Report No. 28069-ZA Zambia Country Economic Memorandum Policies for Growth and Diversification (In Two Volumes) Volume I: Main Report October 20, 2004 Poverty Reduction and Economic Management I Southern Africa Africa Region Document of the World Bank TABLEOF CONTENTS ACKNOWLEDGMENTS vi EXECUTIVESUMMARY vii Why Didn't Zambia Grow Faster in the 1990s? vii Sectoral Opportunities for Growth and Diversijication xii A Policy Agendafor Growth and Diversijication xiv Prospectsfor Growth and Poverty Reduction xv 1. INTRODUCTION 1 2. ECONOMICAND SOCIAL DEVELOPMENT SINCE INDEPENDENCE 3 Economic Performance between 1965 and 2002 3 TheRecord between 1991and 2002 6 3. WHY DIDN'T ZAMBIA GROW FASTERDURINGTHE 1990S? 17 WhatPolicies Worked Well? 17 What Challenges Lie Ahead? 19 4. OPPORTUNITIESAND CONSTRAINTSTO GROWTHINAGRICULTURE 34 Principal Products 36 Key Constraints to Growth in Agriculture 38 Implicationsfor Rural Poverty Reduction 46 5. OPPORTUNITIESAND CONSTRAINTSTO GROWTHOUTSIDE AGRICULTURE 48 Mining 48 Manufacturing 50 Tourism 54 6. PROSPECTSFORGROWTH, DIVERSIFICATION, AND POVERTY REDUCTION THROUGH 2015 58 Background 58 Structure of the Zambian Economy in the Base Year (2001) 59 TheBase Growth Path Simulation 62 Simulation of Alternative Government Policies .65 Simulation of Zambia 's External Economic Environment 76 7. A SUMMARY OF POLICY RECOMMENDATIONS 80 Boxes Box 2-1 Zambia's MainEconomic Policy Regimes, 1964-2002 Box 2-2 Policy and Structural Reforms inZambia, 1991-2002 Box 3-1 The Government's Strategy for Combating HIV/AIDS Box 5-1 Prospects for Copper Production inZambia Box 6-1 Database for the CGE Model Figures Figure2-1 SelectedMacroeconomic Indicators, 1964-2002 Figure2-2 Economic Performance byPolicy Regime, 1964-2002 Figure2-3 Total ExternalDebt, 1970-2002 Figure3-1 Shares of SelectedNontraditional Exports inTotal Exports, 1990-2001 Figure3-2. Indicatorsof Governance inZambia and Sub-SaharanAfrica, 2002 Figure4-1 Agricultural Value-Added andContributionof Agriculture to GDP, 1990-2002 Figure4-2. Value of Agricultural Exports and Proportion of Total Merchandise Exports, 1990- 2000 Figure4-3 Exports ofFreshFlowers andVegetables, 1994-2002 Figure 4-4 Supplyo fFertilizer, 1990/91-1 999/2000 Figure5-1 Structure o fManufacturing, 1991-2002 Figure5-2 GrowthinManufacturingValue-Added, 1992-2002 Figure5-3. Shareof SelectedNontraditionalExports inTotal Exports, 1990-2002 Figure5-4. Arrivals inZambiabyPurposeof Visit, 1997-2001 Figure5-5. Tourist Arrivals and Expenditures, 1997-2002 Figure6-1. Long-Run Poverty Reductionunderthe Base Scenario, 2001-50 Tables Table 2-1 Sources of Growth, by Sector, 1991-2002 Table 2-2 GrowthAccounting Results, 1960-2002 Table 2-3 SelectedMacroeconomic Indicators, 1991-2002 Table 2-4 Trends inSelectedMeasures of Well-Being, 1990-2000 Table 2-5 Percentageo f Population Living Below the Poverty Line,by Region, 1991-98 Table 2-6 Infant andUnder-FiveMortality, by Region, 1980, 1990, and 2000 Table 2-7 School Enrollments by Boys and Girls, 1998 Table 3-1 Constraints to Business Operations inZambia Table 3-2 Stock o f Physical Infrastructure, 1980sand 1990s Table 3-3 Quality o f PhysicalInfrastructure, 2003 Table 3-4 Structure o f Most Favored NationTariffs inZambia, 2001/2002 Table 3-5 Indicators of Trade Performance, 1990-2001 Table 4-1 Typology o f Agricultural Producers inZambia Table 5-1 PotentialGemstone Sales IfCurrent Zambian Best Practice Were Diffused Table 5-2 Average Annual Growth Rates of Manufacturing, 1964-1990 Table 6-1 GDP, Exports, andImports, by Sector Table 6-2 Population Distributionby HouseholdCategory, 2001 Table 6-3 HouseholdConsumptionPatterns, 2001 Table 6-4 Base-Case Scenario Assumptions Table 6-5 HIV/AIDS Scenario Assumptions Table 6-6 Simulation o f Alternative Government Policies: Effects on Growth, Diversification, and Poverty, 2001-1 5 Table 6-7 Education Scenario Assumptions Table 6-8 IncreasesinTFP Growth for Agriculture-Focused Simulations, 2001-1 5 Table 6-9 Output and World Price Changes for the Copper MiningSector, 2001-1 5 Table 6-10 Simulation of ExternalEconomic Environment: Effects on Growth, Diversification, andPoverty, 2001-15 ACKNOWLEDGMENTS This report is prepared by the World Bank in close collaboration with the Government of the Republic of Zambia. From the Bank's side, the report was prepared by a team led by Abebe Adugna (AFTPl) and comprising Douglas Addison (AFTP4), Wendy Ayres (AFTP2), Constantine Chikosi (AFTPS), Steen Jorgensen (SDV), Leonid Koryukin (AFTPl), Zlatina Loudjeva (SDV), Helen Mbao (AFTHl), Alex Mwanakasale (AFTSl), Nyamazana (AFTPl), Roy Pepper (CICFA), Mushiba Fahrettin Yagci (AFTPl), and Plamen Yossifov (Consultant). The team visited Zambia in September and December 2002. Hans Lofgren, Sherman Robinson, and James Thurlow of the International Food Policy Research Institute (IFPRI) prepared a background paper on prospects for growth. Valerie Kozel (AFTHl) contributed to the poverty section of the report, and Ahmet Soylemezoglu (AFTFS) on financial sector issues. Valuable comments were received from Delfin Go, Peter Moll, Jos Verbeek, (all AFTPl), Stefan0 Patemostro (PRMPR), and Hinh Dinh (AFTP3). The Government team was led by MI. James Mulungushi, Director of Planning, MFNP, and consisted of Mr. J.C. Mubanga (Assistant Director of Planning, and main coordinator), Mr. M Lungu, Ms. I.Musonda, Ms.S.M.Sikaona, Mr. John Pycroft, Mr. F. Nkulukusa (all at the Ministryof Finance and National Planning), Mr.M.Mwiinga (Bank of Zambia), Mr. P.Mukuka (Central Statistics Office), and Mr.Siakalenge (Ministry ,of Commerce, Trade and Industry).The team provided detailed comments on the draft report, and subsequently discussed the comments with the Bank team. The report benefited from comments by its peer reviewers, Jeffrey Lewis (PRMTR), Sudhir Shetty (PRMPR), and Quentin Wodon (AFTPM). Yaw Ansu and Hartwig Schafer (Country Directors) and Emmanuel Akpa and Philippe L e Houerou (Sector Managers) provided guidance Dotilda Sidibe (AFTP1) provided desktop support, and Judith Chilufya (AFMZM) assisted the team duringmissions. Financial support from the government of Norway for Poverty and Social Impact Analysis i s gratefully acknowledged. vii EXECUTIVESUMMARY 1. In October 1991Zambia movedto a multipartydemocratic system. Inthe following years, the government implemented a number o f policy and structural reforms, liberalizing exchange and interest rates, simplifying the tariff structure and removing quantitative restrictions on trade, privatizing most state-owned enterprises, and substantially withdrawing from the agriculture sector. 2. Despite these reforms, economic growth has remained lackluster, and poverty and social conditions have worsened. Between 1991 and 2002 real per capita GDP declined at an average annual rate o f 1.5 percent a year. Reflecting the poor economic performance, overall poverty increased from 70 percent in the early 1990s to 73 percent in 1998, the latest year for which data are available. Other social indicators declined as well. Life expectancy declined from 49 years to 38 years. The under-five mortality rate rose from 151per 1,000 live birthsto 162 per 1,000. Net primary school enrollment fell from 77 percent to 67 percent. And adult literacy declined. 3. The failure to generate sufficient growthto improve the well-being of the population has led to questions, both within and outside Zambia, about the wisdom of the reforms pursued in the 1990s. But analysis suggests that it was haphazard implementation rather than the fundamental direction of reforms that was at fault. Commitment to the reforms was weak, partly because many were seen as externally imposed. As a consequence, the government didnot follow through wholeheartedly on several policy, institutional, and governance-related reforms that were critical to restore growth and reduce poverty. 4. There are hopeful signs that increased growth and poverty reduction are within reachinZambia. A participatory and country-owned Poverty ReductionStrategy Paper (PRSP), which i s to guide the dialogue with both internal and external stakeholders and structure external financial support, was completed in 2002. Some real gains-such as the four consecutive years o f per capita growth since 1999-are already visible. Recently, the Government has taken fundamental steps to introduce better fiscal control and improve the budget management processes (e.g. introduction o f MTEF, activity-based budgeting, and retooling o f its financial management system). The Kwacha has been relatively stable, and lending interest rates have been reduced although inflation still remains high. However, these hopeful signs are fragile and needto be sustained and strengthened. At issue i s not a fundamental change in directiOn o f reform but rather a demonstration o f strong ownership and commitment to reforms and implementation o f the programs outlined inthe PRSP. WHYDIDN'TZAMBIA GROWFASTERTHE 1990S? IN 5. Zambia's economy has long been tied to the copper industry, whose purchasing power has been in decline for decades. Historically, copper mining was the workhorse of the economy, the main contributor to exports and growth. But the international purchasing power o f copper began to fall as far back as 1975, and Zambia failed to find a substitute for copper to maintain export growth. Throughout the 1990s the external environment remained unfavorable, ... Vlll as the international price o f copper declined precipitously, especially after 1995. The severe drop in world copper prices, coupled with significant and almost continuous falling production and rising costs o f production, exacerbated the variability inforeign exchange earnings, inimports o f capital and intermediate goods, and in investment and growth. The downward spiral o f Zambia's economy caused by heavy dependence on copper was thus exacerbated by the sharp fall inprices and production in the 1990s. In addition, the delay in privatization o f the copper parastatal (ZCCM) meant a significant drain on its resources throughout the 1990s. 6. But declining copper prices were not the only reasons Zambia's economic performance declinedbetween 1991and 2002. Followingthe deflation ofthe mid-l990s,GDP fell sharply, declining 9 percent in 1994 and 2 percent in 1995. These output losses were exacerbated by the short-run effects o f eliminating unsustainable budget subsidies and scaling back the government's role in agriculture. Trade liberalization had a temporary disruptive effect on many industries, particularly on textiles and chemicals. Specific government policies exacerbated output losses in agriculture and manufacturing, but the across-the-board nature o f the economic slowdown in 1994-95 suggests that it was the price paid for achieving relative price stability. 7. Excludingthe one-timedisruption in real sector activityin 1994-95, real GDP grew at an average annualrate of 3 percent during 1991-2002. Why didn't Zambia grow faster? Estimates put Zambia's annual long-term growth potential at about 5 percent, implyingper capita income growth of 2.5-3.0 percent a year. Why i s this potential not being achieved? Several key problems-macroeconomic mismanagement, lack o f ownership o f reform and poor policy implementation, a weak investment climate, lack o f good governance, the HIV/AIDS pandemic- holdthe answer. MacroeconomicManagement 8. Although the macroeconomic environment has improved considerably since the early 199Os, it remains unfavorable to private investment and growth. At the beginning o f the 1990s real GDP had been stagnant for three consecutive years, inflation exceeded 100 percent, the government deficit was more than 8 percent o f GDP, and scheduled debt service amounted to 61 percent o f export earnings. By 2003 inflation had been reduced to about 17 percent, the government deficit to 6.1 percent, and debt service to about 15 percent o f export earnings. Although gains in price stabilization have been slowly consolidated, the battle i s still far from won, as inflation continues to hover between 20 and 25 percent; real interest rates remain high(about 30percent), andthe exchangerate is volatile. 9. Central to the lack of macroeconomic stability-in particular to the highinflation and real interest rate-is the lack of fiscal control and commitment to fiscal discipline. In principle, since 1993 Zambia has maintained a cash budget system, which prohibits monetary financing o f deficits. Inpractice, it violates its zero monetary financing rule. Typically, spending goes o f f track early in the year and expenditures are squeezed or ad hoc revenue measures are introduced later to make up the shortfall (see World Bank 2003b for a complete discussion o f the public expenditure management and financial accountability issues). Expenditure squeezing has fallen mostly on nonwage and capital expenditures. The government appears to abide by budgetary constraints, but it often fails to pay suppliers, increasing its payment arrears. Despite austerity measures, the government has taken on large quasi-fiscal obligations, which have resulted in huge budget overruns. Arrears accumulation, rising domestic debt, and unbudgeted wage hikes and retirement packages for civil servants have resulted in significant budget deficits and increases in government indebtedness. The latest such overrun i s the wage and salary ix increases awarded to civil servants in April 2003, which put the IMF Poverty Reduction and Growth Facility program in jeopardy and delayed progress toward the Heavily Indebted Poor Countries (HIPC) completion point. In addition, poor public expenditure management-in particular ineffective and unrealistic budgeting, wasteful public procurement, and lack of transparency and timely disclosure of government financial statements-has undermined the budget process. The lapses in fiscal discipline and fiscal management have made it difficult to create a macroeconomic environment conducive to growth and diversification. 10. Zambia's large external and rising domestic debt, combined with budgetary dependence on external financing, has constrained the government's ability to exert monetary control to achieve macroeconomic stability. Throughout the 1990s and untiltoday, Zambia has relied heavily on additional external financing to service its external debt. Poor policy performance has often led to a failure to meet the conditionalities associated with foreign aid. The result i s budget shortfalls, which are subsequently monetized. Government budgets have frequently been prepared under implausible assumptions about the level o f donor assistance and the government's ability to meet aid conditionalities. Consequently, actual disbursements frequently fall short o f pledges and usually arrive later than forecast. 11. The budget's unabated domestic financing requirements, coupled with the need to service high-yielding domestic debt and roll over maturing issues, have kept nominal lending rates high. The government's demand for domestic financing has pushed up the Treasury bill rate, the basis for commercial bank lending rates. Between 1994 and 2002 the real commercial bank lending rate averaged 16 percent, reaching 30 percent in 2002. Highreturns on government debt have caused formal sector finance to switch to Treasury bills, crowding out credit to the private sector. At 8 percent, the ratio of private sector credit to GDP in Zambia i s one o f the lowest in Sub-Saharan Africa. Inrecent years commercial banks have preferred to invest in banks abroad and in government debt. By mid-2003 almost half of commercial banks' aggregate total assets were in either government debt or abroad. 12. The financial sector must become more efficient and capable of supporting private investmentand growth. Key institutional and policy issues that require immediate attention are creating a mechanism to resolve the debt o f failed banks and state-owned non-bank financial institutions; upgrading the human and technological resources o f financial system regulators and supervisors; improving access to financial services, in particular rural financial services; and investing in financial system infrastructure to improve market data and accounting and auditing standards. Lackof Ownership of ReformandPoorPolicyImplementation 13. Zambia's commitment to reforms and policy implementationweakened during the second halfof the 1990s. Institutional reforms-most notably the privatization program and the public sector reform program-have been delayed or not fully implemented. 14. Successive governments have failed to create an environment within which the private sector can be encouraged to diversify away from copper by pursuing profitable opportunities both domestically and in export markets. As far back as the 1980s, several donors, including the World Bank, emphasized the need for Zambia to reduce dependence on copper by providing the right policy and institutional environment for other exports to flourish. The recommendations were not implemented inthe 1980s. X 15. Some limitedprogressinexport diversificationwas achievedinthe 199Os, as a result of the policy and structural reforms of the early 1990s and the availability of affordable export credit provided by the EU and the World Bank. But the extraordinary growth of nontraditional exports through 1997 could not be sustained during 1998-2001, mainly because o f the Asian financial crisis. Although nontraditional exports seem to have recovered in 2002 and 2003, policies for export diversification and to remove the anti-export bias in the economy have still not been mainstreamed. 16. Privatization of the largest parastatal, Zambia Consolidated Copper Mines, was besetby problems and delays. The government's privatization planwas adopted inthe summer o f 1996, but it took nearly three and a half years to complete the privatization. The delay was costly to the Zambian economy, substantially increasing the government's quasi-fiscal deficit and leading to a rapid accumulation o f government arrears. By disrupting donor support, the delay also had a strong negative multiplier effect on the macroeconomic environment. Privatization or commercialization o f the largest parastatals in the energy, telecommunications, and financial sectors has yet to be completed. 17. The government has not fully implemented the Public Sector Reform Program started in 1993. The main obstacle to the reformprogram for much o f the 1990s seemed to be expensive labor regulations. Since the legally mandated retrenchment costs amounted to 10 years' salary for each employee, the most economic approach for the government was to retain workers rather than retrench them. A renewed effort at public sector reform began at the end o f the 1990s. The focus has been to put inplace the basic financial management, accountability and transparency, personnel, and other systems and to begin to reform wages so that key staff can be attracted, retained, and motivated to provide good-quality services while the government continues to right-size and restructure the public service. Considerable capacity building took place inthe early 2000s, but several key policy reforms, such as financial management and public sector restructuringand pay reform, remainunaccomplished. Weak InvestmentClimateandBiasedTradePolicy 18. Zambia's less than adequate physical infrastructure and regulatory and institutionalbarriers continue to constrain investment and growth. The quality of Zambia's infrastructure improved marginally at best duringthe 1990s. And administrative and bureaucratic barriers, including difficulties securing work permits and establishing businesses, continue to discourage potential investors. Too many licenses are required to operate a business, and government agencies take too long to process licenses and other requests from investors. Government services for investors are poorly coordinated, and they are generally unavailable or of poorer quality the farther businesses are from Lusaka. Coupled with the poor macroeconomic management, it i s no wonder that both domestic and foreign private investment declined in the second half o f the 1990s and the productivity o f investments made in the first half o f the 1990s remained depressed. 19. Trade policy has remained largely biased against exports. Despite this, limited export diversification was achieved in the second half of the 199Os, triggered in particular by affordable export credit. The incentives to produce for export markets are lower than those for the domestic market. Reimbursements o f value-added taxes (VAT) and import duties paid by exporters under the duty-drawback scheme-the main export incentive scheme in Zambia-are not functioning well, with long delays. Coupled with the lack o f macroeconomic stability and the poor investment climate, the anti-export bias has further deterred private investment and more vigorous export diversification. xi Weak Governance and Accountability 20. The quality of governance has deteriorated, especially since the second half of the 1990s. During 1996-2002 Zambia performed worse than average for Sub-Saharan African countries on two measures o f governance, corruption and government effectiveness. A recent survey shows that the public widely believes that corruption was worse in the second half o f the 1990s than under previous administrations (Afrobarometer Network 2002). The privatization program has been a particular source o f concern among Zambians, many o f whom feel that it permitted some political officials to amass personal fortunes. 21. The decline in the quality of governance, particularly poor public expenditure management and financial accountability, has significantly reduced government effectiveness. The government has faced, and continues to face, considerable problems ensuring sound public financial management and accountability, which are prerequisites for improved service delivery to the poor and the private sector and ultimately for higher growth. Key problems inpublic expenditure management and accountability include the following: Spending rules and regulations are not enforced. The Ministry o f Finance has not held accountable officers who misspend public funds. The budget is ineffective and unrealistic. Huge deviations occur between budget allocations and expenditures. Supplementary appropriations are large, and spending occurs before it i s approved by Parliament. Public procurement i s wasteful and inefficient. One estimate suggests that the government could save as much as US$50 million a year by implementing good practice procurement rules and regulations (World Bank 2002a). Budget management i s not transparent. Reports are produced late and not disseminated widely. As a result, the public does not understand the tradeoffs that Zambia has to make. Audit systems are not effective, and findingso f audit reports are not adequately followed up on. Institutions for accountability and oversight (the Office o f the Auditor General, the Estimates Committee o f Parliament) are weak and lack the resources to carry out their functions. The HIV/AIDS Pandemic 22. The spread of HIV/AIDS has depleted human capital, lowered worker productivity, increased the demand for health services, and undermined growth. About 1 million adults and 200,000 children in Zambia were living with HIV infection at the end of 2001, and nearly 600,000 children have lost their mother or both parents to AIDS since the beginning of the pandemic. On average HIV/AIDS has lowered the growth rate o f real GDP by an estimated 0.3 percentage points a year since 1992 (Annex B). And this estimate defines only the lower bound: an economy-wide assessment using a computable general equilibrium model for Zambia shows that the disease i s currently reducing growth by about 1 percentage point a year. The human xii development impact o f HIV/AIDS has been even more profound. Life expectancy fell from 49 at the beginning of the decade to just 37 in 2002.' The state and extended families are now responsible for large numbers (nearly a million )of street children whose parents have been killed by HIV/AIDS. SECTORAL OPPORTUNITIES FOR GROWTH AND DIVERSIFICATION 23. Between 1991 and 2002 copper mining output contracted at an average annual rate of 5.1 percent a year. Partly offsetting these losses was growth in other sectors. Agriculture grew by an annual average rate o f 1.0 percent, industry by 1.8 percent, and services by 3.0 percent. With the steep decline in the long-term international purchasing power o f copper and value-added in mining, the need to diversify the economy i s now recognized as a priority. Zambia has several opportunities for growth and export diversification, but it faces significant constraints that reinforce the overarching challenges. Agriculture 24. Several lessons can be drawn from the experience of the 1990s. First, access to markets i s important. Farmers are willing to purchase inputs and invest in agriculture if they are reasonably certain that they can sell their products at a profit. Second, foreign financing of investment played a critical role. Lending interest rates in Zambia have averaged more than 30 percent since interest rates were liberalized in 1992. Nearly all investment in agriculture since then has come from overseas. Access to overseas finance has been instrumental in promoting out-grower programs, particularly in cotton, sugar, horticulture, paprika, and dairy. Third, improving the enabling environment for investors generally rather than providing subsidies for agricultural inputs or marketing services should remain a priority: Private investors who risk their owncapitalhavemuch stronger incentives to seekout profitable activities. 25. T o reduce rural poverty, however, the government needs to do more than promote commercial and export-oriented agriculture. The liberalization reforms o f the early 1990s have stimulated growth in agriculture generally. But 60-70 percent of smallholders have not benefited, primarily because they live far from markets where inputs can be obtained at reasonable cost andwhere output can be sold at a profit. Targeted programs are needed to reduce the isolation o f these farmers and help them move from subsistence farming to more profitable market-oriented agriculture. Sustained investment i s needed in rural infrastructure, research and extension services, rural education and health services, veterinary services, support for expanding microfinance and out-grower schemes, and other measures to stimulate growth o f the rural economy. Mining 26, The copper industry will continue to suffer from declining world copper prices and from Zambia's relatively high production costs. The long-term real price o f copper i s expected to decline by about 2 percent a year, suggesting that prospects for growth through copper will remain modest2 Despite the modest prospects, however, copper will continue to play The CSO data on life expectancy for 2002 shows a much different number:51.9yrs. Government has found a strategic partner to run Konkola Copper Mines on a commercial basis and to attract new investment.This may help introduce new technology and increase output, thereby offsetting the negative effects coming from lower price and relatively high production costs. xiii a vital role in Zambia's economy inthe short and medium term. Prudent management o f foreign exchange earnings and revenues from possible copper price recovery will also remain critical. Given the declining prospects for copper, however, it i s hard to overstate the need for export diversification inthe mediumto long term 27. Zambia's rich reserves of gemstone minerals offer a key opportunity for export diversification and growth in the mining sector. Zambia has the second-largest deposits of emeralds in the world, the largest deposits o f amethyst and aquamarine in Africa, and modest reserves o f tourmaline. Recorded gemstone exports o f US$20.3 million in 2001 represent just a tiny fraction o f the US$600 million in gemstones that could potentially be annually exported. This potential remains un-harnessed due to four key constraints: lack o f good government regulation intracking and monitoring actual and potential investors inthe sector, indisseminating information on available plots for mining, and in recording commercial mining activity in the sector; lack o f a fair value marketing system(gemstone exchange) that promotes legal trading and offers the possibility for enhancing local retention o f gemstone earnings; regular and frequent air transport to the main export market (India) and other supportive infrastructure; and access to affordable finance. Manufacturing 28. Zambia's manufacturing sector is dominated by food, beverages, and tobacco, which account for about two-thirds of totalvalue-added, and textiles andleather industries, which contribute about 17 percent. The policy shocks o f the early 1990s hurt the textiles industry,causing its share invalue-added to decline from 12percent in 1990to 8 percent in 1993, with some recovery to about 10 percent by 1995. The decline occurred because o f increased competition from imports, including imports o f duty-free secondhand clothes. Since 1996 the manufacturing sector has recovered. The food, beverages, and tobacco subsector has grown 3 percent a year, largely as a result o f increasing domestic demand due to population growth. Textiles and garments have grown about 13 percent a year on average, due mainly to their reorientation toward export markets. 29. The key opportunities for export diversification and growthin manufacturing lie in textiles and garments and processed foods. In textiles and garments, cotton yarn and polycotton yam account for more than 90 percent o f textile exports. Although export earnings have tripled since the beginning o f the 1990s, export growth has not been sustained since 1997, partly because o f the decline inthe world price of cotton lint and yam. Although Zambia i s one o f the 19 Sub-Saharan African countries eligible to receive textile and apparel benefits under the US. African Growth Opportunity Act (AGOA), it appears that only one company (Swarp Spinning Mills) has taken advantage o f the textile provisions (by exporting yarn through other eligible countries, such as Botswana, Mauritius, and South Africa). Several other companies (Unity Garments Limited, ZambidChina Mulungushi Textiles, and the Zambia Textiles o f Livingstone) have the potential to benefit from AGOA in 2004 and beyond. Inprocessed foods, sugar accounts for about 90 percent of processed food exports. Most o f the remaining processed food exports consist o f maize meal, wheat flour, and stockfeeds to neighboring countries. Zambia's proximity to Angola and the Democratic Republic o f Congo provides a good opportunity for growth inthis industry. Tourism 30. Zambia has many tourist attractions, including high-quality game parks and Victoria Falls (also known as "Mosi-oa-tunya"). The number o f visitors coming to Zambia xiv grew at an average rate o f 12 percent a year between 1990 and 2002. Foreign exchange proceeds from tourism grew more than 10 percent a year. The potential o f tourism i s immense, but advances in other sectors-both economic and social-must be made if Zambia i s to exploit its tourist assets successfully. 31. Zambia's tourism development strategy is sound, but poor policy and program implementationhas impededthe sector's growth, Constraints to tourism development include the limited international air network and unreliable local air network; the deteriorating state o f infrastructure, including poor tourist access roads, poor public health and safety, and lack o f access to good health facilities near the main tourist sites; and the high cost o f capital and significant administrative barriers, which make starting a business inZambia costly. A POLICYAGENDAFORGROWTH AND DIVERSIFICATION 32. Zambia's growth potential remains much higher than its history would suggest. During 1999-2002, a period o f relative normalcy after the extensive policy reforms o f the early and mid-l990s, real GDP grew an average o f 3.8 percent a year, much higher than the long-term average following independence (1.4 percent) or during 1991-2002 (1.4 percent), but below estimated long-term potential o f 5 percent a year (2.5-3.0 percent per capita). 33. What, then, should Zambia do? The review o f the 1990s suggests the way forward. The key to realizing higher growth and diversification lies inremoving cross-cutting and sectoral constraints. A detailed list o f policy and institutional measures for growth and diversification i s provided in the table at the end o f this section. The most important measures include the following: Improve policy implementation and ownership in key areas. The government must improve policy implementation in several key areas, where adequate strategies and policies are already in place. These areas include the PRSP itself, the public sector reform program, privatization or commercialization o f the energy and telecommunications sector, the fight against HIV/AIDS, development o f the tourism sector, and maintenance and development o f (rural) roads. Maintain policy stability. Evidence suggests that continued and unpredictable government intervention i s disruptive to sustained private sector response. Avoiding sudden surprises and clarifying the role o f government will create the certainty that the private sector needs to invest. Implement soundfiscal policy and management. Lack o f fiscal control i s at the root o f the high inflation (exceeding 25 percent in 2002) and high real interest rate (about 30 percent in 2002). Key priorities must be to reduce domestic debt and domestic financing, eliminate the quasi-fiscal deficit by further rationalizing and privatizing the remaining major public enterprises, and reform public expenditure management, by improving expenditure control and reorienting expenditures toward productive economic services and infrastructure. Improve trade policy for exports. The government must simplify the export incentive system and fully reimburse import duties to exporters in a timely manner, allow duty-free import o f capital goods, operationalize the export processing zones, provide adequate and affordable export credit, take advantage o f trade preferences by raising awareness and improving standards to meet product requirements in the European Union and United xv States, advocate reform in regional trade agreements to harmonize the various commitments and policies, nurture cross-border trade through investments in roads and communications infrastructure, improve policy coordination within the government and public-private partnership in policy design and implementation, and establish local capacity for trade policy analysis and negotiation to enhance competitiveness. Reform thefinancial sector to support private investment and growth. The government must create a mechanism to deal with failed banks and other non-bank financial institutions; upgrade human and technological resources available to financial system regulators and supervisors; improve access to financial services, including rural financial services; and invest in financial system infrastructure and improved market data, accounting standards, and auditing standards. Increase farmers ' access to markets. Measures include policies that promote cross- border trade; rural roads, storage, and marketing infrastructure; rural energy and communications; easily obtainable, accurate, and timely market information; effective legal systems for enforcing contracts; financial services that provide credit, accept savings, and help farmers insure against risks; and measures to help farmers manage risk, such as commodity exchanges. Improve regulation and market linkages in the gemstone sector. Some 460 firms are estimated to hold gemstone mininglicenses, but more than 60 percent o f the licenses are considered dormant. The vast majority of miningplots are licensedto operators who lack the financial or technical capacity to realize the mines' potential. Improved regulation, supervision, and enforcement o f statutory commitments, including termination o f dormant mining licenses, i s required. In addition, the government could facilitate the establishment o f a fair value marketing system, relieving constraints in the supply chain. This couldbe achievedby establishing a private-sector-driven gem exchange. Remove the key administrative barriers to private investment. The government should simplify work permit and license requirements and procedures, improve access to land, and improve the performance o f key government agencies serving private investors. In addition, efforts to improve the tax and customs administration procedures should be continued. Such measures will not only encourage investment but may also help informal businesses formalize their participationinthe economy. PROSPECTSFOR GROWTH AND POVERTYREDUCTION 34. What are the prospectsfor growth andpovertyreduction over the next decade?This report simulates the growth and welfare impact o f several exogenous changes in Zambia's economic environment, as well as the impact o f alternative government policies. For exogenous changes, it considers the growth and welfare impact of altemative copper prices and production scenarios as well as the impact o f HIPC debt forgiveness. For policies that are under the control o f the government, it considers the growth and welfare impact of increased spending on HIV/AIDS, increased spending on education, measures that would help increase productivity in subsistence and export agriculture, and increased investment in transportation and rural roads. What do the simulation results show? Rural households will be cushioned from the negative effects o f the likely decline in copper prices (and output) over the next decade, while urban households will face increasedpoverty. xvi a Unless the funds released by HIPC debt forgiveness are used to finance pro-poor programs, the impact o f debt relief on poverty reduction i s likely to be modest. 