achievements of the project is therefore uncertain. In the absence of evidence of corrective action by GOG, this PPAR assesses overall project sustainability as unlikely 4. Institutional development impact 73. The project had a substantial positive impact on the Minerals Commission. It is clearly the premier public entity dealing with the sector. After close to 15 years of support from and intensive dealings with the Bank under two projects, it has developed into a mature and relatively well-run institution. It has made a major contribution to the impressive growth of the Ghanaian mining sector. However, given the weakness of the new Ministry of Mines (para 75) it needs to be more proactive in steering the proposed fiscal and legal modifications to rapid adoption by Parliament. 74. The Mines Department benefited significantly under the project in the form of vehicles, office, and laboratory equipment. The MD's ability to police and enforce mining regulations has been enhanced by the project. Nevertheless, it is overstretched, given the increase in the number of mines, short of technical staff,45 and underfunded. The present mining regulations are outmoded and in numerous instances inappropriate for surface mining. They have been updated by the MD but have yet to be formally adopted by GOG. The extent of its responsibilities for environmental monitoring is still imprecise and a clear demarcation of roles with the EPA is needed. The case for a merger with the MC appears to have some merits and deserves study. 75. At the Geological Survey Department the project funded a significant amount of airborne geophysical surveys, software, hardware, and consulting services46 for its interpretation. The latter work is still ongoing. The diagnostic studies of the GSD as an institution (also carried out under the project) concluded that it was moribund, in need of radical restructuring and that it should subcontract much of its technical work to specialized, outside firms. The GSD senior staff had a totally different philosophy and believed that in-house capabilities should be developed to do all the analytical work. The restructuring proposals were never accepted by GSD and neither the MC nor the ministry were in a position to impose change. In recent years the GSD has found its financial situation growing increasingly worse as funding from the MDF dried up and the Bank's project closed. It is still greatly overstaffed at the lower skill levels47 and its operating budget from MOF is inadequate for it to function effectively. The ID impact of the project on the GSD is rated as modest. 44. The Region disagrees with this conclusion and believes that the funding of sector bodies is adequate. 45. Hiring and retaining skilled staff is difficult because of its civil service pay scales. The Bank unsuccessfully tried to obtain better service conditions for MD staff. 46. Essentially financed by NDF, which also provided a resident TA adviser to the GSD. 47. About 250 of its 350 staff are in the administrative and support category. 19 76. The project had a negligible ID impact on the supervising ministry (Energy & Mines), which was split into two ministries in mid-2002. The sector now has its own ministry for the first time, but unfortunately it is an empty shell. The split took place without planning or budgeting. At the time of the PPAR mission, it had no office space of its own and virtually no staff. The office equipment funded under the project had been retained by the Ministry of Energy. In these circumstances, the ministry is unable to play a substantive role in policymaking and the minister has to rely on the MC for advice and guidance. 77. Overall the project's ID impact is assessed as substantial because of (i) the positive results achieved in the MC and MD, (ii) the improved legal and fiscal provisions for mining (currently before Parliament) that were drawn up as a direct result of the project and (iii) the project's contribution to the integration of environmental issues in mining sector policy making and supervision. Outcome 78. Despite the failure of the SSM component, the overall outcome of the project is assessed as moderately satisfactory in light of the project's continuing relevance and the substantial progress in its capacity-building objective. Bank performance 79. The project had several shortcomings at entry (paras 58-60) and was insufficiently prepared as regards institutional reforms. Supervision in the initial two years of the project appears to have suffered because the Bank's team gave priority to bringing the previous project to satisfactory closure after many years of unsatisfactory performance. Supervision was then enhanced and strenuous efforts were made during the latter part of project implementation to push the reform agenda and recover from lost time. 80. Given the good results in the closing stages of implementation, the decision by the Bank's country management to withdraw from the sector was premature, particularly as key project activities were still ongoing at the time of closure. Some of the reform momentum built up in the last two years of implementation has thus been lost. There is a large agenda of unfinished policy and regulatory improvements that was launched in the closing years of the project.48 Although NDF is still engaged and the EU is expected to take over as the key donor agency from 2004, it is unclear whether other donors are in a position to carry out thorough and regular supervision. 