Report No. 47969-MW Malawi Country Economic Memorandum Seizing Opportunities for Growth through Regional Integration and Trade Volume I: Summary of Main Finding and Recommendations Poverty Reduction and Economic Management 1 Africa Region Document of the World Bank Table of Contents Executive Summary ........................................................................................................... i A. Introduction ............................................................................................................. i B. Setting the stage ...................................................................................................... i C. Key messages ......................................................................................................... ii 1. Malawi's Growth History and Characteristics ....................................................... 1 A. Growth history ........................................................................................................1 B. What has driven growth? ........................................................................................3 C. The pattern of growth ............................................................................................10 D. The spatial dimension of growth ...........................................................................16 E. Observations regarding growth in Malawi ...........................................................18 2. Opportunities in Agriculture and Agribusiness ................................................... 23 A. Introduction ...........................................................................................................23 B. Changing context and prospects ...........................................................................26 C. The role of regional integration ............................................................................28 D. Lessons from agricultural value chains ................................................................29 E. Agribusiness ..........................................................................................................35 F. Maize, input subsidies, and food security .............................................................39 3. Improving Connectivity and Energy Provision .................................................... 42 A. Improving transport and trade logistics ...............................................................42 B. Improving connectivity through the spatial dimension of growth ........................50 C. Improving energy provision ..................................................................................55 4. The Supportive Macroeconomic Framework ....................................................... 63 A. Introduction ...........................................................................................................63 B. Key macroeconomic policy priorities to sustain and broaden growth .................64 References ........................................................................................................................ 76 List of Tables Table 1.1: GDP and sources of growth ............................................................................... 4 Table 1.2: Export shares to regional trading partners by product (2000-07) .................... 10 Table 1.3: The diffusion of growth out of agriculture ...................................................... 11 Table 1.4: Composition of expenditures ........................................................................... 16 Table 1.5: Cash crop production (%) by agronomic potential and travel time ................. 17 Table 1.6: Industrial structure in Blantyre and Lilongwe ................................................. 18 Table 2.1: Maize, rice, cotton, and tobacco parity price comparison ............................... 30 -3- Table 2.2: Comparison of input price build-up between Malawi and other countries ..... 32 Table 3.1: Distance to the ports and estimated trade volumes .......................................... 42 Table 3.2: Reliability of ports and corridor transit times .................................................. 43 Table 3.3: Transport and logistics cost of major exports, 2004 ........................................ 45 Table 3.4: Need for periodic and routine maintenance of Malawi's road network ........... 46 Table 3.5: Regional trade facilitation initiatives ............................................................... 47 Table 3.6: Physical road conditions by road class ............................................................ 51 Table 3.7: Determinants of farmer's tobacco transport costs ........................................... 52 Table 3.8: Cash crop production by travel time and agronomic potential ........................ 52 Table 3.9: Trucking survey - summary of main findings ................................................. 53 Table 3.10: Determinants of unit transport costs from trucking survey ........................... 54 Table 3.11 ESCOM's existing generation facilities ......................................................... 56 Table 4.1: Key recent macroeconomic developments ...................................................... 64 List of Figures Figure 1.1: Phases of GDP per capita in Malawi 1960-2008 ............................................. 3 Figure 1.2: Per capita maize harvest and GDP ................................................................... 5 Figure 1.3: Exports in constant MK and growth (1960-2007)............................................ 6 Figure 1.4: Exports in real dollars, real kwacha and the real exchange rate....................... 7 Figure 1.5: Real exports per capita and export volume ...................................................... 8 Figure 1.6: Tobacco price and production .......................................................................... 8 Figure 1.7: Malawi's regional exports ................................................................................ 9 Figure 1.8: Macro-environment and investment ............................................................... 11 Figure 1.9: Manufacturing versus imports ........................................................................ 12 Figure 1.10: Import composition and the external balance............................................... 14 Figure 1.11: Effects of aid on growth and competitiveness.............................................. 15 Figure 1.12: Tobacco production intensity and agro-climatic conditions ........................ 17 Figure 2.1: Real agriculture growth by farming sector (const 94 prices) ......................... 24 Figure 2.2: Area under cultivation by smallholders.......................................................... 24 Figure 2.3: Nominal FAO food price index ...................................................................... 26 Figure 2.4: Food commodity price indices ....................................................................... 26 Figure 2.5: Maize, build-up of financial costs along the value chain ............................... 34 Figure 3.1: Transit from the port of Beira to Malawi ....................................................... 43 Figure 3.2: Road quality and traffic volume ..................................................................... 50 Figure 3.3: Projection of peak demand and generating capacity ..................................... 55 Figure 3.4: Sales losses due to power outages and firms with generators ....................... 57 Figure 4.1: Malawi's progress towards a stable macroeconomic environment ................ 65 Figure 4.2: Malawi's lingering macroeconomic challenges.............................................. 68 Figure 4.3: Malawi's fiscal developments......................................................................... 72 -4- List of Boxes Box 1.1: HIV/AIDS: an obstacle or constraint to growth? ................................................ 2 Box 1.2: Is growth constrained by education? ................................................................. 13 Box 1.3: Additional constraints identified by the growth diagnostic .............................. 20 Box 1.4: Growth, gender, and per capita income ............................................................ 21 Box 2.1: Diversification of smallholder agriculture ........................................................ 25 Box 2.2: Are the rising food prices here to stay? ............................................................. 28 Box 2.3: Summary of the main uses and methodology of value chain analysis .............. 31 Box 2.4: Three ways to improve farm-level returns: a simulation .................................. 33 Box 2.5: General constraints to growth of the agribusiness sector .................................. 38 Box 4.1: Financial services in Malawi ............................................................................. 67 Box 4.2: Two illustrative scenarios ................................................................................. 71 -5- ABBREVIATIONS AND ACRONYMS ACE Agricultural Commodity Exchange ADMARC Agriculture Development and Marketing Corporation ADT Average daily Traffic AGOA African Growth and Opportunity Act ADP Agriculture Development Programme AFDB Africa Development Bank AICD Africa Infrastructure Diagnostic AIDS Acquired Immune Deficiency Syndrome BAT British American Tobacco BOP Balance Of Payments CAADP Comprehensive African Agriculture Development Program CAB Current Account Balance CAN Calcium Ammonium Nitrate CEAR Central East African Railways Company CFL Compact Fluorescent Light CIAT International Center for Tropical Agriculture CIF Cost, Insurance and Freight COMESA Common Market for Eastern and Southern Africa DAC Development Assistance Committee DBI Doing Business Indicators DFID Department for International Development DSM Demand Side Management DVA Domestic value Addition ECA Eastern Africa Community ECF Emerging Commercial Farmer ESCOM Electricity Supply Corporation Of Malawi EU European Union FAM Family Run Farm FAO Food and Agricultural Organization FOB Free On Board FSAP Financial Sector Assessment Program FY Fiscal Year GDP Gross Domestic Product GIS Geographic Information Systems GOM Government of Malawi GOT Ginning Out Turn GTZ Gesellschaft für Technische Zusammenarbeit (German) HH HouseHold HIPC Highly Indebted Poor Country HIV Human Immunodeficiency Virus HDM Highway Development and Maintenance Model HRV Hausmann, Rodrik, and Velasco ICA Investment Climate Assessment ICRISAT International Crops Research Institute for the Semi Arid Tropics IDA International Development Agency IFPRI International Food Policy Research Institute -6- IFS International Financial Statistics IPPs Independent Power Producers ISP Infrastructure Services programme IHS Integrated Household Survey IITA International Institute of Tropical Agriculture IMF International Monetary Fund IMPACT International Model for Policy Analysis of Agricultural Commodities and Trade IMT Intermediate Means of Transport ISI Import Substituting Industrialization Kg Kilogram kWh kiloWatt hour LCF Large Commercial Farmer MACE Malawi Agricultural Commodity Exchange MAREP Malawi's Rural Electrification Programme MBS Malawi Bureau of Standards MCA Millennium Challenge Account MCC Millennium Challenge Corporation MDRI Multilateral Debt Relief Initiative MGDS Malawi Growth and Development Strategy MK Malawi Kwacha MOF Ministry of Finance MT Metric Ton MW Mega Watt NPK Nitrogen, Phosphorous, Potassium NSO National Statistical Office OECD Organisation for Economic Co-operation and Development OLS Ordinary Least Squares PA Period Average PC Per Capita PER Public Expenditure Review PPAs Power Purchase Agreements PSRS Power Sector Reform Strategy PVA Poverty and Vulnerability Assessment R2 Regression Coefficient R&D Research and Development RER Real Exchange Rate RSA Republic of South Africa SACMEQ South African Consortium for Monitoring Educational Quality SADC Southern Africa Development Community SANRAL South African National Road Agency Ltd SARNET Southern Africa Regional Research Network SDR Special Drawing Right SPAM Spatial Production Allocation Model SSA Sub-Saharan Africa SV Shipment Value T&D Transmission and Distribution TEU Twenty-foot Equivalent Unit TIR International Road Transit (Transports Internationaux Routiers. French) -7- TOU Time Of Use UN United Nations US United States USD United States Dollar VAT Value Added Tax WDI World Development Indicators WTO World Trade Organization -8- ACKNOWLEDGEMENTS This report was prepared through a unique collaboration among various development partners. At the outset the Millennium Challenge Corporation (MCC), the African Development Bank (AFDB), the UK Department for International Development (DFID) and the World Bank (IDA) decided to pull intellectual and financial resources to analyze Malawi's growth opportunities with the aim to find those core binding constraints that alleviated could sustain and broaden the impact of Malawi's current growth momentum. Although the overall team was led by Jos Verbeek (Lead economist, IDA), each institution was represented in the core team with a coordinator. For the Millennium Challenge Account-Malawi this was Alex Gomani (Chief economist), for AFDB, this was Martha Phiri (Country economist), and DFID was represented in the core team by David Woolnough (Growth Team Leader, DFID Malawi). In addition the team included Nick Lea (Consultant, IDA & DFID), Lucia Hanmer (Growth advisor, DFID), Hans Binswanger-Mkhize (Consultant, IDA), John Keyser (Consultant, IDA), Hardwick Tchale (Agriculture economist, AFTAR), David Rorhbach (Senior agriculture economist, AFTAR), Ron Kopicki (Consultant, IDA), Jason Agar (Consultant IDA & DFID), Alexander Culiuc (Consultant, MCC), Manu Manthri (Economist, DFID-Malawi), Oliver Knight (Energy advisor, DFID), Don Webster (Consultant, DFID), Jean Francois Arvis (Senior transport economist, PRMTR), Patricia Macchi (Junior professional associate, AFTTR), Joao Mabombo (Infrastructure specialist, AFDB), Benson Nkhoma (Infrastructure specialist, AFDB), Khwima Nthara (Senior country economist, AFTP1), Somik Lall (Senior economist, FEU), Mark Pearson (Consultant, DFID), Hyoung Gun Wang (Economist, FEU), Thomas Munthali (ETC, AFTP1), Rob Mills (Economist, AFTEG), Susan Banda (Public outreach coordinator, MCA-Malawi), Zeria Banda (Communications officer, AFREX), Vera Ng'oma (Communications officer, DFID-Malawi), Dotilda Sidibe (Program Assistant, AFTP1), Pauline Kayuni (Program assistant, AFMMW), Victoria Barr (Consultant, AFMMW) and Sara Sundaram (Program assistant, AFTP1). The team received guidance from the heads of mission of each of the institutions involved i.e. Nick Dyer and Gwen Hines of DFID, Frank Kufakwandi of AFDB, Deidra Fair and Sam Kakhobwe of MCC and MCA-M respectively, and Tim Gilbo of IDA. Comments received at various stages by John Panzer (Sector Manager, AFTP1), Sudhir Shetty (Director AFTPM), Tony Venables (Chief Economist, DFID at the initiation) and Alain Winter (Chief Economist, DFID at completion) are gratefully acknowledged and have helped to enrich the report. In preparing the report, Nick Lea in close collaboration with Lucia Hanmer was responsible for the growth analysis drawing on earlier work done by Alexander Culiuc. Hans Binswanger-Mkhize in collaboration with Hardwick Tchale, John Keyser and David Rorhbach were responsible for the analysis of the agricultural sector, while Ron -9- Kopicki and Jason Agar together led the work on agribusiness. Oliver Knight, Rob Mills, Don Webster, and David Woolnough analyzed the energy sector and J-F Arvis, Patricia Macchi, Mark Pearson, Benson Nkhoma and Joao Mabomba were responsible for the discussion of transport and trade facilitation. Somik Lall, Hyoung Wang and Thomas Munthali prepared the spatial contribution to the CEM and Jos Verbeek drafted the macroeconomic section of the report. Invaluable administrative support was received from Pauline Kayuni, Dotilda Sidibe and Sara Sundaram. Vera Ng'oma and Zeria Banda prepared a detailed dissemination strategy. The internal peer reviewers for the report at concept stage were Richard Newfarmer (Special Representative to the WTO and UN, Geneva, EXTGE) and Eduardo Ley (Lead economist, PRMED), while Eduardo Ley was joined by Steven Jaffee (Lead economist, ARD) to review the final document. The team is also grateful to the support it has received from the various agencies of the Government of Malawi. The team would like to thank in particularly Mr. Perks Ligoya, Director, Ministry of Finance, Mr. Ted Sitima-wina, Secretary for Economic Planning and Development, Ministry of Economic Planning and Development, and Mr. MacDonald Mwale, Principle economist, Reserve Bank of Malawi for their feedback and overall assistance. The CEM benefited enormously from the comments and support provided by many stakeholders in Malawi. The initial results were discussed in Malawi in December 2008 and at several consultations, which were organized in Malawi in early 2009 to further discuss initial findings. The final draft version of the report was disseminated at various events across Malawi. These events were attended by Government officials, members of the media, representatives of civil society, development partners, private sector, and staff of the University of Malawi. In January 2010, a final consultation was hosted by the Honourable Minister of Finance and at this event the report was discussed with members of Malawi's Cabinet. The team is very appreciative for being given this opportunity. - 10 - Malawi Country Economic Memorandum EXECUTIVE SUMMARY A. INTRODUCTION i. Malawi needs to focus on exports to maintain and broaden its current inspiring levels of economic growth. The focus of future policy should therefore be on reforms that improve competitiveness in global and regional markets. This does not require a fundamental shift in direction, but instead a rebalancing of policy and expenditures to support an outward-oriented development framework. ii. Until the recent global financial crisis, domestic and regional trends were promising for Malawi: rapid economic growth, strong donor support, increases in foreign direct investment, real potential to upgrade regional infrastructure, and the gradual dismantling of barriers within the region through the COMESA and SADC free trade areas. Despite recent global developments, this report makes the case that, for a small landlocked economy like Malawi, integration into a dynamic and more open Southern Africa region will remain key to building prosperity and improving livelihoods. B. SETTING THE STAGE What drives growth? iii. Malawi's growth is export-led. Despite changes in the structure of agricultural production (from estate to smallholder farming and the liberalization of prices and finance), a longstanding relationship exists between exports and overall GDP. More specifically, there is a strong link between GDP and export revenues expressed in real domestic currency. A central driver of the economy has been the positive multiplier effect of exports, which creates demand for local services and, to a limited extent, domestic manufacturing. iv. The real exchange rate determines the strength of the multiplier effect from exports, by translating US dollar export revenues into local purchasing power. It is important for Malawi to adopt more flexibility in the management of its currency. An appropriately valued currency would narrow the gap between official and parallel exchange rates and curb chronic shortages of foreign exchange. Ensuring that the currency is not over-valued would also enhance export diversification, and improve the response from import -substituting sectors to strong domestic demand. v. While maize consumption is central to the welfare of Malawi's population, maize has contributed only additively to GDP and has been a source of economic volatility, due to its dependence on rainfall. The link between maize and economic growth is weak, -i- largely because the low proportion of maize production sold limits multiplier or spill-over effects. vi. The export of non-traditional products to regional markets is becoming an increasingly important source of export growth. Progress in regional cooperation through SADC, COMESA and bilaterally with South Africa and Mozambique will be key to expanding market access and the flow of inward investment to support diversified exports and economic growth. C. KEY MESSAGES Agriculture vii. A dynamic smallholder farming sector has driven growth in agriculture. Smallholders have responded impressively when offered favourable opportunities and prices. Higher international agricultural prices and relatively strong regional growth are creating new opportunities for commercial crops in Malawi, even though these prospects are clouded by the global economic crisis in the short term. However, if as many expect, world food prices eventually rebound to levels that exceed those of the 1990s, agriculture can continue to be a significant driver of growth, through regional exports as well as import substitution. viii. Regional cooperation in agriculture is central to Malawi's market access, product quality and technological capability. Malawi depends on regional collaboration to upgrade infrastructure to regional and international markets; the harmonization of standards and sanitary measures; protection against plant and animal disease epidemics; and the generation of new technology. ix. The potential for regional trade is threatened by the risk of government intervention, and differences in rules of origin, quality and product standards. Trade in food staples has long been discouraged by national policies that place a high priority on food self-sufficiency. In 2008, Malawi banned all exports of maize, and other SADC and COMESA countries have also intervened in these markets. Consequently, one of the biggest impediments to large-scale investment in regional trading capability remains the unpredictable behaviour of governments, and their tendency to impose export bans whenever they fear shortages in their own markets. x. Recent macroeconomic and agricultural policy decisions, however, have risked deteriorating agricultural incentives in Malawi. These include the lack of flexibility in the exchange rate leading to episodes of overvaluation, as well as continuous and often unanticipated changes in major policies, such as restrictions on private trade in maize and the setting of minimum prices for several crops. Such policy decisions put recent gains at risk and potentially undermine investment in productivity improvements. - ii - xi. Malawi's competitiveness is undermined by low productivity, high cost of inputs, high cost of transport, and high costs along the value chain. Inorganic fertilizer and other agricultural inputs are expensive mainly due to high international and domestic transport costs, as well as high trader margins resulting from transaction risks associated with agricultural input trading in Malawi. The high cost of inputs further leads to sub-optimal uptake of fertilizer and improved seed. Overtime, low uptake of improved technology results in under-capitalization in the sector, which tends to impede further technological and institutional innovation. xii. Much of Malawi's competitive advantage in agribusiness derives from the regulatory regimes and market institutions which have developed for specific subsectors. However, in some subsectors, the regulatory environment has deteriorated and the government has intervened arbitrarily, imposing price and trade controls. A new legislative compact is needed that commits the government to rule-based regulation that is systematic, predicable and based on facts and market intelligence. The key objective should be just enough regulation to insure that each sector is able to respond quickly to competitive challenges. xiii. Food security has improved in Malawi due to the increased use of fertilizer and improved seed made possible by the input subsidy program. However, there is now a need to rebalance agricultural expenditures to include programs supporting crop diversification, small-stock production, commodity risk management and other productivity enhancing investments. Expenditure on the fertilizer and seed program should be brought under control by better targeting toward food-poor households and by more timely procurement. xiv. To increase the income of farmers, policy should move away from unilateral minimum price management toward productivity promotion. Maize prices would be better stabilized through the consistent application of market risk management strategies, than through internal trade and price controls. Agricultural productivity could be improved through a targeted reduction of fuel levies, a reduction in the cost of finance, improved information about credit history, and the linking of agricultural credit with weather insurance. Trade Logistics xv. Malawi's geography is a challenge, given its distance from markets including gateway ports in Mozambique and South Africa. Current transportation costs measured as unit costs per kilometre travelled are not especially high. However, distances are relatively large and the total cost of export transport to final destination probably amounts to close to 15 percent of trade value, while the cost of import transport is approximately 12.5 percent of the value of imports. - iii - xvi. While there is probably limited scope to alter the fundamentals which determine international transport costs, the reliability of trade transport is a constraint to growth which could be influenced at a policy level and should receive more attention. Malawian trade suffers extensive delays at ports (with the exception of Durban) and is subject to a high number of procedures and transit arrangements creating fragmented supply chains. xvii. Policy makers and donors have over-emphasized road infrastructure at the expense of concession efficiency, trade facilitation and regional cooperation. The focus of policy should shift towards the maintenance (roads) and rehabilitation (rail) of current infrastructure, the creation of a single regional transit regime, improvements in regional trade facilitation, and the promotion through regional organizations of increased operational efficiency of ports. xviii. Regional trade facilitation could be improved through the establishment of a regional door-to-door transit regime. This could be achieved through the creation of a consistent SADC/COMESA transit regime in the form of regional carnet-bond system, and the harmonization of road charges within SADC. While Mozambique may not be in a position to join such a system immediately, practical improvements could be negotiated between Malawi, Mozambique, and Zambia for the Beira and Nacala corridors. xix. Domestic trade facilitation could be made more efficient by allowing custom declarations to be filled in and processed online; introducing an authorized operator regime for large importers; introducing tighter regulation of customs brokers; and faster and more transparent issue of trade licenses. Spatial Dimension of Growth xx. Malawi's manufacturing industry is concentrated in Blantyre and Lilongwe. Industry in Blantyre is more high-tech and more externally orientated. Blantyre is physically closer to international markets and firms there specialize in activities benefiting from access to international markets. Firms in Lilongwe, which is located in the agriculture belt, are more integrated into domestic markets through agro-processing. Consequently, reliable power provision, international transport and trade facilitation are relatively more important to firms located near Blantyre than firms located near Lilongwe. It is important to recognize these differences when prioritizing public investments in power, transport and trade logistics. xxi. Spatial analysis also reveals that the production of cash crops is concentrated in areas of good agronomic potential with relatively low transport times to major cities. While Malawi's current transport strategy focuses on tarmac roads, a higher economic return could be gained from prioritizing feeder roads in areas of high agronomic potential, because the quality of feeder roads has a significant impact on farmers' transport costs. - iv - xxii. The domestic trucking industry is artificially protected from international competition on domestic routes, which contributes to significantly higher domestic transport costs and lower farm-gate prices. While international fleets would not be able to enter rural sections of the market because of load sizes and road quality, domestic routes should still be opened to international competition. Energy xxiii. The undersupply and poor quality of electricity in Malawi is a binding constraint on economic growth, and energy generation is near totally dependent on the Shire river. The government needs to invest urgently in additional generating capacity. xxiv. The regulatory framework of the industry requires clarification and reform. Electricity tariffs need to be raised substantially to fund additional investment, and the commercial framework needs to provide sufficient incentives to attract long-term investment from independent power producers. xxv. The industry should investigate ways of prioritizing uninterrupted supply to growth-critical industries, and of creating a tariff which reflects the true cost of prioritization. Spatial analysis indicates that the highest demand for reliable electricity is most likely around Blantyre, where most manufacturing and export-oriented firms are located. Macroeconomic management xxvi. Recent macroeconomic policy, in particular fiscal discipline, has played a significant role in building Malawi's growth momentum and improving the country's growth prospects. However, the domestic supply response has not been sufficiently large for growth of GDP to match the growth of aggregate demand. Macroeconomic management should therefore focus on supporting economic growth. xxvii. In the short run, macroeconomic policy should seek to balance GDP growth and growth in aggregate demand, partly through a reduction in public consumptive expenditures and an adjustment in the exchange rate. This would curb demand for imports of consumables and stimulate exports. In addition, the accumulation of foreign reserves should be a priority, and a re-evaluation of the use of foreign aid to this end should be considered. xxviii. To increase Malawi's productive capacity, strong macroeconomic management will need to be accompanied by public investment to alleviate supply side constraints identified in the areas of energy supply, and (cross-border) transport and trade facilitation. Public expenditure should therefore be focused on infrastructural programs that improve productivity. At the same time, growth of overall public expenditure should be contained, as the size of government has grown large, relative to comparator countries. -v- - vi - 1. MALAWI'S GROWTH HISTORY AND CHARACTERISTICS Growth is not an end in itself. But it makes it possible to achieve other important objectives of individuals and societies. It can spare people en masse from poverty and drudgery. Nothing else ever has. It also creates the resources to support health care, education, and the other Millennium Development Goals to which the world has committed itself. In short, we take the view that growth is a necessary, if not sufficient, condition for broader development, enlarging the scope for individuals to be productive and creative." The Growth Report, Commission on growth and development, 2008 A. GROWTH HISTORY 1.1 Malawi is a landlocked, densely populated country in Southern Africa where nominal annual income currently stands at around US$290 (WDI, World Bank 2008a). The population is highly rural; agricultural production is predominately subsistence and rain fed; and due to having a single rainfall season and limited irrigation, the country experienced food shortages during 1994, 2002 and 2005. Malawi has been severely affected by the HIV/AIDS pandemic1: it has ravaged families, decimated the labour force, and left over 550 thousand children ­ out of a total population of 14 million - orphaned. Furthermore, unlike many of its neighbors, Malawi has not benefited from significant mineral endowments, and its export corridors to the ports in Mozambique have yet to be fully repaired from the damage done by that country's civil war. 1.2 Growth in Malawi is currently strong and historically can be characterized by four distinct phases. The economy grew by over 6 percent in 2006, by over 8 percent in 2007 and is projected to sustain this level of growth in 2008 and 2009. While recent performance is excellent relative to the current global context, the history of output has been volatile and can be classified into four phases: (a) 1960-79 estate-based growth; (b) 1979-89 decline; (c) 1989-03 stagnation due to shocks and transition to smallholder led growth; and (d) 2004-09 recovery. 