0 Targeting traditional and nontraditional crops, which have strong links to foreign markets, i s the best way to enhance growth and household welfare. The pro-poor outcomes o f agricultural expansion would be greatly enhanced if market access were widened on a large scale. 0 Treating H N / A I D S involves tradeoffs: beneficiaries o f H N / A I D S treatment programs gain, while living conditions (per capita consumption and poverty rates) for the population as a whole decline slightly. a Improvingthe condition o f the road network and extending the network into more remote areas will have the greatest impact on poverty. 35. It will be extremely difficult for Zambia to achieve the Millennium Development Goalof halvingpoverty by 2015. Under the base case, which represents "business as usual" for the government (see Chapter 6), extreme poverty will decline only moderately, from 55:9 percent to 49 percent by 2015. A longer simulation suggests that Zambia will not be able to halve poverty until after 2040. However, under a scenario in which the government "breaks with the past" by implementing various pro-poor policy measures-through investment in education, fighting HIV/AIDs, infrastructure development, and improved productivity in export-oriented agriculture-the headcount poverty rate can be reduced by two to three additional percentage points by 2015 for each measure implemented, thereby increasing the likelihood o f Zambia reaching the Millennium Development Goals (MDGs). Pro-poor spending alone, however, will not do the trick. Zambia needs to move to a higher growth path-estimated to be between 8-10 percent per annum compared to the current growth rate o f about 4 percent-to reach the MDGs. Implementation o f the policy reform agenda for growth, detailed below, will go a long way towards reachingthat goal. A PolicyReformAgenda for GrowthandDiversificationinZambia IPolicv,recommendation ITimeline OVERARCHINGRECOMMENDATIONS Improvemacroeconomicmanagement Implement soundjscalpolicy and management. Control expenditure to reduce domestic debt, Short term stop making guaranteed loans to parastatals, commercialize or privatize the remaining major public enterprises, and implement the public expenditure management reforms, as outlined in Zambia Public Expenditure Management and Financial Accountability Review (PEMFAR). Reform thejnancialsector to support private investment and growth. Amend Article 5 o f the Short to Bank o f Zambia Act to give the central bank independence inimplementing monetary and medium financial sector policies. Develop a cost-effect mechanism to resolve the outstanding debts o f term failed banks and other non-bank financial institutions; reform state-owned nonfinancial institutions; improve access to financial services, including rurallagricultural financial services; upgrade human and technological resources available to regulatory and supervisory financial system institutions; and establish a credit referral system to improve the credit culture and move the economy away from a cash- to a credit-based payment system. Improvedonor coordination to increase thepredictability of donorfunding. Regularly review Medium conditions attached to donor funding and monitor policy implementation, designate an term institutional entry point for all donors to facilitate donor coordination, and explore local options and strategies for exiting external aid should the conditions become unattainable. xvii Policy recommendation 1Timeline OVERARCHING RECOMMENDATIONS Improvethe investmentclimate Maintainpolicy stability. Minimize surprise policy changes by maintaining regular contacts and Short term dialogue between the public sector and the business sector. Clearly define the government's role and social protectionactivities, particularly inthe agricultural sector. Remove the key barriers tojrms `competitiveness. Expandthe road rehabilitation program, Short to especially inkey productive areas; implement necessary reforms to improve the performance o f medium public utilities (by applying the lessons learned from commercializing the Zambia Electricity term Supply Company to other state-owned companies, such as Zambia Telecommunications Company, for example); improve the regulatory framework for public utilities inorder to reduce the costs o fcritical utility inputs; andamendthe labor laws to make it less costly for employers to shed labor. Explore increasingjoint regional trade enhancing investments (for bridge, roads, and railways, for example) through the Southern African Development Community. Remove unnecessary administrative barriers to investment. Simplify license requirements, and Short to rationalize regulatory institutions to improve their effectiveness. medium term Create a trade policy that favors exports Improve trade policy administration. Simplify the export incentive system and improve the Short term timeliness o f duty-drawback schemes, ensire that export processing zones are established and operated in a transparent and efficient manner, improve customs administration, take advantage o f the EU and US. markets (through the African Growth Opportunity Act) by improving the quality o f goods produced, improve intragovernment policy coordination and public-private partnership intrade policy design and implementation, andimprove local capacity for trade policy analysis and negotiation. Combat HIV/AIDS Implement programs that decrease the number of new HIV infections and reduce the Short to socioeconomic-impact of HIV/AIDS. Supplement local resoukes with external support to medium implement behavior-change campaigns, expand the condom distribution system, increase access term to voluntary counseling and testing, strengthen community home-based care, implement programs that reduce mother-to-child transmissions, and expand access to antiretrovirals to all districts. Agriculture and the rural sector Improvefarmers' access to markets. Improve rural roads, storage, and marketing infrastructure; Short to energy and communications; access to timely market information; and contract enforcement. medium Help create fimctioning commodity exchanges that do not require lengthy licensingto export or term import major commodities, such as maize. Help create sustainable ryraljnancial services. Help buildthe capacity o fprofitable, Medium- creditworthy microfinance institutions; foster the development o f new approaches to lending, such Long term as the warehouse receipt program beingpiloted inZambia; and extend the proposed credit referral system to agricultural credit. Contain livestock diseases. Make adequate budget provisions for monitoring and controlling Medium major animal diseases, especially inthe Southern, Western, and Easternprovinces. Regulate and term ensure the certification o f private sector veterinarians and service providers. Land Improve access to secure rights to land. Improve access to land o fthose (few) smallholders who Short term are land constrainedby investing ininfrastructure and rural services to increase the attractiveness o fnewly opened settlement areas. Invest in capacity in land administration and dispute adjudicationfor state land before undertaking large-scale titling of land under traditional tenure. Improve the Land Tribunal and xviii Policy recommendation 1Timeline OVERARCHINGRECOMMENDATIONS land administration to overcome the time-consuming and logistically complicated process required o f investors interested insecuring land inZambia. Improve women's rights to inherit and own land. Regulate customary landtenure systems to Medium ensure women's access to land and dispute adjudication mechanisms. term Fertilizer Issue a clear government policy statement onfertilizer. Where rules have beenclear, the private Short term sector response has been good. Consider completely withdrawing from the market. Reduce the cost of fertilizer. Ensurepolicy consistency and clarity, improve competition among Medium suppliers, and reduce the cost o f fertilizer transportationby negotiating favorable terms for term handling fertilizer at shipping ports. Rural roads Improvepolicy andprogram implementation. Define a clear line o f accountability and a clear Medium incentive structure for implementingthe rural roads policy. Improve coordination and definition term o f responsibilities between the government, NGOs, and communities. Harmonize the different rulesfor community contributions to and management of ruralfeeder Medium roads. Develop a clear legal framework for road registration and ownership by communities. term Linkcommunities to the RoadFundfor maintenance andrehabilitation. Increase local authorities 'capacity to design and implement rural road programs. Upgrade Medium staffing and resource allocation for local authorities to design and implement community roads. to long term Nonagriculture Improve regulation of the gemstone sector. Simplify and enforce gemstone sector regulations to Short term minimize smugglingand other associated vices that currently dominate the sector. Promote market linkages in the gemstone sector. Facilitate establishment o f a fair value Short term marketing system for gemstones. Expedite the establishment o f the Gemstone Exchange, as specified inthe Poverty Reduction Strategy Paper. Improvepolicy andprogram implementation in tourism. Accelerate implementationo f Zambia's Short to soundtourism development strategy by improving environmental mitigation practices and medium intragovemment policy coordination and collaboration. term Adopt a regional approach to marketing Zambia's tourist attractions. Work with neighboring Short- countries (through the Southern African Development Community) to establish regional tourism medium circuits. Market Zambia's tourist attractions and products inthe regional context. term Improvephysical infrastructure, public health, and safety at major tourist sites. Improve Medium domestic air transport services to Victoria Falls, road transport services to other tourist sites, and to long health and sanitation services at the main tourist destinations. Monitor and inspect service term standards and quality at the main tourist hotels. 1 1. INTRODUCTION 1.1. Zambia's growth record has been disappointing over the past three decades. Average real GDP growth fell from 1.5 percent a year in the 1970s to 1.4 percent in the 1980s and 0.3 percent in the 1990s. Since 1999 real GDP has grown at an average annual rate o f 3.8 percent, but it remains to be seen whether this rate o f growth can be sustained. High population growth persistently in excess o f real GDP growth resulted in a one-third decline inreal per capita GDP between 1964, the year Zambia gained independence, and2002. 1.2. In the early 1990s the Zambian government embarked on extensive policy reforms that included price, trade, exchange rate, and interest rate liberalization as well as privatization and institutional reforms. Between 1992 and 1995 the government decontrolled the foreign exchange market, introducing a market-determined exchange rate and full convertibility o f the local currency. Restrictions on bank lending and deposit rates were eliminated in 1993. Trade reforms in 1995-98 simplified the tariff structure, removed quantitative restrictions, and transformed the Zambian trade regime into one o f the most outward oriented in the region. Numerous parastatals were privatized. Chapter 2 assesses these policy reforms, as well as the economic and social achievements o f the 1990s. 1.3. Despite these policy reforms, real GDP grew at an average annual rate of just 1.3 percent between 1992 and 2002, while population increased 2.0 percent a year. The reasons for such disappointing performance in the face of sweeping policy reforms are not well understood. Chapter 3 takes a comprehensive view o f what promoted growth and what worked against it in Zambia in the 1990s, drawing lessons for hture growth and diversification and suggesting the way forward interms o f policy and institutional measures. 1.4. Closely related to the lack of growth is Zambia's heavy dependence on copper. Copper and its byproducts generate about 60 percent o f Zambia's foreign exchange earnings in 2002. The decline in the international price o f copper, coupled with a continuous decline in copper production since the 1970s, has contributedto Zambia's lackluster economic performance. The decision by the Anglo American Company to withdraw from copper mining operations in Zambia has given a new impetus to an active government policy toward creating a more diversified economy, in which agriculture, tourism, and light manufacturing industries play greater roles. Chapters 4 and 5 look at the opportunities for diversification and growth in agriculture and non-agriculture, and detail the main constraints as well as policy and institutional measures that can be undertaken to promote growth and diversification. 1.5. Based on a comprehensive assessment of the past, the report looks to the future to try to forecast the prospects for growth and poverty reduction in Zambia over the next decade. Building on the analyses and on some policy recommendations from the preceding chapters, Chapter 6 simulates Zambia's prospects for growth and poverty reduction through 2015. The growth and welfare impact o f several exogenous changes inZambia's economic environment as well as the impact o f alternative government policies are simulated. For exogenous changes, the growth and welfare impact o f alternative copper prices andproduction scenarios as well as the impact o f HIPC debt forgiveness are considered. For policies that are under the control o f the government, the growth and welfare impact o f increased spending on HIV/AIDS, increased 2 education spending, increased productivity in subsistence and export agriculture, and increased investment intransportation and rural roads are compared. 1.6. The Annex to this volume, which i s issued separately, contains more detailed information as well as statistical tables. 3 2. ECONOMICAND SOCIALDEVELOPMENT SINCE INDEPENDENCE ECONOMIC PERFORMANCE BETWEEN 1965 AND 2002 2.1 Landlocked Zambia is rich in agricultural and mineral resources that are largely underutilized. Some 58 percent of Zambia's total land area o f 39 million hectares is classified as having medium to highpotential for agricultural production, but less than half o f potential arable land i s cultivated. The country i s prone to drought due to erratic rainfall, as its abundant water resources remain largely untapped. Zambia has some o f the largest copper and cobalt deposits inthe world, and it was once a low-cost producer. But the aging o f the mines' infrastructure, exacerbated by decades o f underinvestment and rising extraction costs, due to depletion o f existing mines, have raised productioncosts, puttingthe sector's viability at the mercy o f fluctuations inthe international price o f copper. 4 2.2 Between 1964 and 1972 real GDP in Zambia grew at an average rate of 4 percent, buoyed by high international copper prices (Figure 2-1). The resulting influx of foreign currency increased the economy's capital stock significantly, with about a third of GDP spent on investment. Economic expansion was constrained, however, by the low average educational attainment of the labor force.3 During this period political and economic policies were liberal, with little or no state controls (Box 2-1). Figure 2-1. Selected Macroeconomic Indicators, 1964-2002 International Purchasing Power of Copper - 6500 6500 500 , Real per capita GDP 500 6000 6000 450 450 h 5500 5500 400 400 3 5000 5000 P g :::: 350 350 4500 2P 300 300 4000 2 n 9E5 3500 3500 250 250 8 .............. 3000 .u 200 I 200 2500 150 150 2000 100 io0 1500 50 50 1000 1000 n 0 Domestic Savings and Investment 60 60 50 50 +Gross domestic savings as share o f GDP 40 Gross domestic investment as share of GDP --40 c 30 I4 - 30 20 - 20 10 - 10 0 - 0 Source:World Bank2003c, Central Statistics Office 2002, unpublishedWorld Bankdatasets, and staff estimates. Average years of schooling amongpeople 15 to 64 were 2.5 in 1960and 2.8 in 1970(Barro and Lee 1996). 5 Figure2-2. EconomicPerformanceunderDifferent Policy Regimes, 1964-2002 RealGDPunderDifferentPolicyRegimes 2400.................................. 4d2 2000 h 2200................... .............. ......................... -2000 * 1800 - 1800 48 1600 1600 v .{ 1400 1400 1200 1000i 1000 Source: World Bank 2003c, Central Statistics Office 2002, unpublished World Bank datasets, and staff estimates. 2.3 Inefficientstate control of the economy put real GDP on a muchflatter long-termtrend between 1973 and 1984 (Figure 2-2). The industrializationpushthat occurred duringZambia's foray into state ownership and control over the economy resulted in a doubling o f the average physical capital endowment per effective labor unit relative to 1960-72. The increase occurred despite the steep decline in the international purchasing power o f copper in 1975 and its subsequent fluctuation around a much lower level (Figure 2-1). To maintain the high levels o f private and government consumption, the government dramatically increased its foreign borrowing and increased indebtedness. But mounting economic distortions4aused by the government's unwillingness to depreciate the kwacha in response to negative terms-of-trade shocks and heavy reliance on price controls, subsidies, and hightariffs-led to the plummeting o f average total factor productivity and a decline inreal GDP (Figure 2-l).4 2.4 Between1985 and 1990 declining capitalstock, unsustainedstabilizationand structural adjustment policies (Box 2-2), and a build-up of macroeconomic imbalances kept real GDP growth below the levels enjoyed in the past. The unfavorable external environment and the government's half-hearted attempts to break with the past made maintenance o f the physical capital stock infeasible, despite continued foreign borrowing. At the same time, rapid population growth and significant gains in the average years of schooling boosted the stock o f human capital. As a result, the average physical capital endowment per effective labor unit fell sharply in 1985-90. As foreign financing dried up in the late 1980s, the government resorted to domestic money creation, which created inflationary pressures. The combined effect of these developments was to keep real GDP growth depressed. Maintaining an overvalued exchange rate was part o f the government's import-substitution industrialization policy, as it kept the domestic prices of imported capital equipment and intermediate goods low. 6 2.5 The growthrecordin 1991-2002, a period duringwhich the government implemented stabilizationand structural adjustment programs, was mixed. In 1991-95 real GDP decline by an average o f 1.0 percent a year. During 1996-2002 real GDP grew by an average annual rate o f 3.1 percent. The remarkable similarity between real GDP trends in 1964-72 and 1996-200240th periods in which the government pursued free-market policies in relatively stable macroeconomic environmentwoints to the importance o f sound macroeconomic policies and policies that promote private sector investment for growth. THERECORDBETWEEN1991AND 2002 A. Policy and StructuralReforms 2.6 In October 1991 the Movement for Multi-Party Democracy won Zambia's parliamentaryelections on a platformof politicaland economic reforms. The new government's reform agenda had three main goals: restoring macroeconomic stability through monetary and fiscal reforms, facilitating private sector growth by relaxing or removing price and exchange rate controls and import and export restrictions, and shifting from a system o f public monopolies to one o f private and decentralized institutions. 2.7 In the following years the government implemented a number of policy reforms (Box 2-2). Exchange and interest rates were liberalized. The trade reforms adopted in 1995-98 simplified the tariff structure, removed quantitative restrictions, and transformed the trade regime into one of the most outward oriented inthe region. Several state-owned enterprises were privatized, and the government substantially withdrew from the agriculture sector (see Annex A). B. Growth Performance 2.8 At independence in 1964, Zambia was one of the most prosperous countries in Sub- Saharan Africa. Its economy, managed largely by expatriates, was dominated by minerals and mining. In 1965 value-added from mining accounted for almost half o f formal sector GDP. With substantial agricultural and mineral natural resources, the prospects for growth seemed bright. These prospects did not materialize, with the economy growing at an average annual rate o fjust 1.4 percent between 1965 and 2002, or -1.6 percent on a per capita basis (table 2-l).5 The main engines o f growth were the services sector, which accounted for 1.2 percent o f the 1.4 percent real GDP growth; industry, which accounted for 3.3 percent o f the 1.4 percent; and agriculture for 1.9 percent o f 1.4 percent. Offsetting these positive contributions, mining and quarrying accounted for -3.2 percent o f the 1.4 percent real GDP growth. On the demand side, investment and government consumption were the fastest-growing components, with investment growing at an average annual rate o f 6.4 and government consumption growing 3.3 percent a year. Between 1991 and 2002 average real annual growth o f GDP at factor cost was only 1.1 percent (Table 2-1; see Annex B). Agriculture grew 1.3 percent; industry 1.8 percent (with manufacturing growing 2.0 percent, gas and electricity 2.5 percent, and construction -0.7 percent); and services 3.3 percent (Table 2-1). Offsetting these results was an annual decline inthe miningsector o f 5.1 percent. These growth rates are the average of annual rates. 7 8 Table 2-1. Sources of Growth, by Sector, 1991-2002 Average growth rate (percent) Contribution to growth 2002 2002 98 2002 2002 2002 98 2002 GDP at factor cost 1.4 1.1 -0.2 3.8 1.4 1.1 -0.2 3.8 Agriculture, forestry, fishing 1.9 1.3 1.2 1.3 0.2 -0.14 -0.26 0.23 Mining and quarrying -3.2 -5.1 -8.7 2.1 -0.7 -0.74 -0.89 0.00 Industry 3.3 1.8 0.0 5.2 0.7 0.26 -0.10 1.04 Manufacturing 4.0 2.0 0.9 4.1 0.5 0.14 -0.01 0.46 Gas, electricity, water 9.1 2.5 1 .o 5.4 0.2 0.07 0.02 0.17 Construction -1.0 -0.7 4 . 8 7.5 0.0 0.05 -0.11 0.40 Services 3.0 3.3 2.7 4.6I 1.2 1.73 1.06 2.53 a. Calculated from the average o f constant price data for 1965, 1970, 1977, and 1994. Average annual growth rate stimated as first difference inthe natural log ofthe intertemporal variable values. b. The sumofgrowth contributions equals the GDP growth rate for theperiod. The growth contribution is calculatedas the sectoral growth rate weighted by the share ofthat sector inGDP. c. GDP at factor costs equals GDP measured at market prices minus indirect taxes plus subsidies. Excludes bank service charges. Source: Central Statistics Office, 2002, and World Bank staff calculations. 2.9 Total factor productivity-the efficiency with which capital and labor are combined to produce output-declined at an average annual rate of 2.1 percent between 1965 and 2002. Between 1991 and 2002 productivity declined 0.8 percent a year, with most of that decline occurring in 1991-98 (1.9 percent a year). Since 1999productivityis estimatedto have grownby an average of 1.4 percent a year-the first time such improvements have been seen in Zambia since independence (Table 2-2). Table 2-2. Growth Accounting Results, 1960-2002 (percent) 1991-2002 Item 1960-2002 1991-2002 1991-98 1999-02 Average annual growth rates a/ Real output per effective labor unit -1.4 -2.6 -4.1 0.5 Physical capital per effective labor unit bi 0.7 -1.8 -2.2 ,-1.o Total factor productivity -2.1 -0.8 -1.9 1.4 Share in the growth of outputper effective labor unit a/ c/ Physical capital per effective labor unit 3.0 -39.9 -39.9 -39.7 Total factor productivity 97.0 139.9 139.9 139.7 Memo items: Average annual growth rates a/) Output (real GDP) 1.9 1.3 0.2 3.4 Physical capital (unweighted) 5.2 -0.6 -1.1 0.5 Human capital (unweighted) 3.4 4.0 4.6 2.9 a/ Arithmetic averagesof annual values. b/ Weighted by the income share of capital. c/ Adjusted for 3 outliers (1961, 1963, and 1968). Outliersreplacedby average shares for the respective sub-period. See Annex B for details. Source: World Bank 2003c, Zambia, Central Statistics Office 2002 ,Statistical Information Management and Analysis, Barro and Lee dataset on intemational measureso f schooling (www.worldbank.or~research/erowth/ddbarle2.htm),and Bank staff estimates. 9 2.10 Total factor productivity declined in Zambia for about three decades for two main reasons. First, productivity and innovation are human capital intensive. Gross tertiary school enrollment-a rough indicator o f the extent to which the economy produces researchers, scientists, and technicians-has remained very low in Zambia (about 2.35 percent in 2000). And poor management pervaded the Zambian economy for much o f the period, particularly duringthe period o f state ownership. Second, with few domestic innovations and knowledge generation in Zambia, most technological progress had to come through imports o f capital and intermediate goods. However, because o f constraints in foreign exchange eamings stemmingfrom declining copper production and copper prices, there was little capital deepening. Limitedamounts o f machinery and equipment were imported during 1978-1995, which meant little or no opportunity for investors-mostly the state-to modemize and upgrade productive activities. At the same time the manufacturing sector, where productivity gains could have been achieved through sectoral reallocation o f factors o f production, remained dormant (see Chapter 5).6 2.11 During 1991-98 real GDP fell by an average rate of 0.2 percent a year. Poor performance can be traced mainly to the decline inthe mining sector, which contracted at an average annual rate o f 8.7 percent, and poor performance in the agricultural sector, which contracted by 0.2 percent a year. Industrywas stagnant (manufacturing grew 0.9 percent, utilities grew 1percent, and construction contracted 4.8 percent), and services grew 2.7 percent a year. With no growth in industry, the service sector was the only significant sectoral source o f growth offsetting the huge contraction inminingand agricultural output. Deflation in 1993-94 also contributed to steep output losses. 2.12 During 1999-2002 real GDP grew at an average rate of 3.8 percent a year. This period i s unique in Zambia's economic history since independence, as it represents the first period o f uninterrupted positive real per capita GDP growth since 1965 and the only period during which almost all sectors experienced growth each year. Industry grew at an average annual rate o f 5.2 percent (with manufacturing growing 4.1 percent, utilities 5.4 percent, and construction 7.5 percent); the services sector grew 4.6 percent a year; and agriculture grew 1.3 percent a year. The decline in the miningsector was reversed and the sector actually grew, at an average annual rate o f 2.1 percent. The privatization o f key mines and the resulting infusion o f cash from investors allowed mining sector exports to grow at an average annual rate o f 7.9 percent. Growth thus seemed to come from two sources: higher growth rates across all sectors, perhaps suggesting that the reforms o f the early and mid-1990shad begun to pay off, and the reversal o f the declining mining sector, obviously as a result o fthe privatizationo f the Zambia Consolidated Copper Mines. C. Macroeconomic Policies 2.13 Although Zambia made significant progress in structural reforms during the 1990s, full macroeconomic stability has remained elusive. Gains in price stabilization have slowly been consolidated since 1996, but inflation i s 20-30 percent, real interest rates are 6-21 percent, the kwacha i s volatile, and external indebtedness and debt service remain high(Table 2-3). Ineconomic developmentsuccess stories suchas that of Taiwan (China), the sectoralreallocation ofresources from agriculture to industryaccountedfor about half of the rapid growth intotal factor productivity (Dessus and others 1995). 10 Table 2-3. SelectedMacroeconomicIndicators,1991-2002 (percent) Item 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Inflation rate (Consumer Price Index) 92.6 197.4 188.1 53.6 35.2 45.2 24.5 24.4 26.8 30.1 21.7 22.2 Nominal lending rate 36.5 47.0 124.0 74.2 45.5 53.8 46.7 31.8 40.5 38.8 46.2 45.2 Real lending rate -29.1 -50.6 -22.2 13.4 7.6 5.9 17.8 5.9 10.8 6.7 20.1 18.8 Exchangerate(kwacha/US$) 64.6 172.2 452.8 669.4 866 1207 1315 1862 2388 3110 3608 4398 Fiscal deficitbercent GDP) -7.0 -2.5 -5.6 -4.4 -3.8 -0.5 -0.2 -5.2 -3.7 -5.7 -6.9 -6.1 Extemal debt (percent GNI) 232.8 234.0 214.8 218.7 200.7 215.5 170.2 212.0 209.7 216.8 168.8 168.1 Debt service (percentGDP) 19.6 10.7 10.0 11.5 45.7 9.7 5.6 4.6 4.3 5.5 2.9 4.3 Source: World Bank, 2003c. Injlation 2.14 The mountingdomestic public debt has exacerbatedfiscal deficits,withnegative implicationsfor inflation. The budget's unabateddomestic financing requirements, coupled with the need to service high-yielddebt and roll over maturing issues, has kept the cost o f servicing domestic public debt high. The Bank o f Zambia's monetizationo f fiscal deficits i s the main reason why the government has beenunable to reduce inflationto below 20 percent. 2.15 The budget's dependenceon foreign financing, the government's inabilityto adhere to agreements madewith donorsfor foreign aid, and unrealistic budget assumptionshave created inflationarypressures. Govemment budgets are frequently preparedunder implausible assumptions about the level o f donor assistance and the government's ability to meet the conditions for its release. As a result, actual disbursements frequently fall short o f pledges and usually arrive later than forecasted. The result i s persistent budget deficits and heavier use o f Bank o f Zambia funds by the government, which fuels inflation. 2,16 The financial sector liberalization of the early and mid-1990s strained monetary control and fiscal deficits. Financial sector liberalization policies enactedinthe early 1990s opened the Zambian banking sector to new entrants and gave banks discretion in determining their own lending practices. But the process was not accompanied by improvements in risk management practices by banks or the adaptation o f banking supervision procedures by the Bank o f Zambia. Risky lending to attract new customers, lower revenue from foreign exchange operations, lower Treasury bill yields, periodic shortages o f liquidity, limitations in raising capital, and additional competition for small banks from emerging non-bank financial institutions caused the financial situation o f some banks to deteriorate. The resulting turmoil led to the closure o f nine commercial banks between 1994 and 1998, at a cost estimated at 7.6 percent o f average annual 1996-2001 GDP. More than half o f these costs were bome by the government and the Bank o f Zambia, straining monetary control and fiscal deficits. Interest Rates 2,17 Lending interest rates have significantly fluctuated in Zambia since 1994. Between 1994 and 2002 the real commercial bank lending rate averaged 16 percent, reaching as high as 18.8 percent in2002. 11 2.18 The large spreads between lending and deposit interest rates are caused by lack of competition, high levels of unremunerated reserve requirements, and the high degree of dollarization of bank deposits. Because o f the high cost o f operating branches, only three banks own 102 o f the 156 bank branches in Zambia. Domestic banks operate 73 branches. The impact o f unremunerated reserve requirements on the spread depends on banks' market power over depositors and borrowers. Such market power i s perceivedto be highinZambiaa7 2.19 High lending rates have stifled private borrowing and undermined private investment and growth. At 8 percent, the ratio of private sector credit to GDP inZambia is one o f the lowest in Sub-Saharan Africa. In recent years commercial banks have preferred to invest in banks abroad and ingovernment debt: by mid-2003 such investments accounted for almost halfof their aggregate total assets. Another 17.5 percent o f kwacha and foreign currency deposits are held in nonremunerated deposits with the Bank o f Zambia. Commercial banks derive more than half o f their interest income from government securities. About 15-35 percent o f their income comes from foreign exchange activities, including foreign exchange trading and fees on foreign exchange transactions. Exchange rate 2.20 Since the liberalization of the foreign exchange market in late 1992, the kwacha has lost 95 percent of its externalvalue with respectto the U.S. dollar in nominalterms. Inthe short-run the nominal exchange rate reacts to changes indemand and supply conditions. The long-runbehavior of the nominal exchange rate reflects the precipitous decline inthe domestic purchasing power o f the kwacha. 2.21 The supply of foreign exchange depends to a large extent on copper exports, which generate more than half of Zambia's foreign exchange earnings, and on Bank of Zambia interventions in the foreign exchange market. The delayed privatization o f the copper mines and unfavorable external terms o f trade constrained the supply o f foreign currency throughout most of the 1990s. The rules o f the foreign exchange auction organized by the Bank o f Zambia and the behavior o f the auction's participants effectively turned it into a quasi-administrative allocator o f marginal funds by the Bank o f Zambia.* The Bank o f Zambia tended to keep the auction rate lower than the market rate inperiods o f pressure on the exchange rate, which helped smooth short-run fluctuations in the nominal exchange rate. But its actions diminished the efficiency o f the price discovery process, resulting in occasional overshootings and subsequent corrections o f the auction rate, which contributed to the volatility of nominal and real exchange rates. In2003 the foreign exchange auction organized by the Bank o f Zambia was discontinued, replaced by an interbank foreign currency market 'The oligopolistic market structure o f the banking sector i s another possible cause o f the downward rigidity o f bank lending rates. The Zambian banking system includes 14 commercial banks, but 6 are extremely small and 5 hold 75 percent o f aggregate banks' total assets. 8 The Bank o f Zambia's auction was the second-largest segment (more than 10percent) o f the foreign exchange market in Zambia. There was no official surrender requirement, but suppliers o f foreign currency were not allowed to sell more than US$lOO,OOO a week to an individual bank, making the Bank o f Zambia auction the only alternative for large exporters. Often the amounts allocated were substantially different from the amounts initially offered (the Bank o f Zambia received offers from major suppliers the day before the tender day and announced the amount that would be offered at the auction), as the Bank o f Zambia could reject bids without explanation and adjust the size o f its foreign exchange dealings. Buyers tended to bid up to the maximum allowed (40 percent o f the announced amount), leading to a situation in which the Bank o f Zambia effectively determined the foreign exchange supply infull knowledge o f the demand schedule. 