81. Overall Bank performance is rated as satisfactory, albeit only marginally so. 48. GOG intends to address these issues through new legislation that is to be put before Parliament in the next few months. 20 Borrower performance 82. The MC executed its project coordination role satisfactorily. Implementation problems with other sector institutions and conflicts over turf were not solely of its own making. However, it should have been more proactive in attempting to the salvage the SSM component of the project. 83. GOG's performance is assessed as unsatisfactory because it did not provide proper sector leadership in general and specifically failed to impose a solution to the impasse on the reform of the GSD. Its decision to set up a new Ministry of Mines without any resources is not an example of good governance. Finally, the retention of MDF funds by the MOF is undermining the sector entities, undoing the gains of the project and is politically shortsighted. 84. Overall borrower performance is therefore assessed as unsatisfactory. Current Sectoral Issues 85. The experience of the two mining sector projects, assessed in Part I of this report, illuminates several important issues that need to be addressed in Ghana. The following section explores the way resource rents are shared, the development of small-scale mining and the mitigation of its environmental impacts, post-closure economic development of affected areas, and the future of large-scale mining. Sharing of Resource Rents 86. The Constitution of Ghana49 explicitly reserves all mineral rights for the state and does not recognize the rights of landowners to receive any share of benefits flowing from the extraction of mineral resources. Despite the lack of any constitutional provision, the considerable expansion in mining activity in Ghana gave rise to the sentiment that local communities in affected areas should receive a direct share in the royalties paid to GOG by the mining companies. This was also consistent with the practice of payment to local chiefs of "concession" and various other fees by mining companies before the 1962 Minerals Act when landowners still had rights to the subsoil resources. As a result, the Mineral Development Fund (MDF) was set up50 by GOG in 1992 at the initiative of the Minerals Commission (MC). 87. The MDF receives 20 percent of all mining royalties, about US$3.6 million in 2001. Half of this amount is for the mining communities5" and half is for mining sector institutions and mineral-related investment projects. The payment of a portion of royalties from the MDF to the sector entities was a pragmatic way of ensuring that they had adequate funding to operate effectively and to be able to pay their staff more than civil 49. Article 257, clause 6 of the 1992 Constitution vests all minerals in the President, on behalf of, and in trust, for the people of Ghana. 50. By an administrative decision, not by an Act of Parliament. 51. Less 10 percent retained by the Office of the Administrator of Stool Lands since 1999 to cover its administrative charges. 21 service scales. The portion disbursed to mining communities is again subdivided among the local administration (District Assemblies), the Traditional Councils, and the Stool chiefs, with varied ways of using and accounting for the monies. Unsurprisingly, this has given rise to controversy, both over the respective shares as well over the manner in which these bodies use the funds. For instance some District Assemblies simply treat the funds as part of their normal revenues to pay for recurrent expenses. In other cases, the traditional leaders have been criticized for using the MDF money for conspicuous consumption rather than for community development or disbursing it to those most adversely affected by mining. 88. Equally seriously, the flow of mineral royalties back to the producing areas from the MDF has also been seriously disrupted in recent years by the MOF policy of retaining these funds for general budgetary use rather than disbursing them to the MDF. Given that there is already a significant degree of popular resentment against mining, this is a politically risky strategy. A field visit to Wassa District, which contains the largest share of Ghana's mines, confirmed the competition between mining and agriculture for arable land, the poor state of local infrastructure, inadequate public services, and high unemployment. The local economy in Wassa does not appear to have benefited from large-scale mining through sustained economic growth and improved public services. 89. Local people feel no perceptible benefit from the resources extracted from "their" land, despite the sharing of royalties between the central government and the mining communities. Unemployed youth attacked local chiefs and looted or destroyed their palaces in two towns in Wassa West in late 2001. Their main grievances were the lack of jobs and insufficient access to land for cultivation. Despite the fact that the mining companies promptly pay royalties to GOG on a quarterly basis and that they cannot be held responsible for the lack of adequate flowback of these rents to the local communities, they inevitably have to bear the brunt of local discontent. 