1.3 1960-79: Strong growth was founded on export agriculture from estates, with smallholders producing food and supplying labour. During this period the government supported large-scale agriculture through preferential access to land, investment and credit. Estates grew at an average annual rate of 17 percent while smallholder production grew at 3 percent, while smallholder income was supplemented by remittances from migrant labour. The condition of transport infrastructure was relatively good and 95 percent of Malawi's exports were routed via the Mozambican ports of Nacala and Beira. Estate-led growth was made possible by relatively high product prices, the efficient value chain of estate marketing, good transport infrastructure and cheap credit as interest rates were kept low. 1 Malawi's prevalence of the disease in 2007 was estimated at 11.9 percent, the ninth highest in the world. See box 1.1 for a further exposition. -1- Box 1.1: HIV/AIDS: an obstacle or constraint to growth? There can be no doubt of the devastating effect of HIV/AIDS on Malawian society and economy. Malawi's prevalence of the disease in 2007 was estimated at 11.9 percent - the ninth highest in the World (see table 1). It has ravaged families, decimated the labour force, and left over 550 thousand children ­ out of a total population of 14 million - orphaned. How does HIV/AIDS affect growth? Primarily through its attrition of the quality and quantity of the labour force ­ by reducing the number of people employed, by reducing the total pool of skills, and through the increased sickness of those in the workplace as the disease progresses. Secondly through its effect on capital accumulation which is negatively affected by the effect on private savings (due to lower income from the labour force) and a reduction in public investment due to the redirection of government expenditures towards care and treatment. In addition to the above argument, Bell, Deverajan, and Hans Gersbach (2003) emphasize the importance of human capital and transmission mechanism across a generation, which in turn amplifies the devastating impact of HIV/AIDS on growth. It is clear that HIV/AIDS is an obstacle to growth, but is it also a binding constraint? For the disease to be a constraint, an increase in health expenditure should provoke a strong growth response. In the case of Malawi this is unlikely since expenditure on HIV/AIDS is already a substantial proportion of the health spending and the availability of anti-retroviral therapy (ART) compares favourably with higher income peer countries (see table below) such as Angola, Lesotho, Mozambique and South Africa. However, various studies (Canning 2006, Easterly 2006) have demonstrated that the economic return of interventions which prevent new cases of HIV are usually far higher than those of treatment alone. Although to an extent these go hand-in-hand, since the roll-out of treatment programs also creates awareness within a community, the majority of HIV/AIDS expenditure in Malawi has focused on treatment rather than prevention. The persistent high numbers of new cases in Malawi confirms this and implies that changes in high risk behaviour are only being adopted slowly, implying that a higher economic return could be achieved by channelling more resources into prevention programs. Table 1: A cross country comparison Malawi Mozam South (2007) Angola Botswana DRC Congo Lesotho bique Namibia Swaziland Tanzania Uganda Africa Zambia Zimbabwe Adult prevelance rate 11.9 2.1 23.9 1.4 3.5 23.2 12.5 15.3 26.1 6.2 5.4 18.1 15.2 15.3 Health service and care Indicators ARV Sites 163 56 99 209 28 110 211 57 22 204 286 362 322 89 ARV therapy coverage 35 25 79 17 26 24 88 42 28 46 19 Knowledge and behavior a condom (2004) Male 20 16 29.8 40.5 19 74.4 56.2 28.5 41.5 28 36.3 Female 15.9 7.7 22.9 18.7 4.9 65.7 56.6 20.5 41.1 33.1 40.8 Prevention HIV who receive ARV 32 9 > 95 9 5 32 46 64 67 32 34 57 47 29 Life expectancy 50 41 52 47 54 42 50 61 42 50 50 51 43 43 Source: Country Epidemiological Fact Sheet on HIV and AIDS by UNAIDS, WHO, and UNICEF. 1.4 1979-89: Incomes strongly declined following the oil price shock accompanied by severe deterioration in the terms-of-trade. Malawi's terms-of-trade collapsed by 25 percent at the end of the 1970s beginning a long deteriorating trend. The worsening of export prices reduced demand in South Africa for migrant labor, consequently reducing household remittance incomes in Malawi. Furthermore, civil war in Mozambique between 1985 and 1992 damaged transport infrastructure and blocked the ports of Nacala and Beira, which raised transport costs. The favourable conditions enjoyed in previous -2- decades in output prices, transport costs and credit were reversed and incomes correspondingly contracted. Figure 1.1: Phases of GDP per capita in Malawi 1960-2008 Source: NSO, WDI 1.5 1989-2003: Agriculture becomes smallholder-led amidst macro-instability and external shocks. The introduction of legislation in the early 1990s making it legal for smallholders to grow export crops, dramatically shifted the pattern of agricultural production away from estates. From nearly nothing in 1990, smallholders came to produce around 70 percent of the tobacco crop. However, high fiscal deficits combined with exchange rate liberalization rapidly transmitted price instability to the rest of the economy, and inflation reached a high of 83 percent in 1995. Additionally high domestic borrowing caused real interest rates to exceed 20 percent in 2000-04 which strongly crowded out private investment. Volatility in output was further exacerbated by various external shocks including droughts in 1992 and 1994 and an increased influx of Mozambican refugees. 1.6 2004-2009: Stabilization enabled growth to resume and in 2007, after 28 years, incomes per capital recovered their levels of 1979. The change of government in 2004 brought about a rapid turnaround in government finances. In extremely difficult fiscal circumstances, and for the first time since 1994, the government stayed within the planned budget and, as a result, the fiscal position dramatically improved. This led to an increase in donor inflows, which in turn allowed the government to further reduce its need to borrow domestically thus allowing more resources to become available for private sector investment. The following section discusses the drivers of long-run growth with particular attention to the recent recovery. B. WHAT HAS DRIVEN GROWTH? 1.7 Growth in Malawi is dominated by agriculture, and recently has seen an increasing contribution from domestic services. Between 1995 and 2003 agriculture -3- accounted for nearly three-quarters of all economic growth (see Table 1.1). After 1995, there is a general pattern of falling growth rates until reaching a low point during the drought of 2001/02, followed by a gradual resumption output growth. However, growth since 2004 has been structurally different, being composed of increasingly important contributions from distribution, finance, construction and manufacturing. The headline data on trade also shows an equivalent change in dynamic: whilst exports have stayed within the region of 20 percent of GDP, the trade deficit has significantly worsened and since 2003 has exceeded 10 percent of GDP. Table 1.1: GDP and sources of growth 95 96 97 98 99 00 01 02 03 04 05 06 07 National Account ($US m) GDP ($US m) 1,925 3,149 3,669 2,451 2,447 2,402 2,365 2,665 2,425 2,625 2,755 2,917 3,324 Exports, fob ($US m) 444 510 539 539 447 402 427 414 433 499 509 543 706 Imports, fob ($US m) 508 588 697 497 573 460 471 595 684 810 1,006 1,055 1,182 Trade ratios (as percentage of GDP) Exports (fob) 23.1 16.2 14.7 22.0 18.3 16.7 18.0 15.5 17.9 19.0 18.5 18.6 21.2 Imports (fob) 26.4 18.7 19.0 20.3 23.4 19.2 19.9 22.3 28.2 30.9 36.5 36.2 35.6 Trade balance -3.3 -2.5 -4.3 1.7 -5.1 -2.4 -1.9 -6.8 -10.3 -11.8 -18.0 -17.6 -14.3 Growth in GDP (factor cost) 13.8 10.0 6.6 3.0 1.6 0.8 -3.9 1.9 3.9 5.1 2.2 8.6 9.2 Agriculture 9.8 7.8 0.0 5.3 1.6 2.0 -2.4 1.0 2.3 1.1 -3.5 4.3 3.5 Mining, quarrying 0.0 1.5 -0.4 0.1 0.0 0.1 0.1 -0.6 0.2 0.5 0.8 -0.5 0.0 Manufacturing 0.9 -0.1 0.1 0.2 0.2 -0.4 -1.8 0.0 0.4 0.8 0.9 0.7 1.0 Electricity, water 0.0 0.0 0.1 0.1 0.0 0.1 -0.1 0.1 0.0 0.1 0.1 0.1 0.1 Construction 0.1 0.2 0.1 0.0 0.3 0.0 -0.1 0.3 0.3 0.3 0.4 0.4 0.9 Distribution 0.6 -0.2 3.7 -1.6 -0.4 -0.1 0.2 0.3 -0.2 1.4 2.8 1.8 1.3 Transport, communications 0.9 -0.4 0.4 0.0 0.2 -0.2 0.1 0.6 0.4 0.4 0.5 0.4 1.3 Financial, professional 0.7 1.3 2.5 -0.7 0.0 0.2 -0.2 0.5 0.5 0.8 0.7 1.7 1.1 Ownership of dwellings 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0 Government services 1.0 -0.3 0.2 -0.5 -0.2 -1.0 0.1 0.0 0.2 0.2 0.1 0.3 0.6 Source: NSO 1994 series of accounts. GDP is from 2002 series back-adjusted using the 1994 series rates of growth. 1.8 Agricultural production and policy is dominated by maize. Although there has been some diversification and a notable increase in smallholder tobacco production, the area under maize cultivation is still roughly equivalent to that of all other crops combined. Over 60 percent of national calorie consumption derives from maize; 97 percent of farmers grow maize; and over half of households grow no other crop2. Given Malawi's vulnerability to drought, the maize harvest is thus central to the welfare of the population, and one of the most important political and social variables. This is reflected by the prominence of food security in public policy. Through the Fertilizer Subsidy Program, the government has made increasing maize yields the mainstay of agricultural policy, and has designated it a "strategic crop" subject to import and export bans (more recently intervening to set the domestic price). Together with good rains, the increased use of fertilizer has increased yields and in the last three seasons substantially strengthened food security. 1.9 Yet despite the prevailing assumption of the centrality of maize in the economy, the maize harvest drives volatility but does not dominate growth. The per- 2 See GoM's Integrated Household Survey, 2006 and GoM's and World Bank's joint Poverty and Vulnerability Assessment, 2006. -4- capita maize harvest is plotted alongside GDP3 in Figure 1.2. Although it is clear that very poor harvests are associated with contractions in GDP (GDP/cap fell by 8.5 percent in 1992 and by 9.6 percent in 1994) the correlation is weaker in years of good harvest. Clearly, maize contributes to GDP ­ by definition ­ but strong maize harvests alone do not create strong overall growth, although they do generally lead to lower food prices and thus lower inflation. This weak link between maize production and GDP is at odds with the prevailing assumption of growth dynamics in Malawi. Figure 1.2: Per capita maize harvest and GDP 2005 1990 Source: NSO, Ministry of Agriculture and World Bank Staff 1.10 One explanation for this is that maize creates a limited multiplier effect on the rest of the economy due to the low proportion of production, which is sold. The vast majority of maize is grown by smallholders for their own consumption, generating little cash income, and hence, exerting little stimulus on the rural economy through onward purchases. The Poverty and Vulnerability Assessment (PVA) of 2006 estimates that only 10-15 percent of the total crop ever reaches the market and that only 18 percent of smallholders sell more than half of their output. Furthermore, maize is subject to an export ban, which ensures that total demand is inelastic ­ hence even if cash-selling were more widespread, the ceiling on domestic demand would limit its potential as a driver of growth. A further explanation for the weak connection between maize and growth is the low net profitability of production under fertilizer, which, while increasing yield, has little effect on value-added because of the increased cost of inputs. 1.11 The structure of export crop production in Malawi would suggest that exports could strongly affect growth through a domestic multiplier effect. Exports affect growth in three ways: firstly, because they contribute by definition to GDP; secondly, because of the stimulus effect of exporters spending their income in the rest of the economy; and thirdly, because of the need for auxiliary services such as transport and finance. The structure of production in Malawi makes the hypothesis of growth driven by 3 The comparison of production weight against value-added is justified here since this series of National Accounts uses fixed baseline prices to construct GDP. -5- a demand multiplier from exports promising, since exports are highly labour intensive and production is dispersed over a large area. This is particularly true for tobacco which accounts for about 60 percent of export revenue, but also applies to estate crops such as sugar and tea, which disburse cash income to a large labour force. 1.12 Exports expressed in real kwacha show a persistent long-standing relationship with GDP implying a strong multiplier effect from export revenue4. This is consistent with the demand effect of export income acting as a primary driver of growth through its multiplier effects. There are grounds to take this hypothesis seriously in a small open economy such as Malawi. Firstly, the prevalence of subsistence farming restricts the demand effect from smallholder agriculture to the portion of production that is marketed. Secondly, the dominant distribution sector is highly sensitive to export volumes through cash-crop transport and symmetrically through import transport. Lastly, because of the reported sharp increase in retail network sales in years where cash crop income from smallholders is high (see Table 1.3). Figure 1.3: Exports in constant MK and growth (1960-2007) 2007 1978 1993 1960 Source: NSO and World Bank Staff estimates. 1.13 The strength of this effect suggests that real domestic export income is the dominant driver of growth in Malawi. Which way does the causality run? It is unlikely that the relationship between exports and GDP is due to covariant behaviour between exports and the domestic agricultural sector: the per capita maize harvest for example is nearly perfectly independent from the per capita amount of tobacco sold at auction. Neither is it likely that causality runs the other way around - GDP growth causing a rise in real domestic export income ­ because of the significant exogenous role of export prices. The evidence thus suggests that export income expressed in real kwacha 4 The explanatory power of the regression between GDP per capita and export per capita in real kwacha is 79 percent (R2 =0.791). -6- is likely to have been the primary driver of growth, and that this relationship has persisted since 19605. 1.14 The importance of exports expressed in real kwacha highlights the importance of the real exchange rate. Since the bilateral real exchange rate is equal to the ratio of exports expressed in real dollars over exports expressed in real kwacha, a depreciation of the rate increases the real kwacha value of exports relative to their real dollar value. The broad depreciation of the real exchange rate in Malawi since 1980 (shown in Figure 1.4 by the rise in its reciprocal) has thus had a greatly beneficial effect on real kwacha exports: increasing their value ceteris paribus by about 2.5 times. Because of the sensitivity of GDP to export revenues, this depreciation of the rate is very likely to have also had a large beneficial effect on incomes, and more than any other policy instrument, enabled Malawi to grow under such hostile terms-of-trade. Figure 1.4: Exports in real dollars, real kwacha and the real exchange rate Source: NSO and World Bank staff estimates. 1.15 But improvements in the real exchange rate have been volatile and the current dollar peg has created severe appreciation since August 2008. In the 1990s fiscal policy was not tight enough to control the price transmission from imports into domestic prices, thus sharp nominal devaluations caused the real exchange rate to depreciate and then to re-appreciate as the costs of imports fed through into domestic prices. This instability is likely to have disrupted growth by increasing price uncertainty. Partly as a result of this, policy makers have come to use the nominal dollar exchange rate as a stabilization instrument. This was a beneficial anchor while the dollar was also on a depreciating trend, however after its strong appreciation against major currencies since August 2008, this lock will have serious negative consequences for export sector growth unless a new exchange rate policy is adopted. 5 The persistence of this effect over the long-term implies the operation of multiplier effects for both smallholder and estate production. There is some evidence to support this. Firstly much of the demand effect is exerted at the wholesale level for transport and finance. Secondly increased retail activity is observed in districts with large estates such as Thyolo, and around the sugar estates such as Dwangwa. -7- Figure 1.5: Real exports per capita and export volume Source: NSO, Balance of payments and World Bank staff estimates 1.16 If exports have driven growth, what then has driven exports? Exports continue to be dominated by tobacco despite a gradually expanding share of non- traditional products. Since 1994, tobacco has consistently accounted for around 60 percent of export revenue; the other traditional exports of sugar and tea have seen declining shares over the last five years; and there has been a gradual expansion in the share of non-traditional exports. Non-traditional exports accounted for 20 percent of total revenue in 1994 and 29 percent in 2006. Although there has been some recent growth in production volumes, the majority of this has come from tobacco with other traditional exports exhibiting gradual but volatile growth. Figure 1.6: Tobacco price and production 2008 1994 Source: Tobacco Control Commission of Malawi 1.17 Tobacco production responds moderately to the real auction price of the previous year. Data from the Tobacco Control Commission reveals that prices (again in real kwacha) in the preceding year account for just under 50 percent of the production response. This implies an additional growth dependency on real domestic export prices. -8- Smallholders respond to price signals with a price elasticity of supply of 0.61 - a 1 percent rise in the real domestic price at auction yields a 0.61 percent rise in the following year's supply. 1.18 The combination of GDP sensitivity to exports, and export volume sensitivity to prices creates a double (quadratic) dependency on the real exchange rate. A depreciation in the real exchange rate acts on GDP through two channels. Firstly, a real depreciation immediately increases the real domestic incomes received by exporters for their product. This then exerts a multiplier effect on the rest of the economy through the additional real revenue spent. Secondly, a rise in real domestic export prices stimulates additional production in the following season. Increased prices and increased production combine to give a quadratic increase in export revenue, which, as already discussed is transmitted to GDP. The analysis suggests therefore that for Malawi, the real exchange rate has been, and continues to be, a central policy instrument of growth. 1.19 But tobacco-led growth is unlikely to deliver sustained high growth rates over the long-term. Recent growth can in large part be explained by improvements in the real domestic tobacco price. Yet growth based on an expansion of tobacco production does not appear to be a viable strategy for the medium to long term. Malawi's tobacco is effectively a niche product: burley used mainly as a neutral flavored filler for high-end cigarettes. Although growing, there is a ceiling on world demand for such tobacco, which grows at a relatively slow rate. This precludes significant scaling-up of production without bringing about a decline in world prices. A failure to diversify into new export products thus constrains Malawi's growth into a narrow channel and renders the economy vulnerable to adverse price shocks when they occur in the future. Figure 1.7: Malawi's regional exports Regional exports as a share of total exports 40.0 35.0 % o f to ta l e x p o r ts 30.0 South Africa 25.0 Neigb countries 20.0 15.0 SADC 10.0 Zimbabwe 5.0 0.0 2000 2001 2002 2003 2004 2005 2006 2007 Source: UN COMTRADE and World Bank staff calculations 1.20 The notable growth in non-traditional exports has been led by demand in regional markets which have significantly increased their share of Malawi's exports. During the period 2000-06 traditional exports (tobacco, tea, sugar) grew at an average rate of 2.9 percent but non-traditional exports grew at an annualized rate of 14.3 percent (BOP, NSO). Regional export markets account for a high proportion of Malawi's non- traditional exports: whereas 67 percent of exports to the OECD since 2000 have been -9- traditional products (tobacco, sugar and tea) 66 percent of exports to Sub-Saharan Africa are non-traditional. Thus the growing share of trade with regional markets (see Figure 1.7) has underpinned non-traditional export growth and signals the potential of regional trade in further diversification. The nascent sectors of mining and tourism are anticipated to be a source of future export growth, but are subject to similar supply-side constraints that affect manufacturing.6 Table 1.2: Export shares to regional trading partners by product (2000-07) S. Africa Zimbabwe Egypt Kenya Mozambique Zambia Tanzania Traditional Tobacco 2.77 0.74 4.48 0.84 0.10 Sugar 1.12 0.03 1.77 0.30 0.06 0.41 Tea 1.96 0.18 0.76 Non-traditional Apparel 3.46 0.10 0.01 0.18 Cereals (including maize) 0.44 2.37 0.22 0.20 0.04 Oil seed 1.59 0.32 0.03 0.05 0.10 0.06 Cotton 1.45 0.01 0.10 0.17 Rubber 0.35 0.09 0.01 0.16 Printed books 0.61 Heavy Machinery 0.34 0.10 0.08 0.04 Wood & wood articles 0.37 0.03 0.07 0.09 Plastics 0.44 0.11 Milled products 0.20 Vehicles 0.04 0.13 0.03 Soap 0.05 0.13 Furniture 0.01 0.14 Fertilizer 0.11 Beverages & spirits 0.08 Iron & steel 0.01 0.05 Country's total export share (%) 15.30 4.80 4.70 2.90 2.80 1.70 1.00 Source: UN COMTRADE and World Bank staff calculations 1.21 The single largest destination for both total exports and non-traditional exports is South Africa, which provides an indicator of potential areas of diversification. Table 1.2 gives the share of total exports by product for the seven most significant regional markets ­ all of which are either members of SADC or COMESA. South Africa dominates the market for non-traditional exports, in particular apparel, oil seed, cotton, cereals and wood. The statistics for Zimbabwe are skewed by the maize exports of 2007, and there is a broad but relatively small market for non-traditional exports in Mozambique and Zambia. The size of the South African market, its different structure of factor endowments, the relatively low cost of transport from Malawi, and the strength of current investment and trade links, underline the future importance of South Africa for the diversification of Malawi's exports. C. THE PATTERN OF GROWTH 1.22 The hostile macroeconomic environment 1995-2003 led to a collapse in private investment. Nominal lending rates exceeded 45 percent for three-quarters of this period, and inflation peaked at over 90 percent early on in 1995. Although reliable 6 Details of constraints facing mining can be found in Malawi ­ Mineral Sector Review, Report No. 50160- MW (World Bank, 2009) ,while the opportunities and challenges in the tourism sector are discussed in Malawi-Tourism Sector Report (World Bank, forthcoming). - 10 - investment data is difficult to assemble, the most robust sources are shown in Figure 1.8 which confirm a constricting of total investment during macro-instability 1995-2003. The investment recovery (confirmed by a triangulation against capital goods imports) coincides with the sharp correction of real lending rates which fell from 43 percent in the third quarter of 2003 to 22 percent a year later. Similarly domestic credit to the private sector resumed growth in 2004 as fiscal discipline improved and reversed the crowding- out effect. Figure 1.8: Macro-environment and investment Source: IMF, WDI, NSO and World Bank staff estimates 1.23 With stability established, growth diffuses out of smallholder agriculture into non-tradeable sectors. The largest contribution to growth comes from agriculture, nearly all of which is attributed to the smallholder. The sectoral pattern of GDP confirms a lack of transmission during the period of macro-instability 1993-2003, and from 2003 presents growth diffusing out of agriculture through to financial services, distribution, manufacturing, transport & communications and eventually through to construction. Distribution is the largest contributor to this growth - predominantly a passive response to increased international trade and agriculture expansion ­ and financial reform since the 1980s catalyzed growth in the banking sector enabling the number of banks to expand from two in 1994 to nine in 2008. Table 1.3: The diffusion of growth out of agriculture 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Agriculture, Forestry and Fishing 7.8% 0.0% 5.3% 1.6% 2.0% -2.4% 1.0% 2.3% 1.1% -3.5% 4.3% 3.5% Financial and Professional Services 1.3% 2.5% -0.7% 0.0% 0.2% -0.2% 0.5% 0.5% 0.8% 0.7% 1.7% 1.1% Distribution -0.2% 3.7% -1.6% -0.4% -0.1% 0.2% 0.3% -0.2% 1.4% 2.8% 1.8% 1.3% Manufacturing -0.1% 0.1% 0.2% 0.2% -0.4% -1.8% 0.0% 0.4% 0.8% 0.9% 0.7% 1.0% Transport and Communications -0.4% 0.4% 0.0% 0.2% -0.2% 0.1% 0.6% 0.4% 0.4% 0.5% 0.4% 1.3% Construction 0.2% 0.1% 0.0% 0.3% 0.0% -0.1% 0.3% 0.3% 0.3% 0.4% 0.4% 0.9% Producers of Government Services -0.3% 0.2% -0.5% -0.2% -1.0% 0.1% 0.0% 0.2% 0.2% 0.1% 0.3% 0.6% Electricity and Water 0.0% 0.1% 0.1% 0.0% 0.1% -0.1% 0.1% 0.0% 0.1% 0.1% 0.1% 0.1% Private Social and Community Services 0.2% 0.2% 0.0% 0.0% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1% Ownership of Dwellings 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.0% Mining and Quarrying 1.5% -0.4% 0.1% 0.0% 0.1% 0.1% -0.6% 0.2% 0.5% 0.8% -0.5% 0.0% Unallocable Finance Charges -0.1% -0.4% 0.1% -0.3% -0.1% 0.1% -0.5% -0.3% -0.6% -0.7% -0.7% -0.7% Growth in GDP at Factor cost 10.0% 6.6% 3.0% 1.6% 0.8% -3.9% 1.9% 3.9% 5.1% 2.2% 8.6% 9.2% Source: National Statistics Office, National Accounts 1994-2007 - 11 - 1.24 The lack of growth in electricity and water is an early warning signal. Despite the recovery of user industries, sectoral data implies a per capita contraction in electricity and water since 1995. Together with the prevalence of load shedding and frequent interruptions in water supply, this is an early signal of a potentially binding supply constraint. 1.25 Domestic demand has created a strong response from non-tradeables and imports but only a muted response from domestic manufacturing. Although growth resumed in manufacturing in 2003, the sector is yet to reclaim its level of 1995. This muted response is important. Production for the domestic market, ISI (Import- Substituting Industrialization) is naturally protected from external competition by the relatively high costs of importing into a landlocked country7; and its slow growth contrasts sharply with the strong rise in imports. Imports have doubled over the last five years in nominal dollar terms. The sluggish performance of domestic manufacturing cannot therefore be attributed to a lack of aggregate demand (see also section 4) but is potentially constrained by a shortage of skilled workers as the manufacturing sectors requires a different skill mix then agriculture (see Box 1.2). Figure 1.9: Manufacturing versus imports Source NSO National Accounts 1994-07, NSO Balance of Payments 1.26 The recent import boom has been financed by exports, aid, and errors and omissions. The depletion of foreign exchange reserves has also contributed, falling to 1.3 months of import cover by the end of 2008. The combination of exports and official transfers fell short of funding imports in 2004-06, which looks to have been closed by the contribution from errors and omissions. It seems likely that the account includes a proportion of repatriated capital since the flow reverses at the end of the inflationary period from a US$200 million outflow 1998-2001 to a similar sized inflow 2002-06. 