12 managed by commercial banks, to further entrench the mechanism o f a market-determined exchange rate. 2.22 Since the 1990sthe real exchange rate has exhibited highlyvolatileshort-run dynamics arounda slightly downward-slopinglong-termtrend. Misalignment inthe realexchange rate (that is, sustained departure o f the observed real exchange rate from its equilibrium value or trend) constrains economic growth, by distorting prices and leading to the inefficient allocation o f productive resources. Zambia's equilibrium real exchange rate is highly sensitive to the level o f external debt and debt servicing requirements. An increase in the level o f debt and debt service causes the real exchange rate to depreciate. Huge aid flows are likely to lead to overvaluation, reducing the competitiveness o f exports. 2.23 The high volatility of real exchange rates inZambia has hurt the real sector. Exchange rate volatility creates uncertainty for private investors interms o f both the profitability and the cost of investment. Volatile real exchange rates are associated with erratic swings inthe relative profitability o f investment in the traded and nontraded goods sectors (Serven 2002). The cost o f capital goods also becomes uncertain, because o f the high import content o f investment. Exchange rate volatility also makes local banks unwilling to offer credit denominated inforeign currency. On the export side, exports that are invoiced in nonlocal currency are usually negatively affected by exchange rate volatility, since exporters have to absorb currency risk (Qian and Varangis 1992). External debt I Figure 2-3. Zambia's Total External Debt, I 1970-2002 2.24 Zambia's external debt has grown rapidly since the 1970s. Total external debt stood at US$7 billion in 1991 (233 percent o f gross domestic *fi income) and at US$5.11 billion in 2002 7 (168 percent o f gross domestic income). On a per capita basis, these debt levels are among the highest in Sub-Saharan Africa. The government reduced debt service a 4 (II 3 payments from about 20 percent o f GDP d 3 in 1991 to about 4 percent in 2002 by g2 rescheduling some o f its debt.g 1 O , , , , , , , , , , ,0 A 2.25 Zambia's external debt is high L , , , , , , , , , , , , , , and unsustainable. Recent estimates show that the net present value o f debt to Source: World Bank, 2003c exports ratio will be 157 percent even In 1999 the Paris Club agreed to restructure its loans to Zambia by writing off two-thirds of the debt and rescheduling the remaining third on more concessional terms. The Paris Club negotiations have been considered the f r s t steps toward further debt reductions through the Heavily Indebted Poor Country (HIPC) initiative, which covers the bulk of Zambia's multilateral debt stock. Zambia qualified for the HIPC decision point in December 2000. If the decision point becomes irrevocable (that is, if Zambia reaches the HIPC completion point), it will reduceZambia's stock of debt by about two-thirds. As of early 2004, Zambia hadnot reachedthe HIPC completion point. 13 after Zambia reaches the HIPC completion point (World Bank 2002a). A reassessment o f debt sustainability in view o f the new outlook for growth and exports i s necessary to evaluate Zambia's need for hrther debt relief beyondthe completion point to achieve sustainability. D. Poverty and Social Outcomes 2.26 Zambia's disappointing economic performance in the 1990s i s mirrored by its poor and deteriorating economic and social conditions (Table 2-4). When asked to rank the factors that determinate well-being, many Zambians put adequate food at the top o f their list-this in a country that was once among the wealthier inAfrica. Table 2-4. Trends in Selected Measures of Well-Being, 1990-2000 Indicatodmeasure Early 1990s 1998 Income/consumption poverty Overall poverty (percent) 70 73 Extreme poverty (percent) 58 58 Moderate poverty (percent) 12 15 Health and nutrition outcomes Life expectancy at birth(years) 49 38150* Infant mortality rate (per 1,000 live births) 123 110 Under-five mortality rate @er 1,000 live 151 162 births) Stunting (percent, 5 years and under) 39 47 Education outcomes Gross primary enrollment (percent) 99 78 Net primary enrollment (percent) 77 67 Adult literacy rate (15+) Male 76 77 (percent) Female 56 58 Literacy rate 15-24 (percent) Male 79 75 Female 71 66 Selected indicators: access topublic goods and services Medically assisted delivery (percent) 51 43 Mother received tetanus toxoid injection 82 76 (percent) Access to safe Urban 88 86 drinkingwater Rural 28 30 (percent) Access to electricity Urban 44 (percent) Rural 2 - ~ Not available. Source:Poverty numbers are from Zambia Central Statistics Office, PSI 1991 and LCMS 1998. Infant and under-five mortality rates are from Zambia's 1990 and 2000 population censuses. Stunting is based on the Zambia 2001/02 Demographic and Health Survey. Gross and net enrollments for the early 1990s are drawn from World Bank 2003b. Gross and net enrollments are reported for 2001 and based on the Basic Education Sub-sector Investment Program(BESSIP) Core IndicatorsPerformance Report, December 2002. Literacy figures are fromZambia Central Statistics Office, 1990 and 2000 population census Tables. Figures on medically assisted deliveries and tetanus toxiod coverage are based on the 1992 Zambia Demographic and Health Survey and the 2001/02 Zambia Demographic and Health Survey. Figures on access to water and power are based on 1990 and 2000 population census data. Note: * There are two very different figures on life expectancy. The World Bank's estimate for late 1990s is about 38 years. CSO's estimate for 2000, based on 2000 population census data, i s 50 years. A reconciliation o fthe two estimates i s warranted. The Bank and the Govemment teams have agreed to work together to resolve this issue. 14 2.27 Income (consumption) poverty is high, with 70-75 percent of the population falling below the official poverty line. Aggregate welfare levels did not improve during the 1990s (Table 2-5). Ruralpoverty has remained high, drivenby volatility inthe economy (particularly linked to weather and macroeconomic shocks), poor performance in the agriculture sector, and declining output in other sectors. Rural poverty increased from 88 percent in 1991 to 92 percent in 1993, falling to 83 percent in 1998. While poverty i s lower on average in Zambia's cities and towns, the poor make up a significant and rising share of the urbanpopulation. The percentage of poor people in urbanareas rose from 47 percent in 1990to 56 percent in 1998. Poverty i s widespread throughout the country, althoughlevels are lower inmore developedprovinces, such as Lusaka and the Copperbelt. Table 2-5. Percentageof PopulationLivingBelowthe PovertyLine,byRegion, 1991-98 Region 1991 1993 1996 1998 Zambia 70 74 69 73 Residence Urban 47 45 46 56 Rural 88 92 83 83 Province Central 70 81 74 77 Copperbelt 61 49 56 65 Eastern 85 91 82 80 Luapala 84 88 79 81 Lusaka 31 39 38 52 Northern 84 86 84 81 North-Western 75 88 80 76 Southern 79 87 76 76 Western 84 91 84 89 Source: World Bank 2001. 2.28 Rural households spend nearly three-quarters of their budget on food, with a high percentagedevotedto in-kindhome consumption. Urban households spend less on food, but they spend significantly more on basic services, such as education and urban utilities. On average rural families receive just over half their income from farming; livestock and self-employmentbusiness earnings are also important. In contrast, urban households receive nearly half of their income from wages (45 percent in 1998) and one-third from business activities and self employment. The poorest households earn significantly less from the wage sector (25 percent of total earnings), depending more on self-employment and income from farming (inperi-urban locations). 2.29 Lifeexpectancyfell49 in the early 1990s to 38 years in 1998 (Table 2-4), andmortality has risenfor both adults and children(Table 2-6). Infant mortality fell duringthe 199Os, but rates are still above the levels achieved two decades ago. Under-five mortality has continued to rise, reaching 162 deaths per 1,000 live births by the end of the 199Os, substantially above the levels recorded two decades ago. Among the very poor and other vulnerable groups, as many as one-quarter of children do not live past their fifth birthday. And child mortality in some of Zambia's poorer and more isolatedprovinces averageswell over 200 deathsper 1,000 livebirths. 2.30 The level of adult mortality is very high, in large part due to the HIV/AIDSpandemic. By the end o f 2001, 1million adults and 200,000 children inZambia were livingwith HIV infection, and 120,000 people had died of AIDS-related diseases (UNAIDS 2002). By 2010, 35-40 percent of the populationunder 15-some 2 million children-are expected to be orphaned. 15 Table 2-6. Infant and Under-Five Mortality, by Region, 1980,1990, and 2000 Infant mortality rate Under-$ve mortality rate Item (deathdl,000 live births) (deathdl,000 live births) 1980 1990 2000 1980 1990 2000 Zambia 99 123 110 121 151 162 Type region Urban 89 106 91 108 128 126 Rural 106 133 117 132 164 180 Province Central 81 105 100 100 129 144 Copperbelt 87 109 91 97 132 126 Eastern 128 149 129 177 206 196 Luapala 127 161 132 161 199 224 Lusaka 87 106 88 106 129 126 Northern 104 137 130 127 169 180 North-Western 77 103 83 95 126 137 Southern 94 97 93 115 118 138 Western 106 141 140 132 175 201 Source: Population Census o fZambia 1980, 1990,2000. 2.31 The widespread prevalence of HIV/AIDS, particularly in urban areas and among women of childbearing age, puts tremendous strains on a system already failing to deliver basic services and maintain living conditions. Infection rates among the poor are not the only concern: HIV prevalence is particularly highamong educated workers inurban areas, many of whom are an important means o f support for extendedfamilies living inthe countryside. 2.32 HIV/AIDS is by no means the only factor accounting for high mortality. Malnutrition is high and on the rise. By 1998 stunting (low height for age) among young children had risen to 47 percent. Maternal mortality i s also high, estimated in the mid-1990s at 650 deaths per 100,000 deliveries. Access to services that contribute to safe deliveries deteriorated duringthe 1990s. Table 2-7. SchoolEnrollments by Boys and Girls, 1998 Category Rural Urban Age in years Nonpoor Poor Poorest 30 Total Nonpoor Poor Poorest 30 Total 7 45.1 29.9 22.8 28.6 70.5 48.7 25.8 54.0 8 64.2 59.0 38.8 50.9 84.1 74.1 51.5 74.4 9 80.1 72.8 54.1 65.6 92.8 83.7 72.2 85.3 10 80.5 73.2 60.2 68.2 95.3 86.0 78.4 87.9 11 84.1 74.8 70.0 74.1 94.7 89.9 74.7 89.4 12 79.6 83.6 64.8 74.7 91.0 84.5 71.4 84.9 13 82.6 78.2 65.2 72.5 84.7 82.2 73.9 81.7 Gender Male 74.2 65.8 53.6 61.5 86.7 77.6 62.9 78.7 Female 72.3 67.2 51.8 61.0 87.2 78.0 65.1 79.4 Total 73.3 66.5 52.7 61.3 87.0 77.8 64.0 79.1 Source: Zambia, Central Statistical Office, Living Conditions Monitoring Survey 1998. 2.33 School enrollment rates fell during the 199Os, and they are lower for the poor than for the nonpoor (Table 2-7). The discrepancies are narrowing, however, suggesting that sharp 16 discrepancies in literacy between the poorest and wealthiest households, as well as between men and women, have lessened over time. 2.34 Poor households in Zambia have low levels of human and physical capital and lack the means to respond adequately to natural or economic shocks. Policies and programs are neededto address the problem. 17 3. WHY DIDN'TZAMBIA GROWFASTERDURINGTHE 1990s? 3.1. Zambia's lackluster economic performance in the 1990s has raised doubts about the appropriateness o f the policy advice givenby international financial institutions and bilateral donors. This Chapter examines the policies Zambia adopted between 1991 and 2002. It identifies which policies worked well and which didnot and draws lessons about future growth andpoverty reduction. 3.2. Excluding the one-time disruption in real sector activity in 1994-95, real GDP in Zambia grew at an average rate of 3 percent during 1991-2002. Could Zambia have grown faster? Past performance suggests that it could have. During 1999-2002, a period of relative normalcy after the extensive policy reforms of the early and mid-l990s, real GDP grew at an average rate of 3.8 percent a year-a much higher rate than the 1.4 percent average inthe post-independence period or the period between 1991 and 2002. Some estimates suggest that Zambia has the long-term potential to grow at a rate of about 5 percent a year, with annual per capita income growth of about 3 percent (World Bank, 2003d)". 3.3. What explains the gap between historical and potentialgrowth?The literatureon growth shows that policies, institutions, and governance are central to the growth process (Easterly and Rebelo 1993; Fischer 1993; Dollar 1992; Gramlich 1994; Jimenez 1995; Knack and Keefer 1995). A framework in which growth outcomes reflect policies, institutions, governance, and exogenous shocks can shed light on why Zambia did not grow faster. The framework used here assesses Zambia's macroeconomic policies (policies on inflation, interest rates, and exchange rates); investment climate (infrastructure, regulatory and administrative barriers, institutions); trade policy and export diversification; and policy implementation and reform ownership. It assesses Zambia's record on corruption, the rule of law, public expenditure management, and accountability. It also examines the impact of HIV/AIDS and copper price and weather shocks on growth. The report assesses the extent to which such policies and institutions have been supportive of growth and export diversification. Basedon this retrospective assessment, policy recommendations are drawn. But first the good news about what has worked for growth inZambia inrecent years. WHAT POLICIESWORKED WELL? 3.4. Inthe early 1990sZambia initiated a far-reaching program of privatization that picked up in the second half of the decade. Although the government initially opposed privatizationof the state copper mines, it sold the mines in 2000, following growing fiscal pressure and pressure from donors. lo also the See IMF macroeconomic framework for Zambia, where projected growth rate is often about 5 percent. I t should be noted, however, that the 5% does not define the upper bound on growth rate. Indeed, the economy can grow by more than 5 percent if it were, for example, able to attract foreign direct investment (FDI) insignificant quantities. 18 3.5. There is no consensus on the overall achievement of Zambia's privatizationprogram. A recent World Bank study (2002d) concludesthat privatization did stop the financial hemorrhage to loss-making companies from the national budget and enable potentially profitable companies to survive. Butmanypeople inZambia-both supporters and opponents of privatization-seem to think that privatization was done too quickly, without a thorough study of the capacity of new owners to manage the privatized companies and the implications of terminating the government's social obligations, such as providing employment and social safety nets. In response to this concern, the new government has repudiated some of the previous government's policy commitments on privatization and indicatedto donors that it will not privatize the remaining parastatals as planned but instead commercialize them. 3.6. Privatizationof the copper mines unambiguously improvedinvestment in miningand reversedyears of decline in copper output. The steep output decline inthe miningand quarrying sector between 1990 and 2000 was caused by inadequate investment and delays in privatizing the copper mines. By the time Zambia Consolidated Copper Mines was privatized in March 2000, real output from mining and quarrying was just a third of its 1990 level. Copper output grew 15 percent in 2001. Rapid declines in world copper prices during 2000-2002 nevertheless caused Anglo American, the buyer, to leave Zambia inAugust 2002. 3.7. A process of export diversification began during the mid-1990s (Figure 3-1). Diversification has been one of the government's main stated obiectives since the mid-l970s, but little progress was made Figure 3-1. Shares of Selected Nontraditional Exports in Total until the mid-990s. In Exports, 1990-2001 1992 copper and cobalt exports still constituted 12 about 90 percent of merchandize exports. 3.8. Largely because o f the decline in copper exports, the share of ................. nontraditional export in total exports increased, from about 10 percent in 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 the early 1990s to 28 --+-R i t ~ ~ ~ P g R o d U--4--T&le~ d s percent in 1999-200 1, +RecessedFood +RoliaJlturdFKdlKh growing 18 percent a --m--Hsrticulturdproducts terms. year in current dollar Exports of primary agricultural commodities, floricultural products, horticultural products, textiles, and processed food rose significantly. Growthwas stimulatedby introduction of an export credit scheme financed by the European Union and the World Bank, as well as by foreign investment attracted to the floriculture andhorticulture sectors. The increase in exports i s an encouraging development and a strong indication that the economy has responded to the improved policy environment. Zambia's nontraditional exports, however, seem to have failed to sustain growth since 1998 because of various constraints, includingthe Asian financial crisis o f the late 1990sand still unfavorablemacroeconomic environment. 19 WHATCHALLENGESLIEAHEAD? 3.9. Several factors undermined the growth and diversificationpotentialof Zambia in the 1990s and continueto do so today. Key among them are the poor macroeconomic environment, the weak reform ownership and policy implementation, the poor investment climate, trade policy, governance and institutions, the HIV/AIDS pandemic. A. ReformOwnership andPolicyImplementation 3.10. The failure to diversify exports reflects the problems Zambia has had implementing policy since the 1980s. Several World Bank reports on Zambia emphasized the need for export diversification. The 1984 Zambia Country Economic Memorandum, which focused entirely on this issue, offered several recommendations. Why were those recommendations not implemented? Throughout the 1980s high-level politicians, bureaucrats, and party officials all believed that copper prices would eventually recover (West 1992). As a result, Zambia missed an opportunity to diversify the economy into agriculture. Some limited achievements in diversifying exports and reducing dependence on copper for foreign exchange were witnessed in the 1990s, but a well-coordinated vision and strategy for export diversification-including an effective strategy for removing the anti- export bias in the economy-was lacking. Indeed, it was not until the Anglo American Corporation announced its withdrawal from the Zambia Copper Mines in January 2001 that the need for diversification o fthe economy was explicitly put back on the government's agenda. 3.11. Policy implementation remained weak during the second half of the 1990s. Several studies show that the government's commitment to policy reforms waned during the second half o f the 1990s (Rakner, van de Walle, and Mulaisho 1999). Privatization o f the largest parastatals inthe copper, energy, telecom, and financial sectors was beset by problems and delays, the government failed to implement civil service reforms, and despite efforts by donors to help the government improve its public expenditure management, commitment to fiscal discipline remained elusive. Three areas in which the government's lack o f commitment to reform and policy implementation have seriously undermined macroeconomic stability and growth are lack o f fiscal discipline, the slow privatization o f the Zambia Consolidated Copper Mines, and weak implementation o f the public sector reform program. 3.12. Since 1993 Zambia has maintained a cash budgetsystem, which is supposedto prohibit monetary financing of deficits. In practice, the government has violated its zero monetary financing rule (Dinh 2002). Following each episode, the government has reduced expenditures. Since 1994 fiscal discipline has gone o f f track inthe early part o f the year and been reimposed later in the year by cutting expenditures or introducing ad hoc revenue measures. Expenditure squeezing has fallen mostly on nonwage expenditures and capital expenditures. Budgetary constraints appear to be respected, but the government fails to pay suppliers, creating huge arrears. Despite austerity measures, the government has taken on large quasi-fiscal obligations, which have resulted in huge budget overruns. Arrears accumulation and rising domestic debt as well as unbudgeted wage hikes and retirement packages for civil servants have resulted in significant budget deficits. In short, the cash budget, which was meant to curb fiscal deficits and provide a foundation for single-digit inflation andreal interest rates, has remained mere window dressing. Lapses infiscal discipline have in turn made it difficult to create a conducive macroeconomic environment for growth and diversification. 3.13. Implementationof public sector reform has been weak, The government failed to carry through the 1993 Public Sector Reform Program, which promised to cut 25 percent o f the civil service within three years and improve the conditions o f service for the remaining staff in order to 20 retain and attract the best people. Rather than shrinking, the civil service grew 20 percent between 1990 and 2001. While 15,000 contract daily employees with nojob security were dismissed in 1993, no retrenchment o f civil servants had taken place by early 1999. The main obstacle to retrenchement seems to be cost: since the legally mandated retrenchment costs amount to 10 years' salary for each employee, the most economic approach for the government has simply been to retain workers. In addition, retrenchment i s would be politically tremendously unpopular and unemployment would cause severe difficulties for those involved. 3.14. A renewed effort at public sector reformbegan at the end o f the 199Os, supportedinpart by the World Bank's three-phase Public Sector Capacity Building Project (PSCAP). The initial focus has been to put in place the basic financial management, accountability and transparency, personnel, and other systems and begin to reform wages so that key staff can be attracted, retained, and motivated to provide quality services, while continuing to right-size and restructure the public service. While considerable capacity buildingtook place inthe early 2000s (with PSCAP assistance), several key policy reforms-such as public sector rightsizing and pay reform-remain unaccomplished. In short, the public sector reform program, deemed essential to both balancing the budget and enhancing the capacity o f the bureaucracy for service delivery, remains by and large unimplemented. Some observers have regarded this continued failure to address the large and inefficient public sector as the main obstacle to service delivery and private sector-led economic growth (Rakner, van de Walle, and Mulaisho 1999). 3.15. Privatization of the copper mines has been slow. The privatization program gained momentum in 1996, but privatization o f the largest parastatals was beset by problems and delays. The government could not agree on the approach for privatizing the copper mines (selling them as a single unit or unbundling them into smaller units)." The government produced a Zambia Consolidated Copper Mines Privatization Plan, which was implemented in the summer o f 1996, but privatization took nearly three and a halfyears. 3.16. Why was policyimplementationand ownership poor?The government was unilateraland not consultative inits implementation o f economic policy, which didnot bode well for ownership and implementation (see Annex A on the political economy o f reforms). The quality o f public policy as well as the chance for implementation could have been improved by involving all stakeholders in public policy debates. The lesson-for both the government and donors-seems to be that formal structures may not be sufficient in explaining economic policy failures in Zambia and that consultation o f all stakeholders could improve implementation. 3.17. Lack of effective donor coordination also contributed to weak implementation of policiesand programs. A partial reform syndrome has been supported inZambia by the inability of donors to impose conditionalities coherently (Rakner, van de Walle, and Mulaisho 1999; World Bank 2002~). In 1996 bilateral donors' concern with governance issues led them to reduce aid allocations at a time when multilateral donors, particularly the World Bank, believed the government deserved increased support. The different, and often inconsistent, motivations o f donor agencies diluted the impact o f conditionality measures. Lack o f donor coordinationreducedthe government's incentive to implement reforms. l1The differences were so intense that some leaders even lost their positions. A deputy minister was dismissed for openly favoring selling of the mines as a single entity. 21 B. MacroeconomicEnvironment andInvestmentClimate 3.18. Despite a decade of policy and structural reforms and the tremendous progress achieved by the mid-l990s, the macroeconomic environment in Zambia is not yet fully conduciveto privateinvestmentand growth(see Chapter 2). Highinflation, erratic exchange rate movements, and high real interest rates have created uncertainty and risk, discouraging private investment and growth. High real interest rates, combined with the low level o f financial intermediation for the private sector by financial institutions, have severely limited access to affordable long-term finance by investors. 3.19. The share of private investment in total domestic investment has declined in Zambia sincethe mid-l990s, and foreign direct investment has dropped since 1994. l2Both likely reflect the poor investment ~1imate.l~The investment climate reflects the quality o f a country's infrastructure, the enforcement o f the rule o f law and control o f corruption, and the effectiveness o f i t s institutions (Kaufmann, Kraay, and Zoido-Lobat6n 1999; Knack and Keefer 1995). Macroeconomic Instability and the High Cost of Capital 3.20. Zambian firms report that macroeconomic instabilityand the high cost of capitalare the most serious constraints to private sector investment and growth. The World Bank, in collaboration with local partners, carried out an Investment Climate/Administrative and Regulatory Cost Survey to evaluate the investment climate in Zambia in 2003. The survey was administered to about 350 Zambian enterprises, drawn from the list o f companies that paid taxes in2002. The sectors covered included agro-industry, chemicals and paints, construction, mining, metals, paper, wood, leather, and textiles. According to the survey, the top constraints to investment are macroeconomic instability and the high cost o f credit (Table 3-1). Other factors discouraging private investment and growth include regulatory uncertainty and high tax rates, poor physical infrastructure and the high cost o f utilities, bureaucratic barriers associated with investor entry andbusiness registration, customs and tax administration, and poor government agency services. Table 3-1. Constraintsto BusinessOperationsinZambia (percent of firms evaluatingconstraint as "major"or "very severe") Small Large Foreign Constraint Zambia enterprise enterprise Nonexporter Exporter owned Domestic 1Cost of financing 82.1 85.1 80.3 82.2 81.8 15.4 84.9 2 Macroeconomic instability 73.9 81.8 65.2 I 9 60 10.5 15.3 3 Tax rates 57.5 59.7 56.1 59.2 52.7 54.1 58.9 4 Regulatory uncertainty 57 58.4 60.6 59.2 50.9 59 56.2 5 Access to finance 54.1 66.2 42.4 56.6 41.3 41.5 56.9 6 Crime, theft, fraud and disorder 48.8 49.4 53 50.66 43.64 41.54 49.3 7 Corruption 46.4 54.6 39.4 47.4 43.6 50.8 44.5 8 Electricity 39.6 32.5 . 50 34.2 54.6 41 39 9 Anticompetitive behavior 38.7 42.9 28.8 40.1 34.6 34.4 40.4 l2 This section is based on a 2003 World Bank Investment Climate Survey and an Administrative and Regulatory Cost Survey. The results are presented inWorld Bank 2004b and World Bank 2004c. l3 Recent work on economic growth (Easterly and Levine 1999) has shown that there i s surprisingly little relationship between the quantity o f investment and the rate o f economic growth due to poor policy and institutional environment for investment. 22 10Skills andeducation ofworkers 35.8 36.4 37.9 36.8 32.7 31.2 37.7 11Telecommunications 32.9 32.5 37.9 26.3 50.9 29.5 34.3 12Customs andtrade regulation 32.4 24.7 34.9 32.2 32.7 34.4 31.5 13 Transportation 30.4 28.6 30.3 27.6 38.2 29.5 30.8 14Tax administration 27.5 32.5 27.3 27 29.1 34.4 24.1 15Access to landsecurityo ftenure 17.4 22.1 16.7 20.4 9.1 23 15.1 16Laborregulations 16.9 11.7 21.2 15.1 21.8 16.4 17.1 17Business licensing andpermits 10.1 7.8 9.1 11.2 7.3 13.1 8.9 Source: World Bank 2004c. Policy Uncertainty 3.21. Twelve years of liberalization have done little to bridge the continuing mistrust between the private and public sectors in Zambia. Since 1991 the dialogue between the government and the private sector has improved only marginally. The government's reputation as inefficient and interventionist persists. Mutual antipathy between the government and the private sector has negatively affected private sector investment and growth. 3.22. Regulatory and policy shifts increase uncertainty and risk for private sector investment. Policies are inconsistent over time. The immigration laws recently required all non- Zambians to renew their unexpired permits at prohibitive costs. Mineral royalties were recently increased by 150 percent. Some export-oriented agricultural products were given incentives, while others were not. Moreover, subordinate policies are not in synch with broader policies. For example, while the Investment Act indicates that all sectors o f the economy should be open to foreign investors, the work permit regulation has effectively restricted the sectors open to individual investors, andthe immigration policy has remained protective. Poor Physical Infrastructure 3.23. Zambia's physicalinfrastructuredid not improve in the 1990s (Table 3-2). The quality of infrastructure in Zambia remains poor: more than two-thirds o f the road network i s in poor ~ondition,'~the roads outside o f Lusaka are ridden with potholes and nearly impassable in industrial areas, the supply o f electricity throughout Zambia i s unreliable and system losses are significant (11 percent), telecommunications costs are high, and water supply i s erratic in many o f the industrial areas, forcing companies to invest in boreholes. These deficiencies drive up the cost of doing businessand make it difficult for Zambian companies to compete regionally or domestically. Table3-2. Stock ofPhysicalInfrastructure,1980sand 1990s 1980s 1990s Roads Total road length(km) 37,790 37,359 Paved road length(km) 5,923 6,575 Rail route length(km) 1,228 1,273 Telephone Telephone mainlines per 1,000 people 6.94 9.32 l4 The condition o f the paved road network has improved in recent years. As of 2002, 50 percent o f the core roadnetwork was classified as good, 40 percent as fair, and 10percent as poor. 23 Telephone mainlines per employee 17 24 Telephones (number) 79,878 108,784 Telephone mainlines (number) 48,172 75,867 Electricity Electricity production (million kwh) 9,201 7,857 Electricity generating capacity (million kwh) 2,082 2,436 Source: Canning 1998. 3.24. Zambia's electricity generating capacity increased marginally in the 1990s, but the supply of electricity remains unreliable (Table 3-3). The Zambia Electricity Supply Company is under strain to provide consistent and affordable power. Zambian firms must wait 120 days on average to get access to electric power. Power cuts to the industrial areas in the major towns are frequent. Zambian firms recorded an average o f 37.2 power outages in 2003, with large firms averaging 45. Production lost due to these outages average 4.5 percent (5.6 percent for small firms). More than a third o f all firms (and more than 60 percent o f large firms) have their own generators. Table 3-3. Qualityof PhysicalInfrastructure,2003 Foreign Item Zambia Small Large -owned Domestic Exporter Nonexporter Roads Percent of production lost inshipment 3.8 4.7 2.3 3.3 4 2.6 4.2 Electricity Numberofdays to obtain electricity connection 120.7 47 163 313.3 48.5 90 127.6 Number ofpower outages last year 37.2 29.8 44.9 35.4 38 41.7 35.6 Percent of production lost due to power outages 4.5 5.6 4.3 3.7 4.9 3.3 5 Have own generator ( percent) 38.2 27.3 60.6 42.6 36.3 63.6 29 Telephone Number ofdays to obtain telephone connection 132.5 135 21.7 283.3 67.9 102.5 152.5 Water and sanitation Have built own well (percent) 59.9 42.9 81.8 59 60.3 72.7. 55.3 Percent ofproduction lost inshipment 3.8 4.7 2.3 3.3 4 2.6 4.2 Source: World Bank 2004c. 3.25, Opportunities to develop cellular telecommunications have been missed by failure to commercialize the Zambia Telecommunications Company (ZAMTEL). ZAMTEL has a total exchange capacity o f 147,020 lines and 86,895 customers. The two other private cellular phone companies have 250,000 customers. Although its importance i s declining, ZAMTEL still controls key communications gateways. It takes a firm 92 days on average to obtain a cellular telephone line. Cellular phone service inneighboring countries i s far less expensive than inZambia, where a duopoly o fprivate phone producers controls the market. IfZAMTEL were more commercial, it would be able to provide better service at a more competitive price. ZAMTEL's inadequate performance also inhibits the widespread use o f Internet technology. As a result, businesses are unable to cheaply access global information and business networks, which are becoming increasingly important for exporters. Government agencies that are investing in Web sites in order to improve their services, such as the Revenue Authority, do not see a big a payoffdue to the limitednetwork o f domestic users. 24 Regulatory and Institutional Barriers 3.26. About three-quarters of the surveyed enter rises were dissatisfied with the government's regulatory and bureaucratic performance." Problems considered most important are employment/work permits and licensing, tax and customs administration, land and site development, and agency services. 3.27. Procedures for securing self-employment and work permits are cumbersome. In the case o f self-employment permits, the main problems relate to lengthy procedures, contradictions between the Investment Act and the Immigration and Deportation Act, and a highdegree o f discretion in the law, which kels mistrust and lack o f confidence in government by the private sector. Inthe case o f work permits, the key problems have to do with vague and obsolete legislation, which permits considerable discretion in individual cases and gives extensive powers to professional associations in the vetting o f applications. 3.28. The processof businesslicensingand entry is lengthy and cumbersome. The number of licenses required to operate a business has significantly increased in recent years. One Zambian company with eight subsidiaries requiredmore than 100 licenses to continue operating. The number o f licenses, the time it takes to obtain them, the discretion given to the authorities in the decision making process, and the requirements for annual renewal and fees represent major impediments to starting a business. The lack o f a service mentality inmany o f the licensing agencies and the inability o f the Zambia Investment Center to play the "one-stop shop" role givento it inthe InvestmentAct are the key underlying problems. Surveyed firms also reported an increase in new agencies within the government that require licenses. 3.29. Acquiringlandand developinga site involvesa complicatedbureaucracy. There are no defined procedures for investors to seek permission from chiefs, and chiefs generally do not have the skills required to assess applications from investors or the financial resources to purchase such skills. There are also serious human, financial and institutional deficiencies in land administration. The Surveyor General's Office in the Ministry o f Lands has only 15 surveyors, and there are only 24 registered private surveyors in Zambia. The Deeds and Land Registry appears to be severely under- resourced, its lacks space for effective organization o f records, and files are not easily accessible or well organized, so that document searches are difficult and the integrity o f the records cannot be assumed. The Registry indicatedthat its target i s to register a deed inone month, but it acknowledges that the process actually takes much longer. Coordination between local authorities, who have responsibility for zoning o f land, and chiefs i s poor, as i s coordination between local authorities and the Ministryo f Lands. 