90. It is essential that the sharing of resource rents be put on a transparent, equitable, and sustainable footing. The MDF needs to be institutionalized by Act of Parliament to ensure that the MOF cannot easily withhold disbursements. The proportions earmarked for the different types of beneficiaries also need to be reviewed through public consultations in mining areas before passage of the necessary legislation. The draft bill prepared by the MC is only a first step in this direction as it essentially preserves the status quo as regards the sharing of the rent among the different beneficiaries. Development of Artisanal/Small-Scale Mining 91. This segment of the Ghanaian mining industry makes a significant contribution to mineral exports and is likely to remain an important feature of the industry in the long term. It has a high absorptive capacity for unskilled labor, low import content, and makes a big contribution to the sectors net foreign exchange earnings because there are no outflows for foreign debt service, dividends, or expatriate salaries. 92. On the other hand, small-scale mining (SSM) is responsible for considerable environmental degradation, pays virtually no taxes, operates largely in the parallel economy, and is poorly regulated in all its aspects. There would seem to be considerable 22 scope for improving its productivity through mechanization, its health and safety record, and its environmental performance, but these are major challenges that have so far proven beyond the capability of GOG and donor-funded interventions. Environmental Mitigation in the Small-Scale Mining Sector 93. No workable mechanism has yet been devised to ensure that the environmental damage caused by SSM activities is rectified. All too often, such activities are short-lived and highly mobile, making it extremely hard for any effective monitoring to take place. In many instances the mining is undertaken illegally, often on legal concessions held by large mining companies, and even the identities of the miners are unknown. 94. The MC intended to use a portion of the MDF revenues flowing to it for land reclamation projects. However, in effect this would amount to large-scale mining paying for the environmental damaged caused by SSM, which pays no royalties or taxes. It is not consistent with the principle that the polluter should pay. Such royalties should flow directly into community development projects in the areas most affected by large-scale mining. 95. Another attempt to deal with the problem was made via the Precious Metals Marketing Company (PMMC), but this failed. The PMMC was obliged to drop the 4 percent levy for a land rehabilitation fund that it deducted from the gold purchase price offered to artisanal miners because it was unable to compete with unauthorized gold buyers. No attempt has since been made to reintroduce it. Yet the gold and diamond purchasing agent (whether public or private) seems to be the only point of contact with the official economy where the SSM sector can be levied a fee for environmental reclamation. Provided the environmental levy is kept as low as possible and is systematically collected by all registered buying agents, it should not lead to a substantial increase in undeclared sales. However, the difficulty of universal application of the levy and enforcement of collection should not be underestimated. Post-mine Economic Development 96. This topic has received virtually no attention, despite its increasing relevance to Ghana's mining sector. The Resolute Amansie mine is slated for closure in the very near term. The closure of the Prestea underground mine has devastated the local economy of the township and its future is a political "hot potato" for GOG. Residents are clamoring for government assistance and would like to see underground mining restarted. Potential investors interested in surface mining are being rebuffed. Other mines are also likely to close in the medium term. Experience elsewhere52 shows that planning for alternative economic activities needs to begin well before closure takes place. 52. For example at the Ok Tedi and Misima mines in Papua New Guinea, both scheduled for closure by 2010 are already working on closure planning and post-mining local economic development. 23 Future of Large-Scale Mining 97. Gold production has reached a plateau, the current level of exploration activity is low and total production will decline in the near term. Should this be a cause for concern to GOG? It is unclear what its true net benefits are to Ghana. Large-scale mining by foreign companies has a high import content and produces only modest amounts of net foreign exchange for Ghana after accounting for all its outflows. Similarly, its corporate tax payments are low, due to various fiscal incentives necessary to attract and retain foreign investors. Employment creation is also modest, given the highly capital intensive nature of modern surface mining techniques. Local communities affected by large-scale mining have seen little benefit to date in the form of improved infrastructure or service provision, because much of the rents from mining are used to finance recurrent, not capital expenditure. A broader cost-benefit analysis of large-scale mining that factors in social and environmental costs and includes consultations with the affected communities, needs to be undertaken before granting