7 In 2006, the carriage insurance freight charge for imports was around 15% of their FOB value (NSO). - 12 - Box 1.2: Is growth constrained by education? The ICA (World Bank, 2006a) ranks shortage of skilled workers as the eighth most serious obstacle to growth after macro-instability, finance, power, tax, and crime. Other indirect evidence includes the high proportions of formal sector businesses that have non-Malawians in key posts; and high salaries in the financial sector relative to comparator countries reflect artificially high returns to business skills. The evidence signals a strong probability that education currently constrains growth in Malawi, and an even greater probability that it will become a constraint in the medium term. Either way, given the lead times of education reform, it is critical that government and donors intensify their efforts to improve quality and access particularly of primary and tertiary education (see Malawi Education Country Status Report, World Bank, forthcoming) Despite significant investment in school infrastructure, curriculum development and the expansion of Teacher Training Colleges, primary education remains highly stressed in Malawi. The introduction of Universal Primary Education in 1994 caused enrolment to rise by 43 percent with no corresponding expansion in the supply of teachers or classrooms. As a result the ratio of teachers to pupils is the lowest of comparators at 1:76; and classrooms to pupils at 1:107 (Malawi Education Statistics, 2006). Less than half of children complete their first four years in school, also the lowest among peers, and tertiary enrolment is three times lower than the next worst performer in the comparator table. Source: WDI 2005, SACMEQ 2005 Gender equity in education compares favourably with peer countries. Fractionally more girls complete primary school than do boys, and girls account for 46% of secondary school enrolment (World Bank, 2006b). However, the quality of education is extremely weak. Cross-country testing of standard six pupils by SACMEQ reveals Malawi to have the lowest scores of peers in both numeracy and literacy. The chart above plots the percentage of pupils attaining "level 5" in tests (defined as "competent numeracy" and "inferential reading"). The proportion attaining the numeracy level is too low to be visible. Does this weak performance constrain growth, or is it just a symptom of underdevelopment? The degree of Malawi's divergence from comparators both in terms of access and quality signals that education is likely to be part of the cause of low development. There is also direct evidence to support this. Data from the IHS (GoM, 2006) shows that household tobacco income is strongly correlated with the educational level of the household head, conforming the findings of the PVA that better educated farmers are able to understand the risks and farming practices of cash-cropping. The same dataset reveals that household expenditure is strongly associated with the education level achieved by the head of a household. - 13 - 1.27 The rise in imports since 2004 has been accompanied by a change in composition skewed towards consumption. In real dollar terms, consumption imports have increased by 86 percent, intermediate imports by 50 percent (mainly on account of fuel and fertilizer) and investment imports by only 29 percent. In 2006 at over 14 percent of GDP, Malawi's current account deficit was the highest of all comparator countries. Figure 1.10: Import composition and the external balance Source NSO Trade Data, IMF 1.28 The rise in imports reveals a strong growth of consumption demand, which - despite natural protection - has not created a significant response from domestic industry. Taken together with the trade deficit and run down of reserves, this implies that the exchange rate is overvalued. More recently, banks have been forced to ration foreign currency to importers and the spread between formal and informal exchange rates has widened to 30 percent. There has also been some support for an overvaluation hypothesis in the literature (see Rajan and Subramanian8, 2005). The role of aid 1.29 Malawi's level of aid places it among countries where the question of overvaluation due to aid has provoked study. In 2008, aid per capita was $29, aid as a proportion of GDP was 13.4 percent, and aid as a ratio of merchandise exports stood at 65 percent. Since the size of aid inflows relative to the external sector is an initial indicator of the strength of an appreciation effect on the currency, this ratio puts Malawi among countries such as Uganda and Tanzania where the question of whether aid has harmed export competitiveness has been investigated (Asmah & Levin 2008, Adam & Bevin 2006). 1.30 The aid literature predicts that the net effect of aid on growth critically depends on how the aid is spent. For aid to be absorbed requires an equivalent rise in import demand. This can occur either through the state explicitly deciding to purchase 8 Rajan and Subramanian (2005) find Malawi's exchange rate to be overvalued, but others have found that technical properties of the data obstruct the application of particular frameworks to test the validity of Purchasing Power Parity. It is very likely that changes of regime are contaminating the unit-root tests, and the lack of forward exchange markets prevents the use of interest rate parity. - 14 - more imported goods and services which raises the import demand of the public sector; or by an appreciation of the real exchange rate which raises the import demand of the private-sector. If the government uses the exchange rate to absorb aid, then this will cause an appreciation but, if the aid is spent in such a way as to raise long-term productivity, then this will adjust the real equilibrium exchange rate counteracting the effect of the initial appreciation (Adam & Bevin, 2006). In other words, an aid inflow is likely to appreciate the real exchange rate, but if it finances productivity growth then this consequence becomes less relevant since the long-term equilibrium reference rate will also rise. Note that this equilibrium rate moves only with structural changes in productivity, and not with temporary improvements such as those arising from increased use of intermediate inputs financed by a temporary aid flow (such as the productivity gains derived from the current fertilizer subsidy program). 1.31 There is some evidence of real exchange rate appreciation due to the absorption of an increased level of aid. Using data from the Development Assistance Committee (which is constructed for donors rather than recipients) aid as percentage of GDP has had no consistent immediate effect on GDP per capita. This is not to say that aid has not affected growth since aid can raise productivity through many channels and over many different timescales. Aid surged as a proportion of GDP in 1987 and expressed as a proportion of exports, aid significantly increased in 1987 from around 29 percent of exports (1970-86) to an average of 56 percent (1987-2007). This surge is associated with a temporary appreciation of the real exchange rate (RER) which persists until 1994. This can be seen in Figure 1.11 which shows the long-term depreciating trend of the RER since 1980 with a temporary appreciation coinciding with the step change in aid/exports in 1986. This is consistent with aid applying upward pressure on the RER in proportion to the size of the external sector, and the effect being counterbalanced by longer term depreciating forces such as the liberalisation of trade and foreign exchange and adjustment to the terms-of-trade. Figure 1.11: Effects of aid on growth and competitiveness Sources: DAC,IMF, NSO and World Bank staff estimates - 15 - 1.32 An increase in public sector imports have probably mitigated aid-induced appreciation making it unlikely that aid strongly constrains growth through competitiveness effects. There has been a significant expansion of direct importing by the public sector. Aid inflows have created the fiscal space to finance the fertilizer subsidy program, which constituted a huge 4.6 percent of GDP in 2008/09 - equivalent to one third of all aid inflows. Additionally the high proportion of project aid (about 75 percent) over budget aid tends to finance a higher import component. This growth in direct importing by government reduces the need to appreciate the real exchange rate since more aid is absorbed by the public sector. Table 1.4: Composition of expenditures Percentage of GDP 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 Aid inflow to the budget 4.3 4.2 8.2 10.5 12.8 14.4 12.3 16.9 Total expenditures 22.4 18.3 23.0 22.4 28.2 29.6 33.7 36.1 Current Expenditures 16.7 15.1 19.0 17.8 22.2 23.8 28.1 29.1 Fertilizer program 1.7 1.9 2.7 4.6 Capital Expenditures 5.8 3.2 4.0 4.6 5.9 5.8 5.6 7.0 Gross fixed capital formation 3.9 3.1 3.7 4.3 5.2 5.2 5.4 5.0 Source: IMF, Public Finance National Reports, Government of Malawi Budget Documents 1.33 But aid has only weakly supported growth - persistently low levels of public investment imply that aid has not sufficiently strengthened productivity. Total public investment amounts to around 40 percent of aid inflows. This implies that in excess of 60 percent of aid (since some investment will have been financed domestically) is used for recurrent or consumption spending. Although Malawi currently uses its aid for social development, which will have long-term impacts on productivity through improvements in health and education, aid only weakly addresses supply bottlenecks in support of growth such as power, transport or irrigation. D. THE SPATIAL DIMENSION OF GROWTH 1.34 Where is growth occurring? Spatial data shows that the distribution of cash- crop agriculture is strongly affected by agronomic potential. Tobacco production is geographically concentrated in areas of higher agronomic potential9 for cash crops (see Figure 1.12). Also spatial data reveals that areas of higher agronomic potential are relatively more specialized in high-value and intensive crop production, and that crop productivity is higher. 1.35 Crop production intensity is also influenced by market access. Using travel times to the nearest major city (Lilongwe, Blantyre, Mzuzu, and Zomba) Table 1.5 gives the percentage of cash crops produced at different distances from cities and for different levels of agronomic potential. Crop production is concentrated in locations with good 9 Agronomic potential is derived from the IFPRI Spatial Allocation Model as a composite of the maximum potential yield for coffee, cotton, and groundnuts: all of which share similar cultivation conditions as tobacco. - 16 - agronomic potential and production shares increase with lower travel times to urban centers. Access to urban markets increases profitability through lower transport costs and better access to information and intermediate inputs. Closest to urban centres, crop intensities fall sharply as land for agriculture competes with urban land use demand. Figure 1.12: Tobacco production intensity and agro-climatic conditions Source: Integrated Household Survey 2004 and IFPRI Spatial Allocation Model 1.36 It is likely that a reduction in travel time in remote areas would thus encourage households to specialize in cash crops. Relating the proportion of cash crops grown to agronomic potential and travel time using an econometric model reveals that (i) the beneficial effect of agronomic potential diminishes with remoteness from urban centres; (ii) close to urban centres competition for land use discourages cash crop production and consequently the most intensive cash-crop production occurs around 2.2 hours from a major city. The highest return from transport infrastructure investments has thus come from allocating resources to feeder roads in areas with good agronomic potential that are relatively close to markets. Table 1.5: Cash crop production (%) by agronomic potential and travel time Quintiles of travel time to nearest major city 5 (longest) 4 3 2 1 (shortest) 5 (worst) 1.1 3.3 3.5 0.1 0 4 3 0.7 1 2.3 0.8 Quintiles of Agronomic 3 5.4 13.7 1.4 1 0.5 potential 2 10.3 4.1 9.3 2.1 0.2 1 (best) 1.2 7.6 10 15.1 2.2 Data Source: GIS+HDM4 data, IFPRI SPAM, and the IHS2 - 17 - 1.37 Manufacturing industry is concentrated in Blantyre and Lilongwe. Industry in Blantyre is higher-tech and more externally orientated. Blantyre is physically closer to international markets and hence firms specialize in activities benefiting from access to international markets. Lilongwe is located in the agriculture belt and hence firms are more linked to domestic markets through agro-processing. Firms in Blantyre specialize in textiles, chemicals, plastics and metal products; are more export orientated (45 percent of direct sales) use more imports (67 percent of intermediate inputs) and are relatively higher-tech than those in Lilongwe (Malawi Productivity & Investment Climate Survey, 2004). In contrast, firms in Lilongwe produce relatively low-tech products mainly for the domestic market with only 9 percent of sales directly exported. The main industries of Lilongwe are food and beverage products, as well as rubber/plastic products and furniture. Table 1.6: Industrial structure in Blantyre and Lilongwe Blantyre Lilongwe Number of firms sampled 97 32 Exports (proportion of direct sales) 44.7% 8.8% Imports of intermediate inputs 66.7% 43.7% Proportion of firms with generators 55.7% 28.1% Obstacles to doing business (1=none, 5=severe) Electricity 3.6 3.3 Transportation 2.6 2.7 Telecommunications 2.3 1.9 Source: Malawi Productivity and Investment Climate Survey, 2004 E. OBSERVATIONS REGARDING GROWTH IN MALAWI 1.38 Malawi is an export-led economy. Despite changes in the structure of agricultural production, and the liberalization of prices, exchange and finance, a long- standing relationship persists between export revenue expressed in real domestic currency and overall GDP. The driver of the economy has been the multiplier effect from exports which has created growth by stimulating non-tradeable services, and to a limited extent import substituting industrialization. 1.39 Malawi has faced chronic deterioration in its terms of trade since the 1970s but has recovered largely due to depreciation of the real exchange rate and more recently improvements in the tobacco price. The economy has proved highly sensitive to movements in both these variables since improvements (i) cause an immediate rise in real domestic export revenue which multiplies through the rest of the economy; and (ii) incentivize a production expansion by increasing tradeable returns. 1.40 However, there are strong signs that despite progress the real exchange rate is still overvalued. Recent strong growth in the demand for consumer imports has failed to stimulate much of a response from domestic manufacturing which is also constrained by inadequate provision of infrastructural services such as electricity and transport - 18 - including trade logistics. There has been no significant progress in diversifying out of the traditional exports of tobacco, sugar and tea. Recently, there are widespread and persistent shortages of foreign exchange, low reserves and the current account has a higher deficit than any comparator economy. Malawi is in effect running a fixed exchange rate regime which artificially overvalues the kwacha not just from revenues from the booming tobacco sector, or aid, which finances the capital account, but also from its management regime which incentivizes importing at the expense of exporting. · Growth is primarily driven by the multiplier from real domestic export income · Improvements in the real exchange rate since 1980 have raised real export incomes and have been strongly beneficial to growth · Growth transmission has run from agriculture to non-tradeables and only minimally to domestic manufacturing. This is despite strong demand for consumer imports. · The real exchange rate is thus still overvalued and acting as a strong disincentive to invest in the tradeable sector or in import-substituting industrialization. · Macroeconomic instability between 1995 and 2003 suppressed investment through high inflation and cost of credit which prevented growth transmission 1.41 Aid inflows are substantial relative to the size of the external sector and have probably contributed to exchange rate appreciation. This effect has been mitigated by the increase in direct public-sector imports particularly from the fertilizer subsidy. But aid has had a limited effect on long-term productivity since it predominantly finances consumption and recurrent expenditure. Expenditure data from the Ministry of Finance indicates that less than 40 percent of aid is used for investment. Without financing increases in productivity, aid risks doing long-term damage to growth by appreciating the currency and creating disincentives in the export sector. Because of the centrality of the export sector to Malawi's economy and its sensitivity to the real exchange rate, it is vital that the use of aid is realigned to support export-led growth. 1.42 The most binding constraint on the economy is thus the real exchange rate, which can only be ameliorated by a careful managed devaluation to ensure that price stability is maintained. Growth diagnostics using the HRV10 framework further identifies finance (see section 4), power (see section 3.C), education (see Box 1.2) and trade value chains (see section 3.A) as constraints that are currently binding on marginal investment in Malawi (Box 1.3). 1.43 Malawi's growth pattern and the proposed way forward in this report can have significant impact on poverty, in particular through (smallholder) agriculture and benefit gender equality (see Box 1.4). 10 Hausmann, Rodrik and Velasco, 2005, "Growth Diagnostics". - 19 - Box 1.3: Additional constraints identified by the growth diagnostic Financial Binding. High overheads have created high real lending rates which currently Intermediation constrain investment. Very limited access to formal finance for smallholders reduces the production response of rural farmers to price signals (see also Chapter 4, box 4. 1). Power Binding. The lack of reliable power source significantly lowers social returns, and deters new investments in manufacturing and mining. The lack of rural power additionally exacerbates deforestation (see Chapter 3, section C). Education Binding or imminently binding. Extremely weak performance in quality and access, and evidence of high returns to primary education from cash cropping and tertiary from formal sector employment. Moderate signals of skills shortages from business surveys. The long lead-time of reforms suggests that the quality of primary education in Malawi should be addressed with urgency (see also Box 1.2 above). Export value Binding. An increase in farm-gate prices achieved through efficiencies in input or chains product marketing chains stimulates growth by increasing the production response and the domestic multiplier. Efficiency improvements are probable from regional infrastructure improvements, administrative barriers to trade, and the domestic input and marketing chains (particularly tobacco) (see Chapter 2 and Chapter 3 section A and B in particular). Micro policy Secondary. Government policy facing the firm generally performs well against peers with the exception of administrative barriers to trade and the unpredictability of interventions in agricultural prices. Price interventions are likely to have raised ex-ante risk, creating monopoly rents and in turn raising the cost of inputs to farmers. - 20 - Box 1.4: Growth, gender, and per capita income This box provides some background on how growth and per capita income interrelate to gender equality. It will also discuss how the proposed development agenda of this report can contribute to bridging the existing gender gap in Malawi.11 12 Gender equality is an important dimension of development in its own right. However, besides that there is a positive correlation between GDP per capita and gender equality (see figure 1), more evidence is also emerging that gender equality can spur growth and thus poverty reduction. Even though some of the statistical evidence may be subject to further interrogation, the available evidence clearly points to the fact that countries that increase women's access to education (see figure 2), health care, employment and credit, and narrow the differences in pay and employment opportunities between men and women, are able to increase overall growth and reduce poverty (See DFID (2007), Stotsky (2006), and World Bank (2001)). Figure 1: Gender equality and per capita Income Figure 2: Gender inequality in education and per capita income 1 1.4 0.9 ) 1.2 5 ) 0.8 0 5 0 0 2 0 ( 2 ( o i x 0.7 t a 1 e r d t n I n t e n m e 0.6 l l m o MAL p r 0.8 o l n e e v 0.5 y e r D a d d e MAL n 0.6 t a o c l 0.4 e e s R s r s e o d r 0.4 n 0.3 G e G 0.2 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 11 11.5 0.2 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 11 11.5 GDP per Capita (LN, PPP, 2005) GDP per Capita (LN, PPP, 2005) Source: UNDP Human Development Database How can the proposed policies of this report, which aim to support growth, also enhance gender equality? The current situation of an exchange rate that has appreciated sharply during 2008 and has been kept at this appreciated level since is clearly unfavourable for growth but also for closing the gender gap. The unfavourable FX developments can be expected to impact adversely Malawi's ability to diversify its export base, increase the value addition of its exports, and engage in import substitution small manufacturing, and therefore future growth is anticipated to be lower and to create less employment. It is a well documented fact that increasing off-farm employment is beneficial for women in particular if it is in small value addition manufacturing such as garment industry. In addition, when more of the household income accrues to the woman and/or when she is able to exercise greater control over household economic resources, the spending pattern changes and favours health, nutrition, and education of children in the household and therefore improves their future income-earning capacity, their future productivity, and thus the long term fundamentals for growth (see Thomas (2002) and Schultz (2002)). 11 This box draws heavily on DFID (2007), Stotsky (2006), and World Bank (2001). 12 The box does not provide specific advice on how to reduce gender inequality directly, but merely how the proposed development agenda laid out in the report will help reduce Malawi's gender gap. For specific recommendations the interested reader is referred to Malawi's Poverty and Vulnerability Assessment (2006). - 21 - In addition, an appreciated exchange rate raises the relative price of domestic goods and services, which includes health and education. Evidence on gender based differences in the price elasticity of demand for education in particular show that a relative price increase for education would have a disproportional large impact on reducing access of women to education. Given the underinvestment in education an appreciating currency therefore produces a loss to Malawi. Hence, a depreciating currency, which lowers the relative price of domestic goods and services, including health and education, will consequently, increase access to education for women more proportionally than for men. Improving access to finance for women through, for example, micro credit institutions, has shown that women tend to use these resources to improve their standard of living; lending to women has a greater household welfare effect, and that women tend to have a superior repayment record compared to men (see e.g. Kevane and Wydick (2001)). Hence, measures the authorities or financial institutions might contemplate putting forward to improve access could usefully take these findings into account. Improving productivity in agriculture and strengthening the linkages with agro processing while dealing with the supply bottlenecks coming from energy, trade facilitation, and transport, will create employment, improve productivity, and therefore strengthen the growth process. Evidence suggests that increasing employment in export oriented manufacturing has favoured women disproportionally (World Bank (2001)). Hence, the recommendations in this report could be expected to have a positive impact on the gender gap in Malawi. As this would lead to a larger share of resources under control of women and would therefore lead to improved spending within the household which would reinforce the growth process and strengthen Malawi's long term developmental prospects. - 22 - 2. OPPORTUNITIES IN AGRICULTURE AND AGRIBUSINESS A. INTRODUCTION 2.1 Agriculture has been Malawi's main source of growth and agricultural exports the main driver of growth. An increasing share of Malawi's agricultural produce finds its way to regional export markets, and, agricultural exports have become more diverse during this decade. Clearly, the traditional products of tobacco, sugar, and tea, account for the bulk of Malawi's exports, but non-traditional agricultural exports to regional markets have grown by an impressive average annual growth rate of 17 percent in US$ since 200113. 2.2 Malawi's success in agriculture will depend on a variety of factors, some within the sector (such as reducing barriers to regional trade, improving productivity, and updating institutional market arrangements) some general to the economy as a whole (such as macroeconomic and financial sector developments, general trade logistics, transport and power) while others are outside Malawi's sphere of influence such as global price developments of agricultural commodities. This section of the report focuses primarily on those factors that Malawi policymakers can influence, are within the agricultural sector, and which foster agricultural opportunities, through export growth and/or import substitution. 2.3 Low historical levels of agricultural growth has left Malawian farmers among the poorest and most under-capitalized in the world. Farmers still have very little irrigation, extremely low ownership of livestock, operate without animal draft or tractors, and have over-mined their soils and forest resources. Many are unable to afford fertilizers and other inputs without loans or subsidy, and their technology and productivity levels are very low. Most smallholder farmers produce insufficient food to feed their families and are therefore net buyers of maize. Even with the input subsidy program, their investment rates remain inadequate to sustain high growth. 2.4 Nevertheless, smallholders have responded impressively when offered favorable opportunities and prices. This response of smallholder farmers to price incentives will be a critical mechanism in future agricultural growth. Since 1994 when smallholders were for the first time allowed to grow tobacco on their own fields rather than as sharecroppers or workers on estates, farmers have significantly diversified their cropping pattern (see Figure 2.1 and 2.2) so that by 2008, just over half of their cropped area is used to grow maize. While the data that document these trends remain contested, the diversification trends are sufficiently robust that smallholder agriculture has changed from a predominant focus on maize production to one relying more on other crops (see box 2.1). 13 This excludes the export of Maize in 2007 to Zimbabwe by the Government of Malawi. Including the maize export increases the average annual growth rate of non-traditional exports to SADC to 40 percent. - 23 - 2.5 It is likely that in the medium term, international agricultural prices will create new opportunities for commercial crops in Malawi, but in the short-run these prospects have been affected by the global crisis. The impact of the global financial crisis has been widely felt in Africa through a deceleration of economic growth and the reduction in the willingness of some investors to take risks in African countries. In addition, the sharp falls in international commodity prices that took place in 2008 and part of 2009 have hurt mineral exporting countries such as South Africa, Zambia, Mozambique and Tanzania that are among Malawi's export markets. However in the medium term there is general optimism about the prospects for agriculture in sub-Saharan Africa (SSA): in 2001-06 agricultural value added grew at 3.6 percent for the region; and during 2002-08 grew at 5.4 percent annually Figure 2.1: Real agriculture growth by Figure 2.2: Area under cultivation by farming sector (const 94 prices) smallholders Source: NSO National Accounts 1994-2007 Source: Ministry of Agriculture 2.6 However, recent macroeconomic and agricultural policy decisions have risked deteriorating the agricultural incentives in Malawi, and of weakening the production response of farmers to price signals. These include a decision by the government to maintain a parity exchange rate to the US dollar in the face of a sharp appreciation against most of Malawi's agricultural markets: the EU, South Africa, and its neighbors Tanzania, Zambia and Mozambique. This has led to a significant appreciation of the kwacha (see section 4 and figure 4.2 for more analysis) which will deteriorate farm gate prices for both exportables and importables. Malawi farmers will be less able to compete domestically, regionally and globally at these appreciated exchange rates. In addition, the government restricted private trade in maize marketing in September 2008, an issue discussed in the food security situation below. Further adverse policy measures have included the setting of minimum prices for cotton and tobacco. - 24 - Box 2.1: Diversification of smallholder agriculture This box discusses the importance of different non-traditional products in recent growth roughly in order of their contribution to the value of smallholder agricultural output in the harvest years 2006 and 2007, valued at their 1994 prices. On average during this period, maize production contributed around 32 percent of total value of smallholder crop production. Tobacco: While occupying less than 4 percent of the smallholder area, it contributes significantly to smallholder value of output, and has been the leading diversification crop since smallholders were allowed to grow it in 1994. Sweet potatoes and Irish potatoes. While occupying less than 10 percent of cropped area, contribute about as much to the value of agricultural output as maize (around 30 percent, compared to around 15 percent in the mid 1990s). This growth benefited from the introduction of new varieties starting in the 1990s that were disease free and were brought in by the International Potato Center, also working through the International Institute of Tropical Agriculture (IITA) and the Southern Africa Regional Research Network (SARNET). Cassava occupies less that 10 percent of cropped areas. It accounts for a similar level of calorie consumption as potatoes, in terms of value it is a distant third, contributing a bit less than 14 percent to the value of crop output in 2006 and 2007. The diversification into cassava started when biological pest control became available for cassava mealybug in the late 1980s to early 1990s, and with the introduction of improved cassava varieties in the mid 1990s through IITA/SARNET. In groundnuts, Malawi had previously exploited a niche market for larger confectionary nuts, but this was lost owing to low disease resistance and high levels of aflatoxins contamination. Since the early 1990s, however, groundnuts have been growing very rapidly, and in value terms now rank just behind cassava, at a little less than ten percent of value of agricultural output. A disease resistant variety called Chalimbana 2000 and better management to reduce aflatoxin contamination have restored the potential for groundnut production and exportation in Malawi. Pulses a re often inter-cropped with maize and the area and yield figures are therefore difficult to estimate. The data nevertheless suggest that that the production value of pulses has more than quadrupled since the early 1990s. ICRISAT has brought improved Pigeon and Chickpea varieties and CIAT has for the past decade been working directly with breeders at Bunda College towards production of improved bean varieties. Livestock and associated products account for about 6 percent of the value of smallholder agricultural output. Milk production appears to have been stagnant until about 2001, when production started to rise and has since accelerated sharply in the last few years. Average production rose from about 8,000 liters/day in the 1990s to about 18,000 liters/day in 2007. There remains substantial scope for expansion, as 60 percent of dairy products are still imported. A similar situation exists in fruits and vegetables (including bananas and mangoes), which contributed about 5 percent of total value of crop production in 2006 and 2007. Rice remains a minor smallholder crop, grown mostly in areas along the lakeshore, with relatively limited recent growth. It contributes about 2 percent to total value of crop output, and occupies about 3 percent of the area under food crops. Malawi has the potential to produce and export aromatic rice varieties, such as Kilombero and Faya, which could compete favorably with rice from major producing countries, such as Thailand based on qualities other than price. Soybeans were first promoted in the late 1980s, but a price collapse in 1993/94 dampened growth until the late 1990s and early 2000s. There are 2-3 firms involved in soybean oil and soybean cake production and a couple more involved in chicken feed and infant feeding formulas. Following recent price rises, the Government introduced an export licensing scheme for soybean cakes, oil and raw soybeans to protect the domestic livestock feed industry and consumers from higher soya prices. The licensing scheme is in effect an export ban and detrimental to the growth of the sector. During the mid to late 1990s, cotton was an industry in decline, with falling production for over 15 years, largely because of low and declining international prices. However, better prices and an industry driven input and education initiative led to a significant supply response from 2004. Nevertheless, cotton still contributes less than 1 percent to the total value of crop output (in 1994 prices). The cotton sector has around 120,000 smallholder farmers, six ginning companies and three main input providers. While it is very small in terms of total agricultural production, cotton is now the fourth biggest export crop by value. There is also realistic potential to double the volume and value of cotton production in the coming years, partly because of improvements in the international price of cotton. The macadamia and cashew nut sectors have retrogressed because of the poor establishment of the trees and lack of enough throughputs to sustain modest investments in processing. - 25 - B. CHANGING CONTEXT AND PROSPECTS 2.7 In 2008, the index of international real food prices peaked at roughly 140 percent of its 1990 level while prices of individual commodities had risen even more sharply. Since their peak in 2008, prices have fallen sharply but aggregate food prices at the end of 2008 were still more than 40 percent higher than in 1990. This holds true for most of the individual commodity groups. The sharpest increases were in dairy, oils and fats, and cereal prices, while sugar prices and meat prices increased less dramatically. Figure 2.3: Nominal FAO food price Figure 2.4: Food commodity price index indices Source: FAO Source: FAO 2.8 Higher food prices are a double-edged sword as the implications for net sellers are higher incomes but it will hurt all net food purchasers. With more trade liberalization and higher price transmission elasticities, farmers can benefit from global markets, which will provide them with an enhanced capacity to invest and at the national level; higher domestic prices will make import substitution more attractive. However, higher world prices for food will hurt all net food purchasers. Since in Malawi only about 10 percent of farmers are net sellers of maize, almost everybody in Malawi is affected by higher food prices. Ivanic and Martin (2008) have estimated that, if the high food prices prevailing at the beginning of 2008 were to be fully transmitted to Malawi, rural poverty would increase by about 4 percent while urban poverty would increase by about three percent. 