3.30. Zambian firms report taxes as one of the most binding constraints to private sector growth. The key problems relate to too frequent changes in tax policy, the lack o f a clear rationale for the changes an inadequate consultation by the Zambia Revenue Authority, and problems with VAT, including the long time required for registration, lack o f clarity on the deductibility o f various business costs, and delays inreceiving VAT refunds. 3.31. Delays in customs clearance of goods and the lack of transparency in classifying and valuing goods hurts exporters. The delays reflect the lack o f a risk-based system for inspections, poor document handlingpractices, and poor staffing arrangements and opening hours at border posts. The lack o f consistency and transparency in classifying and valuing goods arises because the l5 See World Bank 2004b for details. 25 authorities do not use importers' invoices as the basis for valuation (on the grounds that the incidence o f fraud and noncompliance by importers i s high). In addition, customs administration i s based on overly centralized procedures for issuingexport certificates, requiring exporters from outside Lusaka to visit the capital on a regular basis. Finally, the duty drawback system i s inefficient. The Zambia Revenue Authority claims that claims are paid within 30 days o f being lodged, but businesspeople report substantial delays, arising from questions about the accuracy o f quantities, the costs o f exported goods, the adequacy o f export documents, and the coefficients used for the product. 3.32. The failure to improve the administrative andregulatoryenvironmentfor businesshas many causes, only some of which relate to bureaucratic indifferenceto business concerns and corruption. Inmany cases, delays reflect organizational weaknesses and shortages of financial and humanresources to implement procedures and policies that are inthemselves reasonable. C. GovernanceandAccountability 3.33. Governance in Zambia Figure3-1. Indicatorsof GovernanceinZambiaand Sub- deteriorated, SaharanAfrica, 2002 particularly in the second half of the .,.....,........,. .,.................,... ZFlMBIFI (2002) ~ ~ 1990s. The Heritage Voice and Accountability Foundation ranked Zambia 119th out o f P o l i t i c a l Stability 156 countries in its ranking o f economic Governnent Effectiveness freedom in 2002. Transparency Regulatory Quality International rated Zambia 77th out o f 102 Rule of Lau countries for corruption (Transparency Control o f Corruption International, 2002). 3.34. All but one dimension of governance (regulatory quality) declined in Zambia between 1996 and 2002, according to the World Bank's Governance Research Indicator Country Snapshot (GRICS). Zambia performed worse than the average Sub-Saharan African countries in control o f corruption and government effectiveness (Figure 3-2). Corruption 3.35. The highlevel of corruptionimposesa burden on existing businesses and discourages investment. Despite the government's recent efforts to fight corruption, staff at government agencies reportedly expect some form o f bribe before performing their jobs. About two-thirds of those who answered questions on corruption as part o f the Investment Climate Assessment indicated that they consider corruption to be a serious problem. 26 3.36. Zambians believe that corruption was worse under the Chiluba government than under Kaunda (only 28 percent viewed corruption as having declined) (Afrobarometer 2002). The privatization program has been a particular source o f concern among Zambians, many o f whom feel that it was conducted in a way that permittedsome political officials to amass personal fortunes. The sale o f the Luanshya copper mine i s cited as having lacked adequate transparency in the selection process. Other surveys have shown the occurrence of graft in public procurement (World Bank, 2000). 3.37. Zambia appears to be turning the corner on corruption. It has made the fight against corruption a top priority and rejuvenated institutions such as the Anti-Corruption Commission and the Prosecutions Division. , It has brought to courts cases of suspected corruption as proof of its determination. It will take time, however, before actual results are seen on the ground in terms o f accountability, improved service delivery, and growth. Problems with the Judiciary System 3.38. The judiciary system suffered setbacks in the late 1990s. While the Zambianjudiciary has been able to preserve its independence, its functional capacity i s severely limited by infrastructure constraints and the shortage o f trainedjurists. Inthe World Business Environment Survey o f private businesses, conducted by the World Bank, most respondents expressed moderately high trust in the integrity o f the judicial system but overwhelming dissatisfaction with the timeliness o f the judicial process (World Bank 2000). L o w salaries are the most often cited reason for the government's inability to fill vacancies. Most courts lack adequate facilities for courtrooms. Judicial undercapacity has also resulted ina backlog of cases inthe criminal court system. The delays injustice create a lack o f faith in the system. On a positive note, the promotion o f mediation and arbitration in recent years has had a positive effect on reducingthe case load andthe time taken to resolve court cases. Lack of Government Effectiveness 3.39. Government effectiveness in the 1990s was compromisedby the diminished quality of economic governance, particularly poor public expenditure management and financial accountability. As several World Bank reports demonstrate,I6 the Zambian govemment continues to face problems ensuring sound public financial management and accountability, which are prerequisites for improving service delivery to the poor and to the private sector-and ultimately for improving growth. The key challenges are to strengthen the laws and institutions for accountability, to get the budgetary allocations right and make sure that those allocations go where they are supposed to, and to track and report on utilization o f public funds inan open and transparent manner. 3.40. The laws and institutions for public expenditure and accountability are weak. In a system with democratic control of public expenditure, the parliament typically approves the budget presented by the executive, as well as any major changes needed duringthe budget year. Under the current provisions o f Zambia's Constitution, the Ministryo f Finance can make sweeping changes in budgets and allocations without prior consent o f Parliament. Such flexibility can easily be misused without proper parliamentary oversight. l6 The World Bank has preparedthree major reports on public expenditure management inZambia. The 1998 Fiscal Management Report focused on fiscal sustainability issues. The 2001 Public Expenditure Review focused on budget allocation and execution problems. The Public Expenditure Management and Financial Accountability Report (PEMFAR) provides a comprehensive and integrated view of Zambia's budget management, financial systems and auditing, andpublicprocurementandlays out athree-yearreformprogram. 27 3.41, Even where legal provisions are appropriate, enforcement and accountability has been lacking. The Ministry o f Finance has not held budget controlling officers accountable for their management of public resources. Audit systems to detect improprieties have been ineffective because the Auditor General's recommendations are not enforced, and this lack o f sanctions has emboldened those engaging in improper activities. The lack o f transparency in budget management-caused in part by weak financial reporting-prevents the public from understanding the economic tradeoffs required o f Zambia today. 3.42. Ineffective and unrealistic budgetinghas undermined service delivery and growth. For years budgets have committed more than i s available. Government budgets largely reflected incremental changes from the prior year, without tackling the policy decisions needed to bring costs inline with resources. Evenwhen the budget appears to be inbalance on an aggregate basis, budgets for some ministry activities are insufficient to cover costs. Ultimately, these activities receive a supplemental budget (which i s not approved by parliament ex ante), squeezing out funding for other activities for which funds had beenbudgeted. As a result, actual spending at the end o f the year bears little relationshipto the original budget. 3.43. Budget execution i s weak. The key problems relate to the cash rationing system, lack o f enforcement o f expenditure controls, lack o f proper recording o f commitments, and weak procurement practices and regulations. Cash budget and unpredictability of budget releases. One o f the major consequences o f the cash budget i s that actual expenditures by line ministries bear little relationship to the budget approved by Parliament. Because o f the unpredictable nature o f cash releases from month to month, line ministries have had few means for planning basic operational and management needs. This has resultedin inefficiencies in service delivery in general and procurement of goods and services in particular. The effect on procurement practices has been to encourage an accumulation o f supplier credits, which raise the prices charged to the government and compromise service delivery. Weak enforcement of financial regulations. Many spending departments in Zambia operate as if standing instructions on the cash control mechanism do not exist. Orders are placed and payment vouchers are prepared and kept (often unrecorded) until cash i s available. In some cases, checks are issuedbefore goods are delivered andpresentedto the procurement staff (or stores) for exchange against goods. Control mechanisms designed largely for a manual system are ineffective because o f the volume of transactions and other changes. Lack of commitment control. A chronic build-upo f arrears has undermined the integrity and value of the annual budget. Although financial regulations prohibit line ministries from committing more than they have resources to pay, they continue to do so. Quarterly commitment reports prepared by the Accountant General are supposed to be used to identify overcommitting ministries, but sanctions are rarely forthcoming. The worst problems with arrears are for capital projects. Weakpublic procurement system. Several factors limit effective public procurement. First, a gap between intentions and practice pervades public procurement. Weaknesses in the structure and content o f the legal and institutional framework allow undesirable practices and procedures to go unchecked. The Zambia National Tender Board i s expected to enforce procurement rules, but it i s liberal in permitting exceptions and allowing negotiations to replace clear procurement guidelines. Second, the legal framework lacks robustness and has both structural and content inadequacies. Issues that should be in the Zambia National 28 Tender Board Act are in the regulations or guidelines and vice versa. Third, procurement management i s weak. In the few departments where management exists, procurement files are often incomplete and procurement planning i s largely nonexistent as a tool for conducting efficient and economic procurement. Fourth, no progress has been made in setting up a procurement cadre, despite the fact that more than 600 people have been trained at different levels. 3.45. Effective tracking of spending and transparency have been undermined by the current systems for financial management, which are manual or rely on outdated technology. Budget preparation, budget execution, and payroll management are all separate in Zambia, and different applications in the Finance Ministry do not interact with each other. Inline ministries the situation i s generally worse: the accounting function relies heavily on manual recordkeeping, and ministries submit their monthly expenditure reports on diskettes with the help o f a computer-based financial management system, which records the details o f transactions after payments are made. Several problems have resulted from such system. Expenditure reports are prone to errors, and inconsistencies are difficult to find and reconcile. Expenditure reporting tends to be delayed because of the compilation required. The delays and inconsistencies greatly diminish the ability o f the Ministry of Finance to provide adequate oversight and control over expenditures. Because of these inadequacies, the executive has found it difficult to provide an accurate, complete, and transparent account o f its financial position to Parliament and other stakeholders, including the general public and donors. 3.46. T o improve government effectiveness and service delivery to the poor and the private sector, Zambia must significantly improve its public expenditure management and financial accountability system. Doing so will require a clear break with the past in terms o f government budgeting and accountability. Parliament should be involved in preparing the budget as well as in providing oversight over spending. The process for granting supplementary estimates must be transparent, the government should promote transparency and improve its reporting on public resources, and there must be the political will to enforce Zambia's rules and regulations for public spending and sanction those who violate the ruled7Without such policy measures, it will be difficult for the government to be seen as supporting private investment and growth, much less poverty reduction. D. Trade Policy and Export Diversification 3.47. Since 1992 the government has implemented a comprehensive trade reform program. Import duties and other charges were significantly reduced; import and export licenses and export bans and taxes were eliminated; exchange rates were unified and exchange controls removed, allowing the exchange rate to be determined by the market; a system o f export incentives (duty drawback, manufactures-under-bond) was introduced; and an Export Promotion Board was set up. 3.48. The simple average most favored nation tariff rate is 13.4 percent, with a coefficient of variation of 0.7, indicating modest dispersion of tariff rates. The 2001/2002 tariff has 6,041 lines at the Harmonized System eight-digit level. Only 80 lines bear alternate tariffs. The ad valorem tariff includes four bands: 0, 5, 15, and 25 percent (Table 3-4). The most common rate of 15 percent applies to about a third of all tariff lines. The maximum rate (25 percent) applies to final goods and agriculture-related tariff lines. l7 See World Bank (2003b), whichprovides a three-year policy reformagenda for improvingpublic expenditure managementand financial accountabilityinZambia. 29 Table3-4. StructureofMost FavoredNationTariffsinZambia, 2001/2002 Tarifrate Tariff lines Type ofproducts (percent) 0 20.9 Agricultural machinery, pharmaceuticals, some raw materials 5 15.1 Mining and quarrying, machinery, some raw materials 15 33.0 Intermediategoods 25 3 1.0Final goods including agricultural products Source:WTO 2002. 3.49. In addition to the tariff, Zambia also collects a 17.5 percent VAT on goods and services as well as excise duties on selected products, at rates ranging from 5 percent to 125 percent, depending on whether they are imported or locally produced. Duty and tax concessions and exemptions, which are granted for certain goods and beneficiaries, are widespread. 3.50. Zambia now has one of the most liberal trade regimes in Africa. All quantitative restrictions and export taxes have been eliminated. Import controls are maintained only for environmental, health, and security reasons. Export prohibition exists for certain types o f logs under international agreements and occasionally for grains (during drought years). There are no general licensing requirements for exports, although certain goods, such as fertilizers, live animals, gemstones, and firearms, require special export permits. 3.51. An Export Processing Zone Act was enacted in November 2001 and an Export Processing Zone Authority established in September 2002 (although its operation has recently been suspended due to fear o f revenue losses). Several incentives are also in place-although not working well-to assist exporters: 0 The duty drawback on inputs used for export production i s based on an input-output coefficient system at the individual firm or sector level. Capital goods and mining inputs are not eligible. 0 Goods stored in bonded warehouses or manufactured under bond are exempt from custom duties andtaxes when they are exported. 0 Exporters o f nontraditional products are subject to a concessional income tax o f 15 percent rather than the standard rate o f 35 percent. 0 Exporters are zero rated for VAT refund purposes, but imported machinery and spare parts are subject to VAT. 3.52. Zambia has signed various trading arrangements (see Annex C). It is a member of the World Trade Organization, the Common Market for Eastern and Southern Africa (COMESA), and the Southern African Development Community (SADC). It is a signatory to the Cotonou Agreement and eligible for the United States' African Growth and Opportunity Act (AGOA). It i s negotiating bilateral agreements with Angola, Botswana, the Democratic Republic o f the Congo, Malawi, Mozambique, Namibia, Tanzania, and Zimbabwe. It benefits from nonreciprocal preferential treatment from many industrial countries under the Generalized System o f Preferences including the EU's Everything But Arms (EBA) initiative and AGOA. Under the free trade area agreements, Zambia provides reciprocal duty-free access to COMESA members for all goods and to SADC members for Category A products. 30 3.53. Zambia's trade performance has been mixed. While the non-traditional exports (NTE) reacted positively to policy reforms, mining sector exports declined substantially leading to a fall in export revenue. The share o f exports in GDP fell from 34 percent in 1990-92 to 28 percent in 1999- 2001, Over the same period, Zambia's share inworld exports fell from 0.031 percent to 0.015 (Table 3-5). 3.54. Significant changes have taken place in the direction of trade. The share o f exports going to the EU, one o f the principal trading partners, fell from 27.7 percent in 1990-92 to 24.3 percent in 1999-2001 (Table 3-5), while the share o f exports going to SADC (mainly South Africa) increased from 3.5 percent to 18 percent. Changes have been even more marked in the origin o f imports. The share o f imports from the EU declined from 38.7 percent to 12.7 percent, while the share o f SADC (mainly South Africa) increased from 32.6 percent to 69.7 percent, in the same period. The causes and consequences o f this shift from South-North trade to South-South trade i s an important issue that requires study. Table 3-5. Indicators of Trade Performance, 1990-2001 Item 1990- 1999-2001 Item 1990-92 1999-2001 92 Volume of trade (US$ Direction of exports million) PA) Imports 843.3 929.0 COMESA Exports 1330.2 868.7 SADC 4.3 30.9 Traditional 1296.2 626.3 EU 27.7 21.6 Nontraditional 134.0 242.4 UnitedStates 49.1 24.5 Rest ofthe World 18.9 23.0 Composition of exports PA) Origin of imports Traditional exports 89.9 72.1 (percent) Nontraditional exports 10.1 27.9 COMESA 8.1 7.3 SADC 32.6 69.7 EU 38.7 12.7 UnitedStates 6.5 2.2 HerfindahlConcentration 0.70 0.56 Percentof imports index financed by exports 118 95 Percentage of GDP Share in world trade Imports 40.7 26.9 Imports 0.03 1 0.015 Exports 34.0 28.0 Exports 0.038 0.015 Source: Staff calculations. 3.55. Several factors account for Zambia's poor trade performance and its limited export diversification in the 1990s. The liberalization o f trade affects the volume o f imports immediately, without support from other domestic policies. But it takes time to shift resources to export-oriented sectors, and complementary policies (macroeconomic stability, competitive exchange rate, affordable credit, infrastructure, appropriate sectoral policies) are needed to improve the supply response. Inthe past, trade was liberalized quickly before supply-side issues were adequately addressed. As a result, exports could not respond positively to trade liberalization, and many companies, particularly in textile and manufacturing, had to close under import competition. Partly as a result, the share o f manufacturing inGDP fell inthe early 1990s. 3.56. Market access does not seem to constrain Zambian exports. Zambia's largest export markets are the EU, South Africa, and the United States. Under EBA, Zambia can export most products to the EU duty and quota free, and the sugar quota was recently doubled. While South 31 Africa has quotas for some imports, there i s large untapped potential for Zambia inthe South African market. AGOA provides access to the U.S. market that Zambia has not yet taken advantage of. 3.57. The key supply-side constraints are the lack of good behind-the-border policies and trade policy. Behind-the-border policies include policies that help stabilize the volatile and overvalued exchange rate, reduce high real interest rates, improve the investment climate, and strengthen infrastructure for private sector investment. 3.58. Zambia's trade policy is stillnot export friendly, despite substantialtrade liberalization in the 1990s. Exporters face several constraints, includingthe following: Anti-export bias. Zambia's trade policies provide lower incentives for export markets than for domestic markets. The average trade-weighted duty on imports i s about 12 percent, while reimbursement o f import duties for exporters under the duty-drawback scheme-the main export incentive scheme-is about 4 percent. Inaddition, there are long delays (and arrears) inreimbursingimport duties andVAT payments. Duties on capital goods imports. Under the Investment Act of 1991, capital goods in the export sector were imported duty free. This policy contributed to the surge in nontraditional exports in the mid-1990s. The policy was reversed in 1996 and a 5-15 percent import duty imposed on capital goods. The decision was made on the basis o f revenue considerations rather than the need to diversify exports. Lack of affordable export credit. Real interest rates in Zambia are high (20-30 percent). Apart from two foreign-financed credit lines for pre-shipment financing, there i s no dedicated credit available for exporters. The two credit lines cover less than 10 percent o f total nontraditionalexports. Lack of other export incentives such as EPZs. Other export incentive schemes, such as export processing zones, have been successfully used to increase exports in several countries in the past 10 years, especially in countries where there are transparent governance structure for such schemes. Zambia passed legislation to set up export processing zones only recently. It will take time to develop an efficient export processing zone sector, given the institutional and governance weaknesses inZambia. Failure to take advantage of trade preferences. Substantial nonreciprocal preferences were granted to Zambia under Everything But A r m s (EBA) and AGOA. Awareness o f these benefits i s inadequate, however, and preparatory work to meet the requirements to access these huge markets has been slow. As a result, benefits from these preferential arrangements have been very limited. Zambia has also failed to fully benefit from regional preferential agreements, largely because o f supply-side constraints. Weak policy coordination. Responsibility for policies affecting exports i s shared by the Ministry o f Finance (duty-drawback scheme), the Bank o f Zambia (credit policy), the Ministry o f Commerce (trade arrangements), and the Export Board o f Zambia (trade promotion), with inadequate coordination. Inadequate local capacity for policy analysis. Policymakers do not benefit from good research and policy analysis, because there i s no local capacity in or outside government to conduct such work. 32 0 Weakpublic sector-private sector partnership. Consultation with the private sector on trade and other policies i s very weak. Policies have been established largely by the government anddonors rather inresponse to the needs and concems o fthe private sector. 3.59. To increase export diversification,Zambia's trade regime and complementarybehind- the-border policies need to be improved. Improving export incentives rather than further liberalizing imports will be key to export diversification and growth. E. HIV/AIDS and Other Shocks 3.60. Zambia is among the countries hardest hit by the HIV/AIDS virus in Sub-Saharan Africa. With an HIV prevalence rate of 16 percent, according to the 2002 Zambian Demographic and Health Survey, the disease has become the greatest threat to development, reversing past achievements in life expectancy, child malnutrition, and economic growth. The government needs to take a strong leadership role in combating HIV/AIDS, rather than hiding behind cultural mores and social squeamishness. 3.61. The spread of HIV/AIDS is reducingthe stock of human capital and slowing economic growth(see Annex B). Between 1992 and2002 the average annual growth rate o fhuman capital was 0.5 percentage points lower than its potential value because o f the spread o f HIV/AIDS. The effect on growth has been significant, reducing growth by an average rate o f 0.3 percentage points a year since 1992. The strong effect i s not surprisinggiven that the disease disproportionately kills members of the middle class (Cheru 2000). 3.62. The highprevalence of HIV/AIDS has made it difficultfor privatesector employers to find qualifiedpersonnel. Because trained people die, businesses are forced to train multiple people for key positions, increasing their overall costs. 3.63. The high incidence of HIV/AIDS has enormous implications for farm productivity, health, education, and extension services in Zambia (Pillai, Sunil, and Gupta 2003; Alwang and Siegal2003). Alwang and Siege1(2003) shows that the HIV/AIDS pandemic leads to much less land being cultivated and more production o f staple foods relative to other products, reducing rural households' ability to produce for market and escape poverty. Qualitative research in rural areas shows that HIV/AIDS negatively affects farming and welfare in a number of ways (Milimo et al, 2003): 0 HIV/AIDS has depleted the ranks o f trained agricultural extension workers. Lack o f good extension services i s the most frequently cited constraint to agricultural development. The poor quality o f services i s brought about partly by the untimely deaths o f trained manpower. Staff in the health and education sectors have also been affected, constraining the delivery o f education and health services inrural areas. 0 Instead o f farming, people, particularly women, spend much o f their productive hours nursingthe sick. 0 A considerable amount o fmoney is spent onthe sick. Absent HIV/AIDS, this money would have been put to productiveuse. 0 The addition o f HIV/AIDS orphans into households has reduced the welfare o f rural households by forcing families to crowdtogether insmall rooms. 33 3.64. In light of the growth and welfare impacts of HIV/AIDS, the government of Zambia may have little choice but to expand its current level of prevention and treatment (Box 3-1). In its Poverty Reduction Strategy Paper the government has outlined a series o f actions to be taken in response to this growing threat, including educational programs, community-based home care, and measures targeted at orphans and vulnerable children. The cost o f the drugs necessary to combat the epidemic i s high: estimates from other countries range between 100 and 400 percent o f GDP per capita (IMF, 2001). While there is some scope for private sector involvement, most domestic health spending will have to come from the government, from donors, or both. Box 3-1. The Government's Strategy for CombatingHIV/AIDS The Zambian government has developed four national plans to fight the HIV/AIDS epidemic. The f r s t two were implemented by the Ministry of Health; the third involved all ministries. The current ftamework, coordinatedby the National HIViAIDS Council, adopts a multisectoral approach. erventions outlined under the Poverty Reduction StrategyPaper are divided into two levels: The First-Level Priority Program focuses on reducing new HIV infections (by implementingbehavior-changing communication campaigns and improving the distribution of free condoms) and reducing the socioeconomic impact of AIDS (by expanding access to voluntary counseling and testing, strengthening community home- based care, and providing antiretroviral treatment). The Second-Level Priority Program focuses on children, youth, and women who are at risk from HIV transmission. The measures include improving management and treatment of sexually transmitted disease, expanding access to quality prevention of mother-to-child transmission programs, and expanding existing programs targeting high risk groups. The government i s currently implementing a World Banl-supported project (Zambia National Aids Response and Action) thal seeks to combat HIV/AIDS by adoptingsome ofthe measures inthe two priority programs. Source: Zambia, 2002a; IMF2001. 34 agricultural subsectors during the 1990s, especially since liberalization Figure 4-1. Agricultural Value-Added and Contribution reforms were introduced in 1992. It of Agricultureto GDP, 1990-2002 investigates the key constraints to I improved agricultural performance 1,100 (access to fertilizer, water, financial 35 I services, landand secure land rights, and markets); proposes strategies to overcome them; and offers some ~ conclusions andrecommendations. 4.2 Agriculture can serve as an engine for achieving broad- based growth, reducing poverty, 1990 1992 1994 1996 1998 2000 2002 diversifying production and exports, and improving national 1 -m- Agriculturevalueadded +Contribution to GDP 1 1 and household food security. I 1990 and 2001 (FA0 agricultural http://faostat .fao.org) . database, available at Proportionof Total MerchandiseExports, 1990-2000 Sales of agricultural products accounted for 19 percent of total earnings from I merchandise exports in 1999, up from 200 20% just 2 percent in 1990 (Figure 4-2). 15% They declined in 2000, due largely to c) a drop o f more than 50 percent in the 10% f value of coffee exports and a fall in 5% d exports of cotton lint. Exports of agricultural commodities, vegetables, 0% cut flowers, textiles, and garments I 1990 1992 1994 1996 1998 2000 enjoy quota-free, duty-free access to , --E.Agricultural exports I the European Union market at least I ~ until 2007, when a successor agreement will be negotiated. Source: FA0 agriculturaldatabase andthe World Bank 2003c. 35 4.3 Linkages between agriculture and other sectors are important in Zambia. Agro- processing industries account for about 84 percent o f manufacturing output, and they are more than five times more important than the next largest group, textiles and leather products (both of which rely on agricultural raw materials) (World Bank 2003a). Agro-industries alone contribute about 7.4 percent of Zambia's GDP. 4.4 Large commercial farms in Zambia generally use modern technology, machinery, irrigation, fertilizer, and pesticides (Table 4-1). The farms, located mainly along major transportation routes and near population centers, occupy state land under 99-year leases. They produce most o f the country's agricultural exports and about 80 percent o f milk, 75 percent o f wheat, and 70 percent o f soybeans and poultry. Some are vertically integrated with agro-processing operations (World Bank 2003a). 4.5 Emergent farmers are more commercially oriented than the typical smallholder and grow staple foods and cash crops such as cotton, sugarcane,andtobacco. They occupy both state land under long-term leases and land held under traditional tenure systems. They use animal traction, hybridseed, andfertilizer to grow rainfedcrops. 4.6 Smallholders,who hold nearlytwo-thirds of agriculturalland and a large share of the national herd, tend to grow low-value-to-weight food staples, including about 60 percent of the country's maize, 90 percent o f sorghum, 85 percent o f groundnuts, and virtually all the cassava and other starchy staples (groundnuts, roots and tubers), primarily for their own consumption on less than five hectares. Some smallholders also grow some higher-value cash crops, such as cotton, tobacco and paprika. They use mainly hand tools and animal draft power, generally lack access to irrigation and machinery, and use limited purchased inputs. Consequently, their productivity i s low, and their yields average about 50 percent o f those realizedby commercial farmers. They occupy mainly land heldunder traditional tenure systems, managed by local chiefs. -1 Table 4-1. Typology of AgriculturalProducersinZambia TYPeProducer Number of Averagefarm Size Technologv, Market Location Major producers Cultivation Orientation Constraints Practice Small-scale 800,000 Less than 5 Handhoe, Staplefoods, Entirecountry Remoteness, producers households hectares (with minimalinputs, primarily seasonal labor majority householdlabor home constraints, lack cultivating 2 consumption of input and hectares or less) outputmarkets Emergent 50,000 5-20 hectares Oxen, hybrid Staple foods Mostly along rail Seasonallabor farmers households seed, fertilizer, andcash lines (Central, constraints, lack few with crops, Lusaka, Southem of credit, weak irrigation, mostly primarily provinces), some market householdlabor market Eastem, Westem information orientation provinces Large-scale 700 farms 50-150 hectares Tractors, hybrid Maize and Mostly Central, Highcost of commercial seed, fertilizer, cashcrops Lusaka, Southem credit, farms some irrigation, provinces indebtedness modem mana.gement, hirediabor Large 10 farms 1,000+ hectares High IMaize, cash IMostly Central, Uncertain corporate mechanization, crops, vertical Lusaka, Southem policy operations irrigation, integration provinces environment modem management, hirediabor Source: Adapted from World Bank 2003a. 36 4.7 These characteristics of agriculture have important implications for the design of programs aimed at improving the competitiveness of agriculture, diversifying exports, and reducing poverty. While the liberalization reforms of the early 1990s stimulated growth in agriculture generally, some 60-70 percent of smallholders have not benefited from this growth, primarily because they live far from markets where inputs can be obtained at reasonable cost and where farm output canbe sold at aprofit. Targeted programs will beneededto reduce the isolationof these farmers and help them move from subsistence farming to more profitable market-oriented agriculture. 4.8 Output of agriculturalproductsvaried in the 1990s but showed a positivetrend during the second half of the decade. Some 80-90 percent of crop production in Zambia is rainfed, so some of the fluctuation in output i s due to erratic rainfall patterns during the decade (droughts occurred in 1991/92, 1994, 1995, 1997, 1999, 2000, 2001, and 2002). The dramatic drop in output that occurred after 1993, however, resulted from the government's withdrawal from the highly subsidized marketing of maize and f a m inputs. These subsidies, while contributing to expansion of maize production during the 1980s, also contributed to the fiscal crisis and to macroeconomic instability duringthe same period andwere unsustainable. 4.9 In responseto the new opportunitiesto export and to shifts in relative prices of inputs and outputs, farmers diversified their crops. Commercial and to a lesser extent medium-scale farmers expanded coffee, sugar cane, cotton, and horticulture production. Small-scale farmers started cultivating cotton and burley tobacco in addition to food crops. Many small-scale farmers also diversified their food crop production, shifting some land from maize to cassava, sweet potatoes, groundnuts, sorghum, and millet. Becausethese crops are more resistant to drought, and require few purchased inputs to cultivate, this shift has helped many smallholder households become more food secure and less vulnerable to weather shocks and unreliable marketing agents. Devaluation of the kwacha in 1992 and the removal of barriers to exports and imports encouraged new investment in export-oriented crops; since 1992 agricultural exports have more than tripled in value and become more diversified. Agricultural value-added i s now above the levels achieved in the first part of the decade, despite unprecedented declines in world commodity prices. Moreover, the results were achieved duringa period of extremely highlendingrates. 