future production licenses. Lessons Learned 98. Numerous lessons of broader relevance for mining emerge from the experience of these two projects: > Improving the legal and fiscal framework for private investors can be highly effective in attracting inflows of capital to the mining sector. > Legalization of artisanal mining is desirable but needs to be accompanied by measures to tax and control the environmental damage caused by such mining. > Gold mining projects need to be financially robust under very wide price fluctuations if the risk of losses is to be minimized. > Much more analytical work needs to be done on the socioeconomic issues relating to mine closures and the transition to sustainable, non-mining economic activities. > In countries with an unstable, non-convertible currency, gold and diamonds will always be attractive alternative stores of value and convenient vehicles for tax avoidance. > Financial autonomy for regulatory bodies such as the Minerals Commission is crucial for their effectiveness and sustainability. This can be assured through levies on mining companies and/or license fees, etc. > Effective, transparent, and equitable mechanisms to transfer resources back to communities affected by large-scale mining are essential sociopolitical tools to ensure that tangible economic benefits are felt by local people and that such mining has legitimacy in the eyes of the public. > Improving the productivity of artisanal mining is more complex than just the selection of the most suitable equipment. Financial and managerial arrangements for the dissemination and operation of machinery are equally crucial. 25 Annex A Annex A. Basic Data Sheet GHANA MINING SECTOR REHABILITATION PROJECT (CREDIT 1921-GH) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current appraisal estimate estimate Total project costs 120 107.8 90 Loan amount 40 40.1 100 Cofinancing 34.7 17.8 51 Cancellation Not applicable -- Date physical components completed June 1994 June 1997 Economic rate of return 29% Not provided Cumulative Estimated and Actual Disbursements FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 Appraisal 4.5 15.6 27.0 34.1 37.1 40.0 - - estimate (US$M) Actual (US$M) 4.9 11.2 17.7 22.5 24.1 28.4 32.8 35.7 40.1 Actual as % of 109 72 66 66 65 71 82 89 100 appraisal Date of final disbursement: May 6, 1997 Project Dates Original Actual Identification N/A 01/15/86 Preparation N/A 06/17/87 Appraisal 05/87 10/19/87 Negotiation 11/87 02/25/88 Board Presentation 03/88 05/10/88 Signing 07/14/88 Effectiveness 08/88 03/10/89 Mid-term review N/A 10/07/92 Project completion 06/30/94 06/30/97 Loan closing 12/31/93 12/31/96 26 Annex A Staff Inputs Planned Revised Actual Weeks US$ Weeks US$ Weeks US$(000) Preappraisal 72.8 171.8 Appraisal 23.8 98.6 Negotiations 15.2 37.7 Supervision 116.3 133.7 192.6 557.6 Completion Total 116.3 133.7 304.4 865.7 Mission Data Date No. of Staff days Specializations Performance Rating Types of (month/year) persons in field represented rating trend problems ldentificaton/ 12/85 4 15 FA,Eng,Eng,E Preparation E,Eng,E,Con E,FA FA, Eng,Eng FA,Eng Appraisal Supervision 10/88 2 17 FA,Eng 02/89 1 2 2 03/90 2 7 FA, Eng 3 2 09/90 2 10 FA, Eng 3 2 11/91 1 5 Eng 03/92 06/92 3 14 Eng, OpOff,Con 3 2 10/92 Mid-term review 06/93 1 17 Eng 11/93 3 14 Eng,OpOff,Con 2 2 04/94 3 16 Eng, OpOff,FA 07/94 4 19 Eng,OpOff,FA,Con 12/94 2 12 Eng,OpOff 10/95 2 18 Eng,OpOff 02/96 2 4 Eng,OpOff S S 06/96 1 8 Eng S S 12/96 2 8 Eng,OpOff S S Completion 04/97 2 10 Eng,OpOff S S Other Project Data FOLLOW-ON OPERA4TIONS Operation Credit no. Amount Board date (US$ million) Mining Sector Development and Credit 13.27 06/13/1995 Environment Credit 2743 27 Annex A GHANA MINNG SECTOR DEVELOPMENT AND ENVIRONMENT PROJECT (CREDIT 2743-GH) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current appraisal estimate estimate Total project costs 13.6 13.3 98 Loan amount 12.3 9.4 76 Cofinancing -- 3.6 -- Cancellation Not -- -- applicable Date physical components completed December December 2000 2003 Economic rate of return Not applicable Cumulative Estimated and Actual Disbursements FY96 FY97 FY98 FY99 FYOO FY01 FY02 Appraisal estimate 1 4 7 9 11 12.3 (US$M) Actual (US$M) Breakdown not provided in ICR Total: 9.4 Actual as % of appraisal Date of final disbursement: April 2002 Project Dates Original Actual PCD NA 11/19/92 Appraisal NA 06/14/94 Approval NA 06/13/95 Effeciveness NA 03/04/96 Mid-term review 06/20/98 07/14/98 Closing date 12/31/2000 12/31/2001 28 Annex A Staff Inputs (actual) No. of staff weeks US$ ('000) Identification/preparatio 39.1 136.90 n Appraisal/negotiation 42.8 111.10 Supervision 117.27 392.09 ICR 9.00 38.10 Total 298.17 678.19 Mission Data Date No. of Staff days Specializations Implementation Dev. Types of (monthyear) persons in field represented progress Objectives problems Identfication/ 04/94 5 ME,ES,FA,CON,LA Preparation 12/94 2 ME,ES Appraisal 07/94 5 ME,ES,FA,2 CON 02/96 2 ME,ES Supervision 06/96 2 ME,ES S S 04/97 2 ME,ES S S 07/97 2 ME,ES S S 02/98 2 ME,ES S S 07/98 1 ME S S 09/99 1 MS S S 05/00 2 MS,ES S S 11/00 1 MS S S 05/01 1 MS S S 10/01 1 MS S S Completion 10/01 1 MS ME=Mining Engineer; ES=Environment Specialist; FA=Financial Analyst; CON=Consultant; LA=Lead Advisor; MS=Mining Specialist Other Project Data FOLLOW-ON OPERATIONS Operation Credit no. Amount Board date (US$ million) None