2.9 Yet Malawi will be a net beneficiary of higher international agricultural prices. Since 2006, Malawi has been food self-sufficient or even a minor exporter, and thus the impact of higher food prices on the balance of trade in the short run is negligible. In the medium to long run, the impact of higher prices is likely to stimulate investment, higher input use and technology adoption by the Malawian farmers, starting with those who are either net sellers of food, or sellers of commercial crops such as tobacco, cotton, or oilseeds.14 The multiplier effect from this growth would create growth in rural non- 14 Currently about 52 percent agricultural households sell some portion of output to market (GoM and World Bank, 2006). - 26 - tradeables, raise rural wages and have linkage effects to urban areas, where most industrial production and many services are still linked to farming (see Box 2.2). Medium Term Market Opportunities 2.10 With the exception of sugar, all notable cases of SSA agricultural export success have so far occurred in high value commodities such as tobacco, coffee and tea (Poulton et al, 2007). They are high value because "ideal" agro-ecological conditions or low labor costs are necessary for their production, which limits global supply and provides advantage to Malawian producers. It is therefore not surprising that tobacco, tea, sugar, cotton, coffee and groundnuts are the commodities in which Malawi has achieved its greatest success in exports to global markets, and clearly, they will remain among its most important growth opportunities in the future. 2.11 Recent studies of the prospects of commercial agriculture in SSA point to domestic and sub-regional markets for basic food staples and livestock markets as the main opportunities for SSA producers (Poulton et al, 2007; World Bank, forthcoming). Since SSA is an importer of many of these commodities, SSA producers compete in these markets at the import parity price rather than the lower export parity price. In addition, quality standards are not as high and phytosanitary barriers are generally lower than in global markets. The combined value of domestic and regional markets for food staples within SSA is considerably in excess of its total international agricultural exports (Diao et al., 2003) and will grow significantly, propelled by both population and income growth over time. SSA's demand for food staples is projected to about double by 2020. 2.12 In the livestock sector, import substitution and sub-regional trade opportunities are more important than overseas export opportunities. While the livestock sector has been a successful export sector in parts of Africa, exports have stagnated in recent years and there is little sign that Malawi will be able to maintain let alone increase its share of the explosive growth in world market demand for livestock products. However, a dairy industry is beginning to develop within the country which has been tested and strengthened through intense domestic competition and which is beginning to export some of its products. Competition in the dairy industry has taken place between an emergent formal and an entrenched informal sector and more recently among private dairy operators, which has lead to the strengthening of farm to dairy linkages, the refinement of several innovative business models and the introduction of new technology. - 27 - Box 2.2: Are the rising food prices here to stay? The rising international food prices of the past two to three years have been driven by the sharply accelerating income growth in East and South Asia, and more recently in Africa. Urbanization and global growth means demands for a larger and more varied food supply. The rapidly rising subsidies for maize and oilseeds for ethanol production in the US and Europe have sharply expanded the use of these food crops for fuel production, and will permanently link food prices to the price of energy. On the supply side, investments in agricultural R&D have declined globally as has investment in agricultural development. Finally, higher petroleum prices have permanently increased the costs of agricultural production. Because of all these trends, global grain consumption has exceeded global production in seven of the last eight years. The result has been a drawdown of stocks to critically low levels. While nominal energy prices have more than tripled since 2003, (with petroleum prices rising more than six fold), nominal food prices have "only" doubled. Clearly, the rise in energy prices is much sharper than in food prices. The energy price increases have transmitted themselves to higher fertilizer and pesticide prices, higher costs of running farm machinery, and higher freight costs for inputs and outputs. Prices of the basic raw material for nitrogenous fertilizers almost tripled, which was fully reflected in the price of Urea, and partly in the price of Calcium Ammonium Nitrate (CAN). However the former two prices have come down to their levels in the mid 1990s, while CAN still remains high. Prices of phosphate fertilizers have equally exploded and have not yet come down significantly since the global crisis emerged. On average therefore, fertilizers remain significantly more expensive than until the mid-1990s. OECD (2008) already expected prices to decline in the short run, as they indeed have as a consequence of the economic crisis. Nevertheless, for 2008-2017 the OECD project that prices will remain much higher than in the last decade. In addition, a recent IFPRI (2008) analysis using their IMPACT Model makes much longer-term projections all the way to 2050. It projects that real grain and oilseed prices will not decline from levels they reached in late 2007 and will show a modest increase through 2050. These are the first substantive analyses that seem to predict that the long-term secular decline in grain and oilseed prices may be over for good. C. THE ROLE OF REGIONAL INTEGRATION 2.13 Regional cooperation in agriculture is central to Malawi's market access, product quality and technological capability. Malawi depends on regional collaboration to upgrade infrastructure to regional and international markets; the harmonization of standards and sanitary measures; protection against plant and animal disease epidemics; and the generation of new technology15. Malawi dairy producers, for example, must rely on food sanitation tests performed in laboratories of neighboring countries when they export within the region. In addition, biotechnology research is expensive and has a large front end cost, associated with building up a critical mass of research projects. Therefore, a regional center is far superior to an underfunded, under resourced national institution and indigenous scientific capacity requires trained people. While Malawi needs to have its own agricultural education and training institutions, for advanced training it will have to rely on regional centers of excellence, which have critical mass and necessary financial support. 2.14 Gains from regional agricultural trade depends critically on the progress of regional institutions. Malawi is a member of both COMESA and SADC. While SADC 15 The SADC countries have recently created a new sub-regional organization for agricultural technology which should figure prominently in Malawi's future agricultural development efforts. - 28 - has just established a common market, COMESA is in the process of establishing a customs union with planned implementation late in 2008. Trade within the SADC region currently accounts for about 60 percent of Malawi's imports and 31 percent of the country's exports. Trade to COMESA countries accounts for about 10 percent of imports and 12 percent of the nation's exports. These two trade areas, along with the East African Community (EAC) of Kenya, Uganda and Tanzania, have agreed to harmonize their tariffs and other trade regulations, thus eliminating the main cost of overlapping trade zones. 2.15 However, the potential for regional trade is impeded by the risk of government intervention, and differences in rules of origin, quality and product standards. Trade in food staples has long been discouraged by national policies that place a high priority on food self-sufficiency. In 2008, Malawi banned all exports of maize, and other SADC and COMESA countries have also intervened in these markets. Consequently, one of the biggest impediments to large-scale investment in regional trading capability remains the unpredictable behavior of governments in imposing export bans whenever they fear shortages in their own markets. The ex-ante risk of government intervention has slowed down the development of the nascent commodity exchanges that Malawi has put in place for the sub-regional and for its domestic markets: Agricultural Commodity Exchange (ACE), and Malawi Agricultural Commodity Exchange (MACE). 2.16 Agreements have been reached in the SADC region to harmonize regional seed regulations and establish a regional variety registration system. This aims to facilitate trade of seed across the region through common seed inspection and certification standards across all countries in the region. In addition, SADC countries have collectively agreed on a single standard protocol for seed quarantine and phytosanitary controls. 2.17 While much progress has been made to set up the institutions to deal with issues of regional integration, most of them have remained underfunded. The main reason for this is that the regional institutions produce regional public goods, and are therefore subject to the familiar free rider problem of financing. Except for the largest countries, which have an incentive to supply themselves with these regional public goods, countries have tried to benefit from the investment of others, or hoped for donors to step into the breach. Neither of these expectations is likely to lead to a solution of the underfunding problem and SADC countries need to step up to the challenge. Malawi has a particularly compelling need to be active in this area and exercise a leadership role. D. LESSONS FROM AGRICULTURAL VALUE CHAINS 2.18 Value chain analysis (see Box 2.3) reveals Malawi to have competitive advantage in burley tobacco and cotton but not in maize or rice. The analysis indicates that the advantage in tobacco and cotton derives mainly from low labor cost rather than from high productivity. Malawi should promote maize and rice production mainly as an import substitution strategy and should not aim to become a regular maize exporter. However, export possibilities for rice may be enhanced by focusing on - 29 - aromatic varieties for niche regional markets, and by improving productivity and efficiency of the value chains especially at the assembly and processing levels (see table 2.1). Table 2.1: Maize, rice, cotton, and tobacco parity price comparison Maize is competitive at import parity but not at export Rice with current cost and prevailing prices is not parity competitive Rain fed Irrigated Polished Rice Final shipment USD 261 per USD 257 per Final shipment value USD 570 per ton value (excluding ton ton (excluding international MK 79.8 per kg international MK 37 per kg MK 36 per transport) transport) kg USD 450 per ton Import parity USD 379 per ton Import parity (Thailand) MK 63.0 per kg (RSA) MK 53 per kg Export parity USD 269 per ton Export parity (Zambia) USD 480 per ton (Zimbabwe) MK 37.6 per kg MK 67.2 per kg Cotton is competitive but at the margin, partly because Tobacco is competitive at all levels. of high minimum prices set by GoM and could be smallholder- smallholder- improved by aiming for a better ginning out turn low yield high yield Family Low Family High Final shipment Final shipment USD 3,573 USD 3,786 per USD 1353 per USD 1360 Value value (including (boxed T&T tobacco per ton ton MT lint at gin per MT lint MK 500.2 MK 530.4 per international and including transport) gate at gin gate per kg kg international transport) USD 1012 (Dec. 06) Export parity Export Parity USD 3,650 USD 3,800 USD 1163 (Oct. 07) (cif Northern MK 511.0 MK 532.0 per USD 1361 (June 08) Europe) per kg kg Source: Volume II: Value chain study 2.19 Malawi's competitiveness is undermined by low productivity, high cost of inputs, high cost of transport, and high costs along the value chain. Inorganic fertilizer and other agricultural inputs are expensive mainly due to high international and domestic transport costs as well as high trader margins because of transaction risks associated with agricultural input trading. The high cost of inputs further leads to low uptake of fertilizer and improved seed. Overtime, low uptake of improved technology results in under-capitalization in the sector, which tends to arrest any more technological and institutional innovations. The majority of Malawi's smallholder farmers are trapped in this situation, in which, because of risk aversion, they choose to operate at a sub- optimal subsistence level. 2.20 The analysis shows that farm gate prices in Malawi, contrary to popular opinion, are often higher than in other countries, and there is little scope for further increases via minimum prices. Improvement in total farm income would need to come from productivity improvements and a lowering of the production costs in particular of inputs. Interventions that aim at setting minimum prices may be counter-productive, because apart from dampening trade competitiveness, they threaten the provision of - 30 - important services that are provided by players within the value chain. For example, the minimum price set in cotton has seriously eroded margins for other participants within the value chain and one can expect that, this will lead to a reduction in spending on extension services and access to inputs. Box 2.3: Summary of the main uses and methodology of value chain analysis Value chain analysis determines the private cost and profitability at different stages of the value chain. It can assist policy makers to understand the incentives that market participants in agricultural value chains face, and can identify the sources of high total costs. By comparing results to other countries in the region and beyond, Zambia, Mozambique, Cameroon, Nigeria, Brazil, and Thailand were similar exercises have been undertaken, the analysis provides insights in Malawi's competitiveness in regional and global markets i.e. what are Malawi's best opportunities for exports or import substitution. Value chain analysis encompasses all of the factors of production including land, labor, capital, technology, and inputs as well as all economic activities including input supply, production, transformation, handling, transport, marketing, and distribution necessary to create, sell, and deliver a product to a certain destination. The main stages of the value chain are illustrated below. Input Farm Logistics / Production Assembly Processing Distribution Supply The methodology ensures that all inputs and outputs carry forward their inherited value from the previous stage. The accumulated costs at different stages are the key determinant of trade competitiveness, which depends on the efficiency of input supply, farm production, assembly, processing, and logistics up to final delivery point where the good competes internationally as an export or import substitute. The methodology not only shows if the country is internationally competitive, but also helps identify key stages where costs could most effectively be reduced as a strategy for agriculture growth. Total costs are measured in terms of Domestic Value Added (DVA) and Shipment Value (SV), which constitute the main value chain indicators: Domestic Value Added (DVA) = Domestic costs and mark-ups [1] + Official duties and tax Shipment Value (SV) = Domestic Value Added [2] + Foreign components For any given commodity, trade competitiveness is determined by comparing SV at the final market with an equivalent parity price (either a FOB price for exports or CIF price for import substitutes). By looking at the build-up of SV (and DVA) from stage to stage, the methodology therefore reveals the competitiveness of individual participants. If one stage accounts for a disproportionately large share of final shipment value, interventions focused on that part of the value chain likely also have a disproportionately large impact on the overall competitiveness of the chain 2.21 Productivity gains are probably more important than other cost savings along the value chain; however, there are potential gains from reducing the cost of inputs (see Box 2.4 for a simulation). First, interventions aimed at reducing the - 31 - transport costs, such as for example, rationalization of levies on fuel, reduction in domestic taxes and duties, are critical to reducing fertilizer costs in Malawi, thereby raising competitiveness. Secondly, there is need to consider implementing improvements in the management of fertilizer and inputs supply chains such as timely procurement. Thirdly, there is need for appropriate interventions to improve the development of private traders, thereby enhancing the structure of commodity markets, to ensure that some of the margins that are captured by the traders, in both the input and output market, could be passed on to the producers improving farm-gate prices. 2.22 The low crop yields in Malawi can be attributed to the low levels of fertilizer intensity relative to other countries. In land abundant countries such as Mozambique, the yield difference could also be attributed to differences in the natural fertility of the soils. As shown in Table 2.2, Malawi's costs for fertilizer and other agricultural inputs are generally higher than most comparator countries. 2.23 Fertilizer is the single most important cost component in the production of most arable crops. Fertilizer accounts for 20 ­ 50 percent of the farm-level costs of production for all crops considered in the analysis. In 2007, Malawi's fertilizer cost, estimated at US$ 728.57 per ton, was almost twice as expensive compared to Mozambique, Nigeria, Brazil and Thailand (See Table 2.2). Although Malawi does not impose any direct tax on fertilizer in the form of customs duty or VAT, the analysis reveals that domestic taxes still accounted for around 8 percent of total farm gate shipment value. Specifically, these taxes include VAT on clearing fees, fuel taxes, trading licenses, and profit tax charged on dealer mark-ups. As a strategy to improve agriculture competitiveness, therefore, there may be some scope to reduce these specific charges pertaining to fertilizer import and trade. Table 2.2: Comparison of input price build-up between Malawi and other countries Basal General General Casual Road Fertilizer Herbicide Insecticide Labor Freight MT Liter 1ha cotton 1 day FAM 1MT per Km Malawi (2006) 532.82 - - - - Malawi (2007) 728.57 7.14 14.29 0.71 0.129 Mozambique (2006) 377.60 5.00 25.02 0.48 0.070 Zambia (2007) 540.00 5.00 29.28 1.25 0.100 Nigeria (2006) 295.45 5.15 8.56 2.65 0.053 Cameroon (north) (2007) 702.19 7.00 45.83 3.13 0.115 Brazil (2006) 301.40 3.49 30.18 20.88 0.064 Thailand (2006) 346.60 3.59 n/a n/a 0.027 Notes: Fertilizer price for most common blend in each country; Herbicide price for paraquat, round-up or similar product used for general weed burn-down. Insecticides for cotton = 1ha FAM-high in Malawi and Cameroon; 1ha FAM in Nigeria; 1ha ECF in Mozambique, and Zambia; and 1ha LCF in Brazil. Actual quantities of insecticide will vary. Transport costs in Cameroon for direct shipping method (can go to USD 0.54 per MT per km if use informal roadside freight). Slightly lower rates prevail in southern Cameroon because of proximity to refinery. Input prices in Cameroon are 10-15 percent higher in north than in the south due to transport costs and other fees. - 32 - Box 2.4: Three ways to improve farm-level returns: a simulation The impact of three simulations of changes in key factors affecting Malawi's agricultural competiveness are evaluated. These are (i) a technological improvement leading to a 10 percent increase in yield and a percent improvement in GOT in cotton; (ii) a reduction in farm-level input costs through a 5 percent change in face value of the fertilizer and seed voucher; and a reduction in trader margins and marketing cost through a five percent stepwise reduction in transport cost and trader margins. The impact of yield improvements. The impact of yield increases as a result of technological improvements on producers' net profit and farm-level competitiveness are shown in Table 1. The results generally indicate that, other things being equal, a 10 percent improvement in yield results in raising producers' net profit by as much as 35-55 percent in hybrid maize, about 14-25 percent in cotton, 10-13 percent in rice and between 13-17 percent in burley tobacco. The yield improvements also result in an improvement in farm-level competitiveness ranging from 9 to 22 percent. These results come from simple simulations that assume linearity and do not consider the general equilibrium effects of yield changes on output prices. Perhaps, if such market effects were considered through general equilibrium analysis, the magnitude of the positive impacts would somehow be reduced. However, the results imply that one way to improve farm profits and competitiveness is via technological improvements that improve crop yields. Impact of agricultural input subsidy. This particular simulation is implemented only on the crops that have been included in the current input subsidy program i.e. low input maize, low input burley tobacco and low input cotton. The cost of fertilizer and seed are changed to reflect the value of the subsidy voucher i.e. market price of fertilizer and seed minus the amount paid by the farmer to redeem the voucher. This year, this amount is equivalent to MK800 (about US$ 6) for a 50kg bag of basal and top-dressing fertilizer and zero in the case of hybrid maize and cotton seed. The results shown in Table 1 indicate considerably high gains in farm-level net profits and competitiveness because the subsidy increases farmers' application of fertilizer and use of improved seed at a cheaper cost. All these invariably enhance profits and private competitiveness via reduction in the costs. Of course they do not increase the competitiveness of Malawi as a whole because the costs are still borne by the government. Impact of a reduction in transport costs and trader margins. Trader margins are costs that are incurred mostly at the assembly level. As such, we assume a direct pass-through of the gains in transport cost and trader margin reductions to the farm-gate prices. This implies that if we assume a percentage reduction in the trader margins, such a change is applied directly on the farm-gate price. However, since the transport cost is incurred directly by the farmer, the cost reduction is applied directly to the transport parameter in the farm-level crop budget. The results are as shown in Table 1. Reduction in trader margins considerably increases the producer net profit because it raises the producer price. The impact is directly proportional to the importance of the trader margin in the marketing of respective crops. Where the trader margin is quite high as a proportion of the producer price, as in the case of maize, the impact on producer net profit is also very high. Similarly, a reduction in transport costs improves competitiveness more particularly in commodities that have to be transported from the farm to the market such as cotton and tobacco. T able 1: Im pact of various key cost factors on farmers' net profit and shipm ent value crop Impact on producer's net profit Impact on farm-level shipment value (% from base) (% from base) Base Yield Increased Reduced Base level Yield Increased Reduced level improvement subsidy margins improvement subsidy margins Mai ze FAM-LOW 65.5 55 17 23 158.3 9 -43 n/a FAM-HIGH 35 37 183.9 5 -0.3 Cotton FAM-LOW 173.8 25 3 11 232.7 13 -2 -0.8 FAM-HIGH 14 18 247.7 8 -0.5 Rice FAM-LOW 119.1 10 19 163.1 17 n/a FAM-HIGH 112.0 13 12 153.9 12 n/a Tobacco FAM-LOW 369.9 17 60 14 759.1 22 -29 -3.1 FAM-HIGH 13 12 736.7 15 -2.3 - 33 - 2.24 The cost of fertilizer is mainly due to the procurement price and the high international and domestic transports costs, estimated at 26 percent and 6 percent, respectively. Transport cost is therefore a key determinant of Malawi's agricultural competitiveness, given that most agricultural commodities are primary or semi-processed, and are, therefore, bulky and attract high transport costs, on average. Furthermore, domestic dealer mark-ups are also quite high in Malawi, estimated at 16 percent compared to 5 percent or less in Zambia and Mozambique. The wholesaler and retailer mark-ups together contributed nearly 20 percent to the fertilizer price build-up. 2.25 The factors contributing to the high fuel price in Malawi include the multiple levels of levies and surcharges imposed by various agencies including Malawi Energy Regulatory Authority, Road Fund Administration, Bureau of Standard, etc., on the FOB price of fuel. These levies constitute about 40 percent of the retail pump price. The second and the third major factors are the taxes and duties collected by Ministry of Finance, as well as the profit margin of wholesale and retailers, representing about 13 percent and 12 percent of the retail price, respectively. The insurance and handling charges only add up to about 6 percent of the retail pump price. As a result, Malawi has one of the highest fuel prices in the world.16 A rationalization of the levies and taxes on fuel are likely to lead into significant reductions in transport costs, which is likely to improve trade competitiveness in general, and agricultural competitiveness in particular. Any shortfall in resources to maintain and expand the road network should be compensated through a direct subvention from general resources.17 Figure 2.5: Maize, build-up of financial costs along the value chain Maize, FAM-low Build-up of Financial Costs (USD per MT at import parity) 400 350 300 250 200 150 100 50 0 Seed & fertilizer Farm labor Marketing Profit to farmer Packing & Transportation Loading & Trader's profit costs and at roadside storage overheads at import parity overheads Farm-level Assembly into mill Source: Volume II: Value Chain Study 2.26 Proportionately high margins accrue to traders at the assembly level, especially in maize, rice and tobacco. For maize and rice, there are high margins at the assembly and processors/millers, respectively. For tobacco, there are high transport costs at assembly level, which emanate primarily from a somewhat less efficient marketing 16 See GTZ international fuel prices, 6th edition. 