4.10 Despite liberalization reforms, government intervention in the sector has remained extensive, with uneven implementationof policies and frequent policy reversals. Interventions have been particularly extensive in the maize and fertilizer markets, ostensibly as a way to support smallholder farmers but also as a source of political patronage (Jape and others 2002). These interventions have led to a waste of public resources, low private sector confidence to invest, and a low repayment culture among farmers. PRINCIPALPRODUCTS A. Foodcrops 4.11 Maize is the basic staple in Zambians' diets, accounting for up to 70 percent of total calories consumed(WorldBank 1994). The dominance o f maize has more to do with government strategies than with the suitability of maize as a crop, however. 4.12 The withdrawal of subsidies, which has been politically difficult, has led to a more diversified crop production pattern. Land plantedto maize declined by 22 percent between 1990 and 2002, and maize currently occupies about 50 percent of Zambia's cropped area, down from about 70 percent inthe mid-1980s (Government o f Zambia 2000; Howardand Mungoma 1996). 37 B. ExportCrops 4.13 Fresh vegetables and Figure4-3. Exports ofFreshFlowers andVegetables, flowers are importantand fast- 1994-2002 growing agricultural subsectors. The horticultural 25 - industry has thrived since % 20 -- liberalization. Exports o f % 15 ._ specialty and organic vegetables, ~L summer flowers, and roses which -- E 8 10 totaled US$6 million in 1994, 5 -- rose to about US$42 million in 0 2002 (Figure 4-3)'' Exports o f 1994 1996 1998 2000 2002 fresh flowers and vegetable accounted for about 35 percent o f 1-W- Freshflowers +Fresh vegetablesand h i t total agricultural exports in 2002. Horticulture operations directly Source: ExportBoardofZambia,various years. 4.14 Financialsupport from donors has been instrumentalin promoting horticulture. The European Investment Bank and the World Bank Group have provided low-interest long-term loans for more advanced irrigation systems. The EU's Export Development Program has offered short- term finance to exporters for the purchase o f inputs (including airfreight for exporters) through their producers' associations. It also offers technical assistance to help growers improve the productivity o f their operations and the quality o f their output, identify markets, and promote exports. Zambia's horticultural industrystill commands a small share o f world markets for fresh flowers and vegetables and has plenty o f room to expand. Operations are very capital intensive, however, and the high cost o f capital in Zambia i s a major constraint to expansion. Expansion o f this industry for the medium term will likely depend on the availability o f external finance. C. Cotton 4.15 Cotton is the main cash crop for smallholders, who, produce more than 98 percent of Zambia's cotton. Smallholders are willing to produce cotton not because it is necessarily more profitable than other crops but to diversify their production, gain access to inputs on credit, and benefit from the guaranteed market for cotton, reducing the risks o f farming (Tshhirley, Shaffer, and Zulu2004). 4.16 Lack of government interference combined with companies' willingness and ability to innovate have been key to the success of Zambia's cotton industry. In countries with weak contract enforcement and nonexistent rural financial services, the absence of government interference provides both the opportunity and the incentive to innovate. Faced with the need to ensure that farmers receive timely delivery o f inputs and appropriate extension advice, even while credit recovery rates were falling, some companies (e.g. Dunavant) developed a new approach to deliver inputs, ''Thefigures provided by the Export Board of Zambia for the value of exports have been reduced by 40 percent for vegetables andby 50 percentfor flowers to reflect marketing andtransportation costs. 38 called the distributor system, that has resulted in a high level o f credit recovery. Under this system, the company lends inputs to independent agents (distributors), who on-lend them with extension advice to neighboring farmers whom they know and trust (Poulton and others 2002). Since the start o f the scheme, credit recovery rates have risen from about 65 percent to 85 percent, and cotton yields have risen 33 percent. While the effectiveness o f the innovations over the long term are not yet known, the measures have reversedthe decline that threatenedthe cotton industry. D. Coffee and Sugar 4.17 Zambian commercial farmers are relatively low-cost producers of high-quality arabica coffees. Even with the collapse in coffee prices at the end o f the 1990s, Zambian commercial coffee farming has remained profitable. Smallholder coffee production is, however, marginally if at all profitable at today's low world coffee prices. 4.18 Zambia is one of the world's lowest-cost producers of sugar, and production has expanded rapidly since the state divested its assets in the sector. Zambia Sugar dominates the industryinthe country, but smallholders produce most o fthe cane. E. Livestock andAnimalProducts 4.19 Livestock represents about 35 percent of national agricultural output, and earnings from animal products have been rising. Export earnings from animal products totaled US$3.4 million in 1997, US$4.1 million in 1998, and US$4.4 million in 1999 (Export Board o f Zambia 2000). 4.20 Smallholders can profitablyproduce milk and dairy products, and cattle providemany of them with draft power, contributing to incomes and food security. Households that have access to oxen can cultivate significantly more land and produce a larger variety o f crops than those without such access (Alwang and Siegel, 2003). Rebuilding Zambia's oxen herds and increasing access o f smallholders to draft power would improve food security and reduce poverty among Zambia's rural poor. KEY CONSTRAINTS TO GROWTH AGRICULTURE IN A. Lack of Access to Markets 4.21 Perhaps the most important constraint faced by Zambia's farmers is limited access to markets. The experience o f the 1990s shows that farmers are willing to invest in agriculture and produce goods usingpurchased inputs if they are reasonably certain that they will be able to sell what they produce at a profit. Currently, markets are ensured mainly for the large export-oriented farms and for farmers involved in out-grower arrangements that link them to buyers. By contrast, farmers in remote areas of Zambia have few if any market outlets, and they have been hit hard by the withdrawal o f subsidies for transportation, inputs, and marketing o f maize. The high transportation and transactions costs they face for both inputs and outputs make it difficult for them to produce goods profitably for the market. Often they are hampered by lack o f information about market opportunities and by weak business and negotiating skills. Linking farmers in remote areas to markets i s a challenge that will require a long-term effort. 4.22 More needsto be done to create a market system that benefitsall Zambians. Critical are improving the overall macroeconomic framework and reforming the financial sector so that credit and other financial services become more widely available to farmers, input suppliers, and buyers o f farm 39 output. Additional efforts to facilitate trade are needed, including measures that enable farmers and processors to comply with sanitary and phytosanitary standards o f importing countries. More also needs to be done to improve smallholders' access to basic services, such as education and health services, as well as services that can help them learn about and take advantage o f market opportunities. Much more needs to be done to upgrade rural infrastructure, including roads, storage facilities, markets, and energy and communications systems. B. Lackof Water When andWhere It I s Needed 4.23 Irrigationuse is low. Much of the volatility inagricultural output inZambia is linkedto the low use o f irrigation systems. O f Zambia's 6.3 million cultivated hectares, only about 100,000 hectares (less than 2 percent) are irrigated. About 52 percent o f the irrigated area i s cultivated by large- or medium-scale farmers, with the remainder tended by small-scale farmers (FA0 and the Government o f Zambia 2002). Investment in irrigation i s low despite the abundance o f easily accessible water resources, because population density i s low and uncultivated arable land abundant. Because establishing and operating irrigation systems i s more expensive than expanding agriculture into new lands, historically most countries have adopted irrigation only after agriculture had expanded to marginal lands (Jones 1995). 4.24 The use of irrigation appears to be growing-where it is linked to complementary efforts linkingfarmers to markets. While reliable data are scarce, irrigated production of sugar, coffee, horticulture, and other cash crops appears to have expanded substantially since liberalization. Because a wide variety o f irrigation technologies i s available in Zambia- ranging from large center pivots and overhead sprinklers capable o f watering hundreds o f hectares to low-cost treadle pumps used to irrigate tiny plots-expansion o f irrigation has occurred among all categories o f farmers. Given the expense o f purchasing and operating irrigation systems, both large-scale and small-scale farmers generally use the systems to grow high-value crops. Projects designed to encourage small- scale irrigation have not succeeded without complementary efforts to link farmers to markets. Farmers in Zambia who have produced high-value perishable products (such as tomatoes) but have not had good access to markets have often not been able to sell them. 4.25 Government canmost effectively promote agriculturalgrowthandpoverty reductionby creating an enabling environment for privatesector investment in irrigation. Much experience with irrigation projects worldwide suggests that government does have a role to play in promoting irrigation. But this role has more to do with creating an enabling framework that provides incentives for the private sector to invest in irrigation (while respecting the rights o f upstream and downstream water users and the environment) than with providing systems themselves. Priority measures include the following: 0 adopting a water policy and water laws that allocate transferable water rights to farmers 0 conductingresearch to test small-scale irrigation systems anddisseminatingpromising results 0 providing support to link farmers to markets as part o f the package promoting irrigation use among smallholders 0 linking development o f community-wide irrigationto community-driven development 40 C. LimitedRuralFinancial Services 4.26 The supply of credit to farmers has shrunk dramatically since financial markets were liberalized in 1992, due to the elimination of subsidized credit for smallholders and poor macroeconomic management that keeps real interest rates high and limits the supply of credit to the private sector. Before liberalization smallholders received short-term production credit at subsidized interest rates through three main government-run financial institutions: the Lima Bank, the Credit Union and Savings Association, and the Zambia Cooperative Federation's Finance Services (Wichem, Hausner, and Chiwele 1999). The government spent about 20 percent of its total expenditures for agriculture funding operation of these financial institutions. Although the institutions were supposedto repay the funds providedby the government, they were unable to do so because a highproportion of borrowers defaulted on their loans, either because harvests were poor or because borrowers treated theirs loans as subsidies or grants (Pletcher 2000). As part of the liberalizationreforms, the government stoppedproviding funds for agricultural credit. This ledto the collapse in 1997 of the Lima Bank and of the Zambia Cooperative Federation's Finance Services. The Credit Unionand Savings Association also stoppedproviding short-term rural credit in 1997. 4.27 Since liberalization, real interest rates have remained high and credit available for commercial farming has been scarce. Most of the expansion in commercial farm operations i s being financed through capital provided by intemational development banks and home offices of multinational firms. 4.28 New institutions are emerging to serve resource-poor farmers and rural entrepreneurs. Few formal profit-making financial institutions are interested in serving resource-poor farmers and rural entrepreneurs, because the transactions costs of providing small loans and savings accounts to dispersed customers are high and the risks of dealing with socially distant borrowers who lack collateral significant. However, a number o f institutions are emerging to serve the needs and demandsofrural farmers and entrepreneurs. 0 Out-grower schemes. The most important of these institutions are out-grower schemes, through which a firm supplies agricultural inputs on credit with the expectation that the loans will berepaidat the time the crop is sold. While estimatesvary, it appears that 2 5 4 0 percent o f small-scale farmers may be using such arrangements (Wichem, Hausner, and Chiwele 1999;Tshirley, Shaffer, and Zulu2004)." 0 Microjnance institutions. Still small but growing in importance are microfinance institutions, which provide loans to members of groups, relying on joint liability and community screening and monitoring as substitutes for collateral. Most have received donor support to initiate operations and are runby local or intemational NGOs. They are expected to become self-sufficient within a few years of start-up. 0 Informal credit schemes. Informal credit systems that operate among friends, relatives, neighbors, small groups (such as rotating savings and credit associations), and moneylenders provide the majority of loans in many countries. Indirect evidence suggests that they are important in Zambia, as many smallholders purchase fertilizer from private firms with cash (Govereh and others 2002). l9 Such arrangements are not optimal, because they involve significant transactions costs and limit the options o f small farmers to respond to alternative market opportunities. 41 0 Zambia in 2001." It increases the availability o f credit to farmers by encouraging the use o f Warehouse receipt systems. A transferable warehouse receipt system began operations in stored agricultural commodities as a form o f secured collateral against which commercial banks make loans (Coulter and Poulton 2001). If successful, the system could significantly deepen the operations of the financial sector, improve the efficiency o f agricultural marketing, and reduce food price volatility, benefiting both producers and consumers. 4.29 Increasing the supply of credit will help stimulate agricultural production, but it will not on its own be sufficient to reduce rural poverty or increase rural household food security. Studies o f microfinance operations worldwide reveal that increasing access to credit encourages households to take risks they would otherwise be reluctant to assume (such as growing unfamiliar but potentially profitable crops), maintain consumption o f basic necessities when household income temporarily declines, recover more readily from crop failure, and take advantage o f potentially profitable opportunities. However, supplying credit will not in itself reduce poverty. Resource-poor farmers face complex, multiple constraints in pursuing income-earning opportunities (Zeller and Sharma 2000). Resource-poor rural households are more likely to benefit from access to credit ifthey also have access to high-quality extension advice, education and health services, agricultural market information, and improved rural roads." Attention must also be paid to providing the full range o f rural financial services, including savings and insurance services, rather than focusing narrowly on providing credit. The great seasonal uncertainties associated with rainfed agriculture mean that the poor need access to savings and insurance services that will help them manage risks, not just credit that will help them expand production (Zeller and Meyer 2002). Providing multipleservices will also help increase the viability o f providers by allowing them to diversify their portfolios (Zeller and Sharma 2000). 4.30 Government has an important role to play in fostering the growth of rural financial services. Government must create a regulatory and legal framework that encourages the development o f innovative and sustainable financial services generally and rural financial services in particular (Fries 2001). Doing so involves establishing macroeconomic stability, ensuring that service providers are free to respond to opportunities while maintaining prudential banking practices, and creating a legal framework that permits parties to enforce contracts without undue cost. In September 2000 the Zambian Parliament passed substantive revisions to the Banking and Financial Services Act. The changes inthe act place microfinance institutions squarely within the financial markets in legal terms and provide for their regulation and supervision by the Bank o f Zambia or its designated agents. The Bank o f Zambia is now in the process o f finalizing a regulatory framework for microfinance in Zambia. 4.31 An improved regulatory environment i s not enough. Governments and donors also have a role to play in directly fostering the growth of rural financial service providers by financing pilot programs in rural areas to test various models o f service delivery and funding start-up capital; training boards, managers, staff o f service providers, and groups o f borrowers; and 2o The Zambian Agricultural Commodity Agency was established to certify and regulate warehouse operators issuing transferable warehouse receipts. It i s owned by groups with a stake in the agricultural industry, including farmers, traders, millers, bankers, insurers, warehouse operators, inspection companies, andNGOs. 21 Interestingly, ina study o f the impact o f microcredit inMalawi, farmers who received credit ended up worse off than those that did not, because they used the funds to purchase fertilizer to grow maize, which could not be cultivated profitably, rather than switching to alternative crops that couldbe (Diagne 1998). 42 establishing management information systems (Zeller and Sharma 1998; Meagher and Mwiinga 1999). C. Lack of Access to Fertilizer 4.32 The wide distribution of heavily subsidized fertilizers has been one of the key government policies to improve smallholder incomes and food security. Since the drastic reduction o f the subsidy in the 199Os, however, fertilizer prices have increased and availability declined. Fertilizer i s not available on credit, and it i s and often unavailable on time, even to those who pay cash. This affects cash crops as well as hybrid maize. Expensive or unavailable fertilizer thus poses a threat not just to agricultural productivity and economic growth but also to basic food security. 4.33 Fertilizer use among smallholders in Zambia fell sharply following the liberalizationof the marketin 1992. Fertilizer use I: smallholders declined nearly30 percentbetween 1991/92 and 1999/00, after fertilizer subsidies o f Figure4-4. Supply ofFertilizer,1990191-199912000 50 percent were removed, pan- territorial pricing for fertilizer and 1 255 maize was ended, and transport ~~ ~ I subsidies were eliminated(Govereh 3 235 and others 2002) (Figure 4-4). 215 V Much o f the decline occurred in the .-Bh 195 remote northern regions o f the L 175 country, where use o f fertilizer was .;; 155 no longer profitable. A rural 135 household model for Zambia ( 8 115 Alwang and Siegel, 2003) suggests cI 95 that even a slight price increase will ' cause fertilizer use by low- 75 1990191 1992193 1994195 1996197 1998199 ~ technology (hand hoe) households to drop to zero. However, the Source: Goverehand others 2002. income elasticity o f fertilizer i s not high:a 20percent decline infertilizer prices is associated with only a 6 percent increase inincome. 4.34 Despitethe government'scontinuedinvolvementinthe fertilizer markets, privatesector participationhas grown rapidly. Some 10-12 importers, 500 wholesalers, and 5,000 retailers were operating nationwide by the mid-1990s (Govereh and others 2002). Private traders supplied 65 percent o f fertilizer usedby smallholders in 1999/00, most o f it paid for incash. 4.35 The current fertilizer schemes being operated by the government are free distribution through the Food Security Project and a 50 percent subsidy for fertilizer and seeds for hybrid maize production. The Food Security Project provides a 100 percent subsidy to the most vulnerable.22Under the 50/50 scheme, the government subsidizes 50 percent of the price o f fertilizer and some hybridseeds. The District Agricultural Committees, consisting o f representatives from the private sector, government, and nongovernmental organizations (NGOs), identify eligible beneficiaries. "The primary target groups are vulnerable but viable farmers: households headed by women, households headed by children, orphans, disabled farmers cultivating less than one hectare, households affected by natural calamities, and households with terminally illmembers. 43 4.36 Fertilizer supply under both schemes i s inadequate and unreliable. A Poverty and Social Impact Analysis (PSIA) of the programs, carried out by the World Bank, shows that both programs have had difficulties delivering their products on time and according to demand due to inefficiencies and fiscal constraints (World Bank, 2004d). There are considerable leakages in the 50/50 scheme: anecdotal evidence suggests that less than one out of six societies organized for the purpose of fertilizer distribution, among others, receives the bags of fertilizers promisedunderthe scheme. 4.37 Private firms are able to procure fertilizer at up to one-third less than the government programs. Part of the reason why government pays more i s that it uses cumbersome procurement processesthat force it to purchase fertilizer when world prices are at their peak (World Bank 2004a). In 1999/2000 the government distributed fertilizer at an averagecost ofUS$438 aton, comparedwith the farmgate price of fertilizer of US$356 a ton distributedthrough commercial channels (Govereh and others 2002). Private sector providers, however, are concerned with the unpredictability of the government interventions, whichmake early planning difficult. 4.38 World prices for fertilizers have fallen substantially since 1995, but domestic wholesale prices are still nearly triple international prices, and prices to farmers outside of Lusaka are higher still (World Bank, 2003c, Jayne and others 2003).23 More needs to be done to reduce this wedge between intemational and domestic prices, by reducing transactions and transportation costs throughout the marketing chain. A clear statement by the government on its intended role in marketing fertilizer would help by reducing uncertainty and risk for firms that are currently reluctant to import fertilizer inlarge quantities without knowingthe government's plans. 4.39 Neither the current government programs nor private traders effectively serve resource-poor farmers in remote areas. The use of fertilizer for many farmers in remote areas would not be profitable unless transportation and handling costs for fertilizer fell significantly and farmers were able to access markets for their output. Increasing the profitability o f fertilizer use for smallholders requires several measures. Research has shown that farmers often apply fertilizer inappropriately and therefore fail to achieve maximum returns to fertilizer use (Donovan and others 2000). A more active extension service that supplies information on the dose rates appropriate under various growing conditions i s therefore important. Creating demonstration plots in the districts showingthe retums to different levels of fertilizer would also help. D. Poor Rural Roads 4.40 A major constraint for smallholders is lack of good rural roads. The new transport policy, seeks to rehabilitate and maintain rural roads through the Road Sector Investment Program (ROADSIP), a sector-wide framework' established by the government and financed by donors. ROADSIP links communities to the govemment's main community-driven development program, the Zambia Social Investment Fund (ZAMSIF) for the rehabilitation phase and to the road fund for maintenance. When communities request roads, the roads engineer assists the contracting process and other activities. The communities form an association and register ownership o f the road under their name.24 After the works are completed, the community can apply to the National Roads Fund for cost-sharing inmaintenance (75 percent community, 25 percent road fund). The district engineer certifies the condition of the road on the annual applicationthe community makes to the road fund. 23 Nominal exchange rate depreciation may explain part of the difference. 24The Zambia Roads Act i s under revisionto accommodate this arrangement. 44 4.41 A Poverty and Social Impact Assessment of rural roads shows that provision of rural roads has a significant impact on the poverty and social conditions of smallholders (World Bank, 2004d). Qualitative analysis shows that greater use o f community-based or community-driven programs has benefited participating communities (Milimo et a1 2003). Impacts on livelihoods, services, accountability in resource use, social capital, and gender balance tend to be greater with community-based delivery of roads. ZAMSIF's collaboration with ROADSIP has shown that good community work can be well coordinated with traditional government agencies for greater poverty and social impact on the poor. 4.42 Remoteness leads to less land under cultivation, lower returns per household member, and lower returns to land. According to Alwang and Siegel (2003), net returns are about 10percent lower inhouseholds that are remote. Improved infrastructure may increasenet returns and confidence in markets. Such confidence may, over time, reduce the need for farmers to produce for their own consumption. Removal o f the food security constraint for a typical Zambian smallholder increases well-being by about 20 percent and makes production on less land more feasible. Together the reduction in remoteness and the increase in confidence in the market i s associated with about a 25 percent increase innet retums (Alwang and Siegel, 2003). E. Land Titles and Security 4.43 A dual system of land tenure in Zambia operates side by side, a legacy of colonial era laws. State land occupies about 6 percent o f the total and is governed by the principles o f English statutory law (Subramanian, 1998). Although state land i s technically held in the name o f the president for the benefit o f all Zambians, the government grants automatically renewable 99-year leases (commonly called titles) to all claimants who fulfill survey requirement^.^^ Generalizingabout customary tenure systems i s difficult, but the Zambian customary land tenure systems typically provide strong and individualizeduse rights,but limitedtransfer rights (Smith, forthcoming). 4.44 In general, access to land does not seem to be a significant constraint to agricultural development in Zambia. Both Milimo et a1 (2003) and Skjonsberg (2003) find that only farmers with links to commercial agriculture are concerned with obtaining land title. Alwang and Siegel (2003) suggests that due to low technology and resulting low productivity, a typical household could farm only about as much as it now does. Results from the same analysis suggest that, under existing technologies and facing existing constraints, typical households cannot cultivate more than 3.5 hectares using household labor. Lack o f credit and cash i s a major constraint, but using land as collateral may not increase credit and may have a detrimental social impact if the rain and harvests fail. 4.45 Zambia's new land policy proposes to transfer land from the traditional tenure systems to state title. Inthe draft policy, the government also commits to improved access to land for women and vulnerable groups by reserving and enforcing 30% o f land for them. A Poverty and Social Impact Assessment (PSIA) of this proposed reformwas carried out by the World Bank, the key findings o fwhich are as follows (World Bank, 2004d). 4.46 Conversion of customary land into state land insignificant quantities entails considerablerisks and limited benefits, especially for smallholders: 25Assumingthe lessee is incompliance with the conditions andcaveats containedinthe lease agreements. 45 0 Efforts to increase the amount of land under statutory tenure must be proceeded by improvements in the country's land administration system. The current land administration and dispute adjudication system for state (or titled) land i s weak and ineffective. In the absence of effective dispute resolution mechanisms, titling of customary land could cause conflict by shifting power to a centralized bureaucracy without strong accountability. The weaknesses of the Land Tribunal and the slow courts, i s likely to undermine any large scale titling reform. 0 Removing thepower of chiefs through large scale land-titling would have detrimental social impacts. Traditional authorities handle most of the day-to-day administration andenforce the traditional rules governing access to land-a power that provides the foundation for traditional authority. Transferring land to state control would remove that foundation, giving it to the government, which has very low capacity and presence at the local level. Diminishingthe traditional authority without replacing it with any other working authority could promote social discord. 0 A large scale land titling reform could endanger access to common property and natural resources. It could undermine the traditional system, which provides important safety nets and access to justice. All of this may affect negativelythe poor and the vulnerable. 4.47 What can the government do? The analysis (World Bank 200d) suggests several useful steps can be taken by the Government: e Strengthen the land administration and dispute adjudication systemfor state land. Problems with the existing state land system needto be fixed. Simplified and transparent procedures and improved capacity and efficiency of land administration should be the first steps to enable people on state landwho want to obtain title or transfer landto do so quickly and at an affordable cost. Where capacity and economic conditions are favorable, especially in the high-productivity areas and among emerging farmers, landtitling can then be tested gradually on a pilot basis. e Take actions to ensure that women have access to land within the traditional system. Customary tenure arrangements do not guarantee that women have access to land upon divorce or widowhood. The government might consider different legal ways to have automatic spousal co-ownership or at least inheritance, as other African countries do. Securing the rights of women to land will mean that the state will have to regulate customary land tenure systems and provide women with access to dispute adjudication mechanisms outside of the customary community. It will also require an information campaign to sensitize menand women to their rightsand how to enforce them. e Promote accountability within the traditional tenure system. The chiefs' transparency and accountability needs to be increased, and some internal (e.g. mandatingelections or approval by village committees) and extergal statutes (e.g. being able to enter into contract arrangements) needto be designed. Such statutes couldbe compulsory for a chiefs' authority to be legally recognized. e Encourage households that hold land but do not cultivate to sublet the land to those willing to farm it. According to key informant surveys in the Central and Southern Provinces, substantial high-quality land lies idle even in areas relatively well connected to markets and urban centers. Potential landlords with control over large tracts of land are reluctant to lease 46 land to smallholder households for fear of eventually losing control over the land, especially ifthe sametenant farms it for multipleyears(Zuluandothers 2002). Thishighlights theneed for stronger contract dispute mechanisms and sanctioned contractual arrangements to reduce transactions costs and the risks of leasing or renting land. 0 Open up new lands. Lack of infrastructure, inputs, tools, and technology are major constraints to brining new land under cultivation. Providing basic infrastructure and public services designed to link currently`isolatedareas with existing road and rail infrastructure and with basic public services can help encouragedevelopment and settlement of new areas (Zulu and others, 2002). IMPLICATIONS FOR RURAL POVERTYREDUCTION 4.48 Research shows that for most smallholders-the bulk of Zambia's agriculture sector-- labor is a constraining factor for agriculturalproduction(World Bank 1994, Alwang and Siegel, 2003).26The rural household model results suggest that allocation o f more land to smallholders without addressing labor constraints will have no real impact on productivity in smallholder agriculture. 4.49 Remoteness, or lack of access to markets, is an important factor perpetuatingpoverty amongsmallholders. Milimo et a1(2003) shows that access to markets is worse today than it was in the 1990s in all but one of the 10 communities studies, and it finds strong links between access to markets and ~ell-being.~'Alwang and Siegel (2003) shows that remoteness leads to less land under cultivation, lower returns per household member, and lower returns to land. Skjonsberg (2003) shows the importance of nonfarm income, education, and access to price information for successful farming, and of the problems of distance to markets and the resultingprohibitive costs of transport for both farm inputs and outputs. 4.50 The transitionfrom subsistencefarmingto commercialfarmingwill likely requireother complementary inputs. Most traditional smallholder crops, such as local maize, groundnuts, and cotton, have low input costs, but they also have relatively low returns to labor. Moving from these low-value to higher-value crops, such as paprika and tobacco, requires higher cash input and labor demands. Financial barriers to entry could be daunting for smallholders. Paprika represents an example of a new smallholder crop that proliferated in Zambia in the 1990sbut requiredsubstantial investments in farmer extension, input supply, marketing support, laboratory equipment, processing facilities, construction of rural depots, and negotiation with potential buyers around the world. Similar opportunities exist in other areas, but success depends on whether individual entrepreneurs have the meansto invest and the ability to take risks. 4.5 1 I fZambiais to beginto make inroadsintotransformingits smallholder agriculture,it will needto complementits growth and export diversification policieswith an array of direct povertyalleviationinitiatives,including efforts to enhancehuman capital through education and healthcare, scale up extension services to the poor and remote farmers, andimprove rural roads to 26A multivariate statistical analysis for Zambia shows there is an inverted Ushape relationship between age and expenditure per adult equivalent: middle-age heads o f households are able to spend more, whereas older and younger heads o f households are less well off. This suggests that the constraining variable for agricultural production is the availability o f labor. 27Improvement was due to a unique partnership among commercial farmers, community members, and local government, which may be a model for replication inZambia. 47 connect them to markets and social services. These investments are likely to be large and will require time to reach fruition. Complementary opportunities to stimulate nonagriculturaleconomic activities inruralareas should also be explored extensively.** Even if high seasonal labor demands create bottlenecks at key times, nonagricultural activities provide the opportunity to spread labor more evenly through the year (Alwang and Siege12003). 48 5. OPPORTUNITIES AND CONSTRAINTS TO GROWTH OUTSIDE AGRICULTURE MINING 5.1 Copper mining was the engine of economic activity in Zambia throughout the twentieth century. Although Zambia i s endowed with precious metals, gemstones, and agro-industrial minerals, the mining sector i s dominated by copper. Cobalt acquired some importance beginning in the mid-l980s, as the proportion o f copper ore associated with cobalt increased. Exports o f mineral products, mainly copper, still contribute about 60 percent o f the country's foreign exchange earnings. 