17 See section IV for a discussion on the fiscal framework supportive of growth. - 34 - system. The findings for maize where trader margins and other marketing costs along the chain are very high proportional to the volume of grain handled by traders relative to producers and illustrated in figure 2.5. Maize traders' profit margin (per metric ton) at import parity are equal to approximately 30 percent of import parity price, which is very high compared to the profit that accrues to the producer. Given that on average, traders handle more volume than producers the traders' margin is exceptionally high. 2.27 The higher trader margins are attributed to thin maize markets, resulting mainly from the high transaction risks in the remote areas that often limit competition. As such only, few traders with transport facilities are able to reach remote areas where they reap monopolistic rents. Through appropriate interventions that improve the development of private traders, thereby enhancing the structure of maize and rice markets, it is possible that some of the margins that are captured by the traders could be passed on to the producers thereby improving the farm-gate prices. 2.28 Irrigation can significantly improve income of maize and rice farmers, but does not contribute to improved trade competitiveness due to its relatively high cost. While private irrigation via treadle and motor pumps is an important option for expanding irrigation, investments in small, medium scale irrigation systems remain a mixed story for Malawi. An important bottleneck identified is the management of these systems and recent activities in this area are making an effort to have a larger role in the management of these systems by farmer organizations, through the formation and capacity building of Water User Associations. Furthermore, farmers that engage in irrigation are strongly encouraged and facilitated to produce high value crops and supply organized chains through contract farming or out-grower arrangements. In this way, farmers will be able to sustain the operations and maintenance of the irrigation schemes. E. AGRIBUSINESS 2.29 Agribusiness investment is concentrated in relatively few sectors, where comparative advantage for locally produced commodities has long been established. These sectors include: i) tobacco, ii) tea, iii) sugar, and iv) cotton. Other sectors are of relatively recent significance to the national economy. Private investment in these sectors has helped to open new markets, to develop new products and substitute modern industrial processes for more traditional modes of manufacture. These include: i) paprika, ii) poultry, iii) dairy, and iv) bio fuels. 2.30 The competitiveness of agribusiness often depends on highly specialized regulatory framework applied to a particular subsector. Much of the competitive advantage, which Malawi enjoys is due to the regulatory regimes and the market institutions which have developed for specific subsectors. Hence, much of the work of assuring continued global competitiveness is the work of reforming and adapting existing institutional arrangements to better suit changing and more demanding global market conditions. 2.31 However, in some cases, the regulatory environment has degraded and the government has exercised arbitrary intervention in price and trade controls. For - 35 - example, regulatory controls have broken down in the cotton sector, and the government set minimum prices well in excess of prevailing prices in neighboring countries leading the major buyers to threaten to pull out of the market. As a result, investment sharply declined as did the provision of private extension support. 2.32 A new legislative compact is required in many sub-sectors that commit the government to rule based regulation, which is systematic, predicable and based on facts and market intelligence. The key objective should be the adoption of just enough regulation to assure that each sector is able to enhance its response time to competitive challenges and thus through a continuous upgrading of its value composition can extend it competitive advantages. Regulation that is more heavy-handed may be required in some sectors e.g. tobacco versus relatively light regulation for the dairy sector. However, all agribusiness sectors analyzed require a regulatory reconstitution under which the government itself commits to due process, to transparent decision making and to systemic regulation, which is rule based, as contrasted with arbitrary decision making, which is simply ad hoc and opportunistic. This re-chartering will require a thorough and open dialogue between industry representatives and government regulators and legislators. Tobacco, cotton, and dairy sectors 2.33 In the tobacco sector, an established rules based regulatory framework and the partial autonomy of a knowledgeable regulator (the Tobacco Control Commission) is perceived as a positive influence on private investment and a source of competitive advantage over other regional tobacco producing countries. The Tobacco Auction system is tested and proved but it is also slow to adapt and slow to respond to the needs of global cigarette manufactures for more precisely specified products. 2.34 In the cotton sector, the breakdown of regulatory controls, arbitrary government intervention in prices and the lack of a formal market institution greatly increases risk to investors. The cotton market operates at the forward buying points set up by each of the ginners. It is inefficient as each ginner must provide its own forward buying station in order to compete and it does not effectively incentivize increased quality and productivity though the smart application of inputs. 2.35 In the dairy sector, a lack of regulatory oversight poses no fundamental threat to investors. Neither does a lack of formal market mechanisms deter new market entrants. In this sector, an important institutional benefit exists in the intermediary organizations, which have emerged to assemble milk and to negotiate off take agreements in behalf of farm level organizations. Three regional assemblers have been formed as legal associations in Malawi and they in turn contract for milk sales with dairies and enforce contract terms on their members. Retail food chains and dairy wholesalers offer further market development assistance on the "sell end" of dairy chain. An extensive national coverage of food retailers and dairy wholesalers effectively lowers barriers for new entrants and stimulated competition among dairy industry incumbents. - 36 - Contract Farming 2.36 A key constraint for agribusiness is the problem of how to engage smallholder farmers collectively, verses the high costs of engaging them individually. The resulting coordination failures limit the availability of farm produce and cause its quality generally to be poor. 2.37 It will be important to focus on legitimizing modes of contract farming that encourage long-term investment and productivity growth. By strengthening relationships between sellers and buyers, contracting provides a framework for improving productivity levels and the reliability of commodity supply. When contracts work well, farmers gain improved access to inputs, agricultural credit and extension advice. A main difficulty encountered in contracting arrangements is the negotiation about product price. Agreement has to be reached on how to share the costs of inputs and extension support, and the gains derived from commodity sales. These costs and returns are embedded in the farm-gate or delivery price of the contracted product. 2.38 Side marketing undermines the success of contract farming. Productivity gains are more likely to be achieved in relatively closed marketing systems where side marketing is less likely e.g. sugar and tea. In open marketing systems, as exist for cotton and paprika, contract arrangements quickly break down when competing enterprises offer timely buying or marginally better prices to farmers already contracted by another buyer. In the process, the farmer often sacrifices the future opportunity to achieve larger gains in productivity growth and income by failing to repay an input loan provided by the original contractor and to supply the contracted produce. Trust and confidence among buyers to not engage in side purchasing is therefore a necessary condition for success of contract farming. 2.39 As does the tendency for government to intervene and set commodity prices. This tendency has particularly threatened the viability of contract farming in cotton. In 2008, the government mandated a price of K65 well above prevailing prices in Zambia and Mozambique leading the major buyers to threaten to pull out of the market. The immediate decision was to sharply cut investments in the provision of inputs on credit and private extension support. In effect, farmers may receive a higher price, but productivity and crop quality gains essential for Malawian farmers to remain competitive on international markets are under threat. 2.40 Regulatory frameworks or collectively adopted institutional arrangements need to be strengthened in order to facilitate supply chain performance and to allow long-term relationships to develop between famers and other stakeholders in sub- sector supply chains. It is extremely important to find practical ways to enforce discipline on farm level organizations, to hold them responsible for repaying credits which are advanced to them for inputs and to impose penalties and/or to create incentives to enforce quality controls. These disciplines are material to moving from the production of farm commodities to the production of value-differentiated food products. Commercial experiments attempted in the cotton, tobacco and milk sectors have met with mix success. - 37 - Box 2.5: General constraints to growth of the agribusiness sector This box evaluates in addition to sector specific constraints discussed in this section, how the general growth constraints identified in section I affect the opportunities of agribusiness sector to expand. Transport across national borders poses a substantial barrier to trade. Poor infrastructure, congestion at border and river crossings as well as bureaucratic processing and corruption all adversely affect cross border movements. Both the cost and the timeliness of transport are issues for the tobacco and cotton sectors, as they affect containerized exports via Mozambican, Tanzanian or South African ports (see also section III). Still, informal exports for consumer food products (such as long-life milk) are flourishing to the extent that dairy processors are seriously investigating their own investment in regional trade opportunities. The lack of reliable electricity supply (effecting both quality and availability) has become a major obstacle to investment in new processing facilities. This is a problem, which adversely affects all agribusiness sectors. Most processors have a back up generating capacity because of the unreliability of ESCOM power. However, an emergency conversion to backup power still imposes a increase in costs on agribusinesses. Water supply, particularly in Blantyre poses another serious challenge to food processors, including dairies. One of the Blantyre based diaries has overcome this constraint by investing in its own underground reservoir. The company felt that this investment was necessary even though its plant is located less than 1 km from the main dam and water board headquarters in Blantyre. A particularly challenging problem, which most agribusinesses face, is securing a sufficient number of skilled workers. In the sector, expatriates fill many management positions, including in finance and accounting, operations management and marketing. Universities and technical institutions in Malawi are not effectively addressing the skills gap, which is building in the agribusiness sector. Most companies deal with this challenge by training up their own work forces from a very rudimentary base, for example, by offering instruction in basic hygienic practices and in simple food handling methods. Another across the-board challenge is securing foreign exchange. The problem of limited foreign exchange affects the entire farm to market chain. For example, the rationing of foreign exchange directly affects the sectors ability to expand, as new plant equipment requires access to foreign exchange. The second largest expense item for all industry participants is imported packaging materials, which again requires regular and timely access to foreign exchange. Other product critical inputs are equally essential for extending their product lines and for diversifying into new product categories. As a result, some agribusiness have lost open terms with their suppliers and face added costs when they reverted to more expensive trade finance instruments High domestic interest rates on loans severely limit the ability of Malawi's agribusiness sector from growing to its full potential. The capital structure constraint, associated with a lack of debt financing limits participation in the sector to two main kinds of enterprises: large multinational companies and privately funded companies, able to draw on informal (often family) linkages to have access to capital. 2.41 There is an important role for the public sector in strengthening farmer organizations, market information and technology access: · Farmers organisations. The government can improve the capacity of farmers to negotiate contracting arrangements by strengthening farmer organizations such that they operate effectively as a business. This has the aim that these farm level organizations can become progressively credit worthy, can negotiate contracts effectively using market research and expert advice, can be relied upon to satisfy the terms of contracts in good faith, and will respond to contract incentives designed to improve productivity and the ultimate consumer value of farm production. · Market information. The government can assist all partners in the value chain in price negotiations by providing information about levels and trends in international - 38 - prices, commodity quality differentials, and market prospects. This information needs to be combined, however, with a good understanding of how international prices translate into domestic prices in the context of high marketing costs and trade risks. · Technology access. The government should aim to speed productivity growth by assuring open access to the best available technology on the international market. Public private partnerships may be pursued in testing and disseminating these technologies. F. MAIZE, INPUT SUBSIDIES, AND FOOD SECURITY 2.42 Despite declining relative cropped area, maize remains the single most important source of food security for farmers and public policy. High population density, uncertain rainfall and being landlocked have made ensuring maize security the most challenging issue for Malawian policy makers. In 2007, Malawi's maize production reached an estimated record of 3.4 million metric tons18, compared to the previous production maximum of 1.8 million in 1999/2000, and almost three-fold the drought affected harvest level of 2005. This harvest encouraged the country to export over 300,000 tons of grain to neighbouring Zimbabwe, and another 20,000 tons of grain to Swaziland and Lesotho. This record stands in sharp contrast to the need for 350,000 tons of maize imports in 2005. 2.43 The record levels of maize production can be attributed to favourable rainfall, and the widespread use of fertilizer and improved seed made possible by the input subsidy program. In 2006/07, over 1.5 million households each received vouchers to purchase 50 kg of NPK (Nitrogen, Phosphorous, Potassium) and 50 kg of urea for their maize crop. Each households also received a voucher allowing the purchase of either 2 kg of hybrid maize seed or 4 kg of open pollinated maize seed. Another 400,000 farmers received coupons for maize seed without fertilizer and almost one million farmers received flexible coupons that could be redeemed for any grain or legume seed available in the market. Average maize yields were reported at record levels of about 2.2 tons per hectare. 2.44 The success of the input subsidy program has led to it being established as a recurrent expenditure program in the pursuit of national food security. The payoff to this program seems clear when the costs of the subsidies are compared with the price of imported maize. Due to the high costs of transport from South Africa, import parity prices are commonly US$100 to US$150 above domestic retail prices. When there is severe drought across southern Africa, clogged transport systems raise the costs of import dependence even further. Transport costs alone can rise as high as US$180 per tonne. 18 While average maize yields are consistent with regional trends, many observers doubt that maize harvests were as large as these figures suggest as there are apparently still overestimations in total output due to consistent overestimation of the area under hybrids and composite varieties. The magnitude of over estimation is not known, but some feel that it could be in excess of 20 percent. - 39 - 2.45 However, the input subsidy program needs to be targeted to non-cash crop producers with the most severe cash constraints to achieve the most valued outcomes. Over the past few years, while maintaining its focus on improving food security, the input subsidy program has expanded to encompass fertilizer subsidies for tobacco, seed and agro-chemical subsidies for cotton, and subsidies for grain storage chemicals and for coffee and tea in 2008-09. These have become a significant part (about 20 percent) of the overall program costs, which risk displacing commercial input purchases. Most households growing cash crops such as tobacco and cotton are relatively wealthy and would be expected to purchase seed, fertilizer and agro-chemicals through contract farming or credit supply arrangements. 2.46 Malawi spends almost all its fiscal resources allocated to agriculture on the fertilizer subsidy. This leaves little or no resources to support complementary investment programs in the sector. In 2007/08 Malawi allocated 12.9 percent of its national budget to agriculture, this increased to 14 percent in FY08/09 of which approximately 90 percent was allocated to the fertilizer and seed subsidy. The sharp increase was mainly a consequence of higher fertilizer prices in 2008. However, it is clear that while the government's Agricultural Development Program (ADP) is consistent with a broader approach, public expenditures, policy, and the attention of the Ministry of Agriculture remain almost exclusively focused on maize and inputs, thereby ignoring the much wider set of opportunities in the recent past, the present, and the future. 2.47 Malawi should rebalance its agricultural public expenditures. Even though the recent decline in international prices for fertilizer, the Government could bring down the cost of the fertilizer and seed program by better targeting of the program to food deficit households as well improving the timing of procurement (see section 2. E). The created fiscal space could then be used to expand expenditure programs of those components of the ADP that focus on crop diversification, small stock production, commodity risk management, and other productivity enhancing investments such as domestic and regional research institutions and irrigation. This will improve the farmers' productivity, will make him or her more competitive and ultimately increase income from farming. The maize market 2.48 The combination of rainfall variability and thin maize markets has resulted in some of the most volatile maize prices in SSA. Small changes in trade volume can have a large impact on grain prices, and the recent production success has not translated into greater price stability. When maize shortages appeared in the national market in January 2008, just 8 months after having reported a record harvest in 2007, questions arose about the quality of national maize production estimates. Evidence suggests that the unexpectedly sharp price increases on local markets were aggravated by public sector purchases of maize for the strategic grain reserve and ADMARC reserves, which bid up maize prices in competition with private traders. Speculation increased about possible shortages in the market. Ultimately, government introduced controlled prices and banned maize trade by large traders. - 40 - 2.49 Price and trade controls have instead undermined investment in maize trading which will reduce the supply of maize marketed. Greater trust in markets and use of available risk management instruments would probably have led to significantly lower consumer prices in 2008/09. The controls have instead undermined incentives to invest in expanding commercial trade and grain storage and uncertainty persists about the form and logic of future market interventions. Thus while the subsidy policy has been a success in terms of production, it has not achieved the desired increases in the household food security of all urban and rural households who are net purchasers of maize, and still confronted record high prices. 2.50 Maize markets could be deepened (and price stability improved) by the adoption of a warehouse receipts system. Private traders handle the majority of Malawi's maize trade: thousands of small-scale entrepreneurs buying, storing and selling small stocks of grain. Two main problems undermine these efforts: the high costs and limited availability of trading capital and the risks of high losses of grain in storage. Both access to capital and the reliability of grain storage can be enhanced through the adoption of a warehouse receipts system. With such a system, small and large traders could use the warehouse receipt as collateral for their loans. In addition, this would target improved access to and management of warehouse space in the country. The establishment of warehouse receipts would also add transparency to the grain market. 2.51 The government should hedge the risk of high maize prices with a physical call option when early signs signal the possibility of drought. Following the 2004/05 season drought, the government saved about US$ 60 per ton by establishing a physical call option for imports of maize. At a time when regional grain markets and transport facilities were in short supply, the call option locked in the delivery of 50,000t of grain at a specified strike price. Once this price was publicly announced, the private sector knew the point at which the government would intervene on national markets, thus marking a price ceiling for local markets. In 2005, prices were still rising at the time of contract expiry and so maize was imported at prices agreed much earlier in the year. 2.52 Conversely, in good years, a complementary repurchase agreement (REPO) would allow the government to support post harvest maize prices. Via the REPO agreement, the government purchases the right of first refusal to a stock of grain originating from domestic farmers and held in the country. It thereby provides a strong incentive for private traders to invest in grain stocks that are purchased after a favourable harvest when local prices are low. If the government does not exercise the REPO agreement, the trader can sell the maize on the domestic market or export it. 2.53 The recent purchase of drought insurance in the international weather derivative market also extends this risk management portfolio. The weather derivative offers Malawi a payment if rainfall drops below an agreed threshold necessary for a good harvest. The size of the payment is determined by the severity of the drought. This sort of `insurance' payment will not provide enough money to pay for the full quantity of grain imports likely to be required in the event of severe drought. Nevertheless, the payout could be used to fund a call option locking in regional commitments on maize imports. - 41 - 3. IMPROVING CONNECTIVITY AND ENERGY PROVISION 3.1 Section one of this report identifies that the reliability of power supply deters new investments in manufacturing, including agro processing, and mining. It also identifies and confirms in section 2 that efficiency improvements in transport and the provision of trade logistic services can increase farmgate prices and improve Malawi's competitiveness significantly. Consequently, this section takes a closer look at what the obstacles are in transport and trade logistics (section 3A) and power (section 3C) and how these obstacles can overcome. As resources are generally scarce, spatial analysis and techniques are used to show where public investments say in feeder roads and power, can bring the highest returns to Malawi. These issues are explored in Section 3B. A. IMPROVING TRANSPORT AND TRADE LOGISTICS 3.2 Being landlocked forces Malawi to trade in transit through neighboring countries to access global markets. To some extent, Malawi is better off than many other landlocked countries as it is served by four competing transport corridors to the sea. Malawian traders can thus choose their preferred route depending on the trade- off between cost and time reliability. Each of the corridors has advantages and disadvantages attracting different products depending on their commitments and on the cost constraints. While trading through Durban is more reliable, it is also the most expensive. Railway freight to Nacala, on the other hand, is the cheapest route, although only marginally, to the sea but also the least consistent, while Beira offers mid levels of price and reliability. Most of Malawi's trade is currently routed through Beira and secondly through Durban. Table 3.1: Distance to the ports and estimated trade volumes From From Estimated Total Lilongwe Blantyre Trade Volume Trade Beira 1,194 km 846 km 950,000 t Tobacco, general cargo and most fuel Durban 2,678 km 2,323 km 300,000 t Time-sensitive trade Nacala* 1,085 km (994) 959 km ( 798) 215,000 t Sugar exports, fertilizer and bulk imports Dar-es-Salaam 1,667 km 2,031 km marginal Fuel and fertilizer Note: * Railway distance in parenthesis. Source: Professional bodies, World Bank staff estimates. 3.3 Malawi's trade volumes are relatively small and total annual trade outside direct neighbors is estimated at around 2 million tons. There are no comprehensive statistics of the volume of trade for Malawi except through the professional bodies of the main traded commodities such as sugar, fertilizer and fuel. Combining these different sources with estimates of truck traffic at the borders gives a total trade volume (imports plus exports) of around 2 million tons. Were Malawi a coastal country, this volume of trade would justify a single relatively small port most likely served by feeder services from a regional hub. Even taking into account current volumes of trade from adjacent areas in Mozambique and Zambia does not alter this figure significantly. - 42 - 3.4 The pattern of corridor use was greatly disrupted by the war in Mozambique. Mozambican ports were the traditional outlets for Malawian trade because of their proximity and the efficient railway operations on the Sena line from Beira. During the civil war, the Beira and Nacala routes collapsed, and Durban and Dar-es-Salaam became the gateways for Malawian trade. Currently, after 14 years of peace in Mozambique and the increase in transportation cost (principally due to the increase in fuel) Malawi can again transit through the Mozambican ports. However, trade patterns and thus also logistic patterns have changed as South Africa has become the prime commercial partner of Malawi. Furthermore, the port and shipping industry is increasingly structured in hubs and spokes, which re-enforces the role of Durban as a hub and Nacala and Beira as spokes. Table 3.2: Reliability of ports and corridor transit times Transport Infrastructure Port Port mode condition reliability Port Delay Transit time Beira Road Good/fair Medium 2 weeks 2-3 days Durban Road Good High 1 day One week Nacala Railway Poor Low >3 weeks unpredictable Dar-es-Salaam Road Good /fair Medium 4 weeks Source: Interviews with freight users, July 2008 3.5 The pattern of corridor usage reveals a preference among traders for reliability. In particular, the unreliability of the Nacala corridor both in terms of its transit time and port delays has caused traders to reroute through more predictable and costly routes (see Table 3.2). The logistics performance index (World Bank, 2007) ranks Mozambique and Tanzania as below Malawi - and significantly below South Africa - both in terms of logistics performance and logistics competence. This is confirmed by interviews with port users who report that delays at Beira, Nacala and Dar-es-Salaam can easily exceed two weeks whereas delays at Durban are reasonable but time in transit is longer. With the exception of the port of Durban, which handles around 20 percent of freight, the elapsed time for a cargo from Malawi to being loaded on-board ship is more affected by the time it takes to initiate transit at the port than the time taken in transit to the port. Figure 3.1: Transit from the port of Beira to Malawi (a) Container traffic (b) General cargo traffic 25,000 400,000 20,000 Throughputs in TEUs 300,000 Metric Tonnes 15,000 200,000 10,000 100,000 5,000 0 0 2000 2001 2002 2003 2004 2005 2006 2007 2000 2001 2002 2003 2004 2005 2006 2007 Source: CFM. - 43 - 3.6 Beira: transit volumes have tripled since 2000 and the port is now Malawi's main gateway. The Sena railway line to Beira has not operated since the civil war in Mozambique and therefore, freight is shipped to the port by road. Although the road condition of the Beira corridor is fair to good, the port is not as reliable as Durban, and experiences a typical delay of two weeks. Beira is a shallow port in urgent need of capital dredging, currently only operates at 40 percent capacity and for only six hours per day (unless vessels take the risk of grounding). The port is primarily served by feeder vessels, which transship their cargoes to international ships at Durban. Tobacco is the main commodity exported through this corridor and the port is used to import from Europe and the Far East. In 2007, Beira handled approximate 22,000 Twenty-foot Equivalent Units (TEUs) in containerized cargo and 200,000 tons of general cargo going to Malawi, plus most of the country's fuel imports (300,000 tons) (see Figure 3.1). 3.7 Nacala: although a deep water port, the Nacala corridor is marred by poor port and transit operations resulting in long port delays. The delays at the port are caused by the poor condition of its infrastructure and performance of its concession and customs. The Nacala port, served only by feeder vessels that transship in Durban, is linked to Malawi by rail. The railroad has a potential throughput of one million tons per year, but in 2007, Malawi's throughput was only about 215,000 tons. The CEAR concession is operating well below its optimal level and is facing problems with the condition of the tracks and its rolling stock. The railway line between Nacala and Cuamba (533 km) was upgraded with heavier rails in 1993; however, the 77 km of track between Cuamba and the Malawian border remain in poor condition requiring rolling stock to travel at 10km/hr. Furthermore, road traffic between Nacala and Malawi is low due to the poor condition of the road west of Nampula, and the need for trucks to take long detours which results in a doubling of the distance to the Malawian border. Moreover, during the wet season, the road is not passable. Low cost combined with low reliability mean that the port is used for robust non-time-sensitive products such as sugar, which have low time constraints. The main export is sugar while the main imports are fertilizer and general goods. 3.8 Durban: used by non-traditional exports (e.g. AGOA garments), but also by tobacco and tea, because of consistency and reliability. Traders with reliability needs such as textile exporters unable to afford the cost of delays in the Mozambican routes choose the Durban route. For the Malawian textile industry, final destination freight costs via Durban are equivalent to approximately 15 percent of the value of production (source: interviews). Despite the additional cost incurred if a ship is missed and the cargo needs to catch up in Cape Town, the industry manages to export between 16 to 20 forty- foot containers per month and import around twelve. 3.9 Transportation costs of Malawian trade are not especially high, but differ considerably by goods traded and by corridor used. The total cost of transport and logistics per ton from Malawi to final export destination is given in Table 3.3. The data demonstrates that the tobacco sector has been able to negotiate competitive rates across all three major corridors, but that there is a significant premium on sugar and tea exports - 44 - for routes with greater reliability. Sugar exporters pay an additional 14 percent to route sugar through Beira rather than Nacala, and tea exporters pay an additional 17 percent for Beira and an extra 39 percent to route through Durban. Table 3.3: Transport and logistics cost of major exports, 2004 US$ per ton Nacala Beira Durban Tobacco (Lilongwe to NWC) $ 314 (58) $ 314 (62) $ 349 (149) Sugar (Blantyre to NWC) ** $ 122 (31) $ 138 (48) Tea (Estate to NWC) $ 164 (45) $ 192 (82) $ 228 (127) Notes: * Inland freight charges are show in parenthesis ** Sugar transport costs via Beira are the average of cost with and without backloads Source: Malawi Transport Cost Study, Tera International Group, 2005 3.10 Transport costs imply that the additional import costs due to being landlocked average around US$120/ton or about 12.5 percent of the value of imports. The unit costs of long-haul freight are in the lower range of what can be expected in Sub-Saharan Africa despite limited backloads on most corridors. A recent World Bank survey (Teravaninthorn and Raballand, 2008) found that transport costs from Johannesburg to Lusaka were about 30 percent higher than to Blantyre over a comparable distance, in part because truckers had a faster turnaround from Blantyre due to excessive waiting time at the Zambian border. 3.11 The long-haul trucking markets are liberalized and operate reasonably efficiently. Malawian truckers on long-haul routes are operating under competitive conditions and have good operational performance (120,000 km/year, 2 US$/km of operating costs). They have good maintenance and operational alliances to maximize backload, which is an obvious problem given the imbalance of trade. The logistics of tobacco is the exception since most of the tobacco is transported by individuals operating seasonally using very old transport equipment. On domestic routes, companies like Farmers World closed down their own fleet and used long-term contracts to outsource to professional providers. Nevertheless, the market is very shallow and hence unexpected short-term peaks in demand such as around the time of the tobacco harvest, can exert substantial temporary increases in transport price. 3.12 There is widespread perception among policy makers that the costs of trade transport are excessive and that the solution lies in infrastructure. At the centre of the Malawi Growth and Development Strategy (MGDS) is the assumption that poor infrastructure limits the country's productivity and affects internal and external trade efficiency. The National Transport Policy (2004) reinforces the thrust of the MGDS in putting investment in infrastructure at the center of the growth strategy. For the last decade, the focus of attention has been the Nacala corridor as the primary gateway for Malawi and adjacent regions (e.g. eastern Zambia and Mozambique). 3.13 Yet it is the reliability of transport, which causes exporters and importers to reroute into higher cost routes, and these higher costs which disincentivise trade - 45 - (World Bank, 2007). This is confirmed by the findings of the UN initiative on landlocked countries, the "Almaty Program of Actions" which concludes that corridor reliability eventually becomes more important than the cost of transport itself. In Malawi, the dependence on multiple transit systems through other countries implies that supply chains from regional gateways are to various degrees fragmented. The performance of these supply chains depends on a combination of hard and soft constraints and factors, many of which cannot be addressed in Malawi alone but need regional or bilateral cooperation: · Infrastructure condition on corridors including for accessing activity centres in the landlocked country · Logistics performance of gateway ports and their connectivity to world markets · The implementation of regional transit arrangements · The domestic trade facilitation framework i.e. trade related procedures · The availability of services brokers, freight forwarders and truckers support Transport Infrastructure 3.14 The paved road network in Malawi is currently in good or very good condition but maintenance funding has been inadequate. The 2006 Public Expenditure Review (GoM and World Bank, 2007) recommended that adequate and consistent funding be allocated for periodic and routine maintenance. The current budget allocation funded mostly by the fuel levy (7 US cents per liter) is only sufficient to cover routine maintenance and a fraction of periodic maintenance (see Table 3.4), with an annual budgetary shortfall of approximately US$50 million remaining. Without a substantial increase in the maintenance budget, the road network will fast deteriorate resulting in greatly higher cost of repairs. Additionally, the information generated by the spatial analysis (see section 3.B) serves as a guide for prioritizing new road investments that maximize economic returns. The largest contributions to growth for new roads are likely to come from feeder roads in areas of high cash crop potential when improving market access. Table 3.4: Need for periodic and routine maintenance of Malawi's road network Annual estimate for Annual estimate for routine 08/09 Budget Annual shortfall of periodic maintenance maintenance allocation resources 9 billion MK 1 billion MK 2.3 billion MK 7.7 billion MK Source: European Union 3.15 At present, Malawian current rail freight volumes do not justify major rail investment but mining operations close to the line (such as the coalmine in Moatize) could increase freight volumes to the point where rail investment becomes viable. The government should closely monitor mining sector developments in Mozambique, which may justify the re-establishment of the link between Beira from Blantyre. A - 46 - concessionaire of an efficient freight line today will typically look for annual volumes of between 2 to 5 million tons a year ­ a threshold implied by the costs of routine track maintenance and economies of scale of the rolling stock. Given sufficient volumes, the successful operation of the line would depend critically upon the appointment of a single concessionaire able to operate a general cargo freight service with mineral convoys. This requires an urgent renegotiation of the rail concessions. Expected increases in volume could also positively affect the viability of the Nacala corridor. However, it is important that investment decisions are based on solid information regarding expected volumes and returns from cost savings. Currently, the cost savings of rail compared to transit by road are minimal (at 10 to 40 USD/ton compared with Beira) and rail is only attractive to bulk and time-insensitive goods such as sugar. 3.16 Air and lake transport have not developed. The non-application until recently of the Yamoussoukro agreement on the 5th freedom for air transport limited the connectivity of Malawi ­ although the potential for air-freight remains limited. Lake transport has been a traditional means of transportation, but like the railroad, the concession has not produced the expected results. Given current global and regional patterns of logistics, and the level of services expected by international traders, lake transport is likely only to be a solution for regional domestic transport. Trade Facilitation 3.17 Malawi's membership of SADC and COMESA brings facilitation despite lack of implementation. There have been significant achievement such as the COMESA insurance card, and some degree of integration of the trucking sector. More problematic has been the non-uniform application of road transit charges to foreign trucks. Mozambique applies charges twice as high as the agreed target. Malawian trucks are not picking up significant amount of transit loads in Mozambique consequently leaving the market from Beira to Mozambicans. In practice, the 5th freedom has not benefitted Malawian truckers. See Table 3.5 for a list of ongoing regional policy initiatives that benefit Malawi's transport system. Table 3.5: Regional trade facilitation initiatives Simplification and harmonisation of customs COMESA/SADC project, ongoing procedures and legislation Single administrative document for customs Effective in COMESA countries Harmonisation of IT systems and electronic Asycuda++ available in Malawi, Zambia, Tanzania, other customs management systems systems in South Africa Mozambique Harmonised axle loading Harmonized COMESA, SADC (56 tons GVM) Harmonised Road Transit Charges Objective of 10 US cents per km There is a regional carrier license, which in principle allow Carrier's license freedom of loading including 5th freedom. The COMESA Yellow Card is a vehicle insurance scheme, COMESA Yellow Card Scheme: which covers third-party liability and medical expenses. Not applicable in SADC countries. - 47 - 3.18 However, the current transit system does not facilitate Malawian access. Although there is harmonization of documentation within COMESA there are no "gate- to-door" procedures allowing movement of a container from the port to the clearance center in Blantyre or Lilongwe. Transit is a lengthy sequence with many players. The main observations are the following: · Durban-Malawi. The same information is declared 7 times and in the same format (except through Mozambique). · Beira and Nacala. Transit procedures are manual, different from SADC/ COMESA formats and inconsistent from one corridor to the other. It takes as long to initiate a transit declaration as to clear one for final consumption. · Truckers are dependent on brokers and the degree of professionalism of the small brokers is low. Brokers are not allowed to input into Asycuda. · Bonds are country specific and attached to the broker not the trucker. · Within COMESA, thanks to Asycuda and good border management by the Malawi Revenue Authority (MRA), processing delays at land borders are mainly due to brokers. · The national transit procedures in Malawi from the border to the inland depot are not part of Asycuda and not in a format compatible with final clearance and international transit. 3.19 Huge gains would come from the implementation of a regional transit system similar to the International Road Transit (TIR) system in Europe. Two main issues stand out: first, there is the risk of loss of information or unreliable brokers that can delay a truck in transit on the Durban corridor (both for exports and imports). For example, garment exporters are quite often delayed on this route, which removes their 2-3 day safety margin, and consequently have to pay an extra US$5,000 to reach the same ship in Cape Town. Secondly, the release of goods for transit in Mozambican ports is essentially unpredictable. 3.20 Domestic trade related institutions are providing a relatively good service. The main actor is the Malawian Revenue Authority (MRA) which operates customs clearance. The MRA has maintained a good balance between revenue collection and facilitation, and has implemented relevant IT systems and risk management procedures. Areas for future improvement include allowing users to fill declarations online, and the introduction of an authorized operator regime. Solutions 3.21 Malawi is logistically constrained by the reliability and fragmentation of its supply chains to external markets which leads to higher costs as exporters reroute to more reliable corridors. Improving access and connectivity of the country should remain high on the policy agenda of the government and international agencies. Although the reliability of trade logistics depends largely on decisions made by other - 48 - countries, there are good foundations upon which to implement improvements. The private sector in Malawi is remarkably market orientated; the general business- environment scores highly; and the implementation of customs and revenue collection is relatively efficient with a low level of corruption. Additionally, regional agreements constitute a credible base to lobby for improvements in regional trade facilitation. 3.22 The focus of medium-term policy should be on improving trade transport reliability through multiple measures including: · Management of existing infrastructure · Development of alternative modes of freight · Regional trade facilitation · Domestic trade facilitation 3.23 Management of existing infrastructure. Although most of the paved roads in Malawi are currently in good condition, the amount allocated to routine and periodic maintenance is insufficient to ensure that this continues. The South African National Road Agency Ltd. (SANRAL) estimates that repair costs rise to six times maintenance costs after three years of neglect and to 18 times after five years of neglect. It is strongly recommended that the government put in place adequate plans for periodic maintenance of the main paved road and rail networks; and that it discusses with donors ways of financing the funding gap. 3.24 Development of alternative modes of freight. The government should cautiously explore developing alternative modes of freight whilst avoiding a drain on constrained resources. Firstly, by taking advantage of the current rail concession review to: i) Evolve the concession agreement between Malawi and Mozambique to make it consistent across the two countries and to be able to take advantage of the new freight volumes in Mozambique; ii) Discuss with Mozambique the possibility of opening the rail concession to invite a new world-class commercial operator with suitable competences that could efficiently operate traffic, both general cargo and coal logistics under a single agreement. Secondly, if justified by increases in Mozambique trade volumes, through investment in rail to re-open the Sena line and to maintain the current service on the Nacala line. 3.25 Regional trade facilitation. The government and donors should press for the establishment of a regional door-to-door transit regime. Specifically this would mean: (a) the creation of a consistent SADC/COMESA transit regime in the form of regional - 49 - carnet-bond system19; and (b) the harmonization of road charges within SADC. Although Mozambique may not be able to be among the first countries to join the system, practical improvements could be negotiated between Malawi, Mozambique, and Zambia for the Beira and Nacala corridor. 3.26 Domestic trade facilitation. There are various measures which should be implemented to improve the efficiency of domestic trade including (a) allow custom declarations to be filled in and processed online; (b) introduce an authorized operator regime for large importers; (c) Introduce tighter regulation of customs brokers; and (d) improve transparency and response time for the issuance of trade licenses. B. IMPROVING CONNECTIVITY THROUGH THE SPATIAL DIMENSION OF GROWTH 3.27 Road transport currently plays the dominant role in handling Malawi's domestic and international trade. Malawi's network of 12,500 km of formal roads and 79,000 km of unclassified feeder roads is extensive but has enormous variations in road quality. Roads account for more than 70 percent of internal freight and over 90 percent of the country's international traffic. Formal main, secondary, and tertiary roads are evenly distributed: 3,400 kilometers, 2,800 kilometers, and 3,800 kilometers respectively. However, Malawi also has over 79,000 km of unclassified feeder roads, which form the majority (86 percent) of the national road network. Figure 3.2: Road quality and traffic volume Source: Malawi National Road Authority (road location), HDM4 (road conditions and traffic volume) 19 Note that this project is part of the current North-South corridor initiative. - 50 - 3.28 Transport volumes are (inversely) correlated with road quality ­ better roads are used more by motorized and non-motorized traffic. The spatial data on physical condition and traffic volumes for each class of road is summarized in Figure 3.2 and Table 3.6. Spatial data confirms that traffic volumes are concentrated around the major cities of Lilongwe, Blantyre, Mzuzu and Zomba, and the averages at the level of road class shows that higher traffic volumes are associated with higher road quality in terms of proportion paved and roughness index. Table 3.6: Physical road conditions by road class Length Width Roughness Average Daily Traffic Road class (km) (lanes) (average) Paved All Motorized Main 1,931 1.7 4.4 75.7% 506.9 204.9 Secondary 1,361 1.0 6.9 12.9% 90.2 38.3 Tertiary 1,509 1.0 6.6 5.5% 67.3 11.8 District 149 1.1 5.2 33.6% 182.7 73.1 Other 3,577 1.2 6.1 33.9% 223.1 65.8 Total 8,528 1.2 5.9 35.7% 237.9 83.5 Note: The sample excludes unmatched road segments. Source: Malawi National Road Authority and HDM4 database 3.29 The government has increased its funding of transport infrastructure but implementation has been weak. The Malawi Growth and Development Strategy (MGDS) recognizes that poor infrastructure negatively affects both growth and the internal and external efficiency of trade. Transport investment has become a priority and the government has significantly increased budget allocation to the sector from around US$5 million in 2003/04 to over US$20 million in 2008/09. However, despite increased funding, the 2006/07 Annual MGDS Review reveals that project implementation in the transport sector has been poor and has not reflected economic growth priorities. 3.30 Transport strategy currently focuses on growth promotion through large- scale infrastructure but pays little attention to feeder roads. In 2006, the GoM developed infrastructure plans under the Infrastructure Services Programme (ISP) aimed at supporting economic growth in five prioritized corridors prioritized on the potential value addition in key productive activities. These corridors were Rumphi-Nyika-Chitipa, Ntcheu-Tsangano-Mwanza, Mangochi-Cape McLear, Zomba-Phalombe-Mulanje and Nsanje-Bangula-Tengani. What determines the cost of transport? 3.31 Farmer's tobacco transport costs are influenced by the quality of feeder roads. Using data from the 2004 Integrated Household Survey (IHS) a regression analysis was performed using farmer transport costs for taking tobacco to markets as the dependent variable. After controlling for the strong effects of shipment volume and distance transported, the analysis found that the quality of feeder roads (proxied by the proportion of unpaved road in the journey to the nearest town) was significant at the 5 percent confidence level. Tobacco from farms connected by unpaved roads was more expensive to transport than from farms with better quality feeder roads ­ even if the total - 51 - distance moved was the same. The quality of feeder roads exerts a significant effect on transport prices. Table 3.7: Determinants of farmer's tobacco transport costs Dependent variable: ln(tobacco transport costs) Estimation OLS, with IHS2 sampling weights Sample all tobacco farmers in IHS dataset ln(transport volume) 0.815*** 0.800*** 0.812*** 0.804*** (0.029) (0.029) (0.030) (0.030) ln(transport distance) 0.150*** 0.139*** 0.148*** 0.152*** (0.024) (0.025) (0.031) (0.033) ln(% of unpaved to nearest town) 0.032** 0.014 (0.013) (0.014) ln(road traffic volume, ADT) -0.066*** -0.061*** (0.018) (0.020) Observations 970 920 825 793 R-squared 0.646 0.637 0.665 0.656 Note: 1. Robust standard errors in parentheses 2. * significant at 10%; ** significant at 5%; *** significant at 1% Source: Integrated Household Survey, Malawi National Road Authority and HDM4 database 3.32 The role of feeder roads is confirmed by the proportion of cash crops produced in areas with low travel times to major cities. Using travel times to the nearest major city (Lilongwe, Blantyre, Mzuzu, and Zomba) Table 1.58 gives the percentage of cash crops produced at different distances from cities and for different levels of agronomic potential. Crop production is concentrated in locations with lower travel times and good agronomic potential confirming that improvements in feeder roads in such areas are likely to release additional cash crop production. Table 3.8: Cash crop production by travel time and agronomic potential Quintiles of travel time to nearest major city 5 (longest) 4 3 2 1 (shortest) 5 (worst) 1.1 3.3 3.5 0.1 0 4 3 0.7 1 2.3 0.8 Quintiles of Agronomic 3 5.4 13.7 1.4 1 0.5 potential 2 10.3 4.1 9.3 2.1 0.2 1 (best) 1.2 7.6 10 15.1 2.2 Data Source: GIS+HDM4 data, IFPRI SPAM, and the IHS2 3.33 Transport costs are lower in locations with high traffic volumes, where demand creates competition between transporters. For each of the 564 locations in the IHS dataset, spatial analysis was used to compute the average road traffic volume (ADT) within a radius of 10 kilometers. The results from the regression show that the transport costs are lower in locations with high traffic volumes. This result was significant at the 1 percent confidence level. - 52 - 3.34 A farmer's transport costs of taking tobacco to markets are thus determined by both supply and demand side factors. Poor feeder road quality, which links farms to main roads and low traffic volume (limited competition and less trucking demand) on neighboring main roads raise transport costs. These findings are consistent with Africa wide studies that highlight infrastructure constraints (Pedersen, 2001) low levels of competition between service providers (Rizet & Hine, 1993) and weak infrastructure (Limao & Venables, 2001) as contributors to high transport costs. Domestic and international trucking prices 3.35 Domestic unit transport prices (per ton, per km) are far in excess of international ones. The Malawi route-specific trucking survey (World Bank, 2008b)20 reported that the average unit transport price from rural areas to main cities was MK228 per ton per km; against a unit price of MK12 per ton per km for routes linking the country to international markets. Table 3.9: Trucking survey - summary of main findings Unit price per Trip Price per Annual Average Empty Origin Destination ton per km distance tonne mileage load backhaul (MK) (km) (MK) (km) (tonnes) Agro town City 228.4 85 19,323 34,020 27% 2.5 Agro town Exporting port 10.3 2,273 23,462 82,200 11% 24.6 City Exporting port 12.1 2,012 24,433 71,102 5% 19.9 Source: Malawi route-specific trucking survey, November 2008, World Bank 3.36 What could explain the difference in unit prices between domestic and international routes? Domestic routes are short (on average a mere 85 km), have small loads (on average 2.5 tonnes), have higher proportion of empty backhauls (27 percent), and have segments of their routes which international fleets simply could not operate on. This results in higher fixed and transaction costs, and even if an international operator wanted to compete on domestic routes, the low volumes, short distances and poor roads would result in low expected profits. 3.37 Unit transport costs are affected by truck size, the percentage of empty backhauls, and whether the company is domestic or international. A regression analysis of determinants of unit transport costs is shown in table 3.10. The analysis controls for trip distance and finds that truck size is a significant factor: larger trucks introduce economies of scale and reduce the unit transport price. The percentage of empty backhaul trips has a significant effect on unit prices for international routes but not for domestic truckers considering the low volumes and short average trip distance of 85 km against 2000 km for international routes. Lastly, by using a dummy variable for 20 The survey was commissioned for this CEM and collected 130 surveys from 20 trucking companies and 60 truckers. This extreme disparity in unit prices is a clear indication that transport markets are segmented with limited mobility between segments - 53 - domestic truckers, the regression implies that controlling for truck size, trip distance domestic truckers charge higher prices than those servicing international routes. Table 3.10: Determinants of unit transport costs from trucking survey Dependent variable: ln(price charged to go per ton) Estimation OLS Sample Full sample Domestic truckers Full sample ln(trip distance) 0.668*** 0.986*** 1.023*** (0.017) (0.094) (0.021) ln(average tare weight) -1.092*** -0.711** -0.428 (0.063) (0.302) (0.281) % of empty back- haul 2.354* 0.914 1.028** (0.555) (0.868) (0.120) dummy for domestic 2.417** truckers (0.455) Observations 59 31 59 R-squared 0.710 0.813 0.868 Note: 1. Clustered (by route type) robust standard errors in parentheses 2. * significant at 10%; ** significant at 5%; *** significant at 1% Source: Malawi route-specific trucking survey, November 2008, World Bank Solutions 3.38 New infrastructure investment should be made in feeder roads in areas with good agronomic potential that are close to markets. The quality of the trunk road network is not a major constraint in Malawi, however differences in the quality of feeder roads connecting villages to the main road network have significant bearing on transport costs. Since the intensity of cash-cropping is significantly affected by journey times to the nearest market, the evidence strongly suggests that investment in feeder roads in areas of high agronomic potential relatively near markets will have high impact on economic growth. 3.39 The domestic trucking services industry should be opened to competition for local routes from international firms. The current regulatory framework protects domestic transport from international competition along domestic routes. This protection perpetuates a lack of competition in the domestic transport industry enabling it to charge monopolistic rents. High transport charges directly lower the farm gate prices of agricultural products and hence act as a drag on economic growth. However, although this measure could significantly lower transport costs along trunk routes, it is unlikely to have significantly impact in rural areas since low volumes and poor road quality make it unlikely that international operators will enter these markets. 3.40 To lower transport costs in rural areas, the government should encourage the development of a domestic small vehicle transport sector. As international truckers are unlikely to find these routes profitable, alternate modes of transport need to be identified. It may be useful to consider options for promoting appropriate intermediate means of transport (IMTs) such as such as single-axle tractors, motorcycles/bicycles with - 54 - trailers, animal-drawn sledges or wheelbarrows for connecting rural areas to market towns. Related work by the World Bank's transport department suggests that IMTs can reduce the cost, time and effort of short distance transport, and can increase overall rural mobility. IMTs compare favorably with "head loading" since they have higher load capacity, and enable users to transport at a higher speed. In addition, they require only well maintained tracks and footpaths and can easily be maintained locally at low cost. Efforts on identifying appropriate IMTs are already underway in Malawi. C. IMPROVING ENERGY PROVISION 3.41 Peak electricity demand currently exceeds available generating capacity by a large margin. Malawi generates almost all of its grid-supplied electricity through hydropower plants on the Shire river. Total installed generation capacity is currently around 283 MW, but recent rehabilitation of generation units has reduced available capacity by up to 40 MW. With unsuppressed net peak load estimated to exceed 290 MW, available capacity has been over 40 MW less than peak demand. As a result, load- shedding (planned power outages) are used extensively to ration power. The situation is worsened by frequent unscheduled outages, caused by equipment failures and exogenous shocks such as flooding. Figure 3.3: Projection of peak demand and generating capacity Source: ESCOM 3.42 The Electricity Supply Corporation of Malawi (ESCOM) is a Government- owned electric utility that generates, transmits and distributes electric power. ESCOM's combined technical and non-technical losses amount to around 20 percent of electricity generated. While these losses are not as serious as other utilities in sub- Saharan Africa, there is ample scope to reduce them substantially. 3.43 Less than 6 percent of households are connected to the ESCOM grid and the large majority of these are in urban centres. Only 1 percent of the rural population is connected and a relatively low proportion of electricity is consumed by industry ­ 47 percent compared to Uganda's 73 percent. Average residential consumption, on the other hand, is very high for a low-income country (at around 250kWh/month, compared to - 55 - 71kWh/month in Ghana) and is closer to South African levels. This reflects the relatively high levels of cooking with electricity, driven by low tariffs. Table 3.11 ESCOM's existing generation facilities Original Current Facility Capacity Capacity Configuration Notes Nkula Falls A 24 MW 24 MW 3 x 8MW Commissioned in 1966/67. Nkula Falls B 100 MW 100 MW 5 x 20MW Rehabilitation required 2009-10 Tedzani I & II 40 MW 20 MW 4 x 10MW Rehabilitation due to finish early 2009 Tedzani III 50 MW 50 MW 2 x 25MW Commissioned in 1995/96. Kapichira I 65 MW 65 MW 2 x 32.4MW Commissioned in 2000. Wovwe 4.5 MW 4.5 MW 3 x 1.45MW Commissioned in 1995. Total 283 MW 263 MW Source: ESCOM 3.44 Energy is consistently identified as a key constraint to economic growth in Malawi. The 2008 Business Climate Survey cited erratic electricity supply as the primary obstacle to doing business in Malawi. Poor quality of supply was reported as the fourth more serious obstacle by firms in the 2006 Investment Climate Assessment (ICA, World Bank 2006a); the constraints analysis undertaken for this report identifies it as a binding constraint to growth. The Africa Infrastructure Diagnostic (AICD, World Bank 2008c) of 2008 reports that Malawi has the worst supply reliability record of the 24 countries surveyed and estimates that the hidden cost of inefficiencies in electricity supply are equivalent to 4.4 percent of GDP ­ the most severe in sub-Saharan Africa. 