5.2 Between 1965 and2002 copper production contracted at an average rate of 3.2 percent a year. The decline was particularly severe inthe 1990s, when the sector contracted at an average rate o f 5.1 percent a year. The privatization o f Zambia Consolidated Copper Mines in March 2000 reversed the decline, and for the first time in more than three decades, the sector has grown, at an average rate of 2.1 percent a year since 1999. Today the mining sector contributes 6-9 percent o f Zambia's GDP and accounts for about 10 percent o f formal sector employment. 5.3 International copper prices fell by an average rate of 3.4 percent a year over the past three decades (although there was a stretch in the mid-1990s when the price o f copper was significantly above the long-term trend). The price o f copper i s expected to move cyclically in tandem with world economic demand, but the long-term trend will remainnegative. 5.4 Zambia's copper industry i s not cost competitive, producing copper at an average cost o f US$0.58 a pound. About half o f world copper i s produced at or below US$O.3O-US$O.45 a pound, and the most efficient world producers have productioncosts below US$O.10 a pound. 5.5 World demand for copper is projected to increase about 2 percent a year for the next few years, but the long-term trend in the real price of copper i s expected to continue to decline by about 2 percent a year (see Box 5-1). The prospects for growth through copper are therefore modest at best. Zambia's copper industry will continue to suffer from declining world prices and its high relative costs. Inthe medium to long term, the case for export diversification cannot be overstated. 5.6 Despite unfavorable prospects, copper will continue to play a vital role in Zambia's economy in the short and medium term. The development o f the Zambian copper industrybeyond 2004 depends on the future outlook for the price o f copper, the evolution o f world copper production and consumption, and Zambia's ability to attract new investment. It will also depend on the government's ability to find a strategic partner to restructure Konkola Copper Mines as a going concern. So far the process o f finding such a partner has been slow. Inaddition to finding a strategic partner, management o f the copper sector on a commercial basis, as well as prudent management o f foreign exchange earnings and revenues from possible copper price recovery, will be critical. 49 .Prospectsfor CopperP forecasts three different sc umption. Under the optimistic tion, estimated at about 15 increases modestly, by about 2 price of copper ris nd, and Konkola Copper Mines reduces investment and operating co an attract investor interested in financing Konkola Deep Mining Project intime for it to enter production in2010. This scenario also assumes that there is investor interest insome of the smaller promising ore bodies, such as Kansanshi. Zambia's copper output could rebound by about 100,000 tons by about 2007-08 and increase by about 150,000 tons a year by about 2010, when KonkolaDeep MiningProject comes on line. Under the average scenario, world consumptionof copper increases very modestly, by about 1percent a year, and the price of copper reaches US$l.O7apound in 2005 and beyond. Konkola Copper Mines runs out of copper ore about2010 and i s unableto finance the ola DeepMiningProject, but other investors succeed in bringing into production some of the smaller cop e bodies, such as Kansanshi. Under such a scenario, Zambia's copper output falls between2004 and 2006, rebounds gradually until about 2009, and starts declining againby about 40,000 tons ayear about 2010. Under the pessimistic scenario, world consumptio stagnates at about 15 million tons a year or declines in2003 and for a few years beyond, the pri falls to US$0.95 apound in2005 andbeyond, and although Konkola Copper Mines is restructure ,its miningoperationsceaseinabout 2010, when the ore at existing mines i s exhausted. ze the Konkola Deep Mining Project. Zambia is ableto attract a limited amount of invest me ofthe smaller copper ore bodies. Source: World Bank 2003a. A. Opportunities for ExportDiversification:NontraditionalMining 5.7 Gemstone and nontraditionalmining offers a huge potentialfor export diversification. Zambia has rich reserves o f gemstone minerals, including the second-largest deposits of emeralds in the world, the largest deposits o f amethyst and aquamarine in Africa, and modest reserves o f tourmaline (Tyler 2002). The quality o f deposits, especially o f emeralds, i s judged to be among the best inthe world. 5.8 Interviews with Zambian firms indicate that value could be increased by a factor o f three for emeralds, aquamarine, and tourmaline and by a factor o f six for amethyst if current Zambian best practice were diffused. This means that annual sales could increase from the current recorded sales level o f US$20 million to US$830 million (Table 5-l).. This potential could be realized without huge capital investments, through the use o f retained earnings, learning by doing, and proper regulation. Potential sales come from rough minedproduction and processedgemstones and gemstone material. Table 5-1. PotentialGemstoneSales IfCurrentZambian BestPracticeWere Diffused ($ million) Aquamarine and Item Emeralds tourmaline Amethyst Total Recordedand unrecordedexports - - - 20-100 Potentialraw/ unprocessedmaterial 150-200 30 20 200-250 Potential with value-added 600 90 140 830 -- Notavailable. Source: Tyler 2002. 50 B. Constraints to Growth and Export Diversification 5.9 The key constraints to growth relate to regulation and supervision, policy discrimination, and barriers to investment, such as infrastructure (in particular air transport), bureaucratic barriers and poor agency services, and the highcost o f capital. e Lack of Regulation. The Mines and Minerals Act was passed in 1995, but the regulations accompanying the act have not yet been passed. The absence o f these regulations remains a major concern to local investors and to the legal advisers o f professional foreign investors. Weaknesses are evident in the tracking o f potential investors in the sector, in operating an efficient system o f logging, indisseminating information on available plots for mining, and in recording commercial mining a~tivity.'~Effective monitoring to ensure compliance with licensing rules i s not undertaken. e Lack of Market Linkages. Absence o f a fair value marketing system has constrained the supply chain and limited value-added in the sector. Market linkages are undermined by informal and illegal traders, whose prices may be half those o f legitimate dealers. What the government needs to do i s facilitate the creation o f a regulated but private sector-led gem exchange market that allows for fair and legal trading. e Weak Investment Climate. The broader investment climate affects investment and product development (see Chapter 3). Lacking are effective investment promotion, affordable finance, supportive infrastructure, and security (insome firms 95 percent o f the most valuable gemstones are reportedly stolen) [Tyler 2002). 0 Lack of Technology and Skills. Technology constraints include open pit mining practices (as opposed to underground mining), which are usually more costly and lead to lower employment, shorter lifespan for the mines, greater environmental degradation, and bad blasting practices, which make gemstones crack and substantially reduce their sales value.30 Allowing foreign investors to bring inmanagers and technicians from abroad could be a way to partially address the technology and skills constraints. MANUFACTURING 5.10 Before 1964 Zambia was a source of raw materials and a market for manufacturing activities, After independence Zambia embarked on a strategy o f import substitution, a policy that had dismal results (Table 5-2). Between 1975 and 1995 the manufacturing sector was either stagnant or declining, as most firms could not obtain the necessary imported intermediate materials or spare parts. 29One o f the major exporters notes it has been exporting tourmalines for four years, but the ministry records show no exports, eventhough it i s granted permission to export up to 10million tons a year. 30The largest trader in emeralds reports that 95 percent o f the gemstone material it handles is cracked and that 80 percent ofthe cracking is due to bad blasting. 51 Table5-2. Average Annual GrowthRates inManufacturing,1964-1990 (percent) Industry 1964-70 1970-75 1975-80 1980-84 1985-90 Food, beverages, and tobacco 40 3 -7 21 11 Textiles, garments, and leather 19 7 19 3 2 Wood and wood products 7 11 -8 -10 2 Paper printingand publishing 14 -2 -1 -4 24 Rubber chemicals petroleum 19 31 -5 -3 -2 Nonmetallic products 2 -5 1 3 5 Basic metals and metal products 5 11 -6 2 9 Other manufacturing 17 27 -9 -6 44 Total manufacturing 18 6 0 2 8 Source:Zambia, Central Statistics Office, 2002. 5.11 The dominance of copper mining in the economy has had adverse impacts on manufacturing,for several reasons. First, since many o f manufacturing industries, such as basic metals and metal products were designed to meet the needs of the mining sector, the decline in copper mining also meant the decline in Figure5-1. Structure of Manufacturing,1991-2002 these industries: basic metal and metal products declined from 22 percent o f total manufacturing sector value-added 60% in 1964 to about 12 percent in 1980 and 2 percent in 2002, Much o f the 29 - 4oy0t.~ ........................................................................ decline in basic metals 30% ! ................. (engineering products) - reflects inadequate foreign exchange to import the necessary intermediate inputs and spare parts. Second, 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 export diversification in -Fed,BeveragesandTobacco -Textile, andleather mbtries manufacturing was not vigorously pursued. -wood andwoodproducts -cherrdcals, h h randPlasticproclur There has been a strong -others forward linkage from the manufacturing to the Source: Zambia, Central Statistics Office, 2002 mining sector but very l j l e forward linkage from the mining to manufacturing sectors. This effectively underminedthe role o f manufacturing inthe economy, particularly inexports. 5.12 The manufacturing sector continues to be dominated by food, beverages, and tobacco, which represented about 60 percent o f total value-added inmanufacturing in2002 (Figure 5-1). Most of the food industries are concentrated in producing the staple maize meal (packaged roller and breakfast meal and the more nutritious and less expensive hammer-milled maize meal). The next-largest contributor to value-added are textiles and leather, at about 17 percent o f total value-added. Wood and wood products, chemicals, rubber, and plastic products jointly account for another 17 percent o f value-added. Together the four industries-food andbeverages; textiles andleather; wood and wood 52 products; and chemicals, rubber, and plastic products-account for about 95 percent o f the total value-added inmanufacturing. deterioration in the growth o f the Figure5-2. GrowthinManufacturingValue-Added, 1992-2002 manufacturing sector (Figure 5-2). Value- 50 ........................................................................................ added increased in food and beveragesand wood 1 1 ...................................... and wood products (from 48 percent in 1991 to 64 percent in 1995 for food and beverages and from 2 percent to 8 percent for ~ ~ wood and wood products), All other 1 industries recorded declines intheir shares o f -Manufacturing, total -Food, BeveragesandTobacco 1 value-added. I -......................................................................................... -Textile, andleatherindustries -M o dandw o d products J Source: Zambia, Central Statistics Office, 2002 5.14 The policy 5.4. Since 1996 the manufacturing sector has grown, although the recovery has been uneven. The industries that have significantly recovered are those that have had growing domestic demand due to population growth (food, beverages, and tobacco) and those that have had new entrants with a focus on exports (mainly textiles and leather products). For the food-related industries, value-added grew by an average 3 percent a year since 1996, prompted inpart by growing domestic demand owing to population growth (2 percent). For textiles and leather products, value- added increased by an average 13 percent a year between 1996 and 2002. The reorientation toward export markets helped restore the relative share o f valued-added inmanufacturing to the levels o f the 1980s. The textiles, garments, and leather industries (Lonrho Cotton and later Dunavant) increased the amount o f lint cotton significantly, as did other large spurn yam producers. Food and beverage and wood and wood products have steadily picked up in response mainly to favorable domestic markets. 5.5. Wood and wood products grew about 3 percent a year on average. Chemicals, rubber, and plastic grew 8 percent a year. Other industries-basic metals, nonmetallic minerals, and fabricated 31According to the November 2002 Bank Privatization Mission Report, with better sequencing and safeguards, the privatizationprogram implementedinthe early 1990s could have achieved better outcomes than it did. 53 minerals-remained stagnant or declined (fabricated metals). These industries are historically strongly associatedwith the fortunes of the copper sector. A. Opportunities for Export Diversification:LightManufacturingandProcessedFoods 1997 and US$34.5 million in 2001. Growth Figure5-3. Share of SelectedNontraditionalExportsinTotal has not been sustained Exports, 1990-2002 since 1997 because of various constraints, 60000~......-.-...- .....-.-...-.....-.......-...-.....-.....-................ . .-.- .- .~.~.".I including the decline in .......................................... the world price of cotton lint and yarn, but relative to the beginning of the 1990s export earnings have tripled (Figure 5-3). The share of textiles and ........................................................ garments in nontraditional exports 0 1 ' I stood at 11 percent in ,99" ,99" ,9P ,94" ,9gb ,9P ,9i6 ,94 ,9P ,9P +Q$Q'$Q% 0 2001-the third most important after --Textiles ----*--- agricultural exports and - - ProcessedAgricultural -IC-Engineering and garments processedfoods. Source: Zambia, Central Statistics Office, 2002. 5.8. Processedfoods offer another opportunity for export diversificationin manufacturing. During the 1990s the processed food industry more than doubled its contribution to nontraditional exports earnings from 6 percent in 1990 (US$6.2 million) to 13.8 percent in 2001 (US$43 million). In 2001 processed food export earnings were second only to exports of primary agricultural commodities. Sugar accounts for most of these exports (about 90 percent). Most of the remaining processed food exports consist of maize meal; wheat flour, and stockfeeds to neighboring countries. Zambia's proximity to Angola and the Democratic Republic o f Congo provides an opportunity for exporting processedfoods. B. Constraintsto Growth and ExportDiversification 5.9. Evidence from the recent Zambia Investment ClimateSurvey shows that the three key constraintsto firms are (WorldBank, 2004~): 54 Macroeconomic instability and the high cost ofjnance. About three-quarters of the firms surveyed report the cost of financing to be a major constraint to their operation (World Bank, 2004~). Uncertainty in government policy and regulation. Frequentand unpredictable changes to key policies have had a damaging effect on both foreign and domestic firms. For example, the immigration laws were recently changed requiring all non-Zambians to renew unexpired permits at high cost. Mining royalties were recently increasedby 150 percent without a clear rationale. Firmsreported that such shifts inpolicy exacerbate the risks associatedwith doing business inZambia and create a disincentive to invest. Poor infrastructure and administrative barriers. Poor infrastructure and the high costs of utilities, administrative and regulatory procedures, including multiplicity of licensing, inefficient customs clearance processes, high tariffs for intermediate inputs sourced from non-COMESA countries, and poor tax administration and export duty drawback systems (see Chapter 3). TOURISM 5.10. A window of opportunity exists for Zambia to position itself as one of the premier eco-, cultural, and adventure tourism destinations in the region. The World Tourism Organization projects average annual growth of tourism arrivals in Southem Africa of 5.4 percent (8 percent a year through 2008 and 4 percent thereafter through 2020). Zambian tourism faces serious competitors in the region, but those competitors provide an opportunity for Zambia to become part of regional tourism circuits in Southern Africa as well as a preferreddestination of its own. 5.11. Over the past 5-10 years, Zambia has witnessedrobust increasesin tourist arrivals and investment, which have increased employment and foreign exchange earnings. The share of GDP from hotels and restaurants, a major component of the tourism industry,rose from 2.0 percent in 1994 to about 3 percent in2002. The hotel and restaurant industryaccelerated interms of real value- added, growing at an average annual rate of 5 percent between 1991 and 1998 and 18 percent between 1999 and 2001, up from just 0.8 percent for 1985-90. Bed occupancy has also improved in recent years. The average hotel bed occupancy rate, which shows capacity utilization, grew from 49 percent in 1997 to 51 percent in 2001. The increase was buoyed by the opening of two major hotels inLivingstone by SunInternational, Zambia's hosting of the Organizationof African Unity Summit in2001, andthe occurrence of the solar eclipse in2002. The contribution oftourism to formal sector employment i s believedto have grown inrecent years.32 5.12. Arrivals from abroad are estimated to have grown at an average annual rate of 12 percent between 1990 and 2001, from 141,004 in 1990 to 491,993 in 2001 (Figure 5-4) (World Tourism Organization 2003). In 2002 the figure reached 556,000. Domestic tourism, measured by the number of overnight stays in hotels and similar establishments, also increased, from 99,000 in 1997 to about 185, 000 in 2001 (Zambia, Ministry of Finance and National Planning, Economic Report 2002). 32 Hotels and tourist services create jobs such as waiters, maintenance engineers, and drivers, which are relatively well paid. Indeed, tourism is labor intensive, with two employees required per hotel room in developingcountries, dependingon the type of hotel andthe level of local skills (Christie and Crompton 2001). 55 5.13. Arrivals in Zambia are dominated by business travel, which accounts for 55 percent of all tourist arrivals. The country's tourism potential continues to be largely untapped, with vacationers accounting for only 32 percent of total visitors. Other reasons for travel, including sports, religion, and study, accounted for 12 percent of visits during 1997-2001. The regionaldistribution of tourists i s dominated by visitors from other Africans countries (67 percent), followed by Europeans (25 percent). Tourists from Northand South America accounted for about 7 percent of arrivals. 5.14. Annual total tourist expenditure and foreign exchange Figure 5-4. Arrivals inZambia by Purpose of Visit, earnings from tourism grew more 1997-2001 than 10 percent in recent years (Figure 5-5). In-country tourist 600 expenditure grew from US$75 million in 1997to US$145 million in2002, an 500 average increase of 13 percent a year. .-$400 In 2002 tourist expenditures 300 accounted for about 4 percent of GDP. 84 0 Net foreign exchange earnings from 200 tourism appear significant, ranging 100 from 50-90 percent of gross tourist 0 expenditures (Christie and Crompton 1997 1998 I999 2000 2001 2001). At a conservative estimate of about 70 percent of tourism 0Leisure, recreation and holidays Business and professional expenditure, the foreign exchange 0Other earnings from tourism in Zambia in Source: World Tourism Organization 2003. 2002 were about US$102 million. Tourism dominates in the exports of Figure 5-5. Tourist Arrivals and Expenditures, 1997- nonfactor services and is a major 2002 foreign exchange earner in that category. Between 1997 and 2001 tourism averaged well over 50 percent of Zambia's total foreign exchange earnings from exports of nonfactor services. In 1997-2001 tourism showed growth rates in foreign exchange proceeds exceeding 10 percent.33 L o 1997 1968 1999 2000 2001 2002 A. Opportunitiesfor Growth j --C Internationaltourists (`000 overnightvisitors)' 5.15. Zambia i s rich in intrinsic 1+Tourism expenditure(US$million) tourism assets, which are essential to , Source:World Tourism Organization 2003. establishing a tourism industrythat can compete internationally. Its main assets include the spectacular Victoria Falls, the Lower Zambezi (South Luangwa and Kafue) and eco- and cultural tourism, which includes cultural assets reflecting distinctive local customs and song, dance, art and handicrafts, and museums that reflect the local 33The increaseoccurreddespite the delay inthe privatization of the Zambia ConsolidatedCopper Mines, which significantly slowed donor activities inZambia and must have affectedbusiness travel. 56 cultural or a wider global heritage. Tourism also includes small-scale eco-tourism lodges, where construction i s largely of local wood and thatch, and guests can view game either from the lodge's safari vehicles or on foot, accompaniedby a guide. 5.16. These assets provide a good basisfrom which to launchtourism development, but they are not sufficient in themselves. Natural assets must be accompanied by and packaged with appropriate and competitive built assets-accommodations, restaurants and other types of food service, tourist services, infrastructure-as well as a safe and healthy environment for tourists. The attractiveness-and therefore competitiveness-of the tourism product depends on the quality of these complementary services. B. Constraints to Growth 5.17. Development of the tourism industry requires creation of an environment that will attract investment in tourism and provide better accommodation and services to tourists. This in turn depends on a number of measures, such as good policy and program implementation, marketing, human resource development, upgrading of transport infrastructure, and improvement of health care services. In Zambia the key constraints in the sector have been poor policy implementation, administrative and non-administrative barriers to private sector investment and product development, and the deteriorating state of infrastructure, including poor access roads to tourist sites, limited international air network and unreliable local air network, and insufficient financial resources for startingbusinesses. Policy implementation. The government approved its first tourism policy in 1997. The policy was followed by preparation of,the strategic action plan in 1999, which translated the policy into a program of action for 2000-2005. The strategies focus on three main areas: management and marketing, product planning, and infrastructure development. Poor environmental planning and policy implementation, as well as poor coordination among govement agencies have impeded the sector's growth. The 2002 budget speech emphasized the need for improving auxiliary factors, such as solid waste management, street lights, and human resource development, in order to indirectly promote the development of the sector, but little has beenachieved due to declining andmisallocated budgets. Marketing of tourism. Inadequate marketing of Zambia as a tourist destination remains a constraint to developing the tourism industry. Recently, the Zambia National Tourist Board appears to have taken advantage of the Internet to market the country.34But more aggressive development and marketing of Zambia's tourist attractions may be necessary. Regional and domestic cooperation and marketing can help boost the critical mass of tourists required to make the tourist sector viable. This can be done by promoting regional tourism or byjointly marketing destinations in neighboring countries, Such marketing may attract tourists, boost occupancy rates, and iron out seasonal fluctuations intourism earnings. Zambia already has substantial numbers of regional visitors, but regional tourism could be further encouraged through cooperation with countries in the region. For joint marketing of destinations in neighboring countries, the Regional Tourism Organization of Southern Africa, established in 1995 by the Southern African Development Community and based in Johannesburg ,already functions as a regional marketing organization. Zambia should increase the marketing of its tourism products through this organization. 34 An extensive guide provides essential travel information and descriptions of Zambia's tourist attractions (see www.Zambiatourism.com). 57 Investment climate issues. The broader investment climate affects investment and product development intourism (see Chapter 3). Air transport. Lack of good air transport is a significant barrier to tourism development in Zambia. Scheduled carriers are among the most expensive per kilometer in Southern and Eastern Africa (Christie and Crompton 2001). South Africa i s the regional hub for air transport, outside which service i s less frequent and often erratic. Main carriers flying to Zambia include British Airways, South African Airways, Ethiopian Airlines, and Air Zimbabwe. The high cost of airfares limits the pool of middle-to lower-income tourists able to afford a vacation in Zambia. The local air transport network linking Lusaka to the main tourist attraction of Zambia-Victoria Falls-is served by Zambian Airways, which i s unreliable and has infrequent flights to the area. Safety and hygiene standards. Zambia's hotel accommodations and restaurants do not meet the required standards. Hotel standards are monitored by the Hotels Board of Zambia, but few inspections have been conducted inrecent years. The Zambia National Tender Board is responsible for monitoring the quality of restaurants, but few inspections have been conducted inthe past few years due to lack of transport. Health facilities near tourist sites. Many tourists perceive malaria as a major health risk. Public policy measures that guarantee hotel and service standards and safety, reduce public healthrisks,and improve medical facilities near tourist sites are critical to attracting tourists. Competitiveness of tourism services. Zambia i s generally perceived as a high-cost tourism destination, in large part because air transportation i s likely to constitute a large portion of tourist expenditure (especially for nonregional tourists). Other factors raising the cost of tourism inZambia are high local transportation, electricity, and fuel costs. 58 6. PROSPECTS FORGROWTH, DIVERSIFICATION,AND POVERTY REDUCTIONTHROUGH2015 BACKGROUND 6.1 This Chapter assesses Zambia's prospects for economic growth and poverty reduction between 2001 and 2015, drawing on the analyses and conclusions o f the preceding chapters.35It first considers the growth and poverty implications o f policy changes in four areas considered central to growth in Zambia: increased HIV/AIDS-related health spending, increased education spending to improve worker productivity, increased productivity and export diversification in agriculture, and increasedpublic investment in infrastructure, inparticularrural roads.36This scenario i s referredto as the "breaking with the past" scenario. Growth and welfare outcomes under this scenario are compared with growth under the base-case growth scenario and with the welfare path that represents "business as usual". The Chapter then assesses the impact o f changes in the external economic environment-namely, changes in copper production and copper prices and changes in external debt forgiveness-on growth and poverty reduction. 6.2 Assessment of the growth, sectoral, and welfare implications of Zambia's internal and external economic environment requires a framework that captures their repercussions throughout the economy. The model usedhere i s an economywide computable general equilibrium (CGE) model for Zambia that uses 2001 as its base year. The CGE model i s benchmarked, or calibrated, to the social accounting matrix for 2001. The properties o f the CGE model are described inAnnex E. 6.3 The rest of this Chapter i s organized as follows. First, the Chapter describes the structure o f the Zambian economy based on a social accounting matrix for 2001. Second, it presents the base-case scenario for 200l-15.Third, it analyzes the potential impact o f alternative government policies in accelerating growth and reducing poverty. Fourth, it assesses the impact o f changes inthe external economic environment on Zambia's growth and poverty reduction. Finally, it summarizes the prospects for Zambia's economy duringthe coming decade. Annex E presents additional details on the model assumptions and simulations and the detailed simulation results. 35 This Chapter i s based on a background study carried out for this report by the staffs of the International Food Policy Research Institute, using an updated Social Accounting Matrix and a Zambia Computable General EquilibriumModel (CGE). The background study is entitled "Prospects for Growth and Poverty Reduction in Zambia, 2001-2015", and was done by Hans Lofgren, James Thurlow and Sherman Robinson. 36 These policies are based on the analysis in the preceding chapters and reflect the government's PRSP. The HIV/AIDS and education scenarios address two problems seen as most pressing for Zambia in terms of their effects on human development (Government of Zambia 2002a). Accelerating agricultural growth and investing intransport infrastructure are complementary components ofZambia's rural development strategy. 59 Box 6-1. Database for the CGE Model The CGE model is used to simulate the impact of policies and economic shocks on growth and poverty. Its structure supports analysisoftradeoffs and synergies betweendifferent policies, the consequences of alternative financing mechanisms, and the extent to which foreign debt forgiveness can facilitate the task of reducing poverty. The model database, which captures the structural features of the Zambian economy, consists of a social accounting matrix; base-year and projected values for labor force, population, government policies, foreign savings, foreign borrowing, interest payments on foreign debt, and factor productivity; and a set of elasticities (for trade, production, and consumption). The social accounting matrix was constructed using input-output data, including an earlier social accounting matrix (Hausner 1999), as well as the database assembled by the World Bank for its 2001 Revised Minimum StandardModel of Zambia and the 1998 Living StandardsMonitoring Survey ( Evans, Robinson, and Thurlow 2003) Base-year data on the population of each household group and the labor force are from the 1998 Living Standards Measurement Survey. The population and labor force numbers were scaled to match 2001 totals from other World Bank data (World Bank 2003d). The AIDS-adjusted growth rates from 2001 to 2015 for The Revised MinimuinStandardModel provided the data on foreign debt for the entire planning horizon: the STRUCTUREOF THE ZAMBIAN ECONOMYINTHE BASEYEAR (2001) A. Production and Trade 6.4 The structure of Zambia' production and trade shows strong reliance on the primary sector in production and exports. Agriculture accounts for less than a quarter o f GDP, well below the Sub-Saharan average o f 32.8 percent (Table 6-1).37Mining, which i s dominated by copper, i s the largest export commodity.38Manufacturing accounted for more than 85 percent o f imports in 2001. Most o f these imports are capital goods used for investment. Services form a large part o f the economy, generating more than half o f GDP. The government accounts for only a fifth o f total services produced in the country. Education and health spending are the two largest components o f the government budget, accounting for 18.9 and 14.2 percent o f total government spending, respectively. Government spending on the agricultural sector (4.1 percent of the budget), transportation sector (3.3 percent), and industry (1.8 percent) i s comparatively low. Private services are dominated by the trade, transportation, financial services, andreal estate sectors. 37The average for Sub-Saharan(which excludes SouthAfrica) is taken from World Bank (2003~).Zambia data are from the 2001Zambianprovisional Social Accounting Matrix (Evans, Robinson, and Thurlow 2003). 38 Cobalt has recently become an important export, but since this commodity i s a byproduct of the copper miningprocess, it is tied to the performanceofcopper exports. 60 Table 6-1. GDP, Exports, and Imports, by Sector 2001 (percent) Share of total Exports' share Sector GDP Exports Imports of Imports' share output of demand Agriculturea 22.2 9.1 2.2 16.8 5.7 Staples 18.9 0.6 1.9 1.8 7.0 Traditional 1.3 5.2 0.0 76.6 0.0 Nontraditional 1.2 3.3 0.2 52.2 8.8 Other 0.9 0.0 0.0 0.0 0.0 Industry 25.7 85.8 86.0 28.1 46.3 Mining 9.1 53.3 0.7 85.0 15.1 Manufacturing 11.1 28.0 85.3 16.7 55.7 Other 5.5 4.5 0.0 7.1 0.0 Services 52.1 5.1 11.9 1.6 5.9 Private 41.9 5.1 11.9 2.1 7.4 Public 10.2 0.0 0.0 0.0 0.0 Total 100.0 100.0 100.0 13.4 25.4 a. Staples includes maize, groundnuts, wheat, vegetables, livestock, and other crops. Traditional agriculture include sugar, tobacco, and coffee. Nontraditional agriculture includes cotton and export horticulture. Other agriculture includes fishing and forestry. Source:Authors' calculations from the Zambian Social AccountingMatrix for 2001. 6.5 The agricultural sectors account for a majority of the value-added generated by lower- skilled labor (Annex D). The industrial sectors are the most capital intensive. Inmining almost all value-added i s generated by either the mineral resource or capital. The services sector i s intensive in both capital andhigher-slulledlabor. B. Household Income and Expenditure 6.6 Most Zambian households live in rural areas (Table 6-2). Historically, migration has been encouraged by the long-standing bias against agriculture created by a large mining sector and a protected manufacturing sector, but agriculture still remains an important source of income. Table 6-2. Population Distribution by Household Category, 2001 Household category Population Share of total (percent) Rural 6,507,407 62.6 Small-scale farm 5,724,640 55.0 Medium-scale farm 253,291 2.4 Large-scale farm 9,827 0.1 Nonfarm 519,649 5.0 Urban 3,892,593 37.4 Self-employed low skill 1,713,648 16.5 Private employees 811,704 7.8 Public employees 1,264,103 12.2 Self-employed skilled and employers 103,138 1.o Total 10,400,000 100.0 Source: World Bank 2003c for total population, Zambia Central Statistics Office 2000 for household distribution. 61 6.7 Rural households generate most of their income from returns to lower-skilled employment. The only exception is large-scale farms, which are more dependent on high-skilled labor and capital. Medium-scale householdsare most dependent on primary-educated labor and land, while small households rely on uneducated labor and land. By contrast, urban households generate most of their income from higher-slulled labor and capital. Mining resource returns are earned largely by households headedby public employeesandhigh-skilled, self-employed workers. 