3.45 Power outages impose high costs on the private sector and are consistently reported to have prevented new investment in Malawi. Power outages are estimated to be responsible for median sales losses of 10 percent across Malawi, significantly higher than the low-income country average of 2 percent. For those firms without a generator (around 50 percent of Malawian firms) median sales losses are even higher at an estimated 20 percent. Power outages damage product quality in the dairy industry, require extensive cleaning of equipment in the plastics industry, and result in high losses in thread spinning. The cost of unscheduled outages in the textile industry can often be disproportionate to the duration of the outage; for example, the Mapeto textile factory may lose between 300 and 500 meters of fabric whenever there is a supply interruption. Inconsistent quality of supply was cited by cigarette manufacturer BAT for moving its operation out of Malawi; the proposed extraction of titanium heavy sands from the lake was cancelled for the same reason; and similarly several potential textile investments were never realized. In addition, firms which do decide to invest often have to resort to expensive self generation using diesel imports, harming their competitiveness and creating higher import dependence. - 56 - Figure 3.4: Sales losses due to power outages and firms with generators Median sales losses due to power % of firms with generator outages Malawi Kenya Tanzania Tanzania Madagascar Malawi Kenya Zambia Zambia Uganda Uganda Madagascar South Africa South Africa SSA SSA Low income Low income 0% 2% 4% 6% 8% 10% 0% 20% 40% 60% 80% Source: Malawi Investment Climate Assessment, 2006 Issues 3.46 Net peak demand is projected to grow at over 7 percent per year over the next decade. This incorporates the mitigating effects of a proposed Demand Side Management (DSM) programme ­ peak demand reduction gained by installation of energy efficient light bulbs and the introduction of a Time-of-Use (ToU) tariff to incentivize off-peak power usage by the private sector. ESCOM estimates that net peak demand will reach 334 MW by 2010 and 696 MW by 2020. These figures include Malawi's Rural Electrification Programme (MAREP) currently being implemented by the government, and it is anticipated that electricity access will have risen from the current level of less than 6 percent to 7 percent in 2010 and a planned 30 percent by 2020. 3.47 The significant shortfall in generating capacity is expected to continue in the short-term. Despite the re-commissioning of 40 MW at the Tedzani hydropower scheme, the scheduled rehabilitation of Nkula `B' in 2009-10 will temporarily reduce available capacity by 20 MW. Assuming a required reserve margin of 10 percent of generation capacity, an additional 40 MW is required to sustain supply at an acceptable level of reliability. 3.48 In the medium-term, the capacity shortfall will increase to 55 MW by 2013 - even with planned investments. Assuming an additional 50 MW from the construction of the interconnector to Mozambique by 2012 (financed by the World Bank but currently awaiting Parliament approval) and an additional 65 MW from a second set of turbines at Kapichira, a further 55 MW will still be required to meet projected demand. 3.49 Rehabilitation of existing stations is increasingly critical. In addition to flood damage, generation equipment at Tedzani I and II has also been damaged by increasing levels of sediment load. Siltation has become an increasing problem due to deforestation and soil erosion, and dredging is required at all hydro-electric plants. Aquatic weed clearance is another growing challenge, and recent high water levels in the Shire overwhelmed existing weed and debris-removing machinery. Rehabilitation of Nkula B - 57 - will commence shortly, and older plants such as Nkula A are also in need of routine overhaul. 3.50 Malawi faces very significant risks because it is reliant on a highly undiversified source of power. 98.5 percent of power is generated from hydroelectric plants on the Shire river, which has shown significant long-term variation in its flow. This dependence on hydro-electric generation intensifies Malawi's economic dependence on rainfall patterns and exposes it to the possibility of a drought-induced power crisis. The average flow in the Shire between 1994 and 2002 was less than half the long-run average calculated from available data. Moreover, the Shire has in the past periodically stopped flowing altogether; this occurred in 1915 for a period of 17 years. The implications of this happening today are hard to overstate: Malawi would have an operational hydropower output of 4.5 MW against a demand of over 290 MW. The transmission interconnection with Mozambique can only mitigate this risk to a limited extent. 3.51 Underinvestment in transmission and distribution infrastructure imposes further economic costs. Inadequate capacity of transmission lines, overloaded transformers and other shortcomings results in network failures and poor quality of supply. Low frequency and voltage spikes damage motors and electronic equipment, and prematurely shorten equipment lives. In fact, low-quality and frequently interrupted supply due to network inadequacies are perhaps the greatest concern for existing industrial and commercial operators, many of whom are largely shielded from load shedding. While ESCOM tries to maintain supply to priority HV customers, outages due to equipment failures in the transmission and distribution systems are often unavoidable. The transmission network supplying northern Malawi is particularly weak and insecure, and is a major constraint to economic growth in the region. 3.52 Electricity tariffs are too low to fund new investment. Domestic tariffs (at 3-5 US cents/kWh) are among the lowest in Sub-Saharan Africa, benefitting a small group of relatively wealthier urban households. Industry tariffs are higher, but still comparatively low for the region. As a result, revenue during FY03-07 covered ESCOM operating costs, but made it practically impossible for the organization to undertake any major rehabilitation or expansion in generation or distribution. Additionally poor performance in revenue collection, coupled with devaluation of the kwacha, resulted in ESCOM being unable to service its debt in the 1990s. The Government has since restructured the organization's debt but the net cash flows have been negative during three of the past four years (AICD, World Bank 2008c). As such, ESCOM is unable to borrow for even relatively modest capital projects, and the rehabilitation at Tedzani was undertaken on a high-cost trade-financed basis and took several years to arrange. 3.53 The power sector is not attracting private sector investment. Disincentives to private sector participation include a lack of clarity on sector structure (particularly relating to the role of ESCOM), concerns over the creditworthiness of ESCOM as the - 58 - sole off-taker and tariffs that do not offer the prospect of a sufficient return on investment. Solutions - Immediate 3.54 Invest in two additional generating units at Kapichira. System planning studies indicate that Malawi's least-cost option for increasing system capacity is to expand the existing Kapichira plant, adding 64.8 MW to the system from two additional turbines. All civil works for the second phase of Kapichira were completed under Phase I and the incremental cost of the scheme is therefore quite low. It is estimated that Kapichira II will cost approximately US$55-60 million (excluding engineering, supervision and associated transmission line and substation costs) and that the implementation period would be about three years. ESCOM has recently launched a tender for the work, to be financed by the Government. 3.55 Approve the Transmission Interconnector with Mozambique. Parliamentary approval of the Interconnector Project must be a priority after the elections in May 2009. Given the constrained power supply situation in southern Africa, securing a Power Supply Agreement is also a priority 3.56 Implement demand-side management (DSM) programmes to reduce overall energy needs and, in particular, demand at peak times. Options include: a. Provide immediate financial support for an energy-saving light-bulb programme. ESCOM is planning to substitute over 1 million traditional incandescent bulbs with energy-efficient compact fluorescent light (CFL) bulbs, over the course of three years. The programme could offset peak demand by up to 20 MW at an estimated cost of around US$4.4 million - much less than would be required to install the equivalent generating capacity. In addition, the programme could be rolled out rapidly by bulk procurement from international suppliers. b. Shifting of load from peak times using a time-of-use tariff. Time-of-use tariffs incentivize commercial and industrial customers to reduce power use at peak times and would not require major investment, since many of these customers already have programmable digital meters that can cope with dynamic tariff structures. 3.57 Improve ESCOM's financial position through efforts to improve revenue collection and streamline the cost base. ESCOM's precarious financial situation is exacerbated by the non-payment of electricity bills, particularly by public sector organizations. This causes serious cashflow constraints and restricts the ability of the utility to fund routine maintenance and connect new customers. Launching a new arrears and debt collection programme, including via a public awareness campaign, could help boost the collection rate. On the cost side, there is also significant room to increase the - 59 - efficiency of ESCOM's operations, including through tighter control of spending and improved logistics. 3.58 Reform the electricity tariff to finance significant network investment and incentivize Independent Power Producers (IPPs). An ESCOM proposal for a tariff increase is currently under review (as of March 2009), which involves public consultations and proposes an increase in the weighted average tariff of 54 percent. There is an emerging consensus that any tariff increase for ESCOM must be linked to improved quality of supply and greater operational efficiency at ESCOM. It is possible that higher domestic tariffs may reduce cooking using electricity, hence increasing pressure on biomass fuel supplies. 3.59 Investigate the introduction of a high quality-of-service tariff for industries with critical energy requirements. Because of the significant costs imposed on businesses by outages and poor quality of supply, ESCOM could investigate the possibility of offering a `premium' supply service to industries with a critical need for reliable power. In return for highest priority supply, customers would pay a higher tariff, reflecting the alternative cost to the firm of self-generation of using diesel generators. Service to other customers would be likely to be slightly degraded. Solutions ­ Near term 3.60 Encourage Independent Power Producers (IPPs) to invest in small-scale renewable generation through guaranteed pricing and tax incentives. Isolated mini- grids below 5 MW are permitted according to the Rural Electrification Act. Private investors could be incentivized by (a) setting a standardized "feed-in" tariff (a price which ESCOM would guarantee to pay to the IPP) at a sufficiently high rate to give an adequate return on investment; (b) creating standardized Power Purchase Agreements (PPAs) and a consistent legal framework for IPPs; (c) investigate the possibility of offering tax incentives, such as enhanced capital allowances, to IPPs looking to invest in renewable energy generation. 3.61 The Malawi government should articulate a clear framework for the institutional structure of the sector, including the respective roles of ESCOM and private investors. The Power Sector Reform Strategy (PSRS) approved by the Government of Malawi in 2003 laid out a new structure for the energy sector, in which ESCOM would be `unbundled' and the private sector engaged through long term concessions in transmission and distribution. However, following limited success with this model elsewhere in Africa, a full unbundling may not be the optimal approach for a small system such as Malawi's. Nonetheless, uncertainty remains about which elements of the PSRS will be implemented, which may act as a disincentive to potential investors in possible concessions or PPPs. It is therefore important that the government clarify the future role of ESCOM in power generation, the future corporate structure of the utility and the implementation of the PSRS. - 60 - 3.62 Revisit ethanol fuel mixing, with a view to reintroducing an 80:20 mix. Malawi's fuel is currently a 90:10 blend of petrol and ethanol. A return to an 80:20 blend would reduce petrol imports by around 10 percent. Malawi has surplus ethanol production capacity, which is only being utilized gradually in line with the expansion in sugar production. Even without an expansion in sugar production, an ethanol blend could be reintroduced with a net saving of foreign currency (equal to the import price of petrol less the export price of ethanol). Solutions - Medium Term 3.63 Conduct a full feasibility study into Lower Fufu in order to attract a shortlist of IPPs by 2010. The Lower Fufu Falls on the Rukuru river is considered to be the best option for a new hydro site (in terms of economic, environmental and social impacts), with generation potential of between 70-145 MW. Assuming that a clear framework for IPP investment is in place (as discussed above) the Lower Fufu scheme could be developed by 2015. A feasibility study is an urgent prerequisite, including a detailed environmental and social impact assessment, and may cost between US$5-8 million. A grouping of donor agencies may wish to collaborate on the funding of these studies. 3.64 Transmission and Distribution (T&D) investment programme. According to World Bank estimates, over US$200 million of new investment is needed in Malawi's transmission and distribution networks. The proposed World Bank energy sector project and the Millennium Challenge Corporation's proposed Compact are expected to finance some urgent rehabilitation and upgrade requirements, but further external financing will be needed. Looking further ahead, options for future regional interconnectors, including to Tanzania, may come under consideration as the regional power pools start to integrate. 3.65 Review the impact of the `single-buyer' model on the incentives for IPPs to invest in the sector. Under the power sector structure proposed in the PSRS, the transmission licensee would act as `single-buyer' (that is, all power generated would need to be purchased and wheeled through the transmission company's grid). This `single- buyer' model may act as a disincentive to IPPs, because (a) ESCOM, as the transmission licensee and sole off-taker, may not be considered by external IPP financiers to have sufficient financial strength, and (b) developers may be reluctant to rely on ESCOM to evacuate the power produced. Alternative models that allow IPPs to sell power directly to consumers (such as mining or commercial agriculture firms) could be considered. This points to a possible role for multilateral and donor assistance in providing appropriate guarantees to help bridge the gap between real and perceived risks in electricity investment in Malawi. 3.66 Finance a Malawi-wide micro hydropower resource study and strategic environmental assessment. The micro and pico-hydro potential in Malawi is likely to be economically viable, especially in remote locations. As a first step, reconnaissance resource studies should be undertaken to identify potential energy sources, using stream- flow estimates extrapolated from local gauging stations and rain gauges. The cost of a - 61 - regional resource study is estimated at around US$200,000, and the cost of a country- wide assessment at around US$400,000. 3.67 Finance a high-level wind power resource study followed by targeted anemometer studies in at least five promising locations beginning in 2009. Wind energy in Malawi has not yet been exploited on any significant scale. However, recent experience with hybrid solar PV systems, which incorporate small wind turbines, has been positive. In addition, some commentators claim that the wind regime on Malawi's many hill tops, and on the shores of Lake Malawi, is promising21. A detailed resource study, which would collect primary data from newly-erected anemometers at five key sites, would cost an estimated US$150,000-200,000. 3.68 Donor organizations based in Malawi should investigate the potential for procuring renewable energy supplies directly from IPPs, thus testing out the regulatory framework and providing a case study for other prospective consumers. By soliciting bids for new, renewable generation supplies for their own energy needs, on the back of a fixed-term, premium-priced power purchase agreement, donors could incentivize prospective IPPs to enter the market, using the ESCOM grid to wheel the power. 21 This is corroborated by studies undertaken in Tanzania, which discovered significant wind resource potential inland that should enable commercial exploitation. - 62 - 4. THE SUPPORTIVE MACROECONOMIC FRAMEWORK A. INTRODUCTION 4.1 This section of the report addresses key questions facing Malawian policymakers in the arena of macroeconomic policy. It summarizes macroeconomic developments, recapped in Table 4.1 and Figure 4.1 and Figure 4.2, and analyzes the key policies needed to achieve a macroeconomic environment that supports growth through exports in particular over the medium term. 4.2 There is global consensus that a stable macroeconomic framework is a necessary ingredient for sustained economic growth while an unstable macroeconomic environment affects the poor disproportionately22. A stable macroeconomic environment can therefore also make growth more inclusive and thus improve the impact of growth on poverty. Macroeconomic policy in Malawi must aim to contribute to the effective use of available productive capacity, advance the accumulation of inputs in production, and allow for an efficient allocation of resources. International experience provides ample evidence that creating a macroeconomic environment conducive to growth depends on relatively low inflation and interest rates, low fiscal deficits in combination with low debt distress,23 a competitive exchange rate, and growth of aggregate demand, which does not outstrip expansion of domestic production in the medium term.24 4.3 Recent macroeconomic policy in Malawi, in particular fiscal discipline, have to a great extent contributed to Malawi's current growth momentum and have also improved future growth prospects. Lower budget deficits routinely result in higher domestic savings and this in turn can lead to higher investments. Reduced demand for resources by the government not only increases the supply of capital to the private sector but in general also leads to a lower cost of capital. Lower public indebtedness in turn improves the flexibility of macroeconomic policy, as it strengthens the government's ability to use fiscal policy as a countercyclical macroeconomic policy instrument. 4.4 However, improved macroeconomic management, focused primarily on fiscal discipline and low inflation, has not led to a sufficiently large domestic supply response for GDP growth to match growth of aggregate demand. Domestic demand has grown faster than the economy as a whole, leading to a widening of the external trade balance. These developments have put significant pressure on the availability of foreign exchange, even in the presence of large inflows of foreign aid and debt relief. 4.5 What is needed now is not a fundamental change of direction, but a shift in emphasis in macroeconomic management from attaining macroeconomic stability to also supporting growth. In the short run, macroeconomic policy needs to contain growth in aggregate demand as it has outpaced increases in GDP, partly through an 22 See Easterly and Fisher (2001) for an exposition on the impact of high inflation on the poor. 23 See Kraay and Nehru (2003) for an exposition on the characteristics of debt distress. 24 See for instance the seminal paper by Fisher (1993), as well as Easterly and Rebelo (1994). - 63 - adjustment of the real exchange rate and a reduction of public expenditures. This will also assist in restraining the demand for imports of consumables. In addition, macroeconomic policy also needs to address the accumulation of foreign reserves and as such a reevaluation of the use of foreign aid. For macroeconomic policy to have the desired impact on Malawi's productive capacity and supply potential, policy will need to be complemented by public investments to alleviate the supply side constraints identified in the areas of energy supply, and (cross-border) transport and trade facilitation. This will require a further switching of public expenditures in favor of infrastructural programs that improve productivity while at the same time reducing overall public expenditures as the size of government has grown large compared to its comparators.. B. KEY MACROECONOMIC POLICY PRIORITIES TO SUSTAIN AND BROADEN GROWTH 4.6 The macroeconomic challenges facing Malawi at the beginning of the decade clearly threatened macro stability and thus sustainable growth. The fiscal position was becoming untenable, inflation was high, and the external debt situation was unsustainable. Successful implementation by the authorities of an economic stabilization program resulted in a number of important achievements, including a significant reduction of domestic and external debt, and lower inflation and interest rates. Table 4.1: Key recent macroeconomic developments 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 General 1 .6 - 5 .0 2 .9 5 . 7 5 .4 3 .3 6 .7 8 .6 9 .8 GDP Growth (constant prices) 3 5 .4 2 2 .1 1 1 .5 9 . 8 1 3 .7 1 6 .5 1 0 .1 7 .5 9 .9 Inflation CPI (end of period, year-on-year, % change) 3 9 .5 4 2 .4 4 1 .7 3 9 . 3 2 8 .6 2 4 .4 1 9 .3 1 3 .9 1 1 .3 91 day TB yield (period average, %) 5 3 .1 5 6 .2 5 0 .5 4 8 . 9 3 6 .8 3 3 .1 3 2 .3 2 7 .7 .. Lending rate (period average, %) 1 2 6 .5 1 2 8 .4 1 3 3 .2 1 3 6 .2 1 3 7 .1 1 4 5 .7 1 4 6 .1 1 4 3 .0 1 5 3 .8 Aggregate Demand (% of GDP) 1 0 7 .9 1 0 8 .1 1 1 5 .6 1 1 6 .4 1 1 7 .7 1 2 5 .7 1 2 5 .5 1 1 9 .1 1 2 6 .7 Domestic Demand (% of GDP) 2 0 0 0 / 0 1 2 0 0 1 / 0 2 2 0 0 2 / 0 3 2 0 0 3 / 0 4 2 0 0 4 / 0 5 2 0 0 5 / 0 6 2 0 0 6 / 0 7 2 0 0 7 / 0 8 2 0 0 8 / 0 9 Public Finances (cash basis) by FY 1 8 .3 1 5 .7 1 8 .1 2 3 .0 2 6 .0 2 9 .5 3 1 .7 3 0 .1 3 2 .3 Central Government Revenues (% of GDP) of which 6 .1 4 .5 4 .5 8 .1 8 .6 1 2 .5 1 3 .6 1 0 .9 1 1 .7 Grants (% of GDP) 2 2 .2 2 0 .8 2 6 .0 2 8 .2 2 9 .8 3 0 .2 3 3 .0 3 2 .8 3 7 .9 Central Government Expenditures* (% of GDP) of which 1 9 .0 1 5 .0 2 2 .7 2 4 .1 2 5 .1 2 4 .3 2 7 .2 2 7 .2 3 1 .0 Current Expenditure 3 . 2 5 . 8 3 . 2 4 . 0 4 . 6 5 . 9 5 . 8 5 . 6 6 . 9 Capital Expenditures - 3 . 9 - 5 . 1 - 7 . 9 - 5 . 1 - 3 . 8 - 0 . 8 - 1 . 3 - 2 . 7 - 5 . 6 Central Government Overall Balance (as % of GDP) 3 . 1 - 0 . 1 - 0 . 3 0 . 1 1 . 9 0 . 2 0 . 8 1 . 4 2 . 0 Foreign Loan Financed (% of GDP) - 0 . 8 - 1 . 8 - 3 . 2 1 . 9 2 . 2 3 . 8 2 . 2 - 0 . 6 - 2 . 9 Central Government Primary Balance* (as % of GDP) 5 . 3 6 . 6 1 3 . 8 1 6 . 5 1 6 . 5 1 4 . 6 1 2 . 3 1 7 . 1 1 8 . 0 Domestic Debt (% of GDP) 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 Savings & Investment by CY 1 6 .1 1 7 .1 1 8 .2 2 0 .2 2 2 .0 2 5 .9 2 4 .5 Gross Domestic Investment (% of GDP) 0 .6 0 .7 0 .5 - 5 .5 - 3 .5 6 .8 - 2 .2 Gross Domestic Savings (% of GDP) External Developments 4 4 6 .4 4 8 0 .5 4 7 0 .8 4 8 0 .6 5 1 0 .6 5 5 0 .3 6 0 2 .0 7 9 5 .5 1 0 5 2 .1 Exports of Goods & Services (current US$) 6 3 7 .0 6 7 1 .6 8 8 5 .8 8 7 8 .3 9 7 4 .5 1 2 5 8 .7 1 3 4 5 .1 1 4 3 0 .7 2 0 9 3 .8 Import of Goods & Services (current US$) - 7 .9 - 8 .1 - 1 5 .6 - 1 6 .4 - 1 7 .7 - 2 5 .7 - 2 5 .5 - 1 9 .1 - 2 6 .7 Resource Balance (% of GDP) - 0 .2 - 0 .5 - 8 .1 - 5 .5 - 1 2 .4 - 1 7 .4 - 7 .0 - 2 .4 - 7 .0 Current Account Balance (incl. grants and debt relief, % of GDP) 1 1 .4 9 .5 1 4 .2 1 0 .4 1 6 .2 1 4 .2 1 9 .6 1 7 .9 1 4 .7 Aid inflows (incl. grants and ext. official lending, % of GDP) External Creditworthiness 1 5 6 .7 1 6 0 .5 1 4 3 .3 1 5 8 .9 1 4 8 .1 1 3 6 .8 1 4 .3 1 4 .5 1 6 .0 External Debt (% of GDP) 2 1 .8 2 0 .0 1 8 .3 2 1 .8 2 2 .9 2 1 .5 1 4 .1 3 .2 1 .0 Debt Service due (% of Exports of Goods and Services) 4 .4 3 .3 1 .5 1 .6 1 .5 1 .2 1 .0 1 .3 1 .3 Reserves as months of imports of goods and services * N o t i n c l u d i n g c h a n g e s i n d o m e s ti c a r r e a r s o r b a l a n c e s S o u r c e s : N S O , G o M ; I M F ; a n d B a n k S t a ff e s ti m a t e s - 64 - 4.7 Fiscal policy has significantly improved the macroeconomic environment. Fiscal consolidation has clearly brought Malawi a certain degree of macroeconomic stabilization; in FY03/04 Malawi was on the brink of a fiscal crisis, with over 7 percent of GDP or 27.2 percent of total expenditures allocated to servicing interest on domestic debt. In FY08/09, only 2.7 percent of GDP was allocated to domestic interest payments and domestic debt stood at 18 percent of GDP25. This is a remarkable turnaround and a major achievement by the authorities (see Table 4.1 and Figure 4.1). Figure 4.1: Malawi's progress towards a stable macroeconomic environment Macroeconomic improvements have been led by And caused in part a substantive reduction in fiscal adjustment, thru gains in the primary inflation and subsequently in rates on T-Bills balance, until recently 4.0 Overall balance 80.0 3.0 Primary Balance 91 day T-bill rate 2.0 70.0 Inflation 1.0 60.0 0.0 -1.0 50.0 -2.0 40.0 -3.0 30.0 -4.0 -5.0 20.0 -6.0 10.0 -7.0 -8.0 0.0 1 2 2 2 2 2 2 2 2 2 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 - - - - - - - - - - - - - - - - - - - 9 0 1 2 3 4 5 6 7 8 c n c n c n c n c n c n c n c n c n c / 0 / 0 / 0 / 0 / 0 / 0 / 0 / 0 / 0 / 0 e u e u e u e u e u e u e u e u e u e 0 1 2 3 4 5 6 7 8 9 D J D J D J D J D J D J D J D J D J D * These developments resulted in a reduction of And reaching HIPC and MDRI in 2006 wiped domestic debt and domestic interest payments out much of Malawi's external obligations. Interest Payments (LHS, % of GDP) External debt (percent of GDP, LHS) 8.0 20.0 180.0 25.0 Domestic Debt (RHS, % of GDP) External debt service (percent of exports, RHS) 18.0 160.0 7.0 16.0 20.0 6.0 140.0 14.0 5.0 120.0 12.0 15.0 4.0 10.0 100.0 3.0 8.0 80.0 10.0 6.0 60.0 2.0 4.0 40.0 5.0 1.0 2.0 20.0 0.0 1 2 2 2 2 2 2 2 2 2 0.0 9 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0.0 0.0 9 0 1 2 3 4 5 6 7 8 / 0 / 0 / 0 / 0 / 0 / 0 / 0 / 0 / 0 / 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009* 0 1 2 3 4 5 6 7 8 9 * Source: NSO, IMF and World Bank staff estimates. * denotes estimate 4.8 At the beginning of the decade, annual inflation hovered around 40 percent, but has since declined to 10 percent, as of end 2009. A combination of monetary tightening, fiscal consolidation, and three consecutive successful harvests contributed to this reduction. The decline in inflation has in turn led to a decline in interest rates on 25 The increase 2007/08 is due to the recapitalization of the Reserve Bank of Malawi with government bonds. - 65 - Government T-bills and bonds, as well as falling domestic lending and deposit rates (see Table 4.1.). However, the interest margin, measured as the difference between average deposit and average lending rates, stands at 20 percent and has shown little if no decline over the last decade. This compares unfavorably with other sub-Saharan African countries where the margin is closer to 10 percent. 4.9 Growth of domestic credit responded to the economic upturn and increased to an estimated 7.5 percent of GDP in 2008. Compared to the rest of sub-Saharan Africa, this level is still rather low and signals a low level of financial intermediation. Policies to improve access to banking services and finance should be considered. The Financial Sector Assessment Program (FSAP),26 completed in 2008, articulates various reform proposals that could reduce the high spreads and improve outreach and access. The main findings and recommendations are summarized in Box 4.1. Financial services in Malawi. 4.10 Reaching HIPC completion point in August 2006 significantly reduced Malawi's external indebtedness. Total external debt as a percentage of GDP stood at an unsustainable level of close to 160 percent of GDP in 2000. Qualifying for HIPC and MDRI27 has wiped out most of Malawi's external debt, which is now estimated at less than 20 percent of GDP (end 2009). This has resulted in a reduction in debt service payments from 21.8 percent of exports to an estimated 1.3 percent in 2009 (See figure 4.1). 4.11 Malawi's external financing requirements remain significant and mostly funded by official inflows. Malawi's external financing needs averaged 17.1 percent of GDP for the period 2003/07, increasing to an estimated 18.9 percent of GDP for 2008 due to negative terms of trade shocks. These financing needs were met mostly by foreign donors. Official inflows for the period 2003/07 accounted for 88 percent of total financing requirements. Foreign direct investment amounted to a low 1.1 percent of GDP for the same period. However, 2008 saw a significant increase in private inflows as investment into the new uranium mine materialized. 4.12 The external situation deteriorated as a result of terms of trade shocks in 2007 and 2008. Despite strong export performance in 2008, due to strong increases in tobacco exports fueled by higher prices, Malawi's terms of trade deteriorated sharply. The IMF estimates that the impact of high prices for oil and fertilizer on Malawi's terms of trade led to a deterioration of the current account by 1.5 percent of GDP in 2007 and 3.8 percent of GDP in 2008.28 Given the inflexibility in the exchange rate, this let to a significant decline in international reserve cover during 2008. 