6.8 Rural households spend more of their income on agricultural and food products than do urban households (Table 6-3). Rural households consume fewer utilities as a share of their total spending. Urbanhouseholds have considerably higher consumption shares for services, as do large- scale rural farm households. Table 6-3. Household Consumption Patterns, 2001 (percent) Household category Share of total householdconsumptionspending Agriculture andfood Manufactures Utilities Services Rural 80.2 11.3 0.5 8.0 Small-scale farm 84.5 8.7 0.4 6.4 Medium-scale farm 67.7 18.8 0.5 13.0 Large-scale farm 48.5 26.3 3.6 21.7 Nonfarm 69.9 18.5 0.8 10.9 Urban 48.5 20.6 4.5 26.5 Self-employed low skill 53.2 19.2 4.3 23.3 Private employees 48.9 21.3 3.5 26.3 Public employees 47.9 20.0 4.6 27.5 Self-employed skilled and employers 35.6 25.4 6.1 32.9 Total 59.3 17.4 3.1 20.2 Source: Authors' calculations from the Zambian Social Accounting Matrix for 2001. Government and Rest of World 6.9 The government is dependent on direct taxes for its revenue, and runs a significant current account deficit. In2001 the government generated a majority of its revenue from personal taxes paid by higher-income urban households and from corporate taxes paid by non-agricultural enterprises. Combined current spending exceeded revenues, however, leading to a budget deficit of 3.6 percent of GDP (i.e., negative government savings). Imports greatly exceededexports in 2001 in the current account. Despite some remittances receivedbyhouseholds from the rest ofthe world, the current account balance was dominated by the trade deficit. The debt repayments also worsen the current account, which in2001 equaled20.1percent o f GDP. 6.10 Zambia i s typical of many Sub-Saharan countries in its structure of production, trade, household consumption and government revenue and expenditures. There i s a strong reliance on the primary sector both inproduction and exports. A majority of import demand i s for manufactured commodities, which are largely usedfor investment or intermediate demand. Services, which tend to be non-traded, are dominated by trade and transportation. Public services are also an important source of value-added inthe economy. The distinction between rural and urban households is strong. The former are more dependent on farm-labor and landincome, while the latter receive a majority of the income from capital and higher-skilled labor. Rural households spend more of their income on agricultural and food products, while urban households consume more services. Based on this information describing the Zambian economy, a CGE model i s constructed. 62 THEBASEGROWTH PATHSIMULATION 6.1 1 The World Bank's Revised Minimum Standard Model, a macroeconomic consistency framework, is used to project a macroeconomic growth path for Zambia's economy for 2001-15. Based on this projection, the Zambian CGE model i s used to determine the implications o f this growth path for disaggregated growth, trade, and poverty. 6.12 The basegrowthpathincludes the projectedeffect of HIV/AIDS on the accumulationof humancapitaland overall mortality (Table 6-4). Populationand labor force growth are taken from the IMF (2001) study on the impact of HIV/AIDS on the Zambian economy. All capital depreciates at 4 percent a year, which, given the capital-output ratio, implies a net profit rate o f about 20 percent. Table 6-4. Base-CaseScenario Assumptions Annual Item growth rate Source hercent) Real GDP 4.1 RevisedMinimumStandardModel (World Bank, 2003d) Population 2.0 RevisedMinimumStandardModel (World Bank2003d); IMF(2001) Laborforce Uneducated 2.0 IMF(2001) Primary education 2.0 IMF(2001) Secondary education 1.7 IMF(2001) Postsecondaryeducation 1.7 IMF(2001) Capital depreciationrate 4.0 RevisedMinimumStandardModel (World Bank 2003d) Total factor productivity 2.3 Endogenous Government spending Consumption 3.0 RevisedMinimumStandardModel (World Bank 2003d) Transfers 3.0 Assumed Foreign inflows (foreign savings)a 2.0 RevisedMinimumStandardModel (World Bank 2003d) Interest rate onforeign debt Due 2.8 RevisedMinimumStandardModel (World Bank 2003d) Paid 1.1 RevisedMinimumStandardModel (World Bank 2003d) Copper mining sector Output growth -1.0 Chapter 5 of this report World export price -2.0 Chapter 5 ofthis report a. Foreignsavings (the current account deficit) represents 20.1 percent of GDP inthebaseyear. 6.13 Transfer spending and real government consumption grow at 3 percent a year in real terms. Based on Revised Minimum Standard Model (RMSM) data, real government consumption grows at 3 percent a year. In later simulations, government consumption spending increases as new government policies are implemented. These include health spending for HIV/AIDS treatment, agriculture spending for agricultural investment, and transportation spending for improved rural infrastructure. 6.14 The current account deficit is expectedto graduallydecline over the comingdecade, at a rate of 2 percent a year. Three possible scenarios for the copper export sector are identified: a high, an average, and a low-case scenario (see Box 5-1). In the base growth path, it i s assumed that the average scenario i s realized and that output declines by 1percent a year and world prices by 2 percent a year, Alternative assumptions regarding the copper mining sector are explored later and are contrasted to this base assumption. 63 6.15 Interest payments due on government debt correspond to an annual rate of 2.8 percent. However, in the past, the government has paid interest at a rate of only 1.1 percent. This influences foreign debt growth, as any shortfall in interest payments i s added to the debt in the following year, with implications for future interest payments. 6.16 The base-case scenario assumes that the average-case scenario for copper (Box 5-1) is realized. World copper prices continue their decline from their current rate of US$l.16apound to US$1.07 in 2005, implying an annual decline in copper prices of 2 percent. This decline i s assumed to continue throughout the 15 years of the study. Output i s likely to decline by 100,000 tons during the period of consolidation leading up to 2006. This loss would be regained inthe years until 2010, when output would fall again by 40,000 tons. These projections imply an annual decline in output over the coming decade of 1percent. 6.17 The economy-wide total factor productivity (TFP) growth rate is adjusted to match the GDP growth rate of 4.1 percent for 2001. To attain this rate of GDP growth, annual TFP growth of 2.3 percent i s required. While this TFP growth rate is attainable if recent upward trends in TFP growth persist, it i s higher than the growth rates achieved over the past decade. 6.18 Real annual GDP growth of 4.1 percent matches the projected growth rate from the RevisedMinimum Standard Model (World Bank 2003d). The poverty headcount index for 2001 matches the estimate for extreme (food) poverty for national (55.8 percent), urban (43.4 percent), and rural areas (63.2 percent) for that year. The 1998 household survey indicates that 55.8 percent of the population fall below the national food poverty line (McCulloch, Baulch, and Cherel-Robson 2000), a poverty measure that may be more appropriate to gauge Zambia's progress toward the Millennium Development Goals. The slow decline in copper production and world copper prices causes export earnings to decline, other things equal, requiringmore rapid export growth, slower import growth, or both. As a result, growth in total absorption (the sum of private consumption, government consumption, and investment) falls short of GDP growth by about 1percent. The real exchange rate responds to this pressure by depreciating in order to encourage exports and reduce imports. The depreciation of the real exchange rate slows the growth of real investment, the most import-intensive component of GDP. 6.19 At the sectoral level, mining exports decline while agricultural exports expand. Following the depreciation, there is a shift inproduction toward those sectors that are either relatively strong exporters or that have a high but underutilized propensity to export, including agriculture. Accordingly, there is rapid export growth within the traditional andnontraditional agricultural sectors and overall strong growth in the agricultural sector. At the crop level, small-scale intensive cotton and medium-scale intensive horticulture grow fastest. Both sectors grow by more than 10 percent a year, a rate that i s consistent with recent trends. 64 competitiveness for agricultural exports, Figure 6-1. Long-Run Poverty Reduction under the Base-case this way, the failing scenario, 2001-50 mining sector contributes to faster declines in rural than urban poverty. Within rural areas, 70 ............................................................................................... -Rural households living on Urban -National small and mediuni-scale farms benefit most. However, the model does not disaggregate poverty spatially. Doing so may be important, given that cotton and horticultural production currently take 0 1 place in the eastern, 2001 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 central, and Lusaka regions, where access to markets i s better. The reduction in poverty i s 6.21 Under the base case, Zambia will be unable to meet the Millennium Development Goal of halving poverty by 2015 (Figure 6-1). Poverty will decline only moderately, from 55.9 percent to 49 percent. A longer simulation suggests that Zambia will not be able to halve poverty until after 2040. McCulloch, Baulch, and Cherel-Robson (2000) estimate the necessary GDP growth rate to halve poverty at 7-9 percent.39 6.22 The base growth path suggests that the government needs to adopt additional policies to accelerate growth, direct the gains from growth more effectively toward the poor, o r both. Such policies would have to be implemented in an environment in which external factors continue to force the economy through a process o f structural adjustment. Perhaps the most important of these are developments in the copper mining sector and the possibility o f external debt relief under the enhanced HIPC initiati~e.~' 39McCulloch, Baulch, and Cherel-Robson(2000) base their estimate on growth-poverty elasticities calculated fromthe 1991, 1996, and 1998 householdsurveys. The estimatedgrowth rate neededvaries dependingon the assumptions about the fixed distribution of consumption across households. 40 Receipt of relief is not purely exogenous, as it depends on the government's performance in meeting the HIPC program conditions. 65 SIMULATION OF ALTERNATIVEGOVERNMENT POLICIES 6.23 The scenarios in this section analyze the impactof policy actionsthe government could implement over the coming decade. Four areas for policy changes are considered: increased HIV/AIDS health spending, increased spending on public provision o f education, measures that accelerate agricultural productivity and export diversification, and increased public investment in transport infrastructure. 6.24 For most o f these policy changes, the actions considered generate significant changes in the balance between government spending and revenues, with consequences for domestic government borrowing and other components o f absorption. To explore the impact o f this repercussion, some o f the simulations are implemented both with and without complementary changes 'in direct tax rates. This additional tax revenue from mainly urban households is sufficient to keep the government's domestic borrowing each year at the same level as under the base-case scenario. Furthermore, given that the results o f the HIPC scenario suggest that debt relief on its own i s not likely to greatly affect poverty, simulations inwhich the government funds saved through HIPC relief are used to implement pro-poor policies are presented. A. Increased HIV/AIDS SpendingSimulations 6.25 HIV/AIDS has become the greatest threat to development in Zambia, reversing past achievements in generating economic growth, improving life expectancy, and decreasing child malnutrition (Cheru 2000). In light o f this crisis, the government may have little choice but to expand its current level of AIDS-related health spending. 6.26 Inits Poverty ReductionStrategy Paper, the government has outlineda series of actions to be taken in response to this growing threat, including educational programs, community-based home care, andmeasures targeted at orphans and vulnerable children. While there i s some scope for private sector involvement, it seems likely that a majority o f domestic health spending will come from the public sector. 6.27 The main channels through which HIV/AIDS is assumedto influence the real economy are through the effects on the growth rates of the population,labor force, and TFP. The base- case scenario includes HIV/AIDS. To gauge the extent to which H N / A I D S has reducedGDP growth and increased poverty, the AIDS treatment scenarios increase population, labor force, and TFP growth rates toward the levels that the IMF (2001) estimates would have prevailed in the absence o f HIV/AIDS. The assumed changes in these growth rates are presented in Table 6-7 and discussed in detail inAnnex E. 66 Table 6-5. HIV/AIDS Scenario Assumptions (percent) AIDS treatment Item Base-case scenario (annual growth rate) scenarios (annual growth rate) Total factor productivity (TFP) 2.3 2.8 Population 2.0 3.3 Laborforce Uneducated 2.0 3.3 Primary education 2.0 3.3 Secondary education 1.7 2.5 Postsecondary education 1.7 2.5 Source: Base-case scenario population is from Revised MinimumStandardModel. Labor force and AIDS scenarios are from IMF2001. TFP is from Zambia model. 6.28 Population growth would be significantly higher were it not for the highmortality causedby HIV/AIDS. Mortality is also captured through a decline in labor force growth rates, which are assumedto disproportionately affect lower-skilledworkers. Morbidity i s captured through declines in the TFP growthrate. Beyondmortality andmorbidity, the declines inthe labor force andTFP capture decreased work hours and increased care-giving for non-sick workers. No adjustment i s made for changes in household consumption patterns to account for increased health spending. Higher dependency ratios are implicitly captured in the model, since the population growth rate exceeds the labor force growth rate by ahigher percentagethan under the AIDS treatment scenarios. "Costless" Treatment Scenario 6.29 In light of the objectives stated in the Poverty Reduction paper strategy, the first AIDS scenario assesses the impact of removing the effects of HIV/AIDS without requiringthe government to increaseits level of health spending (Table 6-5). 6.30 Comparing this scenario to the base-case scenario suggests that HIV/AIDS costs the Zambian economy almost 1percent in GDP growth a year. That is, inthe absenceof HIV/AIDS, the Zambian economy would have grown 5 percent a year. Increased factor growth rates and TFP allow production to increase, raising the level of GDP. At the sectoral level, growth acceleratesmost for agriculture, the part of the economy that i s relatively intensiveinlabor with little or no education. 6.31 The removal of HIV/AIDS has little impact on per capita consumption or the total poverty headcount rate. Inreality there are likely to be some differences ininfectionrates across rural andurban areas, with implications for the distribution of gains across rural andurban areas. (This informationwas not available for this study.) With the exception of South Africa, where rural infection rates are highest, other countries inthe regionhave lower infection rates inrural areas. Public-Financed Treatment Scenavio 6.32 A second scenario considers the economic impact of treating half of the infected population, at a cost of about $360 a year per person, with the government bearing the full cost. This amounts to treating 650,000 people, at a total cost of 7.1 percent o f GDP in2001. Such a large increase in spending raises total government expenditure by 50 percent in 2001, more than tripling current government health spending. It i s assumed that the additional health spending will have to 67 continue through 2015. Since only halfthe population i s treated, the acceleration inpopulation, labor force andTFP growth i s halfwhat i s shown inTable 6-5. - E j i m 1 d 00 li U i 4 d 9 g 70 6.33 The increases in the labor force and TFP have a positive effect on GDP growth (4.6 percent comparedwith 4.1 percent in the base case), but the cost of the treatmentprogram greatly outweighs these gains (Table 6-6). Government consumption increases from 3 percent inthe base-case scenario to 6 percent inthe publicly financed scenario (see Annex E). The rate o f real investment growth increases but by less than GDP growth, while household consumption stagnate^.^' However, the increased economic growth caused by a greater supply o f productive factor resources leads to an accelerated investment growth rate o f 2.9 percent. Rising investment demand increases import demand, with exports rising alongside GDP. At the sectoral level, growth acceleration' i s spread across the three main sectors. Agriculture grows faster under the current scenario, although growth is concentrated in export crops. Manufacturing and services benefit from the increases in investment demand and government consumption. 6.34 The burdenof publicly financed AIDS treatment is evident inthe declinesin annual per capita consumption growth, by about 0.5 percent for most householdgroups. Poverty increases moderately, by slightly less than 1percentage point. The results for this scenario show that if government consumption i s increased as required while investment i s protected (in terms o f its absorption share), cuts in private consumption will be necessary. These losses should be compared with the gains o f treatment in the form o f increased longevity and healthier lives for those treated. Balanced Budget Scenario 6.35 The balanced budget scenario considersthe impactof treating the same 50 percent of the population as in the previous simulation but with a compensatingincreasein direct tax rateson domestic institutions. The growth rate of government consumptionstill risesby the same amount as in the previous simulation, but increased revenues through increased direct taxation offset the increase in government spending. The average aggregate direct tax rate on households almost doubles, from an initial average rate o f 8.5 percent to 12.4 percent.42 6.36 The growth impact of this policy is the same as the previous simulation (4.6 percent). The increase in revenues prevents an increased budget deficit from crowding out investment. However, the higher tax rates imposed on the country's higher-income households reduce their level o f savings. Ultimately, investment grows at 2.9 percent a year. The increased tax rates reduce the level o f disposable income for higher-income urban households, reducing the level of private consumption growth. While investment growth raises import demand, it i s offset by falling import demand from urban households. 41 The distribution o f growth between household consumption and investment is influenced by the macro closure rule that investment is a fixed share o f nominal absorption, with balancing adjustments (inthis case an increase) in household savings rates. If, alternatively, household savings rates were fixed, household consumption would increase in the short to medium run, untilthe resulting decline in capital stock growth reduced consumption. 42 The average rate i s calculated based only on the incomes received by households and corporations that pay direct tax. Including all rural households would greatly reduce the average tax rate. The stated tax rate i s based on direct tax collections and not on the statutory tax rate stated in the tax register. All direct tax rates for households are scaled (in this case upward) to generate the required level o f government savings. Given that households with high per capita incomes tend to have relatively high effective tax rates, the distributional impact o f this adjustment mechanism i s progressive. 71 6.37 The national poverty headcount i s higher than under the base case but slightly lower than under the publicly financed program. While all households enjoy more rapid poverty reduction inthe balanced budget scenario than under the publicly financed scenario, rural households benefit most. Rural households benefitboth from the increased incomes from higher labor force growth and from being excluded from the financing o f the treatment program. Urban households suffer an increase in poverty due to the higher tax burden, although lower-skilled self- employed households have higher per capita consumption growth in the balanced budget scenario. HIPC-Funded AIDS Scenario 6.38 In the third AIDS treatment scenario, the funds saved by the government through debt relief are used to finance an AIDS treatment program.43Unlike in the previous AIDS simulations, where 50 percent o f the infected adult population was treated, treatment o f just 14 percent o f the infected population can be financed through debt relief funds. The increases in population, labor force, and TFP are adjusted downward accordingly. 6.39 The growth and poverty impact of this policy is small, given the limited scope of coverage. An increase in government consumption i s still required inthe HIPC-funded scenario, but there is no increase in the government deficit. As a result, private consumption per capita increases for all household groups. Gains are small, however: even if HIPC funds are used exclusively for AIDS treatment, they are able to provide treatment to only a small percentage o f the population. 6.40 Raising government health spending to address HIV/AIDS may involve tradeoffs between gains for direct beneficiaries and losses for those who finance the increases in government spending. Alternative financing options should be explored. Although the use o f HIPC funds to finance spending produces the most pro-poor results, the opportunity cost o f such funds mayprohibit their use for AIDS treatment. B. IncreasedEducation SpendingSimulations 6.41 Expansionof pro-poor education spending remains a priority in Zambia-and it is one of the criteriafor continuedHIPC debt relief. Zambia's education system is characterized by inadequate provision at every educational level, as a result o f insufficient numbers o fteachers, inadequate teaching materials, and other problems. The lack of teachers i s driven by changes in government requirements, poor teacher benefits, and death due to HIV/AIDS. Dropout rates are high, even at low grades, as a result o f the long distances to schools and the opportunity cost of children's time. More than 10 percent of children exit the schooling system between grades 1 and 6 (Government o f Zambia 2002a). 6.42 The government has identified a number o f education initiatives, including increasing investment in basic education; enhancing the existing literacy program; promoting on-the-job skills training; promoting equity, especially gender equity; and investing in secondary and tertiary schooling. Expandedteacher training i s also needed, especially given the fact that a large number o f teachers are infectedwith HIV/AIDS (Government o f Zambia 2002a). 43 The use o f HIPC funds to pay for AIDS treatment is equivalent to the use o f any funds received from abroad. As such, this simulation can also provide insight into donor-assisted AIDS treatment. 72 6.43 The education scenarios simulate the effect that increased government education spending has on the skill structure of new labor entering the labor market. In the first simulation the government i s assumed to finance additional education spending by increasing direct taxes. In the second simulation funds from HIPC debt relief are used. Under the HIPC- funded scenario, the government budget deficit remains fixed and the cost o f the education program i s covered by funds normally allocated to foreign debt. For both scenarios, the growth rates inthe total economy-wide labor force and the total labor force o f each household group are the same as for the base simulation. Changes in labor supply for each skill level are allocated across the different households in proportionto their shares inthe total labor force. Table6-7. Education ScenarioAssumptions Initial work force Base-case Education: Education: (thousands of scenario HIPCfunded people) funded Item Annual percentage growth rate Laborforce growth rate Uneducated 2,071 2.0 -1.3 1.o Primary education 1,500 2.0 4.8 3.1 Secondary education 252 1.7 3.6 2.4 Tertiary education 144 1.7 1.7 1.7 New cost as apercentage of GDP 5.4 2.0 Source: Authors' calculation (see Annex E for details). 6.44 In the publiclyfunded simulation, government spending on education doubles. The additional spending, which equals 5.4 percent o f GDP, i s offset by an increase in direct taxes, while government savings i s fixed at the level o f the base-case scenario. The expansion in education spending under both scenarios allows labor force growth rates among the higher-skill categories to increase. Conversely, since one-third o f additional spending i s allocated to secondary schooling, the supply o f workers with secondary education increases relative to the base case. People are therefore drawn out o f lower education categories and moved into higher ones. 6.45 Both scenarioslead to a moderate accelerationin GDP growth, from 4.1 percentin the base case to 4.4 percent (Table 6-6). Under the publicly funded scenario, government consumption increases sharply. Investment expands inrough proportion to the general expansion o f growth, whereas total household consumption declines slightly. The growth increase i s driven byan increase inTFP, reflecting higher marginal productivities for more educated labor. 6.46 In terms of welfare, the effects of the first scenario are strongly pro-rural: consumption increases in rural areas, while poverty decreases by more than 2 percentage points. By contrast, in urban areas poverty increases slightly and consumption declines. This outcome reflects the combined impact o f changes inthe skill structure and a financing mechanism that favors households with lower initial direct tax rates (including low-income rural households). On the production side, the only major change is more rapid growth in public services, responding to the changes in government demand. 6.47 Under the HIPC-funded scenario, in the absence of the need to devote local resourcesto the program, householdconsumptiongrows more rapidly, in tandem with the rest of the economy, while government consumptionsremainsunchanged. Inrural areas this scenario permits a more rapid but somewhat less pro-poor increase in consumption. In urban areas consumption grows much more rapidly than under the publicly funded scenario, permitting 73 poverty to fall slightly (from 39.percent to 38 percent). Among the production sectors, gains are strongest for agriculture and services. C. Increased Agricultural Productivity/Export Diversification Simulations 6.48 The simulations consider the impact of accelerated TFP growth in selected agricultural sectors. In the first simulation TFP growth increases 0.5 percent a year for all agriculture. The other simulations target a smaller share o f GDP (export agriculture or agricultural processing). Inorder to make these simulations comparable with the first, the rate o f additional TFP growth i s scaled in inverse proportion to the size o f the targeted sector relative to all o f agriculture (Table 6-8). Given that the purpose o f these simulations i s to identify differing growth and poverty-reduction potentials across agricultural sectors, no attempt i s made to account for the costs o f raising TFP under these scenarios. Table 6-8. Increases in TFP Growth for Agriculture-Focused Simulations, 2001-15 Sector Share of agricultural Scaled additional TFP GDP (2001) growth rate Agriculture 100.0 0.5 Staples 85.1 0.6 Traditional 5.9 8.5 Nontraditional 5.4 9.3 Agriculture-processing and textiles 21.2 2.4 Note: Staples include maize, sorghum, millet, and drought-tolerant crops. Traditional agriculture includes coffee, sugar and tobacco. Nontraditional agriculture includes cotton and horticulture. 6.49 Increasing TFP growth raises the level of production in the targeted sectors. Increased supply lowers the prices o f their outputs, raising quantities demanded for agricultural products in both urban and rural areas. Part o f the additional output i s exported. As long as the demand elasticities are sufficiently high, incomes and consumption increase for both agricultural and nonagriculturalhouseholds, with more rapid growth and a larger decline inpoverty. 6.50 Although accelerated agricultural growth generally increases economic growth and lowers poverty, there are considerable differences across the different scenarios. For the first scenario, in which all agriculture i s targeted, growth increases from 4.1 percent in the base case to 4.2 percent. The gains from accelerated agricultural production are spread unevenly across household groups. Small-scale rural households do not benefit, whereas nonfarm rural and urban households benefit through declines in poverty (by 0.5 percent versus the base case). This urbanbias inpoverty reduction i s drivenby market constraints inthe staples sector. 6.51 Increases in TFP growth for the staples sector increase real GDP growth (from 4.1 to 4.2 percent a year) but marginally increase the headcount poverty rate in rural areas compared with the base-case scenario (from 55 percent to 55.4 percent). Inthe context o f . highly inelastic demand, the fall in staples prices, following the large increase in supply, is so large that producer value-added declines. These declines in price are due to the limited ability o f the domestic market to absorb the higher level o f supply as well as limited access to export markets. 6.52 Growth of traditional exports increases from 4.1 percent a year to 4. 4 percent a year, and nontraditional exports grow 4.7 percent. For these commodities, producers have better access to foreign markets, where they can sell their output following the increase in TFP growth. Domestic markets are therefore not flooded by new supply, and prices fall by less in this 74 scenario than they do in the staples scenario. High growth in export supply generates substantial appreciation o f the exchange rate. Falling import prices favor investment, which grows more rapidly under the traditional and nontraditional scenarios than under the base case. 6.53 Rural poverty falls from 55 percent in the base case to 49.6 percent under these scenarios. Rural households are better off due to higher production incomes, while urban households generally benefit from cheaper imports from the appreciated real exchange rate and linkages to agriculture, as nonagricultural sectors respond to increased consumption demand from agricultural households and to increased demand for intermediate inputs from agriculture. The impact on real GDP growth i s modest, rising from 4.1 percent a year in the base case to 4.3 percent. 6.54 The final simulation in this group considers the strength of backward-linkage effects from agricultural processingsectors to agriculture. As with the traditional and nontraditional sectors, the processing sector already has an established foreign market in which it can sell its output. However, the simulated reduction inpoverty marginally favors urban households, which earn increased retums to nonagricultural labor and capital (urban poverty declines from 38.9 in the base year to 38.2 percent, while rural poverty declines from 55.0 percent to 54.4 percent). While backward linkages from the agricultural-processing sectors do stimulate poverty reduction inrural areas, the fall inpovertyresultingfrom accelerated growth inthese sectors is smaller than for agricultural sectors. 6.55 The overall conclusion from these simulations is that agricultural sectors with strong export potential are appropriate targets for efforts to raise productivity and reduce poverty. On a cautionary note, however, it is clear from current regional patterns that the gains from accelerated agricultural growth would not be evenly distributed across rural areas. This dimension should be considered when strategies for rural development and poverty reduction are formulated. For example, horticultural production tends to be concentrated close to the national airport, and cotton production i s currently located in the Eastern and Central provinces. Gains from the nontraditional scenario are unlikely to benefit farmers in other areas o f the country. Similarly, much o f the country's export agriculture i s concentrated near the main rail and road transport routes. It is questionable whether farmers in the more remote Western and Northwestern provinces would be able to gain access to foreign markets. Furthermore, even if foreign staples markets could be identified, the highcost o ftransportation inmore remote areas o f Zambia i s likely to erode the competitiveness o f potential exports. D. Investmentin TransportInfrastructureSimulations 6.56 Investment in transport infrastructure i s key to facilitating farmers' access to input and output markets and to promoting agricultural growth and rural poverty reduction. Transportation costs currently account for 60-70 percent o f the cost o f production in Zambia (Government o f Zambia 2002a). These costs, which are high even by regional standards, contribute to the highprice o f goods and services and hence to poverty. Although Zambia has an extensive road network, the system has deteriorated in recent years (Government o f Zambia 2002a). 6.57 Market access i s especially limited in many rural areas. Only 18 percent o f rural households are within five kilometers o f input markets. Because of lack o f or inadequate roads, few remote households have access to health and education facilities, both o f which are critical for development. 75 Maintenance Scenario 6.58 About 29 percent o f the current paved road network is in need of repair, and much of the system i s no longer maintained. The maintenance scenario simulates the impact o f performing the necessary repairs and maintaining the existing system. It i s difficult to simulate the impact o f investment in transport given that the returns to such investment are often indirect and difficult to quantify. (For informationon the cost o f repairing and maintaining see Annex D.) 6.59 Assuming that only paved and gravel roads need maintenance, the cost of repairing and maintaining the current system would equal 3.5 percent o f GDP for the first three years. After the thirdyear, roadrepairs are assumed to be completed andno new repairs are needed, as long as the system i s maintained. From year four onwards, the cost o f maintenance i s therefore the only additional cost to government, amounting to 1.3 percent o f GDP a year. The depreciation rate for the transport sector is halved, from 4 percentto 2 percent. 6.60 Under this scenario the growth rates for all GDP components and the capital stock increase by about 0.5 percent, and poverty falls by 2.2. percentage points, with a stronger reduction in rural areas. Reducing the depreciation rate o f the transport sector increases its supply, bringing down the cost of trade and transport services, increasing the overall supply of capital, and adding to GDP growth. All household groups gain. Growth i s stagnant for agriculture but increases for services and nonmining industry, in part a reflection that these nonagricultural sectors include the suppliers o f transport services and commodities that are part o f government consumption. Construction Scenarios 6.61 The first two scenarios contrast the effects o f building new feeder roads in rural areas with the effects o f building new pavedand gravel roads inless remote rural andurban areas. The scenarios consider a 10 percent increase in the supply o f feeder or paved roads, with the cost o f these programs calculated usingthe information provided inAnnex E. Inboth simulations there i s a reduction in the transactions costs face by selected sectors in the economy. The benefits o f feeder roads are limited to agricultural production; pavedroads benefit nonagriculturalproduction and export agriculture. The effect on export agriculture reflects the existing concentration o f export agriculture along the country's main road networks. Paved roads are expected to reduce transactions costs in both domestic and export markets, whereas feeder roads reduce only transactions costs indomestic markets. 6.62 Both construction programs increase annual average growth (from 4.1 percent in the base case to 4.2-4.3 percent) and reduce national poverty (from about 49 percent in the base case to about 46 percent). GDP growth under both scenarios i s similar, although the paved roads scenario has a slightly higher growth rate. The two simulations differ strongly in their household effects, however: 0 Feeder roads in remote rural areas affect rural households more than urban households. Rural consumption increases considerably more rapidly under this scenario, with final-year rural poverty declining more than 4 percentage points and urban poverty decliningjust 2 percentage points. 0 Under the paved road construction scenario, urbanhouseholds gain more than rural households in terms o f consumption, while poverty reduction remains unchanged. Poverty falls by 2.5 percentage points in rural areas. Under this scenario GDP 76 growth i s stronger (4.3 percent), while poverty reduction i s weaker relative to the feeder roads scenario. These results highlight the tradeoff betweenpoverty reduction and growth, pointing to the needto consider bothwhen developingpolicies. 0 Under the last construction scenario, HIPC funds are allocated to spending on feeder roads.44At the household level the effects are close to the earlier feeder construction scenario, with a substantial decline in poverty (by about 3 percentage points) and consumption gains for rural households and slightly smaller gains for urban households. SIMULATION OF ZAMBIA'S EXTERNAL ECONOMICENVIRONMENT A. Copper Sector Simulations 6.63 Export prices o f copper have fallen almost 50 percent since the 1995 peak.45Copper mininginZambia has also been adversely affected by the withdrawal o f investment by the Anglo American mining company. The government has not yet found a strategic partner to run the mines on a commercial basis. 6.64 The assumed annual output and price changes for the three scenarios are presented in Table 6-9 and described in detail in Annex E. In the low-case scenario, both output and world prices are expected to decline at an annual rate o f 5 percent for the entire 15-year period. Inthe high-case scenario, prices are expected to increase at an annual growth rate o f 0.5 percent and output i s expected to rise at an average annual rate o f 4.3 percent over the 15-year period. Table 6-9. Output and World Price Changesfor the Copper Mining Sector, 2001-15 (annual percentage change) Average-case scenario Item High-case scenario (base-caseassumption) Low-case scenario World copper export price 0.5 -2.0 -5.0 Copper mining output 4.3 -1.0 -5.0 Source: Calculations based on scenarios from Box 5-1. High-Case Scenario 6.65 The average annual growth rate under this case is slightly higher than under the base case, but export growth and diversification are lower. Relative to the base case, mining exports reverse their decline and begin to grow, albeit at the relatively slow rate o f 0.7 percent a year. This growth in exports relieves some of the pressure on the current account under the base- case scenario, resulting in a slower depreciation o f the real exchange rate. Accordingly, while mining exports grow faster than under the base-case scenario, all other export growth rates decline. The incentive to diversify production and exports away from the mining sector has been partially removed. By contrast, the lower depreciation o f the exchange rate reduces the import price relative to the base-case scenario. Since investment demand i s highly import intensive it grows faster under the high-case scenario. Higher investment compensates for slower export 44The effects of road constructionhave been scaleddownto reflect the volume of available funds underthe HIPC debt relief agreement. 45For an analysis, see Lofgren, Robinson, andThurlow (2002). Priceshadrecoveredas of early 2004. 77 growth and results in a GDP growth rate (4.3. percent) that i s slightly higher than that o f the base-case scenario. 6.66 All households are better off under the high-case scenario, as reflected in more rapid private consumption growth and the lower poverty headcounts (Table 6-10). Urban households benefit the most from improved copper miningperformance, since these (high-skilled employer) households are the main recipients o f miningincome. Low-Case Scenario . 6.67 GDP growth under this case is slightly lower than that under the base case (3.9 percent), but export growth and diversification are higher. Both output and world prices fall rapidly, putting pressure on the current account. The real exchange rate depreciates in order to offset the fall in export earnings by encouraging production o f tradables while discouraging their consumption locally, thereby raising exports and reducing imports. The agricultural sectors, which have the highest export intensities, respond most. Within agriculture, traditional and nontraditional crops sectors grow fastest. Depreciation also raises the price o f imported capital equipment and vehicles, which form the bulk o f fixed investment demand. Accordingly, investment growth declines slightly more strongly than private consumption, with a negative impact on GDP growth. 6.68 The poverty headcount rises compared with the base-case scenario, with rising poverty in urban areas and little or no effect on rural poverty (Table 6-10). Consumption growth slows relative to the base case for urban households and relatively well-off rural households. Even though the overall poverty headcount rises, rural households experience a slight decline in their poverty headcount (not evident from the table), while the urban rate increases by 1 percentage point. The cushioning o f rural poverty i s driven by growth in agriculturalexports as a result o f the more heavily depreciatedreal exchange rate. 6.69 The results of the base and copper mining scenarios suggest that, while falling copper output and prices have a negative impact on economic growth, a further collapse of the copper mining sector does not greatly affect rural households. In fact, the decline inthe copper sector might paradoxically result in a pro-poor contraction o f the economy relative to the base-case scenario.46One way for the government to avoid the declines in the GDP growth rate experienced under the low- and average-case scenarios would be to increase foreign borrowing in order to supplement falling export earnings. However, such a move would bringabout a slower depreciation o f the real exchange rate, diluting any incentives for the country to diversify production o f tradables and exports away from mining. Given Zambia's longstanding dependence on copper mining and the need to overcome the vulnerability caused by threatened closures o f mines and price falls, the government should be hesitant to drastically alter this restructuring process. 46See Baulch and McCullouch (2000) and Ravallion (2003) for alternativedefinitions of pro-poor growth. 78 Table 6-10. Simulation of External Economic Environment: Effects on Growth, Diversification, and Poverty, 2001-15 Initial value Base growth Copper: Copper: low HIPC: (2001) path high case case debt relief Real GDP (Mn kwacha) - Annualpercentage growth rate (2001-201 5) - GDP (factor cost) 11,843 4.1 4.3 3.9 4.1 Real GDP (Mn kwacha) ~ _ _ Annual percentage growth rate (2001-201 5) _ Agriculture 2,634 4.6 4.4 4.6 4.6 Staples 2,236 3.8 4.1 3.7 3.8 Traditional 155 6.2 4.3 6.7 6.0 Nontraditional 137 10.9 7.5 11.6 10.6 Other 106 4.7 4.6 4.7 4.7 Industry 3,039 3.7 4.0 3.1 3.7 Mining 1,072 0.0 1.4 -3.9 0.0 Manufacturing 1,310 5.3 5.0 5.4 5.3 Other 657 4.8 5.0 4.8 4.8 Services 6,170 3.9 4.3 3.8 4.0 Private 4,966 4.1 4.5 4.0 4.2 Public 1,204 3.0 3.5 2.9 3.1 Total 11,843 4.0 4.2 3.8 4.1 Poverty Headcount (percent) Finalpercentagepoverty headcount (2015) Rural 63.2 55.0 53.9 55.0 54.9 Small-scale farm 65.5 57.0 56.0 57.0 57.0 Medium-scale farm 45.4 36.7 35.7 36.7 36.2 Large-scale farm 3.0 2.0 2.0 2.0 2.0 Nonfarm 48.5 43.O 41.1 43.1 42.5 Urban 43.4 38.9 34.9 39.9 38.6 Total 55.8 49.0 46.8 49.3 48.8 Source: Simulationresults. F. HIPC DebtForgivenessSimulation 6.70 Zambia i s heavily burdenedby debt, much of it accumulated duringthe late 1970s, when world copper prices fell and the government borrowed to alleviate what was seen as only a temporary terms o f trade shock. The servicing o f this debt, which exceeds 150 percent o f total export earnings, has become a substantial burden for Zambia, limiting its ability to reduce poverty and improve living standards. 6.71 The annual interest rate due on the debt i s estimated at 2.8 percent, but the government i s estimated to pay only 1percent. As such, the debt increases at a rate equal to the interest differential. In 2000 Zambia was selected as a HIPC country, entitling it to interim debt relief that, upon completion of the agreed program, would lower the total extemal debt by 69 percent. The actual impact o f debt forgiveness i s assumed to be not the full interest due annually butrather the interest the government has effectively beenpaying inrecent years.47 47Apart from this debt reduction, the base-case scenario remains unchanged, implicitly assuming that Zambia is meeting the different conditions o f the debt-relief program, including implementation o f the 79 6.72 The impact of debt relief on growth and poverty is very small (Table 6-10). The direct effect o f debt reduction i s reduced government debt servicing, limiting the tendency o f the real exchange rate to depreciate. This raises government savings, investment growth, and growth in the capital stock, but the effects are very limited. Investment grows more rapidly, raising import demand (a large share o f capital goods are imported) and counteracting the tendency toward appreciation (see Annex D). Marginally higher growth in the capital stock permits real GDP and household consumption to grow more quickly. The national poverty headcount declines marginally (from 55 percent in the base case to 54.9 percent). All households are slightly better off following debt relief, since they enjoy higher income and consumption levels resulting from higher GDP growth. Ultimately, the impact of debt relief i s positive but small, implyingthat debt relief has a very limited impact on poverty unless accompanied by additional and effective measures aimed at poverty reduction. country's poverty-reduction strategy and progress in fighting HIV/AIDS and implementing educational reforms. 80 7. A SUMMARY OFPOLICY RECOMMENDATIONS 7.1 F o r faster growth a n d diversification, Zambia must do several things right: it must improve the macroeconomic environment (Le., reduce inflation to a single digit, reduce the volatility of the real exchange rate, and reduce real interest rate), the business climate (infrastructure, bureaucracy and red-tape, agency services, and policy dialogue with the private sector), its trade policy and export incentives, governance and accountability, and policy implementation. It must also step up efforts to combat the HIVIAIDS scourge, and better manage the risks arising from both price and weather shocks in.agriculture. Inaddition, Zambia must address the key sectoral constraints discussed inchapters 4 and 5. 7.2 There are several specific policy recommendations arising from the analysis of this report. The mostimportantare summarizedinthe tablebelow: A Policy Agenda for Growth and Diversification Implement soundfiscal policy and management. Control expenditure to reduce domestic debt and Short term domestic borrowing, stop making guaranteed loans to parastatals, commercialize or privatize the remaining major public enterprises, and develop a detailed annual work planto implement Wher public expenditure management reforms, as outlined inZambia Public Expenditure Management and Financial Accountability Review (PEMFAR). Reform thefinancial sector to supportprivate investment and growth. Amend Article 5 o f the Short to BankofZambia Act to give the central bank independenceinimplementing monetary and medium financial sector policies. Develop a cost-effect mechanism to resolve the outstanding debts o f term failed banks and other financial institutions; reform state-owned nonfinancial institutions; improve access to financial services, including rural/agricultural financial services; upgrade human and technological resources available to regulatory and supervisory financial system institutions; and establish a credit referral system to improve the credit culture and move the economy away from a cash- to a credit-based payment system. Improve donor coordination to increase thepredictability of donorfunding. Regularly review conditions attached to donor hnding and monitor policy implementation, designate an institutional entry point for all donors to facilitate donor coordination, and explore local options and strategiesfor exiting external aid should the conditionsbecome unattainable. Maintainpolicy stability. Minimize surprise policy changes by maintaining regular contacts and Short term dialogue between the public sector and the business sector. Clearly define the government's role and social protection activities, particularly inthe agricultural sector. Remove the key barriers tofirms competitiveness. Expandthe roadrehabilitation program, Short to especially inkey productive areas; implementnecessary reforms to improve the performance o f medium public utilities (by applying the lessons learned from commercializing the Zambia Electricity term 81 I OVERARCHING RECOMMENDATIONS SupplyCompany to other state-owned companies, such as Zambia Telecommunications Zompany, for example); improve the regulatory framework for public utilities inorder to reduce the costs o f critical utility inputs; and amendthe labor laws to make it less costly for employers to shed labor. Explore increasingjoint regional trade enhancing investments (for bridge, roads, and railways, for example) through the Southern African Development Community. Remove unnecessary administrative barriers to investment. Simplify license requirements, and Short to rationalize regulatory institutions to improve their effectiveness. medium term Create a trade policy that favors exports rmprove tradepolicy administration. Simplify the export incentive system and improve the Short term timeliness o f duty-drawback schemes, ensure that export processing zones are established and Dperated ina transparent and efficient manner, improve customs administration, take advantage o f the EUand U.S. markets (through the African Growth Opportunity Act) by improving the quality D f goods produced, improve intragovernment policy coordination and public-private partnership intrade policy design and implementation, and improve localcapacity for trade policy analysis andnegotiation. Implement programs that decrease the number of new HIV infections and reduce the Short to socioeconomic impact of HIVIAIDS. Supplement local resources with external support to medium implement behavior-change campaigns, expand the condom distribution system, increase access term to voluntary counseling and testing, strengthen community home-based care, implement programs that reduce mother-to-child transmissions, and expand access to antiretrovirals to all districts. I SECTORAL RECOMMENDATIONS Agriculture and the r u r a l sector ~ Improvefarmers ' access to markets. Improve rural roads, storage, and marketinginfrastructure; Short to energy and communications; access to timely market information; and contract enforcement. medium Help create finctioning commodity exchanges that do not require lengthy licensing to export or term import major commodities, such as maize. Help create sustainable ruraljnancial services. Help buildthe capacity o fprofitable, Medium- creditworthy microfinance institutions; foster the development o f new approaches to lending, such Long term as the warehouse receipt program being piloted inZambia; and extend the proposed credit referral system to agricultural credit. Contain livestock diseases. Make adequate budget provisions for monitoring and controlling Medium major animal diseases, especially inthe Southern, Western, and Eastern provinces. Regulate and term ensure the certification o f private sector veterinarians and service providers. Land Improve access to secure rights to land. Improve access to land of those (few) smallholders who Short term are land constrained by investing in infrastructure and rural services to increase the attractiveness of newly opened settlement areas. Invest in capacity in land administration and dispute adjudicationfor state land before undertaking large-scale titling of land under traditional tenure. Improve the LandTribunal and landadministration to overcome the time-consuming and logistically complicated process required o f investors interested insecuring land inZambia. Improve women s rights to inherit and own land. Regulate customary land tenure systems to Medium ensure women's access to land and dispute adjudication mechanisms. term Fertilizer Issue a clear government policy statement onfertilizer. Where rules have been clear, the private Short t e m 82 Policy recommendation I Timeline I OVERARCHINGRECOMMENDATIONS sector response has been good. Consider completely withdrawing from the market. Reduce the cost of fertilizer. Ensure policy consistency and clarity, improve competition among Medium suppliers, and reduce the cost o f fertilizer transportation by negotiating favorable terms for term handlingfertilizer at shipping ports. Rural roads Improvepolicy andprogram implementation. Define a clear line o f accountability and a clear Medium incentive structure for implementing the rural roads policy. Improve coordination and definition term o f responsibilities between the government, NGOs, and communities. Harmonize the different rulesfor comniunity contributions to and management of ruralfeeder Medium roads, Develop a clear legal framework for road registrationand ownership by communities. term Linkcommunities to the RoadFundfor maintenance andrehabilitation. Increase local authorities' capacity to design and implement rural roadprograms. Upgrade Medium staffing and resource allocation for local authorities to design and implement community roads. to long term Nonagriculture Improve regulation of the gemstone sector. Enforce gemstone sector regulations to minimize Short term smugglingand other associatedvices that currently dominate the sector. Promote market linkages in the gemstone sector. Facilitate establishment o f a fair value Short term marketing system for gemstones. Expedite the establishment o f the Gemstone Exchange, as specified inthe Poverty Reduction Strategy Paper. Improvepolicy andprogram implementation in tourism. Accelerate implementation o f Zambia's Short to soundtourism development strategy by improving environmental mitigationpractices and medium intragovemment policy coordination and collaboration. term Adopt a regional approach to marketing Zambia's tourist attractions. Work with neighboring Short- countries (through the Southern African Development Community) to establish regionaltourism medium circuits. Market Zambia's tourist attractions and products inthe regional context. term Improvephysical infrastructure, public health, and safety at major tourist sites. Improve Medium domestic air transport services to Victoria Falls, road transport services to other tourist sites, and to long health and sanitation services at the main tourist destinations. Monitor and inspect service term standards and quality at the maintourist hotels. 83 REFERENCES Afrobarometer Network. 2002. "Afrobarometer Round I:Compendium o f Comparative Data from a Twelve-Nation Survey." Afrobarometer Paper No. 11. Michigan State University, East Lansing. Alwang, Jeffrey, and Peter Siegel. 2003. "Zambian Household Model Revisited." Draft report for the Social Development Department, December 15, World Bank, Washington, D.C. Barro, Robert. 1991. "Economic Growth in a Cross-section of Countries." Quarterly Journal of Economics 106: 407-444. Barro, Robert J. 1998. "Notes on Growth Accounting." NBER Working Paper 6654. Cambridge, MA: National Bureauof Economic Research. Barro, Robert, and Jong-Wha Lee. 1996. "International Measures of Schooling Years and Schooling Quality.'' Amevican Economic Review 86 (2): 218-23. Baulch, R., and N. McCulloch. 2000. "Tracking Pro-Poor Growth." ID21 Insights No. 31. Institute of Development Studies, Universityof Sussex. Bigsten,A,, J. Levin, and L. Persson. 2001. "Debt Relief and Growth: A Study of Zambia and Tanzania." Paper for WIDER Development Conference on Debt Relief, August 17-1 8, Helsinki. Boughton, Duncan, David Tschirley,,Higinno de Marmle, Afonso Osbrio, and Ballard Zulu. 2002. "Cotton Sector Policies and Performance in Sub-SaharanAfrica: Lessons Behindthe Numbers in Mozambique and Zambia." Flash Report 34E. National Market Information System,'Department of Statistics and the Ministry of Agriculture and Rural Development, Directorate of Economics, Mozambique. Brown, Taylor, and Bennett Siamwiza. 2003. "Chiefs, Commoners, and Enclosures in the Gwembe Valley, Zambia." Mimeo. Canning, D. 1998. "A Database of World Stocks of Infrastructure, 1950-95." World Bank Economic Review 12 (3): 529-47. Cheru, F. 2000. "Debt Relief and Social Investment: Linking the HIPC Initiative to the HIV/AIDS Epidemic in Africa, The Case of Zambia." Review of African Political Economy 86: 519-535. Chinene, Vernon R. N.,Fabian Maimbo, Diana J. Banda, and Stemon C. Msune. 1998. "A Comparison of Customary and Leasehold Tenure: Agriculture and Development in Zambia." Land Reform 2: 89-99. Christie, Iain T., and Doreen E. Crompton. 2001. "Tourism inAfrica." Africa Region Working Paper Series No. 12. World Bank, Washington, D.C. Copestake, J., and P. Weston. 2000. "Pitfalls o f Debt Reduction: A Counterfactual Case Study o f Zambia Duringthe Early 1990s." Journal of International Development 12: 585-600. 84 Coulter, Jonathon, and Gideon Onumah. 2002. "Enhancing Rural Livelihoods through Improved Agricultural Commodity Marketing in Africa: The Role of Warehouse Receipt Systems." Food Policy 27: 3 19-337. Coulter, Jonathon, and Colin Poulton. 2001. "Cereal Marketing in Africa." In Takamasa Akiyama, John Baffes, Donald Larson, Panos Varangis, (eds.), Commodity Market Reforms: Lessonsfrom TwoDecades of Experience. Washington, D.C.: World Bank. Craig, J. 2002. "Twilight on the Zambian Copperbelt?" Review of African Political Economy 29 (92): 364-368. David Neven, Anthony Chapato, and Ballard Zulu. 2001. Michigan State University Development Paper 24. Lansing, MI. Dervis, K., J. de Melo, and S. Robinson. 1982. General Equilibrium Modelsfor Development Policy. New York: Cambridge University Press. Diagne, Aliou. 1998. "Impact of Access to Credit on Income and Food Security in Malawi." InternationalFoodPolicy ResearchInstitute. Discussion Paper 46. Washington, D.C. Dinh,Hinh,Abebe Adugna, and BernardMyers. 2002. "The Impactof CashBudgets on Poverty Reduction inZambia." World Bank Policy ResearchWorking Paper 2914. Washington, D.C. Dollar, D. 1992. "Outward-oriented developing economies really do grow more rapidly: evidence from 95 LDCs, 1976-1985." Economic Development and Cultural Change 40 (3): 523-44. Donovan, Cynthia, M. Damaseke, Jones Govereh, and D. Simumba. 2000. "Framework and Initial Analyses of Fertilizer Profitability in Maize and Cotton in Zambia." Food Security ResearchProject Working Paper 5. Lusaka. Easterly, W., and Ross Levine. 1999. "It's not factor accumulation: stylized facts and growth models." Mimeo. World Bank and Univ. of Minnesota. Easterly, William, and Sergio Rebelo. 1993. "Fiscal policy and economic growth: an empirical investigation." Journal of Monetary Economics 32 (December) 417-458. Evans, D., S. Robinson, and J. Thurlow. 2003. "A Poverty-Oriented Social Accounting Matrix for Zambia for 1998." InternationalFoodPolicy ResearchInstitute, Washington, D.C. Export Board of Zambia. 1999. `"Exporter Audit Report." Lusaka. Export Board o f Zambia. 2000. "Report on the Exporter Audit 1999," Lusaka. . 2001. "Report on the Exporter Audit 2000," Lusaka. . 2002. "2001 Exporter Audit Report," Lusaka. Food and Agriculture Organization and Government of Zambia. 2002. "Zambia's Irrigation Potential, Economic Growth, and the Poverty Alleviation Challenge: Next Steps." Lusaka. 85 Fischer, Stanley. 1993. "The Role of Macroeconomic Factors inGrowth." Journal of Monetary Economics 32: 485-5 11. Fries, Bob. 2001, "Basic Guidelines for Effective Rural Finance Projects." ACDWOCA. Available onthe World Bank website. Gramlich, E. M. 1994. "Infrastructure Investment: A Review Essay." Journal of Economic Literature 32 (3): 1176-96. Govereh, Jones, and Thomas Jayne. 2003. "Cash Cropping and Food Crop Productivity: Synergies or Trade-offs?" Agricultural Economics 28 (1) 39-50. Govereh, Jones, James Nyoro, and Thomas Jayne 1999. "Smallholder Commercialization, Interlinked Markets, and Food Crop Productivity: Cross Country Evidence in Eastern and Southern Africa." Paper presented at the fourth Agricultural TransformationWorkshop, June 27-30, Nairobi. Tegemeo InstituteIEgerton University and the Eastern and Central Africa Programme for Agricultural Policy Analysis. 1 Govereh, Jones, Thomas Jayne, David Tschirley, Cynthia Donovan, Jan-Joost Nijhoff, Michael Weber, and Ballard Zulu. 2000. "Improving Smallholder and Agribusiness Opportunities in Zambia's Cotton Sector: Key Challenges and Options." Food Security Research Project Working Paper 1. Lusaka. Govereh, Jones, Thomas Jayne, Jan-JoostNijhoff, Hyde Haantuba, E. Ngulube, A. Belemu, J. J. Shawa, A. K. Banda, Cynthia Donovan, and Ballard Zulu. 2002. "Developments in Fertilizer Marketing in Zambia: Commercial Trading, Government Programs, and the Smallholder Farmer." Working Paper 4. Zambia Food SecurityResearchProject, Lusaka. Haacker, M. 2002. "The Economics Consequences of HIV/AIDS in Southern Africa." IMF WorkingPaper WPiO2138. Hausner, U. 1999. "A Social Accounting Matrix for Zambia." Trade and Macroeconomics Division Discussion Paper No. 49. International Food Policy Research Institute, Washington, D.C. . 2000. "Structural Adjustment, Agricultural Performance and Income Distribution in Zambia: A Computable General EquilibriumAnalysis." Unpublishedthesis submitted to the Faculty of the Graduate School of the Universityof Minnesota. Howard, Julie, and Catherine Mungoma. 1996. "Zambia's Stop-and-Go Revolution: The Impact of Policies and Organizations on the Development and Spread of Maize Technology." Michigan State University InternationalDevelopmentWorking Paper 61, Lansing, MI. Illovo. 2003. Website: www.il1ovosugar.com. IMF. 1997. SelectedIssues and Statistical Appendix. Washington, D.C. , 1999. SelectedIssues and StatisticalAppendix. Washington, D.C. 86 . 2001 Zambia: Selected Issues and Statistical Appendix. (The Implications of ~ HIV/AIDSfor the Zambian Economy). Washington, D.C. Jayne, Thomas, Jones Grovereh, Mairia Wanzala, and Mulat Demeke. 2003. "Fertilizer Market Development: A Comparative Analysis of Ethiopia, Kenya, and Zambia." Food Policy 28: 293-3 16. Jayne, Thomas, Jones Grovereh, Anthony Mwanaumo, James Nyoro, and Antony Chapoto. 2002. "False Promise or False Premise: The Experience o f Foodand Input MarketReform in Eastern and Southern Africa." WorldDevelopment 30 (11): 1967-1986. Jayne, Thomas, Jones Grovereh, Mairia Wanzala, Mulat Demeke, and James Nyoro. Forthcoming. "Fertilizer Market Reform: Tracing Implementation to Effects in Ethiopia, Kenya, and Zambia." Food Policy, Special Issue on Input Use and Market Development in Sub-SaharanAfrica. Jayne, Thomas, Takashi Yaniano, Michael Weber, David Tshirley, Rui Benfica, David Neven, Antony Chapoto, and Ballard Zulu. 2003. "Smallholder Income and Land Distribution in Africa: Implications for Poverty Reduction Strategies." Food Policy 28: 253-275. Jayne, Thomas, Mulinge Mukumbu,Munhamo Chisvo, David Tschirley, Michael Weber, Ballard Zulu, Robert Johansson, Paula Santos, and David Soroko. 1999. "Successes and Challenges o f Food Market Reform: Experiences from Kenya, Mozambique, Zambia, and Zimbabwe.'' Michigan StateUniversity InternationalDevelopment Working Paper 72. East Lansing, MI. Jimenez, E. 1994. "Human and Physical Infrastructure: Public Investment and Pricing Policies in Developing Countries." Policy and Research Working Paper No. 1281. World Bank, Washington, D.C. Jimenez, Emmanuel. 1995. "Human and Physical Infrastructure: Investment and Pricing Policies in Developing Countries." In Jere Behrman and T. N. Srinivasan, eds., Handbook of Development economics. Vol. 3B. Amsterdam: Elsevier Science. Jones, William. 1995. The World Bank and Irrigation. World Bank Operations Evaluation Study. Washington, D.C. Kaufmann, D., A. Kraay, and M. Mastruzzi. 2003. "Governance Matters 111: Governance Indicatorsfor 1996-2002." World Bank, Washington, D.C. Kaufmann, Kraay, and Zoido-Lobath. 1999. "Governance Matters." World Bank, Washington, D.C. Knack, Stephen, and Philip Keefer. 1995. "Institutions and Economic Performance: Cross- Country Tests Using Alternative Institutional Measures." Economics and Politics 7 (3) :207- 227. Lofgren, H., S. Robinson, and J. Thurlow. 2002. "Macro and Micro Effects o f Recent and Potential Shocks to Copper Mining in Zambia." Trade and Macroeconomics Division Discussion PaperNo. 99. International Food Policy ResearchInstitute, Washington, D.C. 87 Lofgren, Hans, , James Thurlow and Sherman Robinson. 2004. "Prospects for Growth and Poverty Reduction in Zambia, 2001-2015", Commissioned by the World Bank, Washington, D.C. McCulloch, N.,B. Baulch, and M. Cherel-Robson. 2000. "Poverty, Inequality and Growth in Zambia during the 1990s." Working Paper. Institute for Development Studies, University of Sussex. Meagher, Patrick, and Doris Mwiinga. 1999. "Zambian Laws Affecting Microfinance: Review and Recommendations for Reform." IRIS CenterAJniversity of Maryland and Zambia MinistryofLegalAffairs. Milimo, J.T., Participatory Assessment Group, and Civil Society Organizations. 2003. "Participatory Poverty Research" Commissioned by the World Bank. Lusaka. Mwanawina, I.and J. Mulungushi. 2002. "Explaining African Growth Performance: The Case Study of Zambia," Mimeo, Lusaka. Mwiinga, Billy, Jan-Joost Nijhoff, Thomas Jayne, Gelson Tembo, and Jim Shaffer. 2002. "The Role of Mugaiwa in Promoting Household Food Security: Why It Matters Who Gets Access to Government Maize Imports." Food Security ResearchProject Policy Synthesis 5, Lusaka. National Economic Diversification Task Force. 2003. "Integrated Diversification Implementation Plan for the Copperbelt Province and the Livingstone Area." Lusaka. Nehru, V., and A. Dhareshwar. 1993. "A New Database on Physical Capital Stock Sources, Methodology, and Results."Revista de Analisas Econdmica 8 (1) 37-59. ParticipatoryAssessment Group. Undated. "Beneficiary Assessment Phase Iand 111." ParticipatoryAssessment Group. 2001. "Impact Assessment for the MPUAssisted Project." Pillai, V. K.,T. S. Sunil, and R. Gupta. 2003. "AIDS Prevention inZambia: Implications for Social Services." WorldDevelopment 31(1): 149-161. Pletcher, James. 2000. "The Politics of Liberalizing Zambia' Maize Markets." World Development 28 (1): 129-142. Poulton, Colin, Peter Gibbon, Benjamine Hanyani-Mlambo, Jonathon Kydd, Marianne Nylandsted Larsen, Wilbald Maro, Afonso Osbrio, David Tschirley, and Ballard Zulu. 2002. "Competition and Coordination in Liberalized African Cotton Market Systems." Department for InternationalDevelopment (U.K.),Social ScienceResearchUnit. Qian, Ying, and Panos Varangis. 1992. "Does Exchange Rate Volatility HinderExport Growth: Additional Evidence." World Bank Policy Research Working Paper No. 911. Washington, D.C. Rakner, L., N. van de Walle, and D. Mulaisho. 1999. "The Politics o f Economic Reform Implementation inZambia." World Bank, Washington D.C. Ravallion, M. 2003. "Pro-Poor Growth: A Primer." World Bank, Washington, D.C. 88 Republic of Zambia. 1997. The Evolution of PovelCty in Zambia 1991-1996. Central Statistics Office, Lusaka, Zambia. Saasa, Oliver S., Dennis Chiwele, Foustin Mwape, and John C. Keyser. 1999. "Comparative Economic Advantage of Alternative Agricultural Production Activities in Zambia." USAID Technical Paper 104. US. Agency for InternationalDevelopment, Washington, D.C. Scott, G. 2002. "Zambia: Structural Adjustment, Rural Livelihoods and Sustainable Development." Development Southern Africa 19 (3): 405-418. Semen, Luis. 2002. "Real Exchange Rate Uncertainty And Private Investment In Developing Countries." World Bank Policy ResearchWorking Paper 2823, Washington, D.C. Skjonsberg, E. 2003. "A Report on 25 Years of Rural Change inKefaVillage, Eastern Province, Zambia." Consultant Report prepared for Social Development Department. The World Bank: Washington, D.C. mimeo. Smith, Robert. 2001. "Land Tenure, Title Deeds, and Farm Productivity in the Southern Province of Zambia: Preliminary Research Findings." Paper presented at the University of Zambia, September 19. Smith,Robert. Forthcoming. "Land Tenure, FixedInvestment, andFarmProductivity: Evidence from Zambia's Southern Province." WorldDevelopment. Subramanian, Jyoti. 1998. In John Bruce, (ed.), "Country Profiles of Land Tenure: Africa, 1996." ResearchPaper 130, LandTenure Center, Universityo fWisconsin. Transparency International. 2002. "Corruption Perceptions Indexandadditional background data," Available at website: www.transparency.org or www.uni-goettingen.de/-uwvw. Tshirley, David, James Shaffer, and Ballard Zulu. 2004. "Cotton in Zambia: An Assessment of its Organization, Performance, Current Policy Initiatives, and Challenges for the Future," Food Security ResearchProject Worlung Paper 10, Lusaka. Tyler Associates. 2002. "Review of Zambia Gemstone Sector andDevelopment Strategy." Draft report prepared for the World Bank. Washington, D.C. UNAIDS. 2002. Report on the GlobalHIV/AIDSEpidemic 2002. Geneva. UNICEF, 2003. Zambia Country Situation Report: July-August 2002. United Nations International Children's Emergency Fund, New York. West, T. 1992. "The Politics of the Implementation of Structural Adjustment inZambia, 1985- 1987." In The Politics of Economic Reform in Sub-Saharan Africa. Final report preparedby theCenter for Strategic andInternational Studies, Washington, D.C. Wichern, Rainer, Ulrich Hausner, and Dennis K. Chiwele. 1999. "Impediments to Agricultural Growth in Zambia." Trade and Macroeconomics Discussion Paper 47. International Food Policy ResearchInstitute, Washington, D.C. 89 World Bank. 1984. "Zambia Country Economic Memorandum: Issues and Options for Economic Diversification." Washington, D.C. , 1993. "Zambia CEM: Prospects for sustainable and equitable growth." Washington, D.C. , 1994. Zambia Poverty Assessment. Report 12985-ZA. Washington, D.C. . 1996. "Zambia: Prospects for Sustainable Growth." World Bank, Economic Management and Social Policy Group, Washington D.C., . 1997. "Second Coffee Project, Implementation Completion Report." Report 16810. Washington, D.C. . 1998. "Fiscal Management Report." No. 18552, AFTP1. Washington, D.C. . 2000. "The WorldBusiness Environment Survey (WBES) 2000." Washington, D.C. , 2001. Zambia: Public Expenditure Review. ReportNo. 22543-212. Washington, D.C. . 2002a. Country Procurement Assessment Review. AFTQK. Washington, D.C. . 2002b. Looking Beyond Copper. Final Report of the World Bank Task Force on the Zambia Copper Crisis, World Bank, Washington, D.C. , 2002c. "Zambia Country Assistance Evaluation." Operations Evaluation Department, Washington, D.C. . 2002d. "Zambia Privatization Review: Facts, Assessments and Lessons." Report preparedat the request of the Ministerof Finance and National Planning, Lusaka, December 2. . 2003a. The Challenge of Competitiveness and Diversijkation. Report 25388-ZA. Washington, D.C. , 2003b. Public Expenditure Management and Financial Accountability Review. PEMFAR Report No. 26262-ZA. Washington, D.C. . 2003c. WorldDevelopment Indicators. Washington, D.C. . 2003d.Extended Revised Minimum Standard Model -(MSM-g. Washington, D.C. . 2004a. "Institutional Analysisof Fertilizer." Social Development Department, World Bank, Washington, D.C. , 2004b. "Zambia Administrative Barriers to Private Investment." World Bank, FIAS, Washington, D.C. , 2004c. "Zambia Investment Climate Assessment." Private Sector Development Unit, Washington, D.C. 90 . 2004d (forthcoming). "Zambia Poverty and Social Impact Assessment of Rural Reforms", World Bank, Washington, D.C> World Tourism Organization. 2003, Compendium of Tourism Statistics, 2003 Edition. World Trade Organization. 2002. TradePolicy Review: Zambia. Geneva. Zambia, Central Statistics Office. 2000. "The 1998 Living Conditions Monitoring Survey", Lusaka. Zambia, Central Statistics Office. 2001. "Agricultural and Pastoral Production, Small and MediumScaleHoldings, 1999/2000." Lusaka. Zambia, Central Statistics Office. 2002. "National Accounts Statistics." Lusaka. Zambia, Government of. 2000. "Agricultural Statistics Bulletin, 1999/2000," Ministry of Agriculture and Cooperatives, Lusaka. Zambia, Government of. 2002a. Poverty Reduction Strategy Paper. Lusaka. ,2002b. Public Sector Reform Programme. Lusaka. Zambia, Ministry of Agriculture, Food, and Fisheries. 2000. "Agricultural Statistics Bulletin, 1999/2000." Lusaka. Zambia, National Tourism Board. 1998. TheDigest of Tourism Statistics 1997-2001. Lusaka. Zeller, Manfred, and Manohar Sharma. 2000. "Many Borrow, More Save, and All Insure: Implications for Foodand Microfinance Policy," Food Policy 25: 143-167. Zeller, Manfred; and Richard Meyer. 2002. The Triangle of Microfinance: Financial Sustainability, Outreach, and Impact. Baltimore: JohnHopkins University Press.