26 The FSAP is a joint undertaking by the World Bank and the IMF. 27 Malawi became eligible for HIPC debt relief in 2000 and reached HIPC completion in August 2006. At that time, Malawi also qualified for MDRI. 28 See IMF Country Report No. 09/16: Malawi - Request for a One-Year Exogenous Shocks Facility Arrangement. - 66 - Box 4.1: Financial services in Malawi29 The financial sector in Malawi is a concentrated system of a small number of largely privately owned banks. Financial intermediation is very weak; the ratio of private credit to GDP is only 6 percent, among the lowest in SSA; Malawi is ranked 34 out of 42 countries and compares poorly with comparator countries, such as Uganda (9.5), Tanzania (15.5) and Zambia (12.0). The financial sector is, however, very profitable; the return on average assets in 2006 was 6.4% in Malawi, more than double both the SSA average of 2.7% and the low-income country average of 3.0% (see figure 1). High profit margins in the non-lending business drive the high profitability of Malawi's banking system, accounting for two thirds of the return on average assets in 2006. Most non-interest income is from commissions and fees and the main portion derives from foreign exchange business. Figure 1: Return on average assets across countries Figure 2: Decomposition of Malawi's interest margin 7.0 6.4 100% 6.0 2005 2006 90% Tax 5.0 4.3 80% Overhead 4.0 3.7 3.4 3.5 3.3 3.1 3.0 70% 2.7 Reserve 3.0 2.4 2.5 requirements 1.7 60% 1.8 Provisions 2.0 1.5 50% 1.0 Profit Margin 40% 0.0 30% M la i M d g sca Tn n Uad S b S h ran A L wIn m H h In m o a za ia u aa gna ig co e a w aaa r 20% co e 10% 0% frica 2005 2006 Source: FSAP (2008) Source: FSAP (2008 The interest margin between lending and deposit rates has been steady at around a staggering 20 percentage points since 1997. This is very high relative to comparator countries: in 2007, the interest spread averaged 21.7 percentage points, second only to Madagascar (28.7) in sub-Saharan Africa. A breakdown of the 2005 and 2006 interest margin reveals that the vast majority of the spread between rates resulted from overhead costs (around 70% in both 2005 and 06) (see figure 2). Malawi's high overhead costs can in part be explained by the relatively small size of its banking system and the lack of significant economies of scale, as well as by a poor legal framework. Employees are an important cost driver, are highly paid, perhaps caused by labor market legislation, yet they are not very productive. Productivity, as measured by the number of accounts per employee, is extremely low; in Malawi, banks averaged 24 loan accounts and 263 deposit accounts per employee, compared to the SSA average of 505 and 742 respectively. Policy actions in the areas of macroeconomic policy, economies of scale, and contractual and information frameworks are advised to be considered to improve the situation. First, maintain macroeconomic stability and stimulate competitive pressure in lending and deposit markets. Since a significant portion of the profits is derived from the foreign exchange market, more competition in this segment would give banks further incentives to develop the lending and deposit markets more aggressively and go down market. Second, to address the challenges of small scale while maintaining competition, market-driven consolidation of small banks into a bigger bank might increase competition by creating a more powerful competitor. Third, several improvements in the contractual and information frameworks can help reduce intermediation cost and increase outreach. For example, shorten the period for foreclosure procedures and establish legal and regulatory framework for credit information sharing. 29 This box draws from the FSAP's Technical Note: Banking sector: the path to a deeper and more inclusive system (2008). See this note also for additional detail and recommendations. - 67 - 4.13 Improved economic management led to a significant increase in donor support. Foreign aid in the budget increased from 14.2 percent of GDP in FY02/03 to an estimated 17 percent of GDP for FY08/09. This has led to a level of aid dependency, where foreign inflows are financing 46 percent of total expenditure and net lending in FY08/09, up from less than 20 percent in FY02/03. The growth in foreign aid is primarily the result of increased project financing though. Figure 4.2: Malawi's lingering macroeconomic challenges Macroeconomic policies have been unable to And the authorities have found it difficult to keep aggregate demand in check increasing by increase foreign reserves 25 percent of GDP since 2000. Aggregate Demand and its components Gross usable reserves (LHS) 300 5 (percent of GDP) Gross Reserves in months of imports (RHS) Domestic Demand Exports 4.5 150.0 250 4 3.5 125.0 200 3 100.0 150 2.5 75.0 2 100 1.5 50.0 1 50 25.0 0.5 0 0 0.0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009* 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009* While the nominal exchange rate has been stable, the bilateral real exchange rates with Malawi's major trading partners have all appreciated, reducing the competitiveness of Malawi's exports and import substituting inductries. Real Bilateral Exchange (annual average, 2007 =100, increase is appreciation) 150 100 50 0 SA-Rand Euro UK-Pound USD 2007 100 100 100 100 2008 114 99 115 104 2009 118 111 143 113 Source: NSO, IMF, and World Bank staff estimates 4.14 This increase in net external financing, driven by aid and debt relief, allowed Malawi to pursue domestic demand-led growth in recent years that cannot be sustained. Between 2000 and 2009, domestic demand as a percentage of GDP grew by more than 19 percent, reaching an estimated 129 percent of GDP in 2009. This growth of domestic demand has fueled import growth, which rose to an estimated 53 percentage of - 68 - GDP in 2009 up from 26 percent in 2000. Even though the increases in domestic demand have not all been absorbed by consumption and has contributed to a projected increase in gross domestic investment, domestic savings remain well below levels that can sustain the current growth momentum. 4.15 Growth of imports of consumer goods exceeds imports of capital goods by a large margin; the share of consumer goods in total merchandise imports increased from approximately 25 percent in 2000 to well over 40 percent in 2006. Malawi's resource balance deteriorated from a deficit of 7.9 percent in 2000 to an estimated deficit of 26 percent in 2009. The surge in demands for imports, in conjunction with the current fixed exchange rate regime, will lead to a further widening of the existing difference between official and parallel exchange rates, which is estimated to have already amounted to approximately 30 percent in the first quarter of 2009. Periodic foreign exchange shortages will occur more frequently as Malawi's foreign exchange reserves and expected foreign exchange inflows will not be able to match the demand for foreign exchange at the current level of the exchange rate. 4.16 To orchestrate a soft landing of the economy a multi prone macroeconomic program needs to be urgently implemented. The priority in the short run should be to allow the currency to depreciate, to contain the growth of government (current) expenditures to well below nominal GDP growth, and not to absorb all foreign aid. A real depreciation should assist the containment of imports and provide boost to exports and the real kwacha revenues it will generate as well as make import competing industries in Malawi more competitive. A reduction of overall government expenditures in conjunction with a continued redirection of resources towards capital expenditure programs will allow resources to flow to the private sector while the change in composition will provide the needed resources to left the identified supply constraints in energy and transport. The temporary use of foreign aid to built reserves will reduce Malawi's vulnerability to terms of trade shocks (see Box 4.2 for two illustrative scenarios). 4.17 Malawi's real exchange rate appreciated sharply during 2008. The strengthening of the US dollar against the currencies of many of Malawi's trading partners and competitors has led to the appreciation of the Malawian kwacha. Malawi's exchange rate regime, recently classified as a de facto conventional peg by the IMF, has negatively affected Malawi's export competitiveness as the real exchange rate appreciated in 2008 by 20, 10, and 40 percent against the South African rand, the euro, and the British pound, respectively (see Figure 4-2). This makes it more difficult for Malawian exporters to compete internationally and for Malawian producers to compete with imports. 4.18 The appreciation of the kwacha, combined with supply constraints, in particular, the unreliable provision of electricity, has made it difficult for domestic producers to increase production to meet the rising demand for manufactured goods. In addition, international price developments of manufactured goods have been more moderate than price developments faced by local producers. This has further eroded the ability of domestic manufacturers to compete with imports to serve the domestic market. - 69 - The kwacha should be allowed to adjust to increase the competitiveness of domestic producers. Clearly, an over-valued exchange rate is not a favorable development for a country whose growth critically depends on exports and whose growth strategy is one of export diversification and value addition.30 4.19 The Government has chosen to absorb and spend the incoming foreign aid, which has made it difficult for the monetary authorities to accumulate foreign reserves. In FY07/08 45 percent of external aid resources are being used in the social development sectors and approximately 5 percent are supporting infrastructure.31 While Malawi received a massive US$2.8 billion in official aid (excluding HIPC and MDRI debt relief), the authorities were only able to add US$105 million to the foreign reserves of the Reserve Bank of Malawi during the period 2003/08 period, while 2009 saw an outflow of US$125 million reducing Malawi's reserve cover to around 0.6 month sof imports.32 A revision of the current aid absorption polices should be implemented to allow reserve accumulation. Import cover of less than one month is well below an advisable level, given Malawi's current exchange rate regime and its vulnerability to terms of trade as well as weather shocks.33 There are of course costs associated with accumulating reserves but, for a country that frequently experiences shocks requiring foreign exchange, much higher levels of reserves are necessary. As well as providing a buffer to manage terms of trade shocks, building foreign reserves is also an effective strategy for insuring against aid volatility.34 4.20 To assist the Government with the alleviation of the identified supply constraints, the Government and donors should jointly agree on a progressive redirection of aid towards investments in energy and transport, including trade logistics. Even tough aid is to a large extent fungible,35 the current allocation of approximately 5 percent of aid directed to infrastructure is not conducive to solving the supply constraints Malawi faces at this stage of its development. 30 See the Malawi Growth and Development Strategy 2006-2011 (GoM 2006b). 31 See the annual debt report for FY07/08 by the Debt and Aid Division of the Ministry of Finance. 32 Note that about half of this increase is due to the release of the SDR 34.7 million under the IMF's ESF arrangement. 33 Note that for that reason the Authorities sought the assistance of the IMF under its Exogenous Shocks Facility (ESF) arrangement in 2008 in the amount of SDR 52.05 million of which SDR 34.7 million was released in December 2008. 34 See Gelb and Eiffert (2006) 35 See Devarajan and Swaroop (1998) for a cross-country analysis of fungibility of aid. - 70 - Box 4.2: Two illustrative scenarios This boxes aims to demonstrate the net effects of the proposed policy shifts by comparing two of many possible macroeconomic scenarios. The scenarios are primarily designed to show the impact of two different broad reforms strategies. Actual outcomes fall within or outside the range described here. The first scenario aims to illustrate a development path for the Malawian economy where the proposed policy changes proposed by this report are vigorously implemented. This includes the following key elements: (i) a real depreciation of the currency; (ii) a high quality fiscal adjustment, which improves the composition of public expenditures and simultaneously reduces the size of the public sector; (iii) a continuation of the strong aid inflows of the last few years that however are not fully absorbed and as such contribute to an increase in foreign reserves; and (iv) a strong supply response resulting from the improvements in the reliability of energy supply and a reduction of the cost of trade. The second scenario aims to demonstrate the implications of policy reform scenario in which the authorities attempt to maintain the current fiscal expenditure policies as well as the current monetary and exchange rate policy stance. Table 1: Two illustrative scenarios Strong Policy Reform Scenario Business as usual Scenario illustrative scenarios 2009 2010 2015 2009 2010 2015 GDP Growth 6.7 7.1 7.0 6.7 6.5 4.0 GDP per Capita (2009 =100) 100.0 105.0 133.4 100.0 104.4 117.3 % of GDP Domestic Demand 114.8 112.8 104.7 114.9 116.4 120.2 of which Investment 27.5 27.0 28.0 27.0 25.0 16.0 Gross Domestic Savings 12.2 14.7 23.3 12.1 8.6 -4.2 CAB excluding grants -16.7 -15.3 -6.8 -16.7 -18.4 -21.6 Exports GS 26.3 29.5 34.3 25.2 23.7 21.1 Import GS 41.1 42.3 39.0 40.1 40.1 41.3 Foreign Reserves (mill. US$ 209.5 252.1 634.9 209.5 229.2 374.9 as months of imports 1.3 1.5 3.0 1.3 1.2 1.2 Real Exchange (2009 =100) 100.0 111.8 126.5 100.0 94.2 79.2 % of GDP 2008/09 2009/10 2014/15 2008/09 2009/10 2014/15 Fiscal Deficit excl. Grants -16.6 -13.9 -11.4 -16.6 -15.4 -15.9 Total Expenditures o.w. 36.0 33.0 29.6 36.0 34.4 34.0 Consumption expenditures 20.5 19.1 17.1 20.5 20.4 19.9 Gross Capital Formation 5.0 6.0 8.5 5.0 5.0 5.0 The results show very different outcomes for key economic variables under the two scenarios with Malawi being able to maintain consistent higher growth rates under the first scenario (see table 1). Under the first scenario, Malawi sustains the current growth momentum, as growth would be driven by exports, which have significant multiplier effects through the whole economy and as such broaden growth. The sustained growth performance arises from a strong private sector response due to favorable policy actions that have restored external competitiveness and a fiscal adjustment that addresses the supply constraints in energy, trade and transport, which would lead to greater confidence in the overall macroeconomic policy framework. This environment would lead to sustained high investment as a share of GDP, which in combination with improved domestic savings would sustain the projected high growth rates. In contrast, average growth under the second scenario would fall back to around 4 percent during the same period due to the inability to shift economic policy towards export promotion and therefore would remain supportive of (non-tradable) sectors which have little spillover effects on the rest of the economy and would continue to require high levels of government subsidies. Hence, public expenditures would remain high and a re-composition of expenditures to deal with supply constraints would not materialize. Consequently, private investment would fall back and export revenues would remain depended on traditional categories such as tobacco and sugar. Clearly, such a macroeconomic policy environment would not be supportive of Malawi's growth and development agenda. - 71 - Figure 4.3: Malawi's fiscal developments Even though the functional composition of expenditure improved with more resources allocated to economic services in particular transport and agriculture, Composition of Public Expenditures (% of total non-interest expenditures) 45 40 35 30 25 20 15 10 5 0 General Social Agriculture Economic Administratio Services and Education Health Other and Natural Transport Other Services n community Resources 2000/01 to 2003/04 41 40 16 11 13 18 7 8 3 2004/05 to 2007/08 38 32 12 11 8 31 13 15 3 The size of government, measured by total And the additional fiscal resources have expenditures as a % of GDP, increased and is facilitated a decline in domestic debt, but have well above the average for Malawi's peers, mostly been absorbed by current expenditure 40.0 programs. Malawi 03/04-07/08 03/04-08/09 03/04-08/09 35.0 Ghana Namibia (% of GDP) (% of GDP) % of total Sources of Fiscal space 13.1 15.5 100.0 Angola Domestic Revenues 4.2 5.7 36.5 30.0 Mozambique Interest payments 4.8 4.3 28.0 Kenya Botswana Grants 2.8 3.6 23.2 Gambia Tanz External financing 1.3 1.9 12.3 25.0 Burkina South Africa Rwanda Zambia Uses of Fiscal space 13.1 15.5 100.0 Ethiopia Madagascar Total Expenditures 9.3 14.2 91.5 20.0 Niger Uganda Current Expenditures 7.7 11.3 73.2 Congo o.w. Fertilizer and seed subsidy 2.7 5.8 37.5 Capital Expenditures 1.6 2.8 18.3 15.0 o.w. Gross fixed capital formation 1.8 1.2 7.8 4.50 5.50 6.50 7.50 8.50 Domestic Debt repayments 3.8 1.3 8.5 GDP per capita (USD, natural log) Source: NSO, IMF, and World Bank staff estimates 4.21 Fiscal consolidation is still vulnerable, as certain Government expenditure programs have grown exponentially over the last few years. In 2007 and 2008, growth of the fertilizer and seed program can be partly explained by extraordinary increases in international prices for the goods procured by the Government. However, further increases are the result of a loosening of the eligibility criteria of the program. Farmers growing cash crops such as tea and tobacco are now eligible for assistance, as well as farmers growing food crops. Clearly, the program is moving away from its original goal of improving food security. The result has been that, in 2008/09, 5.8 percent of GDP or 15.3 percent of total expenditures is estimated to have been allocated to this program alone. - 72 - Maintaining the input subsidy program at this expanded level will make it difficult for the Government to direct resources to priority public investments without jeopardizing recent fiscal consolidation. 4.22 Malawi's expenditure programs are heavily tilted towards recurrent expenditure programs and the country has lacked the needed investment to upgrade and maintain its infrastructure. While foreign aid is available to help finance some of these much-needed investments, the required new investments in energy and maintenance in transport would also need significant contributions from the budget, and a consequent shift in public expenditures from current expenditures ­ the fertilizer and seed program and expenditure on goods and services ­ towards investments and maintenance.36 A redirection of public expenditure towards programs that improve the productivity of the economy should be considered. This report argues that the main constraints in agriculture, transport, and energy to sustain Malawi's current growth momentum require immediate additional resources of approximately 5 percent of GDP annually over the coming three to four years. An additional 1.5 percent of GDP should be allocated to the roads budget37 as to be able to adequately maintain Malawi's current road network. Of the remaining 3.5 percent of GDP, 2.2 percent of GDP would need to be allocated to finance investments in power38 and 1.3 to agriculture. 4.23 In the case of agriculture, the increased spending on the fertilizer and seed subsidy program has crowded out the resources available to the sector to implement the investment priorities identified in the Agriculture Development Program (ADP). To bring the ADP back on track would require an additional 1.3 percent of annually resources to be allocated to agriculture on and above the planned budget for ADP implementation. Part of the savings needed to finance the proposed expenditures, can be generated by containing the fertilizer and seed subsidy program, which was originally costed at less than 2 percent of GDP, while the program is estimated to absorb 5.8 percent of GDP in FY08/09. Other savings to finance the needed increases could be generated by vigorously seeking expenditure savings in non-priority sectors the joint Public Expenditure review of the Government of Malawi and the World Bank (2007) provides a variety of suggestions on where efficiency improvements can be made and savings accumulated. 36 The recently finished PER (GoM and World Bank (2007)) discusses the issues of expenditure composition in more detail and provides several suggestions to improve the allocation of expenditures towards infrastructure. In addition, using the so-called MAMS model, the PER also analyzes the impact on growth of an increase of infrastructural expenditures. It concludes that an increase in infrastructural spending increases growth by 0.3 percent annually while redirecting expenditures to human development decreases growth by 0.4 percent annually. 37 Section 3.1 identified an annual shortfall of Mk7.7 billion in road maintenance which equals 1.5 percent of GDP in 2008. 38 Section 3.3 identifies investment needs in Kapachira, the transmission interconnector, an energy savings light bulb program, and initiation of Lower Fufu, amounting roughly to 2.2 percent of GDP annually over the next three to four years. - 73 - 4.24 Despite the reduction of the fiscal deficit, total expenditures increased from around 21.8 percent of GDP in FY00/01 to an estimated 37.9 percent in FY08/09, and public consumption (public expenditure on wages, salaries, goods and services) has increased by over 10 percent of GDP since FY00/01 to an estimated 19.5 percent of GDP in FY08/09. Even though the current government has made an effort to increase development expenditures, it has achieved only a modest increase of around three percent of GDP in expenditure on gross capital formation during the period FY00/01 - FY08/09. (see Table 4.1 and Figure 4.3). A smaller public sector would provide greater space for private sector development and promote growth. The size of Malawi's public sector as measured by total expenditures and net lending, as a percentage of GDP is relatively large for its income level. The burden of fiscal adjustment will have to be borne initially by expenditures; but once the primary deficit has been brought back in line with a sustainable trajectory of Malawi's debt, reductions in the tax burden in particular on transport related goods and services can also be considered. The government could usefully adopt a fiscal rule which would limit the growth of overall public expenditures to, say, anticipated growth in nominal GDP minus 2.5 percent which will reduce total public expenditures as a ratio to GDP by about 1 percent of GDP annually. Even though this might seem a formidable challenge in the presence of the proposed redirection of public expenditures, the rule does not require a real reduction in spending on existing programs. Maintaining the fertilizer and seed program at current real levels would reduce it by approximately three percent of GDP in FY09/10 and in conjunction with the containment of the fiscal deficit would allow for a further reduction in domestic borrowing translating in a reduction of domestic interest payments by the public sector. Cumulative and in combination with containing real growth of existing current expenditure programs to 4 percent, which equals approximately nominal growth minus 2.5 percent, would create the fiscal space needed to increase public investment and allocate additional resources to road maintenance programs called for. 4.25 Such a program if implemented prudently could lead not only to a soft landing but also lead to an improved savings investment balance for the economy as a whole improving further Malawi's growth prospects. High-growth economies typically maintain prudent macro economic policies and save a substantial share of their income for investment. The Growth Commission39 recently observed that the high- growth economies investigated in the report on average had savings rates of 20-25 percent to finance public and private investment. Malawi's domestic savings rate of around 7 percent is well below this level and a major challenge will be to boost domestic 39 See http:www.growthcommission.org for a full explanation of the scope of the commissions work, its findings and various background papers. - 74 - savings to provide the necessary domestic resources to finance much-needed investment. 40 4.26 A sustained reduction in overall size of the public sector while containing the fiscal deficit and improving the composition of public expenditures through the policies described above would improve the overall macroeconomic policy mix. Different combinations of monetary and fiscal policies can yield the same level of aggregate demand, while resulting in different outcomes for the real exchange rate, real interest rates, the current account balance and composition of output. In the current Malawian context, a combination of tighter fiscal policy and better targeted expenditure policies towards improving productivity of the economy would imply a less appreciated exchange rate, a higher share of investments and a reduction in the current account balance. This would allow aggregate demand to move in line with GDP and improve the growth prospects of the economy as a whole. 40 See also Box 4.1 on financial services in Malawi, which discusses how to improve the functioning of the financial sector as a means to improve intermediation but also access and thus the effectiveness to mobilize savings. - 75 - REFERENCES Adam and Bevan, 2006. "Aid and the Supply Side: Public Investment, Export Performance, and Dutch Disease in Low-Income Countries," World Bank Economic Review, Oxford University Press, vol. 20(2), pages 261-290. Asmah and Levin, 2008, "Aid-Financed Public Investments and the Dutch Disease: Evidence from Tanzania", Working paper No. 3, Örebro University, Örebro. Bell, Deverajan, and Gersbach, 2003, "The Long-Run Economic Costs of AIDS: Theory and an Application to South Africa", World Bank Policy Research Working Paper No. 3152, The World Bank, Washington, D.C. Canning, 2006. "The Economics of HIV/AIDS in Low-Income Countries: The Case for Prevention," Journal of Economic Perspectives, American Economic Association, vol. 20(3), pages 121-142. Department for International Development, 2007, Gender Equality and Growth, Evidence and Action, Department for International Development, London, UK. Devarajan and Swaroop, 1998, "The Implications of Foreign Aid Fungibility for Development Assistance", World Bank Policy Research Working Paper No. 2022, The World Bank, Washington, D.C. Diao, Hazell, Resnick, and Thurlow, 2006. "The Role of Agriculture in Development: Implications for Sub-Saharan Africa", DSG Discussion Paper 29, IFPRI, Washington, D.C. Easterly, 2006, The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill and So Little Good, New York: Penguin Press. Easterly and Fischer, 2001, "Inflation and the Poor," Journal of Money, Credit and Banking, Blackwell Publishing, vol. 33(2), pages 160-78. Easterly, William & Rebelo, Sergio, 1993. "Fiscal policy and economic growth: An empirical investigation," Journal of Monetary Economics, Vol. 32(3), pages 417- 458 Eifert and Gelb, 2006, "Improving the Dynamics of Aid: Toward More Predictable Budget Support," in Budget Support as More Effective Aid? Recent Experiences and Emerging Lessons, ed. by Koeberle, Stravreski, and Walliser. World Bank, Washington, D.C. - 76 - Fischer, 1993, "The Role of. Macroeconomic Factors in Growth," Journal of. Monetary Economics, Vol. 32, pages 485-512. Government of Malawi, 2004, The National Transport Policy, Ministry of Transport, Lilongwe. Government of Malawi, 2006a, The 2004-2005 Integrated Household Survey Report, National Statistical Office, Zomba. Government of Malawi, 2006b, Malawi Growth and Development Strategy 2006-2011, Government of Malawi, Lilongwe. Government of Malawi, 2007, Agriculture Development Programme, Ministry of Agriculture and Food Security, Lilongwe. Government of Malawi, Various years, Public Finance National Reports, Ministry of Finance, Lilongwe. Government of Malawi and World Bank, 2006, "Malawi: Poverty and Vulnerability Assessment Investing in our Future", World Bank Report No. 36546-MW, The World Bank, Washington D.C. Government of Malawi and World Bank, 2007, "Malawi Public Expenditures Review", World Bank Report No. 40145-MW, The World Bank, Washington D.C. GTZ, 2008, International Fuel Prices, 6th edition, Deutsche Gesellschaft fur Technische Zusammenarbeit, Eschborn. Hausmann, Rodrik, and Velasco, 2005, "Growth Diagnostics", John F. Kennedy School of Government, Harvard University, Cambridge, Massachusetts. IMF, 2008, "Malawi - IMF Country Report No. 08/3", International Monetary Fund, Washington, D.C. IMF, 2009, "Malawi - IMF Country Report No. 09/16", International Monetary Fund, Washington, D.C. IMF and World Bank, 2007, "Financial Sector Assessment Program Malawi. Banking Sector: The path to a deeper and more inclusive system", International Monetary Fund and World Bank, Washington D.C. IFPRI, 2008, High Food Prices: The What, Who, and How of Proposed Policy Actions, Policy Brief, IFPRI, Washington D.C. - 77 - Ivanic and Martin, 2008, "Implications of Higher Global Food Prices for Poverty in Low- Income Countries", World Bank Policy Research Working Paper No. 4594, The World Bank, Washington, D.C. Kevane and Wydick, 2001, "Microenterprise Lending to Female Entrepreneurs: Sacrificing Economic Growth for Poverty Alleviation?," World Development, Vol. 29, No. 7, pp. 1225-36. Kraay and Nehru, 2006, "When Is External Debt Sustainable?," World Bank Economic Review, Oxford University Press, vol. 20(3), pages 341-365 Limao and Venables, 2001, "Infrastructure, Geographical Disadvantage and Transport Costs". World Bank Economic Review 15 (3): 451­79. OECD-FAO, 2008, Agricultural Outlook: 2008-2017: Highlights, Paris and Rome, Poulton, Tyler, Dorward, Hazell, Kydd and Stockbridge, 2007, "All-Africa Review of Experiences with Commercial Agriculture: Summary report", Mimeo, World Bank, Washington, D.C. Pedersen, 2001, "Freight Transport under Globalization and Its Impact on Africa". Journal of Transport Geography 9: 85­99. Rajan and Subramanian, 2005, "What Undermines Aid's Impact on Growth?" IMF Working Paper No. 05/26, International Monetary Fund, Washington D.C. Rizet and Hine, 1993, "A Comparison of the Costs and Productivity of Road Freight Transport in Africa and Pakistan." Transport Reviews 13(2). Schultz, 2002, "Why Governments Should Invest More to Educate Girls," World Development, Vol. 30, No.2, pp. 207-25. Stotsky, 2006, "Gender and Its Relevance to Macroeconomic Policy: A Survey, IMF Working Paper No. 06/233, International Monetary Fund, Washington D.C. Teravaninthorn and Raballand, 2008, Transport Prices and Costs in Africa: A Review of the International Corridors, World Bank, Washington, D.C. The Growth Commission, 2008, "The Growth Report: Strategies for Sustained Growth and Inclusive Development", World Bank, Washington, DC. Thomas, 2002, `Health, Nutrition and Prosperity: A Microeconomic perspective.' Bulletin of the World Health Organization 80(2), World Health Organization. - 78 - UNAIDS, WHO, and UNICEF, 2008, Country Epidemiological Fact Sheet on HIV and AIDS. 2008 Update, various countries, UNAIDS/WHO, Geneva, Switzerland. United Nations, "Almaty Programme of Action: Addressing the Special Needs of Landlocked Developing Countries within a New Global Framework for Transit Transport Cooperation for Landlocked and Transit Developing Countries", 2004. World Bank, 2001, Engendering Development, A World Bank Policy Research Report, World Bank and Oxford University Press, Washington, D.C. World Bank, 2006a, Malawi Investment Climate Assessment, World Bank, Washington, D.C. World Bank, 2006b, World Development Indicators, World Bank, Washington, D.C. World Bank, 2007, Connecting to Compete: Trade Logistics in the Global Economy, World Bank, Washington, D.C. World Bank, 2008a, World Development Indicators, World Bank, Washington, D.C. World Bank, 2008b, World Development Report 2009 ­ Reshaping Economic Geography, Oxford University Press. World Bank, 2008c, Underpowered: The State of the Power Sector in Sub-Saharan Africa, Africa Infrastructure Diagnostic, World Bank, Washington, D.C. - 79 -